GSA Special Publication No.12: Eromanga Basin 1986 by GSAustralia

OAVENP0RT ^GWjt

W i &OftAW

M3 ,f HOWittl 50WNS

GOYDER5 LAOOOf

CONTRIBUTIONS TO

CARBON POTENTIAL PRCUM.NAftY SKETCH PLAN Qf AREAS HELD SY * SANTOS

N£. SOfTH

LTD,

AUSTRALIA SHOWING

& S.W. QUEEfs

PRINCIPAL

j M E S O Z O I C STRUCTURES I EQIf ; OwtY , • t **' • O&Q OOvi

Geological Society of Australia Incorporated President:

D. M. Boyd

Vice-Presidents: J. B. Waterhouse, I. R. Johnson Hon. Secretary: Hon.

A. P. Belperio

Treasurer: J. W. Hunt

Hon. Administrative

Officer:

D. H. Probert

Copies of Special Publications and the Society's Journal are available from:

The Business Manager, Geological Society of Australia Incorporated Room 1001, Challis House, 10 Martin Place Sydney 2000, N.S.W. Australia

Printed by Graphic Services Pty. Ltd., Northfield, S.A.

CONTRIBUTIONS TO THE GEOLOGY AND HYDROCARBON POTENTIAL OF THE EROMANGA BASIN Editors: D. I. Gravestock, P. S. Moore & G. M. Pitt

1986 SPECIAL PUBLICATION NO. 12 Geological Society of Australia Incorporated

National Library of Australia ISBN 0 909869 39 1 ISSN 0072 1085

Foreword

Contents

P. S. MOORE:

An exploration overview of the Eromanga Basin Historical Perspective

The Eromanga Basin in the search for commercial hydrocarbons The Eromanga Basin, an overview of exploration and potential Regional Basin Studies

9 25

R. C. SPRIGG:

J. D. ARMSTRONG & T. M. BARR: P. S. MOORE:

Jurassic and Triassic stratigraphy and hydrocarbon potential of the Poolowanna Trough (Simpson Desert Region), northern South Australia 39 Palynology, cyclic sedimentation, and palaeoenvironments in the Late Mesozoic of the Eromanga Basin 53 G. AMBROSE, R. SUTTILL & I. LAVERING: The geology and hydrocarbon potential of the Murta Member (Mooga Formation) in the southern Eromanga Basin 71 B. G. FORBES: Margin of the Eromanga Basin South Australia: a review 85 P. S. MOORE, G. M. PITT & M. E. DETTMANN: The Early Cretaceous Coorikiana Sandstone and Toolebuc Formation: their recognition and stratigraphic relationship in the southwestern Eromanga Basin 97 V. SCHEIBNEROVA: Marine Cretaceous of the Great Australian Basin — foraminiferal and palynological zonations reconciled 115 S. OZIMIC: The geology and petrophysics of the Toolebuc Formation and its time equivalents, Eromanga and Carpentaria Basins 119 A. MCMINN & D. BURGER: Palynology and palaeoenvironments of the Toolebuc Formation (sensu lato) in the Eromanga Basin 139 Structure and Tectonics

D. BURGER:

J. LOCK, C. D. N. COLLINS & D. M . FINLAYSON:

Basement structure and velocities under the central Eromanga Basin from seismic refraction studies 155 The Canaway Fault and its effect on the Eromanga Basin 163 G. W. KRIEG: Stratigraphy and tectonics of the Dalhousie Anticline, southwest Eromanga Basin 175 Petroleum Reservoirs J. PINCHIN & V. ANFILOFF:

0 . J. W. BOWERING & D. M . HARRISON:

The Merrimelia Oil and Gas Field — a case history The Birkhead Formation — a Jurassic petroleum reservoir Source Rock Geochemistry and Organic Petrology

1. M. PATON:

183 195

A. C. COOK:

The nature and significance of the organic facies in the Eromanga Basin 203 Source rock evaluation and maturation history of the central Eromanga Basin ... 221

V. L. PASSMORE & C. J. BOREHAM: J. D. SAXBY:

Geochemistry of oil shale in the eastern Eromanga Basin Organic matter in the Toolebuc Formation

N. R. SHERWOOD & A . C. COOK:

241 255

K. W. RILEY & J. D. SAXBY:

Organic matter and vanadium in the Toolebuc Formation, northern Eromanga Basin and southern Carpentaria Basin 267

M. GLIKSON & G. H . TAYLOR:

Cyanobacterial mats: major contributors to the organic matter in Toolebuc Formation oil shales 273 Organic geochemical facies of the Cretaceous Bulldog Shale, western Eromanga Basin, South Australia 287 Hydrocarbon Generation and Migration D. M. MCKIRDY, J. K. EMMETT, B. A. MOONEY, R. E. COX & B. L. WATSON:

A. J. KANTSLER, A. C. COOK & M . ZWIGULIS:

Organic maturation in the Eromanga Basin

G. M. PITT:

305

Geothermal gradients, geothermal histories and the timing of thermal maturation in the Eromanga-Cooper Basins 323

M. A. HABERMEHL:

Regional groundwater movement, hydrochemistry and hydrocarbon migration in the Eromanga BaSin 353

A. F. WILLIAMS & K. MORIARTY:

Hydrocarbon flushing in the Eromanga Basin — fact or fallacy?

377

Geological Society of Australia Special Publication No. 12, 1-8

An exploration overview of the Eromanga Basin P. S. Moore

Delhi Petroleum Pty. Ltd., 101 Grenfell St., Adelaide, S.A. 5000. ABSTRACT Significant progress has been made since 1977, when Poolowanna 1 flowed oil from the Eromanga Basin sequence. In the last seven years, over 350 petroleum wells have been drilled in the basin and several tens of thousands of kilometres of seismic data have been recorded. The exploration effort has led to the discovery of over 50 hydrocarbon pools. Two major development programmes (the 'Liquids Project' in South Australia and the Jackson Project in Queensland) are operational, with an expenditure to date of over SI.6 billion. Significant progress has also been made since November 1982, when the Eromanga Basin Symposium summarised some of these data. Thus, the following compilation of papers bears little resemblance to the original Symposium proceedings. Of the 26 papers reproduced herein, 4 are entirely new and another 6 appeared only in abstract form at the Symposium. The remainder have been revised. The papers in this volume thus provide a renewed contribution to the geology and hydrocarbon potential of the Eromanga Basin. INTRODUCTION This paper is designed to provide an overview of the stratigraphy and petroleum potential of the Eromanga Basin, in order to place the following 26 articles in perspective. Other major review articles on the Eromanga Basin are those by Sprigg (1986) and Armstrong & Barr (1986), which also appear in this volume. The Eromanga Basin, of Jurassic and Cretaceous age, extends over 1 million square kilometres of central Australia. The basin is productive of both oil and gas (Fig. 1) and is considered to be the most prospective onshore area in Australia for oil exploration. Indeed, exploration levels are high by Australian standards. During 1984 approximately 20,000 line kilometres of seismic were recorded and 100 exploration wells drilled. The liquids development phase will have cost approximately $2 billion by 1986 (Swindon & Moore, in press). With such high levels of expenditure, it is natural that a large amount of information should result on the geology and hydrocarbon potential of the Eromanga Basin. It is estimated that approximately 200 geoscientists are involved with studies of the Eromanga Basin. They are derived from petroleum exploration companies, State Geological Surveys, the Federal Bureau of Mineral Resources, educational institutions and various consultant groups. Each group has its own set of priorities and objectives, and each generally works in a particular part of the basin, since the area is too large to be studied as a whole (at least, not in any detail). Fortunately, the stratigraphy of the Eromanga Basin can be summarised in fairly simple terms, and this summary is presented below. More details are, of course, contained in the 8 papers on Regional Basin Studies. This paper also reviews basic details relating to the nature and location of hydrocarbon discoveries in the Eromanga Basin. Again, more details are contained in papers by Ambrose et al. (1986), Armstrong & Barr (1986), Bowering

& Harrison (1986), Kantsler et al. (1986), Moore (1986) and Paton (1986), all in this volume. Finally, the status of petroleum development and production is presented, since it is here that a major part of the expenditure is committed. SUMMARY OF STRATIGRAPHY AND DEPOSITIONAL ENVIRONMENTS The Eromanga Basin is part of the hydrogeological Great Artesian Basin (Habermehl, 1986). Other constituents are the Carpentaria and Surat Basins, which are linked to the Eromanga Basin across shallow basement ridges (Fig. 2). The Great Artesian Basin is underlain by a variety of rock types constituting older sedimentary basins and cratonised areas. It is overlain by a relatively thin veneer of fluviatile, lacustrine and aeolian sediment assigned principally to the Lake Eyre Basin. The Great Artesian Basin is one of the world's largest artesian systems, occupying an area of 1.7 million square kilometres or one-fifth of the Australian continent. Confined aquifers occur mainly in fluviatile sandstones of Jurassic and Early Cretaceous age. The basin forms a large synclinal structure, uplifted and exposed along its eastern margin and tilted southwest. Recharge occurs mainly in the east, with discharge in the south and southwest (Habermehl, 1986). Stratigraphic relationships, as observed in the subsurface, are summarised in Figure 3. Only the most important formations are listed, since many local names are used to describe facies variations around the margins of the Eromanga Basin. In Figure 3 and in the following text, reference is also made to the Surat Basin, since it is there that many units crop out and have been studied in detail. Furthermore, facies variations between the Eromanga and Surat Basins throw considerable light on the tectonic evolution of the Great Artesian Basin as a whole.

P. S. MOORE Surat and Carpentaria Basins finally were linked in the late Early Jurassic by the development of an extensive, bedloaddominated, sandy fluviatile system which spread over the entire region. Sediment was derived from a generally westerly direction, with the coarsest and mineralogically least mature deposits occurring within the Algebuckina Sandstone (Wopfner et al., 1970). The laterally equivalent Hutton Sandstone is better sorted and rounded, and contains calcrete and silcrete horizons in its upper portions (Gravestock et al., 1983). Further east in the Surat Basin, the Hutton Sandstone consists of lacustrine delta-fill deposits at the base (Wiltshire, 1982) overlain by sandy, bedload-dominated fluviatile deposits (Power & Devine, 1970; Exon, 1976). Throughout the entire Eromanga-Surat Basin area, sediment transport was from the west, northwest and southwest (Exon, 1976; Martin, 1981; Moore, 1986). The great lateral extent of the Hutton Sandstone and other braided-fluviatile units within the Eromanga Basin has puzzled sedimentologists for decades. Exon (1976), studying sequences in the Surat Basin, suggested that their deposition was in response to a sharp, eustatic drop in sea level. He argued that this would cause a sudden change in the base level of erosion resulting in high-energy, braided-fluviatile deposition. Subsequent deposits, formed during periods of stability or rising sea level, would be increasingly finer grained, terminating in lacustrine or swamp conditions. According to Day et al. (1983) this suggestion has considerable merit, because of the tectonic stability of the Australian craton and inferred connection between the low-lying Great Artesian Fig. 1. Location of the Jurassic-Cretaceous Eromanga Basin, central Basin and the open sea to the east. Alternatively, such cycles eastern Australia. Moomba was originally constructed in 1968 may be tectonic in origin, resulting from uplift and erosion for the gathering and processing of natural gas produced from beyond the basin margins (e.g. Gawler Craton, Wopfner et the underlying Permo-Triassic Cooper Basin. Oil and natural al1970), prior to the onset of rifting between Australia and gas liquids processing facilities were added in 1982-83, in order Antarctica. Both cases may have operated during the evolution to produce liquid hydrocarbons from both the Cooper and of the basin. Eromanga Basins. The liquids are transported from Moomba However, regardless of their origin, it is apparent that within to Point Bonython via a 669 km pipeline. Only Eromanga Basin oil flows along the second, recently completed pipeline, the Jurassic and earliest Cretaceous sequences of the Great from Jackson to Moonie. Moonie is the liquids collection Artesian Basin, laterally extensive braided-fluviatile sandstones centre for the Surat Basin, and has had a pipeline link with occur interbedded with floodplain, swamp and lacustrine Brisbane since 1963. deposits. Thus, the braided-fluviatile Hutton Sandstone is overlain by finer-grained floodplain deposits of the Birkhead Sedimentation in the Great Artesian Basin commenced Formation (Paton, 1986) and swamp deposits of the Walloon roughly synchronously in the Surat and Eromanga Basins. Coal Measures. The succeeding cycles (Adori-Westbourne and In the Surat Basin the oldest deposit is the braided-fluviatile Gubberamunda-Orallo; Fig. 3) are similar although less Precipice Sandstone. According to Martin (1981), cross- pronounced, with the latter being confined to the finer-grained bedding orientations indicate transport from the west, while Surat Basin sequence. The Murta Member has no lithological lithological characteristics suggest a metamorphosed equivalent in the Surat Basin, and is lacustrine in origin. This Precambrian source which conceivably included the Willyama unit is discussed in detail by Ambrose et al. (1986). Complex (Fig. 2). Non-marine sedimentation probably persisted until the end The first sediment to be deposited in the Eromanga Basin of the Neocomian, although a marine influence has been was laid down in the Poolowanna Trough (Figs 2, 3). The basal suggested for the upper sandy part of the Cadna-owie unit, known as the Poolowanna Formation (Moore, 1986) Formation and its lateral equivalent (the Bungil Formation) consists of interbedded sandstone, siltstone, shale and thin in the Surat Basin (Day et al., 1983). A major marine coal beds, deposited in a moderate energy, meandering transgression occurred in the earliest Aptian and deposited fluviatile environment with associated floodplains and mudstone throughout the area. Large channels incised into swamps. Synchronously, in the Surat Basin the Evergreen the Cadna-owie Formation at the end of the previous Formation was being laid down under slightly more quiescent regressive phase also appear to have been filled with fineconditions (Porter, 1979; Wiltshire, 1982). Lacustrine deposits grained marine sediment (Moore & Pitt, 1984). in the upper part of the Evergreen Formation contain acritarch The Aptian to Albian marine sequence is up to 700m thick swarms which may indicate a partial connection with the open in the Eromanga Basin and is relatively homogeneous, apart sea, which lay further to the east (Evans, 1962). from the Toolebuc Formation which locally develops oil shale In the late Early Jurassic, deposition expanded beyond the facies (McMinn & Burger, 1986; Ozimic, 1986; Saxby, 1986; Poolowanna Trough and Surat Basin areas, as indicated by Riley & Saxby, 1986) and the Coorikiana Sandstone, which the sporadic development of low-energy, fluviatile deposits is a thin shoreface sandstone (Moore et al., 1986). Overlying in the central Eromanga Basin ('Cuttapirrie beds', Fig. 3; this sequence is the Mackunda Formation, which is a marginal otherwise known as the 'basal Jurassic unit). The Eromanga, marine, regressive sandy siltstone. Finally, non-marine

FOREWORD, EXPLORATION OVERVIEW

Fig. 2. Geological setting. The Eromanga Basin, together with the Carpentaria and Surat Basins, forms part of the hydrogeological Great Artesian Basin. Several Late Carboniferous to Late Triassic sedimentary basins underlie the Great Artesian Basin and partly control its distribution and sediment thickness. Shaded areas represent pre-Carboniferous outcrops or highly deformed, Permo-Triassic rocks. Cross-section AB is orientated roughly east-west and passes through the deepest parts of the Eromanga and Surat Basins.

sedimentation dominated in the Cenomanian, with the deposition of a 1000 m thick coal-bearing, fluviatile Winton Formation.

HYDROCARBON DISCOVERIES

Hydrocarbon discoveries have been made at eight stratigraphic horizons within the Eromanga Basin (Fig. 4;

Table 1). All commercial discoveries are located within the non-marine, Jurassic to earliest Cretaceous part of the sequence (Fig. 2). Most of the traps are anticlinal although some are stratigraphic. The first Eromanga Basin discovery was in 1976, when gas flowed to surface at the rate of 13.9 million cubic feet per day (MMCFD) from the Namur Sandstone Member in the Namur 1 well (Armstrong & Barr, 1986). At the time, it was

P. S. MOORE AGE

EARLY TURONIAN LATE CENOMANIAN K EARLY CRETACEOUS

ALBIAN APTIAN

SUBSURFACE

STRATIGRAPHIC

NOMENCLATURE

EROMANGA BASIN

SURAT

MACKUNDA FORMATION S * A L L A R u MUDSTONE * nnnN*nATTA TM f MARREE f * TQOLEBUC FORMATION F WALLUMBILLA FORMATION # ^ BULLDOG SHALE

MUDST0NE

GRIMAN CREEK

^lllllPwr

BASIN

^ ™ \ TOOLEBUC FORMATION ^ SUP^T SIIT^T™* \ * WALLUMBILLA FORMATION " I ALLARU

BUNGIL FORMATION

CADNA-OWIE FORMATION

\ \

NEOCOMIAN

LATE JURASSIC

MIDDLE

EARLY

JURASSIC

\ S S S I I \ 1 V I \ \

^ MOOGA Z UJ ~Z MP aa o<

MURTA MEMBER^~^» . ^ ^ ^H- «

FORMATION

€

<U)

#•

— . „ 5-^VVESTBOURNE Fjj ^^ADomjs E A D FORMATION

* *

GUBBERAMUNDA SANDSTO 2 — —

"^WALLOON COAL MEASURES

HUTTON SANDSTONE NEBINE

^FORMATIOJ^T'

RIDGE

PATCHAWARRA

AND

NAPPAMERRI

TROUGHS

j —

GAS DISCOVERY

—* EVERGREEN FM •

-/ / /

^V^fRECIPICEte^ r^

TROUGH

• OIL DISCOVERY

— ORALLO FORMATION*

* H

s ' • * V^^XUTTAPIR^BEDSVV^^ S POOLOWANNA r ^ POOLOWANNA

MOOGA SANDSTONE

HOORAY SANDSTONE

NAMUR SANDSTONE MEMBER

* SOURCE ROCK

Fig. 3. Major stratigraphic units within the Eromanga and Surat Basins. Line of section (AB) shown in Fig. 2. Cross hatching indicates significant erosion.

recovery of oil in the drillpipe following a drill-stem test (Table 1). Not all discoveries are presently economically producible and some have yet to be evaluated by appraisal drilling.

believed that the gas has a Permian source and has migrated to its present site up a fault. In 1977, Poolowanna 1 flowed a small amount of viscous, waxy oil from the Poolowanna Formation in the Simpson Desert region (Moore, 1986). Both discoveries aroused interest in the Jurassic sequence, although they were not representative of the discoveries which followed.

The first major oil discovery occurred in 1981, when the Strzelecki 3 well flowed at the rate of 2400 barrels of oil per day (BOPD) from the Hutton Sandstone (Fig. 5a). By then, it was realised that oil could be sourced from within the Jurassic sequence and trapped at several different horizons. Also, it was becoming apparent that Jurassic reservoirs were

FOREWORD, EXPLORATION OVERVIEW

Fig. 5. Geological cross-sections of the Strzelecki, Dullingari, Jackson and Merrimelia Fields. Field locations shown in Fig. 4.

areally quite small (average 1000-3000 acres), requiring a tight (1 km) seismic grid to accurately locate the crest of the anticlines. Exploration in the period 1978-1981 resulted in several successes, the most notable of which was the discovery of a stratigraphic accumulation of oil in the Murta Member in Dullingari (Mount, 1981; Ambrose et al., 1986). The reservoir sand is a lacustrine-beach deposit less than 1 m thick (Fig. 5b). However, the sandstone has excellent reservoir properties, with porosities of about 16%, permeabilities of up to 3 darcies and flow rates in excess of 2000 BOPD (Mount, 1982). In December 1981, the Jackson oilfield was discovered in southwestern Queensland (Halyburton & Robertson, 1984). Oil was found at three stratigraphic horizons, although the largest accumulation occurred in the Hutton Sandstone (Fig. 5c). This was the first oil discovery in the Queensland portion of the Eromanga Basin and the largest discovery to date. At 31st March 1984, total proved and probable reserves in the Jackson Field were estimated by Delhi Petroleum Pty Ltd at 42 million barrels of oil, of which 38 million barrels were contained in the Hutton Sandstone. The Jackson discovery was particularly significant since it established the potential of the Jackson-Naccowlah trend, which flanks the northeastern end of the Nappamerri Trough. Seventeen oil and 3 gas pools have since been discovered along this trend (Table 1), with exploration still in progress. Meanwhile, in South Australia additional wells were being drilled to intersect the crests of structures which previously

had been drilled down-dip and had either proved to be dry or were established Permian gas fields. Thus, small pools of oil were discovered in Dullingari 22, Merrimelia 6, 8 and 9, Strzelecki 3 and 4, Big Lake 26, Narcoonowie 2, and Gidgealpa 17. At August 1984, 58 oil and 8 gas pools had been discovered in the Eromanga Basin, with proved and probable in-place hydrocarbons totalling over 250 million barrels of oil and 75 BCF of gas. Most of the hydrocarbons are contained within structural traps at the top of the Hutton Sandstone, particularly along the Jackson-Naccowlah trend. While the majority of the discoveries are located around the flanks of the mature Nappamerri Trough, recent oil discoveries in Tintaburra 1, Charo 1 and Bodalla South 1 (Fig. 4) have led to a reappraisal of the potential of the Eromanga Basin.

DEVELOPMENT AND PRODUCTION The South Australian 'Liquids Project'

The concept of selling crude oil and natural gas liquids from the Cooper Basin was first suggested in 1970 following discovery of the Early Permian Tirrawarra Oilfield. However, the remote location of the fields, their relatively small sizes and large interfield distances made development uneconomical. In the late 1970s, circumstances altered with a significant worldwide increase in the price of crude oil. This stimulated exploration, which in turn led to the discovery of more oil and wet-gas fields in the Cooper Basin and the

P. S. MOORE

TABLE 1 Eromanga Basin hydrocarbon discoveries Discovery Well

Discovery Date

A. Poolowanna Formation Poolowanna 1

1977 B. Cuttapirrie beds or 'basal Hutton' Cuttapirrie 1 1980 Bodalla South 1 1984 C. Hutton Sandstone Strzelecki 4 1978 Merrimelia 6 1981 Jackson 1 1981 Chookoo 1 1983

Hydrocarbon TVpe

Gas Character Oil Character co2 Gravity Pour Point c3+ w (%) (°API) (°C)

Flow Rate or Recovery

—

Rec 840m O; 840m W

oil oil

50 47

-2 n.a.

—

160 BOPD 2000 BOPD

oil oil oil gas; oil

44 55 40 50

14 -5 20 -25

52 48 40 41

-12 16 18 18

oil

Narcoonowie 2 Naccowlah South 1 Wilson 1 Naccowlah West 1 Challum 1 Tintaburra 1 Bodalla South 1 Kerinna 1 Gidgealpa 17

1983 1983 1983 1983 1984 1984 1984 1984 1984

oil oil oil oil gas oil oil oil oil

D. Birkhead Formation Wackett 1 Strzelecki 3 Nappacoongee 2 Moorari 3 Jackson South 1 Big Lake 26 Merrimelia 9 Narcoonowie 2 Chookoo 2 Jackson 18 Mudera 2 Charo 1 Bogala 1 Graham 1 Gidgealpa 17

1978 1978 1979 1981 1982 1982 1982 1983 1983 1983 1984 1984 1984 1984 1984

gas oil oil oil oil oil oil oil gas oil oil oil oil oil oil

E. Westbourne Formation Jackson 1 Jackson South 1 Wilson 1 Chookoo 1

1981 1981 1983 1983

F. Namur Sandstone Member Namur 1 Strzelecki 4 McKinlay 1 Dullingari 22 Marabooka 2 Merrimelia 8 Narcoonowie 2 Wilson 1 Sigma 1 Biala 1

—

n.a. n.a. 42 48

n.a. n.a. 30 n.a.

—

45 45 51 41 48 53 n.a.

8 11 n.a. 23 9 -5 n.a.

—

2.4

40 47 48 47 37 53

22 3 6 9 n.a. n.a.

—

oil oil oil oil

41 41 42 47

19 18 16 -3

1976 1981 1981 1982 1982 1982 1983 1983 1983 1984

gas oil oil oil gas oil oil oil oil oil

—

48 41 57

17 15 n.a.

—

51 53 54 45 41

-5 -15 -22 10 18

—

Gidgealpa 17

1984

oil

n.a.

—

G. Murta Member Dullingari North 1 Merrimelia 6 Jackson 1 Kihee 1 Nockatunga 1 Gunna 1 Sigma 1 Wilson 1 Tinpilla 1 Naccowlah West 1 Challum 1 Yanda 1 Bogala 1

1979 1981 1981 1982 1983 1983 1983 1983 1983 1983 1983 1984 1984

oil oil oil oil oil oil oil oil oil oil oil oil oil

57 51 51 n.a. n.a. 47 45 47 48 48 60 44 48

-3 -5 0 n.a. n.a. -8 9 -12 n.a. -9 -22 8 0

—

— —

—

3170 BOPD 2738 BOPD 2375 BOPD 7.4 MMCFD; 483 BOPD not tested 1395 BOPD 163 BOPD 144 BOPD 8.9 MMCFD 1750 BOPD 883 BOPD Rec 751m O 3200 BOPD

—

2.2 MMCFD Rec 64m O Rec 61m O, 1005m W Rec 13 bbl O Rec 137m O, 396m MW 150 BOPD Rec 15 bbl O not tested 1.1 MMCFD Rec 6m O Rec 178m O Rec 317m O, 204m W Rec 283m O, 125m MO Rec 138m O Rec 83m O, 120m M

—

1165 BOPD 750 BOPD 707 BOPD 5 BOPD

—

— —

_ —

—

_ —

—

— —

—

13.9 MMCFD 200 BOPD 9 BOPD 1994 BOPD 7 MMCFD 440 BOPD 1275 BOPD Rec 1132m O Rec 350m O 201 BOPD, 1205 BWPD 260 BOPD, 720 BWPD 450 BOPD 443 BOPD; 679 BWPD 338 BOPD Rec. oil Flowed OTS Rec 219m O Rec 722m O Rec 546m O Rec 415m O 120 BOPD Rec 15m O, 91m OM Rec 18m MO Rec 694m O, 415m OM

FOREWORD, EXPLORATION OVERVIEW Limestone Creek 1 1984 oil Biala 1 1984 oil Narcoonowie 3 1984 oil Kobari 1 1984 oil H. Cadna-owie Formation Merrimelia 15 1983 oil Tintaburra 1 1984 oil I. Coorikiana Sandstone Strzelecki 8 1982 gas Marabooka 3 1984 gas Abbreviations: n.a. is not available Bbl is barrels Cond. is condensate BOPD is barrels of oil per day BWPD is barrels of water per day

42 42 56 43

18 18 n.a. n.a.

46 n.a.

4 n.a.

—

7 0.7 n.a. n.a. MMCFD is million cubic feet of gas per day Rec is recovered O is oil W is water —

discovery of the Strzelecki and Dullingari Oilfields in the overlying Eromanga Basin. Early in 1980, a consortium of eleven companies announced its decision to proceed with the 'Liquids Project'. These companies are party to the Cooper Basin Unit Agreement, which was established in 1976 to control the orderly development of the Cooper Basin gas reserves. The companies are Alliance Petroleum Australia Pty Ltd, Basin Oil N.L., Bridge Oil Developments Pty Ltd, Bridge Oil Ltd, Crusader Resources N.L., Delhi Petroleum Pty Ltd (part of the CSR Group), Reef Oil N.L., Santos Ltd, South Australian Oil and Gas Corporation Pty Ltd, Total Exploration Australia Pty Ltd and Vamgas Ltd. The Liquids Project is an integral part of continuing petroleum development in both the Cooper and Eromanga Basins. As well as the production of crude oil, the Liquids Project involves the separation of natural gas liquids (ethane, propane, butane and condensate) from gas bound for the Sydney and Adelaide markets. The rate of production of natural gas liquids therefore depends in part upon the rate of consumption of natural gas. Prior to liquids development, only five dry gas fields were in production (Gidgealpa, Moomba, Big Lake, Namur and Delia). In its initial phase, the Liquids Project involves the development of an additional 14 Cooper Basin gas fields, and 6 oil fields. Of the 6 oil fields, Tirrawarra, Moorari and Fly Lake produce from the Cooper Basin whereas Strzelecki and Dullingari produce from the Eromanga Basin. The sixth field is Merrimelia, which produces oil from both Cooper and Eromanga Basin sequences (Figs 4, 5). For the Eromanga Basin fields, treatment stations at Dullingari, Strzelecki and Merrimelia cool and separate the produced fluid into oil, water and gas. The main treatment plant for the Liquids Project is situated at Moomba, in the Strzelecki Desert, 800 km northeast of Adelaide. The Moomba complex has been designed to process 25.4 million cubic metres of raw natural gas per day, 7250 kilolitres of natural gas liquids per day and 30,000 barrels of crude oil per day. Ethane, extracted from natural gas, is being stored underground at Moomba in partially depleted gas wells while its marketability is investigated. Remaining natural gas liquids are commingled with the stabilised crude oil and piped 659 km to Port Bonython along a 355 mm diameter underground pipeline. The pipeline, with 4 pump stations installed, has a throughput capacity of 80,000 barrels of hydrocarbon liquids per day. Production of crude oil and condensate began early in 1983 and LPG production commenced in mid 1984. When facilities are fully operational the initial production of crude oil and

—

332 BOPD Rec 15.5 bbl O Rec 666m O, 35m m 53 BOPD Rec 9m O, 12m OM Rec 12m O, 18m M 0.38 MMCFD 0.65 MMCFD MO is muddy oil MW is muddy water OM is oily mud OTS is oil to surface

condensate will amount to about 10 million barrels per annum, while LPG will be produced at the rate of about 550,000 tonnes per annum. The crude oil and most of the condensate is sold to Australian refineries. The LPG supplies a small domestic market and is also exported. Ethane is used as plant fuel at Port Bonython but eventually may be used as a feedstock for a petrochemical plant.

The Jackson Project in Queensland

The size of the Jackson Oilfield, discovered in southwestern Queensland in 1981, led to a decision to quickly put the field into production. The nearby Jackson South Field, discovered in February 1982, was included in this development programme. Some of the more recent discoveries in the area are also being linked to the production facilities. Others will have separate facilities, depending upon the economics of the development, which will include the size of the discoveries and their distance from Jackson. Meanwhile, as exploration continues in the Naccowlah Block and adjacent licence areas, field facilities at Jackson and a pipeline from Jackson to Moonie (Fig. 1) have been constructed to enable the oil to be transported to Brisbane for refining. The six companies responsible for field development are Ampol Exploration Ltd (7.5%), Claremont Petroleum NL (10%), Delhi Petroleum Pty Ltd (upstream operator; 32%), Oil Company of Australia NL (2.5%), Santos Ltd (downstream operator; 40%) and Vamgas Ltd (8%). The cost of development is presently estimated at $62 million, and has included the drilling and completion of 31 development wells and installation of gathering, treating, storage and infrastructure facilities. In the Jackson Field, oil is produced from the Murta Member, Westbourne Formation and Hutton Sandstone. The Hutton Sandstone contains the majority of reserves and typically flows at rates in excess of 1000 BOPD per well. In the Jackson South Field, three wells produce oil from the Westbourne Formation; these are linked to the Jackson facilities by a 5.5. km pipeline. Construction of a 323 mm diameter, 780 km underground pipeline from Jackson to Moonie commenced in April 1983 and was completed in February 1984 at a cost of approximately $120 million. Initial design capacity was 16,000 barrels of oil per day. However, this can be increased to 55,000 barrels of oil per day with the installation of extra pumps. The high pour points of Hutton and Westbourne oils (25 °C and 20 °C respectively) and their waxy nature provide significant pumpability problems. Although no treatment is needed during the hot Australian summer, pour point depressants are added in spring, autumn and winter. This

P. S. MOORE

increases the production cost of Jackson crude by about $0.50 per barrel. During winter, air temperatures in the Australian desert can drop to 0°C at night and, although the pipeline is buried, additional modification of the flow properties of the oil is necessary. This is achieved by producing low pour point (0°C) oil from the Murta reservoir and adding it as a 10% diluent.

subsurface sections. Even in the most prospective central portion of the Eromanga Basin, drilling density is low, with less than 6 wells per thousand square kilometres. However, exploration activity is increasing. During 1984 approximately 20,000 kilometres of seismic data have been recorded and 100 exploration wells drilled. This activity reflects a common belief that the Eromanga Basin is the most prospective onshore area in Australia for petroleum. Future exploration is likely to be concentrated in northeastern South CONCLUSIONS Australia and particularly southwestern Queensland, where Despite recent progress, exploration in the Eromanga Basin success has been greatest. The presence of two oil pipelines is still at a very early stage. For example, less than 600 will also promote exploration in these areas, making it more petroleum wells penetrate the sequence, with large areas almost attractive to explore for the small to medium-sized (1-30 totally unexplored. Outcrops on the basin margins yield only million barrels) oil fields typical of the basin. Based on limited data because the thin, gently dipping and deeply successes to date and on petroleum potential outlined by the weathered strata bear little resemblance to the more complete following papers, many more discoveries can be expected.

REFERENCES

AMBROSE, G., SUTTILL, R. & LAVERING, I., 1986: The geology and

MOORE, P. S., PITT, G. M . & DETTMAN, M . E., 1986: The Early

Cretaceous Coorikiana Sandstone and Toolebuc Formation: their recognition and stratigraphic relationship in the southwestern Eromanga Basin; in This volume. MOUNT, T. J., 1981: Dullingari North 1, an oil discovery in the Murta Member of the Eromanga Basin. APEA J., 21 (1), 71-7. MOUNT, T. J., 1982: Geology of the Dullingari Murta Oilfield, in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin DAY, R. W., WHITAKER, W. G., MURRAY, C. G., WILSON, I. H . & Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. GRIMES, K. G., 1983: Queensland geology. A companion Soc. Aust., Adelaide. volume to the 1:2,500,000 scale geological map (1975). Geol. OZIMIC, S., 1986: The geology and petrophysics of the Toolebuc Surv. Qld, Publ., 383. Formation and its time equivalents, Eromanga and Carpenteria EVANS, P. R., 1982: Microfossils associated with the ~Bundamba in This volume. Group" of the Surat Basin, Queensland. Aust., Bur. Miner. PATON,Basins; I. M., 1986: The Birkhead Formation—A Jurassic petroleum Resour., Geol. Geophys., Rec., 1962/115 (unpubl.) reservoir; in This volume. EXON, N. F., 1976: Geology of the Surat Basin in Queensland. Aust. PORTER, C. R., 1979: Fluvio-deltaic deposition in Lower Jurassic Bur. Miner. Resour. Geol Geophys. Bull. 16. sediments of the Surat Basin, Queensland. APEA J., 19 (2), GRAVESTOCK, D. I., GRIFFITHS, M. <fc HILL, A., 1983: The Hutton 37-50. Sandstone—two separate reservoirs in the Eromanga Basin, POWER, P. E. & DEVINE, S. B., 1970: Surat Basin, AustraliaSouth Australia. APEA J. 23(1), 109-19. subsurface stratigraphy, history and petroleum. Am. Assoc. Pet. HABERMEHL, M. A., 19.86: Regional groundwater movement, Bull. 54, 2410-37. hydrochemistry and hydrocarbon migration in the Eromanga RILEY,Geol. K. W. & SAXBY, J. D., 1986: Organic matter and vanadium Basin; in This volume. in the Toolebuc Formation, northern Eromanga Basin and HALYBURTON, R. V. & ROBERTSON, A. L., 1984: Geology of the southern Carpentaria Basin; in This volume. Jackson Oil Field. APEA J., 24 (1), 259-65. SAXBY, J. D., 1986: Geochemistry of oil shale in the eastern Eromanga KANTSLER, A. J., COOK, A. C. & ZWIGULIS, M., 1986: Organic Basin; in This volume. maturation in the Eromanga Basin; in This volume. SPRIGG, R. C., 1986: The Eromanga Basin in the search for M C M I N N , A. & BURGER, D., 1986: Palynology and commercial hydrocarbons; in This volume. palaeoenvironments of the Toolebuc Formation (sensu lato) in SWINDON, V. G. & MOORE, P. S. in press: Exploration and the Eromanga Basin; in This volume. production, Eromanga Basin, Central Australia. Am. Assoc. Pet. MARTIN, K. R., 1981: Deposition of the Precipice Sandstone and Geol., Proc. 1984 Beijing Petrol. Symp. evolution of the Surat Basin in the Early Jurassic. APEA J., WILTSHIRE, M. J., 1982: Late Triassic and Early Jurassic 21 (1), 16-23. sedimentation in the Great Artesian Basin; in Moore, P. S. & MOORE, P. S., 1986: Jurassic and Triassic stratigraphy and Mount, T. J. (compilers) Eromanga Basin Symposium, summary hydrocarbon potential of the Poolowanna Trough (Simpson papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide. Desert region) northern South Australia; in This volume. WOPFNER, H., FREYTAG, I. B. & HEATH, G. R., 1970: Basal MOORE, P. S. & PITT, G. M . 1984: Cretaceous of the Eromanga Jurassic—Cretaceous rocks of western Great Artesian Basin, Basin—implications for hydrocarbon exploration APEA J., South Australia: stratigraphy and environment. Am. Assoc. Pet. 24 (1), 358-76. Geol., Bull., 54, 383-416. hydrocarbon potential of the Murta Member (Mooga Formation) in the southern Eromanga Basin; in This volume. ARMSTRONG, J. D. & BARR, T. M., 1986: The Eromanga Basin: an overview of exploration and potential; in This volume. BOWERING, O. J. W. & HARRISON, D. M., 1986: The Merrimelia Oil and Gas Field—a case history; in This volume.

Geological Society of Australia Special Publication No. 12, 9-24

The Eromanga Basin in the search for commercial hydrocarbons Reg C. Sprigg

C/~ Arkaroola Sanctuary, via Port Augusta, S.A. 5700.

ABSTRACT

The Eromanga (Great Artesian) Basin is devoid of obvious hydrocarbon seepages at the surface. Most of its deeper sediments are non-marine. None of its widespread thousands of artesian water bores produced a single flow of oil or gas prior to the 1970s. However, small volumes of methane were known to escape with artesian water from the Coonanna and Yandama bores lying to the northeast of Lake Frome, the Patchawarra Bore and others. Early geologists regarded the Great Artesian Basin as a structurally stable Mesozoic development; a simple saucer-like depression interrupted only by buried basement ridges and local warpings. The significance of Mesozoic-Tertiary growth structures mapped during early hydrological investigations (Jack, 1925) went unrecognised or was misreported. The 'fortuitous' discovery of crude oil in 1925, the first in the Eromanga Basin, was made in the Longreach town water bore. The concept of natural conversion of terrigenous coaly matter to petroleum seemed unacceptable to geologists prior to the 1960s despite widespread minor showings of oil and gas throughout thick non-marine sediments in the Eromanga Basin. Still, geological and geophysical exploration (gravity and magnetic) expanded in the region during 1930-1950. Reports stemming from hydro-geochemical studies suggested that the northeastern portion of the Great Artesian Basin in South Australia could host 'Jurassic' oil, a conclusion encouraged by numerous hydrocarbon indications .in Queensland artesian wells. The Mesozoic oil discovery in Rough Range, Western Australia, in 1954 provided real exploration impetus, leading in South Australia to the formation of Santos Limited. Subsequent mapping and drilling of anticlines near Oodnadatta, Haddon Downs, Innamincka and elsewhere, led to the discovery of minor indications of Jurassic oil, and Permian gas at Innamincka. Low oil prices and poor test results along with later Permian gas discoveries, deflected interest from the Mesozoic. Commonwealth Government exploration subsidies in the late 1950s and 1960s greatly encouraged application of more sophisticated oil exploration technology which culminated in widespread hydrocarbon discoveries across Australia as well as grossly improving basin understanding generally. By now Australia's sedimentary coaly detritus was being regarded seriously as potential source material for petroleum generation. Oil was noted to occur where the coals are of lower rank (80-85% carbon). Significant Eromanga Basin oil was first tested from Poolowanna 1, in the Simpson Desert in 1977. Still it was not until 1978 that the first commercial oil discovery was made in Strzelecki 3 which tested 2400 barrels per day (B.P.D.) from the Hutton Sandstone, soon after which Dullingari North 1 flowed 450 B.O.P.D. from the Murta Member of the Mooga Formation. Now, sands all the way from the basal Jurassic into the Cretaceous Coorikiana Sandstone and Toolebuc Formation are recording oil and gas flows or encouraging showings. A broad Jurassic-Cretaceous 'oil window' appears now well-defined in relation to temperature-depth parameters. Structural timing in relation to migration is undoubtedly important. It is now clear that sedimentation climaxed in Late Cretaceous time, providing sufficient depth of burial to generate and expel hydrocarbons. The full significance of hydrodynamic flushing in stratigraphic entrapment, or alternatively oil escape, has still to be assessed, but fossil oil seepages preserved in Lake Eyre mound springs indicate that flushing could be significant. Import parity pricing of crude oil has greatly stimulated ongoing oil search. Continued sound pricing will undoubtedly result in more widespread discovery.

R. C. SPRIGG western Eromanga Sub-basin of Mott. Evans (1946) isolated in the south as the Frome Embayment, and Sprigg et Exploration for petroleum in the Great Artesian Basin a lobe (1958) proposed an extension west beyond the Peake and virtually commenced with the fortuitous discovery of natural alDenison Ranges as the Arckaringa Basin. gas in the course of water drilling at Roma, Queensland, in This confusion of basins, sub-basins, infra-basins, 1900. At about this time, thousands of artesian wells were embayments, geobasins and even miogeosynclines continues being drilled far in the centre of the continent. This very fact of multitudinous flowing water wells, some down to two to the present. However, general consensus seems now to kilometres (Patchawarra and Springleigh Bores) without accept the redefinition of Senior et al (1978) that the significant oil discovery was to inhibit the systematic Eromanga Basin represents the preserved limits of Jurassic exploration of the basin for decades to follow. A number of and Cretaceous sediments in the central Australian depression, water wells, such as in the Frome Embayment (early 1900s) delineated in the north by the Euroka Arch, the east by the and at Patchawarra (1914) did, however, flow enough gas to Nebine Ridge and to the south and west by older sedimentary burn continuously, but the gas was disappointingly 'dry'. basins or crystalline basement. It was S.A. Government Geologist, the inimitable Dr R. Lockhart Jack who, in the early 1920s, first investigated the EARLY OIL DISCOVERY AND NEW IDEAS central Great Artesian Basin (as he then named it) in detail. It was Lander Oil Company which first recovered 'thick He mapped and described a broad Tertiary anticline about black oil' to the surface at Longreach in 1927 from 987 m, Cordillo Downs and defined a prominent angular immediately above granite bedrock as the result of deliberate unconformity at the Cretaceous-Tertiary boundary which test drilling. This gave birth to renewed expectations for the indicated continuing structural development. This discovery of commercial the Eromanga Basin. breakthrough went almost unheeded for another quarter of Queensland Minister of Mines,oiltheinHon. A. J. Jones, returned a century. samples of the fresh oil to Brisbane, by which time it had The discovery of waxy oil above granitic basement in the congealed to 'axle grease'. Deputy Government Geologist Ball Longreach town water bore in 1925 led to the first true wildcat (1927, p. 358) reported further that 'the occurrence of drilling in the Eromanga Basin, first by Lander Oil Company, petroleum or paraffin wax . . . from Longreach Bore No. 2, then by Oriomo Oil Ltd (a subsidiary of Oil Search Ltd), but while perhaps not of immediate economic significance, is of without success. considerable geological interest, and may have important During the 1930s, Commonwealth Government consultants commercial results . . . it exists below in liquid state'. stimulated increasing interest in the Great Artesian Basin, but Government Geologist Moreton further added (1927, p. 317) their reports proved cautious and mostly unenthusiastic. that this 'afforded the most significant proof to date of a fact Incredibly, and despite a visit to the structurally developed already recognised that the Artesian Basin in Queensland is Cordillo Downs area in 1927, Commonwealth Government not devoid of petroleum contents'. Ball, at this time, in Geologist, Dr W. G. Woolnough seemed not to be conscious company with Commonwealth Government Geologist, W. G. of Jack's (1925) findings, and reported that Tertiary duricrust Woolnough, also reconnoitred extensively in the neighbouring there was flat-lying. Near Innamincka, he noted (Woolnough, Thomson River headwaters. For the first time, and in true 1927, p. 131) 'inliers of gneissic granite surrounded by the explorer fashion, a number of dip-reversals were observed in Cretaceous sediments of the Great Artesian Basin'. This was outcropping Cretaceous shale. Professor Steel of Queensland yet another serious setback for exploration. University in a lecture to the Royal Society of Queensland At about the outbreak of World War II, both Shell on 10th October 1927 (Qld Gov. Min. J. 1927, p. 387) stated Development (Queensland) and Zinc Corporation commenced that 'the idea that there was oil in Queensland . . . was not more systematic geological appraisals of large portions of the a new one, but recent happenings had converted a probability basin. After considerable drilling during the next decade, into something of a certainty . . . but reports of oil had to principally by Zinc Corporation and Frome Broken Hill be examined with the utmost care; they must be looked at Company, interest again waned. It was not until the discovery with suspicion and sifted most thoroughly. Mr Henderson of oil at Rough Range in Western Australia in 1954 that in his capacity of Government geologist, no doubt would be exploration effort generally took a new turn. This excited the able to say that many samples of oil had been handed to him interest first of entrepreneurs and wildcatters, then of for examination, which had never been put under the ground Government agencies, and finally the more established oil by the Almighty'. companies. In this phase, Santos Ltd (formed in 1954) 1927 was a time of increasing interest by the Queensland assumed a leading role. Mines Department in reports of German wartime (1914-18) success in the production of oil from coal. The Queensland Government Mining Journal (1927, p. 61, 65, 390) in that year DEFINING THE EROMANGA BASIN carried numerous reports dealing with the artificial Pittman (1895) appears to have been first to recognise the 'transformation of coal into oil'. Chemist-Engineer Bradfield widespread development of 'Cretaceous water-bearing rock' (1943, p. 48) (author of the scheme to turn Queensland rivers in eastern Australia, which he simply named the Artesian inland), specificially warned geologists and others not to Basin'. Later (Pittman, 1915, p. 5) referred to it as the Great overlook Australia's great coal measures as providing potential Australian Artesian Basin. The 'Great Artesian Basin' was source material for subterranean oil! Steeped as most named by Jack (1925, 1930). Mott (1952) later introduced the petroleum geologists then were in the reigning belief that oil term 'Eromanga Sub-basin' in summarising the region's generation was limited almost exclusively to marine source widespread oil and gas showings. He separated the northern rocks, this advice went almost unheeded. The significance of Carpentaria Basin by the shallow Euroka Shelf, and the Surat Cosmo-Newberry's discovery in 1908 (see Ward, 1944, p. 29) 'Sub-basin' by the Eulo Shelf and Nebine Ridge. At about of traces of free oil in Triassic shale cores from below 330 the same time, Whitehouse (1954) introduced the term 'Austral metres in Leigh Creek No. 2 coal bore, in northern South Geobasin', of which his Thompson Sub-basin equated the Australia too, went unrecognised. 10

INTRODUCTION

E R O M A N G A O I L SEARCH By 1930, serious oil search had retreated to the Roma vicinity of Queensland, where Oil Search Ltd continued to operate. With deepening world recession prior to World War II, it was a fact of life that, despite thousands of water wells drilled across the Great Artesian Basin, no significant hydrocarbon discovery had yet been made. Few Australian geologists were then brave enough to predict real prospectivity. The view grew that Australia was indeed too old and too stable, its basins too shallow and its sediments of the wrong type. The absence of surface oil seeps, or of significant 'shows' in wells, was thought to weigh heavily against discovery. To this must be added the lack of adequate exploration technology, the tyranny of distance, inhospitable terrain, low oil prices and plentiful imported, low-cost oil.

THE WARTIME ADVENT OF MAJOR COMPANIES AND EARLY EXPLORATION GEOPHYSICS Shell Development Queensland Pty Ltd (Shell) was the first major oil company to undertake serious exploration in Queensland. From 1939 to 1943, Shell undertook regional geological mapping, gravity and magnetic surveys and scout drilling (in the Roma-Tambo region), mostly in areas lying east of the scope of this presentation. Shell's gravity-magnetic reconnaissance surveys extended far west to include Eromanga and Windorah, and were, in those days, an epic of exploration geophysics. Low-key investigations continued up to 1951, with integration of data from more than 1000 water wells into gravity-magnetic interpretations (Shell, 1951). Broad structural undulations evident in the Mesozoic sequence were generally attributed to compaction over uneven basement, or to gentle warping. At about this time, Zinc Corporation Pty Ltd commissioned a geological review (de Verteuil, 1940) of the oil prospects of northeastern South Australia and adjoining Queensland and New South Wales. He saw little or no petroleum prospectivity for the marine Cretaceous or for the nonmarine Jurassic. He did, however, draw attention to traces of free oil in the Leigh Creek Triassic Coal Measures and to small quantities of bituminous material in Permian shale encountered in Lake Phillipson bore (Arckaringa Basin). He concurred with discouraging reports by former 'geologists of repute and experience' and concluded (op. cit., p. 18) that 'the oil prospects of the region under review are considered too poor to warrant any recommendation for further action. A fundamental defect is the lack of any positive evidence of petroliferous character in the geological succession'.

THE EARLY POST WORLD WAR II YEARS Re-awakening of post-war exploration interest was brought about by the enthusiasm of W. S. Robinson of the Zinc Corporation, supported by A. J. Keast, (Sir) Maurice Mawby and others of the company staff. Mawby (1944), in a then confidential report entitled 'A summary of the possibilities of obtaining commercial natural gas supplies in Australia', drew together much new field data on the occurrence of methane gas escaping from artesian water bores in the Frome Embayment. He drew particular encouragement from water bore information plotted by Kenny (1934) of the N.S.W. Geological Survey, that delineated the Joulnie Anticline in Mesozoic strata, lying some 130 km northwest of energydeficient Broken Hill. Out of this came Osborne's (1945) 'Report on the oil and gas possibilities of the Frome Embayment in N.S.W. and S.A.' Osborne detailed

approximately 20 water wells emitting methane gas from the Jurassic Walloon Formation in the Joulnie area. In the absence of any indications of higher hydrocarbons in the escaping gases, he concluded (op. cit., p. 2) that the region's 'oil prospects are very remote, but that the chances of natural gas concentrations are sufficiently good to justify an exploration for structural traps with the object of testing by drilling'. Operations were subsequently taken over by Frome Broken Hill Company (Frobilco; a consortium of Zinc Corporation, Standard Vacuum and British Petroleum) and extended north over almost 500 000 sq km of the Eromanga Basin. Under contract, the Bureau of Mineral Resources (B.M.R.) first carried out a detailed gravity survey over the Joulnie anticline, then Frobilco imported a Sullivan 300R drilling rig and a gravity-magnetic team from New York headed by geophysicist Bob Sauve. In 1947, the team set out on an ambitious triangular grouping of station traverses from Broken Hill to Patchawarra Bore, back to Moolawatana (Flinders Ranges) and across the Frome Embayment to base. Side traverses took in Cordillo Downs and more of the Frome Embayment (Fig. 1). The survey was a milestone in early Australian geophysical exploration and served to delineate areas of potentially deep basin sedimentation separated by prominent basement ridges. Kaufman & McPhail (1948) of the Vacuum Oil Company, N.Y., subsequently co-ordinated all available geological and geophysical data and concluded that a good correlation existed between Bouguer gravity and basement depth determined from water-well data. Structure was interpreted to be due more to basement relief than to subsequent tectonic folding. The density contrast between basement and overlying sediment was deduced to be 0.4 gm cm 3, giving an interpreted maximum sedimentary thickness of about 3000 m located south of Cooper Creek. Magnetic data were less amenable. The Bouguer gravity maps clearly outlined 'highs' consistent with the later-defined Innamincka and subsidiary Packsaddle anticlines and another in the Dullingari vicinity. The regional 'low' south of Innamincka is now known to relate to the deeper (Permo-Triassic) Nappamerri Trough. From 1947 to 1951, Frobilco drilled seven stratigraphic wells in the Frome Embayment and across the Birdsville Track, but without encouragement. Surprisingly, the enormous Innamincka structure (90 x 40 km and with 200 m of structural 'closure' in outcrop) failed to excite serious exploration interest, nor did Jack's Cordillo anticlines. Reeves and Evans (personal communications) did plan an aerial inspection of the Innamincka 'Tent Hills' with their spectacular radial drainage but were turned back by a severe dust storm.

THE STIMULATION OF THE ROUGH RANGE OIL DISCOVERY The non-commercial Rough Range oil discovery in W.A. in 1954 triggered Australia's wildest-ever oil exploration boom. A rash of newly-formed Australian exploration companies, and later, American 'wildcatters', moved in to explore, or to profit from share market speculation. The South Australian front-runner was Santos Ltd, formed by John Bonython and Bob Bristowe with Reg Sprigg as geological consultant. Preliminary interest was to investigate minor oil showings in the Wilkatana artesian water bore near Port Augusta. The longer-term plan was to take up and investigate as much of the western Great Artesian Basin as became available. Sprigg was a former Assistant Government Geologist with the S.A. Mines Department, well acquainted

R. C. SPRIGG

Fig. 1. Locality map, central Eromanga Basin, showing location of Frobilco gravity-magnetic traverses (1947) and selected well locations.

with predecessor Jack's published work, and with Frome Broken Hill's operations in the area. He knew, too, of Departmental Hydrologist, Dr Ivan Chebotarev's (1952) deductions from hydrological investigations that the most promising environment for oil generation in the Great Artesian Basin, in his view, appeared to be in the extreme northeastern corner of South Australia. For this, Chebotarev drew on his extensive knowledge of oil-field waters in Russia. On behalf of Santos, in 1955, Sprigg's private consulting and contracting company, Geosurveys of Australia Pty Ltd (Geosurveys) took opportunity, as the full South Australian

portion of the Great Artesian Basin became available, to extend gravity surveys from Innamincka to Betoota, Birdsville and Marree. It was during these operations that the deeper potential significance of Jack's (1925) Cordillo anticlines with their internal unconformities (Fig. 2A, B) finally hit home forcefully to Sprigg, but at a time when Santos' limited finances were already over-extended. Inspection of the latest topographic maps in relation to sketchy outlines of the distribution of the Eyrean silcrete 'duricrust' that included the MacGregor and Grey Ranges in Queensland and the Innamincka 'duricrust outlier' drove home the possibility that

EROMANGA OIL SEARCH

B Fig. 2. A, B. Tertiary-Mesozoic 'folding' and internal unconformity in the Cordillo Downs region as revealed by Jack (1925).

R. C. SPRIGG

these could together constitute strings of outcropping anticlines that confined the major modern drainage lines (e.g. Cooper, Diamantina) to intervening youthful synclinal troughs or valleys. Old ties with the S.A. Lands Department Aerial Photographic unit remained strong, so that despite Santos fund restrictions, it proved possible to put Geosurveys' geologist, Heli Wopfner, aboard to aerially inspect the Cordillo-Betoota-Curalle duricrust areas in detail; this was accomplished with spectacular success. The Betoota and Curalle 'ranges' with their strong radial drainage patterns were indeed elongate, domed, anticlines with duricrust limbs dipping up to 10 to 25 degrees. Santos directors, now suitably excited, permitted the company's light aircraft to ferry Sprigg and Rudi Brunnschweiler to investigate also the expansive Innamincka 'outlier' with even more stimulating results. Sprigg's (1958) publication of these and later, more detailed Geosurveys' aerial and photogeological observations had a spectacular effect on American independent operators and 'wildcatters'. At the instigation of Dr Martin Glaessner, Santos in 1957 brought Dr A. I. Levorsen from Tulsa to review company progress. His opinions confirmed the growing SantosGeosurveys view that investigation of the Mesozoic Great Artesian Basin must now take precedence over Cambrian studies at Wilkatana. His classic Statement to Santos chairman Bonython, namely 'It's a case of Cadillacs versus Chevvies, John' won the day, whereupon Levorsen recommended inviting Delhi Taylor Oil Corporation of Tulsa to earn a 50% interest in the areas by taking over ongoing exploration interests and costs to an agreed stage. After initial haggling, it took only a single flight of senior technical staff over the Innamincka Dome to conclude such a farm-in agreement.

TRACKING DOWN GEOLOGICAL STRUCTURE

Early geologists correctly viewed the Great Artesian Basin primarily as a broad epeirogenic downwarp. Whitehouse (1954) even calculated a predicted average rage of central basin subsidence since the Triassic, of 2 mm per century. It was Jack (1925) who, in mapping the Cordillo 'uplift', provided the first clues that the Tertiary duricrust itself was gently folded (Fig. 2A, B). Geosurveys geologists in 1957 extended this idea to major anticlines about Betoota, Curalle and Innamincka (Fig. 3). Sprigg further recognised that the Cooper, Diamantina, Farrers, Wilson and Kyabra drainage channels occupied broad, youthful synclines and that the intervening ridges were actually variously eroded" anticlines, developed and preserved in resistant silcrete. This highlighted a further fact that most deep artesian bores in these areas had been drilled in the board synclinal flats between the rough anticlinal ridges and consequently had even less chance of fortuitously encountering hydrocarbon accumulations. Several bores along the Birdsville Track Ridge appeared to be notable exceptions. Sprigg personally extended such photointerpretation over several tens of thousands of square kilometres of southwest Queensland and outlined more than 70 potential anticlines and domal culminations. As Vine (1966, p. 112) observed much later, many of the structures revealed by seismic surveys could well have been located by surface mapping, as indeed they already were. 'Creekology' had already had its field day! Both Sprigg (1958) and Wopfner (1960) concluded that the Cordillo-Betoota grouping of structures probably related to fault-shearing in bedrock. Sprigg (1961, p. 53) further drew attention to major basement shear lineaments around the basin perimeter that could intersect beneath the deeper reaches

of the Eromanga Basin and be capable of inducing some of the observed en echelon folding in the overlying sediments as well as developing a central accentuated downwarp about their intersections (Fig. 4). Following these heady early days, seismic geophysics took over and outlined numerous additional fold structures, some buried far out beneath the surrounding featureless deserts. Some 'folds' proved to reflect compaction over pre-existing bedrock topography, some are growth or drape folds, some are horsts, while still others relate to normal faults at depth. King & Falvey (1977), in discussing the strongly asymmetric, block-faulted Canaway structure, noted that this ancient feature effectively separated the Permo-Triassic Galilee and Cooper Infra-basins. Erosional 'baldheading' at significant horizons and late structural development of the more prominent anticlines appears generally to have opposed effective hydrocarbon entrapment in many cases. This is a problem experienced elsewhere in the world as Levorsen was at pains to warn Santos-Geosurveys personnel from the outset. Lewis Weeks frequently propounded that many anticlines within the broader basin downwarps are initiated by tension faulting of the basement during continued subsidence. Many such anticlines tend to be asymmetric with central fault blocks. However, this seems not to take into account the upward curvature of the earth against which sagging (of the arch) would produce compression rather than extension and be far more likely to facilitate en echelon shear-faulting in basement rocks. The Cork group of en Echelon faults which delineate the southeast termination of the Georgina Basin, and which line up generally with the Paralana fault system of the Flinders Ranges (Fig. 4), appear to be of this association. Translatory movement along any of these could readily account for any of the en Echelon folds of the Curalle-Betoota and related systems. Of particular interest is the recent advent of Landsat satellite imagery for use in the interpretation of linear and curvilinear surface features. Computer enhancement involving 'contrast stretching' has dramatically increased scene contrast and provided improved geological mapping capability. By such techniques, it is also reputed that subtle reflectance differences have been already noted over some Australian oilfields. Alteration of soil or rock in areas of diffuse tonal anomalies above known oilfields has been attributed to gas seepage. Landsat D, launched in August, 1982, promises still greater advances as the new 'imagers' are specifically designed to produce even more useful geological data. The folding of mid-Tertiary silcrete, and the overturning of Tertiary formations along the northeastern margin of the Flinders Ranges (Woolnough, 1927, p. 32), clearly indicates the extent of continuing instability in and about the southern margin of the Eromanga Basin. So, too, does the outbreak of several generations of mound-springs about the southwestern perimeter of the basin in Late Cainozoic time (see also Krieg, 1982, this volume). The extensive seismic activity (up to 6.2 on the Richter Scale) in the Poeppels Corner vicinity of the Simpson Desert (Stewart & Denham, 1974, p. 335), and topographic upstepping along interdune corridors along major known faults, here and in the Frome Embayment (Figs 4, 7) further attest to continuing movement.

THE WILDCATTERS MOVE IN

Santos Oodnadatta 1, drilled in 1957 on a low amplitude anticline immediately north of Oodnadatta, was the first of the 'post-Rough Range' era of wells in the Eromanga Basin.

EROMANGA OIL SEARCH

Fig. 3. Structure on base of duricrust mapped by Geosurveys/Santos geologists in 1957 (after Sprigg, 1958).

R. C. SPRIGG

Fig. 4. Principal fault lineaments projecting beneath the Great Artesian (Eromanga) Basin and presumed to contribute to accentuated central basin formation and folding (Sprigg, 1967).

At 405 m depth, traces of oil were encountered in cores from the marine Cretaceous section. The well terminated in steeplydipping, clean white sandstone of presumed Ordovician or Late Devonian age. The oil 'shows' gave the first indication that the Mesozoic in the western Eromanga Basin could be capable of trapping oil. On 28th March, 1959, Delhi Australian Petroleum spudded Innamincka 1, its first Australian wildcat well on Innamincka Dome as part of its farm-in commitment. The drill site was determined by surface mapping (Fig. 5) and the well was completed 234 days later at a total depth of 3 853 m (K.B.). It established 1 950 m of Mesozoic section overlying 100 m of Permian sandstone carrying minor amounts of wet (condensate-rich) gas. The Permian section rested with strong angular unconformity on thick, presumed Devonian red-beds. The well encountered organically-lean, marine Cretaceous sediments from 547 to 1 347 m (K.B.) thereafter mainly sandstone, yielding occasional traces of methane gas and sporadic oil traces down to the Permian strata. The 'shows' were considered to be 'the result of the decay of brackishwater plant material or humus, inconsequential as an important petroleum source' (Ryan, 1961, p. 19). Various drillstem tests flowed only formation water, but at least one

sidewall core in a poorly-consolidated sandstone at 1 460 m provided 'fair to good yellow fluorescence with a good flowing cut'. It was regarded as 'a good oil show' and it lay within the Late Jurassic section. Although technically 'dry', the well provided the first demonstration that the Late Jurassic sequence here was potentially oil generating. In addition, it provided the first direct evidence of a Permian infra-basin (Cooper Basin) and also demonstrated the top Blythesdale Formation to approximate the 'C Horizon' seismic reflector. The rig was transferred to the Betoota 1 site in southwest Queensland to test an elongate NNE-SSW trending anticlinal dome clearly defined at the surface both by 'creekology' and by steeply dipping limbs. The well was completed at 2 994 m (K.B. depth) after passing through 1 730 m of Mesozoic sediments into steeply dipping ?Devonian conglomerate and sandstone. Harrison et al. (1961) reported 703 m of Jurassic sediments with several weak oil shows that failed to flow when tested. Dead (i.e. residual) oil was encountered in the preJurassic sequence. Disclosure of Permian sediments below the Eromanga Basin in Innamincka 1 opened new exploration horizons. Interest in the Mesozoic accordingly diminished promptly, even though

EROMANGA OIL SEARCH

Fig. 5. Innamincka Dome mapped by R. B. Wilson and B. Fitzpatrick (Sprigg, 1958).

seven wells later in Gidgealpa 1, drilled off-structure, a strong oil show in the Walloon Formation provided 'conclusive proof that parts of the Mesozoic in this region contain fluid hydrocarbons. . . that under more favourable conditions of entrapment . . . may hold commercial hydrocarbons' (Delhi Aust. Pet., 1966). Gidgealpa 2 drilled at the apex of the seismically defined structure in 1963, struck gas in the Permian sequence, which proved potentially commercial on test. All effort now centred on Permian appraisal to satisfy firstly Adelaide's immediate natural gas requirements and then Sydney's. The Moomba Permian gas field was discovered in 1968 and became the centre of gas treatment and distribution. Pipelines to both Adelaide and Sydney were completed in 1969 and 1976 respectively. Santos Ltd shareholders received their first 'gas-generated' dividend only in 1978, 24 long years after the company was founded. The low price of oil (approx. $1.85 per barrel) obviously excited little further interest in the 'Walloon' oil prospects for some time.

THE FIRST TENTATIVE SEISMIC SURVEYS Although the B.M.R. had operated reflection seismic surveys in and around Roma since 1948, it was not until 1957 that, at Santos-Geosurveys' request, it was able to extend operations into the Eromanga Basin. The first survey was over a low amplitude anticline north of Oodnadatta on which Santos Oodnadatta 1 had revealed academic amounts of Cretaceous oil earlier that year. 'Closure'

was confirmed and seismic depth to the top of the Blythesdale Formation was estimated by reflection at 395 m and by refraction, at 380 m. This compared very satisfactorily with the drilling intersection of 388 m. The team then transferred to Haddon Downs anticlines. The surveys confirmed structure at depth and identified a relatively high speed refractor at 2300 m that later proved to be ? Devonian red-beds. Lucky Strike Oil Company, based in Brisbane, appears to have been the first private company to operate a seismic survey in Australia. In 1958, Geosurveys purchased the equipment and established a base at Innamincka under contract to Delhi Australian Petroleum. Operations confirmed major closure at depth and also outlined a westerly satellite roll-over consistent with the subsequently named Packsaddle Anticline. In 1959, Geosurveys, on behalf of 'wildcatter' L. H. Smart Oil Exploration Company, transferred seismic operations to the Grey Ranges west of Quilpie. A number of outcropping anticlines were confirmed at depth and closure demonstrated. These included Canaway, Pinkilla and Chesson, and it was shown that the strongly asymmetric Canaway anticline gave way to faulting at depth; to the west, possible Permian units appeared to wedge-in. The team then extended operations to Longreach on behalf of Longreach Oil Ltd and followed a prominent basement ridge which nosed to the southwest but failed to develop dip reversal. Wedging-in of probable Permian rocks from this direction was also indicated. The team transferred in 1962 to the Simpson Desert in the Northern Territory and soon after, French Petroleum Company

R. C. SPRIGG

(Australia) Ltd instigated its more extensive reflection and refraction seismic surveys immediately to the south. The highlight of this operation was the completion of a highly informative seismic cross-section across the full width of the Pedirka Basin. In 1959, the South Australian Department of Mines purchased its own seismic equipment, and extended and improved coverage of the Innamincka Dome. The survey confirmed earlier drilling indications that the structure tended to 'bald-heading' at the Permian levels, but that the total sedimentary section thickened markedly down-flank to the south, confirming Frome-Broken Hill Company's earlier predictions. The stage was now set for numerous international seismic teams to expand widely into the Eromanga Basin. Namco International Corporation teams established camp 40 km south of Innamincka and the Delhi-Santos history of drilling and Permian gas discovery that followed is well-documented (e.g. Battersby, 1976). Thereafter, State and Commonwealth seismic crews concentrated more on regional surveys, reconnoitring of neglected areas, or on experimental surveys. The extension of the Commonwealth Government Petroleum Search Subsidy Acts to cover geophysical surveys as well as drilling operations at this time of declining activity gave new impetus to competitive contract operations throughout the Eromanga Basin. By the early 1960s, seismic teams were operating across the length and breadth of the Eromanga Basin. French Petroleum Company was by now efficiently probing the central Simpson Desert via both reflection and refraction surveys, and accurately correlating major velocity discontinuities with known and predicted stratigraphy. All companies were soon mapping the widespread 'C Horizon' (top Transition beds'), the 'P Horizon' (near top Toolachee Formation), the 'Pz' (top older Palaeozoic) and the 'Z Horizon' (presumed basement) reflectors. A range of structure-contour and isopach maps were basic preparations. 'Tramline' seismic sections in some regions revealed prolonged structural stability consistent with even, overall basin subsidence, whereas 'fanning' of reflectors off some structures revealed continuing or episodic structural growth. The latter was to prove particularly important in assessing timing of structural development in relation to predicted hydrocarbon migration. With the discovery of deeper (and hydrocarbon-bearing) infra-basins, higher input frequencies tended intentionally to be filtered out, somewhat at the expense of shallower reflections. Now mugh later, with the increasing concentration on shallower Jurassic-Cretaceous structure consistent with greater oil potential, more effort is being re-directed to improving resolution within the shallower Cretaceous sequence, the better to pin down timing of structural development in relation to the predicted time of maximum oil generation and migration. During the last two decades of seismic exploration, the technology has advanced from wriggle-trace and simple amplitude 'picks' through variable-density and related presentations, and a whole range of computer enhancement is applied at all stages of operation. Close-traverse spacing or '3-D' techniques have also been applied. Well logging, too, has undergone dramatic improvement since the early Innamincka days. A whole range of new and more sensitive tools has been developed, and methods of presentation are upgraded consistently. Drilling mud control has improved equally, as has the reliability of down-hole testing, sidewall coring and other techniques. Automatic recording of a range of drilling parameters and radically

improved gas monitoring has added new reliability to exploration generally. Undoubtedly, potentially significant oil and gas horizons will have been missed in the past, due to either poor technology, preconceived notions, failed equipment, human error or, of course, rated economic significance. THE COMMONWEALTH PETROLEUM SEARCH SUBSIDY ACTS A N D A.P.E.A. Successive Commonwealth and State Governments have had a long and varied history in encouraging oil search. This has included direct participation, the provision of incentives and monetary rewards, advances for drilling, provision of plant and equipment, subsidies of various kinds, supportive geological and geophysical surveys, provision of topographic maps and air photographs, tax concessions of various kinds and oil and gas price fixing. In 1946, the Commonwealth Government established the Bureau of Mineral Resources (B.M.R.) in part to actively explore for oil and minerals as well as to carry out geological and geophysical surveys and other functions. Earlier policies of direct financial assistance to exploration companies for drilling were abandoned. In 1949, the Menzies Government decided that the B.M.R. itself should no longer drill for oil but should rather supply industry with support services. 'It is fair to say! stated B.M.R. Deputy Director, Dr N. Fisher (1975, p. 38), 'that most success has been achieved where vigorous private enterprise exploration companies have been encouraged to operate under favourable Government climate'. Still, Australian exploration remained desperately slow by world standards. For a time following the initial 1954 euphoria of Rough Range, the industry enjoyed new impetus, but this was shortlived. Much of Australia was indeed still little-known geologically. 'Alluvium' on even the best geological maps simply covered untold depths of ignorance and negative thinking. In 1957, when Santos invited international consultant Dr A. I. Levorsen to Australia, Dr H. G. Raggatt, Head of the B.M.R., wisely took the opportunity to avail of Levorsen's extensive experience and opinion. From this emerged the Petroleum Search Subsidy Acts, representing one of the most enlightened support schemes in oil exploration history. Much of the exploration success in the subsequent 1960s relates directly back to this. The first of the P.S.S. Acts was introduced in 1957, in time for Innamincka 1. It provided subsidy at the rate of 50 per cent for stratigraphic drilling only. Wildcat drilling increased rapidly in response, but was not always preceded by desirable geophysical checking. A modified P.S.S. Act two years later provided in addition, a 50 per cent subsidy for acceptable geophysical survey costs, and for the logging of bore holes other than water bores. In 1961, amendments provided for detailed structural drilling and test drilling. Then, in 1962, rather prematurely following the Cabawin gas and Moonie oil discoveries, the maximum drilling subsidy was reduced to 30 per cent. The Moonie Mesozoic oil discovery had, however, provided a turning point. New momentum developed. 1964 proved to be the most significant discovery year in Australia to that time. Alton, Conloi and Gilmore discoveries were made in Queensland, Mereenie in the Northern Territory, Gidgealpa in South Australia, and Barrow Island and Yardarinna in Western Australia. At the same time, new, advanced technology, available literally 'off the shelf' in the northern hemisphere revolutionised Australian exploration efficiency. Amendments

EROMANGA OIL SEARCH then still further reduced the level of subsidy; first to limited areas around discovery wells, finally back to a maximum of $200 000 for any single drilling operation, to finally the Acts' discontinuation in 1973. The Acts had indeed achieved the hoped-for increased level of exploration and, in particular, assured proper recording and early public release of muchneeded stratigraphic and structural information. Until the late 1950s, the Australian exploration front was one of generally rugged individualism and mostly remote, if watchful, interest by international oil companies. Australia was certainly not prime oil country and early indications were that it would be extensively gas prone. In 1959, a group of Australian enthusiasts banded together to form the Australian Petroleum Exploration Association, or A.P.E.A., as it soon became known. A highlight of its activities was the immediate institution of annual conferences open to all concerned. The publication in 1961 of the first A.P.E.A. Journal incorporating all conference papers, proved a major stimulus to the industry and attracted significant overseas interest. The first papers were primarily geological and technical, but new exploration information was being circulated publicly and promptly for the first time. During subsequent years, the new information did much to fill in empty places on the Australian geological map and permit a more 'three-dimensional' understanding of continental geology. The papers supplemented and extended the more academic publications of the Geological Society of Australia and kindred organisations. A forum was provided for new ideas and the reporting of new technologies. Some papers offered a measure of prediction. All in all, the annual A.P.E.A. Journals have come to accumulate a wealth of progressive geological, geophysical and drilling information and provide invaluable insight into, and a history of, the modern Australian petroleum exploration industry. Its continuing stimulus to ongoing exploration and governments alike is incalculable. In 1967, the Professional Division of A.P.E.A. was formed and, by 1974, membership across Australia approached 1000. At this stage, in order to preserve its non-political nature, and later to cater for the industry's professional and more scientific needs, the organisation was separated as the Petroleum Exploration Society of Australia (P.E.S.A.).

AN EVOLVING MESOZOIC STRATIGRAPHY

The Innamincka and Betoota wells threw new light on the stratigraphic history of the central Eromanga Basin. Two thousand metres of Jurassic and Cretaceous strata were revealed and a preliminary fourfold subdivision of the Jurassic emerged. The search for more extensive marine influence continued. Tanner (1966, p. 116) drew attention to the 'Mesozoic basin (as) similar in lithology and stratigraphic history to the shallow cover of the Great Plains area in North America: a transgressive Jurassic to Lower Cretaceous sandstone-shale sequence, a thick grey marine Lower Cretaceous shale with very minor carbonates, and a regressive continental sandstonesiltstone phase of Upper Cretaceous age'. The underlying Triassic sediments merely infilled graben that developed during the post-orogenic phase of the Bowen Orogeny. Much later, it has become popular to try to relate phases of downwarping, marine flooding and regional volcanism (in the northeast) to episodes of Australasian and world-wide plate tectonics. Parallels in intercontinental geology were emphasized also by the recognition of the prominent 'fish scale' horizon (Toolebuc Formation) encountered in Santos Oodnadatta 1 and Delhi-

Santos Mornington Island 1 as equivalent to comparable deposits in North America. Although French consultants, Trumpy et al. (1960, p. 40), saw the marine Cretaceous as offering the only source-rock potential in the Eromanga Basin, most others saw it merely as cap-rock to deeper reservoir potential. The search for deeper Jurassic or older marine accumulations for a time became almost an obsession. Vine (1966, p. 110) in emphasizing the value of gamma-ray well logging noted that 'high radioactivity associated with three argillaceous units within the Jurassic sequence, possibly indicates marine incursions. Also the presence of glauconite and acritarchs suggests marine conditions'. At least they would indicate brackish waters or paralic seas. Exon & Senior (1976, p. 47) summarised the broader geotectonic controls. They noted that clastic quartz-rich sediments dominated the non-marine Jurassic through into the marine Aptian, but that andesitic debris dominated the marine and non-marine Albian and Cenomanian. Furthermore, glauconie (a green clay mineral, seldom true glauconite) and montmorillonite dominated the clay fraction of paralic and marine sequences. The Toolebuc Formation is dominantly a black carbonaceous and bituminous shale with siltstone and limestone lenses and coquinites averaging 15 m in thickness over great areas. Its carbonate content diminishes to the south-southwest. It was deposited in shallow seas with connections to the ocean via the north, under sheltered, reducing conditions. The limestones developed as bioherms colonised by pelecypods. Veevers et al. (1982) in relating the Eromanga Basin to the tectonic framework of Australia emphasized that the hiatus immediately preceding Jurassic deposition was followed by widespread dolerite intrusion in Tasmania and the then conjoined Antarctica. 'By 170 Ma ago, the arc . . . had established itself in its final, most easterly position in Australia —along the Queensland coast. . . And from this arc . . . on two occasions the volcanogenic sediments reached westward into the Eromanga Basin: first, in the middle Jurassic, with the Birkhead Formation representing the distal part of the Injune Creek Group, and secondly in the Early Cretaceous with the Rolling Downs Group. The great volume and extent of the Rolling Downs Group, including the Winton Formation, indicates a terminal climax in the volcanic arc' (Veevers et al., 1982, p. 294). Nugent (1969) provided an extended understanding of Jurassic stratigraphy across the central Eromanga Basin. Predominantly a thick, freshwater sandstone accumulation, he noted (op. cit., p. 99) it is broken by mappable shalesiltstone intervals with minor coals that wedge-out westward. The sandstone is clean, angular to sub-angular, quartzose and porous, and the shale-siltstone lithology is grey, micaceous, carbonaceous and pyritic in part. Nugent concluded that the only effective method of determining the age of the sequence is by palynology involving a modification of Evans' (1966) zones. Nugent (1969) defined two main Jurassic depocentres, one about Windorah, coinciding with the pre-existing PermoTriassic (Cooper) 'low', and one further west about MokariPoolowanna. In the southwest, the sequence is almost continuously sandy, whereas away from the basin margin, it is broken by two shale-siltstone intervals, thickening eastwards as the Birkhead and Westbourne Formations. The underlying Lower Jurassic Hutton Sandstone is partly equivalent to the oil-bearing Chandos beds to the east and the Evergreen Formation and Precipice Sandstone. The Adori Sandstone

R. C. SPRIGG

wedges in eastwards between the Birkhead and Westbourne Formations in the Surat Basin. The shaly Murta Member at the top of the otherwise sandy Mooga Formation in turn gives way into the 'Transition beds' or the Cadna-owie Formation. Overall, Jurassic sedimentation is dominated by fluviatile processes, whereas the finer grained Birkhead and Westbourne Formations relate to lower energy lacustrine conditions. A somewhat revised understanding (Bowering, 1982; Moore, 1982, this volume) resulted from the drilling of the Poolowanna discovery well. Nugent's unnamed lower Hutton equivalent to the southwest has become the Poolowanna Formation, while the overlying Jurassic sequence, which is entirely sandy, has been assigned to the Algebuckina Sandstone of Wopfner et a/. (1970). The Poolowanna Formation is recognised widely across the Simpson Desert region and onto the Birdsville Track Ridge (Fig. 6) and is equated in part to the Evergreen Formation and Precipice Sandstone of the Surat Basin Queensland. The organic shales are believed to have sourced the Poolowanna and Cuttapirrie oils. :

OIL GENERATION AND THE TIMING OF MIGRATION

Early international geologists visiting Australia all but insisted on a marine source for copious oil generation in the light of their basic northern hemisphere experience. General acceptance also of alternative non-marine sources, particularly in paralic deltaic environments that favoured reducing conditions, mostly gained favour in the 1960s (e.g. Hedberg, 1967). Brooks (1970) of the C.S.I.R.O. was able to come out clearly in favour of regarding Australian coaly detritus as potential sourcing material for petroleum generation—something that Bradfield (1927) and others were advocating almost half a century earlier. Brooks (1970, p. 35) concluded that 'petroleum hydrocarbons are not normal constituents of recent sediments, but only appear when a certain stage of diagenesis is reached, through deeper burial . . . Peat and brown coal contain the same type of hydrocarbons as are present in land plants, but the composition of coal hydrocarbons changes abruptly in the sub-bituminous to high-volatile, bituminous coal range.

This is because petroleum-type hydrocarbons are formed at this stage from precursors which are components of waxy leaf cuticles, pollen and spore coatings, by chemical reactions in which oxygen groups are removed from long chain acids, alcohols and ester waxes'. Brooks et a/. (1971, p. 121) extended these deductions and, in relation to the Cooper Infra-basin, noted 'there appears to be a relation between low grade metamorphism of the coaly matter and the nature of hydrocarbons in the reservoirs. Liquid hydrocarbons are not found in areas and at depths where the coals are at the highrank bituminous stage (88-89% carbon, dry mineral-free); there, methane is the main hydrocarbon.present. Oil occurs in association with coals of lower rank (80-85% carbon, dry mineral-free) and it seems possible that underground gasification of liquid hydrocarbons has occurred under natural conditions during advanced coalification'. Powell & McKirdy (1972) confirmed the probably terrigenous origin of Cooper Basin oils. They noted (p. 125) that 'Australian oils are light by world standards. They have A.P.I, gravities greater than 35°, low sulphur and asphalt contents, and are of paraffinic or naphthenic base'. Despite this apparent relatively simple relationship, elsewhere in the world it is clear that some coal measures have generated mainly gas while others have sourced large oil accumulations. The record shows for example, that only minor amounts of oil are associated with Late Palaeozoic coals despite their extraordinary abundance. Thomas (1982, p. 164) in turn drew attention to the fact that the composition of Australian coals (as elsewhere) has varied significantly 'through geological time as a result of differences in climate, geological setting, depositional environment and stage of floral evolution. Consequently, most Australian pre-Jurassic coal measure sequences are deficient in exinite macerals and are therefore mainly gas-prone. In contrast, Jurassic to Tertiary coal-rich sequences often contain abundant exinite and may have substantial potential to generate oil in commercial quantities . . . there appears to be a close relationship between the occurrence of waxy, land-plant-derived crudes and coaly sediments of Cretaceous and Tertiary age. This is thought to be a result of the dominance of conifers in swamp floras of those periods, together with the evolution of angiosperms (flowering plants) in the Late Cretaceous'.

POOLOWANNA TROUGH

oolowanM*

fO-

birdsville

track ridge

Fig. 6. Diagrammatic summary of principal stratigraphic units across the Eromanga Basin. Modified and expanded after Bowering (1982 figs 5, 13).

EROMANGA OIL SEARCH Further, Thomas (1982) noted that land plants' contribution to oil formation relates particularly to accumulation in the paralic-deltaic or marginal marine environment, where reducing conditions obtain. Such relate particularly to the Jurassic to Early Cretaceous siltstone-shale sequences of the Eromanga Basin that tongue in from the east. Given the evidence of suitable source material, generation and migration of petroleum fluids is primarily a function of time, temperature and depth of burial. In the absence of a conflicting hydraulic gradient, it is reasonable to expect that the general expulsion of fluids due to sedimentary loadcompaction would take place primarily vertically and, or radially from a depocentre towards shelf areas and basin margins, as suggested by Bowering (1982). In the Eromanga Basin, then, it seems likely that migration of Mesozoic hydrocarbons might have been initiated in the three deeper depocentres about Tanbar, Yanpurra and Poolowanna. Each extends below 2400 m sub-sea, and each contains organically rich, fine-grained sediments of the Poolowanna and, or Birkhead Formations and others. Greater, more rapid deposition of sediments climaxed in Late Cretaceous time, when burial would have been sufficiently deep to generate hydrocarbons and activate their expulsion. Primary migration would have then taken place upwards and, or radially outwards in a ground water environment that was all but static—at least until the MidTertiary when (?OHgocene) silcretes were being formed. Basinmargin uplift and causally related folding in the central region of the Eromanga Basin followed, possibly in the Late Miocene. This finally gave way to tilting southward towards Lake Eyre, Lake Frome and the Flinders Ranges during the Late Cainozoic. A major change in hydrodynamics of the artesian water system was then initiated with flow now directed strongly towards the general southwest. Several generations of mound spring activity (L. Eyre, L. Frome vicinities) and mud volcanoes (Cunnamulla) followed. Samples obtained by the S.A. Department of Mines in 1958 when drilling the Coward Springs mounds (see Bowering, 1982, p. 234) have revealed the presence of 'hydrocarbons from an external source', which presumably had migrated out of the adjoining Eromanga Basin sediments. The prominent C to C peaks on gas chromatograms of the Coward Springs samples are interpreted to represent hydrocarbons 'with a marine source or a mature oil, or both' (A. C. Cook, pers. comm. to Bowering (1982, p. 234)). The littoral marine Cadna-owie Formation (Transition beds) is believed to have been their source, although long distance migration from the deeper Poolowanna Formation is also possible. The lack of earlier oil discovery in drilling the well-defined central great domal structures and the restriction of 'shows' within them principally to 'tight' sands, encouraged the belief that laterally-moving artesian waters must indeed have flushed most hydrocarbons out of the more porous and permeable sections. Cameron (1959), in discussing artesian hydrodynamics, subscribed generally to this view, but calculated that, even with strong lateral flow, even severe tilting of the oil-water interfaces could not have 'emptied' the major anticlines of their accumulated hydrocarbons. Vine (1976, p. 308) persisted that the failure of 20 000 water wells plus several major exploration tests to make a significant hydrocarbon strike must point to effective flushing of hydrocarbons. Increased sand content westward in the direction of flushing could only assist the process. Despite this, he conceded that facies changes could still present 14

efficient stratigraphic trapping mechanisms in basin margin situations. Bowering (1982, p. 233) more recently recomputed the likely angle of tilt of an oil-water contact using Hubbert's (1953) equation. Employing an average S.G. for Eromanga oil of 0.70 and an average value for the regional hydraulic gradient of 1 in 3000, he arrived at a mere four minute slope of the oilwater interface within an open trap structure. Under such conditions it would be impossible to flush oil from anything but the flattest anticline. An absence of local generation then, or reservoir breaching by erosion or faulting, or alternatively, bad timing of suitable structural trap development in relation to hydrocarbon generation and migration, would appear to be governing factors in the absence of significant oil accumulations. Senior & Habermehl (1980) have produced the most upto-date model for the migration of hydrocarbons under hydrodynamic drive in the Eromanga Basin and also their possible accumulation and entrapment in suitable structural or stratigraphic situations. Habermehl (1982) reported that gas samples taken from about 50 long-established and 'randomly' drilled artesian water wells in the central basin revealed significant methane concentrations (in solution) in the main Early Cretaceous to Jurassic aquifer (Cadna-owie Formation-Hooray Sandstone). Methane ranged up to about 100 000 microlitres per litre of water, while the sum of the hydrocarbons ethane to heptane, reached 2250 microlitres per litre. 'Several samples contain relative high values for the hydrocarbons ethane to heptane, which suggests that the ground water acts as a migration agent for hydrocarbons derived from the adjoining source rocks or from existing reservoirs, (Habermehl, 1982, p. 312). This would presumably suggest a mechanism for actually depleting existing reservoirs. In addition, it could possibly overcome some of the problems of large, early-formed structures that are deficient of hydrocarbons such as Innamincka, as well as across the Birdsville Track Ridge. Whatever the role that such 'solution transport' presents in hydrocarbon migration, most geologists still assume that oil translation is effected predominantly in the immiscible liquid phase. The problem remains in the present author's mind however: just what energy source (or sources) is available to supply the relatively large forces necessary to override the various intergranular resistances obtaining in a relatively consolidated porous sediment? The latter must include friction, surface tension, capillarity, adhesion, osmosis and even inertia. Relative buoyancy of the lighter immiscible petroleum phases certainly offers a major vertical driving force which in confined, inclined bedding situations can also be resolved laterally as well as upward. But what happens when buoyancy forces are exceeded by the intergranular forces previously mentioned? Particularly, what would be the case in flat-lying porous beds where dips of less than one degree obtain? Compaction is one obvious alternative of some efficacy. A very major energy source rarely if ever taken into consideration is that of earthquakes and microseisms—the very forces that move even mountains. Such phenomena undoubtedly generate major inertial forces, and on a regional scale. The resultant omni-directional shaking, heaving and jarring must surely induce gross differential motion into any sand-water-immiscible fluid phase system. Where the fluids are of significant density contrast, then liquid phase separation must be facilitated, buoyancy stimulated, droplets coalesced increasingly, and differential fluid translation affected. The

R. C. SPRIGG

overall effect in any confined sediment must surely amount to accelerated up-dip displacement of the more buoyant immiscible fluid phases. It is a fact that the world's largest oil and gas accumulations are concentrated for good reason in seismically active areas, past or present. The efficiency of hydrocarbon migration and accumulation in these situations appears to be akin to tapping a cylinder full of water-saturated sand that carries in addition suspended globules of less-dense immiscible liquid. Such globules are immediately mobilised and coalesced by the tapping action, and in a system that is otherwise relatively static, these would move upward along even low-angle, imposed gradients, quite efficiently. A subsidiary question arises in the case of the Eromanga Basin—an area traditionally considered to be unusually stable technically. That this indeed was not so in Early Tertiary time is revealed by the marked uplift and folding that resulted in erosional unconformities both above and below the horizon of the silcrete surface. Even today, 'creekology' in the area east of Lake Frome demonstrates that ongoing faulting affects modern sand-dune trends and drainage lines quite noticeably. In the Simpson Desert, accurate survey levellings, along and across interdune corridors, reveal elevational anomalies that similarly reflect deeper folding (? part compaction) and faulting that are also confirmed by reflection seismic surveys to continue fully to the surface. The northern extension of the Poolowanna fault across the Northern Territory-South Australian border in particular is a known locus of significant ongoing seismicity (Fig. 7). Moreover, here the interdune corridor floors slope anomalously back to the northwest in sympathy with deeper, proven down-faulting,. This reversal of the regional south slope of the desert floor appears to be observed nowhere else in the Simpson Desert beyond the influence of the fossil gypsum dune and lagoonal systems of the Poeppels-Warburton association. In retrospect it would appear then, that the most efficient barriers to physically migrating Jurassic-Cretaceous hydrocarbons would be where stratigraphic, structural or combination traps were already developed or developing by Late Cretaceous time at about the peak of the oil generation

Fig. 7. Known earthquakes, Australian region 1900-1971, magnitude 4.0 or greater, showing concentration of epicentres immediately west of Poeppels Corner in vicinity of Poolowanna Fault (after Stewart & Denham, 1972).

and migration phase. Isopach thinning over Mesozoic structures present at about the Toolebuc Formation level would appear to best fit these requirements. Later-developed structures would largely have missed out.

PRODUCIBLE 'EROMANGA' OIL AT LAST

The dream of 'Eromanga' oil took exactly half a century to materialise. While the first barrel of waxy crude was produced from the basal Jurassic at Longreach in 1927, the Poolowanna 1 discovery in the Simpson Desert did not eventuate until 1977. Discovery of a potentially commercial flow had to wait another year. The first commercial oil flow was 2400 B.P.D. from the Hutton Sandstone in Strzelecki 3 in 1978. In 1979, oil flowed at 450 B.P.D. from the Murta Member of the Mooga Formation in Dullingari North 1. A series of discoveries followed including Big Lake, Cuttapirrie, Jackson, Marabooka, McKinlay, Merrimelia, Moorari, Namur and Wackett, in reservoirs all the way through the sequence from the basal Jurassic into the Early Cretaceous Coorikiana Sandstone. Although the best developments to this time appear to be within the Hutton Sandstone, almost the entire stratigraphic sequence now appears to be prospective from the Early Jurassic Poolowanna Formation to the Early Cretaceous Coorikiana Sandstone and Toolebuc Formation. So far, the Toolebuc Formation has produced only oil showings (Santos Oodnadatta 1, in 1957). The Birkhead Formation has not yet produced commercial oil but Nappacoongee 2 (1979), Moorari 3 (1981) and Big Lake 26 (1982) tested promising oil indications. A broad Jurassic-Cretaceous 'oil window' now seems to be quite well defined in relation to temperature-depth parameters, so that vitrinite reflectance relationships can be applied regionally as a reliable indicator of source-bed maturity and consequent oil generation potential. Reflectance values in the range of about 0.6 to 1.0 per cent appear to define the oil window providing however, that the total organic carbon content (T.O.C.) of the sediments involved is reasonable (i.e. greater than about 1%). The new oil discoveries emphasize clearly the importance of timing of trap development in relation to oil generation and migration. Flushing by the movement of artesian water seems not to be a serious deterrent to accumulation in structural traps, but is likely to be an important factor in stratigraphic entrapment. The occurrence of waxy oil in Poolowanna 1, below 2350 m in the Poolowanna Trough, would appear to indicate substantially lower thermal gradients in that region as also appears to be the case in the Tirrawarra Field (a Permian oil reservoir). The presence of a dominantly sandstone sequence representing the bulk of Jurassic accumulation in the more westerly expression of the Eromanga Basin would appear to significantly reduce the trapping potential in that direction. Despite this, the presence of oil 'shows' recorded in Santos Oodnadatta 1 (Toolebuc and ?Cadna-owie Formations: 1957) does suggest continuing up-dip migration of oil towards the basin margins and likely seepage escape, not yet detected. Much subtler trapping mechanisms may be required in these situations. The industry generally waits with interest to learn to what extent the more 'paralic' environments represented at the base of the Hutton Sandstone, Poolowanna, Birkhead and Westbourne Formations and the Murta Member of the Mooga

EROMANGA OIL SEARCH Formation have specifically generated oil. The extent to which the various structures are filled to spill-point and the extent to which stratigraphic factors play a significant role in entrapment will also be matters of concern.

CONCLUSIONS

It is easy to be knowledgeable and clever in retrospect. For more than half a century, a succession of explorers and wildcatters has observed and commented variously on the widespread, teasingly small indications of Mesozoic oil and gas in wells distributed widely throughout the Eromanga Basin. Still, the very existence of many thousand deep artesian water wells without significant oil discovery intimidated all but the most enthusiastic 'believer'. Similarly, geological preconceptions, negative thinking, an absence of obvious surface oil seepages, apparent suitable structure, or of deeper marine section, poor pricing and so forth, all mitigated against early success. It seems that Wallace Pratt's observation still stands. T h e presence of oil in the ground is not enough, Gold is where you find it according to the old adage but, judging from the record of our experience, oil must be sought first of all in the minds of men' (Pratt, 1940). The original SantosGeosurveys team certainly had the inspiration and the combined Delhi-Santos consortium is now achieving success. Looking back, it is obvious that the giant Innamincka and Betoota domes provided hopes that never quite materialised. Permian gas then saved the day for the overlying Eromanga Basin by extending operations into the 'O.P.E.C! era when oil prices soared from $1.85 per barrel in 1971 to $11.63 in 1973 and to $34.00 by the end of 1981. The world energy picture had changed forever and the Delhi-Santos consortium could now justify seeking considerably smaller accumulations. The 1977 Poolowanna 1 discovery inflamed new hopes and now with liquids pipelines through from Moomba to Stony Point and Jackson to Moonie, exploration for oil has taken on a new complexion. Once again, it is demonstrated that, given attractive pricing (be it for oil or gas), and favourable outlets and markets, the Australian exploration industry has

ACKNOWLEDGEMENTS

Most of the information employed in this study is from published works and my own daily diaries. Acknowledgement is made specifically for the provision of early company reports by Shell Development (Queensland) Pty Ltd, the Frome Broken Hill Co. Pty Ltd and Geosurveys of Australia Pty Ltd, now a subsidiary of Beach Petroleum N.L. Appreciation is expressed also to Beach Petroleum N.L. and officers of Delhi Petroleum Pty Ltd for assistance in finalizing this summary for publication. Personally, I wish to emphasize past valued associations with Sir Harold Raggatt, Jack Raynor and others of the B.M.R., also Sir Maurice Mawby, Harry Eyans and Bob Lesley of Frome Broken Hill Co Ltd, Dr Ivan Chebotarev of the South Australian Geological Survey, Dick Mott of the Queensland Geological Survey, and many other leaders in government and industry; also consultants Dr Frank Reeves (Virginia, U.S.A.), Dr H. I. Levorsen (Tulsa, U.S.A.), Dr Lewis G. Weeks (Connecticut, U.S.A.) and last but not least, with entrepreneur and prominent South Australian, John L. Bonython, A.O. Together, John and I set out in 1954 in the face of considerable official apathy and scepticism, not only to further the search for Cambrian oil in South Australia and the Northern Territory, but also to explore the central Great Artesian Basin in detail. Special appreciation is accorded to the hundred or so geologists, geophysicists and engineers employed by Geosurveys of Australia Pty Ltd across the length and breadth of Australia between 1954 and 1961, many of whom made significant contributions to the geological understanding of the continent and its resources.

REFERENCES

B A L L , L. C., 1 9 7 2 : The search for oil. Qld Gov. Min. J. 28, 3 5 7 - 8 . BATTERSBY, D . G., 1976: Cooper Basin gas and oil fields; in Knight,

C. L. (ed.) Economic geology of Australia and Papua New Guinea, 3, Petroleum. Australas. Inst. Min. Metall. Monogr. 7, 321-68. B O W E R I N G , O . J. W . , 1 9 8 2 : Hydrodynamics and hydrocarbon migration—a model for the Eromanga Basin. APE A J. 22(1),

226-53. B R A D F I E L D , J. J. C., 1943: Oil from coal. Qld Geogr. J. 47-52. BROOKS, J. D., 1970: The use of coals as indicators of the occurrence

of oil and gas. A PEA J. 10, 35-40. W. R . & RIGBY, D . , 1971: The natural conversion of oil to gas in sediments in the Cooper Basin. A PEA

BROOKS, J. D . , H E S P ,

J. 11, 1 2 1 - 5 . C A M E R O N , J., 1959: Hydrodynamics and drape folding. Qld

Gov. Min. J. 60. I., 1952: Hydrological and thermal aspects of petroleum occurrence. Am. Assoc. Pet. Geol. Bull. 36, 688-99. CHEBOTAREV, I., 1955: Metamorphism of natural waters in the crust of weathering. Geochim. Cosmochim. Acta, 8, 22-48, 147-70, 198-212. DELHI A U S T R A L I A N PETROLEUM LTD., 1966: Delhi-Santos Gidgealpa no. 1 well, South Australia. Bur. Min. Resour. Aust. Pet. Search Subs. Acts Publ. 73.

CHEBOTAREV,

the drive and ability to commit many hidden oil and gas fields to meaningful production. Wallace Pratt's immortal words (in Knowles, 1978, p. 334) were never truer, 'Oil in the earth is far more abundant and far more widely distributed than is generally realised. The prime requisite to success is freedom (and ability . . . RCS) to explore'.

D E VERTEUIL, J. P., 1 9 4 0 : Report

on oil prospects in the northwest region of New South Wales and the northeast region of South Australia comprising part of the Great Artesian Basin. Zinc Corp. Pty Ltd (unpubl.). EVANS, H. J., 1 9 4 6 : Report on natural gas exploration, Frome embayment New South Wales and South Australia and adjoining areas. Frome Broken Hill Company (unpubl.). EVANS, P. R., 1966: Mesozoic stratigraphic palynology in Australia. Australas. Oil Gas J. 6, 58-63. E X O N , N. F. & SENIOR, B. R., 1976: The Cretaceous of the Eromanga and Surat Basins. BMR J. Aust. Geol. Geophys. 1, 33-50. F I S H E R , N . H., 1 9 7 5 : Some thoughts about incentives. A PEA J. 15, 38. FORBES, B. G . , 1959: Boring on mound springs near Coward Springs. S. Aust. Dept Mines, Min. Rev. Ill, 1 4 5 - 4 9 . H A B E R M E H I , M. A., 1982: Aspects of regional groundwater movement and hydrocarbon migration in the Eromanga Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust & Pet. Explor. Soc. Aust. Adelaide. H A R R I S O N , J., W A R N E R , R. A . & G I B S O N , A . R., 1961: Delhi-FromeSantos Betoota no. 1, Queensland. Well completion report. Bur. Min. Resour. Aust. A*/. Search Subs. Acts Publ. 10.

R. C. SPRIGG

HEDBERG, H. D., 1967: Geological controls in petroleum genesis. Proc. Seventh World Petrol. Congr. 2.

RYAN, J., 1961: Delhi Frome Santos Innamincka no. 1 well, South Australia. Bur. Min. Resour. Aust. Pet. Search Subs. Acts Publ. 9.

HUBBERT, M. K., 1953: E n t r a p m e n t of petroleum under hydrodynamic conditions. Am. Assoc. Pet. Geol. Bull. 37, 1954-2056. JACK, R. L., 1925: Some developments in shallow water areas in the northeast of South Australia. S. Aust. Geol. Surv. Bull. 11, 13-7.

S E N I O R , B. R., M O N D , A . & H A R R I S O N , P. L., 1978: G e o l o g y o f t h e

JACK, R. L., 1930: Geological structure and other factors in relation to underground water supply in portions of South Australia. S. Aust. Geol. Surv. Bull. 14, 9-17.

S H E L L ( Q U E E N S L A N D ) D E V E L O P M E N T L T D , 1951: General

KAUFMAN, F. & M C P H A I L , M . R . , 1948: Coordination

and geophysical Great Artesian (,unpubl.).

geological

data from the southwestern portion of the Basin, Australia. Frome Broken Hill Co.

KENNY, E. J., 1934: A geological reconnaissance of the West Darling District. N.S.W. Geol. Surv. Bull. 36. KING, D. & FALVEY, M., 1977: A seismic survey in the Cooper Basin, O l d . A PEA

J. 17,

LESLEY, R. B., 1976: Eromanga Basin; in Knight, C. L. (ed.) Economic geology of Australia and Papua New Guinea, 3, Petroleum. Australas. Inst. Min. Metall. Monogr. 7, 265. MAWBY, M., 1944: A summary of the possibilities of obtaining commercial natural gas supplies in Australia. Zinc Corp. Pty Ltd {unpubl.). MCKIRDY,-D. M., 1982: Aspects of the source rock and petroleum geochemistry of the Eromanga Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 258-9. MOORE, P. S., 1982: Mesozoic geology of the Simpson Desert region, northern South Australia; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 46-57. P. S.,

PITT,

DETTMANN,

E.,

1986:

The

Coorikiana Sandstone and Toolebuc Formation: their stratigraphic relationship in the southern Eromanga Basin; in This volume. M O R E T O N , H . C . , 1927: T h e s e a r c h f o r oil. Qld

Gov.

Min.

J. 28,

317.

MOTT, W. D., 1952: Oil in Queensland. Qld Gov. Min. J. 53, 848-60. NUGENT, O. W., 1969: Sedimentation and petroleum potential of the Jurassic sequence in the southwestern Great Artesian Basin. A PEA

J. 9, 9 7 - 1 0 7 .

OSBORNE, N., 1945: Report on oil and gas possibilities of the Frome Embayment, South Australia. Frome Broken Hill Company (unpubl.). PlTTMAN, E. F., 1895: Note on the Cretaceous rocks in the northwestern portion of New South Wales. Aust. Assoc. Adv Sci. 6. PITTMAN, E. F., 1915: The composition and porosity of the intake beds of the Great Artesian Basin. Geol. Surv. N.S.W. Bull. 1-15. PORTER, C. R., 1978: The Pedirka Basin—a preliminary exploration review. Aust. Pet. Explor. Assoc. seminar, Adelaide {unpubl.). POWELL,

MCKIRDY,

and petroleum potential, central Eromanga Basin, Queensland, Australia. BMR J. Aust. Geol. Geophys. 5, 47-55. Eromanga Basin. Aust. Bur. Min. Resour. Geol. Geophys. 167.

Bull.

report

investigations and operations carried out by the company in the search for oil in Queensland, 1950-51. (unpubl.). SPRIGG, R. C., 1958: Petroleum prospects of the western parts of the Great Australian Artesian Basin. Am. Assoc. Pet. Geol. Bull. 42,

2465-91.

SPRIGG, R. C., 1961: On the structural evolution of the Great Artesian Basin. APEA J. 1, 37-56. SPRIGG, R . C . , & STAFF (GEOSURVEYS OF AUSTRALIA LTD), 1958: T h e

78-84.

KRIEG, G. W., 1982: Stratigraphy and tectonics of the Dalhousie Anticline, southwest Eromanga Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 145-58.

MOORE,

SENIOR, B. R . & H A B E R M E H L , M . A . , 1980: S t r u c t u r e , h y d r o d y n a m i c s

M.,

1972:

The

characterisation of Australian crude oils. APEA

geochemical

J. 12, 125-31.

PRATT, W., 1940: Oil in the earth. Univ. Kansas Press, Lawrence, Kansas. QLD Gov. MIN. J., 1927: The search for oil, 93-4. QLD Gov. MIN. J., 1927: Motor fuels from coal, 61. REEVES, F., 1951: Australian oil possibilities. Am. Assoc. Pet. Geol. Bull. 35, 2479-525.

Great Artesian Basin in South Australia; in Glaessner, M. F. & Parkin, L. W. (eds) The geology of South Australia. Geol. Soc. A u s t . J. 5, 8 8 - 1 0 1 . S T E E L , B. D., 1927: T h e s e a r c h f o r o i l . Qld

Gov.

Min.

J. 28,

386-9.

STEWART, 1. C. D. & DENHAM, D., 1974: Simpson Desert earthquake, Aug. 1972. Geophys. J. R. Astr. Soc. 39, 335. TANNER, J. J., 1966: Distribution of Paleozoic rocks beneath the Great Artesian Basin, Queensland. APEA J. 6, 116-20. THOMAS, B. M., 1982: Land-plant source rocks for oil and their significance in Australian basins. APEA J. 22, 164-78. THOMAS, N. M., 1960: Geothermal studies, Great Artesian Basin, Queensland, South Australia and New South Wales, Australia. Frome Broken Hill Company Pty Ltd {unpubl.). TRUMPY, D., GUILLEMOT, J. & TISSOT, B., 1960: Petroleum

prospects

in Australia: preliminary review of Australian sedimentary basins. Bur. Etudes Geologiques Inst. Francais Petrole. (Rep. to Aust. Gov., unpubl.). VACUUM O I L COMPANY P T Y L T D , 1948: Coordination

geological

and geophysical data from southwest portion of the Great Artesian Basin, Australia. Frome Broken Hill Company Pty Ltd (unpubl.). VEEVERS, J. J., J O N E S , J. G . & P O W E L L , C . M C A . , 1982: T e c t o n i c

framework of Australia's sedimentary basins. APEA

J. 22,

283-300.

VINE, R. R., 1966: Recent geological mapping in the northern Eromanga Basin. APEA J. 6, 110-5. VINE, R. R., 1976: The Eromanga Basin; in Knight, C. L. (ed.) Economic geology of Australia and Papua New Guinea, 3, Petroleum. Australas. Inst. Min. Metall. Monogr. 7, 306-9. WADE, A., 1950: Fifty yeats of searching for oil in Queensland. Qld Gov. Min. J. 51. WARD, L. K., 1944: The search for oil in South Australia. S. Aust. Geol. Surv. Bull. 22, 4-40. WHITEHOUSE, F. M., 1954: The geology of the Queensland portion of the Great Australian Artesian Basin. Appendix G; in Artesian water supplies in Queensland. Dept. Co-ord. Pub. Works, Qld, Rept, A56-1955, 1-20. WILLIAMS, G. D., 1966: The Great Artesian Basin—origin and history. APEA J. 6, 88-92. WOOLNOUGH, W. G., 1927: Presidential address. Roy. Soc Proc. 61, 31-2.

NSW

WOPFNER, H., 1960: On some structural developments in the central part of the Great Artesian Basin. Roy. Soc. S. Aust Trans 83 179-93. W O P F N E R , H . , FREYTAG, I. B., & H E A T H , G . R., 1970: B a s a l J u r a s s i c -

Cretaceous rocks of western Great Artesian Basin, South Australia: stratigraphy and environment. Am. Assoc. Pet Geol Bull.

54,

383-416.

Geological Society of Australia Special Publication No. 12, 25-38

The Eromanga Basin. An overview of exploration and potential J. D. Armstrong & T. M. Barr SANTOS Limited, 39 Grenfell Street, Adelaide, S.A. 5000.

ABSTRACT The Eromanga Basin, of immense area and volume, has been explored for hydrocarbons since 1924. Covering 1 000 000 square kilometres it has recently become prominent in the Australian petroleum scene with 11 significant discoveries being made since 1975. The hydrocarbon potential of the Eromanga Basin was largely ignored in the past by the majority of earth scientists. In reviewing the literature only a few authors, for example R. C. Sprigg and G. D. Williams, expressed a positive view of the prospect of finding hydrocarbons in the Eromanga Basin. The Namur 1 gas discovery in 1976 confirmed the optimism of Sprigg and Williams to some extent, but it wasn't until substantial oil flows were obtained from Strzelecki 3 in 1978, and from Dullingari North 1 in 1979 that attention began to be more seriously focused on the basin. Discoveries have followed since 1978 with finds such as Wackett, Dullingari, Cuttapirrie, Merrimelia, Jackson and Jackson South. With the completion in 1982 of a liquids pipeline from Moomba to Port Bonython, 3.5 million barrels of oil were produced in 1983 from fields in the South Australian Eromanga Basin. The existence of these facilities will provide a significant impetus to exploration in the reduction of the lead time between discovery and production. If it is assumed that there could be an undrilled prospect every 500 square kilometres over the central 50 per cent of the basin, then the Eromanga Basin could contain 1000 prospects yet to be investigated. The 1000 prospects represent some 2.3 million metres of drilling and using the cost of $740 per metre, which is a cost that makes allowance for both seismic and drilling, then this equates with exploratory work worth $1.7 billion. This exploration is significant from a commercial viewpoint in that Australia's economy has remained relatively buoyant in the last 10 years due to the substantial production from the Gippsland Basin. However, this production will decline during the last 2 decades of the present century and there will be an ever-widening gap between the demand for petroleum-based fuels and indigenous production. Successful exploration of the Eromanga Basin could contribute to offset the anticipated decline in Gippsland Basin production over the next 20 years. For example, applying a 1 in 20 success to the hypothetical 1000 undrilled prospects, and assuming that half the discoveries were oil, with Jackson Oil Field size recoverable reserves, then there could be some 600 million cubic metres of recoverable oil to be discovered in the basin. The importance of this potential to the Australian economy in maintaining a high level of domestic oil production cannot be understated. The Eromanga Basin is large and petroleum has accumulated in almost all of the Jurassic reservoirs in the section at one location or another. If exploration is to be encouraged, exploration companies need security of tenure of exploration permits and continued world market pricing for the petroleum produced. To achieve success, earth scientists should look forward with a breadth of vision commensurate with the potential of the basin.

INTRODUCTION The Eromanga Basin (Fig. 1) is the largest onshore accumulation of sedimentary rock in Australia and in areal extent is approximately half as large as the Australian continental shelf. Its petroleum potential has in the past been neglected and it is only in recent times that the Eromanga Basin has been recognised as a significant exploration and production area. This paper will attempt to assess the Eromanga Basin in terms of what has been discovered, what

the potential of the Eromanga Basin is, and how this potential can be realized in the future. The Eromanga Basin is located centrally in the Australian continent, spanning three States and the Northern Territory, with an areal extent of 1 000 000 square kilometres. It has been an important source of water for many years and studies of the hydrogeology have been an important feature of the geoscience literature in the past. The Eromanga Basin is serviced with a petroleum production centre at Moomba,

J. D. ARMSTRONG & T. M. BARR

Fig. 1. Location map, Eromanga Basin.

which is located in the northeast corner of South Australia and is the focal point for Cooper and Eromanga Basin gas production from which sales gas is supplied to both Sydney and Adelaide. Production of oil from both the Cooper and Eromanga Basins commenced in 1983 with the completion of a liquids pipeline between Moomba and Port Bonython (Fig. 1).

STRUCTURE

The Eromanga Basin was first defined by Mott (1952) as extending south of the Euroka Arch, west of the pre-Permian rocks of eastern Australia and separated from the Surat Basin by the Nebine and Eulo Ridges to the southeast. The basin outcrops in the southwest and is bounded by the pre-Permian rocks of the Flinders and Olary Ranges. It is underlain by a number of older Palaeozoic basins (Fig. 2) which have influenced the deposition of Eromanga Basin sediments. A contour map drawn at the approximate JurassicCretaceous boundary (Wiltshire, 1979) depicts the Eromanga

Basin as a broad downwarp (Fig. 3). At this regional mapping scale the main structural features of the Eromanga Basin can be seen. In its eastern half northeasterly trends are apparent, influenced by the Permian Cooper Basin and its internal structural features with the Patchawarra and Nappamerri Troughs, and the Innamincka Dome, recognizable in the overlying Eromanga Basin (Fig. 4). In the western sector of the Eromanga Basin, the major influence is the Poolowanna Trough which is a large synclinal area separated from the Cooper Basin by the northeast-trending Birdsville Track Ridge. This area has been influenced by both the older Pedirka and Simpson Desert Basins, as the broad structural low coincides with these basins, indicating a depocentre which was controlled by the underlying structural trends. In the eastern sector of the Eromanga Basin the Canaway Ridge (Fig. 4) is a major structural feature trending northsouth and coincident with the northern boundary of the Permian Cooper Basin (Fig. 2). The Canaway Ridge effectively truncates the deeper segment of the Eromanga Basin on its eastern side (Fig. 3).

EROMANGA BASIN EXPLORATION AND POTENTIAL

Fig. 2. Older basins beneath the Eromanga Basin.

STRATIGRAPHY, SOURCE AND RESERVOIR POTENTIAL The traditional Eromanga Basin sequence ranges from Early Jurassic to Late Cretaceous in age, however Wiltshire (1982) suggested that it also includes Late Triassic sediments which have been discovered more recently in Poolowanna 1 and Cuttapirrie 1. The simplified stratigraphic sequence is shown on Figure 5. In general terms the Jurassic section consists of continental deposits of quartzose sandstone, carbonaceous siltstone and minor coal attaining a maximum thickness of 1200 metres. The Early Cretaceous (Neocomian) is a transitional paralic sequence of quartz-rich sandstone and siltstone (Senior et al.y 1975). Marine conditions were established in the Early Cretaceous (Aptian) and fine-grained glauconitic and pyritic sandstones were deposited in a shallow marine environment.

Marginal marine conditions and finally non-marine conditions prevailed in the Late Cretaceous during which a sequence of siltstone, sandstone and shale with minor coal interbeds was deposited. The stratigraphic nomenclature has been greatly modified since its establishment by Whitehouse (1954) and papers presented in this volume address these changes. An informal dual nomenclature is used herein, one for the South Australian sector and one for the Queensland sector. These are based on general usage within the DelhiSantos group. Hydrocarbon discoveries within the Eromanga Basin sequence have shown that generally the Jurassic-Cretaceous sequence has sourced hydrocarbons. Within the JurassicCretaceous sequence there are three essentially fine-grained carbonaceous rock units: (1) within the Jurassic Birkhead Formation, (2) the Westboume Formation, and (3) the Early Cretaceous Murta Member of the Mooga Formation.

J. D. ARMSTRONG & T. M. BARR

The mid-Jurassic Birkhead Formation, an organic-rich shale (Mudge, 1980; Mclntyre, 1980), has sourced hydrocarbons and six discoveries have been made within the unit. The Early Cretaceous Murta Member of the Mooga Formation is host to discoveries at Dullingari and Jackson, with significant oil generation. Geochemistry of Murta Member samples from the Dullingari Field indicates that the dispersed-organic-matter (DOM) content exceeds 2.8 per cent (Mclntyre, 1980) with exinite being the dominant maceral. The source rock data available from the Murta Member of the Jackson Field indicate that between 1.5 and 3.5 per cent DOM is present in this sequence, suggesting petroleum generative capacity (Ambrose et al.y 1982, this volume). The Murta Member is a good source rock, resulting from deposition in a relatively low energy, reducing, lacustrine environment. However, Kantsler et al. (1982) suggest that the Dullingari Murta oil

is sourced from older rocks. The fact that the shale above the Murta oil reservoir was an adequate seal, suggests that the underlying shale also would have prevented vertical migration of oil into the Murta reservoir. The Early Jurassic reservoirs which flowed oil at Cuttapirrie and Poolowanna indicate that either the underlying Triassic (Barr & Youngs, 1981) or the Early Jurassic (Thomas, 1982) are likely to have been the source rocks. While the Triassic sequence is conventionally assigned to older basins, its source potential is obviously significant to the Eromanga Basin. The fluvial regime responsible for most of the Jurassic sequence provided conditions for the deposition of reservoir rocks. The Hutton Sandstone is a good reservoir exhibiting porosities of up to 20 per cent with permeabilities of up to 1 darcy. A flow from Strzelecki 3 of 359 cubic metres per day

EROMANGA BASIN EXPLORATION AND POTENTIAL

is evidence of the ability of the reservoir to flow oil. Similar rock characteristics apply in the Namur Sandstone Member of the Mooga Formation, which is described as a gritty, poorly sorted, tabular cross-bedded sandstone (Ambrose et al., 1982, this volume), and a flow of 289 cubic metres per day from Dullingari 22 is indicative of the excellent reservoir properties. The Murta Member reservoirs at Dullingari and Jackson resulted from the extensive reworking of shoreline sediments, and the main producing sand in the Dullingari Field exhibits porosities of around 15 per cent, and permeabilities which vary, but which have been measured as high as 3 darcies.

EXPLORATION HISTORY

The search for hydrocarbons in the Eromanga Basin seems to have begun in Queensland in 1924. Mott (1952) records

a well 'Blume's Malta', sunk in an area east of Tambo, which had been drilled to explore for hydrocarbons, and recovered traces of yellow oil and grease at 987 metres. Later, in 1927 the Lander Oil Company drilled holes in the Longreach area to follow up 'shows' from the town's water supply bore, and recovered oil from 987 metres. These two occurrences were the beginning of an exploration effort which has spanned 60 years but still remains in its infancy. Sprigg (1982, 1983, this volume) records the earliest years of exploration of the basin and shows that the period between 1924 and 1954 was one where little exploratory drilling was carried out; rather the very basics of exploration were attempted, surface mapping, some gravity and magnetic work and seismic traversing to confirm surface features. He also makes the point that cheap imported oil negated the local

J. D. ARMSTRONG & T. M. BARR

30 AGE

SOUTH

OUEENSLAND

AUSTRALIA

WINTON BLANCHEWATER

FORMATION

MACKUNDA

—J

ALLARU

<111

JURASSIC JLY MID LATE

CRETACEOUS

OODNADATTA

FORMATION

COORIKIANA

SANDSTONE

BULLDOG

SHALE

TRANSITION

BEDS

MURTA

MCJCXJIA

NAMUR

FORMATION SANDSTONE

basal

JURASSIC

FORMATION

TRANSITION

MEMBER

HUTTON

FORMATION

WALLUMBILLA

FORMATION BIRKHEAD

FORMATION MUDSTONE

TOOLEBUC

MEMBER

SANDSTONE

FORMATION

WESTBOURNE FORMATION ADORI SANDSTONE BIRKHEAD FORMATION HUTTON

BEDS

MURTA MEMBER HOORAY SANDSTONE

SANDSTONE "basal

JURASSIC

Fig. 5. Generalized stratigraphy—Eromanga Basin.

exploration effort, and the remoteness of prospective parts of the Eromanga Basin discouraged explorers. A disincentive during this period was the fact that the massive amount of drilling for artesian water had not discovered appreciable oil and therfore was seen by some as proof that the Eromanga Basin was devoid of hydrocarbons. However, reasonable 'shows' were recorded by drillers and have been consistently referred to by Sprigg. Moreover Sprigg (1958b) significantly observes that pastoralists selected water bore sites along major water courses which were invariably lying along synclines. This was important inasmuch as it implied that most of the water-well drilling had not evaluated the hydrocarbon potential of the Eromanga Basin. Indeed today it has become obvious that successful wildcat petroleum wells have to be located very specifically at the structural culmination. The distribution of 'shows' recorded both in artesian and petroleum wells is illustrated on Figure 6. A concentration of 'shows' around Longreach reflects the density of the earliest exploration efforts. In South Australia, the recorded 'shows' in the Lake Frome area were a source of encouragement. The acquisition of a large portion of the Eromanga Basin by SANTOS in 1954 and a farm-in by Delhi International Petroleum in 1958 began what Reg Sprigg (1982; 1983) quotes as 'the long lead up to a commercial discovery'. This progress commenced with the drilling of a stratigraphic hole Oodnadatta 1. The well was located in the western part of the Basin and it encountered hydrocarbon shows in the Eromanga Basin sequence. Since then numerous wells have recorded significant hydrocarbon shows in the Eromanga Basin sequence, however the overall scientific opinion which prevailed until 1980 was that the Eromanga Basin had either been flushed or had not generated oil, and the chances of discovery were minimal. A review of the literature is enlightening, with Frank Reeves, an eminent American

geologist, publishing 'It is evident that commercial accumulations of oil will not be found in Mesozoic formations within the Artesian Basin' (Reeves, 1951, p. 2519). He based this thesis primarily on the lack of marine sediments and on the fact that some 5000 water wells had not found any hydrocarbons. This opinion contrasts with that of Sprigg (1958a, p. 17) 'In the face of all recent evidence, the potential for oil in the untested parts of the Great Artesian Basin makes it a very encouraging prospect'. The recent evidence Sprigg referred to at that time was the large surface mapped anticlines, the marine Cretaceous sediments, the porosity of the aquifers and the oil and gas 'shows'. Later in 1965, following the discovery of Permian gas, Sprigg's opinion had remained unchanged: 'commercial oil discovery remains a definite probability' (Sprigg, 1965, p. 167). G. D. Williams in 1966 showed similar optimism and drew the analogy that the Surat Basin had shown its potential and therefore 'it is suggested that the anticlinal folds in the centre of the Eromanga Basin and its western rim also have good potential' (Williams, 1966, p. 92). This optimism was not shared by E. R. Smith who in 1977 in assessing Australia's onshore basins stated 'It is difficult to recommend further exploration in the Eromanga Basin considering past results' (Smith, 1977, p. 21). Smith's primary reason for this statement was the 'flushing theory' which is now known to be not applicable to the basin (Moriarty & Williams, 1982; Williams & Moriarty, this volume). This same thesis persisted until 1978 when Senior et al (1978, p. 46) stated 'Prospects of finding commercial hydrocarbons in the Eromanga Basin in light of the present knowledge are poor'. Gidgealpa 2, drilled in 1964 had a substantial impact on the petroleum development of the region. This well was drilled updip from Gidgealpa 1 and discovered Permian gas, and with that encouragement the SANTOS-Delhi consortium concentrated on the Permian sediments below the Eromanga

EROMANGA BASIN EXPLORATION AND POTENTIAL

Basin for the next 16 years. Insofar as the considered opinion was that the Eromanga Basin held little hope for hydrocarbons, a great majority of wells up to the 1976 Namur Jurassic gas discovery lacked two very basic attributes fundamental to the evaluation of an exploration well. In most wells no geologist was assigned to the well prior to intersecting the Permian, and basic electric logging suites were not run over the Eromanga Basin interval. The result was incomplete evaluation of the Mesozoic sequence. In summary, the overall prognosis, until the first discovery of Eromanga oil at Strzelecki, was pessimistic. This pessimism prevailed in spite of the evidence of 'shows', source rocks and known structures.

DISCOVERIES The first commercial discovery of petroleum in the Eromanga Basin was in 1976 when Namur 1 produced gas. This well, primarily drilled for a Permian target was located on the crest of an elongate anticline (Fig. 7). Testing established a gas flow of 0.4 million cubic metres per day from the Namur Sandstone Member of the Mooga Formation, and additional drilling defined reserves of 1.21 billion cubic metres of gas in place (Table 1). Poolowanna 1 has significance in that it was the first well to recover oil from the Jurassic in an appreciable quantity. The well was drilled in 1977 and recovered oil on test from a basal Jurassic section informally named the 'Poolowanna

TABLE 1 C h a r a c t e r i s t i c s o f E r o m a n g a B a s i n d i s c o v e r i e s t o Discovery Date

Field

1976 1977 1978 1978 1981 1982 1979 1982 1979 1980 1981 1981 1981 1981

Namur Poolowanna Wackett Strzelecki Strzelecki Strzelecki Dullingari Dullingari Nappacoongee Cuttapirrie Marabooka Moorari McKinlay Merrimelia

1982 1981

Merrimelia Jackson

1981 1982

J a c k s o n Sth. Big L a k e

Formation

Namur Mbr. Basal Jurassic Birkhead F m . H u t t o n Sst. Namur Mbr. C o o r i k i a n a Sst. Murta Mbr. Namur Mbr. Birkhead F m . Basal Jurassic Namur Mbr. Birkhead F m . Mooga Fm. Murta Mbr. Namur Mbr. H u t t o n Sst. Birkhead F m . Murta Mbr. Westbourne Fm. H u t t o n Sst. Westbourne Fm. Birkhead F m .

Abbreviations MSCM : thousand standard cubic metres M M C M D : m i l l i o n c u b i c m e t r e s g a s per d a y CMPD : c u b i c m e t r e s oil per d a y MMCM : million cubic metres oil BCM : billion cubic metres gas GIP : g a s in p l a c e OIP : o i l in p l a c e * N o flow to surface

1982.

Reservoir Depth

Closure Area

Accum. Area

Max. Pay Thickness

metres

sq. km

metres

- 1630 -2469 - 1573 - 1600 - 1341 -926 - 1310 - 1402 - 1542 -2404 - 1407 -2103 - 1200 - 1524 - 1562 - 1821 - 1820 -975 - 1219 - 1310 - 1188 - 1918

10.0 46.3 51.0 17.0 15.9 15.9 46.5 46.5 5.5 11.8 4.3 15.9 30.3 50.6 50.6 50.6 50.6 2.0 25.7 25.7 25.7 32.3

7.8

21 None 10 17 15 3 1 12 1 4 6 5 3 1 2 10 3 2 5 30 5 2

—

1.6 2.7 1.9 —

36.4 3.6 —

3.8 3.3 2.8 — —

1.2 1.6 2.0 —

4.2 5.3 5.3 0.9

N O T E : T h e proved, probable reserve e s t i m a t e s a r e in p l a c e f i g u r e s a s at J u l y , 1 9 8 2 a n d r e f l e c t t h e estimates accepted by S A N T O S Ltd and the authors and not necessarily those o f the participating c o m p a n i e s in t h e v a r i o u s f i e l d s .

Oil Characters Pour Pt Gravity "API

°C

—

co2

G a s Characters Cond. c3 +

4.5

2.25

Porosity

29.2

0.17

—

2.25

8.35

209.8

—

0.17

6.58

84.6

54 53 45 53

-3

—

4.12

5.64

117.8

51 41 51 51 55 53 51 41 40 40 48

—

15 -5 -5 -5 -5 0 19 20 14 9

—

0.15 0.17 0.17 0.20 0.14 0.18 0.12 0.10 0.20 0.10 0.15 0.15 0.20 0.22 0.13 0.15 0.21 0.19 0.13 0.13

—

— —

—

Proved/Probable GIP/OIP (BCM)/(MMCM)

45 41

—

Test Rate

m ' gas 0.4 M M C M D 28.6 C M P D * 2.2 M M C M D 81.0 C M P D 111.0 C M P D 9.1 M C M D 71.5 C M P D 317.9 C M P D 3.1 C M P D * 25.4 C M P D 0.2 M M C M D 15.9 C M P D * 1.3 C M P D * 4.7 C M P D * 69.9 C M P D 435.6 C M P D 11.9 C M P D * 54.0 C M P D 214.6 C M P D 377.0 C M P D 101.7 C M P D 23.8 C M P D *

1.22 BCM v. small 0.54 BCM 3.6 M M C M 1.4 M M C M small 1.7 M M C M 1.5 M M C M v. small 0.6 M M C M 0.5 BCM 0.4 M M C M v. small small 0.16 M M C M 0.9 M M C M small 0.01 M M C M 0.84 M M C M 6.13 M M C M 0.79 M M C M 0.08 M M C M

> 33

z o Rp H

70 70

EROMANGA BASIN EXPLORATION AND POTENTIAL

beds' (Wiltshire, 1978). Commercial production has not been possible from the well as it appears that the oil accumulation is relatively small and the oil flow preceded a strong water flow. In 1978 Strzelecki 3 was drilled again primarily with Permian objectives but with a secondary Jurassic target. The well intersected oil shows in both Namur Sandstone Member and Hutton Sandstone reservoirs. An oil flow of 348 cubic metres per day was obtained from the Hutton Sandstone and established the Eromanga Basin as an oil province. This field has been subsequently developed and came on stream in 1983, with production from both reservoirs. Reserves in this field are estimated to be 5 million cubic metres of oil in place (Table 1). Dullingari North 1 was drilled as a gas appraisal well on a separate structural culmination associated with the Permian Dullingari gas field. It was drilled in 1979, and prior to setting an intermediate casing string, a reasonable oil show was recorded. This show was tested and subsequent cased-hole testing produced an oil flow of 71 cubic metres per day. Appraisal and development drilling has shown that the oil reservoir is thin (less than 0.6 metres thick) but extensive. Good oil flow rates of up to 317 cubic metres per day have been obtained from this reservoir. The known limits of the field are located within the structural closure of the Dullingari feature, however there is a possibility that the accumulation may be in part stratigraphically controlled (Mount, 1981). Reserves are estimated at 1.7 million cubic metres (Table 1). Cuttapirrie 1 was drilled on a pronounced fault-controlled anticline near the edge of the Patchawarra Trough (Fig. 7). This well, drilled in 1980, was the next significant discovery, intersecting a thin oil column near the base of the Hutton

Sandstone at a stratigraphic level similar to that from which Poolowanna 1 recovered oil in 1977. Moorari 3 in 1980 tested a thin sand within the Birkhead Formation and recovered 62 barrels of oil. In 1981 Marabooka 1, drilled on a simple domal anticline located on the Nappacoongee-Murteree High flowed gas from the Namur Sandstone Member at 0.2 million cubic metres per day, and in the same year Strzelecki 4 confirmed that the Namur Sandstone Member was oil productive in the Strzelecki Field with an oil flow of 111 cubic metres per day. Merrimelia 6 drilled near the structural culmination of the prominent Merrimelia Anticline in 1981 flowed oil at varying rates from the Murta Member (2 cubic metres per day), Namur Sandstone Member (70 cubic metres per day) and Hutton Sandstone (435 cubic metres per day). Previous wells (Merrimelia 1-5) had been located on the flanks of this major anticline because the prospective Permian sediments were considered to be absent from the top of the structure. Jackson 1 also drilled in 1981 on an elongate anticlinal closure is the largest Eromanga Basin oil field discovered to date. It found significant reservoirs in the Westbourne Formation and the Hutton Sandstone, which flowed oil at 214 cubic metres per day and 377 cubic metres per day respectively. Oil-in-place reserves for Jackson are estimated at 7 million cubic metres and for Jackson South 0.8 million cubic metres (Table 1). In 1982, Dullingari 22 discovered a new oil reservoir in the Namur Sandstone Member at the culmination of the large Dullingari anticline. Dullingari 22 intersected 6.7 metres of oil pay and was tested at 317 cubic metres per day. Big Lake 26 intersected a thin reservoir in the Birkhead Formation and

J. D. A R M S T R O N G & T. M . B A R R

flowed at 24 cubic metres per day. Merrimelia 9 also flowed 12 cubic metres per day from this formation. Possibly the most important discovery in 1982 was that of Strzelecki 8 which produced gas at a rate of 9915 cubic metres per day from a thin Cretaceous Coorikiana Sandstone reservoir. This flow is the first petroleum to flow from the marine Cretaceous of the Eromanga Basin and points to the potential of the younger Cretaceous which at the moment is not being as actively explored as the basal Cretaceous and Jurassic strata. In the Queensland sector of the Eromanga Basin, the stratigraphic nomenclature used is shown on Figure 8. The four discoveries made in Queensland are in the Murta Member, Westbourne Formation, Birkhead Formation and Hutton Sandstone. This distribution shows that the Jurassic sequence as a whole is prospective. It is interesting to note that all discoveries have been made along the prominent Innamincka Trend which extends eastwards where it is referred to as the Wackett-Jackson High (Fig. 7). This trend flanks the Nappamerri Trough which has been the focal point for the hydrocarbons source. In the South Australian sector, discoveries and 'shows' are illustrated on Figure 9 and discoveries are again concentrated in the Jurassic section, but spread throughout the stratigraphic column, indicating that this section overall is highly prospective. In order to assess these discoveries we can make some observations which characterise them. 1. All were found as the result of drilling on seismically defined four-way structural closures and, with the possible exception of the Dullingari-Murta Field, subsequent drilling has confirmed that the discoveries are simple structural traps. 2. The discoveries are of small size, with an average area of accumulation less than 4 square kilometres, and average pay thickness less than 6.1 metres (Table 1).

WINTON

1). 6. The gas has a low C 0 2 content (2.7%) and is relatively condensate rich (0.005 cubic metres oil/cubic metre gas). 7. W i t h the exception o f Cuttapirrie, Moorari and Poolowanna, the discoveries to date have flanked the Nappamerri Trough. In light of this evidence some obvious conclusions can be drawn about the type of exploration required to find hydrocarbons in the Eromanga Basin. Four-way dip closed structural features defined by suitable seismic control remain the principal targets for wildcat exploration. The seismic coverage is relatively dense in the South Australian 'Cooper sector' of the Eromanga Basin reflecting the impetus of discovery and the development of underlying Permian gas fields (Fig. 10). In this area 1 km seismic grids have been shot to enable efficient gas development but in general the line density is approximately 2-3 km, so that considering the small size of the targets the seismic coverage is still not adequate to define all the possible structural traps in that part of the Eromanga Basin. In Queensland seismic coverage is most dense in the area corresponding to the Palaeozoic Adavale Basin, with a line density of around 5-10 km. In the Pedirka region of the Eromanga Basin the line density is around 12-15 km and obviously is less than the required density to achieve reasonable definition of the

FORMATION

MACKUNDA

FORMATION

ALLARU

MUDSTONE

TOOLEBUC

FORMATION

WALLUMBILLA

FORMATION

TRANSITION ^

3. The discoveries are only filled to a fraction o f the total closure, around 20 per cent suggesting that either the generated volume of oil is small or that the structural trap is larger now than at the time of migration and original accumulation. 4. The predominant hydrocarbon found in the Eromanga sequence is oil, in a ratio of 4:1 against gas (Table 1). 5. The oil tends to have a high gravity of around 47° A P I , a pour point of 11 °C, a low gas/oil ratio of 17.83 cubic metres gas/cubic metre oil, and a high wax content (Table

BEDS

MURTA

HOORAY WESTBOURNE ADORI

MEMBER

+ Jackson 54m 3 Per Day

SANDSTONE FORMATION

• Jackson 185m3Per Day Jackson South 103m 3 Per Day

SANDSTONE

BIRKHEAD

FORMATION

HUTTON

SANDSTONE

basal

JURASSIC

Fig. 8. Queensland stratigraphy and discoveries.

# Wackett 62 103m3Per Day #

Jackson 377m3 Per Day

EROMANGA BASIN EXPLORATION AND POTENTIAL BLANCHEWATER

FORMATION

OODNADATTA

FORMATION

COORIKIANA

SANDSTONE

BULLDOG

SHALE

TRANSITION

BEDS

MOOGA

fe.

MURTA

MEMBER

Strzelecki 8 103m3PerDay

• Dullingari 72m3 Per Day Merrimelia Namur 394 103m3PerDay Marabooka 198 103mrPerDay Strzelecki 111m3Per Day,Dullingari 317m3Per Day,Merrimelia 70m3Per Day Mc K inlay Moorari,Big Lake Nappacoongee, Strzelecki, Merrimelia

• NAMUR SANDSTONE MEMBER • • FORMATION ^ • BIRKHEAD FORMATION

HUTTON

SANDSTONE

basal

JURASSIC'

• Strzelecki 382m3Per Day,Merrimelia 435m3Per Day

• Cuttapirrie 16-25m3PerDay Poolowanna

Fig. 9. South Australian stratigraphy and discoveries.

structural configuration of possible traps. In general the seismic coverage within the Eromanga Basin is inadequate for an assessment of the potential of the basin. Explorers in the Eromanga Basin also have to address the problem of the small volume of oil present compared with the total volume of trap available. It is suggested that either the volume of oil generated is small, or the timing of generation with respect to structural development, are the controlling factors. It is interesting to note that Pitt (1982) suggests that generation started no earlier than the MidTertiary and concludes therefore that 'young' structures are prospective. This view is somewhat different from that expressed by Cook et al. (1982) that 'the onset of significant hydrocarbon generation followed the rapid deposition of the Cenomanian Winton Formation'. Habermehl (1982) also suggests generation 'since Late Cretaceous time' although Poll (1981) contends that the failure to discover hydrocarbons in very young structures appears to suggest that generation predated the Early Tertiary. The view that generation of petroleum from Jurassic source rocks took place during the Jurassic and Cretaceous, influences exploration strategy for petroleum in the basin. A target must exhibit structural closure prior to Cretaceous deposition.

EXPLORATION—THE FUTURE The Eromanga Basin has contributed to Australia's domestic oil production. In 1983, 0.55 million cubic metres of oil were transported through the Moomba-Stony Point pipeline, produced from the Merrimelia, Dullingari, and Strzelecki Fields. The establishment of this facility should prove to be a major factor in the ongoing exploration of the basin in that new discoveries can be brought to production relatively quickly. This had an impact in 1983 when the SANTOS-Delhi group expended several tens of millions of dollars in both seismic and drilling exploration to search for Jurassic oil. Part of this exploration work was aimed at reevaluating existing structures which are known to be gas productive in the Permian, but have not been evaluated exhaustively in the Eromanga Basin section.

To achieve reasonable success in the Eromanga Basin, it is estimated that 250 geoscientists with 100 support staff are required to define drillable prospects seismically, evaluate the geology, and supervise seismic crews and drilling rigs. Such staff need to be adequately trained with a perception of the required balance between the expansion of knowledge and activities which will directly lead to discovery of petroleum. There are also a number of areas of geological thinking which could be pursued in the ongoing assessment of the Eromanga Basin. An understanding of the distribution of the prospective rock units within the Eromanga Basin appears not to have been completed for the whole sequence. Such understanding will be valuable in the process of ranking prospects. The necessity of correct structural timing has been discussed and this problem, associated with a better understanding of the time of primary oil generation, has not to our knowledge been adequately investigated. Present day structural definition has also been discussed and the necessity of accurate seismic modelling is paramount when the oil columns are generally less than 15 metres thick. It is appropriate to speculate about the potential of the Eromanga Basin. A simple approach is to estimate the likely success upon drilling an assumed suite of prospects. If an undrilled prospect exists every 500 square kilometres over the prospective 50 per cent of the basin, then there could be 1000 prospects yet to be investigated. At 2200 metres per well, their testing would involve 2.3 million metres of drilling. Our experience suggests that a drilling cost including provision for seismic and other costs is some $740/metre. Expenditure on 2.3 million metres would be $1.7 billion (1982 dollars). The significance of this lies in the potential oil discoveries as they might contribute to fill the gap between Australian petroleum consumption and indigenous production (Fig. 11). This gap is estimated to be some 0.5 billion cubic metres between 1982 and 2000, excluding heavy oil imports. Assuming a 1 in 20 discovery rate for the 1000 prospects, and assuming that half of the discoveries are oil with Jackson Field size reserves, then the programme would discover 80 million cubic metres of recoverable oil, or about 16 per cent of the 1982-2000 Australian shortfall. For the purpose of

J. D. ARMSTRONG & T. M. BARR

QUEENSLAND NEW SOUTH WALES

Km 100

140°

SCALE

100

200Km

148°

Fig. 10. Seismic coverage—Eromanga Basin.

constructing Figure 11, it was assumed that about 1.5 Jacksonsized fields are discovered each year in the Eromanga Basin from 1984 until 1999 and that each field is on stream 24 months after discovery. On Figure 11, Curve A represents a projection of indigenous production without the known Eromanga Basin fields which have come on stream in 1983-1984. Curve B incorporates those Jurassic fields. Curve C represents the inclusion of the hypothetical Eromanga fields as outlined above. Thus, in 1990, when the shortfall is some 16 million cubic metres, it is estimated that the Eromanga Basin could contribute 3 million cubic metres to reduce this volume.

On a broader scale it is evident that to fully satisfy the 1982-2000 gap, there is a need for five additional increments each the size of the hypothetical Eromanga Programme outlined, These are shown on Figure 11, numbered 1 to 5 and are required to be available for production in a staggered fashion from 1986. While this is not expected at this time to come from the Eromanga Basin, it does demonstrate the scope of the effort required. It can therefore be demonstrated that the Eromanga Basin will have an impact on the national petroleum scene, and this impact will be dependent on the future exploration efforts of the companies and their exploration staff.

EROMANGA BASIN EXPLORATION AND POTENTIAL

1960

1970

1990

1980

2000

YEAR ENDING JUNE Fig. 11. Australian oil demand and production forecast.

REFERENCES AMBROSE, G . , SUTTILL, R . & LAVERING, I., 1982: A review o f t h e

in the Early Jurassic of the Eromanga Basin. APEA J. 21(1), 60-70. DENMEAD, A. L., 1961: Occurrence of petroleum and natural gas in Queensland. Qld. Geol. Surv. Publ. 299. HABERMEHL, M. A., 1982: Aspects of regional groundwater movement and hydrocarbon migration in the Eromanga Basin: in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide. KANTSLER, A . J., C O O K , A . C . & ZWIGULIS, M . , 1982: M a t u r a t i o n

patterns in the Eromanga Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide. MCINTYRE, S., 1980: The hydrocarbon potential of the Jurassic and Lower Cretaceous sediments from five wells in the Eromanga Basin, South Australia. B.Sc. (Hons) thesis, Univ. Adelaide (unpubl.). MORIARTY, K . C . & W I L L I A M S , A . F., 1982: H y d r o c a r b o n f l u s h i n g

in the Eromanga Basin—fact or fallacy? in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide. MOTT, W. D., 1952: Oil in Queensland. Qld Gov. Min. J. 53, 848-61. MOUNT, T. J., 1981: Dullingari North 1, An oil discovery in the Murta Member of the Eromanga Basin. APEA J. 21(1), 71-7.

MUDGE, W. J., 1980: The hydrocarbon potential and environmental interpretations of the Jurassic-Lower Cretaceous section in five wells in the Eromanga Basin. B.Sc. (Hons) thesis, Univ. Adelaide (unpubl.). PITT, G. M., 1982: Geothermal gradients in the Eromanga BasinCooper Basin region; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide. POLL, J. K., 1981: The significance of the southwest Eromanga Basin oil and gas discoveries. APEA J. 21(1), 33-8. SENIOR, B. R . , M O N D , A . & HARRISON, P. L . , 1978: G e o l o g y o f t h e

Eromanga Basin. Aust. Bur. Miner. Resour. Geol. Geophys. Bull. 169. SPRIGG, R. C., 1958a: A new look at the Great Artesian Basin. Aust. Oil Gas J. 5, 13-7. SPRIGG, R. C., 19586: Petroleum prospects of western parts of Great Australian Artesian Basin. Am. Assoc. Pet. Geol. Bull. 42, 2465-91. SPRIGG, R. C., 1965: Progress of exploration for petroleum in the central and western Great Artesian Basin. Commonw. Min. Mettall. Congr. Proc. 5, 167-77. SPRIGG, R. C., 1982: The long lead-up to commercial oil discovery in the Mesozoic Eromanga Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide. SPRIGG, R. C., 1983: The long lead-up to commercial oil discovery in the Mesozoic Eromanga Basin. Pet. Explor. Soc. Aust. J. 2, 8-18.

J. D. ARMSTRONG & T. M. BARR

E. R., 1977: Australia's onshore basins: their petroleum prospects and future exploration programmes. A PEA J., 16(1), 17-23.

SMITH,

T H O M A S , B. M., 1982: Land-plant source rocks for oil and their

significance in Australian basins. A PEA J. 22(1), 164-76.

W H I T E H O U S E , F. W . , 1954: The geology of the Queensland portion

of the Great Artesian Basin. Appendix 4 in Artesian Water supplies—Queensland. Dept. of the Coordinator General of Public Works Queensland, Report A.56, 1955.

WILTSHIRE, M . J., 1978: Poolowanna

Delhi Pet. Pty Ltd (unpubl.).

WILTSHIRE, M . J., 1979: Eromanga

No. 1, well completion report.

Basin, Australia, petroleum prospects. Aust. Mineral Foundation, Adelaide {unpubl.). M. J., 1982: Revision of Eromanga Basin limits; in Moore, R S. & Mount, T. J. (compilers) Eromanga Basin Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide. W I L L I A M S , G. D., 1966: The Great Artesian Basin—origin and history. APEA J. 6(1), 88-92. WILTSHIRE,

Geological Society of Australia Special Publication No. 12, 39-51

Jurassic and Triassic stratigraphy and hydrocarbon potential of the Poolowanna Trough (Simpson Desert region) northern South Australia P. S. Moore

Delhi Petroleum Pty Ltd, 101 Grenfell Street, Adelaide, S.A. 5000. ABSTRACT

The Simpson Desert covers nearly 100 000 km in central Australia, straddling the borders between South Australia, the Northern Territory and Queensland. It roughly coincides with a major subsurface sedimentary depression known as the Poolowanna Trough, which accumulated sediments during the Palaeozoic, Mesozoic and Cainozoic. Triassic sediments in the Poolowanna Trough are assigned to the Simpson Desert Basin. They are separated from underlying Palaeozoic sediments by a major unconformity representing at least all of the Late Permian. Two Triassic formations are recognised. The older Walkandi Formation is a shaly and silty redbed sequence; its type section is defined in Poolowanna 1, from 2768 m to 2902 m. The conformably overlying Peera Peera Formation, of Middle to Late Triassic age, is a sequence of interbedded grey shale, siltstone and fine-grained sandstone of lacustrine and low-energy, meandering-fluvial origin. A disconformity representing the Latest Triassic and earliest Jurassic separates the Peera Peera Formation from Eromanga Basin units, the oldest of which is the Poolowanna Formation, of Early to Middle Jurassic age (J1 to J4 spore-pollen zones). The Poolowanna Formation consists of sandstone with common interbeds of grey siltstone, shale and coal. The type section occurs between 2387 m and 2593 m in Poolowanna 1. A meandering-fluvial environment of deposition is suggested, based on the lithological association and the presence of fining-upward cycles. The conformably overlying and, in part, laterally equivalent Algebuckina Sandstone is a thick sequence dominated by fine to coarse-grained, porous sandstone of braided-fluviatile origin. With only 20 petroleum exploration wells drilled in the Simpson Desert region to date, the area remains largely unexplored. Future exploration will probably be difficult and expensive compared with that in the Cooper Basin region to the east. However, the Poolowanna Trough has considerable potential for small discoveries, based on the combination of source-rock quality, maturity and reservoir quality. The Poolowanna Formation is the main exploration target; it is the most important source-rock horizon in the region, containing up to 15 per cent total organic carbon, with exinite a minor but significant component. The formation should be early mature in the Poolowanna Trough and is believed to have generated the viscous, high wax oil found in Poolowanna 1. Important secondary targets for hydrocarbon exploration in the Poolowanna Trough are the Peera Peera Formation and Algebuckina Sandstone. 2

INTRODUCTION

The Poolowanna Trough is a major subsurface sedimentary depression located principally in the Simpson Desert region of South Australia, but extending into the Northern Territory and southwestern Queensland (Figs 1, 2). It contains sediments of Late Palaeozoic, Mesozoic and Cainozoic age, with nearly 3000 m of sediment preserved in the deepest, central portion of the trough. This paper reviews the Triassic and Jurassic stratigraphy of the Poolowanna Trough, and comments on the hydrocarbon potential of the area.

REGIONAL STRATIGRAPHIC NOMENCLATURE

Stratigraphic nomenclature for the Late Palaeozoic and Mesozoic of the Poolowanna Trough is shown in Figures 3

and 4. Within the trough, portions of four sedimentary basins are superimposed (Fig. 1): (a) Pedirka Basin (Permo-Carboniferous); (b) Simpson Desert Basin (Triassic); (c) Eromanga Basin (Jurassic-Cretaceous); and (d) Lake Eyre Basin (Tertiary). The Pedirka Basin is restricted to the western portion of the Poolowanna Trough and extends northwestwards from there; it contains the Crown Point and Purni Formations, which are latest Carboniferous to Early Permian in age (Youngs, 1975a, 19756; Cooper, 1981). The Simpson Desert Basin (Smyth & Saxby, 1981; Moore, 1982) contains the Walkandi and Peera Peera Formations and is of Triassic age. It is separated from the underlying Pedirka Basin by an unconformity spanning the late Early Permian

P. S. MOORE

134°

138

136°

V 4 /

£ i '

N.T. PEDIRKA BASIN A

-26-

140°

v»ii

MD // • '

QLD. / /

"'Nv \ s »

Tht

- / - - - r — q ^ r > 0

APO Ku

SIMPSON

AD • Birdsville

- - " # Cu

COOPER BASIN

Pn -28"

•Moomba Ranges

- - - ' ' I — 100

—J kms

z in

Fig. 1. Location map, Poolowanna Trough. The trough contains sediments of Permo-Carboniferous (Pedirka Basin), Triassic (Simpson Desert Basin), Jurassic-Cretaceous (Eromanga Basin) and Tertiary (Lake Eyre Basin) age. Well abbreviations, in alphabetical order: AD = Adria Downs 1, Co = Colson 1, Cu = Cuttapirrie 1, Er = Erabena 1, HR = Hale River, Ku = Kuncherinna 1, Ma = Macumba 1, MC = Mt Crispe 1, MD = McDills 1, Mo = Mokari 1, Oo = Oodnadatta 1, P n ^ P o o n a r u n n a 1, Po = Poolowanna 1, Pu = Purni 1, Th=Thomas 1, Wa=Walkandi 1, We=Weedina 1. Line of cross-section, Witcherrie 1 to Pandieburra 1, shown in Figure 2.

and Late Permian. Whereas the Walkandi Formation is confined to the central, deepest portion of the Poolowanna Trough (Fig. 4), the overlying Peera Peera Formation extends eastwards onto the Birdsville Track Ridge; the preserved basal portion passing laterally into the upper part of the Nappamerri Formation of the Cooper Basin (Youngs & Boothby, 1982). A relatively short break representing the latest Triassic and earliest Jurassic separates the Simpson Desert Basin from the overlying Eromanga Basin (Fig. 3). Jurassic sediments of the Eromanga Basin are mostly braided fluviatile sandstones, however several facies changes occur from west to east across the area. Where it crops out around the southwestern basin margins, the Jurassic is represented by the high-energy, braided-fluviatile Algebuckina Sandstone (Wopfner et al., 1970). Basinwards, the lowermost Algebuckina Sandstone becomes shaly and coal bearing and passes laterally into the Poolowanna Formation, which is essentially confined to the Poolowanna Trough. Further east, the Algebuckina Sandstone becomes shaly and silty in two other, higher intervals which in the region of the Birdsville Track Ridge (Fig. 4) are recognised as the Westbourne and Birkhead Formations. Thus, major facies changes, necessitating changes in stratigraphic nomenclature, are recognised across the area.

The Jurassic-Cretaceous boundary lies within the upper part of the Algebuckina Sandstone in the subsurface (Moore et al., 1986) with the overlying Cadna-owie Formation being Neocomian in age (Fig. 3). The Aptian to Cenomanian sequence above the Cadna-owie Formation, which shows only minor lateral facies variations, consists of marine mudstone (Wallumbilla Formation, Toolebuc Formation and Allaru Mudstone) overlain by a paralic sandy unit (Mackunda Formation) and capped by the coally, fluviatile and paludal Winton Formation (Moore & Pitt, 1985). Mesozoic and older units of the Simpson Desert region are typically overlain by up to 200 m of Tertiary, Lake Eyre Basin strata, which in turn are capped by a veneer of Recent aeolian sandstones. The complex stratigraphy of the Lake Eyre Basin has been discussed in detail by Wopfner & Twidale (1967).

WALKANDI FORMATION Definition The Walkandi Formation (Moore, 1982) is the lower of two units defined within the Triassic Simpson Desert Basin (Figs 3, 4). It is recognised in Erabena 1, Poolowanna 1 and Walkandi 1 in South Australia and also occurs in Thomas 1

POOLOWANNA TROUGH STRATIGRAPHY

CONTOUR INTERVAL OS^SECS

TIME STRUCTURE i

Fig. 2. 'Z' horizon (top pre-Permian basement) and *C' horizon (top Cadna-owie Formation) time structure maps, Poolowanna Trough and adjacent areas, South Australia. Heavy lines are faults.

in the Northern Territory. The formation was previously termed 'unnamed redbeds' by Wiltshire (1978a, 19786). The type section extends from 2768 m to 2902 m in Poolowanna 1 and corresponds to Wiltshire's (1978a) original concept of the unit (Fig. 5).

Base of the formation Structural and stratigraphic relationships between the Walkandi Formation and underlying units are still only partly understood. In South Australia, the Walkandi Formation commonly overlies steeply dipping, black, pyritic shales of

P. S. MOORE

EUROPEAN STAGES

AGE

BIOSTRATIGRAPHIC UNITS

POOLOWANNA TROUGH STRATIGRAPHY

BASIN NOMENCLATURE

SENONIAN

CAMPANIAN SANTONIAN CONIACIAN TURONIAN C striatus

CENOMANIAN ALBIAN

Subzone

R pannosus

Zone

C. hughesi

Subzone

WALLUMBILLA

FORMATION

HAUTERIVIAN

wont

haggiensis

CADNA-OWIE

Su b z o n e

VALANGINIAN BERRIASIAN

australiensis

upper

BASIN

BARREMIAN NEOCOMIAN

EARLY

CRETACEOUS

LATE

MAASTRICHTIAN

LATE Middle EARLY

PLIENSBACHIAN

BAJOCIAN

a-b

J4 J 2-3 POOLOWANNA

HETTANGIAN

LATE

Basal Bundamba Assemblage

NORIAN

MID.

LADINIAN

Ipswich Asse mblage

PEERA PEERA

M o o l a y e m ber Assemblage

Tr-3

Tr 3

a-b

FORMATION

c-d

ANISIAN SCYTHIAN

LATE

b J- 1

Tr 2 Tr 1 TATAR I AN

KAZANIAN

P O O L O W A N N A

SINEMURIAN

CARNIAN

TRIASSIC

UPPER J 5 - 6

BATHONIAN

RHAETIAN

UPPER STAGE 5

KUNGURIAN

EARLY

PERMIAN

SANDSTONE

SIMPSON DESERT BASIN

JURASSIC

CALLOVIAN

ALGEBUCKINA

OXFORDIAN

TROUGH

lower

TITHONIAN KIMMERIDGIAN

EROMANGA

S u bzone

ARTINSKIAN

LOWER STAGE 5 STAGE 4

SAKMARIAN

STAGE 3

PURNI FM.

STAGE 2

CROWN

PT. F M .

PEDIRKA BASIN

STAGE 1 Fig. 3. Biostratigraphy, European stage correlations and basin nomenclature of the Poolowanna Trough, based on spore-pollen analysis of M a c u m b a 1, Walkandi 1 and Poolowanna 1.

POOLOWANNA TROUGH STRATIGRAPHY

possible Ordovician age. However, in Erabena 1, the formation is separated from deformed Ordovician basement by a thin sequence of silicified redbeds, tentatively assigned to the Devonian Finke Group. This Devonian sequence, although apparently flat-lying in Erabena 1, has a dip of 25 0 in nearby Macumba 1 suggesting possible deformation during the Alice Springs Orogeny. In Thomas 1 in the Northern Territory, the Walkandi Formation is separated from steeply dipping basement rocks by 40 m of grey-green shale, siltstone and sandstone of unknown age (Wiltshire, 1982A).

Top of the

formation

The Walkandi Formation is overlain conformably by the Peera Peera Formation, with the top of the Walkandi Formation taken to be the highest stratigraphic occurrence of red shale and siltstone (Fig. 5). Based on drilling and seismic evidence, the upward transition from oxidised to reduced sediments occurs at a similar seismic horizon, and thus presumably stratigraphic position, across the basin.

Age The only reliable ages presently recorded from the Walkandi Formation are derived from sidewall-core samples at 2844 m and 2851 m in Poolowanna 1. These yielded poorly-preserved, impoverished palynofloras consisting mostly of gymnospermous pollen {Falcisporites sp) which could not be more accurately dated than Triassic in age (Price, 1978a). While 1 the sequence is lithologically similar to parts of the Nappamerri Formation in the Cooper Basin (Papalia, 1969), the age relationship between the two units is unclear. In

addition, the two formations are separated by a 150-200 km wide zone of non-deposition and erosion across the Birdsville Track Ridge.

Lithology and environment of

deposition

The Walkandi Formation consists of interbedded shale, siltstone and minor sandstone. The finer-grained lithologies are variable in colour, with pale-grey, grey-green, maroon, brown and brick red varieties recorded. Sandstone interbeds are pale grey-green, very fine to fine-grained and well sorted with a silica cement. Porosity and permeability in the interbeds are typically low. A basal sandy unit in Poolowanna 1, extending from 2866 m to 2902 m, is absent from all other wells. Only one core has been cut in the Walkandi Formation to date, from the type section at Poolowanna 1, between 2797 m and 2800 m. In this core, the siltstone is dominantly green, and exhibits poorly developed bedding. Finer, shaly lithologies are red and display shrinkage features interpreted as desiccation cracks (Fig. 6). The combination of colour mottling, lack of bedding and extremely poor sorting in some of the redbeds suggests that the rocks were subjected to pedogenesis. A shallow, ephemeral, lacustrine environment is envisaged for the Walkandi Formation, with deposition occurring slowly, under fairly stable tectonic conditions. Similar sediments are present in the Nappamerri Formation of the Cooper Basin (Papalia, 1969; Youngs & Boothby, 1982) and elsewhere in the Lower to Middle Triassic in Australia suggesting that the climate was conducive to redbed formation.

P. S. MOORE

DRILLERS DEPTH (METRES)

Z (0 < CD

^ i k - e

t! i aS

2 i 11

1 z

m Z O) Sujjii

FORMATION

t 3

ALGEBUCKINA MIDDLE JURASSIC EARLY

2500

JURASSIC

2400

EROMANGA BASIN

SANDSTONE 2387

POOLOWANNA FORMATION

2900 2902

LATE TRIASSIC MIDDLE TRIASSIC

2800

7EARLY TRIASSIC

2768

BASIN 2700

SIMPSON DESERT

2593 2600

ORD

PEERA PEERA FORMATION

WALKANDI FORMATION

BASEMENT

Fig. 5. Type sections, Walkandi Formation, Peera Peera Formation and Poolowanna Formation, P o o l o w a n n a 1.

POOLOWANNA TROUGH STRATIGRAPHY

Top of the formation

The Peera Peera Formation is overlain unconformably by the Poolowanna Formation. Somewhat different palynological age determinations at the top of the Peera Peera Formation between Poolowanna 1 and Macumba 1 (Price, 1978a, 19786) suggest that erosion of the sequence may have occurred prior to the onset of Jurassic deposition. However, due to limited data and the lack of characteristic marker horizons within the Peera Peera Formation, this apparently variable degree of erosion is difficult to confirm from seismic and drilling results.

Age

The youngest age so far recorded from the Peera Peera Formation was obtained from a sidewall core at 2292 m in Macumba 1,1.2 m below the top of the formation. According to Price (1978a, p. 10), the plant microfossil assemblage has basal Bundamba affinities and indicates that the sampled horizon is 'Late Triassic in age no older than the upper part of the Ipswich Coal Measures . . . The sampled horizon is equivalent to the basal part of the Woogaroo Sub-group which de Jersey (1971) regards as being Rhaetian in age'. By comparison, the uppermost part of the Peera Peera Formation in Poolowanna 1 appears to contain a basal Bundamba or lower Ipswich palynoflora, and thus may be slightly older (Fig. 3). The oldest definitive date presently recorded from the Peera Peera Formation was obtained from a sidewall core at 2425 m, in the lower part of the formation, in Macumba 1. The sample, dated as Middle Triassic by Price (19786), contains a Moolayember palynoflora. Similarly, much of the Peera Peera Fig. 6. Variegated, shaly Walkandi Formation at 2799 m in Formation in Poolowanna 1 is broadly assignable to unit Tr3 Poolowanna 1. Dark areas are red, due to hematite cement. (Price, 1978a, Fig. 3). Pale areas are green. Note poor sorting, very poor visual porosity and shale-infilled, anastomosing, subvertical fractures interpreted as due to subaerial desiccation associated with pedogenesis. Core diameter = 10 cm.

PEERA PEERA FORMATION Definition

The Peera Peera Formation defined herein, is the upper unit within the Triassic Simpson Desert Basin (Figs 3,4). It is welldeveloped in Erabena 1, Macumba 1, Poolowanna 1, Walkandi 1 and Kuncherinna 1 in South Australia, and extends east across the Birdsville Track Ridge, occurring as a thin unit in Pandieburra 1. The formation has also been intersected in Colson 1 and Thomas 1 in the Northern Territory (Wiltshire, 1978c, 1982a). The Peera Peera Formation was originally termed 'Peera Peera beds' by Wiltshire (1978a), however subsequent drilling has enabled more complete definition of the unit. The type section is here designated as occurring between 2593 m and 2768 m in Poolowanna 1 and corresponds with Wiltshire's (1978a) original concept of the unit (Fig. 5).

Lithology and environment of deposition

In Poolowanna 1 and Walkandi 1 in the central, deepest portion of the Poolowanna Trough, a three-fold division of the Peera Peera Formation is recognised. The basal unit consists of grey shale and siltstone, with minor thin sandstones and rare coal. The middle unit is sandy and displays common fining-upward cycles (Fig. 7). The upper unit is characterised by black, silty, highly carbonaceous shale. Sandstone throughout the formation is typically fine- to very finegrained, with low porosity and permeability. The presence of

Base of the formation

In the centre of the Simpson Desert Basin, the Peera Peera Formation rests conformably on the underlying Walkandi Formation. Further west, in Macumba 1 and Colson 1, the Peera Peera Formation rests unconformably on the Early Permian Purni Formation. To the east, on the Birdsville Track Ridge, available data suggest that Peera Peera Formation rests directly on steeply dipping ?Ordovician sediments.

Fig. 7. Abundant intraclasts of coal and carbonaceous shale in finegrained sandstone of the Peera Peera Formation at 2348 m in Macumba 1. The sandstone occurs near the base of a welldeveloped fining-upward sequence. Core diameter = 10 cm.

P. S. M O O R E

coal and fining-upward cycles in the Peera Peera Formation, combined with an absence of marine microfossils suggests that deposition occurred on the floodplain of a meandering fluvial system. Dark shales may represent lacustrine sedimentation.

POOLOWANNA FORMATION Definition The Poolowanna Formation is the lowermost unit of the E r o m a n g a Basin in the Simpson Desert region (Figs 3, 4). It is recognised throughout the area, extending as far west as M a c u m b a 1 and Colson 1, and as far east as the Birdsville Track Ridge. The Poolowanna Formation was originally termed 'Poolowanna beds' by Wiltshire (1978tf), due to uncertainty regarding the age and lateral extent of the unit. Subsequently, Moore (1982) divided the unit into an upper and a lower part based on palynological evidence. However, new palynological evidence suggests that this subdivision is unnecessary (Price, pers. comm., 1982). The type section is designated as occurring between 2387 m and 2593 m in Poolowanna 1 and corresponds with Wiltshire's (1978tf) original concept of the unit (Fig. 5).

Base of the formation In all areas where the Poolowanna Formation has been encountered, it rests u n c o n f o r m a b l y on the Peera Peera Formation, as discussed above. The basal portion of the sequence is typically sandy and the acoustic impedance change across the unconformity is represented by a seismic reflection of moderate strength.

Top of the formation In most areas, the Poolowanna Formation is overlain with apparent conformity by the Algebuckina Sandstone. However, on the Birdsville Track Ridge the Poolowanna Formation is overlain c o n f o r m a b l y by the H u t t o n Sandstone (Fig. 4). The top of the formation is taken at the top of the stratigraphically highest, m a j o r shaly unit beneath the Algebuckina Sandstone (Fig. 5).

Age Originally it was suspected that a m a j o r hiatus existed within the Poolowanna Formation, with the lower half being assigned to the J1-J3 spore-pollen zones and the upper half to a 'lower J 5 - 6 ' zone (Moore, 1982; Price, 1978tf, 19786). However, it is now recognised that p a l y n o m o r p h s which constituted the 'lower J 5 - 6 ' zone represent a distinctive

palynofacies within the J2-4 interval (Price, pers.

comm.

1982). Thus, the Poolowanna Formation is now considered to range in age from Early to Middle Jurassic (J1-J4 palynological zones). This change is very important, since the previous interpretation had m a j o r consequences regarding structural evolution of the Birdsville Track Ridge and stratigraphic entrapment of hydrocarbons on the flanks of the ridge, as Wiltshire (19826) emphasised.

Lithology and environment of deposition T h e Poolowanna Formation consists of interbedded sandstone, siltstone, shale and coal. Sandstone interbeds are predominantly fine-grained and in most cases easily distinguished from the overlying, coarser grained Algebuckina Sandstone. Shale and siltstone interbeds are mid to dark grey and are typically evenly laminated. Coal seams form a minor component of the sequence and are generally less than 0.5 m

thick. Fining-upward cycles are c o m m o n , and the formation is interpreted as of meandering or anastomosing fluvial origin, with minor associated floodplain sediments. The Poolowanna Formation is distinguished lithologically f r o m underlying Triassic sediments by a higher sand-shale ratio, and presumably represents a higher energy environment of deposition. To the west, approaching the outcropping margins of the E r o m a n g a Basin, the Poolowanna Formation becomes very sandy and passes laterally into the basal Algebuckina Sandstone. This basinward transition from presumed braided fluviatile sediments into the Poolowanna Formation further supports an interpretation of moderate to high-energy (meandering or anastomosing-channel) fluvial deposition for the Poolowanna Formation. Deposition is suggested to have occurred on a moderate palaeoslope with only limited development of floodplain sediments. T h e nature and extent of the Poolowanna Formation east of the Poolowanna Trough are poorly known. In the vicinity of the Birdsville Track Ridge, wells are widely spaced and it is difficult to ascertain whether the Poolowanna Formation pinches out or passes laterally into more sandy sediments of the H u t t o n Sandstone. East of the Birdsville Track Ridge, a comparable facies is developed sporadically at the base of the Jurassic E r o m a n g a Basin sequence. O n e such local development is represented by the oil-bearing 'basal Hutton Sandstone' in Cuttapirrie 1 (Barr & Youngs, 1981).

ALGEBUCKINA SANDSTONE Definition The Algebuckina Sandstone (Wopfner et at., 1970) is a thick, sandy sequence of Early Jurassic to Early Cretaceous age. It crops out a r o u n d the southwestern margin of the E r o m a n g a Basin and extends into the subsurface as far east as the Birdsville Track Ridge. In the N o r t h e r n Territory, the equivalent sandy unit in outcrop is the De Souza Sandstone (Wells et al., 1970). T h e type section of the Algebuckina Sandstone is located 0.8 km southwest of the disused Algebuckina railway siding (Wopfner et al., 1970, fig. 1). Here, the f o r m a t i o n is quite thin, incomplete, and ill-defined with respect to stratigraphic range.

Base of the formation In outcrop, the Algebuckina Sandstone rests unconformably on Precambrian and Palaeozoic formations. Passing eastwards into the subsurface, towards the centre of the Poolowanna Trough, the Algebuckina Sandstone rests on progressively younger sediments. For example, in Mt Crispe 1 and Witcherrie 1, the f o r m a t i o n rests u n c o n f o r m a b l y on the Crown Point Formation, whereas in P u r n i 1 and Mokari 1, it is underlain by the Purni F o r m a t i o n (Fig. 4). In Colson 1, M a c u m b a 1 and wells east thereof, the Algebuckina S a n d s t o n e rests c o n f o r m a b l y on the Early Jurassic Poolowanna Formation.

Top of the formation In outcrop a r o u n d the southwestern margins of the Eromanga Basin, the Algebuckina Sandstone is overlain, often disconformably, by the Cadna-owie F o r m a t i o n (Wopfner et al., 1970). Ambrose (1980) and A m b r o s e & Flint (1982) have shown that in a few localities near the Peake and Denison Ranges, the contact with the overlying Cadna-owie Formation is a silicified, leached unconformity surface. This relationship has not been identified in the m u c h thicker and presumably more complete subsurface sections. In the Simpson Desert

47 POOLOWANNA TROUGH STRATIGRAPHY region, the top of the Algebuckina Sandstone is taken at the HYDROCARBON POTENTIAL top of the last major sandstone unit; this is overlain with apparent conformity by grey shales and siltstones of the basal Walkandi Formation Cadna-owie Formation. The hydrocarbon potential of the Walkandi Formation is considered to be low, mainly due to the oxidised nature of Age the sediments. Coals are extremely rare, and dispersed organic In outcrop, the Algebuckina Sandstone has been dated as matter (DOM) is sparse, being mainly represented by inertinite Late Jurassic to possibly earliest Cretaceous (Harris, 1970; (Smyth & Saxby, 1981). Furthermore, the sequence is Wopfner et al., 1970). However, in the subsurface a more predominantly fine-grained with sandstone interbeds complete sequence is preserved. For example, in Witcherrie 1, displaying low porosity and permeability. Purni 1 and Mokari 1, the Algebuckina Sandstone rests With only three wells having penetrated the Walkandi unconformably on Pedirka Basin sediments and has an age Formation to date, very little is known about the lateral range of earliest Jurassic (J1 or J2-3 spore-pollen zones) to variability of the unit, so it is possible that more prospective Early Cretaceous. Further east, the basal Algebuckina sequences may still be encountered. The formation is predicted Sandstone becomes increasingly shaly and silty, and finally to be mature for oil generation (Ro max > 0.7 per cent) passes into the Poolowanna Formation. Thus, in the central throughout its area of distribution (Fig. 9), but due to the Poolowanna Trough, the Algebuckina Sandstone has an age type of preserved organic matter, would most probably yield range from Middle Jurassic (J4) to Early Cretaceous (Fig. 3). gas. The exact position of the Jurassic-Cretaceous boundary within the Algebuckina Sandstone is poorly defined, due to the sandy nature of the sequence, the absence of dinoflagellates and Peera Peera Formation The Peera Peera Formation has fair hydrocarbon potential. the broad range of the Cicatricosisporites australiensis sporeThe sequence is dark and rich in organic matter (Fig. 10) with pollen Subzone (Moore & Pitt, 1985). a total organic carbon (TOC) content as high as 5 per cent. Dispersed organic matter (DOM) increases up the section from Lithology and environment of deposition 1 per cent near the base to 4 per cent at the top (Smyth The Algebuckina Sandstone is a thick unit of fine- to &about 1981). The DOM is largely cutinite, which, according coarse-grained, poorly to moderately sorted sandstone. Shale to Saxby, (1982, table 5), has its main oil generation phase in and siltstone interbeds are uncommon, thin and laterally theCook range Ro max = 0.7-0.9 per cent. discontinuous. Although little is known about the Vitrinite reflectance data from the Peera Peera Formation sedimentology of the Algebuckina Sandstone 'in the Poolowanna Trough, a single core near the base of the (Kantsler et al., 1983; Fig. 9) suggest that the unit should be formation in Poolowanna 1 exhibited abundant, large-scale oil mature. Indeed, fluorescence has been encountered in most cross-stratification (Fig. 8). This is consistent with outcrops wells penetrating the Peera Peera Formation. A drillstem test (Wopfner, 1964; Wopfner et al., 1970) and with core data (DST 3) in Poolowanna 1 (Fig. 5) yielded gas at a rate too obtained from shallow stratigraphic drilling by the South small to measure and was accompanied by a very small Australian Geological Survey (Freytag, 1966; Thornton, 1974; amount of oil which was later extracted from the drilling mud Griffiths, 1980; Northcott, 1980). The absence of a marine (Wiltshire, 1978a). The oil is a light, mature variety (42° API fauna, combined with the fairly unidirectional, cross-stratified gravity; - l ° C pour point) which McKirdy (1981) interpreted character of the sandstones suggest a fluviatile origin. The as derived from woody-herbaceous DOM. However, the Peera paucity of shale interbeds and their poor lateral continuity Peera Formation is very rich in inertinite (Cook, 1982; Smyth 1981) and is considered to contain primarily gaswhere present suggest a braided fluvial origin. Based on shale- & Saxby, source rocks with a modest oil yield (Figs 10, 11). sandstone ratios within the Algebuckina Sandstone and prone Moreover, the formation contains few porous and permeable laterally equivalent units, it appears that sediment was being sandstone units of reservoir quality, and is laterally very supplied from both the Gawler Craton and the Amadeus variable. Intensive exploration may well be needed in this Basin regions. sequence before a commercial discovery is made.

Poolowanna Formation

Fig. 8. Medium- to coarse-grained, porous, cross-stratified Algebuckina Sandstone at 2177 m in Poolowanna 1. Most of the smaller, dark spots represent intergranular porosity. Core diameter = 10 cm.

The Poolowanna Formation contains the richest known source rocks in the Simpson Desert region. It is over 200 m thick in the centre of the Poolowanna Trough (Figs 4, 5) and contains up to 15 per cent TOC (Fig. 10). Thin coal seams are common in the sequence, and supplement the dispersed organic matter present in the shales. Source rock characteristics of the Poolowanna Formation have been described by Cook (1982, this volume). The organic facies is dominated by inertinite, which is particularly abundant in the DOM, although coals are generally richer in exinite. The dominant exinite type in the coals is cutinite, whereas sporinite is dominant in the DOM. According to Cook (1982, table 5) sporinite and cutinite have their main oil generation in the range Ro max = 0.7-0.9 per cent. This suggests that significant oil generation is likely to have occurred in the central, deeper parts of the Simpson Desert Basin, with limited generation toward the margins (Fig. 9).

P. S. MOORE

Fig. 9. Vitrinite-reflectance profiles, Poolowanna Trough and adjacent areas. Line of cross-section shown in Figure 1. Vitrinite levels known only at wellsites; profile between wells is interpretive, based on timing and magnitude of structural growth.

Since exinite is concentrated in coals in the Poolowanna Formation, an important consideration is whether the coals can yield liquid hydrocarbons. While it has been commonly stated that coals do not generate oil because of their high internal surface area and high sorptive capacity, Cook (1982) has provided arguments refuting this claim. Furthermore, source rock analyses (Fig. 10) suggest that the Poolowanna Formation organic matter has a moderate C 1 5 + hydrocarbon yield on extraction and has therefore a modest oil yield. Kerogen analyses (Fig. 11) also point to gas-prone Type III source material, with some Type II influence (cf. Tissot & Welte, 1978). The best evidence for oil generation is provided by the results of drilling, with many of the wells in the Poolowanna Trough displaying hydrocarbon fluorescence throughout the formation. In Poolowanna 1, DST 2 tested several sandstone units in the middle part of the formation (Fig. 5) and recovered 1675 m of fluid, which consisted of roughly equal amounts of oil and water. A subsequent cased-hole test (DST 5) flowed oil to surface at the rate of 4 barrels per hour, although after a few hours the oil became mixed with water. The oil was a waxy, paraffinic crude (37° API gravity; 41 °C pour point) which solidified as it reached the surface. According to McKirdy (1981), the oil is a mature variety which has probably been subjected to removal of low molecular weight hydrocarbons, including gas, by water washing. McKirdy (1981) has argued that the oil source was from within the Poolowanna Formation, not from the underlying Peera Peera Formation, as Smyth & Saxby (1981) suggested. The reservoir properties of the Poolowanna Formation are variable, with the majority of sandstone beds in the sequence being fine- to very fine-grained (Fig. 12) with low porosity and permeability. Silica cementation is a particular problem in the deeper parts of the Poolowanna Trough, as exhibited by the very low flow rates recorded from drill stem tests in

Fig. 10. Source-rock-richness plot, Jurassic and Triassic sediments from three wells in the Poolowanna Trough.

Poolowanna 1. However, the sequence becomes sandier, less silicified and coarser grained towards the shallower basin margins, so it is likely that a situation can be found which

POOLOWANNA T R O U G H S T R A T I G R A P H Y

1-5 ^

X I

100 1 o 5

05-

POOLOWANNA FORMATION 0-05

0-10 0-15 ATOMIC

0 20 %

0-25

0-30

Fig. 12. C o m p a r i s o n of sandstones from Poolowanna Formation and Algebuckina Sandstone. Left: well-sorted, very fine-grained, low porosity, silica-cemented s a n d s t o n e f r o m P o o l o w a n n a Formation at 2569 m in Poolowanna 1. Right: Poorly, sorted, fine- to coarse-grained, p o r o u s A l g e b u c k i n a S a n d s t o n e at 2365 m in P o o l o w a n n a 1. C o r e diameter = 10 cm.

potential. However, despite this, the Algebuckina Sandstone has so far failed to yield any sign of hydrocarbons. / I

If Moriarty & Williams (1982) are correct, it is unlikely that hydrocarbons have been flushed from the area by artesian flow. Explanations of the lack of hydrocarbon shows include the possibilities that: (a) hydrocarbons generated within the Poolowanna Formation are trapped entirely within that unit; (b) significant hydrocarbon accumulations are present within the Algebuckina Sandstone, but drilling has so far failed to intersect them; (c) very little hydrocarbon generation has occurred.

15"

100 1 o

\\ -• " ^ 1

PEERA PEERA FORMATION

i 005

i i 0-10 0-15 ATOMIC

• "M-

0-5 -

i 0-20 %

i 0-25

• 0-30

LEGEND 4 • • •

Evolutionary paths of the principal types of kerogen Walkandi - 1 Erabena - 1 Kuncherinna - 1

Fig. 11. Kerogen analysis (Van Krevelen diagrams) of Peera Peera a n d P o o l o w a n n a F o r m a t i o n s f r o m three wells in P o o l o w a n n a Trough.

provides an adequate combination of mature source, seal, reservoir porosity and permeability.

Algebuckina

Sandstone

The Algebuckina Sandstone is a m a j o r artesian aquifer, consisting of fine- to coarse-grained sandstone with fair to good porosity. Shales of the overlying Cadna-owie Formation probably form an adequate seal to the unit, while the underlying Poolowanna Formation has proven source

There is little evidence to suggest which alternative is correct, but on the basis of known source quality and maturation of the Poolowanna Formation in the central Poolowanna Trough, it is suggested that there is a reasonable likelihood that a few small oil accumulations may be present at the top of the Algebuckina Sandstone, particularly in the deeper portions of the trough.

STRUCTURAL DEVELOPMENT AND TIMING OF HYDROCARBON GENERATION The Poolowanna Trough has a complex and varied structural history. In the western part of the trough, seismic evidence suggests that Permian sediments occur in a structural setting similar to the Cooper Basin. Early Permian structuring is evident, with small horsts and grabens suggestive of a tensional tectonic regime. Following deposition of Early Permian Pedirka Basin sediments there was an important period of Late Permian structuring, as indicated by erosion of the sequence in structurally high positions, and by the large time break separating these deposits from overlying Mesozoic strata. The entire sequence was again deformed in the mid-Tertiary. In the central and eastern portions of the Poolowanna Trough, Permian sediments are absent and there is very little sign of Permian or Mesozoic structural development. Instead, the main phase of structuring appears to have occurred in the mid-Tertiary. According to Wopfner & Twidale (1967), Wopfner (1974) and Moore & Pitt (1984), widespread epeirogenic movements affected the Lake Eyre Basin probably

P. S. MOORE

Miocene tectonic event which essentially created the Poolowanna structure. Thus, although some hydrocarbons were probably generated prior to the structure being formed, the critical combination of mature source rocks and an adequate trap has been in existence for approximately 25-30 million years. Related, but shallower structures towards the margins of the Poolowanna Trough have similar maturation scenarios, although for many of these shallower structures the Poolowanna Formation has only attained the stage of initial oil generation (Ro max = 0.6-0.8 per cent).

CONCLUSIONS

With only 20 petroleum exploration wells drilled in the Simpson Desert region to date, the area remains largely unexplored. Future exploration will probably be relatively difficult and somewhat expensive compared with the high success rate achieved in the Cooper Basin region. However, the Poolowanna and Peera Peera Formations are rich in organic matter, a small but significant proportion of which is exinite. In addition, the source rocks are early-mature Fig. 13. Subsidence—maturation history, Poolowanna 1. Present day throughout most of the area (Ro max averages 0.8 per cent), geothermal gradient: 3.9°C/100 m. Known maturation level and have generated heavy, waxy oil in Poolowanna 1. successfully modelled using constant gradient of 3.0°C/ In the central, deeper parts of the Poolowanna Trough, 100 m. Vitrinite reflectance data measured from well. generation of liquid hydrocarbons from the Peera Peera and Methodology after Waples (1980). Poolowanna Formations is considered to have commenced the Late Cretaceous, after the Winton Formation was in the Late Oligocene or earliest Miocene, prior to deposition in Prime conditions for oil generation were attained of the Miocene Etadunna Formation. This situation is deposited. in the Early Tertiary, immediately prior to the main phase demonstrated at Poolowanna (Figs 9, 13). The structure is of structural Thus, the combination of (a) an elongate, anticlinal dome, with strong fault-control on the mature sourcedevelopment. rocks and (b) structural relief capable of western margin. The main fault extends from basement into trapping hydrocarbons has been existence for about 25 the Cretaceous Cadna-owie Formation and, indeed, may million years. This is consideredin sufficient time for the extend upwards as far as the Toolebuc Formation; generation, migration and entrapment of hydrocarbons, as stratigraphically higher horizons right up into the Winton Formation show flexure rather than faulting. The timing of the presence of oil in Poolowanna 1 testifies. the main phase of deformation is thus clearly post-Winton Formation and is assigned a mid-Tertiary age on the basis of ACKNOWLEDGEMENTS evidence discussed above, and presented by Moore & Pitt The author wishes to acknowledge discussions and helpful (1984). criticism by A. Cook (Wollongong University), K. Hoolihan Using the Lopatin (1971) technique of maturation analysis (Santos Ltd), N. Moriarty (Delhi Petroleum Pty Ltd), P. (Fig. 13), it is apparent that the Peera Peera and Poolowanna Moignard (S.A. Oil and Gas Corp.), G. Pitt (Western Mining Formations in Poolowanna 1 were incapable of generating Corp.) and P. Price (CSR Oil and Gas Division, Brisbane). liquid hydrocarbons until after the Winton Formation was This paper is published with the permission of Delhi deposited. Initial generation occurred in the Late Cretaceous Petroleum Pty Ltd and associates in the Pedirka Block (AAR and Early Tertiary, with significant oil generation beginning Ltd, Bridge Oil Ltd, Santos Ltd, Vamgas Ltd and Western in the mid-Tertiary, immediately prior to the Oligocene- Mining Corp.).

REFERENCES

AMBROSE, G. J., 1980: Southern Eromanga Basin excursion. S. Aust. Geol. Surv. Rep. 7 9 / 1 4 3 (unpubl.J. AMBROSE, G. J. & FLINT, R. B., 1982: M e s o z o i c stratigraphy—SW

margin of the Eromanga Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 376-7.

BARR, T. M . & YOUNGS, B. C., 1981: Cuttapirrie 1, an oil discovery

in the Early Jurassic of the Eromanga Basin. A PEA J., 21(1),

60-70.

COOK, A. C., 1982: Organic facies in the Eromanga Basin; in Moore, P. S. & Mount. T. J. (compilers) Eromanga Basin Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 234-57. COOPER, B. J., 1981: Carboniferous and Permian sediments in South Australia and their correlation. S. Aust. Geol. Surv., Q. geol. Notes, 79, 2-6.

DE JERSEY, N. J., 1971: Triassic miospores from the Tivoli Formation and Kholo Sub-group. Qld. Geol. Surv. Publ., 353.

FREYTAG, I. B., 1966: Proposed rock units for marine Lower Cretaceous sediments in the Oodnadatta region of the Great Artesian Basin. S. Aust. Geol. Surv., Q. geol. Notes, 18, 3-7. GRIFFITHS, M., 1980: Toodla No. 1 well completion

Dept. Mines. Ener. Rep. 80-128

(unpubl.).

report. S. Aust.,

HARRIS, W. K., 1970: An upper Jurassic microflora from the western margin of the Great Artesian Basin, South Australia. S. Aust. Geol. Surv., Q. geol. Notes, 35, 3-8. KANTSLER, A . J., COOK, A . C. & ZWIGULIS, M . , 1982: M a t u r a t i o n

patterns in the Eromanga Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 284-95.

POOLOWANNA TROUGH STRATIGRAPHY LOPATIN, N . V., 1971: Temperature and geologic time as factors in coalification. Akad. Nauk SSSR Izv. Ser. Geol. 3, 9 5 - 1 0 6 .

(Russian). geochemistry and source-rock potential of the Arrowie, Pedirka, Cooper and Eromanga Basins, central Australia. Delhi Petroleum Pty Ltd (unpubl.). MOORE, P. S., 1982: Mesozoic geology of the Simpson Desert region, northern South Australia; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 46-57. MOORE, P. S. & PITT, G. M . , 1984: Cretaceous of the Eromanga Basin—implications for hydrocarbon exploration. A PEA J., 24(1), 38-56. MOORE, P. S. & PITT, G. M. 1985: Cretaceous subsurface stratigraphy of the southwestern Eromanga Basin: a review; in Lindsay, J. M. (ed.) Stratigraphy, palaeontology, malacology. S. Aust. Dept. Mines & Ener. Spec. Publ. 5, 269-86. MOORE, P. S., PITT, G. M . & DETTMANN, M . E., 1986: The Early Cretaceous Coorikiana Sandstone and Toolebuc Formation: their recognition and stratigraphic relationship in the southwestern Eromanga Basin; in this volume. MORIARTY, K. C. & WILLIAMS, A. F., 1982: Hydrocarbons flushing in the Eromanga Basin—fact or fallacy; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 313-28. NORTHCOTT, 1. W., 1980: Coongra No. 1 well completion report. S. Aust. Dept. Mines Ener. Rep., 8 0 / 5 4 (unpubl.). PAPALIA, N., 1969: The Nappamerri Formation. APEA J. 9(2), 108-10. PRICE, P. L., 1978a: Palynological laboratory report No. 13/103; in Wiltshire, M. J., Well completion report, Poolowanna No. 1. Delhi Petroleum Pty Ltd (unpubl.). PRICE, P. L., 19786: Palynological laboratory report No. 13/104; in Wiltshire, M. J., Well completion report, Macumba No. 1. Delhi Petroleum Pty Ltd (unpubl.). SMYTH, M. & SAXBY, J. D, 1981: Organic petrology and geochemistry of source rocks in the Pedirka—Simpson Desert Basins, central Australia. APEA J. 21(1), 187-99. THORNTON, R. C. N., 1974: Oodnadatta Town Bore No. 2 well completion report. S. Aust. Miner. Resour. Rev. 141, 51-63.

MCKIRDY, D. M . , 1981: Petroleum

TISSOT, B. P. & WELTE, D. H . ,

1978: Petroleum formation and occurrence: a new approach to oil and gas exploration. SpringerVerlag, New York. WAPLES, D. W., 1980: Time and temperature in petroleum formation: application of Lopatin's method to petroleum exploration. Am. Assoc. Pet. Geol. Bull. 64, 916-26.

WELLS, A . T., FOREMAN, D. J., RANFORD, L. C . & COOK, P. J., 1970:

Geology of the Amadeus Basin, central Australia, Aust. Bur. Miner. Resour. Geol. Geophys. Bull. 100. report, Poolowanna No. 1. Delhi Petroleum Pty Ltd (unpubl.). WILTSHIRE, M . J., 19786: Well completion report, Macumba No. 1. Delhi Petroleum Pty Ltd (unpubl.). WILTSHIRE, M . J., 1978c: Well completion report, Colson No. 1. Beach Petroleum Pty Ltd (unpubl.). WILTSHIRE, M . J., 1982a: Thomas No. 1, well completion report. Argonaut International Corporation (unpubl.). WILTSHIRE, M. J., 19826: Late Triassic and Early Jurassic sedimentation in the Great Artesian Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide. WOPFNER, H . , 1964: Permian-Jurassic history of the western Great Artesian Basin. Roy. Soc. S. Aust. Trans. 88, 117-29. WOPFNER, H . , 1974: Post-Eocene history and stratigraphy of northeastern South Australia. Roy. Soc. S. Aust. Trans. 98, 1-12. WOPFNER, H . & TWIDALE, C. R., 1967: Geomorphological history of the Lake Eyre Basin; in Jennings, J. N. & Mabutt, J. A. (eds) Landform studies from Australia and New Guinea. Aust. Nat. Univ. Press, Canberra. WOPFNER, H . , FREYTAG, I. B. & HEATH, G. R., 1970: Basal JurassicCretaceous rocks of western Great Artesian Basin, South Australia: stratigraphy and environment. Am. Assoc. Pet. Geol. Bull. 54, 383-416. YOUNGS, B. C., 1975a: The Early Permian Purni Formation of the Pedirka Basin. 5. Aust. Geol. Surv. Q. geol. Notes, 54. YOUNGS, B. C., 19756: The geology and hydrocarbon potential of the Pedirka Basin. S. Aust. Geol. Surv. Rep. Invest. 44. YOUNGS, B. C. & BOOTHBY, P. G., 1982: The Late Permian-Triassic rocks of the southwestern Cooper Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, WILTSHIRE, M . J., 1978a: Well completion

76-8.

Geological Society of Australia Special Publication No. 12, 53-70

Palynology, cyclic sedimentation, and palaeoenvironments in the Late Mesozoic of the Eromanga Basin D. Burger Bureau of Mineral Resources, P.O. Box 378, Canberra City, A.C.T. 2601.

ABSTRACT Cyclic sedimentation in the Great Artesian Basin in Queensland is expressed by alternating deposition of sandstone and mudstone, and eight sedimentary cycles are defined in the Eromanga Basin. Palynological and palaeontological evidence suggest that these cycles may be associated with recurrent global eustatic sea level fluctuations during the Jurassic and Early Cretaceous, in the sense that high energy deposition (sand) and low-energy deposition (silt, mud) occurred in phase with low and high eustatic sea levels. The ages of nonmarine Jurassic and Neocomian sedimentary units are reassessed from their associations with the global sea level movements. Palaeogeographic reconstructions based on evidence from marine palynomorphs (dinoflagellates, acritarchs) illustrate the extent of successive marine incursions into the basin during Cretaceous high sea level phases, and their impact on regional facies boundaries.

INTRODUCTION The Eromanga Basin of the Great Artesian Basin in northeastern and central Australia contains the most complete onshore Jurassic and Lower Cretaceous sedimentary sequence of the entire continent (Fig. 1). In the deepest parts of the basin sections approximately 2.5 km thick have been measured. During this time the interior of the Australian Plate formed a remarkably stable tectonic province, which resulted in closely comparable sedimentary sequences being accumulated in the adjacent Surat and Carpentaria Basins. The sedimentary sequence of the Eromanga Basin in Queensland and Northern Territory has been reviewed by Senior et al. (1978), and their stratigraphic nomenclature is adopted here with slight revisions, which are based on the following studies. Recently, Senior et al. (1975) have redefined certain Cretaceous formations through parts of the Eromanga and Carpentaria Basins. Vine (1970), Vine & Paine (1974), Exon & Senior (1976), Smart et al. (1978), Burger & Senior (1979), and Burger (19826) have shed more light on the depositional history at the northeastern and northern basin margin. Hind & Helby (1969) reviewed the geology of the basin in northern New South Wales. Subsequent drilling by the New South Wales Department of Mines has led to revisions of the stratigraphic nomenclature there (Morgan, 1978; Byrnes, 1980; McMinn & Burger, 1986). Age determinations of sedimentary rocks based on the invertebrate fossil record are restricted to the late Early Cretaceous (Aptian and Albian). Ammonites, bivalves, and other shelly fossils were reviewed by Day (1969); Haig (1979) presented a very detailed account of benthonic and planktonic foraminifera in Queensland, and Ludbrook (1966) studied foraminifera in South Australia. The present paper discusses the palynological contribution towards dating and correlating

sedimentary rock units in Queensland, New South Wales, and Northern Territory. Palynological summaries of the South Australian portion of the basin are being written by other workers; only Santos Oodnadatta 1 well, which relates to the Cretaceous palynological and geological picture in New South Wales and Northern Territory, and DFS Innamincka 1, whose palynology was discussed by Evans (in Ryan, 1961) are included in this paper. Drill holes referred to in this paper are plotted in Figure 1, and listed in Table 1 with the relevant palynological references. Various geographic locations and structural features mentioned hereafter are shown in Figure 2.

THE PALYNOLOGICAL RECORD Spores and pollen grains The complexes of spores and pollen from the Great Artesian Basin show strong affinities of the regional vegetation with the Late Mesozoic Indo-European Floral Province (Burger, 1981). Owing to equable climatic conditions many elements identified here also occur in the Mesozoic of other Gondwanic and Laurasian continents, notably eastern Siberia (Vachrameev, 1964), South Africa (McLachlan & Pieterse, 1978), and India (Bose & Dev, 1961). Plant life of inland Australia was dominated by gymnosperm (conifer) forests, and the coastal vegetation carried diverse associations of ferns, such as Cyatheaceae, Osmundaceae, Gleicheniaceae, and several hepatics and mosses. The first angiosperm elements very slowly infiltrated the lowland vegetation during the early Albian (Dettmann, 1973, 1981; Burger, 1980a, 1981). Changes in Australian Mesozoic plant life are recorded in literature by documenting earliest and ultimate geological

D. B U R G E R

GULF OF

CARPENTARIA

COOPER BASIN

SOUTH

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15

AUSTRALIA

BMR Mossman 1 BMR Croydon 1 BMR Gilberton 2 BMR Richmond 3 GSQ Hughenden 7 BMR Hughenden 1 - 1 A GSQ Manuka 1 BMR Boulia 3 - 3 A Conorada Goroonoo 1 BMR Tangorin 1 Exoil Brookwood 1 QDM Aramac 1 BMR Muttaburra 1 BMR Longreach 5 BEA Coreena 1

16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

BMR Jericho 11 FDNL Alice River 1 LOL Saltern Creek 1 APC Newlands 1 WOL 3 (Warbreccan) Alliance Yongala 1 Alliance Chandos 1 DPS Betoota 1 Santos Oodnadatta 1 (SA) BMR Hay River 12 ( N T ) BMRTambo6 BMR Augathella 3 APC Westbourne 1 BMR Augathella 6 BMR Charleville 5

31 32 33 34 35 36 37 38 39 40 41 42 43 44 45

BMR Toompine 1 BMR Eulo 2 BMR Bulloo 1 DPS Innamincka 1 GSQ Eromanga 1 BMR Urisino 1 OM Wanaaring ODH 1 OM Yantabulla ODH 1 OM Bellfield DOH 1 D M Weilmoringle OOH 1 NAI Whyenbirra 1 BMR Wyandra 1 DON Maranda 1 APC Thunderbolt 1 AAP Mayneside 1

Fig. 1. Mesozoic Great Artesian Basin and Permo-Triassic Cooper and Galilee Basins, northeastern and central Australia. Locations of drillholes in Table 1.

PALYNOLOGY AND CYCLIC SEDIMENTATION

T A B L E 1. Alphabetical list of drillholes mentioned in text, with relevant palynological bibliographic references. Numbers correspond to plots

in Fig. 1.

17 12 27 29 39 23 8 11 33 22 30 15 2 35 32

Alice River 1 Aramac 1 Augathella 3 Augathella 6 Bellfield DDH 1 Betoota 1 Boulia 3-3A Brookwood 1 Bulloo 1 Chandos 1 Charleville 5 Coreena 1 Croydon 1 Eromanga 1 Eulo 2

a . Burger,, 1968 b . M o r g a n ,, 1978 c. E v a n s , 1966b d. E v a n s , 1966c e . E v a n s , 1966d

fl gl fr m b m m fk 0 ef m 1

•k * *

f . B u r g e r , 1973a g- B u r g e r , 1977 h . B u r g e r , 1982a i . B u r g e r , 1982b 1979 j . McKellar

3 25 6 5 34 16 14 7 43 45 1 13 19 24 9

Gilberton 2 Hay River 12 Hughenden 1-1A Hughenden 7 Innamincka 1 Jericho 11 Longreach 5 Manuka 1 Maranda 1 Mayneside 1 Mossman 1 Muttaburra 1 Newlands 1 Oodnadatta 1 Ooroonoo 1

k . Burger & K e m p , 1972 1 . Burger & Senior, 1979 m . McMinn & B u r g e r , 1983 n . E v a n s , in L a i n g , 1966 o . Terpstra & B u r g e r , 1969

occurrences of transient spore and pollen species in the palynological record. Some of those changes clearly were of a local restricted nature, whereas others took place across the entire continent. The importance of those observations for biostratigraphic application is obvious, since palynological studies of the late Pleistocene vegetation history in Europe and North America, supported by radiocarbon datings have shown that under favourable conditions plants may colonise vast areas within a geologically negligible time span. This section briefly summarises the biostratigraphic record in the Great Artesian Basin and its chronostratigraphic implications. The Jurassic sequence of spores and pollen grains in Queensland is subdivided into five zonal intervals or 'palynological units' J1 to J6 (Figs 3, 9), which were proposed by Evans (1966a) and subsequently slightly modified by Burger & Senior (1979). Unit boundaries are defined by successive first appearances of well-known and widely distributed spore and pollen species. In their review of Cretaceous palynology in Australia, Dettmann & Playford (1969) commented on the 'remarkable overall uniformity' of Early Cretaceous spore and pollen sequences in eastern Australia, and subsequent studies have amply confirmed this. Eight zonal intervals have been outlined in the Neocomian to Cenomanian record of the Great Artesian Basin in Queensland (Fig. 4), and several of those intervals have been observed also in the Otway Basin and Western Australia.

Marine phytoplankton

Despite their restricted nature, dinoflagellates and acritarchs have proven their value in biostratigraphic studies of the Great Artesian Basin. Burger (1982c) outlined three Neocomian dinoflagellate zonal intervals from the Gilbert River Formation in the Carpentaria Basin (Fig. 5). Morgan (1980c) distinguished one Aptian and two Albian dinoflagellate zones from the marine Rolling Downs Group in the Eromanga Basin (Figs 4, 5), and Burger (1980c) identified those zones also in the Surat Basin.

Isochronous palynological

intervals

fh m fk ij f m a * 1 df h 1 df bmp bf

Morgan (1980c) analysed dinoflagellate, foraminiferal, and megafaunal published records from the Aptian and Albian

4 18 26 10 44 31 36 37 20 40 28 41 42 38 21

Richmond 3 Saltern Creek 1 Tambo 6 Tangorin 1 Thunderbolt 1 Toompine 1 Urisino 1 Wanaaring DDH 1 WOL 3 (Warbreccan) Weilmoringle DDH 1 Westbourne 1 Whyenbirra 1 Wyandra 1 Yantabulla DDH 1 Yongala 1

i fl ar 1 kl m m b df b cr f q b efn

P- Dettmann & P l a y f o r d , 1969 q . B u r g e r , iji Senior et a l . , 1969 r . Evans & B u r g e r , in Exon et a l . , 1972 * under study

in Europe and Australia. He demonstrated the timeconcordant nature of several dinoflagellate events in the Great Artesian Basin as compared with Europe, and on that basis dated his dinoflagellate zonal intervals as shown in Figure 5. Burger (1982c) estimated the ages of the Neocomian zones in the Carpentaria Basin by a similar comparative study with the European record (Fig. 5). A similar approach of dating spore and pollen zones would lead to unreliable conclusions, as provincialism imposed by geographic and climatic barriers has occurred throughout the history of land plants. However, the present evidence leads to the conclusion that, within the Great Artesian Basin, certain zonal boundaries in fact represent time-parallel horizons. Burger (1980c, 1982c) and McMinn & Burger (1986) found no diachronous relationship between several Aptian and Albian spore and pollen zones and the abovementioned dinoflagellate intervals across northeastern and central Australia. Likewise, the time-parallel nature of the Neocomian zonal boundaries is suggested by the fact that ages of sporepollen zonal intervals shown in Figure 4 confirm indirect age determinations of their counterparts in nonmarine sequences of the Otway and Gippsland Basins (see Dettmann, 1963; Evans, \966e\ Burger, 1973c, 1982c; Dettmann & Douglas, 1976). There is no direct evidence for the ages of palynological units J1 to J6, and several authors (De Jersey & Paten, 1964; Evans, 1966c; Reiser & Williams, 1969; McKellar, 1974, 1978) have compared them with corresponding palynological intervals which are associated with Jurassic marine invertebrate faunas in Western Australia (Balme, 1957, 1964; Filatoff, 1975). Because of what has been said with regard to the Cretaceous zonal intervals, those authors' implicit acceptance of isochronous Jurassic units is followed here in principle. Jurassic stratigraphic palynology of the Great Artesian Basin is at present under review, and it may be necessary to discuss some aspects of the following summary again at a later date.

CYCLIC SEDIMENTATION AND ITS CAUSES

Senior et al. (1978) described the standard sedimentary column of the Eromanga Basin in Queensland, which is shown in Figures 3 and 4, as a series of coarse sandstone bodies separated by finer-grained, silty and argillaceous intervals (see

D. B U R G E R

Fig. 2. Isopach map, H u t t o n Sandstone, 100 m intervals. Marginal-facies deposits at the n o r t h e a s t e r n margin depicted in Fig. 6. Locations of townships and structural units mentioned in text.

below). A similar sequence of alternating sandstone and mudstone units has been mapped in the Surat Basin by Exon (1976). Exon (1980) and Exon & Burger (1981) interpreted those recurrent changes of high-energy to low-energy deposition as the result of widespread sea-level fluctuations, and this paper offers palynological and other evidence to show that contemporaneous cycles in the E r o m a n g a Basin may be related to the same causes. Several authors, who advocated the occurrence of synchronous global rises and falls of the sea level in the geological past, believed them to be related to movements of continental plates, possibly of isostatic origin, compensating volumetrically for expanding and subsiding mid-oceanic ridges (Hallam, 1963; Pessagno, 1972; Rona, 1973). Vail etal (1977) distinguished six global eustatic cycles during the Jurassic and three during the Early Cretaceous, and presented a curve of 'relative changes in sea level', indicating a series of asymmetrical cycles, each including a gradual rise and subsequent rapid fall of the sea level (Fig. 3). Subsequent criticism led Vail & Todd (1981) to draw up a slightly modified

curve of 'relative changes of coastal onlap' for the Jurassic and earliest Cretaceous, in which the global cycles are still recognised. The asymmetry of the cycles seems to be borne out by asymmetric cyclic sedimentation in the Surat Basin during the Jurassic. In Exon & Burger's (1981) view, a rapidly falling sea level started off high energy deposition by braided rivers, sometimes after a phase of widespread erosion. A gradually rising sea level resulted in more sluggish regional drainage (meandering streams, parallel bedding-planes), and finally deposition of silt and m u d in lakes and swamps, with local accumulation of organic matter, leading to the f o r m a t i o n of coal. According to Vail et al. (1977) the global sea level was higher during the Cretaceous. C o o p e r (1977) described a succession of global marine transgressive phases, and produced a curve of eustatic sea level fluctuations, of which the lower part is shown in Figure 4. In the Great Artesian Basin these fluctuations are apparent as recurrent marine transgressions and regressions, with deposition of siltstone

EROMANGA BASIN

Palynological

SURAT BASIN

Units

Global sea

(Queensland)

level

movements

JURASSIC INTERNATIONAL STAGES

Ma 130

(EARLY Murospora

f/orida

CRETACEOUS)

Mooga Sandstone

Tithonian Hooray Sandstone

-140 ;

Unit J 6

Westbourne

Oxfordian

Gubberamunda Sst^ T

:: W e s t b o u r n e Springbok

Adori Sandstone

Kimmeridgian

Orallo Formation

-149-

150

Callovian

Sst

Unit J5 Birkhead Formation-

Unit J 4

g | 0 u r o m b a h Beds^v

-160

Bathonian

Walloon Formation

Bajocian 171-

Mutton Sandstone

-170

Aalenian Toarcian

Unit J2-3 Evergreen Formation

Pliensbachian

180

Unit J1 Sinemurian

Precipice Sandstone Sandstone

Si Its

tone

Mud

stone

and

conglomerate

Hettangian Rising level

sea (LATE

-192

190

TRIASSIC)

L 200

Fig. 3. Palynological correlation of Jurassic formations, Eromanga and Surat Basins. Time relationships between sedimentary cycles and global eustatic sea level movements. See Vail et a/. (1977) for global sea level curve. Absolute ages (Ma) based on Van Hinte (1976a).

E R O M A N G A BASIN

Append ?

Global sea level movements

S U R A T BASIN

Palynological zonal intervals

Queensland

CRETACEOUS INTERNATIONAL STAGES

distocarinatus ? ?

Phimopollenites pannosus

Coptospora

paradoxa

Crybe/osporites

striatus

Osm unda cidites dub/us

11111

•Griman Creek Formation —_-_Surat

Siltstone-^u

^Coreena Member Doncaster Member

Oo iMinmi Member • :

Fora m in isporis asymmetricus

; Nullawurt Sandstone Member

£ DK3

Forminisporis wonthaggiensis

Kinguli Members 5A

Cicatricosisporites austra/iensis

: Mooga Sandstone

Rising sea level Fie 4

Palynological correlation of C r e t a c e o u s f o r m a t i o n s , E r o m a n g a and Surat Basins. T i m e relationships between sedimentary cycles and global eustatic sea level movements. See Vail et al. (1977) for Late Jurassic-Berriasian sea level curve, a n d C o o p e r (1977) for V a l a n g i n i a n - C e n o m a n i a n sea level curve. Absolute ages (Ma) based on Van Hinte (19766).

Palynological dinoflagellates

Oodnadatta

Formation

5-31 £

iana

Endoceratium ludbrookiae

zonations spores and pollen

Phim.

CRETACEOUS INTERNATIONAL STAGES

pannosus

Copt, Pseudoceratium turneri

Global sea level movements

paradox a Cryb.

striatus

Bulldog

2 Cyc/osporites

Shale Odontochitina operculata

hughesii

5O •c >

Wallumbilla

z a

Fm O.

operculata

asymmetricus

C o f f i n Hiil M e m b e r DK3

o Foram. wo nth a ggiensis

i Yappar

m §

Cicatric. austra/iensis

Member

no record Unit J 6 (EULO Q U E E N GROUP)

Rising sea level

Fig. 5. Palynological associations of Neocomian sedimentary rocks (southern Carpentaria Basin) apd Aptian-Albian sedimentary rocks (Santos Oodnadatta 1). See Figs 2, 3 for global sea level curve and geological time scale.

D. BURGER

and mudstone, and clean to lithic sandstone respectively, without apparent asymmetry. Exon (1976, 1980) and Exon & Burger (1981) described Jurassic sedimentary Cycles 1 to 4 and Lower Cretaceous Cycles 5 and 6 from the Surat Basin. Exon & Burger (1981) matched Cycles 1 to 4 against the Jurassic sea level curve of Vail et al. (1977) as shown in Figure 2. Cycles 5 and 6 are here matched against Cooper's (1977) sea level curve (Fig. 4), which broadly resembles that of Vail et al. (1977) for the Early Cretaceous, but shows more definition.

SEDIMENTARY CYCLES IN THE EROMANGA BASIN Morgan (1980&) discussed Aptian and Albian marine and nonmarine episodes in several Australian sedimentary basins, including the Eromanga Basin, and demonstrated that they occurred in synchrony with global eustatic high and low sea levels. The present author entirely agrees with this concept and extends it to include the basal Cretaceous (Neocomian) and Jurassic of the Eromanga Basin, thus expanding on Exon & Burger's (1981) similar study of the Surat Basin. Eight sedimentary cycles are outlined in the Eromanga Basin; Jurassic Cycles 1, 2, 3, Late Jurassic-basal Cretaceous Cycle 4-5A, and Cretaceous (BarremianAlbian) Cycles 5B, 6, 7 and 8. Two Cretaceous cycles are distinguished in Santos Oodnadatta 1 (South Australia), and two Cretaceous cycles in New South Wales. Numbers 1-6 correspond to the system of cycles for the Surat Basin described by Exon & Burger (1981). As in the Surat Basin, a full cycle starts with basal arenaceous sediments corresponding to a low global sea level, and ends with fine-grained, silty or argillaceous units deposited during high global sea level. Palynological correlation of Jurassic and Cretaceous sedimentary cycles in the Surat and Eromanga Basins, and the association of each cycle with global eustatic sea level movements are shown in Figures 3-5.

possibly Newlands 1, I believe that it may be related to a high global sea level during the Toarcian. Cycle 2 includes the upper sandy part of the Hutton Sandstone and the argillaceous Birkhead Formation (Fig. 3). Palynological unit J4 occurs in the Hutton Sandstone of Coreena 1, Hughenden 7, and the Birkhead Formation of Thunderbolt 1, Maranda 1, Chandos 1, Newlands 1, Brookwood 1, and Westbourne 1. Palynological unit J5 has been identified in the Birkhead Formation of the Tambo and Jericho areas (see Fig. 2) and the northern Surat Basin (Evans, 1966b, Burger & Senior, 1979), and tentatively in WOL 3 (Warbreccan), Newlands 1, and Mayneside 1. This cycle is related to global eustatic cycles J2.1 and J2.2 during the Aalenian to Bathonian. A fall of sea level at the end of cycle J2.2 may have been too slight to induce regressive conditions in the basin. Cycle 3 includes the Adori Sandstone and the argillaceous Westbourne Formation of the Injune Creek Group (Fig. 3). It falls within palynological unit J5 in the Longreach and Jericho areas (Burger & Senior, 1979). Unit J5 also occurs in the Westbourne Formation of Westbourne 1, and undifferentiated units J5-6 are present in the Westbourne Formation-Adori Sandstone interval of the Tambo, Augathella, and Jericho areas (Evans & Burger, 1972), and in the Westbourne Formation of Newlands 1 and Mayneside 1. More specific data may be derived from that interval by further examination of the continuously cored sections of GSQ Eromanga 1 and GSQ Manuka 1. The cycle is related to Callovian eustatic cycle J2.3. Palynological evidence has established the Ronlow Beds, Blantyre Beds, and lower Eulo Queen Group, which are shown in Figure 6, to be broadly coeval with the Westbourne Formation (Burger, 1977, 1982b; Burger & Senior, 1979). In GSQ Hughenden 7 the Injune Creek Group cannot be subdivided, and a direct correlative of the Westbourne Formation may not be present (McKellar, 1979).

Late Jurassic to Aptian Cycles 4-5 Early to Middle Jurassic Cycles 1 to 3 At present, Cycles 1, 2 and 3 have been recognised only in the Surat Basin and the eastern part of the Eromanga Basin. Initial data published by Gravestock (1982) and Moore (1982) show that periodic sedimentation and erosion occurred during the Jurassic in South Australia, but it is too early to suggest detailed correlations between the Queensland and South Australian sequences. Cycle 1 starts with coarse sands of the basal Hutton Sandstone, and is taken to end with an argillaceous sequence, which has been found within the lower Hutton Sandstone in a number of deep wells (Fig. 3). This mudstone is correlated with the Evergreen Formation in the Surat Basin. The cycle is probably incomplete, as in the Surat Basin it includes the Precipice Sandstone, which lies within palynological unit Jl. This unit has been identified in the basal Hutton Sandstone in the Tambo area in the northeastern Eromanga Basin (Evans & Burger, 1972), and in South Australia (Moore, 1982). Assemblages not older than palynological units J2-3 have been recovered from the basal Hutton Sandstone in more central areas of the basin (Brookwood 1, Chandos 1, Westbourne 1, Hughenden 7, Yongala 1, possibly also WOL 3 and Newlands 1), and this suggests that sedimentation started later there. The argillaceous interval which ends Cycle 1 may not form a continuous layer, but indicates one or several episodes of tranquil deposition. From its correlation with palynological units J2-3 in Westbourne 1, Eromanga 1, Yongala 1, and

In the Surat Basin, Cycle 4 includes the Gubberamunda Sandstone and Orallo Formation (Fig. 3). It is associated with palynological unit J6 and corresponds to the OxfordianKimmeridgian/Tithonian global eustatic cycle J3.1. Cycle 5 is compound (Fig. 4); Subcycle 5A consists of the Mooga Sandstone and Kingull Member of the Bungil Formation, and lies within the Cicatricosisporites australiensis and Foraminisporis wonthaggiensis Subzones of the Murospora florida Zone. It corresponds to two episodes of high global sea level, one in the Berriasian (global cycle J3.2) and one in the Valanginian-Hauterivian. Subcycle 5B includes the Nullawurt Sandstone Member and Minmi Member of the Bungil Formation, and the Doncaster Member of the Wallumbilla Formation; it is associated with the Barremian Foraminisporis asymmetricus Subzone of the Murospora florida Zone, and the Aptian Osmundacidites dubius Zone. In the Eromanga Basin global eustatic cycles J3.1 and J3.2 had less impact, and only two sedimentary cycles are distinguished. Cycle 4-5A commences with the Hooray Sandstone and ends with the silty lower Cadna-owie Formation (Figs 3, 4). The Hooray Sandstone contains Late Jurassic palynological units J5-6 in Newlands 1, Mayneside 1, Yongala 1, Eromanga 1 and Whyenbirra 1, as well as in the Tambo and Jericho areas (Evans & Burger, 1972; see Fig. 2). The basal Cretaceous Cicatricosisporites australiensis Subzone of the Murospora florida Zone has been identified in the Hooray Sandstone of GSQ Eromanga 1, Newlands 1,

zones

(Osmundacidites

dubius) asymmetricus

Foraminisporis wonthaggiensis Cic.

APC Newlands

LOL Saltern Creek

BEA Coreena

QDM Aramac

BMR Hughenden 1-1A

BMR Richmond 3

BMR Gilberton 2

australiensis

Unit J 6

Units J 4 - 5

4 3

Sandstone conglomerate Siltstone

GSQ Hughenden 7

EROMANGA BASIN (Queensland)

Mudstone

CARPENTARIA BASIN

Murospora florid a

Foram.

Bed. Cycles

PALYNOLOGY AND CYCLIC SEDIMENTATION

Spore-pollen

Fig. 6. Palynological correlation of depositional and erosionai events, northeastern and northern Eromanga Basin. See Fig. 1 for section location.

D. BURGER

Whyenbirra 1, Saltern Creek 1, Brook wood 1, and some boreholes in the Tambo-Jericho area (Burger, 1973c). The lower Cadna-owie Formation is associated with the Foraminisporis wonthaggiensis Subzone of the Murospora florida Zone in Yongala 1, GSQ Eromanga 1, and Manuka 1. The presence of dinoflagellate zonal interval DK3 (see Burger, 1982a) in the formation of Manuka 1 confirms a Hauterivian age for the end of Cycle 4-5A. The younger Cycle 5B starts with the Wyandra Sandstone Member of the Cadna-owie Formation, and ends with the argillaceous Doncaster Member of the Wallumbilla Formation (Fig. 4). The Wyandra Sandstone Member is associated with the Foraminisporis asymmetricus Subzone in GSQ Eromanga 1, Chandos 1, Mayneside 1, Innamincka 1, and probably also Bellfield DDH 1 in New South Wales, and for that reason may be correlated with the Nullawurt Sandstone Member of the Bungil Formation in the Surat Basin. The Doncaster Member is widely associated with the Osmundacidites dubius Zone (Burger, 1968, 1977, 1980c, 1982b\ Exon & Burger, 1981). In the Eromanga Basin it lies almost entirely within the Odontochitina operculata Dinoflagellate Zone (Morgan, 1978; Burger, 1982b), and in the Surat Basin only in the upper part of the zone (Burger, 1980c), which confirms the spore and pollen evidence that the base of the member is diachronous as shown in Figure 3. To the northeast, the sequence contemporaneous with Cycles 4-5B shows no distinct cyclic character (Fig. 6). The Cadna-owie Formation is replaced laterally by the Hooray Sandstone and in places has disappeared by erosion (Burger & Senior, 1979). The beginning of Cycle 5B is not apparent because of the marine character of the 'Ronlow beds' or 'Hooray Sandstone' in Aramac 1, and the Gilbert River Formation in Hughenden 1-1A, Richmond 3, and Gilberton 2 within the Foraminisporis asymmetricus Subzone. Northward across the Euroka Arch, the Gilbert River Formation and Wallumbilla Formation appear to represent a gradual change from terrestrial to open-marine environments (Cycle 5A-B) during the Neocomian and Aptian (Figs 5, 6, 7). In northern New South Wales the Hooray Sandstone gradually overlapped Palaeozoic and older granite and metamorphic basement rocks (Fig. 8). The oldest documented Hooray interval includes the Cicatricosisporites australiensis Subzone; no record exists locally of Jurassic palynological unit J6 (Morgan, 1978). The marine Doncaster Member of the Wallumbilla Formation forms the end of Cycle 5A-B.

Middle Cretaceous Cycles 6 to 8

Cycle 6 is not well-defined in the Surat Basin (Fig. 4). It starts with the nonmarine basal Coreena Member of the Wallumbilla Formation within the early Albian Crybelosporites striatus Zone, and its upper limit is blurred by a very gradual retreat of the sea when the Surat Siltstone and Griman Creek Formation were deposited in the interval of the Coptospora paradoxa Zone (Burger, 1980c; Exon & Burger, 1981). In the Eromanga Basin Cycle 6 is confined to the Coreena Member, whose upper part has a marine regressive character, indicating the beginning of Cycle 7 (Fig. 4). Unlike the Surat Basin, both the Crybelosporites striatus and Coptospora paradoxa Zones are confined to the Coreena Member in the northeastern and eastern regions (Aramac 1, Longreach 5, Jericho 11, Augathella 6, Charleville 5, Wyandra 1), and in the southern part of the basin (Eulo 2, Bulloo 1) as well as northern New South Wales (Fig. 8). The time-transgressive nature of the top of the member with regard to the palynological sequence is therefore clear (Fig. 4).

In the Bulloo Embayment in northern New South Wales, shown in Figure 8, the Coreena and Doncaster Members of the Wallumbilla Formation appear to pass laterally into the marine Bulldog Shale (McMinn & Burger, 1986). In Santos Oodnadatta 1 in South Australia the Bulldog Shale includes the Cyclosporites hughesii and Crybelosporites striatus Subzones of the Dictyotosporites speciosus Zone (Fig. 5), and represents the final stage of what may probably have been a single Cycle 5A-5B-6. (Relations between the formations and spore-pollen zones are shown in Figure 8). Cycle 7 is probably represented in the Surat Basin only as the nonmarine sandy Griman Creek Formation (Fig. 4). In the eastern and southern Eromanga Basin it begins with the correlative sequence, the sandy upper part of the Coreena Member, and ends with the marine argillaceous Toolebuc Formation-Mackunda Formation sequence, which contains the Phimopollenites pannosus spore-pollen Zone and the Endoceratium ludbrookiae Dinoflagellate Zone (Tambo 6, Longreach 5, Augathella 6, Jericho 11, Charleville 5, Toompine 1, Bulloo 1, and several boreholes in New South Wales, see Figs 4, 8). In the northern regions the Wallumbilla Formation represents continuous marine argillaceous deposition from the Aptian to the middle Albian (Hughenden 7, Manuka 1, Ooroonoo 1, Boulia 3, Croydon 1). Palynological evidence, however, indicates local regressive conditions indicating the beginning of Cycle 7 in the interval of the Coptospora paradoza Zone (McMinn & Burger, 1986). Farther south and west, Cycle 7 starts with the Coorikiana Sandstone, which includes assemblages of the Coptospora paradoxa Zone in Urisino 1, Betoota 1, Oodnadatta 1, and Hay River 12 (Figs 5, 7). The overlying argillaceous marine sequence, marking the end of the cycle, includes the Urisino Beds, Wooldridge Limestone Member correlative (see McMinn & Burger, 1986), and the lower Oodnadatta Formation. The upper brackish to nonmarine part of that formation (Mount Alexander Sandstone Member) forms a gradual transition into the next Cycle 8. Cycle 8 commences in Queensland and South Australia with the sandy nonmarine Winton Formation, which probably represents a low global sea level during the earliest Cenomanian (Fig. 4). Dettmann & Playford (1969) associated the formation in South Australia with the Phimopollenites pannosus Zone. The correlative Normanton Formation, the youngest Mesozoic unit in the Carpentaria Basin, was tentatively placed within the succeeding Appendicisporites distocarinatus Zone by Burger (1973&). Marine Cenomanian mudstones on Bathurst Island, north of Darwin, also seem to fall within that zone (Norvick & Burger, 1975). Palynomorph index species are still poorly documented from the Winton Formation in Queensland, and considerable proportions of reworked spores have been observed in many Winton assemblages. More study is needed to ascertain zonal associations of the formation in Queensland.

OBSERVATIONS ON PALAEOENVIRONMENTS

The Hutton Sandstone represents the initial stage in the history of the Eromanga Basin. Quartzose sandstone containing silicified wood and plant debris was laid down by braided rivers. Isopachs of the formation (Fig. 2) suggest that a system of river valleys extended east from South Australia, bounded by the Boulia, Thargomindah, Eulo, and Cunnamulla Shelves (see also Wiltshire, 1982). A second drainage system seems to have originated in the north somewhat later. The loci of maximum thickness are closely

Spore-pollen

zones

Santos

BMR

Oodnadatta

Hay River 12

Gilberton 2

Mossman

Sed.

2 s

Cycles

z o

Phimopollenites

pannosus

Coptospora paradox

Crybe/osporites

striatus

Osmundacidites

dubius

Murospora florida

O H< o r

... . W;':^ + + + Cadna-owie + + + + Fm

+ .

+ +

+ + + + River.V + + + + / Croydon-Smithb'ne J basement high + +

+ +

• • .. . Formation : X ./ •

/ + 4 - - 1 - - I . 4 . 4 - 4 - 4 -

Sandstone conglomerate

— EROMANGA

Si/tstone Mudstone

O cn m .o

(SA)

(NT)

I I

BASIN

CARPENTARIA

(Queensland)

BASIN

o z

Basement

Fig. 7. Palynological correlation of Cretaceous formations, northwestern Eromanga and southern Carpentaria Basins. See Fig. 1 for section location.

Os LU

Spore-pollen zones Dettmann a Playford 1969 Burger 1 973a , 1 980a

BMR Urisino 1

Sed. Cycles

Wanaaring DDH 1

Yantabulla DDH 1

Weilmoringle DDH 1

Phimopollenites pannosus Coptospora paradoxa Crybelosporites st riat us O.

dubius

asymm.

Cyc/ospor. hughesii

Foram.

wonthaggiensis

Cicatric.

australiensis

^ —

Bulloo e m b a y m e n t +

Mooga

Cunnamulla + + shelf . -

Thargomindah shelf

Sandstone conglomerate SUtstone Mudstone +

EROMANGA

BASIN

(New South Wales)

Basement

Fig. 8. Palynological correlation of Cretaceous formations, southern Eromanga Basin. See Fig. 1 for section location.

SURAT BASIN (Queensland)

PALYNOLOGY AND CYCLIC SEDIMENTATION Sed. Spore and pollen cycle zonation distocarinatus P. pannosus

Cryb. striatus

i >

Dinoflagellate zonation (no record)

E. ludbrookiae P. turneri 0. operculata DK3 DK2

C. austra/iensis

4 3 2 1

Unit J6

DK1

Unit J5 Iff

Unit J4

(no record)

Units J2-3 Unit J1

Aequitriradites spp. Ast. asteroides Cal/ialasp. dampieri Camarozon. c/ivosus Cic. austra/iensis

CO LU

wonthaggiensis

!! i i

Apteodin. granulatum Canningia co/liveri Dingodinium cerviculum Endocerat. ludbrookiae Muderongia tetracantha Odont. opercu/ata Pseudocerat. turnerii

asymmetricus

1 1

1 11

Laevig. belfordii Microfasta evansii Murospora florida PhimopoH. pannosus PHosispor. notensis Reticulatispor. pudens

Early

F. F.

1 1

C. paradoxa

6 5

l 1

Cryb. stylosus Cyc/ospor. hughesii Diet, speciosus F. asymmetricus F. wonthaggiensis /nap. turbatus

Middle

0. dubius

Late

JURASSIC

CRETACEOUS

(8)

Stratigraphic distribution of microfossils

Cic. hughesii Classopollis spp. Clavatipollenites spp. Cont. cooksonae Coptospora paradoxa Cryb. striatus

Age

20/A/24

Fig. 9. Spore-pollen and dinoflagellate zonal intervals in the Great Artesian Basin, and stratigraphic distribution of selected zone index palynomorphs.

aligned with the long axes of the Cooper and Galilee Basins (Fig. 1), which may have been lines of continuing subsidence. The two river systems were separated by a topographic high, probably arising from contemporaneous movements of the basin floor in the region of Longreach. Both systems discharged into the Surat Basin, and from there, united, into the open ocean to the east via the so-called Brisbane and Toowoomba Straits. The only suggestion of a widespread change of the environment at that time is the occurrence of an argillaceous interval within the Hutton Sandstone. Although probably a correlative of the Evergreen Formation, it has no formal status, and the western limit of the mappable Evergreen Formation is indicated in Figure 2. There is no evidence that this mudstone was deposited offshore (Wiltshire, 1982). Oolite beds containing swarms of spinose acritarchs are so far known only from the Evergreen Formation in the Surat Basin. They disappear further west, and this suggests that the sea had brief access to the Surat Basin only, via the eastern corridors (Exon & Burger, 1981). Temporary brackish conditions at the end of global eustatic cycle J2.2 are indicated by the occurrence of rare spinose acritarchs in the argillaceous Birkhead Formation of Cycle 2 in Brookwood 1, the Corfield area (Burger & Kemp, 1972), and probably Hughenden 7. Similar organisms have been recorded in the argillaceous Westbourne Formation of Cycle 3 in Tambo area (Evans, 1966b) as the only known regional indication of high sea level during global cycle J2.3. Contemporaneous nonmarine arenites (Eulo Queen Group) occur on the flanks of the Euroka Arch and CroydonSmithburne Basement High further north (see below).

Burger & Senior (1979) detected widespread erosion or nondeposition in the Longreach, Jericho, and Muttaburra areas, which involved removal of part of the Hooray Sandstone during palynological unit J6 (Fig. 6). Exon et al. (1972) observed the lower part of the formation missing from the outcrop sequence in the Tambo area. Unit J6 is probably present in the Hooray Sandstone of Mayneside 1, Newlands 1 and Whyenbirra 1, but is partly missing from the fossil sequence in GSQ Eromanga 1. In New South Wales the Hooray Sandstone overlying basement may be locally not older than basal Cretaceous. It seems likely, therefore, that a low global sea level during eustatic cycle J3.1 gave rise to a complex palaeogeographic mosaic of deposition and erosion on an old Middle Jurassic topography, and deposition of the Hooray Sandstone was probably initially restricted to certain fluvial channels. During the Cretaceous the sea level was higher, and successive episodes of high and low global sea level were manifested as regional marine transgressions and regressions. A series of palaeogeographic maps has been drawn up for various Cretaceous episodes on the basis of periodic recurrences and relative abundances of marine fossils in the palynological sequence (Figs 10-13). The maps also include data from the Carpentaria Basin, which provide evidence of periodic sea level movements during the Neocomian (see Burger, 1982<z, b). The oldest Cretaceous dinoflagellates so far known from the basin have been found in the Gilbert River Formation and Cadna-owie Formation in the north (Mossman 1, Gilberton 2, Manuka 1, see Fig. 10A). A bivalve fauna recently collected from a comparable stratigraphic position within the Gilbert

D. B U R G E R

Foraminisporis asymmetricus Subzone

Foraminisporis wonthaggiensis Subzone

SOUTH AUSTRALIA

Control Land

Great Basin

sections Artesian margin

Offshore area

Fig. 10. Distribution of land and sea in the Great Artesian Basin during the Valanginian/Hauterivian (A), and the Barremian (B), reconstructed from palynological evidence.

Lower Osmundacidites Zone

Upper Osmundacidites Zone

dubius

GULF OF

CARPENTARIA

SOUTH AUSTRALIA Sandstone

SOUTH AUSTRALIA Control Great Basin

Mudstone Offshore

dubius

GULF

sections Artesian margin

area

Fig. 11. Distribution of land and sea in the Great Artesian Basin during the early Aptian (A), and the late Aptian (B), reconstructed from palynological evidence.

PALYNOLOGY AND CYCLIC SEDIMENTATION

Fig. 12. Distribution of land and sea in the Great Artesian Basin during the late early Albian (A), and the middle Albian (B), reconstructed from palynological evidence.

River Formation in outcrop east of Croydon township may well be connected with the same marine episode (Burger, 1980c). Spinose acritarchs and sporadic dinoflagellates indicating brackish near-shore environments of deposition (according to Burger, 1980a) occur in the Cadna-owie Formation farther south (Yongala 1, GSQ Eromanga 1, Alice River 1, Wanaaring 1, Weilmoringle 1). These occurrences all fall within the Foraminisporis wonthaggiensis Subzone and indicate the first brief incursion of the sea connected with Cycle 5A during the Valanginian/Hauterivian. The sea entered the basin through a channel east of Croydon, the narrow neck of the basin being blocked by a land mass known as the Croydon-Smithburne Basement High west of Croydon (Burger, 1982a). Arms of the sea must intermittently have reached as far south as New South Wales. At the same time the sea briefly entered the Surat Basin through the eastern corridors but was probably blocked by the Nebine Arch (Burger, 1980a). A Barremian marine episode connected with Cycle 5B is indicated in Figure 10B by the presence of dinoflagellates in the Gilbert River Formation (Gilberton 2, Hughenden 1-1A), Hooray Sandstone (Aramac 1, Augathella 3, Whyenbirra 1), possibly also the basal Wallumbilla Formation in Ooroonoo 1. Brackish environments prevailed in the Cadna-owie Formation to the south (Weilmoringle 1, Wanaaring 1). Marine and brackish sandstone overlapping the Boulia Shelf ('Longsight Sandstone') may have been deposited at the same time (Burger & Mond, 1973). Simultaneous marine developments within the Bungil Formation in the western Surat Basin reinforce the notion that a marine gulf may have extended south onto the Cunnamulla Shelf, while the eastern accesses were temporarily blocked, possibly as a result of Neocomian structural activity (Burger, 1980a).

Abundant and diverse dinoflagellate assemblages in the Doncaster Member of the Wallumbilla Formation in Queensland and New South Wales, and the lower part of the Bulldog Shale in South Australia represent a marine transgression which may have extended across more than half of the continent. During an early Aptian high global sea level, the sea flooded the Eromanga Basin (Fig. 11 A). The contemporaneous Minmi Member of the Bungil Formation represents brackish-marine conditions only in the northwestern part of the Surat Basin; elsewhere it is nonmarine, suggesting that the sea was barred by the Nebine Arch (Burger, 1973a, 1980a). During a late Aptian high sea level phase the sea covered the entire Great Artesian Basin, and briefly entered the Murray Basin in New South Wales (Fig. 11B). Westerly connections probably existed with epicontinental seas in South Australia (Eucla Basin) and Western Australia (Officer and Canning Basins, see Jackson & van de Graaff, 1981; Forman & Wales, 1981). In the Northern Territory brackish-marine and nonmarine conditions existed, while the marine Bulldog Shale ('Rumbalara Shale') was deposited further south (Burger & Mond, 1973). An early Albian low global sea level (Cycle 6) has been observed as a thin interval lacking marine microfossils (basal Coreena Member), and by regressive conditions (Jones Valley Member, not shown here) in the Hughenden area (Morgan, 19806; Burger, 1982a). The Nebine Arch re-emerged as dry land, and the eastern access to the open sea was permanently blocked (Burger, 1980a). Marine conditions of Cycle 6 occurred again in the late early Albian (Figure 12A). The Coreena Member interval within the Coptospora paradoxa Zone represents brackish to nonmarine environments of deposition towards the east and south (Augathella 6, Wyandra 1, Whyenbirra 1, Eulo 2,

D. BURGER

68 Lower Phimopollenites Zone

pannosus 500 km

28 °

SOUTH AUSTRALIA

CONCLUSIONS

L Mudstone

Great Basin

Sandstone

Control

Offshore

contact between the Coreena Member and the overlying Toolebuc Formation (Ozimic, 1986), attesting to minor erosion during maximum retreat of the sea (lower Cycle 7). The Toolebuc Formation in Queensland, the Urisino Beds in New South Wales, and the correlatives of the Wooldridge Limestone Member in South Australia were reviewed by McMinn & Burger (1986) who argued that those units fall within the lower part of the Phimopollenites pannosus Zone (Figs 7, 8), and showed that the three units were deposited in a shallow epicontinental sea, which inundated most of the Eromanga Basin in the early late Albian (Fig. 13). Contemporaneous brackish-marine greensands (Coreena Member) in Bulloo 1 and Eulo 2 indicate high areas to the southeast. The sequence associated with the Phimopollenites pannosus Zone has not been preserved in the Surat Basin (Burger, 1980c). A gradual retreat of the sea during the late Albian is clearly marked by reduced relative numbers of dinoflagellates in the Mackunda Formation. The Winton Formation includes fluvial and lacustrine sands of Cycle 8. Freshwater bivalves and the absence of dinoflagellates confirm prevailing terrestrial environments, which are found also in the Normanton Formation in the Carpentaria Basin.

Artesian margin section

Margin of Toolebuc Fm and Urisino Beds

area

Fig. 13. Distribution of land and sea in the Great Artesian Basin during the late Albian, reconstructed from palynological evidence.

Weilmoringle 1, Wanaaring 1). West of approximate longitude 144° east, siltstone and mudstone of the contemporaneous Wallumbilla Formation and Bulldog Shale in Hughenden 7, Boulia 3, Urisino 1, and Oodnadatta 1 indicate quieter marine deposition in the deeper parts of the basin. A phase of low global sea level (Cycle 7) in the middle Albian (Fig. 12B) is connected with the brackish to nonmarine glauconitic 'greensand' facies of the upper Coreena Member in Queensland and New South Wales (Jericho 11, Augathella 6, Charleville 5, Wyandra 1, Toompine 1), the Coorikiana Sandstone to the west (Urisino 1, Betoota 1, Oodnadatta 1, Hay River 12), and the sandy Trimble Formation in Cape York Peninsula (not shown here). Deposition of mudstone was restricted to the northern regions (Boulia 3, GSQ Eromanga 1, Manuka 1, Hughenden 7). In many boreholes along the eastern basin margin, pebbly horizons indicate a scoured

The foregoing palynological analysis, supported by lithological and palaeontological evidence, has traced the history of part of the Eromanga Basin from initial fluvial deposition in a few river valleys during the Early Jurassic to marine deposition of maximum extent, covering a quarter of the Australian continent during the Middle Cretaceous. It has tried to (a) interpret and chronologically order geological events in different regions, and (b) reconstruct palaeoenvironments at several intervals during the Early-Middle Cretaceous, showing the extent of successive marine transgressions during global eustatic high sea level episodes. Some of the problems still awaiting solution are: (a) Jurassic and Cretaceous stratigraphy of the Northern Territory; (b) stratigraphy of the region to the west of Hughenden, where interdigitating Jurassic and Cretaceous sequences of the Eromanga and Carpentaria Basins across basement highs are still poorly understood; and (c) stratigraphy of the Cretaceous sequence overlying the Eulo, Cunnamulla, and Thargomindah Shelves in southern Queensland and northern New South Wales.

ACKNOWLEDGEMENTS The author is indebted to Drs B. E. Balme and N. F. Exon for reading the manuscript and offering critical comments. The figures were made by the B.M.R. Cartography Section. Permission for publication was granted by the Director, Bureau of Mineral Resources.

REFERENCES BALME, B. E., 1957: Spores and pollen grains from the Mesozoic of Western Australia. Aust. CSIRO Coal Res. Sect. Techn. Commun. 25.

BURGER, D., 1968: Palynology of marine Lower Cretaceous strata in the northern and eastern Eromanga Basin, Queensland. Aust., Bur. Miner. Resour., Geol. Geophys., Rec., 1968/62 (unpubl.).

BALME, B. E., 1964: The palynological record of Australian PreTertiary floras; in Cranwell, L. M. (ed.), Ancient Pacific Floras. Univ. Hawaii Press, 49-80. BOSE, M. N. & DEV, S., 1961: Studies on the fossil flora of the Jabalpur Series from the South Rewa Gondwana Basin.

BURGER, D., 1973a: Palynological zonation and sedimentary history of the Neocomian in the Great Artesian Basin, Queensland. Geol. Soc. Aust., Spec. PubL, 4, 87-118.

Palaeobotanist,

57-64.

BURGER, D., 19736: Palynological observations in the Carpentaria Basin, Queensland. Aust., Bur. Miner. Resour., Geol. Geophys., Bull., 140. 27-44.

PALYNOLOGY AND CYCLIC SEDIMENTATION BURGER, D., 1977: Palynomorphs from Eromanga Basin formations

in QDM Aramac 1 well. Qld Gov. Min. J., 78, 331-6.

BURGER, D., 1980a: Palynological studies in the Lower Cretaceous

of the Surat Basin, Australia. Aust., Bur. Miner. Resour., Geol. Geophys., Bull., 189.

BURGER, D., 1980&: Early Cretaceous (Neocomian) microplankton

from the Carpentaria Basin, northern Queensland. Alcheringa, 4, 263-79. Queensland. Aust., Bur. Miner. Resour., Geol. Geophys., Prof. Opinion, 80.022 (unpubl.). BURGER, D., 1981: Observations on the earliest angiosperm development with special reference to Australia. IV Int. Palynol. Conf., Lucknow (1976-77), 3, 4 1 8 - 2 8 . BURGER, D., 1982A: A basal Cretaceous dinoflagellate suite from north-eastern Australia. Palynology, 6, 161-92. BURGER, D., 1982b: Palynological examination of Late Mesozoic sediments in GSQ Hughenden 7, and notes on geological events in the northern Eromanga Basin. Qld Gov. Min. J. 83, 421-32. BURGER, D. & KEMP, E . M., 1972: Notes on the Carboniferous to Cretaceous palynology of the Buchanan, Muttaburra, Tangorin, and Hughenden 1:250 000 Sheet areas, Queensland. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1 9 7 2 / 9 9 (unpubl.). BURGER, D. & M O N D , A., 1973: Geological and palynological observations on the Cretaceous of the northwestern Eromanga Basin, Queensland and Northern Territory. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1973/102 (unpubl.). BURGER, D. & SENIOR, B. R., 1979: A revision of the sedimentary and palynological history of the northeastern Eromanga Basin, Queensland. Geol. Soc. Aust. J. 26, 121-32. BYRNES, J. G., 1980: BMR Urisino No. 1 bore and its bearing on prospects for oil shale in the Eromanga Basin, N.S.W. N.S.W. Geol. Surv. Rep. 1 9 8 0 / 1 7 7 (unpubl.). COOPER, M. R., 1977: Eustacy during the Cretaceous: its implications and importance. Palaeogeogr. Palaeoclirnatol. Palaeoecol. 22, 1-60. CREER, K. M. 1970: Review and interpretation of palaeomagnetic data from the Gondwanic continents. 2nd Internat. Gondw. Symp. Proc. and Pap. 55-72. DAY, R. W., 1969: The Lower Cretaceous in the Great Artesian Basin; in Campbell, K. S. W. (ed.), Stratigraphy and Palaeontology, Essays in honour of Dorothy Hill. ANU Press, Canberra, 140-73. DE JERSEY, N. J. & PATEN, R. J., 1964: Jurassic spores and pollen grains from the Surat Basin. Qld Geol. Surv. Publ. 322. DETTMANN, M. E., 1963: Upper Mesozoic microfloras from southeastern Australia. R. Soc. Vic. Proc. 77, 1-148. DETTMANN, M . E., 1973: Angiospermous pollen from Albian to Turonian sediments of eastern Australia. Geol. Soc. Aust. Spec. Publ. 4, 3-34. DETTMANN, M. E., 1981: The Cretaceous flora; in Keast, A. (ed.), Ecological Biogeography of Australia. Dr W. Junk bv Publishers, The Hague, 357-75. DETTMANN, M . E . & DOUGLAS, J. G., 1976: Palaeontology; in Douglas, J. G. & Ferguson, J. A. (eds), Geology of Victoria. Geol. Soc. Aust. Spec. Publ. 5, 1 6 4 - 9 . DETTMANN, M . E . & PLAYFORD, G., 1969: Palynology of the Australian Cretaceous: a review; in Campbell, K. S. W. (ed.), Stratigraphy and Palaeontology, Essays in honour of Dorothy Hill. ANU Press, Canberra, 174-210. EVANS, P. R., 1966a: Mesozoic stratigraphic palynology in Australia. Australas. Oil Gas J. 12(6), 5 8 - 6 3 . EVANS, P. R., 1966ft: Palynological studies in the Longreach, Jericho, Galilee, Tambo, Eddystone, and Taroom 1:250 000 Sheet areas, Queensland. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1966/61 (unpubl.). EVANS, P. R., 1966c: The palynology of Amerada Newlands No. 1 well, Queensland. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1966/186 (unpubl.). BURGER, D., 1980C: Report on field trip to Croydon,

EVANS, P. R., 1966d: Palynological comparison

of the Cooper and Galilee Basins. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1966/222 (unpubl.). EVANS, P. R., 1966c: Mesozoic stratigraphic palynology of the Otway Basin. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1966/69 (unpubl.). EVANS, P. R. & BURGER, D., 1972: Palynology of shallow stratigraphic boreholes and oil exploration wells; in Exon, N. F., Galloway, M. C., Casey, D. J. & Kirkegaard, A. G., Geology of the TamboAugathella area, Queensland. Aust. Bur. Miner. Resour. Geol. Geophys. Rep. 143, 82-101. EXON, N. F., 1976: Geology of the Surat Basin in Queensland. Aust. Bur. Miner. Resour. Geol. Geophys. Bull. 166. EXON, N. F., 1980: The stratigraphy of the Surat Basin, with special reference to coal deposits. Coal Geology, 1, 57-69. EXON, N. E & BURGER, D., 1981: Sedimentary cycles in the Surat Basin, and global changes of sea level. BMR J. Aust. Geol. Geophys. 6, 153-9. EXON, N. E & SENIOR, B. R., 1976: The Cretaceous of the Eromanga and Surat Basins. BMR J. Aust. Geol. Geophys. 1, 33-50. EXON, N . E , GALLOWAY, M . C., CASEY, D. J. & KIRKEGAARD, A . G.,

1972: Geology of the Tambo-Augathella area, Queensland. Aust. Bur. Miner. Resour. Geol. Geophys. Rep. 143.

FILATOFF, J., 1975: Jurassic palynology of the Perth Basin, Western

Australia. Palaeontogr. Abt. B, 154, 1-113. D. J. & WALES, D. W., (compilers), 1981: Geological evolution of the Canning Basin, Western Australia. Aust. Bur. Miner. Resour. Geol. Geophys. Bull. 210. FREYTAG, I. B., 1966: Proposed rock units for marine Lower Cretaceous sediments in the Oodnadatta region of the Great Artesian Basin. 5. Aust. Geol. Surv., Q. geol. Notes, 18, 3-7. GRAVESTOCK, D. I., 1982: Jurassic to Lower Cretaceous stratigraphy of the Eromanga Basin, South Australia—problems and progress in subsurface correlation: in Moore, P. S. & Mount, T. J. (compilers), Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 79-91. GRIMES, K. G., 1972: The Mesozoic and Cainozoic geology of the Cloncurry 1:250 000 Sheet area, Queensland. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1972/57 (unpubl.). H A I G , D. W., 1979: Cretaceous foraminiferal biostratigraphy of Queensland. Alcheringa, 3, 171-87. HALLAM, A., 1963: major epeirogenic and eustatic changes since the Cretaceous, and their possible relationships to crustal structure. Am. J. Sci. 261, 397-423. HIND, M. C . & HELBY, R. J., 1969: The Great Artesian Basin in New South Wales; in Packham, G. H. (ed.), The Geology of New South Wales. Geol. Soc. Aust. J. 16, 481-97. JACKSON, M . J. & VAN DE GRAAFF, W. J. E., 1981: Geology of the Officer Basin, Western Australia. Aust. Bur. Miner. Resour. Geol. Geophys. Bull. 206. LAING, A. C. M., 1966: Completion report Alliance Yongala No. 1 well, A.P. 98P, Queensland. Alliance Oil Dev. Aust. N.L. (unpubl.). LUDBROOK, N. H., 1966: Cretaceous biostratigraphy of the Great Artesian Basin in South Australia. S. Aust. Geol. Surv. Bull. 40. MCKELLAR, J. L., 1974: Jurassic miospores from the upper Evergreen Formation, Hutton Sandstone, and basal Injune Creek Group, north-eastern Surat Basin. Qld Geol. Surv. Publ. 361. M C K E L L A R , J. L., 1978: Palynostratigraphy of samples from GSQ Eddystone 1. Qld Gov. Min. J. 79, 424-34. M C K E L L A R , J. L., 1979: Palynostratigraphy of core samples from GSQ Hughenden 7. Qld Gov. Min. J. 80, 295-302. MCLACHLAN, I. R. & PIETERSE, E., 1978: Preliminary palynological results: site 361, Leg 40, Deep Sea Drilling Project. Deep Sea Drill. Proj. Initial Rep. 40, 857-81. M C M I N N , A. & BURGER, D., 1986: Palynology and palaeoenvironments of the Toolebuc Formation (senus lato) in the Eromanga Basin; in this volume. FORMAN,

D. BURGER

MOORE, P. S., 1982: Mesozoic geology of the Simpson Desert region,

northern South Australia; in Moore, P. S. & Mount, T. J. (compilers), Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 46-57. MORGAN, R., 1978: Early and Middle Cretaceouspalynostratigraphy of Australia. Ph.D. thesis, Univ. Adelaide (unpubl.). MORGAN, R., 1980a: Palynostratigraphy of the Australian Early and Middle Cretaceous. N.S.W. Geol. Surv. Mem. Palaeontol. 18. MORGAN, R., 1980&: Eustacy in the Australian Early and Middle Cretaceous. N.S.W. Geol. Surv. Bull. 27. NORVICK, M. S. & BURGER, D., 1975: Palynology of the Cenomanian of Bathurst Island, Northern Territory, Australia. Aust. Bur. Miner. Resour. Geol. Geophys. Bull. 151. OZIMIC, S., 1986: The geology and petrophysics of the Toolebuc Formation and its time equivalents, Eromanga and Carpentaria Basins; in This volume. PESSAGNO, E. A., 1972: Pulsations, interpulsations and sea-floor spreading. Geol. Soc. Am. Mem. 132, 67-73. REISER, R . E & WILLIAMS, A. J., 1969: Palynology of the Lower Jurassic sediments of the northern Surat Basin, Queensland. Qld Geol. Surv. Publ. 399. RONA, P. A., 1973: Relations between rates of sediment accumulation on continental shelves, sea-floor spreading, and eustacy inferred from the central North Atlantic. Geol. Soc. Am. Bull. 84,

SMART, J., 1976: Stratigraphic correlations between the older units

of the southern Carpentaria and northern Eromanga Basins. Qld Gov. Min. J. 77, 171-8.

SMART, J., GRIMES, K. G., DOUTCH, H . F. & PINCHIN, J., 1980: T h e

Mesozoic Carpentaria Basin and the Cainozoic Karumba Basin, north Queensland. Aust. Bur. Miner. Resour. Geol. Geophys. Bull. 202. TERPSTRA, G. R. J. & BURGER, D., 1969: Micropalaeontology and palynology of samples from BMR Bulloo No. 1 Scout Hole. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1969/39 (unpubl.). VACHRAMEEV, V. A . , 1964: Jurassic and Early Cretaceous floras of Eurasia and the palaeofloristic provinces of this period. Geol. Inst. Acad. Sci. USSR, Trans. 102. VAIL, P. R. & TODD, R. G., 1981: Northern North Sea Jurassic unconformities, chronostratigraphy and sea-level changes from seismic stratigraphy; in Cling, J. V. & Hobson, C. D. (eds), Petroleum geology of Continental Shelf of north-west Europe. Institute of Petroleum, London, 216-35. VAIL, P. R., MITCHUM, R . M . & THOMSON, S., 1977: Seismic stratigraphy and global changes of sea level, Part 4: Global cycles of relative changes of sea level; in Payton, C. E. (ed.), Seismic stratigraphy—applications to hydrocarbon exploration. Am. Assoc. Pet. Geol. Mem. 26, 8 3 - 9 7 . VAN HINTE, J. E., 1976A: A Jurassic time scale. Am. Assoc. Pet. Geol. 2851-72. Bull. 60, 4 8 9 - 9 7 . RYAN, J. C., 1961: Innamincka No. 1 well, South Australia. Aust. Bur. VAN H I N T E , J. E., 19766: A Cretaceous time scale. Am. Assoc. Pet. Miner. Resour. Geol. Geophys. Aust. Petrol. Search Subsidy Geol. Bull. 60, 4 9 8 - 5 1 6 . Acts, Publ. 9. VINE, R. R., 1970: Explanatory notes, Richmond 1:250 000 SENIOR, B. R., EXON, N. F. & BURGER, D., 1975: The Cadna-owie geological map, sheet SF54-4. Aust. Bur. Miner. Resour. Geol. and Toolebuc Formations in the Eromanga Basin, Queensland. Geophys. Qld Gov. Min. J. 76, 4 4 5 - 5 5 . VINE, R. R. & PAINE, A. G. L., 1974: Explanatory notes, Hughenden SENIOR, B. R., INGRAM, J. A . , THOMAS, B. M . & SENIOR, D., 1969: 1:250 000 geological map, sheet SF55-1. Aust. Bur. Miner. The geology of the Quilpie, Charleville, Toompine, Wyandra, Resour. Geol. Geophys. and Cunnamulla 1:250 000 Sheet areas, Queensland. Aust. Bur. W I L T S H I R E , M. J., 1982: Late Triassic and Early Jurassic Miner. Resour. Geol. Geophys. Rec. 1969/13 (unpubl.). sedimentation in the Great Artesian Basin: in Moore, P. S. & Mount, T. J. (compilers), Eromanga Basin Symposium, SENIOR, B. R„ M O N D , A. & HARRISON, P. L., 1978: Geology of the summary.papers.. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Eromanga Basin. Aust. Bur Miner. Resour. Geol. Geophys. Adelaide, 68-75. Bull. 167.

Geological Society of Australia Special Publication No. 12, 71-84

The geology and hydrocarbon potential of the Murta Member (Mooga Formation) in the southern Eromanga Basin G. Ambrose1, R. Suttill2 & I. Layering3 1 2 3

Lasmo Energy (Aust) Ltd, G.P.O. Box 976, Brisbane, Qld 4001. S.A. Oil and Gas Corp., 60 Hindmarsh Square, Adelaide, S.A. 5000. Esso Australia Ltd, 127 Kent Street, Sydney, N.S.W. 2000; present address: Bureau of Mineral Resources, P.O. Box 378, Canberra, A.C.T. 2601.

ABSTRACT In the southern Eromanga Basin the Murta Member of the Mooga Formation comprises a mainly fine-grained lacustrine sequence intervening between braided-fluvial sediments of the Namur Sandstone Member and the overlying, marginal marine Transition beds'. The sequence intertongues with the Namur Sandstone Member on a regional scale and the upper contact with the 'Transition beds' is gradational. Regional reduction in thickness and sand content from the north-northeast to the southwest reflects a depositional pattern whereby the 'Namur' braided fluvial regime contributed large volumes of sediment to the 'Murta' lake on the northern and eastern lake margins. The Jackson Field oil discovery occurs in this area and sandy proximal delta and sub-lacustrine fan sequences have been cored there. The sequence prograded towards the south-southwest and fine-grained sediment and occasional coarse-grained sands were transported large distances into the lake by density flows (i.e. turbidity currents). The Dullingari Field, which is a structural-stratigraphic trap, occurs in this setting. The main reservoir is an elongate shoreline or bar sand; turbidite flows and delta distributaries provide secondary reservoirs. Considerable potential remains for additional oil discoveries in the Murta Member. Exinite-rich shales, which are the source for Dullingari and Jackson oils, are widespread and reservoir quality sands are relatively abundant, especially adjacent to the northern and northeastern lake margins. Overall the style of sedimentation lends itself to stratigraphic entrapment of hydrocarbons, but at this stage of exploration combined structural-stratigraphic plays of the Dullingari type are the most viable targets.

INTRODUCTION This paper describes the regional geology of the Murta Member, Mooga Formation, in the area of the southern Eromanga Basin (Fig. 1). This is a mainly fine grained, Late Jurassic-Early Cretaceous sequence intervening between terrestrial sediments of the Namur Sandstone Member (Late Jurassic-Early Cretaceous) and the overlying marginal marine 'Transition beds' (Early Cretaceous). Nugent (1969) described the Murta Member as a lacustrine facies of the Mooga Formation and inferred lateral interfingering with fluvial sandstones of the underlying Namur Sandstone Member. From that time the Murta Member attracted little attention until 1979 when oil was discovered in Dullingari North 1. A regional depositional model for the Murta Member, including a detailed lithofacies interpretation for the sequence at Dullingari, was undertaken by Ambrose (1980). Mount (1981tf, b, 1982) also discussed aspects of the Dullingari discovery and described several alternatives for the origin of the 'Murta' depositional facies in this area. In 1981 the discovery of oil in Jackson 1 in Queensland further enhanced the prospectivity of the unit.

As a result of these discoveries and an expanded regional drilling programme there has accumulated a considerable amount of new information on the Murta Member and a review is appropriate. This paper includes a detailed study of the Murta lithofacies and oil occurrences in the Dullingari and Jackson Fields, largely based on core data, and a regional interpretation drawn mainly from electric log correlations. In addition, reference sections are described for the sequence in the Dullingari and Jackson Fields.

PALYNOLOGY AND STRATIGRAPHY The Murta Member and 'Transition beds' are represented by spores of the Murospora florida Zone which are mainly Neocomian in age. Burger (1973) subdivided this zone into the Cicatricosisporites australiensis, Foraminisporis wonthaggiensis and F. asymmetricus Subzones (ascending order). Only the lowest subzone is represented in the Murta Member. Price (1983) suggests the lower C. australiensis Subzone is confined to the Namur Sandstone Member, whereas the upper part of the subzone embraces sediments of the Murta Member and basal 'Transition beds'. Although

G. AMBROSE, R. SUTTILL & I. LAYERING

Fig. 1. Location Murta Member study area, southern Eromanga Basin.

there is no clear evidence of depositional environment, the presence of the fresh-water alga Botryococcus, the general absence of morphological diversity in recorded dinoflagellate assemblages (Price in Lawrence & Holland, 1980) and the absence of marine fossils suggest non-marine conditions. The first indications of marine influence are in the F. wonthaggiensis Subzone flora in the middle part of the 'Transition beds' and numbers of dinoflagellates and acritarchs increase in the younger F. asymmetricus Subzone. Core studies indicate a gradational upward passage from the Murta Member into the basal Transition beds'. The Murta Member intertongues with the Namur Sandstone Member locally (e.g. over the Gidgealpa Field) and also on a regional scale particularly on the northern, northwestern and eastern margins of the basin. It is thus proposed that the unit retain its status as a Member of the Mooga Formation. Stratigraphic nomenclature adopted herein is shown in Figure 2.

DULLINGARI FIELD Lit

hofades

Lithofacies vary considerably on a regional scale, hence the Jackson and Dullingari Fields are discussed separately. At Dullingari the Murta Member consists of several laterally continuous sedimentary cycles which are readily correlated between wells. A considerable thickness of Murta Member is cored in Dullingari 9 and it is proposed that a reference section be assigned to this well. The reference section description (Fig. 3) draws on data from other wells (e.g. neighbouring Burke Field) for those parts of the section which were either not cored or where core recovery was poor.

The Namur Sandstone Member generally consists of gritty, poorly sorted, tabular cross-bedded sandstone deposited in a braided fluvial environment (Nugent, 1969). The upward passage into the Murta Member is commonly marked by 3-15 m (10-50 ft) of ripple-laminated and bioturbated siltstone and thin sandstones. The base of the Murta Member is defined by the last of this sandstone-siltstone sequence which is a low energy fluvial-floodplain facies marking gradual degradation of the 'Namur' braided-fluvial regime. The Murta Member is subdivided into four units (Fig. 3). Unit 1 An upward-coarsening sequence dominated in the lower part by laminated to thinly bedded argillaceous siltstones. Laminated to thinly bedded, poorly sorted silty sandstones, often cemented by calcite and siderite, become more common up through the sequence. The following features suggest these beds were deposited on a slope by sediment-laden density currents: (1) sharp based, graded beds display cut-and-fill and ABE Bouma sequences (Bouma, 1962; Fig. 4a). (2) recumbent slump folds and sedimentary dips of up to 15° are common; (3) 'streak out' ripples observed in some beds are attributed to frictional drag of silt- and sand-laden currents on soft but cohesive sediments of underlying beds; (4) small scale deformation features abound (e.g. microfaults, water escape structures and load casts) reflecting relatively rapid sedimentation and subsequent loading. These sediments were deposited on a gently shelving lake floor, below wave base, by currents bearing mainly fine sand and silt which are in places rippled indicating current velocities

MURTA MEMBER GEOLOGY AGE

PEDIRKA BASIN REGION

RECENT TO LATE CRETACEOUS

SOUTHERN COOPER BASIN REGION Su p e r f i c i a l

Toolebuc

Formation

Subgroup

Mudstone

Wallumbilla

Formation

Marree

Allaru

EARLY CRETACEOUS

Deposits

and Winton

Mackunda

Formation

NORTHERN COOPER BASIN REGION

Fornlation

Allaru Mudstone Oodnadatta Toolebuc

Formation

Coorikiana Sst. Bulldog

Formation

Wallumbilla

Formation

Shale

Cadna-owie Formation Transition

Mt. Anna Sandstone Member (Restricted To S.W. Basin Margin

Algebuckina Sandstone JURASSIC

Poolowanna

Mooga

Fm.

Murta Namur Sandstone Member

Formation

Westbourne Adori Birkhead

Formation

Hutton

Sandstone

basal PERMIAN TRIASSIC

Pedirka

Basin

Sediments

beds Member

Cooper

Formation

Sandstone

Jurassic

Basin

Sediments

Fig. 2. Regional nomenclature, southern Eromanga Basin.

Of 10-20 cm s e c (Jopling & Walker, 1968). Some of the thicker flows may have resulted from catastrophic events such as floods. Thin, graded silt laminae were deposited by low energy flows, possibly generated seasonally by periods of high rainfall. Ungraded argillaceous beds are pelagic in origin, deposited by either density overflows (surface currents) or interflows (thermocline undercurrents) (see Fig. 12). Bioturbation is ubiquitous and trails and burrows, probably formed by soft-bodied worms, are often infilled with later sediment thus preserving them on the soles of beds. Graded beds generally increase in thickness and sand content towards the top of Unit 1 culminating in the development of the '49-0' sand (Fig. 3). This sand consists of up to 9 m of very fine to occasionally medium-grained, micaceous, silty sandstone often containing plant fragments and other carbonaceous material in the upper part. Carbonaceous ripple laminae are characteristic (Fig. 4b) and include 'type A' climbing ripples (Jopling & Walker, 1968) reflecting a combination of relatively high flow velocity, coarse grain size and a high rate of bedload to suspended load; deposition on a proximal delta slope is implied. Delta distributary channels, averaging 2-3 m in thickness, are recognised in a number of wells by a diagnostic 'blocky' gamma-ray log signature. This facies was cored in Nappacoongee 2 and consists of ripple-laminated, fine- to medium-grained sand with a sharp base and gradational upper contact. In summary, Unit 1 is a progradational lacustrine-delta sequence formed in a low-energy environment where the lack of tides, effective waves, or wind-generated currents has done little to redistribute deltaic sediment. Under such conditions a highly constructive, lobate or possibly elongate delta would -1

form, with lateral shifting of delta distributaries resulting in the lakeward construction of the delta front by overlapping lobes of sediment. The distal delta deposits comprise mainly thin, graded beds deposited by low-energy density flows, complemented by minor pelagic deposits. This is characteristic of deposition in an open lake where the absence of a permanent hypolimnion (i.e. density stratification) reduces the importance of pelagic sedimentation. Instead, due to the density contrast between lake water and sediment-laden river water, deposition via density underflows predominates with the coarsest sediment load deposited on the proximal delta slope. This depositional model closely resembles that adopted by Sturm & Matter (1978) for sedimentation in Lake Brienz, Switzerland.

Unit 2

A sequence of siltstones and fine sandstones similar to the middle-upper part of Unit 1. Graded beds are characteristic and sometimes form thin upward-coarsening sets. Prolific bioturbation in this unit and the occasional presence of wave ripples and coal laminae indicate deposition in a nearshore, shallow-water environment.

Unit 3

An upward-coarsening sequence, 6-9 m thick, made up of graded beds which increase in sand content and thickness upwards through the unit. A prograding sub-lacustrine fan is interpreted. Distal sediments at the base of the fan sequence consist of laminated dark silty shale with thin intercalations of silty, fine-grained sandstone (Fig. 4c). There is an upward passage into graded siltstone beds, 5-20 cm thick, which contain BCE and BCDE Bouma sequences (Fig. 4d). Sandy, proximal fan sediments at the top of the cycle comprise several thin sand bodies, 10-50 cm thick, separated by thin siltstones

G. AMBROSE, R. SUTTILL & I. LAVERING DULLINGARI 9

JACKSON

basal TRANSITION BEDS

basal TRANSITION BEDS deep lacustrine?

TOP MURTA MEMBER fining upward lacustrine sequence

upward coarsening lacustrine sequence proximal fan deposits capped by delta front sands

bay-marsh facies

overbank splay facies?

delta front DST3 11311 Flow 1600 BWPD 1140m

TOP NAMUR SANDSTONE MEMBER

braided fluvial

LEGEND

m m m

SAND SILT SHALE BIOTURBATION SLUMPED CONVOLUTED

/ZV7T

RIPPLE

WWW

TABULAR

PLANAR

BEDDING

CROSS-LAMINATION CROSS-BEDDING BEDDING

CLIMBING RIPPLES h

ROOTS

Fig. 3. Murta Member lithofacies, Dullingari 9, Jackson 1.

(Fig. 4e). These sands, which are collectively referred to as the '48-6' sand (Fig. 3), have sharp basal contacts often paved with flat, ovate shale intraclasts a few mm in size. Lateral continuity of individual flows is limited and erosion and amalgamation of individual beds are common (Fig. 4f). Current lineations and flute casts occur on bedding planes and abbreviated Bouma sequences containing A, B and C subdivisions are characteristic. The presence of occasional sandstone and shale intraclasts, up to 15 cm in size, testifies to a relatively high energy environment for the upper fan

sequence. Secondary upward-coarsening cycles, superimposed on the general trend, represent stacked fan lobes which were abandoned as density currents shifted laterally across the lake floor. The fan sequence differs from the underlying delta front deposits in Unit 1 (i.e. the '49-0' sand) in that the former result from intermittent influx of sediment carried by decelerating currents. The delta front sands are believed to have been deposited by a fairly continuous density underflow. On a regional basis the '48-6' fan sequence can be correlated using gamma-ray logs over an area of at least

MURTA MEMBER GEOLOGY 5000 km (Fig. 6). This broad, uniform fan development indicates a lake floor with very little topographic relief. A gradation to silty, distal sediments to the southwest suggests a primary source on the north-northeast margin of the lake. The basinward progradation of the '48-6' fan sequence was basically a shallowing trend which culminated locally in the development of a thin shoreline or bar sand ('48-5' sand, Fig. 3). The sand is up to 2 m thick and forms an elongate belt running northwest from Dullingari through Corkwood and possibly on to Coonatie (Figs 5, 6). Lithology generally varies from moderately well sorted fine- to medium-grained sand at the base to poorly sorted, very coarse-grained, gritty sand at the top; massive and flat bedding are common. Sonic logs indicate that effective porosity is concentrated in the lower portion of the sand while the upper coarse-grained part of the unit is strongly cemented and generally tight. There is no evidence of emergence, such as root traces or soil zones, indicating either erosion of subaerial facies or deposition at or just below low-water mark. The latter seems more likely and hence the '48-5' sand was probably deposited above wave base but below low water mark and may, in fact, have formed some distance offshore. On a regional scale the sand forms an elongate trend straddling the '48-6' fan sequence but its exact extent and dimensions are not fully defined (Fig. 6). As a result of the doubt surrounding the exact origin of '48-5' sand, it will be hereafter referred to as a shoreline or bar sand. Unit 4 This unit marks a return to offshore sedimentation as the lake transgressed the '48-5' shoreline-bar sand. Graded siltsand beds predominate and AE Bouma sequences, up to 50 cm thick at the base of the unit, grade upwards into silty CE Bouma sequences less than 2 cm in thickness. Bed thickness and grain size decrease gradually upwards into the basal 'Transition beds' which are characterised by thin graded siltstones in which the lower contacts are often paved with flat, ovate shale intraclasts. This is a low energy facies deposited in deeper water, prior to flooding of the basin by the sea. There is no evidence of marine influence in the basal 'Transition beds' suggesting the effects of the marine transgression only became apparent higher in the sequence. A thin Murta Member siltstone sequence, probably equivalent to Unit 4, blankets the Namur Sandstone Member on the basin margins. These sediments were deposited over a wide area as the Murta lake level rose prior to the marine transgression which came from the north via the Carpentaria Basin. 2

understory and a ground cover of ferns. Lycopods and some ferns may have formed in low-lying swamps which may have also included reed-like plants. A temperate moist climate is implied. The organic matter in Units 2 and 3 occurs largely in thin beds deposited by density flows and hence the composition and abundance of organic matter is variable. The D.O.M. contents vary up to about 3 per cent with some layers rich in transported vitrinite and inertinite, while exinite (mainly alginite and sporinite) is locally abundant. Source rocks in Unit 4 display similar variability but inertinite is predominant perhaps reflecting a relatively oxygenate'd environment. This trend continues into the basal 'Transition beds' where D.O.M. contents decrease to about 1.5 per cent and inertinite is again the dominant maceral. Reflectance measurements of vitrinite, from a number of Dullingari wells, range from 0.60-0.66 per cent indicating sufficient maturity for the generation of liquid hydrocarbons.

Reservoir rocks

There are three main reservoir sands in the Murta Member at Dullingari, namely the '49-0', '48-6' and '48-5' sands (Fig. 3). Consistent differences in the quality of these reservoirs may be largely attributed to variations in depositional environment (Ambrose, 1982). (1) The '49-0' sand was deposited on a proximal delta slope, largely below wave base, and is generally fine-grained to occasionally medium-grained with no evidence of reworking. The sand appears to be oil saturated over much of the field but calcite cementation is common and permeabilities are low (generally less than 0.5 millidarcys). Distributary channel sands provide better quality reservoirs and this facies variant has produced small oil flows (less than 100 barrels of oil per day— BOPD) in Dullingari 18, 31 and 32. This reservoir may be stimulated at some stage during production of the field. (2) The '48-6' sand is very fine- to coarse-grained and variably cemented by quartz and calcite. Core analysis indicates occasionally high porosities but permeabilities are consistently low (generally less than 0.5 md and rarely up to 1.0 md). The reservoir consists of several thin sand flows, interpreted as proximal turbidites deposited at the head of a lacustrine fan. These beds are irregularly eroded and amalgamated and this, combined with the variability of secondary cementation, results in erratic reservoir performance. Pressure studies indicate the '48-6' sand is contributing to oil flows in some wells, but permeability is Source rocks The source of the oil at Dullingari was from shaly units streaky and unpredictable. (3) The '48-5' shoreline-bar sand is by far the most in the Murta Member, the best source rocks occurring in the lower part of Unit 1. The total dispersed organic matter favourable reservoir in the field. Prolonged winnowing has (D.O.M.) exceeds 2.8 per cent for this part of the sequence produced a thin but highly porous sand which reaches in Dullingari North I (Mclntyre, 1980). Exinite is the dominant permeabilities of up to 3 darcys in the lower part. Pay zones maceral and consists of mainly sporinite and alginite, of less than one metre have flowed at rates exceeding 2000 the latter reaching a maximum of 25 per cent of the total BOPD testifying to the quality of this reservoir. It is considered D.O.M. Resinite and cutinite are minor components. A that quartz-calcite cements, which are particularly prominent relatively low energy, reducing environment is implied which in the upper, coarse-grained portion of the sand formed prior is consistent with a distal deltaic facies. These sediments to migration of oil into the reservoir. Petrographic studies of the '48-5' sand show that primary contrast with argillaceous siltstones from the upper Namur Sandstone Member which contain mainly vitrinite formed in porosity has been eliminated as a result of diagenesis. All the an oxygen-rich, fluvial environment. D.O.M. contents generally effective porosity within the '48-5' sand is of secondary origin. decrease upwards through Unit 1, and vitrinite and inertinite The earliest diagenetic event following burial, is compaction dominate the sandy, proximal delta facies at the top of the of the sediments accompanied by development of quartz unit. Spore assemblages from the upper part of Unit 1 reflect overgrowths on detrital quartz grains, overgrowths sometimes significant input from forest (gymnosperms/tree ferns) interfingering. The net effect is a constriction of many of the vegetation. Price (1983) envisages the parental vegetation of pore throats. This phase of diagenesis is most marked in the the palynmorphs as open coniferalean forest with a tree fern clean, well-sorted parts of the sand. In the silty, matrix-

G. AMBROSE, R. SUTTILL & I. LAYERING

Fig 4a

Fig 4c

Fig 4b

Fig 4d

•2

•3

Fig 4e Fig. 4. (a) turbidite, middle-upper part of Unit 1, Dullingari 9. Note graded bedding and flame structures at base. Depth 1503.2 m. (b) delta-front sand ('49-0' sand) top of Unit 1, Dullingari 9. Depth 1500.5 m. (c) slumped silty shales, base of Unit 3; note recumbent slump folds and abundant microfaulting (photograph from Burke 2 core). Depth 1537.1 m. (d) vertically stacked, graded fine sand-silt turbidites, middle part of Unit 3 (photograph from Burke 2 core). Depth 1536.1 m. (e) graded sandy turbidite beds ('48-6' sand) containing A, B and C Bouma sequences (photograph from Burke 2 core). Depth 1533.1 m. (f) amalgamated turbidite sand flows at top of '48-6' sand. Line of amalgamation marked by pseudo shale clasts which are remnants formed by erosive effect of successive sand flows (photograph from Burke 2 core). Depth 1533.4 m.

supported sediment at the base of the '48-5' sand, little quartz cementation has occurred. The next major stage of diagenesis is the introduction of a carbonate cement (calcite) which filled all remaining primary porosity. Calcite cement has in part replaced the margins of quartz grains and their overgrowths. Detrital feldspar grains also show partial or total replacement and alteration by kaolinite booklets and by calcite. An earlier, but minor, event is the formation of kaolinite booklets in intergranular pores. The current excellent reservoir properties of the '48-5' sand result from dissolution of much of the carbonate cement, and creation of secondary porosity. The secondary porosity is primarily intergranular, mimicking the residual primary porosity which existed after the development of quartz overgrowths. Some of the pores and importantly the pore

Fig 4f throats are increased in size because of dissolution of the replacive calcite, leading to the formation of oversized pores (where calcite which replaced feldspars has been dissolved), and also as a result of the formation of corroded margins to quartz grains. The introduction of hydrocarbons into secondary pores then followed. Prediction of the extent and location of carbonate cement is difficult as no clear pattern is evident at this stage. As a result, one or two wells have encountered a thick '48-5' sand section but have given restricted oil flows due to cementation by calcite.

Oil entrapment

The mechanism of oil entrapment at Dullingari is complex and includes structural, stratigraphic and diagenetic components. The '48-5' shoreline-bar sand is, at least in part, a stratigraphic trap considering that a zero sand edge has been established for this reservoir on the western and southern margins of the field (Fig. 5). Clearly, local structure has partly controlled shoreline geometry and more importantly, porous sand development. However, isopach mapping of sub-units within the Murta Member shows no evidence of major structural growth during deposition, but subtle upward movements would have allowed sufficient winnowing to form the thin pay sands. In addition, as previously described, diagenesis has played an important role in restricting reservoir quality over the field.

JACKSON FIELD Lithofacies

Jackson 1 is chosen as the reference section in this area and core data from Jackson wells 1, 2 and 3 are incorporated

G. AMBROSE, R. SUTTILL & I. LAYERING in the lithofacies description. The sediments here contain a higher proportion of silt and sand than at Dullingari reflecting proximity to source areas on the lake's eastern margin. The sequence is subdivided into three units, each denoting a separate lithofacies (Fig. 3). These units are not intended as direct correlatives of the units at Dullingari. Unit 1 An upward-coarsening sequence of which only the upper portion has been cored. The lower part of the unit is interpreted, from the gamma-ray log, as a fine-grained

136°

135°

137°

138°

lacustrine sequence similar to that in the lower Murta Member at Dullingari. There is a gradation upwards into silty sandstones which are interpreted as proximal delta deposits, probably deposited below wave base by density currents. Bioturbated, ripple-laminated, micaceous, silty sandstones cored near the top of the unit are probably distributary mouth bar deposits. Overlying silty sands, which display abundant roots in the upper part, are interpreted as overbank splay or possibly channel-fill facies. This facies is succeeded by shallow marsh and bay deposits which comprise mainly siltstone containing occasional rootlets and minor coal laminae. Unit 1 thus represents a prograding lacustrine delta sequence culminating at the top in shallow marsh and bay deposits. Unit 2 This unit is similar to the '48-6' fan sequence at Dullingari in that it has a broad upward-coarsening gamma-ray log motif on which are superimposed secondary upward-coarsening cycles. The latter are capped by planar-bedded, calcite cemented conglomerate containing intraformational sandstone clasts and large, woody plant fragments. Graded beds up to 50 cm thick are common and contain mainly BCD and BDE Bouma sequences. Intervals of fine, silty sand near the top of the unit are heavily bioturbated and contain abundant, wispy carbonaceous material. The environment is one of upward shallowing, deltaic sedimentation, and the reservoir sand capping this sequence may have been modified by wave action. It is proposed that the braided fluvial 'Namur' regime fed relatively coarse

[ 2 3 PROXIMAL LACUSTRINE

F ^ l INTERM. LACUSTRINE

COOPER

THIN BLANKET OF MURTA SILTS

y O O O O O O ft O O O O O O O O f

-tt—o--©—c o oT / o o o o o o o o o o o o. o . o ~ u cofr coo uo uo U j U

o o o o o b o ^ r o o o oBASIN DM /Jss/ o o o o o oj^^o o o o o o o 0 0 0 0

48-5'SHORELINEBAR SAND

PEDIRKA

143°

A.T.P. 2 5 9 P.

142°

141°

i E^j DISTAL LACUSTRINE TRANSITIONAL ZONE TO _ BRAIDED FLUVIAL S.A. RE.L. 5 / 6

140°

139°

—I

I BRAIDED FLUVIAL

S c S r S o o o o o o^fT o° 9— o o o o o o o o o " " o o o o o o o o o y o o o o o cAf o o o o PS^-O^ - o o0 o

•D

BASIN N

ARCKARINGA

\ BASIN

135°

1360

\ 137°

138°

139°

140°

141°

Fig. 6. Generalised regional lithofacies interpretation, M u r t a Member, southern E r o m a n g a Basin.

142°

143°

MURTA MEMBER GEOLOGY detritus into the 'Murta' lake from the east and, in response, fan-deltas and deltas built basinward, prograding towards the south and west (Figs 6, 12).

Unit 3 This sequence marks a return to mainly fine-grained sedimentation and is tentatively correlated with Unit 4 at Dullingari. Graded beds are characteristic and these generally decrease in thickness and sand content upwards. Small and large scale slumps are common, reflecting deposition by density currents on a slope with the sequence building basinward by a process of lateral accretion. One feature commonly seen in these sediments is thin, ripple-laminated silt to fine sand beds consisting of either a row of ripple crests sitting on an individual lamination or single ripple crests draped by silty shale. The beds have sharp bases and the ripples are asymmetrical and apparently unidirectional, suggesting they formed at the thin edge or tail of a density flow. The upper contact with the Transition beds' is gradational and, as seen at Dullingari, it is marked by an upward decrease in bed thickness and grain size.

Source rocks Available source rock data indicate a variety of source material. The lower shaly part of Unit 1 has not been sampled

but comparison with the Dullingari area indicates the possibility of alginite-rich source rocks in this part of the sequence. The marsh-bay facies at the top of Unit 1 contains up to 3.5 per cent D.O.M. and, as expected in this environment, vitrinite is dominant over exinite and inertinite. Silty shales in Unit 2 contain about 3.0 per cent D.O.M. with vitrinite dominant over exinite. As at Dullingari, the basal Transition beds' and upper Murta Member contain an average D.O.M. of 1.5-2.0 per cent with inertinite the dominant maceral. Vitrinite and exinite are minor components and the latter comprises mainly alginite with minor sporinite and cutinite. Reflectance measurements of vitrinite show the Murta Member to be marginally less mature than at Dullingari (reflectance varies from 0.49-0.55%) but still capable of generating liquid hydrocarbons.

Reservoir rocks Reservoir rocks are relatively abundant in the Murta Member at Jackson reflecting the high sand content of the sequence, a relatively shallow depth of burial and limited diagenetic effects. Sands in the upper part of Unit 2 flowed 338 BOPD in Jackson 1 (Fig. 3), and permeabilities average 48 md. The permeability of thin conglomerate beds in Unit 2 is generally inhibited by quartz-calcite cements. Deltaic sands at the top of Unit I have an average permeability of 60 md but are water-saturated.

G. AMBROSE, R. SUTTILL & I. LAYERING

REGIONAL RELATIONSHIPS

The Murta Member was deposited in a broad, asymmetric basin formed by compaction of the underlying Cooper Basin sequence; the main depocentre lies adjacent to the northeast margin of the underlying Cooper Basin. Salient features of the regional geology are summarised by three cross-sections (Figs 7-9). (1) Figure 7 is a southwest-northeast section running from Dullingari 22 to Ingella 1. The Murta Member increases in thickness and sand content from Dullingari northeast towards the Tartulla-Wackett area, but is reduced to a thin siltstone facies in Ingella 1. It is significant that the stratigraphic interval between the top Transition beds' and top Westbourne Formation is approximately constant in thickness between Tartulla 1 and Ingella 1, and thinning of the Murta Member to the northeast is attributed to lateral interfingering with the Namur Sandstone Member. (2) Figure 8 is a north-south section from Morney 1, across the Merrimelia-Innamincka 'high', to Cherri 1 near the basin's southern margin. A relatively thick, sandy Murta Member in the Beanbush-Paning area resembles that in the TartullaWackett region. The sequence is reduced to a thin siltstone facies to the north in Curalle 1 and Morney 1 as a result of interfingering with the Namur Sandstone Member. To the south, in Cherri 1, the unit is relatively thin and silty and although there is probably some minor interfingering with

Fig. 8. Stratigraphic cross-section, Cherri 1-Morney 1.

the Namur Sandstone Member near the southern basin margin, this is difficult to verify by log correlations. An anomalously thin upper Murta Member section in Innamincka 3 suggests minor structural growth during deposition in that area. (3) Figure 9 traces the Murta Member from the area of the Cooper Basin across the Birdsville Track Ridge, to the Poolowanna Trough. The '48-5' shoreline-bar sand at Dullingari can be traced northwest to Corkwood 1 and may correlate with a similar sand in Coonatie 2, which is similar in both log signature and stratigraphic position indicating either: (a) the presence of a continuous thin shoreline-bar sand between Dullingari and Coonatie, or (b) isolated development of winnowed sands formed in response to local structural movements. On the northwest margin of the Cooper Basin interfingering of the Murta Member with the Namur Sandstone Member is demonstrated between Cuttapirrie 1 and Coongie 1. To the west, Early Cretaceous sediments thin over the Birdsville Track Ridge reflecting structural growth during this period, probably via compaction of thicker sequences deposited to the east. Across the Birdsville Track Ridge, in the Poolowanna Trough (Simpson Desert region), the Murta Member is relatively thin and silty.

MURTA MEMBER GEOLOGY

/ // /# // // // ^ /

Fig. 9. Stratigraphic cross-section, Poolowanna 1-Dullingari 22.

Regional isopach and sand: shale ratio maps highlight the general asymmetry of the 'Murta' lake setting and show a gradual trend of decreasing thickness and sand content from the north and east to the southwest (Figs 10, 11). This trend is accommodated in a model for regional deposition whereby the 'Namur' braided fluvial regime contributed large volumes of sediment to the 'Murta' lake on the northern, northwestern and eastern basin margins (Figs 6, 12). The coarsest sediment load was deposited on fan-deltas, deltas and proximal sublacustrine fans (e.g. as seen in Jackson cores) which prograded to the south-southwest. Finer-grained sediment and occasional coarse-grained sands were transported large distances into the basin by density currents traversing a gently shelving lake floor. The '48-6' fan at Dullingari is an example. Although sediment input occurred to varying degrees around the circumference of the basin, the southern and southwestern basin margins were passive with relatively little sediment input. Basal units of the Eromanga Basin sequence outcrop along the southwestern margin of the basin. Upper units in the Algebuckina Sandstone (Namur Sandstone Member equivalent) are interpreted by Wopfner et al. (1970) as lacustrine bars and shoreline deposits which may represent marginal facies equivalents of the Murta Member. A regional disconformity separates this unit and the overlying Cadnaowie Formation (Transition beds' equivalent) on the southwestern basin margin (Ambrose & Flint, 1982). On at least part of the southern basin margin this break in sedimentation appears to be associated with deep weathering on a low relief landscape. Although there was some erosion, this margin contributed relatively little sediment to the north. It is unlikely the Mount Anna Sandstone Member (Wopfner

et al., 1970), a fan-delta facies of the Cadna-owie Formation, contributed sediment to the 'Murta' lake from the west since it is not encountered in the subsurface east of the Muloorina Ridge (Fig. 6).

HYDROCARBON PROSPECTIVITY

There is considerable potential for additional oil discoveries in the Murta Member and the regional depositional model outlined herein aids in delineating the more prospective areas. With reference to this model (Fig. 6) the hydrocarbon potential of the unit is summarised as follows: (1) Distal lacustrine facies: this area contains adequate source rocks but reservoir sands are sparse. Thin, distal turbidite sands within the sequence are possible reservoirs and shoreline facies may exist around the southern lake margin. Overall, however, the prospectivity of the unit in this area is regarded as moderate to poor. (2) Intermediate lacustrine facies: the Dullingari Field occurs in this area which is regarded as highly prospective. Good oil shows have been recorded in a number of wells (e.g. Tirrawarra and Merrimelia Fields) but where tested the sands have proved to be tight. Exinite-rich source rocks are relatively plentiful and although reservoir quality sands are not abundant, a number of possibilities exist. For example, shoreline-bar sands at Dullingari may extend northwest to Corkwood and on to Coonatie. In addition delta front, distributary channel and turbidite facies would provide adequate reservoirs where diagenesis has been restricted. Certainly, the style of sedimentation lends itself to stratigraphic entrapment of hydrocarbons, but it is considered that at this stage of exploration, combined structural-

G. AMBROSE, R. SUTTILL & I. LAYERING

Fig. 10. Isopach map Murta Member, southern Eromanga Basin. Isopach interval 15 m.

PEDIRKA

V — J RE.L. 5/6 S

BASIN

ARCKARINGA \

BASIN

ft;

\ J BASIN

139°

140°

Fig. 11. Sand:shale ratio map, Murta Member, southern Eromanga Basin.

141°

142°

143°

MURTA MEMBER GEOLOGY

SOUTHERN EROMANGA BASIN—MURTA MEMBER GENERALISED FACIES RELATIONSHIPS

Fig. 12. Generalised depositional model, Murta Member, southern Eromanga Basin.

stratigraphic plays of the Dullingari type, are the most promising.

thus proposed that the Murta Member retain its status as a member of the Mooga Formation.

(3) Proximal lacustrine facies: this area is characterised by high sand: shale ratios in which the sands generally have excellent reservoir properties. Some exinite-rich source rocks are present but the presence of the alginite-rich variety is yet to be confirmed. The Jackson Murta oil discovery occurs in this area and the potential for further hydrocarbon discoveries is high. An overriding consideration in assessing the potential of the Murta Member is depth of burial. It is noteworthy that silty micaceous sands, which are good reservoirs in the Jackson Field, have far higher permeabilities than similar sands in the Dullingari Field (e.g. '49-0' sand). The depressed reservoir quality at Dullingari (excluding the '48-5' shorelinebar sand) is due to increased diagenetic effects related at least in part, to a greater depth of burial.

A reference section is assigned to Dullingari 9 where the sequence is subdivided into four units. The basal unit (Unit 1) is an upward-coarsening lacustrine delta sequence capped by delta front sands and occasional distributary channel deposits. Lower distal deposits contain minor pelagics but consist mainly of thin, graded silt-fine sand beds deposited by density currents which traversed a gently shelving lake floor.

CONCLUSIONS In the study area the Murta Member (Mooga Formation) was deposited in an asymmetric basin formed by compaction of the underlying Cooper Basin sequence. It is a mainly finegrained lacustrine sequence intervening between braided fluvial sediments of the Namur Sandstone Member (Mooga Formation) and the overlying, marginal marine Transition beds'. The lower contact is transitional and marks a period of degradation of the 'Namur' braided fluvial regime prior to the onset of lacustrine conditions. Core data indicate a gradational upward passage from the Murta Member into the basal Transition beds' and there is no conclusive evidence of marine influence until about midway through the latter unit. Log correlations show the sequence intertongues with the Namur Sandstone Member on a regional scale and it is

Unit 2 contains similar graded fine sand and silt beds deposited in relatively shallow water. This unit is succeeded by a sub-lacustrine fan sequence comprising graded beds which show an upward increase in thickness and sand content (Unit 3). The basinward progradation of the fan sequence, which extends over an area of about 5000 km 2 , was basically a shallowing trend which culminated locally in the development of a thin, elongate shoreline-bar sand. Graded beds characterise Unit 4 which marks a return to offshore sedimentation. The sequence fines upwards into the basal Transition beds' probably reflecting deeper water conditions prior to flooding of the lake by the sea. On a regional scale the Murta Member decreases in thickness and sand content from the north-northeast to the southwest. This trend reflects a regional depositional pattern whereby the 'Namur' braided fluvial regime contributed large volumes of sediment to the 'Murta' lake on the northern and eastern basin margins. Sandy proximal delta and fan sequences, together with minor bay-marsh facies have been cored in the Jackson Field. The sequence prograded from this area towards the south-southwest and fine-grained sediment and occasional coarse-grained sands were transported large distances into the basin by density flows (i.e. turbidity currents).

G. AMBROSE, R. SUTTILL & I. LAYERING

To date the Murta Member has hosted two commercial oil discoveries. The largest is the Dullingari-Murta Field which is a combined structural-stratigraphic play. Lenticular winnowed shoreline-bar sands provide the main reservoir in the field and subtle structural movements have apparently controlled porous sand development later modified by complex diagenetic processes. Proximal turbidite sand fk)ws and delta distributary channels have only produced small oil flows, largely due to the presence of quartz-calcite cements and limited reservoir continuity. The Jackson Field discovery occurs in a relatively sandy sequence deposited close to source areas. The main reservoir is a proximal delta front deposit, possibly modified by wave action. The sources of oil at Dullingari and Jackson are interpreted to have been silty shales within the Murta Member which are locally enriched in exinite (mainly alginite, sporinite and cutinite), particularly in the lower part of the sequence. With reference to the regional depositional model (Fig. 6) the hydrocarbon potential of the Murta Member is summarised as follows:

(1) Distal lacustrine facies: contains adequate source rocks but reservoir quality sands are sparse. Prospectivity is regarded as moderate to poor. (2) Intermediate lacustrine facies: source rocks are present and potential reservoirs although not abundant, include shoreline, delta front, distributary channel and turbidite facies. Prospectivity is regarded as high. (3) Proximal lacustrine facies: reservoir quality sandstones including delta front and proximal lacustrine fan facies, are relatively abundant and prospectivity is regarded as very high. Overall, it is considered that combined structuralstratigraphic plays of the Dullingari type are viable targets, especially along the '48-5' shoreline-bar sand trend. Structural plays will probably be more rewarding to the north and east where reservoir quality sands are relatively abundant.

REFERENCES AMBROSE, G. J., 1980: A regional study of the Murta Member with emphasis on depositional environments. Santos Ltd, Adelaide (unpubl.).

MOUNT, T. J., 1981a: Dullingari North No. 1, an oil discovery in the Murta Member of the Eromanga Basin. APE A J. 21(1), 71-7.

AMBROSE, G. J., 1982: Dullingari Adelaide (unpubl.).

MOUNT, T. J., 19816: Corkwood Pty Ltd (unpubl.).

Murta oil reserves. Santos Ltd,

AMBROSE, G . J. & F L I N T , R . B., 1982: M e s o z o i c s t r a t i g r a p h y S.W.

margin of the Eromanga Basin; in Moore, P.S. & Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 376-7. BURGER, D., 1973: Spore zonation and sedimentary history of the Neocomian Great Artesian Basin, Queensland. Geol. Soc. Aust. Spec.

1; proposal to drill. Delhi Petroleum

MOUNT, T. J., 1982: Geology of the Dullingari Murta Oilfield. Delhi Petroleum Pty Ltd (unpubl.). NUGENT, O., 1969: Sedimentation and petroleum potential of the Jurassic sequence in the southwestern Great Artesian Basin. A PEA

J. 9(1),

97-106.

Pub I. 4, 8 7 - 1 1 8 .

BOUMA, A. H., 1962: Sedimentology of some flysch deposits: A graphic approach to facies interpretation. Elsevier, Amsterdam. J O P L I N G , A . V. & W A L K E R , R . G . , 1968: M o r p h o l o g y a n d o r i g i n o f

ripple-drift cross-lamination, with examples from the Pleistocene of Massachusetts. J. Sediment. Petrol. 38, 971-84. LAWRENCE, A . P. & H O L L A N D , J. B., 1980: Dullingari

completion

report. Delhi Petroleum Pty Ltd

No.

well

(unpubl.).

MCINTYRE, S., 1980: The hydrocarbon potential of the Jurassic and Lower Cretaceous sediments from five wells in the Eromanga Basin, South Australia. B.Sc. (Hons) thesis, Univ. Adelaide (unpubl.).

PRICE, P. L., 1983: Dullingari No's 11, 13, 14, 15 and 21. Mines Administration Pty Ltd, Palynological Report 13/146 (unpubl.). STURM, M. & MATTER, A., 1978: Turbidites and varves in Lake Brienz (Switzerland): deposition of clastic detritus by density currents; in Matter, A. & Tucker, E. (eds) Modern and ancient lake sediments. Int. Assoc. Sedimentologists, Spec. Publ. 2. W O P F N E R , H . , FREYTAG, I. B. & H E A T H , G . R . , 1970: B a s a l J u r a s s i c -

Cretaceous rocks of western Great Artesian Basin, South Australia: stratigraphy and environment. Am. Assoc. Pet. Geol. Bull. 54, 383-416.

Geological Society of Australia Special Publication No. 12, 85-95

Margin of the Eromanga Basin South Australia: a review B. G. Forbes

S.A. Department of Mines and Energy, 191 Greenhill Road, Parkside, S.A. 5063.

ABSTRACT

In publications prior to 1978 the term Great Artesian Basin was commonly used to refer to the Eromanga Basin in South Australia. Some important early contributions were by Hudleston and Howchin (palaeontology), Jack (stratigraphy and structure), and Sprigg (petroleum potential). Current stratigraphic nomenclature is based on geological mapping by Freytag, Forbes, Wopfner and others and the biostratigraphy of Ludbrook. Stratigraphic units in common use on Geological Survey maps are (from oldest to youngest): Algebuckina Sandstone (Jurassic, main artesian aquifer); Cadna-owie Formation (Neocomian to Aptian); Bulldog Shale, marine Aptian to Albian; Oodnadatta Formation, marine Albian claystone, sandstone and limestone, with Coorikiana Member and Mount Alexander Sandstone Member; Winton Formation (Cenomanian). Although Bulldog Shale and Oodnadatta Formation have been grouped together as Marree Subgroup, it is probable that the original Marree Formation did not include the upper sandstone (Mount Alexander Sandstone Member) in the Oodnadatta Formation. It is recommmended that Marree Subgroup refer to the sequence in the original Marree Formation, with the exception of the basal Trinity Well Sandstone Member.

INTRODUCTION

first to describe the stratigraphy and structure of the Great Artesian Basin was Jack (1930) who used the stratigraphic subdivision of Whitehouse (1928) namely "Jurassic sand", Morven Formation ("transition series"), Roma Formation, Tambo Formation and Winton Beds, which approximates closely the current lithostratigraphic subdivision. He was followed by Ward (1946). In 1955 Woodard and Glaessner & Rao published details of the lithology and fossil plant remains of the sandy sequence below the marine Cretaceous in the northern Flinders Ranges. HISTORICAL The present era of exploration for oil and gas was ushered Studies prior to the current era of geological mapping in in by R. C. Sprigg and his colleagues of Geosurveys of the Great Artesian Basin in South Australia were all related Australia Limited who in their summary of the geology of to the search for groundwater. These include geological the Great Artesian Basin in South Australia (Sprigg, 1958) reconnaissance, palaeontology and investigation of noted that traces of oil and gas were present throughout the groundwater. Rawlinson (1878) quoted by Ward (1946, p. 42) basin and the petroleum potential was being investigated. appears to have made the earliest prediction of large supplies Geological Survey mapping of the margin of the Eromanga of underground water in an interior basin. Between 1894 and Basin began with the production in 1961 of the Gardiner, 1895 Mesozoic molluscs and foraminifera from the Lake Eyre and Paralana one-mile geological maps region were collected during reconnaissance survey and drilling Moolawatana et al., 1961c, b, c) showing the edge of the by the South Australian Government. The fossils were (Campana northeastern Flinders Ranges. Compilation was by R. P. Coats described by Hudleston (1884, 1890) and Howchin (1886, and mapping by Coats, Thatcher, Campana and 1893*7, b, 1895). A collection of geological specimens, Webb. These maps usedHorwitz, Queensland stratigraphic including Mesozoic fossils, from South Australia was exhibited nomenclature of Blythesdalethe Group, Roma Formation and by the Government Geologist at the Colonial Exhibition in Tambo Formation. The Callanna, Marree and Wilpoorinna London in 1886. In 1887 Scoular made general geological observations and collected fossils northwest of Marree. Brown one-mile geological maps appeared in 1963 and used similar (1892*7, b) reported on his wide-ranging reconnaissance around nomenclature (Webb, et al., 1963; Coats et al., 1963; Forbes the northern edge of the Flinders Ranges and the Lake Eyre & Coats, 1963). region during which he made extensive collections of fossil The first use of local stratigraphic nomenclature on Survey material and noted mound springs and anomalous boulders maps of the Eromanga Basin appeared on the 1:250 000 of quartzite and porphyry within the Cretaceous shales. The geological maps MARREE (Forbes et al., 1965) and In this review of the geology of the margin of the Eromanga Basin, emphasis will be on exposure and surface expression of the basin and the aspects which most concern a geologist engaged in regional geological mapping. Figure 1 is a general locality map. Prior to 1978 the term Great Artesian Basin was used in publications of the South Australian Department of Mines and Energy to refer to the Eromanga Basin.

B. G .

FORBES

OLD r

MUSGRAVE

EROMANGA

BLOCK

BASIN •Oo'dnaddtta'iK OFFICER BASIN

> Moomba

•

£ . C 6 o b e r Pedy :

Andamooka •

ADELAIDE

AJlU GEOSYNCLINE '

s j

GAWLER

S.S

r ,

J- S

V'VsV"

CRATON

STUART

J u r a s s i c - C r e t a c e o u s outcrop Palaeozoic Late P r e c a m b r i a n - C a m b r i a n Older P r e c a m b r i a n and volcanics Eromanga Basin margin •2

Type Section

KILOMETRES

SHELF

•i Mt. Alexander Sandstone Member of the Oodnadatta Formation •2 Wooldridge Limestone Member of the Oodnadatta Formation •3 Oodnadatta Formation •4 Coorikiana Member of the Oodnadatta Formation •5 Algebuckina Sandstone and Cadna-owie Formation • 6 Bulldog Shale • 7 Mt. Anna Sandstone Member of the Cadna-owie Formation •8 Mt. Howie Sandstone •9 Attraction Hill Sandstone Member of the Marree Formation •10 Wilpoorinna Breccia Member of the Marree Formation • n Lower part of the Marree Formation •12 Upper part of the Marree Formation and the Blanchewater Formation • 13 Parabarana Sandstone

Fig. 1. Eromanga Basin in South Australia: regional geology. Terms refer to original definitions. Note that the following revisions are suggested later: Mt Alexander Sandstone, Coorikiana Sandstone, Marree Subgroup, Wilpoorinna Breccia.

EROMANGA BASIN MARGIN, SA OODNADATTA (Freytag et al. 1967). Ludbrook (1966) produced a major contribution on Cretaceous biostratigraphy of the Great Artesian Basin in South Australia and provided a comprehensive biostratigraphic framework for both surface and subsurface studies. Wopfner, Freytag & Heath (1970), in another major contribution, extended the stratigraphy established in the Oodnadatta region over a great part of the Eromanga Basin in South Australia. The more recent Survey maps and reports are summarised in Figure 2. 9

STRATIGRAPHY

The common nomenclature used on recent Geological Survey maps of the Eromanga Basin in South Australia is shown in the left hand part of Figure 3.

Algebuckina Sandstone

composed of well-rounded quartz pebbles. The sandstone is cross-bedded, contains plant fossils and is thought to represent a continental, mainly fluvial environment. Ambrose (1980) suggests a braided-stream environment. Exposures of the Algebuckina Sandstone are best seen in the OODNADATTA and WARRINA regions, but basal sandstones thought to be equivalent have been recognised along a great part of the marginal Eromanga Basin (Fig. 4) and include the Village Well Formation in the MARREE area. Thickness data in the northeast were gathered from Nugent (1969: Birkhead and Mooga Formation and equivalents). Areas of non-deposition or erosion occur around the northeast Flinders Ranges, southeast and southwest of Lake Eyre and in the far west. In the northwest, in the ABMINGA area it is difficult to distinguish the Algebuckina Sandstone from the overlying Cadna-owie Formation. Field observations by Ambrose (1980) and others suggest an interval of widespread chemical weathering of the Algebuckina Sandstone prior to deposition of the Cadna-owie Formation.

The name Algebuckina Sandstone was first used in its current sense on the OODNADATTA 1:250 000 geological map (Freytag et al., 1967) and was formally defined by Wopfner et al. (1970). The type section is 0.8 km southwest of the disused Algebuckina railway siding. Age is thought Cadna-owie Formation The name Cadna-owie Formation was used on the to be Late Jurassic to possibly Early Cretaceous, based on microfloral studies (Harris, 1970)*. The lithology is typically OODNADATTA 1:250 000 geological map (Freytag et al., white, kaolinitic, quartz sandstone and conglomerate 1967) and the formation was defined by Wopfner, Freytag & Heath in 1970. The type section is 4 km west-southwest of * This refers to a locality near Ingomar Homestead, not the type the disused Algebuckina railway siding. Age is thought to be section, which is Harris (in Wopfner et al., 1970). The Ingomar Neocomian to Aptian. Lithology is typically fine- to mediumgrained, brownish sandstone. It is in parts calcitic, feldspathic, microflora is Late Jurassic.

B. G. FORBES

CENOMANIAN

Winton

Formation

Mt. Howie Ss

7 • Blanchewater Formation

Mt. Alexander Ss Member-

ALBIAN

Oodnadatta Formation Wooldridge Ls Member •Coorikiana Member

Marree Attraction Hill Ss Member

ALBIAN

APTIAN

Formation Bulldog Shale r Wilpoorinna Breccia J Member

APTIANNEOCOMIAN

\m. Anna Ss Member

Parabarana and Trinity Well Ss Pelican Well Formation

Cadna-owie Formation ? NEOCOMIAN -JURASSIC

V i l l a g e Well Formation

Algebuckina Ss

Fig. 3. Eromanga Basin in South Australia: stratigraphic units. Left hand side shows original terminology for c o m m o n l y used stratigraphic units of Jurassic-Cretaceous in South Australia. On right are map units originally used on M A R R E E and COPLEY. Note that Marree Formation now revised to Marree Subgroup is not everywhere divisible into upper and lower members and is overlain by a probable equivalent of the Mt Alexander Sandstone.

silty and micaceous, and also contains pebbles and boulders of quartzite. Pyrite is commonly present and may occur as large concretions. Carbonaceous shale is present in some areas. The Mount Anna Sandstone Member of the Cadna-owie Formation appears to occur mainly in the upper part of the Cadna-owie Formation. It has its type section at Mount Anna and is characterised by rounded pebbles of porphyritic rhyolite in a medium- to coarse-grained feldspathic sandstone. The Cadna-owie Formation is more widely distributed than the Algebuckina Sandstone and represents shallow marine to fluvial conditions (Fig. 5). Thickness data in the northeast were gathered from Nugent (1969). In the MARREE area it is represented by the Pelican Well Formation and Trinity Well Sandstone. Wopfner et al. (1970) noted a trough of fluvial Mount Anna Sandstone northwest of Lake Torrens. This, and evidence of northeasterly-directed palaeocurrents, support the derivation of porphyry boulders from the Gawler Range Volcanics. In the Mount Painter region Giles & Teale (1979) noted that prophyry boulders in the Cretaceous were locally derived. Boulders of Devonian quartzite have been,noted by Campbell et al., (1977). The Cadna-owie Formation may lie transitionally above the Algebuckina Sandstone or may overlie a low-angle unconformity.

Bulldog Shale The Bulldog Shale was defined by Freytag (1966) at a reference section 8 km south of Bulldog Creek, east of the Peake and Denison Ranges (Fig. 6). It is of Aptian and Albian age (Ludbrook, 1966, 1978, 1980; McNamara, 1980; Morgan, 1980) and is characterised by dark fossiliferous shale, claystone

and concretionary limestone. Thickness in Oodnadatta 1 is 170 m. Flint et al. (1980) have described fossiliferous Devonian quartzite boulders from the base of the Bulldog Shale. These are thought to have possibly originated as a reworked submarine debris-flow deposit, related to basin margins and basement highs. Ultimate derivation is thought to have been from the Amphitheatre Group, N.S.W., via CarboniferousPermian glacial ice. Around the northeast Flinders Ranges the Bulldog Shale appears to pass imperceptibly up into the Oodnadatta Formation of similar lithology, but over most of the area the base of the Oodnadatta Formation is marked by its sandy Coorikiana Member.

Oodnadatta Formation The Oodnadatta Formation of Albian age is lithologically similar to the Bulldog Shale. At the type section at Mount Arthur (Freytag, 1966) it is over 140 m thick. The basal Coorikiana Member ('Coorikiana Sandstone Member', Pitt, 1978) of the Oodnadatta Formation has a type area in Coorikiana Creek 40 km southwest of Oodnadatta. It is composed of fine-grained feldspathic and glauconitic sandstone with some coarser-grained and pebbly lenses. Coorikiana Sandstone is the name used for this unit by Thomson (1980) and is now the preferred term for the Coorikiana Member (Moore & Pitt, 1982). The Wooldridge Limestone Member of the Oodnadatta Formation occurs typically over about 8 m thickness in Wooldridge Creek, northwest of Oodnadatta, where it is

EROMANGA BASIN MARGIN, SA

Fig. 4. Eromanga Basin in South Australia: Algebuckina Sandstone thickness and outcrop distribution. Dotted line shows very approximately limits of areas of absence of Algebuckina Sandstone.

composed of calcareous and sandy siltstone with limestone concretions. It contains an ammonite fauna of Late Albian age (Ludbrook, 1966, 1978; McNamara, 1980). The Mount Alexander Sandstone Member of the Oodnadatta Formation (Freytag, 1966) was defined from a type section 49 m thick at Mount Alexander, north-northeast of Oodnadatta. It is characterised by very fine-grained glauconitic sandstone with shaley coarse siltstone and ferruginous carbonate beds. It is tentatively correlated with the lower part of the Blanchewater Formation (MARREE area) and also the Mackunda Formation. For this reason it is possibly best separated from the Oodnadatta Formation (Moore & Pitt, 1985).

Marree Subgroup The term Marree Subgroup, which first appeared on the South Australia 1:1 000 000 geological map of 1980, has been used to refer to the sequence, base of Bulldog Shale to top of Oodnadatta Formation. It should be noted that this application, which has not yet been formalised in the literature, is not the same as Marree Formation which terminated at the base of the Blanchewater Formation below the base of the Winton equivalent in the Marree area. Marree Formation, in

its original definition (Forbes, 1966) also included at its base a sandstone (Trinity Well Sandstone Member of the Marree Formation) which is probably equivalent to the Mount Anna Sandstone Member of the Cadna-owie Formation. It is suggested here (with Moore & Pitt, 1985) that Marree Subgroup be used in place of Marree Formation, as described below. The type area of the lower Marree Formation is in the southeast of the MARREE map area, near Village Well. Relationships are shown in Figure 7. The Trinity Well Sandstone Member of the Marree Formation has a sharp basal conglomeratic contact on pebbly and carbonaceous shale of the Pelican Well Formation and grades up into siltstones of the Marree Formation containing Aptian microfossils (Ludbrook, 1966). In spite of this relationship it is reasonable to accept the proposal of Wopfner et al. (1970) that the Trinity Well Sandstone Member is probably equivalent to the Mount Anna Sandstone Member of the Cadna-owie Formation. If the Trinity Well Sandstone is separated from the Marree Formation, the base of the Marree Formation becomes more readily mappable as the base of a shale sequence. In places near Marree, the base of the Marree Formation is composed of a shaley marine Aptian breccia, the Wilpoorinna Breccia Member, resting directly on the Proterozoic.

B. G. FORBES

The type section of the uppermost Marree Formation and overlying Blanchewater Formation is provided near Reedy Springs, southeastern MARREE region. Relationships are shown in Figures 8 and 9. The top of the Marree Formation (now top of Marree Subgroup) was chosen at the incoming of the lower sands of the Blanchewater Formation, probable equivalent of the Mackunda Formation (Exon & Senior, 1976), because this appeared to be the most readily mappable boundary. Winton Formation was presumed to be represented in the upper part of the type section of the Blanchewater Formation but no mappable boundary was recognised. In its eastern type region the Marree Formation is not divisible into upper and lower members because of the absence of suitable marker beds such as the Coorikiana Sandstone.

Mount Howie Sandstone

From this, it would seem that if the term Marree Subgroup is to be formalised, it should be applied only to a shaley sequence equivalent to Bulldog Shale plus the Oodnadatta Formation without the Mount Alexander Sandstone.

CONCLUSION

Above the Mackunda Formation equivalent as described at Reedy Springs, Winton Formation is widely exposed, but will not be described further here. The Mount Howie Sandstone was described and defined by Wopfner (1963) in the Cordillo region as a fluviatile unit incised into the Winton Formation. It is composed of white cross-bedded sandstone and shale-pebble conglomerate. Age may possibly be Turonian. Very similar cross-cutting, fluvial, kaolinitic, sandy beds have been noted northeast of Marree (Forbes, 1972). This unit, if correctly identified, has thus, approached quite close to the basin margin.

Although there is some duplication of stratigraphic names, the various units can generally be recognised and correlated. Lithologies characteristic of the Algebuckina Sandstone are

EROMANGA BASIN MARGIN, SA

Fig. 6. Eromanga Basin in South Australia: Bulldog Shale to Winton Formation. Distribution of outcropping Cretaceous above Cadna-owie Formation.

recognisable, though not continuously, over wide areas. The Cadna-owie facies is widespread and generally recognisable. The mainly shaley Bulldog-Oodnadatta sequence is the most extensive in area, while it seems the equivalent of the Mackunda Formation may be fairly continuous in both outcrop and in drillholes as indicated by Ludbrook (1966). It is recommended that in future mapping Marree Subgroup should refer to the same sequence as Marree Formation and

an endeavour be made to map the Mackunda Formation. Following the suggestions of Moore & Pitt (1982, 1985) it is agreed that preferred nomenclature within the Marree Subgroup be (from below) Wilpoorinna Breccia, Bulldog Shale, Coorikiana Sandstone and revised Oodnadatta Formation. Above the Oodnadatta Formation are the probably equivalent Mount Alexander Sandstone and Mackunda Formation, underlying the Winton Formation. The

B. G. FORBES

GEOLOGICAL SKETCH PLAN showing location of typesections near Village Well. © V i l l a g e Well Formation and Pelican Well Formation © L o w e r part, Marree Subgroup

QUATERNARY

CRETACEOUS \ ] P e l i c a n Well (Cadna-owie

Alluvium Gravels

V i l l a g e Well F o r m a t i o n (Algebuckina Sandstone)

TERTIARY SiIcrete

PRECAMBRIAN Wilpena Group : Q u a r t z i t e , si It s t o n e , m a r b l e

CRETACEOUS R V V S

Formation Formation)

JURASSIC-CRETACEOUS

M a r r e e Subgroup ( B uj lil d o g S h a l e ) T r i n i t y W e l l Sandstone (Mount Anna Sandstone Member { )

Ironstone

Fig. 7. E r o m a n g a Basin in South Australia: type sections near Village Well (location 11 in Fig. 1). Lower Marree S u b g r o u p a n d underlying Mesozoic, southeast o f Marree.

EROMANGA BASIN MARGIN, SA

Reedy Springs

i^Jfe

/ /

I Kn

n / f l ^

J ^ T x nA &?|TYPE SECTION]))]

y v C V ;

v KS \

^£

Blanchewater

W e / / v > > 0

IBLANCHEWATER • Happy Thoughts

100

Well\\

\ \ \ \ \

200

I—

KILOMETRES

1 Slope deposits Creek alluvium

1 1 1 1 Sand Quaternary sand and gravel

Tertiary silcrete Eyre Formation Blanchewater Formation M a r r e e Subgroup

Fig. 8. Eromanga Basin in South Australia: geology of Reedy Springs area (location 12, Fig. 1). Type section of upper part of Marree Subgroup and Blanchewater Formation, southeast of Reedy Springs on Blanchewater 1:63 360 sheet. Intersecting faults (F) have springs along them and delineate a depressed block of Blanchewater Formation. Reedy Springs occur within a minor dome bounded to the east by steeply-dipping, monoclinally folded silcrete. Adapted from Forbes (1966).

B. G. FORBES

WNW

ENE Quaternary clay Silcrete \ outcrop FAULT

Quaternary gravel'

Probable base of Winton Formation

Ironstone

Marree

Blanchewater

Subgroup

Fossiliferous claystone, shale, with large calcareous concretions 0

Eyre

Formation

Sandstone,sandy shale

100

200 METRES

300

Fig. 9. Eromanga Basin in South Australia: diagrammatic section southeast of Reedy Springs, MARREE region (location 12, Fig. 1). Type sections of upper Marree Subgroup (originally Marree Formation) and Blanchewater Formation. Probable equivalence is: (a) Wooldridge Limestone Member of Oodnadatta Formation, (b) Mount Alexander Sandstone and Mackunda Formation, (c) Winton Formation. Adapted from Forbes (1966).

term 'Blanchewater' has been omitted from Table 1 below but ACKNOWLEDGEMENTS may be needed if it becomes impractical To map the I am grateful to colleagues of the S.A. Geological Survey, Mackunda-Winton boundary. Dr D. Gravestock, Dr N. H. Ludbrook, Dr P. S. Moore and referees for helpful comments during preparation of the paper. TABLE 1. Suggested nomenclature. Mount Howie Sandstone Winton Formation Mackunda Formation, Mount Alexander Sandstone MARREE SUBGROUP: Oodnadatta Formation.(revised) Coorikiana Sandstone Bulldog Shale Wilpoorinna Breccia Cadna-owie Formation Algebuckina Sandstone

AMBROSE, G. J. & FLINT, R. B., 1981: Explanatory

REFERENCES

notes, Billa Kalina, 1:250 000 geological map, sheet SH53-7. S. Aust. Geol. Surv. AMBROSE, G. J., 1980: Southern Eromanga Basin excursion. S. Aust. Dept Mines Ener. Rep. 79/143 (unpubl.).

ALLCHURCH, P. D., WOPFNER, H . , HARRIS, W. K. & MCGOWRAN,

B., 1973: South Australian Department of Mines Cootanoorina No. 1 well. S. Aust. Geol. Surv. Rep. Invest. 40. BENBOW, M. C., 1982: Explanatory notes Coober Pedy 1:250 000 geological map, sheet SH53-6. S. Aust. Geol. Surv. BROWN, H . Y. L., 1892A: Further geological examination of Leigh Creek and Hergott districts. S. Aust. Pari. Pap. 23, 1-3. BROWN, H . Y. L., 18926: Government Geologist's report on country in the neighbourhood of Lake Eyre. S. Aust. Pari Pap. 141, 1-5. BRUNNSCHWEILER, R. O., 1959: New Aconerceratidae (Ammonoidea) from the Albian and Aptian of Australia. Aust. Bur. Miner. Resour. Geol. Geophys. Bull. 54, 5-19. CAMPANA, B., COATS, R. P., HORWITZ, R. C . & THATCHER, D., 1961a: Gardiner 1:63 360 geological map. S. Aust. Geol. Surv.

CAMPANA, B., COATS, R . P., HORWITZ, R. C . & THATCHER, D.,

19616: Moolawatana 1:63 360 geological map. S. Aust. Geol. Surv.

CAMPANA, B., COATS, R. P., HORWITZ, R. C. & THATCHER, D., 1961C:

Paralana 1:63 360 geological map. S. Aust. Geol. Surv.

CAMPBELL, K. S. W., ROGERS, P. A . & BENBOW, M., 1977:

Fossiliferous Lower Devonian boulders from the Cretaceous of South Australia. S. Aust. Geol. Surv. Q. geol. Notes, 62, 9-13.

COATS, R. P., HORWITZ, R. C . & WEBB, B. P., 1963: Marree 1:63 360

geological map. S. Aust. Geol. Surv. DALY, S. J., 1981: Stratigraphy of the Tarcoola 1:250 000 map sheet area. S. Aust. Dept Mines Ener. Rep. 81/5 {unpubl.). EXON, N. F. & SENIOR, B. R., 1976: The Cretaceous of the Eromanga and Surat Basins. BMR J. Aust. Geol. Geophys. 1, 33-50. FLINT, R. B., AMBROSE, G . J. & CAMPBELL, K. S. W., 1980:

Fossiliferous Lower Devonian boulders in Cretaceous sediments of the Great Australian Basin. R. Soc. S. Aust. Trans. 104, 57-66.

FORBES, B. G., 1966: The geology of the Marree 1:250 000 map area.

S. Aust. Geol. Surv. Rep. Invest. 28.

EROMANGA BASIN MARGIN, SA FORBES, B. G., 1972: Possible post-Winton Mesozoic rocks northeast

of Marree South Australia. S. Aust. Geol. Surv. Q. geol. Notes, 41, 1-3.

FORBES, B. G. & COATS, R. P., 1963: Wilpoorinna 1:63 360 geological

map. S. Aust. Geol. Surv.

FORBES, B. G., COATS, R. P., WEBB, B. P. & HORWITZ, R. C., 1965:

Marree 1:250 000 geological map, SH54-5. S. Aust. Geol. Surv.

FREYTAG, I. B., 1966: Proposed rock units for marine Lower

Cretaceous sediments in the Oodnadatta region of the Great Artesian Basin. 5. Aust. Geol. Surv. Q. geol. Notes, 18, 3-7. Oodnadatta 1:250 000 geological map, SG53-15. S. Aust. Geol. Surv. GILES, C. W. & TEALE, G. S., 1979: The geochemistry of Proterozoic acid volcanics from the Frome Basin. S. Aust. Geol. Surv. Q. geol. Notes, 71, 13-8. GLAESSNER, M. F. & RAO, V. R., 1955: Lower Cretaceous plant remains from the vicinity of Mount Babbage, South Australia. R. Soc. S. Aust. Trans. 78, 134-40. HARRIS, W. K., 1970: An upper Jurassic microflora from the western margin of the Great Artesian Basin, South Australia. S. Aust. Geol. Surv. Q. geol. Notes, 35, 3-8. HOWCHIN, W., 1886: On the fossil foraminifera from the Government boring at Hergott township, with general remarks on the section and on other forms of microzoa observed therein. R. Soc. S. Aust. Trans. 8, 79-93. HOWCHIN, W., 1893cr: Notes on the Government borings at Tarkaninna and Mirrabuckina, with special reference to the foraminifera observed therein. R. Soc. S. Aust. Trans. 17, 346-9. HOWCHIN, W., 18936: A census of the fossil foraminifera of Australia. Aust. Ass. Adv. Sci. 5, 348-73. HOWCHIN, W., 1895: Two new species of Cretaceous foraminifera. R. Soc. S. Aust. Trans. 19, 198-200. HUDLESTON, W. H., 1884: Notes on some Mollusca from South Australia obtained near Mount Hamilton and the Peak Station. Geol. Mag. n.s. dec.III; 1, 339-42. HUDLESTON, W. H., 1890: Further notes on some Mollusca from South Australia. Geol. Mag. n.s. dec.III, 7, 241-6. KRIEG, G. W., in press: Explanatory notes, Dalhousie 1:250 000 geological map, sheet SG53-11. S. Aust. Geol. Surv. JACK, R. L., 1930: Geological structure and other factors in relation to underground water supply in portions of South Australia. S. Aust. Geol. Surv. Bull. 14. LUDBROOK, N. H., 1966: Cretaceous biostratigraphy of the Great Artesian Basin in South Australia. S. Aust. Geol. Surv. Bull. 40. LUDBROOK, N. H., 1978: Australia; in Moullade, M. & Nairn, A. E. M. (eds) The Phanerozoic geology, of the World II, The Mesozoic, A. Elsevier, Amsterdam. LUDBROOK, N. H., 1980: A guide to the geology and mineral resources of South Australia. S. Aust. Dept. Mines Ener. MCNAMARA, K. J., 1980: Heteromorph ammonites from the Albian of South Australia. R. Soc. S. Aust. Trans. 104, 145-59. MOORE, P. S. & PITT, G. M., 1982: Cretaceous of the southwestern Eromanga Basin: stratigraphy, facies variations and petroleum potential. Eromanga Basin Symposium, summary papers, 127-44. MOORE, P. S. & PITT, G. M., 1985: Cretaceous subsurface stratigraphy of the southwestern Eromanga Basin: a review. Geol. Surv. S. Aust. Spec. Publ. 5, 269-86. FREYTAG, I. B., HEATH, G. R. & WOPFNER, H., 1967:

MORGAN, R., 1980: Eustacy in the Australian Early and Middle

Cretaceous. N.S.W. Geol. Surv. Bull. 27. NUGENT, O. W., 1969: Sedimentation and petroleum potential of the Jurassic sequence in the southwestern Great Artesian Basin. APEA J. 9, 97-107. RAWLINSON, T. E., 1878: Subterranean drainage in the interior. Adelaide Phil. Soc. Trans. Proc. & Rep. 1877-78, 124-6. REYMENT, R. A., 1964: Albian ammonites from Fossil Creek, Oodnadatta, South Australia. R. Soc. S. Aust. Trans. 88, 21-36. REYNER, M. L., 1955: The geology of the Peake and Denison region. S. Aust. Geol. Surv. Rep. Invest. 6. SCOULAR, G., 1887: Sketch of the geology of the southern and western parts of the Lake Eyre Basin. R. Soc. S. Aust. Trans. 9, 39-54. SPRIGG, R. C. & STAFF (Geosurveys of Australia Ltd), 1958: The

Great Artesian Basin in South Australia; in Glaessner, M. F. & Parkin, L. W. (eds) The geology of South Australia. Geol. Soc. Aust. J. 5, 88-101. THOMSON, B. P. (compiler) 1980: South Australia, 1:1 000 000 geological map. S. Aust. Dep. Mines Ener. THORNTON, R. C. N., 1974: Oodnadatta Town Bore 2 well completion report. S. Aust. Dep. Mines Ener. Rep. 74/178 (unpubl.). THORNTON, R. C. N., 1975: Manya No. 1 and Maria No. 1 well completion report. S. Aust. Dept Mines Ener. Rep. 75/106 {unpubl.). TOWNSEND, I. J., 1976: Stratigraphic drilling in the Arckaringa Basin, 1969-1971. 5. Aust. Geol. Surv. Rep. Invest. 45. TRUELOVE, A. J., 1980: Mesozoic stratigraphy of the Frome Embayment. S. Aust. Dept Mines Ener. Rep. 80/116 {unpubl.). WARD, L. K., 1946: The occurrence, composition, testing and utilisation of underground water in South Australia and the search for further supplies. S. Aust. Geol. Surv. Bull. 23. WEBB, B. P., HORWITZ, R. C. & COATS, R. P., 1963: Callana 1:63 360 geological map. S. Aust. Geol. Surv. WHITEHOUSE, F. W., 1928: The correlation of the marine Cretaceous deposits of Australia. Aust. Ass. Adv. Sci. Rep. 18, 275-80. WILLIAMS, A. F., 1970: Reconnaissance of the lower boundaries of the Bulldog Shale and Coorikiana Sandstone in the western portion of the Great Artesian Basin, South Australia. S. Aust. Dept Mines Ener. Rep. 70/113 {unpubl.). WILLIAMS, A. F., 1975: Explanatory notes, Noolyeana 1:250 000 geological map, sheet SG53-16. S. Aust. Geol. Surv. WILLIAMS, A. F., 1976: Explanatory notes, Lake Eyre 1:250 000 geological map, sheet SH53-4. S. Aust. Geol. Surv. WOODARD, G. D., 1955: The stratigraphic succession in the vicinity of Mt Babbage Station, South Australia. R. Soc. S. Aust. Trans. 78, 8-17. WOPFNER, H., 1963: Post-Winton sediments of probable Upper Cretaceous age in the central Great Artesian Basin. R. Soc. S. Aust. Trans. 86, 247-53. WOPFNER, H. & CORNISH, B. E., 1967: S.A.G. Fortville No. 3 well completion report. S. Aust. Geol. Surv. Rep. Invest. 29. WOPFNER, H., FREYTAG, I. B. & HEATH, G. R , 1970: Basal JurassicCretaceous rocks of western Great Artesian Basin, South Australia: stratigraphy and environment. Am. /4ssoc\ Petrol. Geol. Bull. 54, 383-416. YOUNGS, B. C., 1975: The geology and hydrocarbon potential of the Pedirka Basin. S. Aust. Geol. Surv. Rep. Invest. 44.

Geological Society of Australia Special Publication No. 12, 97-114

The Early Cretaceous Coorikiana Sandstone and Toolebuc Formation: their recognition and stratigraphic relationship in the southwestern Eromanga Basin P. S. Moore1, G. M. Pitt2 & M. E. Dettman3 1 2 3

Delhi Petroleum Pty Ltd, 101 Grenfell St, Adelaide, S.A. 5000. Western Mining Corporation Ltd, 168 Greenhill Road, Parkside, S.A. 5063. CSR Oil and Gas Divisions, 10 Eagle Street, Brisbane, Qld, 4000; present address University of Queensland, St Lucia, Qld 4067.

ABSTRACT The Coorikiana Sandstone and Toolebuc Formation are thin, laterally persistent and lithol'ogically distinctive units within the Early Cretaceous sequence of the southwestern Eromanga Basin. They are important units for correlation and represent significant events in the depositional history of the basin. The Coorikiana Sandstone is confined to a zone approximately 100 km wide, extending for over 1000 km around the southwestern margin of the Eromanga Basin. The lithological uniformity of the Coorikiana Sandstone over such a large area, its coarsening-upwards nature, lateral continuity and the gross shape of the component sand body suggest that it is a regressive, marine, shoreface deposit. The formation is assigned to the Coptospora paradoxa spore-pollen Zone and the Pseudoceratium turneri V dinoflagellate Subzone, which accumulated during the middle Albian, towards the close of a period of mild tectonism. The stratigraphically higher Toolebuc Formation is well developed in Queensland and northeastern South Australia, but is absent from the southern margin of the Eromanga Basin due to a facies change, as it grades laterally into shales indistinguishable from the vertically adjacent Marree Subgroup. The formation is characterised by dark grey to black mudstone which is richly fossiliferous and carbonaceous in parts, and which produces a distinctive, high gamma-ray wireline-log response related to scavenged uranium. The Toolebuc Formation accumulated in a basinal setting (not necessarily in deep water) where a slow rate of deposition and strongly reducing conditions promoted the accumulation of organic matter. The unit is assigned to the C. paradoxa sporepollen Zone and the Endoceratium ludbrookiae 'a' dinoflagellate Subzone in the southwestern Eromanga Basin and is accordingly considered to be of latest middle to late Albian age. A younger age assessment (Phimopollenites pannosus spore-pollen Zone; late Albian) for central Queensland sections suggests that either the Toolebuc Formation is slightly diachronous, or the distribution of Phimopollenites pannosus is facies controlled. Detailed wireline-log correlations show that the Toolebuc Formation lies approximately 50-120 m above the Coorikiana Sandstone in the few areas where these two units overlap. Previous confusion over the stratigraphic position of the Toolebuc Formation has resulted from failure to recognise that there are several other, similar gamma-ray anomalies in the Lower Cretaceous sequence. These stratigraphically lower anomalies are typically associated with dark, carbonaceous lithologies similar to those of the Toolebuc Formation, and which possibly formed in a similar depositional setting.

INTRODUCTION The Coorikiana Sandstone and Toolebuc Formation are two important marker units within the otherwise largely monotonous, marine, Lower Cretaceous shale sequence of the Eromanga Basin. The nomenclature of this sequence has recently been discussed and reviewed by Moore & Pitt (1982, 1985), who rationalised South Australian terminology and rendered it more compatible with that of the Eromanga Basin in Queensland. The present detailed discussion of these two critical units is intended to provide further supportive data for those reviews. Location of the study area is shown in Figures 1 and 2.

In previous years, the Coorikiana Sandstone and Toolebuc Formation were often regarded as equivalent, following the tentative.correlation of Freytag (1966). More recently the Toolebuc Formation, or facies variants thereof, have been depicted as immediately overlying the Coorikiana Sandstone. However, stratigraphic cross-sections presented herein show clearly that the units are not equivalent and are typically separated by 50-120 m of vertical section. Furthermore, the areal distributions of the Toolebuc Formation and Coorikiana Sandstone are almost mutually exclusive. Recent intense petroleum exploration in the Cooper Basin area has provided a large store of data in the form of wireline

P. S. MOORE, G. M. PITT & M. E. DETTMAN

logs, cuttings, conventional and sidewall cores and various analytical studies. The areal coverage of available logs is such that correlations can be made with some precision between wells from less than 1 km to rarely more than 30 km apart, in a broad, east-west transect across the southwestern Eromanga Basin. Step-by-step wireline-log correlation must be the first stage towards resolving the stratigraphic problems of the Lower Cretaceous of the Eromanga Basin. Only when log-based correlations and regional cross-sections are established and integrated within a biostratigraphic framework, can interpretations on such aspects as diachroneity, for example, be considered. Wireline logs being simply petrophysical measurements, reflect the nature of, or such changes in, strata that would in outcrop form the basis for lithological mapping. From this it follows that detailed correlation of log sections, such as are available from the Lower Cretaceous of the Eromanga Basin, leaves little room for alternative litho-correlations.

STRATIGRAPHIC NOMENCLATURE OF THE SOUTHWESTERN EROMANGA BASIN Figure 3 shows the stratigraphic nomenclature of the Cretaceous sequence of the southwestern Eromanga Basin as proposed by Moore & Pitt (1982, 1985). In summary, the Rolling Downs Group in Queensland (Exon & Senior, 1976), with the exception of the uppermost Winton and Mackunda Formations, is equivalent to the Marree Subgroup in South Australia (Forbes, 1982). Within the deeper parts of the southwestern Eromanga Basin, the presence of the Toolebuc Formation allows subdivision of the Marree Subgroup into the Allaru Mudstone, Toolebuc Formation and Wallumbilla Formation. Closer to the southwestern depositional margin of the basin, the Coorikiana Sandstone is present, but the Toolebuc Formation is not. There, the Oodnadatta Formation, Coorikiana Sandstone and Bulldog Shale are recognised. Generally, these units are readily identified on wireline logs from petroleum wells, as shown for two widely spaced wells in Figure 3.

COORIKIANA SANDSTONE Introduction The Coorikiana Member of the Oodnadatta Formation was defined by Freytag (1966) from Coorikiana Creek, south of Oodnadatta in central South Australia. The unit was later redesignated the 'Coorikiana Sandstone Member' (Pitt & Barnes, 1973), after mapping of the Cretaceous sequence in the Stuart Range area. Detailed mapping was made difficult by deep weathering, poor outcrop and low dip. It remains uncertain whether the sandstone in outcrop is a single sand body or a series of generally laterally-equivalent sandstone lenses. Meanwhile, Forbes (1966) and Ludbrook (1966) recognised the Attraction Hill Member in the Marree area nearly 200 km east of Oodnadatta. Both Ludbrook and Freytag suggested a correlation of the two members, via a glauconitic sandstone unit within Oodnadatta 1.

Fig. 2.

Principal well localities and lines of section, southwestern Eromanga Basin.

COORIKIANA SANDSTONE AND TOOLEBUC FORMATION RELATIONSHIPS

Fig. 3.

Cretaceous stratigraphic nomenclature of the southwestern Eromanga Basin showing relationship between formation boundaries, gamma-ray (G) and borehole compensated sonic (SLS) logs.

Recently Moore & Pitt (1982) proposed elevation of the Coorikiana Sandstone Member to formation status, following on from the use of the name 'Coorikiana Sandstone' (incorporating the Attraction Hill Member) by Thomson (1980). Formation status has since been applied by Forbes (1982), Ambrose et al. (1982) and others.

Outcrop lithology and distribution The type locality at Coorikiana Creek is a creek-bed exposure of some areal, but little vertical stratigraphic extent; the sequence is virtually flat-lying. The Coorikiana Sandstone consists of fine-grained, richly glauconitic, feldspathic and lithic sandstone, with minor, grey siltstone interbeds. Sandstone beds are evenly laminated and cross-stratified, calcareous in part, with minor fawn-weathering limestone concretions (Freytag, 1966). West of the type locality, lithologies equated with the Coorikiana Sandstone were mapped on the Murloocoppie 1:250 000 geological sheet (Pitt & Barnes, 1973). Here, they crop out as a thin sandstone sequence within vertical 'breakaway' exposures consisting otherwise of bleached, silty, Marree Subgroup shale. The sandstone varies from fine to coarse-grained, sometimes gritty with rare pebbles. Low-angle cross-stratification and bioturbation are both common. While occasionally comprising a single thin 0.5 m grit bed, the unit typically forms a protruding ledge in the escarpment of some 2-8 m thickness which passes rapidly upwards, and more gradually downwards, into silty shale. A direct outcrop

connection from the type area to other exposures has not been observed. The Coorikiana Sandstone in the Marree area (previously known as the Attraction Hill Member) has a similar outcrop pattern and appearance. Dips are low, outcrop is discontinuous and the unit comprises richly glauconitic, fine-grained sandstone. The sandstone appears to be laterally extensive and is the only significant sandy unit within the Marree Subgroup in this area (Forbes, 1966). Two other areas of sandstone outcrop within the Marree Subgroup deserve mention. Sandstone crops out beneath, and to the west of the Wooldridge Limestone Member type section at Wooldridge (or Fossil) Creek, northwest of Oodnadatta (Freytag et al., 1967) and scattered outcrops of cross-bedded glauconitic sandstone occur in a band between the Peake and Denison Ranges and Lake Eyre. The mapped distribution of the latter (Williams, 1976) suggests two or more beds within more shaly rocks, reminiscent of the sequence interpreted on wireline logs of the southern Cooper Basin area, described below. Both occurrences are directly equated with the Coorikiana Sandstone. In all areas, shales of the Marree Subgroup occur conformably above and below the Coorikiana Sandstone.

Subsurface correlations and log signature An essential feature of the stratigraphic scheme of Moore & Pitt (1982, 1985) is the extension of the Coorikiana

100

P. S. MOORE, G. M. PITT & M. E. DETTMAN

Sandstone into the subsurface. Although outcrop is sparse and the sand units may well be lenticular, the Coorikiana facies can be traced with some confidence for 600 km from the type locality, near Oodnadatta, to north of Marree, where Townsend (1971) recognised a correlative in Kopperamanna Bore. From Kopperamanna, it is only 80 km to the nearest wells in the southern Cooper Basin region. Here, in a broad east-west swath, extending 250 km from Lake Hope 1 in South Australia to Yanko 1 in Queensland, logs, sidewall cores and cuttings record a prominent, coarsening-upward sequence of glauconitic sandstone, up to 18 m thick, 50-120 m stratigraphically below the level of the Toolebuc Formation. This unit is correlated with outcropping Coorikiana Sandstone and bears the same name (Moore & Pitt, 1982, 1985). A subsurface reference section is designated in Dullingari 11, from 1123 m to 1131 m (Figs 3, 4).

The Coorikiana Sandstone in the southern Cooper Basin region has a distinctive appearance on wireline logs; a 'blocky', low gamma-ray response is accompanied by a relatively high sonic velocity which increases towards the top of the formation. From cuttings examination, the sequence grades from sandy siltstone at the base to fine-grained, moderately sorted sandstone at the top. An increase in calcite cement accompanies the increase in grain size and is reflected in the 'saw-tooth' sonic response. Several similar coarsening-upward sequences may occur above, or less commonly below, the Coorikiana Sandstone in the subsurface of the Cooper Basin area (Fig. 4). These units are generally thinner, finer-grained (dominantly siltstone) and laterally discontinuous. They are excluded from the Coorikiana Sandstone in the subsurface, although we suspect that in some areas of outcrop, these units may have been included in what has been mapped as Coorikiana Sandstone. The Coorikiana Sandstone has also been recognised in the subsurface in Oodnadatta 1 (Freytag, 1966, and others). The unit, alternatively known as the Terebratella Beds' (Sprigg, 1958) and the 'unnamed greensand member' (Ludbrook, 1966), extends from 131 m to 137 m depth and comprises an upwards-coarsening sequence of glauconitic sandstone. The sandstone is massive or very thingly bedded, very fine- to medium-grained, moderately sorted and very friable. Several carbonate-cemented horizons, 10-30 cm thick, occur in the upper half of the unit while the top 15 cm is a sandy limestone. The Coorikiana Sandstone in Oodnadatta 1 is the only sandstone of any significance in the middle and lower portions of the Marree Subgroup. Thus, it has an identical lithostratigraphic setting and a very similar lithology to other subsurface occurrences of the formation described above, and is also of a comparable age, within the resolution of the limited data available. On this basis, we tentatively support Freytag's (1966) original correlation with the type section. The Coorikiana Sandstone has also been described from Kopperamanna Bore (Townsend, 1971), and Tickalara and Kalladeina Bores (Morgan, 1980tf) north of Marree. Again, lithology, stratigraphic setting, and position within the Marree Subgroup are similar to nearby known occurrences of the Coorikiana Sandstone and we accept these correlations. Glauconitic sandstone units regarded by McMinn (1983) and McMinn & Burger (this volume) as possible correlatives of the Coorikiana Sandstone, occur in Urisino 1 and Birrigoolpa 1 (northern New South Wales) and Hay River 12 (Northern Territory). A glauconitic, sandy unit in Betoota 1 is also referred to as Coorikiana Sandstone by McMinn & Burger (this volume), some 250 km beyond the northern limit of the Sandstone as identified by our well-to-well correlations (Figs 2, 13). In our opinion, the geographic isolation of these occurrences is such that they should not be called Coorikiana Sandstone at this stage. Age

Fig. 4.

Coorikiana Sandstone; reference section, Dullingari 11.

Age control on the Coorikiana Sandstone in its type area is derived from megafaunal determinations on the adjacent shaly formations. Based on these determinations, Ludbrook (1966) assigned the unit to the lower part of the Albian. However, precise determination of the Aptian-Albian boundary was hindered by a paucity of ammonites in the upper part of the Bulldog Shale, due to the restricted environment of deposition at this level. In Oodnadatta 1, the Coorikiana Sandstone has been assigned to the Pseudoceratium turneri 'c' dinoflagellate Subzone (Morgan, 1977, 1980a, 19806) and the Coptospora

COORIKIANA SANDSTONE AND TOOLEBUC FORMATION RELATIONSHIPS 101 paradoxa spore-pollen Zone (Dettman & Playford, 1969). from the Boulia area of the northwestern Eromanga Basin. These authors determined a middle Albian age for the Vine & Day (1965) and Vine et al. (1967) changed the name formation, an assignment which we accept (Fig. 5). to Toolebuc Limestone', on the basis of its carbonate content Scheibnerova (1980) indicated that the Coorikiana Sandstone in outcrop. Subsequently Senior et al. (1975), noting the is above her Lingulogavelinella frankei foraminiferal datum heterogeneity of the unit, renamed it Toolebuc Formation and and suggested a younger, late Albian, age for the formation designated the type section in BMR Boulia 3A, near the on the basis of seemingly inconclusive and equivocal data (see original study area of Casey (Fig. 1). Morgan, 1980a; Playford et al., 1975). Stratigraphic relationships of the Toolebuc Formation (Fig. Age determinations of the Coorikiana Sandstone, derived 3) are discussed by Moore & Pitt (1985). The unit is overlain from recent drilling in the study area (Fig. 2), are summarised conformably by the Allaru Mudstone and underlain by the in Figures 6 and 7. The Coorikiana Sandstone lies within the Wallumbilla Formation. It occurs throughout a large portion C. paradoxa spore-pollen Zone and the P. turneri 'c' of the Eromanga Basin and extends into the Carpentaria dinoflagellate Subzone, as indicated by data from McKinlay Basin; distribution to the northeast and northwest is limited 1, Marabooka 1, Mudera 1 and Kidman 2. Determinations by the outcrop of the unit, while to the east, south and of P. turneri 'c' from immediately above the formation in southwest a facies change renders the Toolebuc Formation McKinlay 1 and Mudera 1 suggest that the Coorikiana indistinguishable from the enclosing shales. The Toolebuc Sandstone is entirely within that subzone. To the west, in Formation gamma-ray log anomaly as a result diminishes in Kalladeina Bore, the Coorikiana Sandstone has the same C. amplitude, until eventually, it merges with the background. paradoxa-P turneri *<? biostratigraphical relationships. To the Senior et al. (1975) and previous workers such as Smart east, however, in Tickalara Bore, the formation persists into (1972) recognised that the Toolebuc Formation consists of the younger E. ludbrookiae 'a' Subzone (Morgan, 1980a). calcareous and bituminous siltstone, black labile sandstone Substantial diachroneity spanning the P turneri 'b* to E. and shale. Hence, in Queensland, Senior et al. (1975, p. 452) ludbrookiae 'a' Subzones is implied by the datings of the proposed 'to use the name Toolebuc Formation for the supposed, but by no means certain, Coorikiana Sandstone calcareous heterogeneous sequence which overlies the litho-equivalents in northern New South Wales and the relatively non-calcareous siltstone and mudstone of the Northern Territory. In BMR Urisino 1, glauconitic sandstone Wallumbilla Formation and underlies the relatively nonunits are as old as the P. turneri 'b' Subzone (McMinn, 1983; calcareous mudstone of the Allaru Mudstone'. The strong McMinn & Burger, this volume), whereas in BMR Hay River gamma-ray anomaly was considered by Senior et al. (1975) 12, a sandstone unit, designated by McMinn & Burger (this and Exon & Senior (1976) to be characteristic of the volume) as Coorikiana Sandstone, is within the E. formation. ludbrookiae 'a' Subzone.

Definition and mappability

TOOLEBUC FORMATION Introduction

We consider that, in the subsurface, the Toolebuc Formation should be defined from wireline logs and mapped with reference to the top and bottom of that gamma-ray anomaly which, considered in the context of the entire marine sequence and by regional correlation, is the most prominent and

The Toolebuc Formation was originally designated by Casey (1959) as the Toolebuc Member' of the Wilgunya Formation,

SUBSURFACE STRATIGRAPHIC NOMENCLATURE

AGE

SPORE-POLLEN DINOFLAGELLATE ZONES

ZONES

A. distocarinatus WINTON

MACKUNDA

early

APT IAN

NEOCOMIAN

SUBGROUP

middle

MARREE

ALBIAN

late

FORMATION E. ludbrookiae

FORMATION

OODNADATTA

ALLARU

FORMATION

TOOLEBUC

P. pannosus a

MUDSTONE FORMATION C. paradoxa

COORIKIANA

SST. P. turneri WALLUMBILLA

BULLDOG FORMATION SHALE

CADNA-OWIE

FORMATION

D. speciosus

CENOMANIAN

C. striatus

b a

e C. hughesi

0 . operculata

b a

F. w onthaggiensis

Fig. 5. Cretaceous stratigraphic nomenclature and zonation, southwestern Eromanga Basin.

PK 298

102

PACKSADDLE

McKINLAY

MUDERA

MARABOOKA

KIDMAN

TOOLACHEE

NACCOWLAH

WAREENA

WINTON FORMATION

MACKUNDA FORMATION

OODNADATTA

ALLARU

FORMATION MUDSTONE

TOOLEBUC FORMATION

COORIKIANA SANDSTONE METRES

BULLDOG

WALLUMBILLA

SHALE FORMATION

CADNA-OWIE FORMATION SPORE

POLLEN

FORMATION

DETERMINATION

BOUNDARY

MOOGA FORMATION

R S. MOORE, G. M. PITT & M. E. DETTMAN

UNDIFFERENTIATED

Fig. 6. Lower Cretaceous spore-pollen determinations, various wells. Pp = Phimopollenitespannosus; Cp = Coptospora paradoxa; Cs = C. striatus; Ch = C. hughesi; Fw = Foraminisporis wonthaggiensis; Ca = Cicatricosisporites australiensis.

WALKANDI

FORMATION

MACKUNDA

FORMATION

MUDSTONE

TOOLEBUC FORMATION

COORIKIANA SANDSTONE

METRES

SHALE

FORMATION

CADNA-OWIE

FORMATION

Fig. 7. Lower Cretaceous dinoflagellate determinations, various wells. La = Endoceratium ludbrookiae 'a' Subzone; Ta = Pseudoceratium turneri 'a' Subzone; Tb = turneri 'b' Subzone; Jc = P. turneri 'c' Subzone; Ob = Odontochitina operculata 'b' Subzone; Oc = O. operculata 'c' Subzone.

WINTON

COORIKIANA SANDSTONE AND TOOLEBUC FORMATION RELATIONSHIPS

FORMATION UNDIFFERENTIATED

MOOGA BOUNDARY FORMATION

DETERMINATION DINOFLAGELLATE

WALLUMBILLA BULLDOG

ALLARU OODNADATTA

1 WAREENA 1 NACCOWLAH 9 TOOLACHEE 2 KIDMAN 1 MARABOOKA 1 MUDERA 1 McKINLAY 3 PACKSADDLE 1 WALKANDI

103

104

P. S. M O O R E , G. M. P I T T & M. E. D E T T M A N

persistent anomaly of a number that may be locally present. Our reasons are as follows: (a) the top and bottom of the gamma-ray anomaly have defined the boundaries of the Toolebuc Formation for two decades (see Senior et al., 1978); (b) our evidence shows that the acceptance by past workers of a correlation between the log anomaly and typical Toolebuc Formation lithologies is correct; and (c) such boundaries, based on interpretation of several hundred wireline logs from petroleum wells, are easily mappable. Ozimic (1982a, fig. 1; 1982c) suggested that the Toolebuc Formation be confined to areas where the unit is shale oilbearing. Since shale oil content is determined by chemical analysis, such a technique would render the formation unmappable using normal geological techniques. Furthermore, Ozimic's interpretation is inconsistent with the original concept of the formation proposed by Senior et al. (1975).

fossils, such as Inoceramus, are rare. T h e mudstone is soft to firm, evenly laminated and very clayey. It corresponds to a marked gamma-ray anomaly (200 API units). A low sonic velocity is attributed to the abundance of carbonaceous matter. (b)

Dark grey fossiliferous mudstone or coquinite (moderate gamma-ray response combined with high sonic velocity; electrofacies C).

Toolebuc Formation lithology and the nature of gamma-ray anomalies Between the coaly Winton Formation and the Cadna-owie Formation, generally one, often two, and sometimes more gamma-ray anomalies are present within shaly lithologies of the Marree Subgroup and Rolling Downs Group. However, the persistence and amplitude of the uppermost gamma-ray anomaly throughout much of the Eromanga Basin in South Australia and Queensland leave no doubt that it can be correlated regionally as the Toolebuc Formation especially if done so within the context of the whole Cretaceous section. The Toolebuc Formation anomaly in the study area is generally a 'sawtooth' gamma-ray log signature rising to 100-200 A P I units above the 40-50 A P I shale background. Peak values may be considerably greater. The base of the anomaly is typically sharp, while the top displays a gradual d e c l i n e to n o r m a l A l l a r u M u d s t o n e b a c k g r o u n d . Superimposed on the positive gamma-ray anomaly are 'negative spikes' known from core and cuttings data to be due to bedding-plane concentrations of Inoceramus shells (N. Ludbrook, pers. comm., 1981). The correlation of gamma-ray anomaly and Toolebuc Formation lithology has been widely accepted, however Ozimic (1982a, p. 229) suggested that 'the gamma-ray anomaly possibly rises in "the succession from the southern to the northern parts of the Eromanga Basin, from the upper part of the Coreena Member equivalent to the upper beds of the Toolebuc Formation' (see also McMinn, 1983). This contention is based largely upon a gamma-ray anomaly associated with glauconitic sandstone beneath dark shale beds and minor limestone bands in Urisino 1, northern New South Wales. However, we have found no supportive evidence within our study area and suggest that the anomaly in question in Urisino 1 is not a correlative of the Toolebuc Formation anomaly. This view is supported by the results of Ramsden et al. (1982) who show that the Urisino 1 anomaly is mainly due to potassium, whereas the Toolebuc gamma-ray anomaly results from uranium. Our studies of the Toolebuc Formation in PELs 5 & 6 and ATP 259P have led to the recognition of three lithotypes, which can be correlated with gamma-ray and sonic log responses (Fig. 8): (a) Black calcareous mudstone (high gamma-ray, low sonic velocity response; electrofacies D). The black mudstone contains a b u n d a n t carbonaceous matter and various fossil fish remains. Large calcareous

Fig. 8.

Lithology, gamma-ray and sonic (G-BHCS) log responses of Toolebuc Formation and enclosing strata, Morney 1, southwestern Queensland. Fades A: mid-grey siltstone; relatively low gamma-ray response (Wallumbilla Formation, Allaru Mudstone). Fades B: dark grey mudstone; moderate gamma-ray response. Fades C: Dark grey fossiliferous mudstone and coquinite. Dominant macrofossil is Inoceramus (I). Moderate gamma-ray response with high sonic velocity. Fades D: Black calcareous mudstone. Common skeletal fish fragments (F). Very high gamma-ray, low sonic velocity responses.

COORIKIANA SANDSTONE AND TOOLEBUC FORMATION RELATIONSHIPS

105

Abundant Inoceramus shells, lying parallel to bedding, suggest that the water was stratified with a permanent are a feature of this lithotype. The mudstone is firm, halocline below a layer of fresher water. calcareous and clayey. It is distinguished in the Although oil shales appear to be restricted to Queensland subsurface by a moderate to high gamma-ray response (Saxby, 1982; Ozimic, 1982a, b, c) the carbonaceous facies (100-200 API units) and a relatively high sonic velocity. of the Toolebuc Formation with its high gamma-ray response The elevated sonic velocity is a response to the abundant is well-developed in the northern parts of our study area calcareous fossil content, (Fig. 2). The dark colour and the absence of both a benthonic (c) Dark grey mudstone (moderate gamma-ray response; fauna and bioturbation suggest strong oxygen depletion in electrofacies B). the lower part of the water column and possibly also in the The mudstone is firm, clayey and evenly laminated. It substrate. The calcareous nature of the black mudstone, an is represented by a moderately high gamma-ray response abundance of carbonaceous matter and paucity of terrigenous (80-140 API units), with no significant variation from material suggest slow deposition, possibly with more the sonic-log values of adjacent formations. Electrofacies hospitable conditions for organisms occupying the upper B is most commonly developed in the upper part of the portion of the water column. Toolebuc Formation, but may also be present near the Cook (1982) suggested that the oxic-anoxic boundary during base, as at Morney 1. It represents a transitional facies Toolebuc Formation deposition lay at, or just above, the between the main anomaly and gamma-ray readings sediment-water interface, controlled by the development of recorded from adjacent formations (electrofacies A; Fig. algal-fungal mats. The mats were suggested to be the source 8). of abundant bituminite, which characterises the Toolebuc This correlation of Toolebuc Formation lithologies and Formation organic-rich facies. gamma-ray-sonic log responses is based on 31 sidewall-core samples from Morney 1, a full-hole core from Gilpeppee 2, Age various sidewall cores from other wells, cuttings descriptions, The Toolebuc Formation in Queensland is richly and wireline-log responses. Similar results are described by fossiliferous, containing shelly faunas, fish fragments, Senior et al. (1975) and Balfe (1978) for the Toolebuc Radiolaria and coccoliths of Albian age (Day 1969; Ramsden, Formation in Queensland. In addition, an independent study 1983). In more southernly regions of the basin however, faunal of the organic matter in the Cretaceous of the southwestern and floral data are limited (Moore & Pitt, 1985). ForaEromanga Basin by Cook (1982) has confirmed the presence minifera from Queensland were assessed by Haig (1979) of a distinctive Toolebuc Formation organic-rich lithotype as late middle to early late Albian, largely on the basis of corresponding to the gamma-ray anomaly. Acid insoluble planktonic forms that designate his Hedbergella infracretacea organic matter is almost entirely of algal origin (sapropel and Zone. Planktonic forms are scarce in more southerly regions dinoflagellates); land plant debris is rare to absent. where benthonic forms diagnostic of Haig's (1979) Two visually similar, but more poorly defined gamma-ray Ammobaculites association predominate. On the basis of her anomalies also occur in the Lower Cretaceous sequence foraminiferal studies, Scheibnerova (1983) asserted that the stratigraphically below the Toolebuc Formation. The Toolebuc Formation belonged to her HaplophragmoidesAssemblage uppermost occurs just above the Coorikiana Sandstone. It Trochammina-Textularia-ceT3itobuV\mimd shows characteristic Toolebuc-like' gamma-ray-sonic log Subzone (Subzone C). However, none of Scheibnerova's signatures, with gamma levels peaking at no more than 100 studied sections containing this subzone includes the Toolebuc API units against a 40-50 API unit background. The base Formation. Therefore age implications advanced by her of the anomaly defines horizon 'K ' (Fig. 3). It may occur cannot be regarded as applicable to the formation. where the Toolebuc Formation and its associated anomaly Dinoflagellates occur abundantly in the Toolebuc are absent and in this context may be misidentified as Toolebuc Formation but published data on the forms represented and Formation. Examples from Beanbush 1 (1647 m) and their biostratigraphical significance are confined to several Dullingari 11 (1092 m) are shown in Figure 3. scattered borehole sequences in the northern Eromanga Basin. A second gamma-ray anomaly occurs at a significantly At the reference section, in BMR Boulia 3A, and in BMR lower stratigraphic level, near the base of the Bulldog Shale Dobbyn 1, the Toolebuc Formation is within and just above and laterally equivalent Wallumbilla Formation (Fig. 11). The the base of the Endoceratium ludbrookiae 'a' Subzone of log response at this level is similar to that previously discussed, latest middle to late Albian age (Morgan 1980a, b\ McMinn, although anomalously low sonic velocities in some areas 1983). Other E. ludbrookiae 'a' determinations of Toolebuc suggest overpressuring, this interpretation being supported sediments are reported from BMR Augathella 6 and BMR Jericho 11 (Burger, 1981) and from CSR Borehole BHC 3017 by engineering data. From cuttings data, and by analogy with known lithologies (25.25 m-40.75 m) near Julia Creek (this paper). These of the Toolebuc Formation, both high gamma-ray intervals northern Toolebuc assemblages display moderate to high are interpreted to reflect organic-rich mudstones. This diversity and are further characterised by high frequencies of interpretation is supported by the mapping of an organic-rich Diconodinium spp. and Spiniferites spp. Farther south, in GSQ Eromanga 1 the Toolebuc dinocyst basal mudstone member of the Bulldog Shale in outcrop in assemblages examined by us from 466 m-470 m are the Coober Pedy area (Pitt & Barnes, 1973). dominated by Diconodinium spp. and cannot be determined more precisely than Pseudoceratium turneri 'c Endoceratium Environment of deposition ludbrookiae a' Subzones. Similar Diconodinium-rich The environment of deposition of the Toolebuc Formation assemblages lacking subzonal indices were recovered from the Toolebuc Formation in Wareena 1 and Naccowlah 1 (Filatoff has been discussed by many workers (see e.g. Moore & Mount, 1982*7, 1982/?; this volume, for recent work). The Toolebuc & Price, 1981; Dettmann & Price, 1981; present study, Fig. Formation is considered to have been deposited in a restricted- 7). To the west in Walkandi 1, the formation is referred to marine environment. Ozimic (1982c) and Glikson (1982) the E. ludbrookiae 'a* Subzone on the basis of Endoceratium 3

106

P. S. MOORE, G. M. PITT & M. E. DETTMAN

ludbrookiae which occurs within, below and above the (b) that the stratigraphic distribution of P. pannosus is more facies-dependent than previously thought, although formation (Dettmann et al., 1983; present study, Fig. 7). Dettmann (1973) and Burger (1980) have noted higher Collectively these dinoflagellate determinations imply that the incidences of angiospermous pollen in near-shore Toolebuc Formation falls within the E. ludbrookiae 'a' environments than those more distant from the shore. Subzone and is therefore of latest middle to late Albian age. In terms of spore-pollen biostratigraphy, some sections of Further possible causes for varying age determinations the Toolebuc Formation have been allocated to the could be due to differences in sub-surface samples (cores, Coptospora paradoxa Zone (middle Albian) and others to sidewall cores, cuttings) or differing techniques applied to the succeeding Phimopollenites pannosus Zone (late Albian- sample processing. Cuttings samples in particular need to be ?Cenomanian). Examination of these determinations reveals assessed with special caution in a biostratigraphic scheme that that identification of the P. pannosus Zone has not been based is based principally on first appearances of diagnostic species. on consistent criteria. As originally defined by Dettman & Age determinations from supposed Toolebuc Formation Playford (1969) and subsequently applied by Playford et al. equivalents in basin margin regions should be considered even (1975) the C. paradoxa/P. pannosus boundary is delimited carefully. For example, in Oodnadatta 1, a fossiliferous by the first appearance of Phimopollenites pannosus. The more from 97 m to 104 m has been correlated with the type immediately preceding C. paradoxa Zone lacks P pannosus shale Wooldridge Limestone Member by many authors, and with although other superficially similar angiosperm pollen are the Toolebuc Formation. This fossiliferous shale (commonly frequently encountered in the upper part of the zone. referred to as 'Wooldridge Limestone Member') is within and McMinn & Burger (this volume) recognise the P pannosus situated 15-20 m above the base of both the P. pannosus Zone Zone on the basis of its nominate species and other, (Playford et al., 1975) and the E. ludbrookiae 'a' Subzone presumably undescribed, angiosperm pollen species. (Morgan, 1977, 1980a). However, after careful examination Previously Burger (1981, 1982) and McMinn (1983) identified of the Oodnadatta 1 core, we conclude the interval 97 m to the P. pannosus Zone from the presence of 104 m bears only minor lithological resemblance to the Microfoveolatosporis canaliculars, observed by us from near Toolebuc and even less to the type Wooldridge the base of the C. paradoxa Zone. Thus, Burger's {op. cit.) LimestoneFormation Member. Moreover, it is only one of several and McMinn's {op. cit.) referral of the Toolebuc Formation similar horizons overlying the Coorikiana in Boulia 3A, Jericho 11, Augathella 6 and GSQ Hughenden lithologically Sandstone in Oodnadatta 1. Thus, we see no justification for 7 to the P. pannosus Zone must be viewed with caution. referring interval to either the Wooldridge Limestone Morgan (1980Z?) also allocated the Toolebuc Formation to the Member this or to the Toolebuc Formation. Consequently P. pannosus Zone (Morgan, 19806, figs 15, 16, 19, 20, Oodnadatta 1 cannot be regarded as providing valid evidence summary tables). The palynological bases of these (see regarding the age of either unit. Morgan 1980*7, charts 15, 16) relate to two subsurface sections A similar situation occurs in northwestern New South Wales containing the Toolebuc Formation, namely Conorado where the Toolebuc Formation is absent. Ozimic (1982c) and Ooroonoo 1 and Dobbyn 1. In the former, the C. paradoxa Zone is shown by Morgan (1980a, chart 15) to be represented McMinn (1983) referred the interval 102 m to 122 m in Urisino 1 to the Toolebuc Formation but noted that the section lacks in the Allaru Mudstone immediately above the Toolebuc Formation which is therefore no younger than that zone. In the characteristic gamma ray anomaly of the formation. Dobbyn 1 the Toolebuc Formation (Morgan 1980a, chart 16) Burger (1982; see also Scheibnerova, 1983) designated that lacked diagnostic spores and pollen; samples immediately section Wooldridge Limestone Member after correlating it below (Wallumbilla Formation) and above (Allaru Mudstone) with the supposed 'Wooldridge Limestone Member' in provided C. paradoxa and P. pannosus zonal determinations Oodnadatta 1. Clearly, Burger's (1981) and McMinn's (1983) respectively. Hence there is no determination of the P P. pannosus determinations of the Urisino section should not pannosus Zone per se by Morgan from the Toolebuc be invoked to argue the age of the Toolebuc Formation or Formation; moreover his Conorada Ooroonoo 1 data clearly the Wooldridge Limestone Member. imply correlation of the formation with the C. paradoxa Zone Resolution of biostratigraphic problems relating to the (see Fig. 16). Toolebuc Formation requires further well-controlled sampling and detailed palynological analyses. Meanwhile, caution Allocation of the Toolebuc Formation to the C. paradoxa Zone is supported by data from several southern Eromanga should be exercised when describing units as time-equivalents Basin wells. Determinations from Wareena 1 (Filatoff & Price, of the Toolebuc Formation, or when suggesting that the 1981), Naccowlah 1 (Dettmann & Price, 1981) and Walkandi Toolebuc Formation is isochronous throughout the area of 1 (Dettmann et al., 1983) demonstrate that the Toolebuc distribution. Formation is within the upper part of the C. paradoxa Zone and that the C. paradoxa-P. pannosus boundary is located within the basal part of the Allaru Mudstone (Fig. 6). To the REGIONAL LOG CORRELATIONS AND THEIR north, a continuously cored Toolebuc section examined from SIGNIFICANCE Eromanga 1, 466 m-470 m, yielded spore-pollen assemblages Introduction diagnostic of the C. paradoxa Zone. The value of log correlations in tracing Eromanga Basin It is thus evident that the palynostratigraphic findings of Burger (1981, 1982) and McMinn (1983) in relation to marginal stratigraphy lies in (a) the considerable density of available Toolebuc sections (Boulia 3A, Augathella 6 and Jericho 11) wells, (b) the relatively sensitive response of wireline logs to imply a younger age than the data from more distal Toolebuc lithological changes in an apparently rather uniform sequence, sections (Walkandi 1, Naccowlah 1, Wareena 1 and Eromanga and (c) the distinctive log responses of the Toolebuc Formation 1) viz. P pannosus Zone vs. C. paradoxa Zone respectively. and Coorikiana Sandstone. These lead to very reliable lithostratigraphic correlations, which in conjunction with This apparent age discrepancy could suggest: (a) that the Toolebuc Formation may be slightly diachronous palynological data are of considerable value in elucidating the depositional history. within the Eromanga Basin; or

COORIKIANA SANDSTONE AND TOOLEBUC FORMATION RELATIONSHIPS The logs and log correlations figured here are extended between two well-established datum levels: top of the Cadnaowie Formation and base of the Winton Formation. This is necessary to allow an interdependent number of stratigraphic ties to be drawn through the sequence and thus strengthen the overall lithostratigraphic correlation. A sense of 'geometric perspective' is thus gained which is valuable when attempting correlation of a single unit.

Tie-lines used for correlation In Figure 3, a number of datum levels are identified which are based on consistent changes in wireline-log signature. Three categories are recognised: (a) regional datum levels such as the top of Cadna-owie Formation, which are confidently recognisable throughout the major part of the Eromanga Basin and may be followed on our sections, and in part, on the sections of Exon & Senior (1976). (b) semi-regional datum levels, such as the top of Coorikiana Sandstone, and the Toolebuc Formation gamma-ray peak, which are identifiable over large areas but which are not present over the whole basin. (c) facies-controlled horizons, such as the base of Mackunda Formation or top of Toolebuc Formation, which approximately parallel other datum levels, but which also show some facies control. It is unfortunate that past published sections have been limited to gamma-ray logs only, since sonic logs greatly facilitate the interpretation of the lithological sequence.

Regional datum levels (i)

C' datum: top of the Cadna-owie Formation; equivalent to the ' C seismic horizon (Fig. 3). 'O' datum: top of the basal, organic-rich member of the Bulldog Shale and Wallumbilla Formation. The unit 'C' to 'O' is recognised on logs throughout a major part of the Eromanga Basin. 'O' marks a return from elevated gamma-ray values and low sonic velocities to those of the 'normal' overlying sequence (Fig. 3). (iii) 'D' datum: this horizon is recognised primarily on sonic logs. It marks the change upward within the Bulldog Shale and Wallumbilla Formation from a sonic signature which has a lobate base line, to one with a flat baseline (Fig. 3). The horizon is generally best shown in sections containing Toolebuc Formation rather than Coorikiana Sandstone. Other relevant features are: (a) the 'C' to 'D' isopach is remarkably constant at 150 to 180 m. Thickness variations in the interval *C horizon to base Toolebuc Formation are accommodated by the sequence above 'D' datum. (b) 'D' datum corresponds with a subtle maximum observed on the gamma-ray log, between 'C' and base Toolebuc Formation. As such it occurs about 20-40 m below Exon & Senior's (1976, figs 7-11) top Doncaster Member tie-line. Therefore, pre-Toolebuc Formation thickness changes are largely accommodated within the Coreena Member.

(ii)

107

localised areas (e.g. Cuttapirrie 1 to Arrabury 1) there is no doubt of the correlation and identity of this shale. However, other high gamma-ray shales in the southern Cooper Basin area may be stratigraphically lower, immediately overlying the Coorikiana sequence. We suggest that the lower gamma-ray anomaly of Herbert (1980) is similar and approximately equivalent, but not necessarily a direct correlative, (iii) 'T 2 ' datum: an intra-Toolebuc Formation datum, defined by the gamma-ray maximum.

Facies-controlled horizons (i) (ii)

'K,' horizon: base of Coorikiana Sandstone. 'T,' horizon: base of Toolebuc Formation, often sharply defined on gamma-ray and sonic logs. (iii) T 3 ' horizon: top of Toolebuc Formation, defined by the return to normal gamma-ray activity. This horizon is analogous to 'O' datum and, similarly, is often not sharply defined. (iv) A' horizon: top Allaru Mudstone, defined by the lowest major high-velocity sonic spikes and an irregular gamma-ray trace, due to carbonate-cemented sandstones, which are identified with the Mackunda Formation. (v) 'M' horizon: base Winton Formation. By comparison with the log response of the Mackunda Formation, the overlying Winton Formation has alternating highvelocity spikes due to carbonates and low velocity spikes due to coals, and a subtly less irregular gamma-ray trace. In Figure 3, Beanbush 1 and Dullingari 11 show the datum levels particularly well. Other wells may show the various datum levels less clearly. Certain of these horizons are employed in Moore & Pitt (1985).

Section 1: Purni 1 to Beanbush 1 (Fig. 9) This section extends from the western Pedirka region, across the Birdsville Track Ridge (upon which Pandieburra 1 was drilled), to Beanbush 1 in the Patchawarra Trough area of the Cooper Basin. In the central Pedirka region (Simpson Desert) a major gamma-ray anomaly characterises the Toolebuc Formation. Thickness of the Toolebuc Formation at Kuncherinna 1 is 45 m; nearby, in Putamurdi 1 (not shown) the formation reaches its maximum known thickness of 52 m. Further west, the Toolebuc Formation thins considerably (e.g. Erabena 1, Purni 1) until, in wells west of Purni 1, it cannot be identified. Nor is it recognised where the correlative section crops out in the Dalhousie region (Krieg, 1982). Although the stratigraphic section above the Toolebuc Formation in the Simpson Desert area can be roughly subdivided into Allaru Mudstone, Mackunda Formation and Winton Formation, unit boundaries are unclear. For this reason the interval is presently shown as undifferentiated (Fig. 9). Correlation of the pre-Toolebuc section suggests that most of the thinning of the Wallumbilla Formation is accommodated by the base-of-Toolebuc Formation (T, to 'K 3 ' horizon interval.

Section 2: Beanbush 1 to Morney 1 (Fig. 10) Semi-regional datum levels (i) (ii)

'K ' datum: top of Coorikiana Sandstone. 'K ' datum: the base of a high gamma-ray shale which occurs just above the top of the Coorikiana Sandstone. The gamma-ray-sonic log response in this shale resembles that of the Toolebuc Formation. In relatively

In this section, the Winton and Mackunda Formations are clearly defined by their gamma-ray and sonic log responses. In Gilpeppee 2, Curalle 1 and Morney 1, in the northern Cooper Basin area, the Toolebuc Formation anomaly is prominent and well-developed. As with Section 1 (Fig. 9), thickness changes in the Wallumbilla Formation are

108

Fig. 9.

P. S. MOORE, G. M. PITT & M. E. DETTMAN

Gamma-ray, sonic log correlations. Section 1: Purni 1 to Beanbush 1. Location of cross-section shown in Fig. 2.

interpreted to take place largely in the interval immediately underlying the Toolebuc Formation.

are not apparent and all intra-Marree Subgroup features are difficult to recognise and correlate.

Section 3: Beanbush 1 to Weena 1 (Fig. 11)

Section 4: Spencer 2 to Toolachee 9 (Fig. 12)

In contrast to section 2, this section, which extends southwards from Beanbush 1, shows facies changes in the Aptian-Albian section. The gamma-ray anomaly of the Toolebuc Formation persists to the south, but becomes more subdued until in the Gidgealpa area the anomaly is reduced to a small, eventually imperceptible rise above background. South of Gidgealpa 16, the Toolebuc Formation anomaly and associated lithologies are absent. A second gamma-ray anomaly, resembling that of the Toolebuc Formation, occurs at the 'K 3 ' level in Beanbush 1. This feature may be confidently correlated westwards, eastwards and southwards from Beanbush. In the GidgealpaMoomba-Wooloo 1 area the 'K3' anomaly is prominent, whereas the Toolebuc Formation anomaly is either subdued, or absent. Moreover, the 'K 3 ' anomaly in this area overlies carbonate-cemented sandstone units which correlate in their lower part with the Coorikiana Sandstone. Only in this transect of those presented here, are both the Toolebuc Formation and Coorikiana Sandstone present in vertical section.

Throughout this section, the Toolebuc Formation is absent, while the low-gamma-ray signature of the Coorikiana Sandstone is well-developed. An isopach map of the Coorikiana Sandstone (Fig. 13) shows that the formation is confined to a 40 km wide, linear trend subparallel to the southern Eromanga Basin margin. However, carbonate-rich, fine-grained, silty equivalents of the Coorikiana Sandstone may be correlated much further northwards to Beanbush 1, for example, where sonic peaks below horizon K3 represent equivalents of the Coorikiana Sandstone.

South of Wooloo 1, towards the southern basin margin, the distinctive characteristics of the Coorikiana Sandstone

Section 5: Toolachee 9 to Ashby 1 (Fig. 14) The line of Section 4 is here extended eastwards. The Coorikiana Sandstone is present throughout, and in most wells the K3 horizon may be recognised.

Section 6: Ashby 1 to Ingella 1 (Fig. 15) Passing northwards from Ashby 1, the Coorikiana Sandstone becomes finer grained until, in Naccowlah 1, it is unrecognisable. However, as the Coorikiana Sandstone fades out, the Toolebuc Formation gamma-ray anomaly appears, occurring approximately 50 m higher in the stratigraphic

COORIKIANA SANDSTONE AND TOOLEBUC FORMATION RELATIONSHIPS

Fig. 10. Gamma-ray, sonic log correlations. Section 2: Beanbush 1 to Morney 1. Location of cross-section shown in Fig. 2.

Fig 11- Gamma-ray, sonic log correlations. Section 3: Beanbush 1 to Weena 1. Location of cross-section shown in Fig. 2.

110

p. S. M O O R E , G. M . P I T T & M . E . D E T T M A N

SPENCER-2

WOOLOO-1

NAMUR-1

MCKINLAY-1

TOOLACHEE-9

WINTON FORMATION

MACKUNDA FORMATION

OODNADATTA FORMATION

COORIKIANA SANDSTONE

METRES

BULLDOG

SHALE

CADNA-OWIE FORMATION

Fig. 12.

Gamma-ray, sonic log correlations. Section 4: Spencer 2 to Toolachee 9. Location of cross-section shown in Fig. 2.

WELCOME LAKE

MERRIMELIA COOPERS CK

BURLEY

TINDILPIE

JACK LAKE ORIENTOT YANKO HUME ROSENEATH .WOLGOLLA ASHBY

TICKALARA

WILLS

PELKETA

BRUMBY

BELAH >NARYILCO

PINNA TILPAREE

A TP 259P MULGA

PE.L. 5 8 6 SOUTH

AUSTRALIA QUEENSLAND

KUMBARIE

TINGA TIN6ANA GunnA WCCNA

CHERRI

CONTOUR

Fig. 13.

NEW INTERVAL

Smttrei

Isopach map, Coorikiana Sandstone, southern Cooper Basin region.

SOUTH KIlOMETKS

WALES

COOR1KIANA SANDSTONE AND TOOLEBUC FORMATION RELATIONSHIPS TOOLACHEE-9

KIDMAN-2

DULLINGARI-1 1

ASHBY-1

Fig. 14. Gamma-ray, sonic log correlations. Section 5: Toolachee 9 to Ashby 1. Location of cross-section shown in Fig. 2.

Fig. 15- Gamma-ray, sonic log correlations. Section 6: Ashby 1 to lngella 1. Location of cross-section shown in Fig. 2.

P. S. MOORE, G. M. PITT & M. E. DETTMAN

112 Conorada

Ooroonoo

BIOSTRATIGRAPHIC

1 BMR

400m

Dobbyn

SUMMARY

1 THIS

PAPER

MORGAN 1980(B)

P. P ALLARU MUDSTONE

\ -C. p . \

\ TOOLEBUC rFORMATION"1^

WALLUMBILLA FORMATION "

P. pannosus Zone

P. P. Not

C. p.

C. sT

-o

P. pannosus Zone

Definitive

C. paradoxa Zone

C. S. -C. S.

150m

650m

C. striatus Subzone

-25 50 METRES

Fig. 16. Toolebuc Formation spore-pollen biostratigraphy interpreted by Morgan (19806) and herein, for Conorado Ooroonoo 1 and BMR Dobbyn 1.

Albian C. striatus/P. turneri 'b' Subzones are extremely thin or missing, and in sections across the Birdsville Track Ridge and to the east, there is marked attenuation of the middle Albian C. paradoxa/P. turneri 'c' Subzones beneath the Toolebuc Formation. Morgan (1980&) argued that early-middle Albian sedimentation in the Eromanga Basin could be Discussion of sections It can be concluded from the above data that within the explained entirely by a transgressive/regressive cycle (cycle E) section from 'C' horizon to Winton Formation, a major and and that tectonism other than slow and uniform downcorrectable gamma-ray anomaly (the Toolebuc Formation warping need not be considered. Our data imply that anomaly) extends through the northern portion of the study tectonism may have influenced sediment accumulation within area, but disappears to the south. Stratigraphically lower in the southern Eromanga Basin during early-middle Albian. the section, a similar, second gamma-ray anomaly However, caution is needed when ascribing palynological (base = K ) is often developed. This is similarly interpreted breaks in the sequence to disconformities, since local loss of as a thin organic-rich shale unit, often underlain by carbonate- stratigraphic section can be due to listric, keystone faulting cemented sandstone units, which are correlated in part with (Moore & Pitt, 1984). Fortunately, major faults of this nature are easily recognisable on modern seismic sections. the Coorikiana Sandstone. At the base of the Marree Subgroup, an organic-rich mudstone is interpreted as a third, 'Toolebuc-like' gamma- CONCLUSIONS ray and sonic feature on wireline logs. Our correlations and The Coorikiana Sandstone is defined from outcrop, and the sections of Senior et al. (1975) show this unit to be basin- is correlated into the subsurface via Oodnadatta 1, wide. We consider the 'unnamed transitional unit! described Kopperamanna Bore and Lake Hope 1. It has a consistent by Pitt (1978) in the Coober Pedy area, to be the equivalent middle Albian age over much of its area of distribution, being unit in outcrop. assigned to the C. paradoxa spore-pollen Zone and the P. Logs of specific wells show minor high gamma-ray turneri 'c' dinoflagellate Subzone. The coarsening-upwards Toolebuc-like' signatures at other levels in the Aptian-Albian nature of the unit, its distribution around the margins of the section, reflecting local developments of organic-rich muds basin and its considerable lateral extent suggest that the developed in response to strongly reducing conditions. Careful Coorikiana Sandstone is a regressive marine, shoreface examination of gamma-ray and sonic logs suggests that deposit, that accumulated in response to mild tectonism thickness changes of the Wallumbilla Formation and Bulldog and/or a eustatic fall in sea level. Shale are ascribed to thickness changes in the upper half of The Toolebuc Formation has its type section in the those formations between 'D' and T,' horizons. A phase of subsurface in central Queensland and can be easily correlated gentle structuring of the approximate age of the Coorikiana into South Australia. While oil shales appear to be absent Sandstone (early-middle Albian) is thereby implied. from the southwestern Eromanga Basin, all of the other Furthermore, palynological data support the presence of distinctive features of the Toolebuc Formation such as its dark disconformities within the upper part of the Wallumbilla colour, abundant Inoceramus, skeletal fish remains, marly Formation. For example, in Walkandi 1 (Figs 6, 7) the early character, high concentration of carbonaceous material and section. The gamma-ray anomaly and associated dark, carbonaceous facies are well-developed in these-Queensland wells, and can be easily correlated with the type-section of the Toolebuc Formation at Boulia 3A.

113

C O O R I K I A N A SANDSTONE A N D T O O L E B U C F O R M A T I O N R E L A T I O N S H I P S high gamma-ray response are developed. The Toolebuc Formation is latest middle or late Albian in age. In our study area, it lies within the C. paradoxa spore-pollen Zone and the E. ludbrookiae 'a' dinoflagellate Subzone. Results from central Queensland (Burger, 1981, 1982) suggest that the formation lies within the younger P. pannosus spore-pollen Zone. This minor discrepancy can be explained in a variety of ways, and raises the possibility that the Toolebuc Formation is locally diachronous. Alternative interpretations are that the stratigraphic distribution of P. pannosus is facies controlled, or that the age anomaly is an expression of different workers' interpretations and techniques. Further sampling and discussion are required to solve this problem. Using outcrop and subsurface data, we have shown that the Toolebuc Formation and Coorikiana Sandstone are almost mutually exclusive in their areal distribution, with the Toolebuc Formation developed away from the southwestern Eromanga Basin margin and areas of sediment input, and the slightly older Coorikiana Sandstone developed in a

nearshore, regressive shallow shelf environment. In the process, we hope that we have demonstrated that detailed wirelinelog correlation is an essential prerequisite for interpreting the sedimentology and stratigraphy of the Eromanga Basin. Data presented here are based on examination of nearly 500 petroleum wells and provide a substantial framework for further studies.

ACKNOWLEDGEMENTS The authors wish to acknowledge discussions and helpful criticism from G. Krieg, B. Forbes, D. Gravestock (South A u s t r a l i a n D e p a r t m e n t of Mines a n d Energy), V. Scheibnerova, A. McMinn (New South Wales Department of Mineral Resources), S. Ozimic, D. Burger (Bureau of Mineral Resources), P. Price (AAR Limited), A. Williams, T. Mount (Delhi Petroleum Pty Ltd), W. Harris, C. Foster (Western Mining Corporation), N. Ludbrook and B. Senior. The paper is published with the permission of Delhi Petroleum Pty Ltd and its associates in PELs 5 & 6 and ATP 259P.

REFERENCES AMBROSE, G . , SUTTILL, R . & LAVERING, I., 1982: A r e v i e w o f t h e

Early Cretaceous Murta Member in the southern Eromanga Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 92-110. BALFE, P. E. 1978: Stratigraphic drilling report—GSQ Manuka 1. Qld Gov. Min. J. 79, 258-70. BURGER, D., 1980: Palynological studies in the Lower Cretaceous of the Surat Basin, Australia. Aust. Bur. Miner. Resour. Geol. Geophys. Bull. 189. BURGER, D., 1981: Palynology of three NERDDC stratigraphic bores in the Eromanga Basin: preliminary report. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1981/32 (unpubl.). BURGER, D., 1982: Palynological examination of Late Mesozoic in G.S.Q. Hughenden 7, and notes on geological events in the northern Eromanga Basin. Qld Gov. Min. J. 83, 421-32. BURGER, D., 1986: Palynology, cyclic sedimentation, and palaeoenvironments in the Late Mesozoic of the Eromanga Basin; in This volume. CASEY, J. N., 1959: New names in Queensland stratigraphy; northwest Queensland. Aust. Oil Gas J. 5, 31-6. COOK, A. C., 1982: Organic facies in the Eromanga Basin; in Moore, P. S. & Mount,. T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 234-57. DAY, R. W., 1969: The Lower Cretaceous of the Great Artesian Basin; in Campbell, K. S. W. (ed.) Stratigraphy and palaeontology: essays in honour of Dorothy Hill. ANU Press, Canberra. DETTMANN, M. E., 1973: Angiospermous pollen from Albian to Turonian sediments of eastern Australia; in Glover, J. E. & Playford, G. (eds) Essays in honour of Isabel Cookson. Geol. Soc. Aust. Spec. Publ. 4, 3-34. DETTMANN,

E. &

PLAYFORD,

G.,

1969:

Palynology

the

Australian Cretaceous: a review; in Campbell, K. S. W. (ed.) Stratigraphy and palaeontology, essays in honour of Dorothy Hill. ANU Press, Canberra. DETTMANN, M . E . , FILATOFF, J. & P R I C E , P. L . , 1983:

Palynological

Laboratory Report No. 13/150, Walkandi No. 1. Mines Administration Pty Ltd {unpubl.). DETTMANN, M. E. & PRICE, P. L., 1981: Palynological laboratory report 13/132, Naccowlah 1. Mines Administration Pty Ltd {unpubl.). EXON N F. & SENIOR, B. R., 1976: T h e Cretaceous of the Eromanga

and Surat Basins. BMR J. Aust. Geol. Geophys. 1, 33-50. FILATOFF J & PR^E, P L., 1981: Palynological laboratory report 13/132, Wareena 1. Mines Administration Pty Ltd {unpubl.).

FORBES, B. G., 1966: The geology of the Marree 1:250 000 map area. S. Aust. Geol. Surv. Rep. Invest. 28. FORBES, B. G., 1982: Margin of the Eromanga Basin South Australia: a review; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 110-26. FREYTAG, I. B., 1966: Proposed rock units for marine Lower Cretaceous sediments in the Oodnadatta region of the Great Artesian Basin. S. Aust. Geol. Surv. Q. geol. Notes, 10. FREYTAG, I. B., H E A T H , G . R . & W O P F N E R , H . , 1967:

Oodnadatta

1:250 000 geological map. S. Aust. Geol. Surv. GLIKSON, M., 1982: Cynobacterial mats and other bacteria: major contributors to the formation of Lower Cretaceous Toolebuc oil shales; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 203. HAIG, D. W., 1979: Cretaceous foraminiferal biostratigraphy of Queensland. Alcheringa, 3, 171-87. HERBERT, C., 1980: Prospects for oil shale in the Eromanga Basin, New South Wales. N.S.W. Geol. Surv. Q. geol. Notes, 40, 2-6. KRIEG, G. W., 1982: Stratigraphy and tectonics of the Dalhousie anticline, southwest Eromanga Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 145-58. LUDBROOK, N. H., 1966: Cretaceous biostratigraphy of the Great Artesian Basin in South Australia. 5. Aust. Geol. Surv. Bull. 40. MCMINN, A., 1983: The Toolebuc Formation in the Eromanga Basin of New South Wales. N.S.W. Geol. Surv. Q. geol. Notes, 50, 1-7. MCMINN,

BURGER,

D.,

1986:

Palynology

and

palaeoenvironments of the Toolebuc Formation {sensu lato) in the Eromanga Basin; in This volume. MOORE, P. S. & MOUNT, T. J. (eds), 1982a: The Eromanga Basin Symposium. Geol. Soc. Aust. Abstracts, 5. MOORE, P. S. & MOUNT, T. J. (compilers), 1982b: Eromanga Basin Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide. MOORE, P. 'S. & PITT, G. M., 1982: Cretaceous of the southwestern Eromanga Basin: stratigraphy, facies variations and petroleum potential; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 127-44. MOORE, P. S. & PITT, G. M.,

1985: Cretaceous

subsurface

stratigraphy of the southwestern Eromanga Basin: a review; in Lindsay, J. M. (ed.) Stratigraphy, palaeontology, malacology. Papers in honour of Dr Nell Ludbrook. S. Aust. Geol. Surv. Spec. Publ. 5, 269-86.

114

P. S. MOORE, G. M. PITT & M. E. DETTMAN

M O R G A N , R . , 1977: New dinoflagellate zones and a depositional

model for the Great Artesian Basin. N.S.W. Geol. Surv. Q. geol. Notes, 28, 10-8. M O R G A N , R . , 1980a: Palynostratigraphy of the Australian Early and Middle Cretaceous. N.S.W. Geol. Surv. Palaeont. Mem. 18. M O R G A N , R . , 19806: Eustacy in the Australian Early and Middle Cretaceous. N.S.W. Geol. Surv. Bull. 27. O Z I M I C , S., 1982c: Depositional environment of the Toolebuc Formation and its equivalents, Eromanga Basin, Australia; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 200-2. O Z I M I C , S., 19826; The significance of gamma-ray anomalies in the Cretaceous Toolebuc Formation facies and in their lateral equivalents, Eromanga and Carpentaria Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 229-31. O Z I M I C , S., 1982c: Depositional environment of the oil sh^le hearing Cretaceous Toolebuc Formation and its equivalents, Eromanga Basin, Australia Proc. 15th Oil Shale Symp., Golden, Colorado, U.S.A. Colorado School of Mines Press, Colorado. P I T T , G. M . (compiler), 1978: Explanatory Notes, Murloocoppie 1:250 000 geological map sheet SH53-2. S. Aust. Geol. Surv. P I T T , G. M . & B A R N E S , L . C., 1973: Murloocoppie 1:250 000 geological map. S. Aust. Geol. Surv.

PLAYFORD, G . , H A I G , D. W . & D E T T M A N N , M . E . 1975. A m i d -

Cretaceous microfossil assemblage from the Great Artesian Basin, northwestern Queensland. Neues Jahrb. Geol. Palaont. Abh., 149, 333-62. P R I C E , P. L., 1979A: Palynological laboratory report no. 13/119, Packsaddle No. 3. Mines Administration Pty Ltd (unpubl.). RAMSDEN, A. R., 1983: Microscopic petrography of oil shales at Julia Creek, northwestern Queensland. Geol. Soc. Aust. J. 30, 17-23. R A M S D E N , A . R . , D I C K S O N , B. L . & M E A K I N S , R . L., 1982: Origin and significance of the Toolebuc gamma-ray anomaly in parts of the Eromanga Basin. Geol. Soc. Aust. J. 29, 2 8 5 - 9 6 .

R., 1982: Geochemistry of oil shale in the eastern Eromanga Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 217-28. SCHEIBNEROVA, V., 1980: Comparative foraminiferal biostratigraphy of the Santos Oodnadatta No. 1 section, South Australia. N.S.W. Geol. Surv. Rec. 19, 81-138. SCHEIBNEROVA, V., 1983: Revised correlation of Cretaceous lithostratigraphic units in New South Wales. N.S.W. Geol. Surv. Q. geol. Notes, 52, 1-7. S E N I O R , B. R . , E X O N , N. F. & B U R G E R , D., 1975: The Cadna-owie and Toolebuc Formations in the Eromanga Basin, Queensland. Qld Gov. Min. J. 76, 445-55. S E N I O R , B. R . , M O N D , A . & H A R R I S O N , P. L . , 1978: Geology of the Eromanga Basin. Aust. Bur. Miner. Resour. Geol. Geophys. Bull., 167. SMART, J., 1972: The terms Toolebuc Limestone and Kamileroi Limestone. Qld Gov. Min. J. 73, 280-6. S P R I G G , R. C., 1958: The Great Artesian Basin in South Australia: in Glaessner, M. R. & Parkin, L. W. (eds) The geology of South Australia. Geol. Soc. Aust. J. 5. T H O M S O N , B. P., (compiler) 1980: South Australia, 1:1 000 000 geological map. S. Aust. Geol. Surv. T O W N S E N D , I. J. 1971: Recent logging of New Kopperamanna Bore. 5. Aust. Geol. Surv. Q. geol. Notes, 39, 8-10. VINE, R. R. & DAY, R. W., 1965: Nomenclature of the Rolling Downs Group, northern Eromanga Basin, Queensland. Qld Gov. Min. SAXBY, J.

J. 66, 4 1 6 - 2 1 .

VINE, R . R., DAY, R . W., CASEY, D. J., MILLIGAN, E . N . , GALLOWAY, M . C . . & EXON, N . F., 1967: Revised nomenclature of the Rolling

Downs Group, Eromanga and Surat Basins. Qld Gov. Min. J. 68, 144-SI. Notes, Noolyeeana 1:250 000 geological map sheet SG53-16. S. Aust. Geol. Surv.

W I L L I A M S , A . F., 1976: Explanatory

Geological Society of Australia Special Publication No. 12, 115-118

Marine Cretaceous of the Great Australian Basin— foraminiferal and palynological zonations reconciled Viera Scheibnerova Geological Survey of New South Wales, Geological & Mining Museum, 36 George Street, Sydney, N.S.W. 2000. ABSTRACT Foraminiferal faunas from the uppermost parts of the marine Cretaceous of the Great Australian Basin show that the duration of marine sedimentation was the same in New South Wales as the rest of the Basin. Palynological determinations on certain boreholes in New South Wales show that the last palynological zone was either Crybelosporites striatus or Coptospora paradoxa, and that the youngest zone of the marine section, the Phimopollenites pannosus Zone, is missing. An investigation of the depth of the weathered profile in relation to the depth distribution of positive palynological samples in twelve sections has shown that the base of the weathered zone is relatively deep in the earlier studied New South Wales sections, while being relatively shallow elsewhere. It is suggested that the palynological record was lost due to weathering. The weathered sediments contained foraminifera which elsewhere occurred in the pannosus zone. Thus foraminiferal faunas remain reliable for stratigraphic appraisal of the weathered Cretaceous sediments when palynomorphs are absent. These findings have an important bearing on the recognition of the Toolebuc event, which is considered widely distributed in the Great Australian Basin, and which occurs in a variety of fades..

INTRODUCTION Marine Cretaceous sediments are widespread in eastern and southern Australia and Queensland, forming the well-known structural unit, the Great Australian Basin. During the AptianAlbian and Early Cenomanian most of eastern Australia formed a series of sedimentary basins inundated by a relatively shallow epicontinental sea. The environment within these was remarkably uniform most of the time, although there are important exceptions to this generalization. Foraminiferal studies on four boreholes drilled in northwestern New South Wales (DM Bellfield 1 and la; DM Yantabulla 1; DM Wanaaring 1; DM Weilmoringle 1), showed that in all these sections a complete succession of foraminiferal zones and subzones occurred. Most importantly the top part of the Cretaceous section in all bores contained the same assemblage, that of the L. frankei Subzone. (Scheibnerova, 1975, 1976, 1980; Scheibnerova & Byrnes, 1977). On the other hand, the youngest palynological zone recorded was the striatus Zone in Bellfield and Yantabulla and the paradoxa Zone in Weilmoringle and Wanaaring (Morgan, 1980), although Burger (1969) had determined elements of the paradoxa Zone (K2A) at 53.34 m (175 0 and 45.72 m (150') in DM Bellfield DDH 1. The youngest productive samples of Morgan (1980) in these bores were 39.8 m (130.7'), 53 m (175 0, 54-4 m (177 0 and 53 m (175 0 respectively, this reflecting the depth of weathering. The average sample interval for the palynological logs was about 15.2 m (50 0 in each case, whereas the microfossil sample intervals were as follows: DM Bellfield DDH 1 and la: 300 (162 positive) samples in about 1 m intervals

between 10 m and 305 m; DM Wanaaring DDH 1: 280 (128 positive) samples in about 1 m intervals between 6 m and 305 m; DM Weilmoringle DDH 1: 300 (163 positive) samples between 6 m and 293 m; and in DM Yantabulla DDH 1: 250 (131 positive) samples between about 10 m and 290 m. EFFECT OF WEATHERING ON MICROFOSSIL PRESERVATION More recently, samples became available from two new artesian bores drilled by the Water Resources Commission. One of these sections, Urella Downs WRC 30944, reached a depth of some 540 m (1800 feet). The foraminiferal faunas here are represented by the same sequence of zones and subzones known to occur in the sections drilled close by. McMinn (1983) studied a series of samples from the upper part of both Birrigoolpa and Urella Downs boreholes for palynomorphs. Below the base of the weathered zone he encountered elements of the pannosus Zone. This prompted an investigation of the depth of the weathered profile in relation to the distribution of positive palynological samples. A summary of this information is provided in Figure 2. The first positive palynological and microplankton samples in all studied sections including the New South Wales boreholes come from below the base of the weathered zone, which varies considerably in thickness. In the four earlier studied New South Wales boreholes this left 40-50 m of the youngest Cretaceous sediments without a palynological record. Thus, because there was no record of floral elements indicative of the younger zones in the upper part of the boreholes, it was

140

125°

REFERENCE Od Or T Wa Y Be We UD Bo Bir

-25

Santos Oodnadatta No1 Conorada Ooroonoo No 1 BMR Tickalara DDH 1 DM Wanaaring DDH 1 DM Yantabulla DDH 1 DM Bel I field DDH 1 DM Weilmoringle DDH 1 and Keats High WRC 30944 Urella Downs BMR Boulia 3 and 3A WRC 30961 Birrigoolpa

DSDP Site 258 (Naturallate Plateau) Kilometres

125' I

140

11417

SURFACE WEATHERING AND MICROFOSSIL PRESERVATION

concluded that sediments of the corresponding age were also absent (Morgan, 1977, fig. 3). From Figure 2 it is apparent that the critical parts of the sections of boreholes Wanaaring, Yantabulla, Bellfield and Weilmoringle lie largely within the weathered zone. Whereas foraminiferal assemblages of the youngest Cretaceous zone have been recovered from the weathered zone, the absence of palynological data can obviously not be taken as indicating absence of sediments which correlate with the upper late paradoxa or pannosus age, since these are recorded in those holes (Urella Downs, Birrigoolpa) in which the appropriate sediments occur below the base of the weathered zone, along with foraminifera of the same age as those in the upper parts of the other four boreholes. Thus the conclusion of Morgan (1977) that sediments correlating with the pannosus age are absent cannot be regarded as well-founded; the foraminiferal evidence suggests that it is incorrect. Weathering was most damaging to stratigraphic analysis in those sections where, due to condensed sedimentation, the thickness of Cretaceous sediments is relatively small but the base of the weathered zone deep. The absence of the youngest palynological zone is directly due to this phenomenon. Foraminifera, which are less susceptible to weathering, remain, and the weathered zones contain foraminifera which elsewhere occur with the youngest of palynological zones {pannosus Zone) and form the basis of Subzone 6 (Scheibnerova, 1983). The foraminiferal faunas thus show that the extent of Cretaceous marine sedimentation was the same in all New South Wales sections, and the same as in other parts of the Great Australian Basin.

AREAL DISTRIBUTION OF THE TOOLEBUC EVENT

The above discussion has an important bearing on the problem of the distribution of the Toolebuc Formation and

117

Toolebuc event (Scheibnerova & Byrnes, 1977). The Toolebuc event is characterized by diminished deposition and reducing bottom conditions in the type area of the Toolebuc Formation. Strongly reducing bottom conditions excluded benthic foraminifera and consequently only planktic foraminifera occur in these sediments in the type area. The thickness of the marine Cretaceous sediments varies greatly within the Great Australian Basin from about 740 m down to about 250 m depending on the rate of subsidence during deposition and basement topography. This includes the essentially non-marine Winton Formation which can reach a thickness of some 300 m in the type area. The uniformity of facies is remarkable only in the lower part of the Cretaceous sequence. However, there are appreciable regional differences in facies at the time of the deposition of the Toolebuc Formation (the Toolebuc event). During the Toolebuc event the areas along the axis of the basin received black shales with carbonaceous limestones while elsewhere the unit is not accompanied by black shales, and deposition continued in the mudstone facies, indistinguishable from the underlying sequences. The event represents a phase of slow deposition during one of the widespread transgressions in Cretaceous time with notable influx of planktic foraminifera, fish and ammonites (Day, 1969). The Toolebuc event is often associated with a subsurface gamma-ray anomaly due to uranium and actinium absorbed on carbonaceous matter, sometimes bituminous (Reynolds, 1968; Jesson & Radeski, 1964; Senior et al. 1975). The Toolebuc Formation has been correlated lithologically with the Wooldridge Limestone Member of South Australia by Senior et al (1975) and with the Kamileroi Limestone (Smart, 1972) in the Carpentaria Basin. This correlation is supported by palaeontological evidence. Where known, the Toolebuc Formation in its typical lithology or its associated facies (e.g. Wooldridge Limestone) is preceded by the pannosus Datum horizon. In BMR Tickalara 1 in southwestern Queensland the pannosus Datum y

118

V. SCHEIBNEROVA

horizon occurs at about 110 m depth (the section has not been affected by weathering), but none of the lithologies typical for the Toolebuc Formation has been recorded above it. This prompted the conclusion that the typical Toolebuc Formation lithology occurs only in restricted areas and elsewhere lacks the characteristic features such as planktic foraminifera, black shale and limestone, abundant carbonaceous matter, gammaray anomaly recorded on wireline logs and Inoceramus prisms (Scheibnerova & Byrnes, 1977). It was suggested that the equivalent of the Toolebuc Formation must also occur in New South Wales. This suggestion has now been confirmed by further drilling in New South Wales in boreholes BMR Urisino 1, WRC Birrigoolpa 39061 and WRC Urella Downs 30944. In these sections McMinn (1983) determined the pannosus Zone, although no lithology typical of the Toolebuc Formation has been observed. In these sections the term Toolebuc event is more appropriate since the event is only reflected in the foraminiferal and microfloral assemblages and sudden increase of fish remains. However, in the light of the new evidence the Toolebuc event as distinguished by Scheibnerova & Byrnes (1977) in the four New South Wales sections occurs within the weathered zone and not within the striatus or paradoxa Zones. In all studied sections two zones of increased accumulation of fish remains have been recorded (Scheibnerova, 1980; in press). Freeman (1964) used the so-called ''Fish Scale'Zone" (a correlative of the Toolebuc Formation) in defining a basal zone of the subsurface Tambo Formation. I suggest the

informal term 'fish zone' since fish scales only form an insignificant part of the fish remnants in it. Far more abundant are parts of fish skeletons (vertebrae and teeth). The lower fish zone is closely associated with the L. albiensis albiensis Datum horizon, always being preceded by it, while the upper fish zone appears shortly after the L. frankei Datum horizon. The upper fish zone is associated with the Toolebuc Limestone in the type area; it is of interest that the lower fish zone in at least one Queensland section (Conorada Ooroonoo 1 at 600 m) is associated with another limestone horizon so far unnamed. This limestone horizon is strongly radioactive and was misinterpreted as the 'Toolebuc Formation' by McPhee (1963). The true Toolebuc Formation (limestone) in the Ooroonoo section occurs at about 300 m (being preceded by the pannosus Datum horizon) and does not show enough radioactivity to cause a gamma-ray anomaly. In the Oodnadatta section the lower fish zone occurs between 260 m and 230 m and the upper one between 185 and 45 m. The Wooldridge Limestone Member occurs within the upper fish zone and an unnamed limestone occurs between 240 and 220 m just above the Lingulogavelinella albiensis albiensis Datum horizon. Neither of the above limestone horizons occurs in the New South Wales sections, although stratigraphically equivalent sequences are present.

ACKNOWLEDGEMENT

Permission to publish this paper was given by the Secretary, New South Wales Department of Mineral Resources.

REFERENCES

1969: Palynological observations of Triassic and Cretaceous strata at the border of Queensland and New South Wales. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1969/96 (unpubl.). DAY, R. W., 1969: The Lower Cretaceous of the Great Artesian Basin; in Campbell, K. S. W. (ed.) Stratigraphy and Palaeontology. Essays in honour of Dorothy Hill. Aust. Nat. Univ. Press, Canberra, 1 4 0 - 1 7 3 . F R E E M A N , R. M., 1964: Oil exploration in the western Great Artesian Basin. Proc. Aust. Inst. Min. Met. 211, 85-114. JESSON, E. E. & RADESKI, A., 1964: Great Artesian Basin, bore logging, Queensland 1962. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1964/103 (unpubl.). M C M I N N , A., 1983: The Toolebuc Formation in the Eromanga Basin of New South Wales. Q. geol. Notes, Geol. Surv. N.S.W. 50, 1-7. M C P H E E , I., 1963: Conorada Ooroonoo No. 1, Queensland. B.M.R. Geol. Geophys. Pet. Search Subsidy Acts Publ. 23, 1-30. M O R G A N , R., 1977: New dinoflagellate zones and a depositional model for the Great Australian Basin. Q. geol. Notes, Geol. Surv. N.S.W. 28, 10-18. M O R G A N , R., 1980a: Palynostratigraphy of the Australian Early and Middle Cretaceous. Mem. Geol. Surv. N.S.W. Palaeont. 18, 153 pp. M O R G A N , R., 1 9 8 0 6 : Eustacy in the Australian Early and Middle Cretaceous. Bull. Geol. Surv. N.S.W. 27, 105 pp.

B U R G E R , D. M . ,

M. A., 1968: Explanatory Notes on the Bedourie 1:250 000 Geological Sheet. Aust. Bur. Miner. Resour. Geol. Geophys. 54-1, 19 pp. V., 1975: Correlation of foraminiferal and palynological datum horizons, zones and subzones in the Cretaceous of the Great Australian Basin. Q. geol. Notes, Geol. Surv. N.S.W. 21, 8-17. SCHEIBNEROVA, V., 1976: Cretaceous foraminifera of the Great Australian Basin. Mem. Geol. Surv. N.S.W. Palaeontology, 17, 277 pp. SCHEIBNEROVA, V., 1980: Comparative foraminiferal biostratigraphy of the Santos Oodnadatta No. 1 section, S.A. Rec. Geol. Surv. N.S.W. 19(1), 81-139. SCHEIBNEROVA, V., (in press): Biostratigraphy of Marine Cretaceous of the Great Australian Basin. Geol. Rec. Geol. Surv. N.S.W. SCHEIBNEROVA, V. & BYRNES, J. G., 1977: Correlation of Cretaceous lithostratigraphic, lithological, palynological, and foraminiferal units in the Great Australian Basin. Q. geol. Notes, Geol. Surv. N.S.W. 27, 1-11. S E N I O R , B. R . , E X O N , N. F. & B U R G E R , D., 1975: The Cadna-owie and Toolebuc Formation in The Eromanga Basin. Qld Govt Min. REYNOLDS,

SCHEIBNEROVA,

J. 76, 4 4 5 - 5 5 . SMART, J., 1972: The terms Toolebuc Limestone and

Limestone. Qld Govt Min. J. 73, 280-86.

Kamileroi

Geological Society of Australia Special Publication No. 12, 119-137

The geology and petrophysics of the Toolebuc Formation and its time equivalents, Eromanga and Carpentaria Basins. S. Ozimic

Bureau of Mineral Resources, Geology and Geophysics, P.O. Box 378, Canberra City, A.C.T. 2601.

ABSTRACT

The widespread Early Cretaceous oil shale-bearing Toolebuc Formation in the northern Eromanga and southern Carpentaria Basins consists of various proportions of calcareous oil shale and coquinite. To the south the formation passes laterally into time-equivalent sequences lacking oil shale; the Wooldridge Limestone Member of the Oodnadatta Formation and the Urisino beds, which consist of coquinite, limestone, siltstone and sandstone. The oil shale appears to have resulted from the accumulation and preservation of planktonic and mat-like algae in a relatively deep (below wave base) restricted, 'positive water balance marine basin' that shallowed southwards. The normal anoxic conditions were periodically interrupted as benthonic shelly scavengers became established in parts of the basin. Oil shales in petrophysically logged bore holes are invariably associated with a marked gamma-ray anomaly. The reverse is not true; gammaray anomalies occur within the Wooldridge Limestone Member, and below the Urisino beds. The conditions favouring oil shale deposition were apparently terminated with a return of normal marine conditions, probably the result of increase in saltwater inflow with rising sea level and possibly due to variations in conditions limiting productivity in the euphotic zone. The estimated Toolebuc Formation productive oil shale and shale oil resources in the Eromanga and Carpentaria Basins are 3.5 x 10' m and 245 x 10 m respectively. 2

INTRODUCTION

This paper presents the results of a geological and petrophysical study of the Toolebuc Formation and its time equivalents in the Eromanga Basin and southern Carpentaria Basin (Fig. 1). The study forms part of a research project— 'Oil Shale Methodology—being carried out jointly by the Bureau of Mineral Resources, Geology and Geophysics (BMR), and the Commonwealth Scientific and Industrial Research Organization (CSIRO), (Ozimic & Saxby, 1983). The project was partly funded by the National Energy Research Development and Demonstration Council (NERDDC). One of the project's major objectives was to investigate the geological and petrophysical controls on the nature and distribution of oil shale within the Toolebuc Formation. To understand the constraints on accumulation of oil shale over large areas it was necessary to investigate the differences between the oil shale-bearing and adjacent barren sequences. In the Eromanga and southern Carpentaria Basins, the sequences of interest are the kerogenous (oil shale-bearing) Toolebuc Formation, its stratigraphic correlatives, the time equivalent Woolridge Limestone Member of the Oodnadatta Formation, and the Urisino beds; the underlying Coreena Member of the Wallumbilla Formation and lower part of the Oodnadatta Formation; and the overlying Allaru Mudstone and younger Oodnadatta Formation (Fig. 2). The study involved petrological examination of cores, cuttings and outcrop samples, qualitative evaluation of

petrophysical logs, palynological age determination on which the basinwide correlation is based, and a literature review. The stratigraphy and geological history of the studied sequences are described with reference to facies and thickness maps (Figs 1, 3) and the structure is illustrated by a structure-contour map (Fig. 4). Figure 1 shows the data control based on petroleum exploration wells, stratigraphic holes, water bores and holes drilled as part of the 'Oil Shale Methodology' project. Each well, hole and bore is identified on Figure 1 by a number and a reference given in Table 1.

BASIN SETTING

The Eromanga and Carpentaria Basins contain a sequence of continental Jurassic and Lower Cretaceous rocks 2000-3000 m thick. The geology and stratigraphy of the Eromanga Basin have been discussed by Senior et al. (1978) and of the Carpentaria Basin by Smart et al. (1980). The Er-omanga Basin connects with the Carpentaria Basin to the north across the Euroka Arch. It also connects with the Surat Basin to the east across the Nebine Ridge (Fig. 1); Surat Basin rocks are probably all older than the Toolebuc Formation. In the Eromanga and Carpentaria Basins the Toolebuc Formation and its equivalents are thin units within a succession up to 2000 m thick, which consists of deltaic, marine and lacustrine mudstone, siltstone, sandstone, coquinite, limestone and minor coal, deposited during marine advances and regressions in Early to mid-Cretaceous times.

S. OZIMIC

120

T o o l e b u c Formation Fades

Fades

GULF

CARPENTARIA CORAL

CARPENTARIA

W o o l d r i d g e Limestone Member

BASIN Urisino beds Anticline Fault

Townsville

Outcrop Fades

boundary

Present

basin

Extent

of the Toolebuc

and its

equivalents

Control points

margin Formation • Mount Isa

with reference

^51

Petroleum

exploration

® 10

Stratigraphic

hole

0105

Water

Reference

numbers well

BoulTa\\

bore section

~ /

3ABoulia

[

NT|QLD

Ju}P0117 Rone ~ J \STORMHILL J FAULT ~

Bedourie

[reach 6

EROMANGA:

basin: 'ooldrii

,v/:Q(»arleyille

(6 Creek

Oodnadattap'61

Urisino

Fig. 1. Location and facies map, Eromanga and southern Carpentaria Basins (refer Table 1 for borehole references).

GENERAL CHARACTERISTICS OF TOOLEBUC and calcareous shale interbedded with coquinite, and is FORMATION associated with a moderate gamma-ray anomaly. Senior et In northern parts of the Eromanga Basin and southern Carpentaria Basin the Toolebuc Formation is the major calcareous unit of the succession. It was a product of Early Cretaceous transgression from the north, which caused a change from subaerial and paralic conditions to a shallow marine environment in which a predominantly mudstone sequence with many minor calcareous beds was laid down. The type section of the Toolebuc Formation has been defined in BMR Boulia 3A by Senior et al. (1975) as the intersection between 25.3 and 35.8 m (Fig. 1). In its type section the Toolebuc Formation consists of black bituminous

al. (1978) attributed the anomaly to uranium-bearing phosphate minerals associated with shell debris. The anomaly was correlated with the Toolebuc Formation throughout the Eromanga and southern Carpentaria Basins without being related to a particular lithology. As described in this paper the Toolebuc Formation occurs in the northern Eromanga and southern Carpentaria Basins (Fig. 1). It comprises two facies, each consisting of varying proportions of kerogenous and calcareous shale (oil shale) interbedded with coquinite. It crops out along the margins of the basins as calcareous rubble, in which the oil shale component cannot be recognised due to weathering.

TOOLEBUC FORMATION PETROPHYSICS

121

Fig. 2. Stratigraphy and correlation of Toolebuc Formation, Wooldridge Limestone Member and Urisino beds. (Locations): 1—Wooldridge Creek, S.A.; 2—BMR Urisino 1 well; 3—BMR Boulia 3A well; 4—BMR Longreach 6 well.

Roughly south of a line joining Charleville and Bedourie in Queensland the formation apparently passes laterally into the time-equivalent non-oil shale-bearing sequences of the Wooldridge Limestone Member of the Oodnadatta Formation and the Urisino beds (Fig. 1).

these correlations and demonstrates age equivalence of this unit with the oil shale-bearing Toolebuc Formation, Urisino beds and Wooldridge Limestone Member of the Oodnadatta Formation.

Stratigraphic nomenclature problems

STRATIGRAPHIC AND FACIES VARIATIONS Toolebuc Formation

In the course of the 'Oil Shale Methodology' project, difficulties arose from the complexity of stratigraphic nomenclature that has been developed in South Australia and Queensland; partly to identify different facies provinces in the Eromanga Basin, and partly through lack of co-ordination of the mapping programs. One of the casualties of this development is the Toolebuc Formation. Over the years several suggestions have been made as to where the Toolebuc Formation fitted in the sequence in northern South Australia by Ludbrook (1966), Freytag (1966), Exon et al. (1972), Scheibnerova (1978), Ozimic (1982), and Moore & Pitt (1982), however no formal agreement has been reached. In this paper the term Wooldridge Limestone Member refers to a non-oil shale-bearing rock unit intersected within the Oodnadatta Formation in numerous petroleum exploration wells and water bores in northern and central South Australia and in southeastern Northern Territory. The unit was correlated with the oil shale-bearing Toolebuc Formation in Queensland and with the Urisino beds in Queensland and New South Wales, on the basis of lithologies, petrophysical characteristics and relative stratigraphic position. Palynological review of the Albian rock sequence in the Eromanga Basin by McMinn & Burger (this volume) supports

The Toolebuc Formation consists wholly of facies A and facies B (Fig. 1), which are each characterised by the same lithologies, age, sedimentary and biogenic features, and boundary characteristics. However, a marked difference in the ratio of coquinite to kerogenous shale between the two facies is the basic criterion for their differentiation. Facies A (Ozimic, 1982) is defined in the type section. It extends over two-thirds of the northwestern Eromanga Basin and most of the southern Carpentaria Basin (Fig. 1), and covers an area of 0.39 x 10 knr. It is thin, with an average thickness of 20 m; and reaches a maximum thickness of 35 m in the Winton area (Fig. 3). It crops out along the margins of both basins mainly as calcareous rubble; the oil shale component is strongly weathered and not recognisable. To the south beyond the 'Charleville-Bedourie' line (Fig. 1) the facies can no longer be recognised. It apparently interfingers with the non-oil shale-bearing, time-equivalent Urisino beds and Wooldridge Limestone Member of the Oodnadatta Formation. To the east, probably in the vicinity of the Berry Anticline and the 'Canaway Fault Zone' (Fig. 4), facies A passes into facies B. 6

122

S. O Z I M I C

TABLE 1. Stratigraphic control points Control Point No. Borehole Reference 1 Westmoreland 1 Burt (1976) 2 Westmoreland 2 Burt (1976) 3 Westmoreland 3 Burt (1976) 4 Croydon 1 Burt (1976) 5 Croydon 4 Burt (1976) 6 Dobbyn 1 Burt (1976) 7 Dobbyn 2 Burt (1976) Burt (1976) 8 Dobbyn 3 Burt (1976) 9 Dobbyn 4 Burt (1976) 10 Millungera 1 Burt (1976) 11 Millungera 2 Burt (1976) 12 Millungera 4 Burt (1976) 13 Millungera 5 14 Julia Creek 1 Burt (1976) Burt (1976) 15 Julia Creek 2 16 Julia Creek 3 Burt (1976) Burt (1976) 17 Julia Creek 4 Burt (1976) 18 Richmond 1 Burt (1976) 19 Richmond 2 20 Richmond 3 Burt (1976) 21 Manuka 1 Balfe (1978) 22 Boulia 1 Burt (1976) 23 Boulia 2 Burt (1976) 24 Boulia 3 Burt (^976) Burt (1976) 25 Boulia 4 Senior et al. (1975) 26 Boulia 3A 261 Boulia 10B Gibson (1982) Senior & Hughes (1972) 27 Springvale 6 Senior & Hughes (1972) 28 Springvale 6 Gibson (1982) 29 Springvale 6 30 Longreach 6 Ozimic (198la) 31 Jericho 11 Ozimic (1981a) 32 Springvale 5 Pan Pacific (1982) 33 The Brothers 1 Magnier & Sweeney (1965) 34 Tambo 37 Ozimic (1981a) 35 Tambo 41 Ozimic (1981a) 36 Tambo 42 Ozimic (1981a) 37 Tambo 44 Ozimic (1981a) 38 Betoota 1 Delhi-Frome-Santos (1961) 39 Cothalow 1 Lewis (1961) 40 Gumbardo 1 Phillips-Sunray (1963) 41 Leopardwood 1 McDonagh et al. (1966) 42 Gilmore 1 Lewis & Kyranis (1965) 43 Log Creek 1 Kyranis & McDonagh (1966) 44 Etonvale 1 Lewis & Kyranis (1962) Netzel (1967) 45 Stafford 1 46 Bonnie 1 Kyranis (1966a) 47 Augathella 5 Ozimic (1981a) 48 Augathella 6 Ozimic (1981a) Ozimic (1981a) 49 Augathella 7 Ozimic (19816) 50 Charleville 5 Stephenson (1982) 51 Charleville 6 Stephenson (1982) 52 Charleville 7 Stephenson (1982) 53 Toompine 1 54 Urisino 1 Ozimic (1981a) Amerada (1966) 55 Hale River 1 Harrison et al. (1963) 56 Pandieburra 1 Cadart (1969) 57 Budgerygar 1 Cadart (1968) 58 Thunda 1 Hess (1957) 59 Oodnadatta 1 Harrison & Higginbotham (1964) 60 Gidgealpa 1 Griffiths (1980) 61 Toodla 1 62 Barcoo Junction 1 Sykes (1981) 63 Eromanga 1 GSQ (1981) 64 Naryilco 1 Campbell et al. (1963a) Ryan (1961) 65 Innamincka 1 Harrison & Greer (1963) 66 Dullingari 1 Greer & Harrison (1963) 67 Orientos 1 Campbell et al. (19636) 68 Putamurdie 1 Minad (1963) 69 Penrith 1

Control Reference Borehole Point No. Aquitaine (1965a) 70 Fermoy 1 Ozimic & Pain (1983) 71 Hay River 12 Ozimic & Pain (1983) 72 Finke 2 Amerada (1965) 73 McDills 1 Habermehl & Morrissey (1983) 74 RWB 391 Aquitaine (19656) 75 Mayneside 1 Evans (1966) 76 Newlands 1 77 Stormhill 1 Warns (1970a) Wiltshire (19736) 78 Ban Ban 1 Meyers & Spencer (1970) 79 Muttaburra 1 Amerada (1967) 80 Thunderbolt 1 Watson (19736) 81 Clyde 1 82 Beryl 1 Minad (1964) Casey (1969) 83 Tangorin 1 84 Clayton Bore SA Geol. Survey 85 Well C.K. Bore Ludbrook (1966) Habermehl & Morrissey (1983) 86 RWB 1474 Habermehl & Morrissey (1983) 87 RWB 378 Habermehl & Morrissey (1983) 88 RWB 2977 Habermehl & Morrissey (1983) 89 RWB 1043 Habermehl & Morrissey (1983) 90 RWB 2655 Habermehl & Morrissey (1983) 91 RWB 3264 Habermehl & Morrissey (1983) 92 RWB 2552 Habermehl & Morrissey (1983) 93 RWB 379 H a b e r m e h l ^ Morrissey I (1983) 94 RWB 2549 Habermehl & Morrissey (1983) 95 RWB 3268 96 Chandos South 1 Laing & Cadart (1970) 97 RWB 1674 Habermehl & Morrissey (1983) Habermehl & Morrissey (1983) 98 RWB 1676 99 RWB 4333 Habermehl & Morrissey (1983) Habermehl & Morrissey (1983) 100 RWB 4339 Habermehl & Morrissey (1983) 101 RWB 14486 102 Thomas 1 Wiltshire (1982) 103 Towerhill 1 Amoco (1966) 104 Brookwood 1 Pemberton (1963) 105 RWB 1645 Habermehl & Morrissey (1983) 106 RWB 1652 Habermehl & Morrissey (1983) 107 RWB 1653 Habermehl & Morrissey (1983) 108 RWB 2348 Habermehl & Morrissey (1983) 109 RWB 3860 Habermehl & Morrissey (1983) 110 RWB 13234 Habermehl & Morrissey (1983) 111 RWB 13944 Habermehl & Morrissey (1983) 112 Ooroonoo 1 Conorada (1960) 113 Belmore 1 Warris (19706) 114 Glenaris 1 McDonagh (1967) 115 Saltern Creek Mott & Associates (1964a) 116 Longreach Oil 4 Mott (19556) 117 Longreach Oil 1 Mott (1955a) 118 Marchmont 1 Mott & Associates (1964c) 119 Cumbroo 1 Campe (1969) 120 Fairlea 1 Garrett & Patterson (19686) 121 Carlow Kyranis (19666) 122 Barcoo 1 Garrett & Patterson (1968a) 123 Boree 1 Gerrard (1964) 124 Colson 1 Wiltshire (1978) 125 Canary 1 Green et al. (1963) 126 Yongala 1 Laing (1966) 127 Yongala 1 Laing (1966) 128 Galway 1 Jacque & Sweeney (1966) 129 Merrimelia 1 Delhi-Frome-Santos (1965) 130 Bury 1 Patterson (1966) 131 Kumbarie 1 Casey (1970) 132 Karmona 1 Wiltshire & Imbert (19776) 133 Gurra 1 Bowering (1970) 134 Gilpeppee 1 Mulready (1970a) 135 Durham Downs ]i Wiltshire (1973a) 136 Coongie 1 Pemberton (1970) 137 Cherri 1 Pexa Oil N.L. (1970) 138 Brumby 1 Lunt (1972) 139 Binerah Downs 1 Hanlon (1966)

123 TOOLEBUC FORMATION PETROPHYSICS upper boundary of facies A in most reference sections is Control marked by an obvious lithological change from coquinite Point No Borehole layers to the overlying massive montmorillonite-rich Allaru Reference Mudstone. The upper contact boundary of facies B is generally 140 Yanpurra 1 LeGay Brereton (1971) gradational from kerogenous shale to the massive Allaru 141 Toolachee East 1 Delhi-Santos-Vamgas (1972) Mudstone. 142 Tinga Tingana 1 Delhi (1968) 143 Tanbar 1 Aquitaine (1976) The totally marine fauna, presence of fine laminations and 144 Tallalia 1 Pemberton & LeGay Brereton (1971) lack of sharp erosional boundaries indicate quiet marine 145 Roseneath 1 Jacque et al. (1970) conditions during deposition of the Toolebuc Formation. The 146 Packsaddle 1 Mulready (19706) great extent of the Toolebuc Formation (0.484 x 10 km ), 147 Nappacoongee 1 Delhi (1966) and absence of sand bodies or reef structures that could have Hardy (1970) 148 Murteree 1 formed lagoon barriers, suggest the existence of a large body Delhi-Santos (1966) 149 Mt Howitt 1 of water. The nature of the organic matter, and regular fine Jacque (1966) 150 Mokari 1 laminations, could not have developed on the bottom of a Magnier (1964) 151 Witcherrie 1 GSQ (1960-65) 152 Warbreccan 1 Toolebuc sea' in which the water was agitated by waves or Mott & Associates (19646) 153 Hulton 1 currents. Preservation of organic matter was possible because Magnier (19646) 154 Purni 1 of poor oxygenation and a probable lack of scavengers in the Wiltshire & Imbert (1977cr) 155 Barrolka 1 lower part of the water column, and from the absence of waves Hematite (1974) 156 Weston 1 with possible bases 50 m below the surface. Recognition of Warris (1969) 157 Rand 1 telalginite and lamalginite (Hutton & Cook, 1983) suggests 158 Lovelle Downs 1 Watson (1973o) algal productivity in the euphotic zone and growth of algal mats on the seabed. The preservation of dead algal matter Facies B (Ozimic, 1982) is defined in the reference section can be related to an oxidising-reducing boundary probably BMR Longreach 6 (Fig. 5). This facies extends along the situated immediately below the base of the living algal mat eastern part of the northern Eromanga Basin (Fig. 1) and layer, and keeping pace with its upward growth. Bubela (1980) covers an area of 0.094 x 10 km . It appears to be an found experimentally that algal mats constitute an effective extension of facies A from west to east across the Berry boundary between the oxidising environment of the water in Anticline and the Canaway Fault Zone (Fig. 5) with an average which they grow and the strongly reducing environment of thickness of 12 m; it reaches a maximum thickness of 20 m the substrate (e.g. sediments or dead algal mats). Mat (Fig. 3). To the south this facies apparently grades into the destruction can, however, result from environmental stresses time equivalent Urisino beds. such as salinity, temperature, and light changes. The Toolebuc Formation lithologies consist of black to dark The coquinite layers represent recurrent conditions suitable grey, laminated kerogenous and calcareous shale (oil shale) for the establishment of specialised, low oxygen-tolerant, interbedded with light grey- calcite laminae, which when large-sized benthonic shelly faunas. Periodic disappearance abundant, forms coquinite (Fig. 6a-d). The kerogenous shale of those faunas is speculatively attributed to the relative displays abundant fine dark brown organic laminae set in a productivity of phyto- and zooplankton, which could have microcrystalline calcite and clay matrix. The matrix rarely depleted oxygen levels in the lower water column, and, or exceeds 10 per cent of the total rock. Pyrite is common and contaminated the water column with plankton debris, so that occurs as well-developed framboids. Foraminifera and conditions were unsuitable even for low oxygen-tolerant filter phosphatised skeletal fish remains occur throughout. feeders, such as Inoceramus and Aucellina. Kerogenous laminae are typically about 10 mm thick, and The evaluated coquinite/kerogenous shale ratios and their are opaque or translucent dark reddish-brown in thin sections. areal variation suggest that living conditions were more The coquinite comprises multiple laminae of crystalline favourable along the western basin margin, whereas along the calcite, which are shells of the pelecypods Inoceramus and eastern basin margin the growth of shelly fauna was severely Aucellina (Fig. 6b-d). The layers are up to 4.0 mm thick and restricted, probably from very low oxygen levels. show tightly interlocking prismatic calcite crystals that have grown perpendicular to the surface of the pelecypod shells. Wooldridge Limestone Member In core samples the laminae are generally regular, partly The Wooldridge Limestone Member of the Oodnadatta recrystallised and some show development of ellipsoidal micro-concretions. In outcrop samples the laminae (shell Formation was defined by Freytag (1966), with the type section fragments) extend laterally up to 0.5 m and retain an almost designated at Wooldridge Creek in South Australia (Fig. 1). It is of the same age as the Toolebuc Formation (McMinn constant thickness between 2.5 and 4.0 mm. this- volume) and covers approximately 0.237 x The estimated coquiniterkerogenous shale ratio of the two &10 Burger, km in the southwestern part of the Eromanga Basin facies varies considerably. For facies A it ranges from 1.50 (Fig. 1). It reaches a maximum thickness of 25 m in the in Julia Creek area to 0.70 in the Boulia area. For facies B the ratio ranges from 0.24 at Barcaldine, to 0.05 at Charleville. northeastern corner of South Australia (Fig. 3), average The contact boundary of the two facies with the underlying thickness being 14 m. To the north, Wooldridge Limestone Wallumbilla Formation is generally regarded as conformable Member apparently passes laterally into the Toolebuc (Senior et al., 1978). However, a minor break in sedimentation Formation facies A, and to the east into the time-equivalent where facies B rests on an erosional surface on top of the Urisino beds. In the subsurface the member consists of glauconitic Coreena Member implies a disconformable contact. Lithologies of the two facies at the lower boundary vary from calcareous siltstone and sandstone interbedded with coquinite gradational (from underlying indistinctly bedded silty (Harrison et al., 1963; Campbell et al., 19636; Wiltshire, 1982). mudstone into finely laminated kerogenous shale) to sharp Beds are up to 0.5 m thick. Fossils include pelecypods, (a conglomerate bed at the top of the Coreena Member is foraminifera, fish remains (Scheibnerova, 1978), spores, pollen abruptly overlain, by finely laminated kerogenous shale). The and dinoflagellates (Dettmann & Playford, 1969; Burger,

TABI.L 1. Stratigraphic control

points

124

S. OZIMIC GULF

CARPENTARIA CORAL CARPENTARIA \

Thickness contour fm)

Fault

mm Outcrop

Fades boundary

BASIN

Townsvillei^

Present basin margin Extent of the Toolebuc Formation and its equivalents Control points with reference numbers £f5i Petroleum exploration well ® 1o

Stratigraphic hole

O105 V\/ater bore tx

Reference section

Boulia

NT|QLD Bedourie

/ / / / / 6 EROMANGA*

BASIN . y Charleville ^ ) i

Oodnadatta

Fig. 3. Isopach map of Toolebuc Formation, Wooldridge Limestone Member and Urisino beds.

1981). The siltstone is grey or greenish, calcareous and in places volume) palynologically correlated this sandy siltstone unit contains coaly material. The sandstone consists of very fine with the upper part of the Wallumbilla Formation in to coarse angular to sub-angular quartz and feldspar, Queensland. The occurrence of calcareous siltstone with coaly cemented by calcite. Both siltstone and sandstone contain up material, together with glauconitic sandstone and coquinite, to 20 per cent glauconite pellets. The coquinite beds are light possibly indicates fluctuating energy conditions in a grey, show reworking, and in places contain ellipsoidal transitional depositional environment with both non-marine concretions. and marine influences. The water body must have been The upper boundary of the Wooldridge Limestone Member sufficiently alkaline for the precipitation of calcite cement, is taken to be the top of a coquinite layer that is overlain by but slightly reducing at times, suitable for carbonaceous silty montmorillonite-rich shale. In places this boundary material to be preserved. grades from calcareous siltstone into silty micaceous shale. The lower boundary is taken to be the base of a coquinite Urisino beds layer that overlies a sandy siltstone unit at the base of the The Urisino beds were defined by Ozimic (1982) in BMR Oodnadatta Formation (Fig. 2). McMinn & Burger (this Urisino 1 stratigraphic hole (Fig. 7), and have also been

TOOLEBUC FORMATION PETROPHYSICS

GULF

125

CARPENTARIA CORAL

—Q— Structural contour (m)

CARPENTARIA

Datum: Mean sea level

—|—

Anticline

—^—

Sync/ine

—|—

Monocline

BASIN

Fault Townsvilleis

mm Outcrop Fades boundary '

Present basin margin Extent of the Toolebuc Formation and its equivalents Control points with reference numbers £f5i

Petroleum exploration well

®io

Stratigraphic hole

Oi°5 Water bore ex

Reference section

NT|QLD I CANAWAY^ FAULT ZONE'

^EROMANGA

BASIN Charleville

Oodnadatta^

Fig. 4.

Structure on top of Toolebuc Formation, Wooldridge Limestone Member and Urisino beds.

identified in a number of other B M R stratigraphic holes (Stephenson, 1982); however, they have not been recognised in outcrop. The Urisino beds in the southeastern Eromanga Basin correlate stratigraphically with the Toolebuc Formation, and the Wooldridge Limestone Member (McMinn & Burger, this volume). To the north along the eastern part of the 'Bedourie-Charleville' line the beds apparently pass into facies A of the Toolebuc Formation, and to the west into the Wooldridge Limestone Member (Fig. 1). The unit reaches a maximum thickness of 25 m in South Australia, the average thickness being 16 m (Fig. 3) covering an area of 0.228 x 106 km 2 . In B M R Urisino 1 the sequence is 20 m thick (Fig. 7) and consists of siltstone, glauconitic sandstone and a few thin beds

of unfossiliferous limestone (Ozimic, 1981cr; Byrnes, 1980). Individual beds are up to 3 m thick. The siltstone and sandstone contain glauconite pellets and coaly matter, and appear to be cemented by siderite and minor pyrite. The limestone bed at the base of the Urisino beds consists of lightgrey, anastomosing, 2 to 50 mm thick bands of calcite. Fossils include foraminifera, fish remains (Scheibnerova, 1981), dinoflagellates, spores and pollen (McMinn, 1980; Burger, 1981). No megafossils have been found. The upper boundary is taken to be the top of a limestone bed (Fig. 7) which is overlain by a mudstone defined as the basal Allaru Mudstone. The lower boundary is taken as the base of an unfossiliferous limestone bed overlying a sandy siltstone unit that is thought to be equivalent in age to the

S. OZIMIC

126

—1] Sands and silts

Conglomerate

Shale -^-t] Kerogenous shale, minor coquinite

Sandstone 'wv* Erosiona! surface

Fig. 5. Composite well log, BMR Longreach 6.

Coreena Member of the Wallumbilla Formation in Queensland, and to the lowest part of the Oodnadatta Formation in South Australia (Fig. 2). The predominance of silt-sized detrital material, carbonaceous matter, and the diagenetic minerals siderite and pyrite, suggests deposition of the beds under reducing low-energy conditions in a deltaic environment.

Coreena Member

The Coreena Member of the Wallumbilla Formation was defined by Vine et al. (1967) near the Cofeena Station, Longreach 1:250 000 geological sheet area. In the course of the 'Oil Shale Methodology' project's drilling, the base of

the Coreena Member was not intersected; maximum thickness reached was 34.4 m in BMR Augathella 6 (Ozimic, 198k). Project holes penetrated massive, light-grey to greenish, generally homogeneous volcanogenic and in places, calcareous sandstone. The sandstone is glauconitic, carbonaceous, pyritic and sideritic. Notable variations in this member from south to north are: (1) diminishing grain size (coarse to fine), (2) diminishing scale of cross-bedding (macro to micro), (3) diminishing content of argillaceous matrix, (4) diminishing content of siderite and carbonaceous material, and (5) increasing calcite cement.

TOOLEBUC FORMATION

PETROPHYSICS

( b )

(c^

127 ( d )

Fig. 6. Cored lithologies and sedimentary structures of Toolebuc Formation, (a) finely laminated kerogenous shale (BMR Augathella 6). (b) finely laminated kerogenous shale with sparse crystalline calcite laminae (white), (BMR Boulia 3A). (c) irregularly banded kerogenous shale and recrystallised coquinite, showing small scale intrabed deformations due probably to differential compaction (Shell Dev. (Aust.) Julia Creek 4). (d) strongly deformed, irregularly banded coquinite with laminae of kerogenous shale, showing soft sediment deformation, small scale faulting and uneven laminae parting (Shell Dev. (Aust.) Julia Creek 4). T h e u p p e r b o u n d a r y of t h e C o r e e n a M e m b e r is taken as the t o p of either a massive s a n d s t o n e , or a layer of p h o s p h a t i s e d pebbles, which are overlain a l o n g t h e western margin of t h e b a s i n by either k e r o g e n o u s shale or coquinite, defined as Toolebuc F o r m a t i o n facies B, a n d by unfossiliferous limestone a n d s a n d y siltstone of the U r i s i n o beds. T h e g a m m a - r a y log shows a distinct decrease in r a d i a t i o n o n entering the m e m b e r f r o m the overlying unit. A strong increase in resistivity indicates the presence of low salinity pore fluids. T h e n o r t h w a r d c h a n g e s s u g g e s t d e p o s i t i o n in a n e n v i r o n m e n t a f f e c t e d by a n o r t h w a r d decrease in fluvial influences a n d current activity, probably towards a m o r e open sea T h e a b u n d a n c e of early diagenetic minerals reflects

c o n s i d e r a b l e r e d u c i n g c o n d i t i o n s at or b e l o w t h e w a t e r / s e d i m e n t interface. T h e n o r t h w a r d increase in calcite cement, interpreted as having been precipitated directly f r o m calcium-rich m a r i n e waters, s u p p o r t s the suggested d e p o s i t i o n a l setting.

Allaru

Mudstone

T h i s unit was m a p p e d by Vine et al. (1967), in n u m e r o u s p a r t s of the R i c h m o n d a n d W i n t o n geological sheets. In the c o u r s e of the 'Oil Shale M e t h o d o l o g y ' project's drilling only the lower part of the Allaru M u d s t o n e was p e n e t r a t e d ; m a x i m u m thickness of 76 m was obtained in B M R Urisino 1 (Ozimic, 1981^7).

128

S. O Z I M I C

S. P.

c r

LITHOLOGY

Lateral I8'8"

GAMMA-RAY API Units Radiation intensity increases

DENSITY

DESCRIPTION OF LITHOLOGY

increase SUPERFICIAL

DEPOSITS

Ferruginous soil and oeolian sand.

- l/\A/v\>v/>>\/

SILCRETE

AND

CLAYSTONE

Grey and yellow silcrete, containing pebbles of clear quartz. Claystone, white to yellow, massively

bedded

with iron rich bands, up to 15mm t h i c k . Vertical fractures are common. vxrv^^vxz Grey and yellow, massively bedded. Iron bands and vertical fractures as above.

CLAYSTONE AND SANDY

SILTSTONE

Yellow and light grey claystone and siltstone with laminae of fine sand. Iron bands and vertical fractures are common. Black, organic rich laminae occur throughout this interval.

CLAYSTONE Dark grey, massively bedded.

Laminae of

organic rich material present throughout this interval

Unfossiliferous and recrystallised with claystone inclusions SANDSTONE AND SILTSTONE Glauconitic carbonaceous and calcareous -122m SANDY

SILTSTONE

Massive, dark grey with laminae of organic rich material SANDSTONE Greenish, grey medium grained. Some beds of claystone.

CLAYSTONE Massive, grey and sideritic.

LT.D.I49m

Si/crete Sands and

Sandstone silts

Claystone Fig. 7.

Limestone Erosiona!

surface

Composite well log, BMR Urisino 1.

The unit is a massive blue-grey mudstone, which grades in places to grey siltstone. It contains fragments of cone-in-cone limestone and sandy, pyritic siltstone. Carbonaceous material, mainly plant remains, is present throughout; fine-grained pyrite and calcite laminae are c o m m o n . Burrowing, slump

and scour structures, and micro-cross-bedding are common. Bivalves are a b u n d a n t in places. The lower b o u n d a r y of this unit is taken as the contact of mudstone with the top of a bed of coquinite or recrystallised limestone defined to be part of the underlying Toolebuc

TOOLEBUC FORMATION PETROPHYSICS 129 Formation or one of its stratigraphic correlatives; in some Member, and Urisino beds are shown in Table 2. These areas the boundary is gradational. In many bore holes it is features have been compiled from numerous petroleum determined from the character of wireline logs, particularly exploration wells, stratigraphic holes and water bores that the gamma-ray log, which commonly shows a marked increase intersected the studied sequences. Laboratory-determined in radiation on entering the Toolebuc Formation. features are based on core analyses carried out by the BMR The marine fossils, together with burrows, lamination and as part of the Oil Shale Methodology project (Ozimic & Saxby, micro-cross-bedding suggest deposition below wave-base in 1983). a sea in which bottom currents operated with varying In the past the petrophysical evaluation of the Toolebuc velocities. The development of pyrite may indicate reducing Formation and its stratigraphic time-equivalents was generally conditions below the water/sediment interface. dependent on gamma-ray logs. This log was used extensively for correlation purposes, because of a marked gamma-ray anomaly thought to have been synonymous with the oil shaleSTRUCTURE bearing Toolebuc Formation. The available geological and geophysical data (Senior et In this study different wireline logs were evaluated and al., 1978; Smart et al., 1980) show that the known Early utilised in: (1) locating bed boundaries, (2) identifying Cretaceous sediments including the Toolebuc Formation were lithologies, (3) correlating lithofacies, and (4) assessing the deposited in a single depression that constituted the Eromanga relationship between specific lithologies and wireline log and Carpentaria Basins. Structure contours of the top responses. Toolebuc Formation and time-equivalent units (Fig. 4) serve to illustrate present structural style in the two basins. The main structural elements presented in Figure 4 are based on Gamma-ray response interpretations by Senior et al. (1978), and by Smart et al. In the past, the presence of the Toolebuc Formation in bores (1980). and wells has been suspected from a marked isolated peak, Overall, the Eromanga Basin succession is little-deformed, or set of peaks on gamma-ray logs (Senior et al., 1978; but drape growth-folds occur in a few places over basement Herbert, 1980). The anomaly was indeed found to correlate ridges and blocks, and rare monoclines in the upper part of with cored Toolebuc Formation during drilling throughout the sequence grade down into faults. The Carpentaria Basin the central and northern Eromanga Basin, and southern succession is not deformed except locally along its western Carpentaria Basin. Senior et al. (1978) attributed the anomaly and southern margins. Major structures affecting the to uranium-bearing phosphate minerals associated with shell depositional and post-depositional history of the Toolebuc debris; however, recent detailed studies at Julia Creek and Formation and time-equivalent units appear to be the Nebine Boulia (Ramsden, 1980; Ramsden et al1980) have confirmed Ridge, and 'Canaway Fault Zone' (Fig. 4). that uranium associated with both organic matter and The Nebine Ridge apparently rose spasmodically during phosphatic fish remains is responsible for gamma-ray the Cretaceous and influenced sedimentation intermittently anomalies. (Exon & Senior, 1976). This hypothesis is supported by pinchIn the southern part of the Eromanga Basin a similar out of the Toolebuc Formation against the western flank of anomaly occurs within the Wooldridge Limestone Member, the Nebine Ridge (Ozimic, 1981&), indicating that the and below, but not within the Urisino beds (Ozimic, 1982) transgressive Toolebuc sea lapped on but did not cross the (Fig. 8). Burger (1981) found the age of sediments giving the Nebine Ridge into the Surat Basin. The 'Canaway Fault Zone' gamma-ray anomaly below the Urisino beds to be slightly consists of a line of faults and fault-induced anticlines trending older than the Toolebuc Formation. It is therefore unwise to northeast along the Canaway Ridge, Canaway Fault, Stormhill assume that a gamma-ray log anomaly invariably indicates Fault and Beryl Anticline (Fig. 4). There is a vertical the presence of the Toolebuc Formation or its equivalents. displacement of approximately 50 m of the Toolebuc Examination of gamma-ray logs of a number of petroleum Formation across this zone which may be the boundary exploration wells, stratigraphic holes and water bores in South separating the Toolebuc Formation facies A and B, although Australia, South Wales, Northern Territory and there are no notable variations in thickness of the two facies Queensland New show also that: (Fig. 3). The distribution and minor thickness variations of (1) the gamma-ray log character along the eastern and the Toolebuc Formation and time-equivalents lead to the western margins of the northern Eromanga Basin is conclusion that the present-day Euroka Arch did not exist serrated, displays a moderate radiation response and at that time. changes to single or double-peaked, with strong Isopachs of the Wallumbilla Formation (Senior et al., 1978) radiation response in the centre of the northern suggest prior and possibly contemporary sinking of the Eromanga Basin and in the whole of the southern Eromanga and Carpentaria Basins to the south and north Carpentaria Basin (Fig. 8), of the Jurassic watershed now incorporated in the Euroka (2) the pattern of the gamma-ray log opposite kerogenous Arch. Furthermore, the Eromanga Basin sagged in broadly shale does not bear a simple relationship to oil shale the same area where the thickest Jurassic sequence and most grade, of the Late Palaeozoic Cooper and Adavale Basin sequences (3) the gamma-ray anomaly usually occurs opposite occur. The distribution and thickness variation of the Allaru Toolebuc Formation kerogenous shales which may or Mudstone (Senior et al., 1978) supports the concept of dual may not contain phosphatised fish remains, basin subsidence on both sides of the Euroka Arch. (4) the gamma-ray anomaly within the Wooldridge Limestone Member occurs opposite glauconitic, carbonaceous and silty sandstone beds, and PETROPHYSICS (5) the coquinites within the Toolebuc Formation and the Wooldridge Limestone Member have very low gammaMajor petrophysical features which characterise the main ray responses. lithologies of the Toolebuc Formation, Wooldridge Limestone

TABLE 2. Petrophysical features of Toolebuc Formation and time equivalents. Rock units

Toolebuc Formation

Lithologies G a m m a Ray

Kerogenous, calcareous shale (oil shale)

U )

Density ( pb)

Wireline logs Neutron Resistivity (|0N) (Rt)

Spontaneous potential (SP)

Interval transit time (At)

Laboratory core analysis Porosity Bulk Density (qualitative) (gm cm 3 )

Moderate to High

Low to Moderate

Low

Moderate to Slow

Low

1.6-2.2

Coquinite

Low

High

Moderate to High

Low to Moderate

Fast

Moderate

2.30-2.45

Wooldridge Limestone Member

Coquinite

Low

High

Low to Moderate

Fast

N/A

Low to Moderate

Moderate

Moderate to High

Moderate

N/A

Urisino beds

Unfossiliferous limestone

Low

High

Moderate to High

Low to Moderate

Fast

Low

2.35-2.45

Calcareous, carbonaceous, glauconitic silty sandstone

Low

Moderate

Low to Moderate

Moderate

2.25-2.40

Calcareous, glauconitic, silty sandstone

131

TOOLEBUC FORMATION PETROPHYSICS Rock Units

b-log

8 -log

« -log

fV=I

8 -log

ALLARU MUDSTONE AND ITS EQUIVALENTS TOOLEBUC FORMATION AND ITS EQUIVALENTS

COREENA MEMBER AND ITS EQUIVALENTS

LVJ

[AaJ

400 km

Coquinite/Limestone Kerogenous shale (oil shale)

Fades

boundary Shale Silty sands Glauconite-bearing

sands

Conglomerate v-ww Fig. 8.

Erosional

12/Q/40

surface

Stratigraphic distribution and character of gamma-ray logs, Eromanga Basin. (Localities: 1—BMR Urisino 1 well; 2—Delhi—FromeSantos Betoota 1 well; 3—Shell Dev. Aust. Boulia 3 well; 4—BMR Tambo 41 well; 5—Shell Dev. Aust. Julia Creek 1 well).

Density log response Response of both the long- and short-spaced density log to lithologies of the Toolebuc Formation and its equivalents appears to be less useful than that of the gamma-ray. The density log's response was most notable opposite layers that were lithologically homogeneous (Table 2), thicker than 0.2 m, and that were bounded by lithologically contrasting layers. For example, the Toolebuc Formation kerogenous shale could not be identified solely from density logs. On the other hand, coquinite layers thicker than 0.2 m showed a definite high density log response. The failure to identify kerogenous shales from density logs is interpreted to be due to masking effects caused by organic and inorganic shale components of differing bulk densities.

Neutron log response The neutron log curve shows a distinct positive deflection opposite most coquinite, limestone and sandstone beds (Table 2). This deflection is interpreted to indicate that these layers contain appreciable amounts of formation fluids and are therefore porous and probably permeable. Opposite kerogenous shales the neutron log response is usually nondescriptive and close to the shale-line.

Resistivity and Spontaneous Potential Resistivity readings over the Toolebuc Formation are variable and generally much lower than those of the Woolridge

Limestone Member and Urisino beds. Notably, increases in resistivity occur opposite most coquinite, limestone and sandstone beds complementing the pattern displayed by the density and neutron logs (Table 2). The patterns on the spontaneous potential log however, are less definitive and do not readily help identify specific lithologies (Table 2). In general, the log's usefulness lies mainly in determining the lower boundary of the Toolebuc Formation and its equivalents, with the underlying silty sequences.

Interval transit time The sonic log was employed mainly for identification of lithologies and location of bed boundaries. Overall the log was most useful in determining high density layers (e.g. coquinite, limestone) but less useful for determining kerogenous shale interbeds. Calcareous matter present in the kerogenous shales showed consistently moderate to slow interval transit time values (Table 2).

GEOLOGICAL HISTORY AND ENVIRONMENTAL SYNTHESIS The Toolebuc Formation, Wooldridge Limestone Member and Urisino beds apparently were deposited: (1) in a region 40° to 70° south of the Early Cretaceous palaeo-equator (Fig. 9a), (2) in a late Albian transgressive epicontinental sea, towards its southern boundary,

S. OZIMIC 0

SOUTH PACIFIC

300 km

OCEAN

50° S

60° S

CLARENCEMORETON '.BASIN 28°-

Conglomerate deposits on top of Coreena M e m b e r

Outflowing fresher (less dense) water Inflowing 50°S

saline (denser)

water

J Land

Palaeolatitude

•

SD, Julia

BMR, Urisino 1

Fades

j Predominant

clastic

Predominant

chemical

(b)

sedimentation sedimentation

SOUTH

NORTH

boundary

Creek 1

CARPENTARIA BASIN

EROMANGA BASIN

EUROKA ARCH (NARROWS)

Fig. 9. (a) Toolebuc Sea', palaeolatitudes and current patterns, (b) Hypothetical cross-section showing possible conditions during deposition of the Toolebuc Formation and time equivalents.

TOOLEBUC FORMATION PETROPHYSICS (3) on a northerly dipping slope, and (4) in a quiet, alternately anaerobic to aerobic and marine to brackish environment. The Toolebuc Formation facies are thought to have been deposited in the deeper, anoxic part of the Eromanga Basin, and the time-equivalent sequences mainly in oxygenated shallower areas (Fig. 9b).

GULF

133

Before the sediments of these units were deposited, and following a major middle Albian marine regression from the Eromanga and Carpentaria Basins, there was an onset of fresh-water conditions during which the regressive sequences of the upper part of the Coreena Sandstone Member of the Wallumbilla Formation were subaqueously and, or subaerially irregularly eroded and reworked (Day, 1969). Conglomeratic

CARPENTARIA CORAL

CARPENTARIA Thickness

contour

iASIN;

(m)

Anticline Fault

Townsville

Outcrop Facies

boundary

Present

basin

margin

Extent of the Toolebuc and its equivalents

Formation \

NT|QLD

N^ Bedourie ROMANGA BASIN

Charfeville

Oodnadatt;

0-50m 50—200m

j Weathered J

zone.

Possible

open-cut

Possible

in-situ

Non-productive mining

12/0/61

area Productive

200m

retorting

area

Fig 10 Thickness of sediments on top of the Toolebuc Formation and time equivalents. (Shaded areas indicate Toolebuc Formation oil shale production potential).

134

S. OZIMIC

deposits at the top of the Coreena Sandstone Member occur mainly along the eastern margin of the Eromanga Basin (Fig. 9a) and are interpreted to have been derived from the adjacent and rising Nebine Ridge. Absence of conglomeratic deposits at this stratigraphic level elsewhere in the Eromanga and southern Carpentaria Basins possibly indicates that the Nebine Ridge had only local structural influence. The sea re-entered the basin from the north in the early part of the late Albian (Burger, 1980), but did not cross the Nebine Ridge. The conditions envisaged are those of a 'positive water balance basin' (Demaison & Moore, 1980) in which saline water entered from the north and circulated anticlockwise, while fresh water from the hinterland flowed northwards out of the region (Figs 9a, b). The sea was most likely stratified, with a permanent halocline below a layer of fresh water. At the same time, the uranium which is the major source of the gamma-ray anomaly associated with the Toolebuc Formation and the Wooldridge Limestone Member, was probably transported into the depositional area in a soluble state by run-off waters from adjacent landmasses that were flowing into the Toolebuc Sea'. Such conditions'would have: (1) enhanced prolific productivity in the euphotic zone; (2) favoured permanent or intermittent oxygen depletion in the lower parts of the sea water; (3) limited the establishment of normal benthonic marine fauna; (4) favoured the preservation of organic matter; (5) enhanced reduction of uranium (from the soluble hexavalent state to the insoluble tetravalent precipitate), and absorption of the radioactive precipitate by organic matter in the prevailing anaerobic environment. Conditions which favoured kerogenous shale deposition were apparently terminated by an increase in saltwater inflow into the depositional basin as a result of rising sea level, and possibly as a result of variations in conditions limiting productivity in the euphotic zone. Deposition of the Toolebuc Formation in an oxygendeficient environment would appear to coincide with similar deposits in other ocean basins that existed within the same general mid-Cretaceous period (Schlanger & Jenkyns, 1976). The widespread global distribution of these blackish sediments in various palaeo-oceanographic and palaeobathymetric settings required that different processes were responsible for black shale formation at different places. However, if these black shale events were the consequence of a single overriding factor, then it was most likely an ultimate response of the increase of both terrestrial and marine biota to changes in atmospheric climate (Habib, 1982).

TOOLEBUC FORMATION OIL SHALE AND SHALE OIL POTENTIAL The Toolebuc Formation productive oil shale covers an area of 0.484 x 106 km 2 (Fig. 10) (excluding the weathered, nonproductive zone from surface to 50 m). The oil shale ranges in thickness from 6.5 to 7.4 m, has a specific gravity of 1.7, and yields on average 37 litres of oil per tonne. On this basis, the total potential shale oil resources of the Toolebuc Formation have been estimated at 245 x 109 m3 (Ozimic &

Saxby, 1983). Approximately 20 per cent could possibly be produced by open-cut mining at depths from 50 to 200 m (Fig. 10). The remainder could potentially be produced by in situ retorting at depths greater than 200 m. Grade of oil shale has a large bearing on mining, retorting, and waste disposal costs; in recent years, grades of 40 to 60 L/t have been proposed as lower economic limits for some deposits. Incorporation of a lower limit of 40 L/t would decrease the potential shale oil resource considerably. A lower limit of 60 L/t would rule out most of the resource and probably leave only small pockets at locations such as Julia Creek, Croydon, Dobbyn, and Boulia. These estimates of resources should not, therefore, be regarded as final. Reliable data are still sparse, and more drilling is needed, especially where the Toolebuc Formation is deeper than 200 m.

SUMMARY The lateral and vertical distribution, and lithological makeup of the three time equivalent units reflect: (1) interplay of basin geometry; (2) stratification and circulation of water; (3) source and rate of sediment supply; (4) alkalinity of sea water for precipitation of calcium carbonate; (5) suitability or otherwise for growth of phytoplankton and algal mats, and preservation of their dead remains; (6) degree of absorption of radioactive elements at the time of deposition of organic matter. Where kerogenous shales (oil shales) have been sampled in bore holes they are invariably associated with a marked gamma-ray anomaly. That the reverse is not true is demonstrated by the occurrence of a gamma-ray anomaly within the Wooldridge Limestone Member, and below, but not within the Urisino beds. Direct identification of Toolebuc Formation kerogenous shales from wireline logs, may therefore be unreliable, and it should not be assumed that a gammaray anomaly invariably indicates the presence of the Toolebuc Formation or its correlatives.

ACKNOWLEDGEMENTS This paper is published with the permission of the Director, Bureau of Mineral Resources, Geology and Geophysics, Canberra. Support was provided by the National Energy Research Development and Demonstration Programme, supervised by Mr H. F. Doutch (BMR) and Dr J. Saxby (CSIRO), Oil Shale Methodology Project leaders. Many thanks are due to numerous workers from: State Geological Surveys of Queensland, South Australia and New South Wales; University of Wollongong; Australian National University; Commonwealth Scientific and Industrial Research Organisation, and Bureau of Mineral Resources, Geology and Geophysics. Among those workers I owe a special debt of gratitude to D. Burger, J. G. Byrnes, A. C. Cook, R. W. Day, H. F. Doutch, D. Gibson, M. Glikson, M. Griffiths, P. J. Hawkins, A. C. Hutton, N. H. Ludbrook, A. McMinn, T. Ramsden, J. Saxby, V. Scheibnerova, T. Stephenson and G. E. Wilford. Assistance from L. Pain during the preparation of this paper is acknowledged.

REFERENCES ALMOND, C. S., 1983: Stratigraphic drilling report G.S.Q. Eromanga—1. Qld Gov. Min. J. 84, 514-23.

AMERADA, 1966: Well completion report, Hale River No. 1. Amerada Pet. Corp. Aust. Ltd (unpubl.).

AMERADA, 1965: Well completion report, McDills No. 1. Amerada Pet. Corp. Aust. Ltd (unpubl.).

AMERADA, 1967: Well completion report, Thunderbolt Amerada Pet. Corp. Aust. Ltd (unpubl.).

No. 1.

TOOLEBUC FORMATION PETROPHYSICS AMOCO, 1966: Well completion report, Towerhill No. 1. Am. Oil Co. (unpubl.). AQUITAINE, 1965a: Well completion report, Fermony No. 1. Aust. Aquitaine Pet. Pty Ltd (unpubl.). AQUITAINE, 19656: Well completion report, Mayneside No. 1. Aust. Aquitaine Pet. Pty Ltd (unpubl.). AQUITAINE, 1976: Well completion report, Tanbar No. 1. Aust. Aquitaine Pet. Pty Ltd (unpubl.). BALFE, P. E., 1978: Stratigraphic drilling report—G.S.Q. Manuka 1. Qld Gov. Min. J. 79, 258-70. BOWERING, O. J. W., 1970: Well completion report, Gurra No. 1. Pexa Oil N.L. (unpubl.). BRERETON LeGAY, 1971: Well completion report, Yanpurra No. 1. Flinders Pet. N.L. (unpubl.). BUBELA, B., 1980: Some aspects of interstitial water movements in simulated sedimentary systems. BMR J. Aust. Geol. Geophys. 5, 257-63. BURGER, D., 1980: Preliminary palynology notes on B.M.R. Urisino—1 N.S.W. Aust. Bur. Miner. Resour. Geol. Geophys. Professional Opinion Geology 80.016 (unpubl.). BURGER, D., 1981: Palynology of three NERDDC stratigraphic bores in the Eromanga Basin; Preliminary Report: Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1981/32 (unpubl.). BURT, D., 1976: Final report, Authorities to Prospect 1426M-1432M inclusive. Shell Development (Aust.) Pty Ltd (unpubl.). BYRNES, J. G., 1980: BMR Urisino No. 1 Bore and its bearing on prospects for oil shale in the Eromanga Basin, N.S.W N.S.W. Geol. Surv. Rep. 80/27 (unpubl.). CADART, M. L., 1968: Well completion report, Thunda No. 1. Alliance Oil Dev. N.L. (unpubl.). CADART, M. L., 1969: Well completion report, Budgerygar No. 1. Alliance Oil Dev. N.L. (unpubl.). C A M P B E L L , I. R . , H A R R I S O N , J. & H I G G I N B O T H A M , G . T., 1963CR:

Well completion report Naryilco No. 1. Delhi-Santos (unpubl.). C A M P B E L L , I. R . , H A R R I S O N , J. & J O H N S O N , J. L . ,

19636:

Well

completion report, Putamurdie No. 1. Delhi-Santos (unpubl.). CAMPE, G. C., 1969: Well completion report, Cumbroo No. 1. Alliance Oil Dev. N.L. (unpubl.). CASEY, D. J., 1969: Explanatory notes, Tangorin 1:250 000 geological map. Aust. Bur. Miner. Resour. Geol. Geophys. CASEY, D. J. , 1970: Final report on Kumbarie No. 1 well. Pexa Oil N.L. (unpubl.). CONORADA, 1960: Well completion report, Ooroonoo No. 1. Conorada Pet. Corp. (unpubl.). DAY, R. W., 1969: The Lower Cretaceous of the Great Artesian Basin; in Campbell, K. S. W. (ed.). Stratigraphy and palaeontology; essays in honour of Dorothy Hill. Aust. Nat. Univ. Press, C a n b e r r a , 140-73.

DELHI, 1966: Well completion report, Nappacoongee No. 1. Delhi Pet. Pty Ltd (unpubl.). DELHI, 1968: Well completion report, Tinga-Tingana No. 1. Delhi Pet. Pty Ltd (unpubl.). DELHI-FROME-SANTOS; 1961: Well completion report, Betoota No. 1. Aust. Bur. Miner. Resour. Geol. Geophys. Petrol. Search Sub. Acts Publ. 10. DELHI-FROME-SANTOS, 1965: Well completion report, Merrimelia No. 1. Delhi Pet. Pty Ltd (unpubl.). DELHI-SANTOS, 1966: Well completion report, Mount Howitt No. I. Delhi Pet. Pty Ltd (unpubl.). D E L H I - S A N T O S - V A M G A S , 1972: Well completion report, Toolachee East No. 1. Delhi Pet. Pty Ltd (unpubl.). DEMAISON, G. J. & MOORE, G. T., 1980: Anoxic environments a n d

oil source bed genesis. Am. Assoc. Pet. Geol. Bull. 64, 1179-209. DETTMANN,

PLAYFORD,

G.,

1969:

Palynology

the

Australian Cretaceous: a review; in Campbell, K. S. W. (ed.). Stratigraphy and palaeontology; essays in honour of Dorothy Hill. Aust. Nat. Univ. Press, Canberra, 174-210.

135

EVANS, P. R., 1966: The palynology of Amerada Newlands No. 1 well, Queensland. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1966/186 (unpubl.). EXON, N. F. & SENIOR, B. R., 1976: The Cretaceous of the Eromanga and Surat Basins. BMR J. Aust. Geol. Geophys. 1, 33-50. FREYTAG, I. B., 1966: Proposed rock units for marine Lower Cretaceous sediments in the Oodnadatta region of the Great Artesian Basin. S. Aust. Geol. Surv. Q. geol. Notes, 18, 3-7. GARRETT, G. W. & PATTERSON, W. A., 1968a: Well completion report, Barcoo No. 1. Aust. Sun Oil Co. Ltd (unpubl.). GARRETT, G. W. & PATTERSON, W. A., 19686: Well completion report, Fairlea No. 1. Aust. Sun Oil Co. Ltd (unpubl.). GERRARD, M. J., 1964: Well completion report, Boree No. 1. Aust. Sun Oil Co. Ltd (unpubl.). GIBSON, D., 1982: Shallow stratigraphic drilling, northwest Eromanga Basin. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1982/16 (unpubl.). GREEN,

C.,

HAMILTON,

D. &

KYRANIS,

N.,

1963:

Well

completion report. Phillips-Sunray stratigraphic drilling Boulia area, ATP 54P Qld. Black Mountain 1, Bean tree 1, Canary 1, Elizabeth Springs 1. Phillips Pet. Co. (unpubl.). GREER, W. 1. & HARRISON, I., 1963: Well completion report, Orientos No. 1. Delhi Pet. Pty Ltd (unpubl.). GRIFFITHS, M., 1980: Well completion report, Toodla No. 1. S. Aust. Dep. Mines Ener. Rep. 80-128 (unpubl.). GSQ, 1960-1965: Occurrence of petroleum and natural gas in Queensland. Qld. Geol. Surv. Publ. 229, with supplements 1-5. HABERMEHL, M. A. & MORRISSEY, I. A., 1983: Index of

geophysical

well logs acquired by BMR from waterwells in the Great Artesian Basin, 1960-1975. Aust. Bur. Miner. Resour. Geol. Geophys. Rep., 241 (unpubl.). HABIB, D., 1982: Sedimentary supply origin of Cretaceous black shales; in Schlanger, S. O., & Cita, M. B. (eds) Nature and origin of Cretaceous carbon-rich fades. Academic Press, London, 113-27. HANLON, P. E., 1966: Well completion report, Binerah Downs No. 1. Clarence River Basin Oil Explor. Co. N.L. (unpubl.). HARDY, J. W. J., 1970: Well completion report, Murteree No. 1. Pursuit Oil N.L. (unpubl.): HARRISON, I. & GREER, W. I., 1963: Well completion report, Dullingari No. 1. Delhi Pet. Pty Ltd (unpubl.). HARRISON, J., CAMPBELL, I. R . & HIGGINBOTHAM, G . T., 1963:

Well

completion report, Pandieburra No. 1. Delhi Pet. Pty Ltd (unpubl.). HARRISON, J. & HIGGINBOTHAM, G. T., 1964: Well

completion

report, Gidgealpa No. 1. Delhi Pet. Pty Ltd (unpubl.). HEMATITE, 1974: Well completion report, Weston No. 1. Hematite Pet. Pty Ltd (unpubl.). HERBERT, C., 1980: A brief review of oil shale in New South Wales. N.S.W. Geol. Surv. Rep. 1980/016, (unpubl.). HESS, A., 1957: Well completion report, Santos Oodnadatta No. 1. S. Aust. Dep. Mines Ener. open file, 260/SR11/5/3A. (unpubl.). HUTTON, A. C. & COOK, A. C., 1983: Organic petrography of oil shale samples from the Toolebuc Formation; in Ozimic, S. & Saxby, J. Oil shale methodology; an examination of the Toolebuc Formation and the laterally contiguous time equivalent units, Eromanga and Carpentaria Basins. Aust. Bur. Min. Resour. Geol. Geophys. Rec. 1983/80 (unpubl.). JACQUE, M 0 1966: Well completion report, Mokari No. 1. French Pet. Co. (Aust.) (unpubl.). JACQUE, M. & SWEENEY, P. J., 1966: Well completion report, Galway No. 1. French Pet. Co. (Aust.) Ltd (unpubl.). JACQUE, M., STRUBEL, P. & SWEENEY, P., 1970: Well

completion

report, Roseneath No. 1. Total Explor. Aust. Ltd (unpubl.). KYRANIS, N., 1966a: Well completion report, Bonnie No. 1. Phillips Pet. Co. (Aust.) Ltd & Sunray DX Oil Co. (unpubl.). KYRANIS, N., 19666: Well completion report, Carlow No. 1. Phillips Aust. Oil Co. & Sunray DX Oil Co. (unpubl.).

S. O Z I M I C

136

Well completion report, Log Creek No. 1. Phillips Pet. Co. (unpubl.). L A I N G , A . C. M . , 1 9 6 6 : Well completion report, Yongala No. 1. Alliance Oil Dev. N.L. (unpubl.). LAING, A. C. M., 1967: Well completion report, Yongala No. 2. Alliance Oil Dev. N.L. (unpubl.). LAING, A. C. M. & CADART, M. C., 1970: Well completion report, Chandos South No. 1. Alliance Oil Dev. N.L. (unpubl.). LAING, S., 1982: Authorities to prospect 2298M & 3000M. Pan Pacific Co., (unpubl.). L E W I S , J. H . , 1961: Well completion report, Cothalow No. 1, Queensland. Phillips Pet. Co. (unpubl.). L E W I S , J. H . & K Y R A N I S , N . , 1 9 6 2 : Well completion report, Etonvale No. 1. Phillips Pet. Co. (unpubl.). L E W I S , J. H . & K Y R A N I S , N . , 1 9 6 5 : Well completion report, Gilmore No. 1. Phillips Pet. Co. & Sunray DX Oil Co. (unpubl.). L U D B R O O K , N . H . , 1 9 6 6 : Cretaceous biostratigraphy of the Great Artesian Basin in South Australia. S. Aust. Geol. Surv. Bull. 40. LUNT, C. K., 1972: Well completion report, Brumby No. 1. Delhi Pet. Pty Ltd (unpubl.). M A G N I E R , P., 1 9 6 4 : Well completion report, Witcherrie No. 1, South Australia. S. Aust. Dep. Mines Ener. Env. 349 (unpubl.). KYRANIS, N . & M C D O N A G H , G., 1966:

MAGNIER, P., 19646: Well completion report, Purni No. 1, South Australia. S. Aust. Dep. Mines Ener. Env. 352 (unpubl.). MAGNIER, P. & SWEENEY, P., 1965: Well completion

report,

The

Brothers No. 1. French Pet. Co. (Aust.) Pty Ltd (unpubl.). P., 1 9 6 7 : Well completion report, Glenaras No. 1. Phillips Pet. Co. (Aust.) Ltd & Sunray DX Oil Co. (unpubl.).

MCDONAGH, G.

A. 1 9 6 6 : Well completion report, Leopardwood No. 1. Phillips Pet. Co. (Aust.) Ltd (unpubl.).

M C D O N A G H , G . P., K N U T H , B. W . & P E T E R S O N , W .

A., 1980: Preliminary palynology of BMR (Jrisino 1, Great Australian Basin. N.S.W. Geol. Surv. Reps. GS1980/115 & GS1980/270, (unpubl.).

MCMINN,

McMlNN, A. & B U R G E R , D., 1986: Palynology of the Toolebuc Formation (sensu lato) in the Eromanga Basin; in This volume. N . A . & S P E N C E R , L., 1 9 7 0 : Well completion report, Muttaburra No. 1. Pursuit Oil N.L. (unpubl.).

MEYERS,

MINAD, 1963: Well completion report, Penrith No. I. Mines Admin. Pty Ltd (unpubl.). MINAD, 1964: Well completion report, Beryl No. 1. Mines Admin. Pty Ltd (unpubl.). W . D . 1955a: Well completion report, Longreach No. 1. Longreach Oil Ltd (unpubl.).

MOTT,

D . , 1 9 5 5 6 : Well completion report, Longreach No. 4. Longreach Oil Ltd (unpubl.).

MOTT, W.

M O T T , W . D. & ASSOCIATES,

1964a: Well completion report, Saltern

Creek No. 1 (unpubl.). M O T T , W . D. & ASSOCIATES, 1964b:

Well completion report, Hulton

No. 1 (unpubl.). MOTT,

ASSOCIATES,

1964C:

Well completion

report,

Marchmont No. 1 (unpubl.). 1970a: Well completion report Gilpeppee No. 1. Alliance Oil Dev. N.L. (unpubl.).

M U L R E A D Y , J. N . ,

Well completion report, Packsaddle No. 1. Alliance Oil Pet. Aust. N.L. (unpubl.).

M U L R E A D Y , J. N . , 1 9 7 0 6 :

OZIMIC, S., 1982: Depositional environment of the oil shale-bearing Cretaceous Toolebuc Formation and its equivalents, Eromanga Basin, Australia. Proc. 15th Oil Shale Symposium. Colorado School of Mines, Colorado, U.S.A. O Z I M I C , S . & P A I N , L., 1 9 8 3 : Stratigraphic drilling in the Early Cretaceous Rolling Downs Group, Simpson Desert, N.T. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1 9 8 3 / 8 0 , (unpubl.). O Z I M I C , S . & S A X B Y , J . , 1 9 8 3 : Oil shale methodology; an examination of the Toolebuc Formation and the laterally contiguous time equivalent units, Eromanga and Carpentaria Basins. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1 9 8 3 / 8 1 , (unpubl.). Well completion report; Bury No. 1. Phillips Aust. Oil Co. & Sunray DX Oil Co. (unpubl.). PEMBERTON, R. L., 1963: Well completion report, Brookwood No. 1. Exoil N.L. (unpubl.). PEMBERTON, R. L., 1970: Well completion report, Coongie No. 1. Flinders Pet. N.L. (unpubl.).

PATTERSON, W . A . 1966:

PEMBERTON, R. L. & LeGAY BRERETON, J., 1971: Well

completion

report, Tallalia No. 1, Queensland. Flinders Pet. N.L. (unpubl.). N.L., 1970: Final report on Cherri No. 1 well, South Australia. French Pet. Co. (Aust.). (unpubl.). PHILLIPS-SUNRAY, 1963: Well completion report, Gumbardo (unpubl.). RAMSDEN, A. R., 1980: Geochemistry, mineralogy and petrology of the type section of the Toolebuc Formation in BMR Boulia 3A stratigraphic drill hole. Aust. CSIRO. Inst. Earth Res. Restrict. Invest. Rep. 1197R, (unpubl.).

PEXA OIL

R A M S D E N , A . R . , D I C K S O N , B. L . , M E A K I N S , R . & B E N N E T , A . J. R.,

1980: Origin and significance of the gamma-ray anomaly, BMR Urisino-1 stratigraphic drill hole. Aust. CSIRO Inst. Earth Res. Restrict. Invest. Rep. 1204R, (unpubl.). RYAN, J. C., 1961: Well completion report on Innamincka No. 1 well, South Australia, of Delhi-Frome-Santos. Aust. Bur. Miner. Resour. Geol. Geophys. Petrol. Search Subs. Acts Publ. 9. S C H E I B N E R O V A , V., 1978: Comparative foraminiferal biostratigraphy of the Santos Oodnadatta No. 1 section, South Australia. N.S.W. Geol. Surv. Miner. Resour. Rec. 19(1) (unpubl.). Micropalaeontology of BMR Urisino-1, northwestern New South Wales. N.S.W. Geol. Surv. Rep. G S 1 9 8 0 / 4 1 2 (unpubl.).

S C H E I B N E R O V A , V., 1981:

C., 1 9 7 6 : Cretaceous oceanic anoxic events—cause and consequences—Geologie en Mijnbouw, 55, 1 7 9 - 8 4 .

S C H L A N G E R , S. O . , & J E N K Y N S , H .

SENIOR, B. R. & HUGHES, R. J., 1972: Shallow stratigraphic

drilling

of radioactive anomalies at Eyre Creek, Springvale Sheet area, western Queensland 1972. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1 9 7 2 / 1 9 (unpubl.). R., E X O N , N. F. & B U R G E R , D., 1975: The Cadno-owie and Toolebuc Formations in the Eromanga Basin, Queensland. Qld Gov. Min. J. 76, 445-55.

S E N I O R , B.

B. R . , M O N D , A . & H A R R I S O N , P. L . 1 9 7 8 : Geology of the Eromanga Basin. Aust. Bur. Miner. Resour. Geol. Geophys. Bull. 167.

SENIOR,

SMART, J., G R I M E S , K . G . , D O U T C H , H . F . & P I N C H I N , J., 1980: T h e

Carpentaria and Karumba Basins, North Queensland. Aust. Bur. Miner. Resour. Geol. Geophys. Bull. 202. A. E., 1982: Stratigraphic drilling in the Cretaceous Toolebuc Formation in the western Eromanga Basin, 1981. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1982/25 (unpubl.).

STEPHENSON,

NETZEL., R. K., 1967: Well completion report, Stafford No. 1. Phillips Aust. Oil Co. & Sunray DX Oil Co. (unpubl.).

SYKES, I. G., 1981: Well completion report, Barcoo Junction No. 1. Queensland. X.L. Pet. ( N . S . W . ) Pty Ltd. (unpubl.).

1981a: Stratigraphic drilling in the Cretaceous Toolebuc Formation in the southern eastern Eromanga Basin, 1980. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1981/12 (unpubl.).

V I N E , R . R . , D A Y , R . W . , C A S E Y , D . J . , M I L L I G A N , E . N . , GALLOWAY,

O Z I M I C , S.,

O Z I M I C , S., 1 9 8 1 6 : Stratigraphic drilling in the Cretaceous

Toolebuc Formation in the Charleville District, southern Queensland. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1 9 8 1 / 3 8 (unpubl.).

M . C . & E X O N , N . F., 1967: Revision of the nomenclature of the Rolling Downs Group in the Eromanga and Surat Basins. Qld Gov. Min. J. 68.

WARRIS, B. J., 1969: Well completion report, Rand No. 1. Longreach Oil Ltd (unpubl.).

TOOLEBUC FORMATION PETROPHYSICS 1970a: Well completion Longreach Oil Ltd (unpubl.).

report, Stormhill No. 1.

WILTSHIRE, M . J., 1 9 7 3 6 :

B. H . , 1 9 7 0 6 : Well completion Longreach Oil Ltd (unpubl.).

report, Belmore No. 1.

WILTSHIRE, M . J. & IMBERT, M . ,

WARRIS,

B. J.,

WARRIS,

1973a: Well completion report, Lovelle Downs No. 1. Hematite Pet. Pty Ltd (unpubl.).

WATSON, P. J.,

Well completion report, Clyde No. 1. Hematite Pet. Pty Ltd (unpubl.).

W A T S O N , P. J . , 1 9 7 3 6 :

J . , 1 9 8 2 : Well completion report, Thomas No. 1. Argonaut Int. Corp. (unpubl.).

WILTSHIRE, M .

1973a: Well completion report, Durham Downs No. 1. Delhi Pet. Pty Ltd (unpubl.).

WILTSHIRE, M . J.,

137

Well completion report, Ban Ban No. I. Henry B. Kelsey Co. (unpubl.). 1977a: Well completion report, Barrolka No. 1. Aust. Aquitaine Pet. Pty Ltd (unpubl.).

M . J . & I M B E R T , 1 9 7 7 6 : Well completion report, Karmona No. 1. Aust. Aquitaine Pet. Pty Ltd (unpubl.).

WILTSHIRE,

WILTSHIRE, M. J., 1978: Well completion report, Colson No. 1. North Broken Hill Ltd & Beach Pet. N.L. (unpubl.). W O P F N E R , H . , F R E Y T A G , I. B. & H E A T H , G. R . ,

1970: Basal JurassicCretaceous rocks of western Great Artesian Basin, South Australia: stratigraphy and environment. Am. Assoc. Pet. Geol. Bull. 54, 383-416.

Geological Society of Australia Special Publication No. 12, 139-154

Palynology and palaeoenvironments of the Toolebuc Formation (sensu lato) in the Eromanga Basin A. McMinn1 & D. Burger2 1 1

Geological Survey of New South Wales, Geological and Mining Museum, 36 George Street, Sydney, NSW 2000. Division of Continental Geology, Bureau of Mineral Resources, Constitution Avenue/A nzac Parade, Canberra, ACT 2601.

ABSTRACT The middle and upper Albian rock sequence in the Eromanga Basin has been examined palynologically in connection with the presence of oil shales in the Toolebuc Formation. The examination includes biostratigraphy, palaeoenvironments, and thermal maturation. The Toolebuc Formation and its correlative strata, the Urisino beds in New South Wales and the Wooldridge Limestone Member correlative interval in South Australia, here referred to collectively as 'Toolebuc Formation (sensu lato)\ are shown to form a single time rock unit of early late Albian age, which is associated with the Phimopollenites pannosus spore-pollen zone and the Endoceratium ludbrookiae dinoflagellate zone. The record of dinoflagellates suggests four successive marine episodes within the studied interval. Two of these appear to be major basin-wide events which we think correspond in time with eustatic phases of high sea level, the Toolebuc Formation being deposited during the last phase. Two other episodes seem to be of local impact only and may not be of eustatic origin. Spore coloration indicates the Toolebuc Formation to be thermally immature, and the free oil in the formation is thought to be of secondary origin.

INTRODUCTION Palynological study of the Lower Cretaceous (Albian) Toolebuc Formation and its correlative sedimentary units in the Eromanga Basin, the Wooldridge Limestone Member equivalent, and Urisino beds is a contribution to the National Energy Research, Development, and Demonstration Council (NERDDC) Oil Shale Methodology Project, a joint undertaking by the Bureau of Mineral Resources and the Commonwealth Scientific and Industrial Research Organisation (CSIRO) Fossil Fuels Division, who have selected the Toolebuc Formation for a pilot study. In the present paper the term Toolebuc Formation (sensu lato)' is used to refer collectively to the three rock units mentioned, in anticipation of a proposal (Ozimic, in prep.) to formally extend the term Toolebuc Formation' to include all three units. Relevant aspects of the geology, petrochemistry, and palaeontology of the formation have been compiled by Ozimic (1982,1986). To obtain study material, the Toolebuc Formation and portions of adjacent strata were drilled and continuously cored in a series of shallow stratigraphic holes along the eastern, northwestern, and southern parts of the basin. The stratigraphy of these boreholes has been described by Burger (19746) and Ozimic (1986). Preliminary palynological accounts of some NERDDC Project boreholes have been written earlier by McMinn (1980a, 19806) and Burger (1981), and more detailed discussions on these are included here. We examined 8 boreholes from as wide an area as possible within the time available (Fig. 1). All except BMR Boulia 3 were drilled with NERDDC Project funds. They are: BMR Augathella 6, BMR Boulia 3, BMR Charleville 3-3A, BMR Charleville 5, BMR Hay River 12, BMR Jericho 11, BMR Toompine 1 and BMR Urisino 1.

All but 2 boreholes are sited in Queensland. BMR Urisino 1 yielded information from the Bulloo Embayment, which is augmented by data from other boreholes drilled to test the presence of oil shale in northwestern New South Wales (Byrnes, 1980; McMinn, 1983). One of these, WRC Birrigoolpa (30961) to the southwest of BMR Urisino 1, is included in this paper. Only a single borehole from Northern Territory (BMR Hay River 12) was examined, and no boreholes were drilled for the NERDDC Project in South Australia. Fortunately, detailed palynological and micropalaeontological data are already available from Santos Oodnadatta 1, which intersected Lower Cretaceous sediments near the western margin of the basin.

STUDY METHODS Our contribution to the NERDDC Project is an attempt to formulate a uniform palynological approach to oil shales in general, and the subjects which we regard as essential to this aim are biostratigraphy, palaeoenvironments, and maturation. Biostratigraphy The aim of biostratigraphic study is to establish the position of the oil shale within the appropriate fossil zonal scheme to determine its age. Boundaries of palynological zones referred to in this paper are defined by the earliest stratigraphic appearance of zonal index species. Arguments for accepting limits of zonal intervals of spores, pollen and dinoflagellates as time-parallel surfaces in the rock sequence have been advanced by Dettmann & Playford (1969) and Burger (1980, 1986). We processed core and cuttings samples from all boreholes to retrieve spores, pollen grains, and phytoplankton

A. McMINN & D. BURGER

140

SYDNEY

DSDP258 Margin of Great Artesian Basin

VVWW/

Northern and eastern limits of Eromanga Basin Limits of occurrence and fades boundaries of Toolebuc Formation fs.l.J

HOBART

Fig. 1. Geographic domain and facies provinces of the Toolebuc Formation (sensu lato) in the Great Artesian Basin, and locations of petroleum wells, water bores, and NERDDC Project boreholes mentioned in text.

(dinoflagellates and acritarchs). Spores and pollen grains are not related biologically to phytoplankton, and so the zonal systems of the two groups of fossils provide an independent check to the stratigraphic conclusions drawn.

Palaeoenvironments

Palaeoenvironmental study helps in understanding a whole range of factors which leads to the formation of oil shales, in that it enables reconstruction of a three-dimensional framework of the palaeoenvironment of a sedimentary sequence. Palynology deals with spores and pollen, as well as phytoplankton, and these groups exhibit different origins and dispersal patterns. Spores and pollen grains released seasonally by land plants in great quantities are ultimately buried in sediments both onshore and offshore. During the Cretaceous, dinoflagellates were apparently restricted to offshore (saline) environments, and their cysts are recovered

mostly from marine sediments. Morgan (1975) reported nonmarine dinoflagellates from the basal Cretaceous Cadnaowie Formation in New South Wales, but recent (unpublished) finds in this formation in Queensland suggest that the fossils represent a marine influx. Acritarchs represent a group of aquatic organisms which during the Cretaceous very probably adapted to a wide range of environments (Burger, 1980). It is clear that statistical palynology, in which relative abundances of these organisms in an assemblage are calculated may provide a means of measuring the local conditions at the time of deposition of the sampled rock interval. Statistical analysis of this kind can also play a useful role in regional correlations of rock sequences in the subsurface. Burger (1974a, 1980) interpreted the presence or absence of dinoflagellates in palynological assemblages to indicate successive marine transgressions and regressions in the Early Cretaceous of the Surat Basin. Exon & Burger (1981)

TOOLEBUC FORMATION PALYNOLOGY

141

confirmed the reality of these episodes by correlating them with eustatic sea-level movements (see below). Evidence that sea-level movements also caused periodic changes of the environment in the Eromanga Basin has been offered by Morgan (19806) and Burger (1982, 1985). We present more detailed statistics of dinoflagellate abundance from various boreholes as an indication of changes in the environment which occurred simultaneously in different parts of the Eromanga Basin.

in BMR Boulia 3A stratigraphic drillhole, which was described by Burger (19746). The new data from the recent NERDDC drilling program led Ozimic (1986) to consider the formation as constituting two laterally intermerging rock units, the Toolebuc Formation and Urisino beds, extending across the Eromanga and southern Carpentaria Basins and representing different facies provinces (Fig. 1). In the southwestern part of the basin Ozimic distinguished a more calcareous 'Wooldridge Limestone' facies province, derived from the Wooldridge Limestone Member (Oodnadatta Formation), Maturation which is known only in outcrop at the type section in For a long time attention has been paid to various aspects Wooldridge or Fossil Creek, about 45 km northwest of of chemical degradation of particulate organic matter with Oodnadatta township. increasing temperature, to assess the petroleum source The Toolebuc Formation consists of calcareous and potential of sedimentary rocks. Several methods have been kerogenous oil shale, mudstone, and minor coquinite. It developed to correlate the degree of thermal organic alteration extends as such north of a line approximately through with the stage of hydrocarbon generation. Evans & Staplin Charleville and Bedourie townships, and encroaches onto the (1971) subdivided progressive degrees of organic southern Carpentaria Basin. In the east the formation rests metamorphism with increasing temperature into immature, on the Coreena Member of the Wallumbilla Formation, which mature and metamorphic thermal facies, the mature facies includes glauconitic sandstone and siltstone. In many being that in which hydrocarbon generation is most likely to boreholes the Coreena Member-Toolebuc Formation contact occur. Palynologists have used the colour of fossil spores and is a scoured surface, marked by one or more pebble horizons. pollen grains (reflecting the degree of carbonisation of the In the northern and northwestern areas the formation rests wall) as a rapid means of interpreting degrees of organic on mudstone (Ranmoor Member, Wallumbilla Formation maturation. Staplin (1977) described the colour of 'exinite' undifferentiated). The formation is overlain by the argillaceous (including spore and pollen walls) respectively as: 'faint yellow Allaru Mudstone (Senior et al., 1978; Smart et al., 1980). to yellow' (immature), 'yellow to brown' (mature), and 'dark South of a line drawn from Charleville in the east to brown to black' (metamorphic). We have made a rapid check Bedourie in the west, the Urisino beds extend across the central of various organic components along similar lines to test the and southern areas of the basin. They consist of nonfeasibility of this type of examination against the results of bituminous siltstone mudstone and glauconitic sandstone, the petrochemical analyses of the Toolebuc Formation oil with minor limestone,and and their reference section is the interval shales. between 102.0 m and 122.0 m depth in BMR Urisino 1 (Ozimic, 1986). TOOLEBUC FORMATION: GEOLOGICAL Freytag (1966) described the Wooldridge Limestone Member SETTING in Wooldridge Creek as a fossiliferous siltstone, partly sandy The Toolebuc Formation in Queensland was discussed by and calcareous, with numerous limestone concretions. Both Senior et al. (1975), who proposed the type section of the he and Wopfner et al. (1970) traced the unit into Santos formation as the interval between 25.3 m and 35.8 m depth Oodnadatta 1 (Fig. 3), but several workers dislike carrying SOUTH AUSTRALIA

QUEENSLAND

THOMPSON, 1980 WOPFNER MOORE AND PITT, SENIOR ET AL, 1975 ET Al, 1970 1982 FREYTAG. 1966

z z Ui < US

HAIG.1979

LUDBR00K, 1966

MORGAN, 1980;

THIS PAPER

w w w v w v ^/V\AAAAAAAAA \AAAA/WWW\AAn WINTON FORMATION Mt Alexander Sst Member

WINTON FORMATION

WINTON FORMATION OODNADATTA FORMATION

Woold - ToolebucX Fm / ridge Lst M Coorikiana Member BULLDOG SHALE

Coorikiana Sandstone BULLDOG SHALE

No record

Nowcord

b-c

MACKUNDA FORMATION Hedbergella de/rioensis

ALLARU MUDSTONE TOOLEBUC FORMATION

< o _l CO

Coreena/ Ranmoor Member

Hedbergella infracretacea

Hedbergella planispira No record

Neobuhmma australiana

I * Qj fS

V. howchini — T. flosculus

No record

Fig 2 Correlation o f A l b i a n rock sequences, Great Artesian Basin, with foraminiferal and palynological biostratigraphic units.

Cryb. striatus

142

A. McMINN & D. BURGER

SANTOS OODNADATTA 1 DETTMANN AND

PLAYFORB, 1969 | PLAYFORDl w&iiffl

morgan,

t m

PERCENTAGE OF DINOFLAGELLATES

30.5

? EPISODE 4

Phimopollenites pannosus

Endoceratium ludbrookiae a

(EPISODE 3B)

EPISODE 3A

124.1 128.0

152.4

164.6 m

EPISODE 1

209.7

60%

• H i

Total depth 4 0 2 . 9 m Stratigraphy see A Sprigg, 1 958 B Ludbrook, 1966 C Freytag, 1966 D Griffiths, 1980

Pebble-breccia horizon

Sandstone ~_z~_q

SUtstone Mudstone

I I I

Limestone

Coal

Fig. 3. Stratigraphy of Santos O o d n a d a t t a l (S.A.) according to different authors. Local stratigraphic distribution of palynological and foraminiferal zonal intervals. Samples analysed by A. M c M i n n .

TOOLEBUC FORMATION PALYNOLOGY the concept of the Wooldridge Limestone Member into the subsurface stratigraphy. Thompson (1980), in his generalised stratigraphic nomenclature for the Eromanga Basin in South Australia, avoided the term Wooldridge Limestone. Moore & Pitt (1982) used the term Toolebuc Formation for this unit, and traced it in many wells from southwestern Queensland into northeastern South Australia, above the southern Cooper Basin as far south as Moomba, where it occurs stratigraphically above the Coorikiana Sandstone, and to the west as far as Oodnadatta. In a few South Australian petroleum wells, including Santos Oodnadatta 1 (Fig. 3), a silty interval has been logged between the Wooldridge Limestone Member correlative and the Coorikiana Sandstone (Fig. 2). Such an interval is observed also separating the Urisino beds and Coorikiana Sandstone in BMR Urisino 1 (Fig. 6). Problems concerning rock nomenclature arising from this and other features have been discussed by Ozimic (1986).

TOOLEBUC FORMATION: PALAEONTOLOGICAL EVIDENCE

A brief summary is given here of the evidence for the age of the Toolebuc Formation (sensu lato), based on study of various groups of fossils from the Early Cretaceous of Queensland, South Australia, and New South Wales.

Invertebrate macrofaunas

Day (1969) reported shelly faunas from the Toolebuc Formation in the Richmond area, including the ammonite

143

The Geological Survey of New South Wales drilled 4 shallow stratigraphic boreholes intersecting marine Lower Cretaceous strata near the southern margin of the Eromanga Basin in which Morgan (1978) recognised the Doncaster and Coreena Members of the Wallumbilla Formation. An initial sedimentological, palynological, and micropalaeontological appraisal of these boreholes has been offered by Byrnes et al. (1975). In that paper, Scheibnerova located the lower limit

of her benthonic Lingulogavelinella frankei Zone (L. frankei

Datum horizon) within the basal Coreena Member in DM Wanaaring 1, near Wanaaring township. She thought the horizon to be late Albian, and rejected as too young a K-Ar age of 96.6 ± 2.5 Ma, which Byrnes obtained from glauconitic sands in the same interval of that borehole. Lingulogavelinella frankei has been recovered from Albian sediments of the Great Artesian Basin in the Richmond and Augathella areas, and also in South Australia (Scheibnerova & Byrnes, 1977; Scheibnerova, 1980). The stratigraphic position of the L. frankei Datum horizon as determined in some bores in central Queensland led to anomalous time relationships between the Albian foraminiferal and palynological biostratigraphies, and this apparent anomaly has been investigated (Scheibnerova, 1983, this volume). Ludbrook (1966) and Scheibnerova (1980) examined the foraminifera from Santos Oodnadatta 1, and their findings are summarised in the borehole data below.

Plant microfossils

The record of plant microfossils which we have examined

genera Labeceras, Myloceras, and Appurdiceras. The fauna in various boreholes is briefly summarised, and includes the

indicated a restricted environment and was dated Albian. Ammonites from the underlying Ranmoor Member of the Wallumbilla Formation, including Anoneceras, Beudanticeras, and Brewericeras, were thought to be late early Albian. Faunal assemblages from the Toolebuc Formation near Tambo and Augathella closely resembled those from Richmond. Faunas from the overlying Allaru Mudstone contained species of the ammonite genera Prohysteroceras, Labeceras, and Myloceras, which Day regarded as early late Albian. Reyment (1964) described an ammonite fauna from Wooldridge Creek, South Australia, which contained the

genera Labeceras, Myloceras, and Falciferella. He correlated

this assemblage with the varicosum-equatoriale Zone of the standard English Gault succession. Ludbrook (1966) listed the additional nautiloid and belemnite genera Eutrephoceras and Dimitobelus from the same locality, and concurred with a late Albian age for this fauna.

Foraminifera

Haig (1979) described Aptian and Albian foraminifera, including many agglutinated forms with siliceous walls, as the Ammobaculites benthonic association, and regarded it as characteristic of shallow land-locked seas. Few of the species of this association are well enough documented to be significant time indicators, and Haig's work on planktonic foraminifera is of more consequence to the present study, although this group of fossils contains far fewer species. Haig (1979) distinguished three assemblage zones which together form the Hedbergella planktonic association. His middle or Hedbergella infracretacea Zone includes abundant H.

infracretacea, common H. delrioensis and H. planispira, and

very rare H. punctata, which suggests a late Albian age for the associated Toolebuc Formation and basal Allaru Mudstone (Fig. 2).

concurrent sequences of dinoflagellates and spores and pollen grains. Morgan (1978, 1980ff) studied dinoflagellate sequences of the Rolling Downs Group in 11 boreholes in the Eromanga and southern Carpentaria Basins. He recognised his Pseudoceratium turneri Zone from the Coreena Member of the Wallumbilla Formation in Queensland and New South Wales, and in the Bulldog Shale and Coorikiana Sandstone in Santos Oodnadatta 1, South Australia. He recorded his Endoceratium ludbrookiae Zone, which succeeds the P. turneri Zone, from the Oodnadatta Formation, including the Wooldridge Limestone Member correlative in Oodnadatta 1, and the upper part of the Coreena Member, Toolebuc Formation, and Allaru Mudstone in Queensland (Fig. 2). By virtue of their planktonic habit, dinoflagellates may be used in certain circumstances for intercontinental biostratigraphic correlation. Morgan (1980tf) compared stratigraphic ranges of a number of species in Queensland with those in Aptian and Albian sequences of the Tethys Province, including type sections in western Europe and in Canada, which had been studied, among others, by Clarke & Verdier (1967), Millioud (1969), and Verdier (1975). On the basis of these comparisons Morgan concluded that his Pseudoceratium turneri 'b' and 'c' Subzones were early and middle Albian respectively, and his Endoceratium ludbrookiae Zone middle to late Albian. More direct evidence for the age of the last zone comes from DSDP Leg 26 site 258 on the Naturaliste Plateau, off southwestern Australia (Fig. 1). Morgan (1978) identified the Endoceratium ludbrookiae Zone between core 15 and core 21. Thierstein (1974) studied the nannoplankton from that section, and dated cores 20 and 21 as middle Albian on the presence of his Prediscosphaera cretacea Zone, and cores 15 to 19 as late Albian on the presence of his Eiffelithus turriseiffeli Zone.

144

A. M c M I N N & D. B U R G E R

The Early Cretaceous sequence of spores and pollen grains has a much more regional character. The zonal intervals used in this paper have been instituted by D e t t m a n n & Playford (1969), and they have been dated indirectly f r o m their associations in time with the Cretaceous faunal sequences reviewed above (Fig. 2). T h e Crybelosporites striatus Zone, the reference section of which is the interval between 213.4 m and 195.7 m depth in the Bulldog Shale of Santos O o d n a d a t t a 1 well, has been identified in the Coreena and R a n m o o r Members of the Wallumbilla Formation in Queensland and New South Wales. T h e zone is of early Albian age (Evans & Hawkins, 1967; Burger, 1968, 1973; D e t t m a n n & Douglas, 1976). T h e succeeding Coptospora paradoxa Zone has its reference section between 136.2 m and 181.7 m depth in the Bulldog Shale of O o d n a d a t t a 1 well. T h e zone has been identified in the Coreena and R a n m o o r Members in Queensland and New South Wales, and has been dated as approximately middle Albian (Burger, 1968, 1980, 1982; Dettmann & Douglas, 1976; Morgan, 1978). The reference section of the Phimopollenites pannosus Zone is within the Oodnadatta Formation (including the Wooldridge Limestone Member correlative) between 124.1 m and 26.5 m in Santos O o d n a d a t t a 1 (Playford et al., 1975). Equivalent spore and pollen assemblages occur in the Allaru M u d s t o n e to W i n t o n F o r m a t i o n interval in Queensland, and the zone has been dated as late Albian (Dettmann & Playford, 1969; Burger, 1973, 1986). The position of the Toolebuc Formation (sensu lato) in this zonal scheme is discussed below. Stratigraphic positions of palynological zonal intervals in various boreholes are shown in Figures 3—6. Geological ranges of some i m p o r t a n t zoneindicative spores, pollen grains, and dinoflagellates are indicated in Figure 7.

BOREHOLE DATA This section presents brief descriptions of the palynological sequences in 10 boreholes specified above, and which are illustrated in Figures 3-6.

BMR Augathella 6 This borehole, approximately 36 km northwest of Augathella (lat. 25°36'13"S, long. 146 °05 '35 "E), was fully logged by BMR. T h e interval f r o m 30 m to total depth, including the lower part of the Allaru Mudstone, Toolebuc Formation, and Coreena M e m b e r of the Wallumbilla Formation was continuously cored. Fifty-two samples were taken from 48.40 m to total depth for palynology, and 15 selected samples are shown in Figure 4. Recovery of spores and pollen f r o m the Coreena Member

was satisfactory. The Coptospora paradoxa Zone was identified between 69.10 m and 84.40 m. An assemblage from 87.40 m, which lacked C. paradoxa, might represent the preceding Crybelosporites striatus Zone. Samples from below 90 m yielded few p a l y n o m o r p h s and could not be dated. The oldest specimens of Phimopollenites pannosus were observed in the upper part of the Coreena Member at 66.70 m depth. Dinoflagellates were present in nearly every sample. A graph representing dinoflagellate percentages in Figure 4 shows their low a b u n d a n c e below 65.80 m, and their assignment to the Pseudoceratium turneri Z o n e is only tentative. T h e assemblages between 48.40 m and 65.80 m represent the

Endoceratium ludbrookiae Zone. Scheibnerova (1981) recovered planktonic and benthonic f o r a m i n i f e r a f r o m 30 m to 66.4 m. She recovered

Lingulogavelinella frankei 49.2 m and 55.6 m.

from the Allaru Mudstone between

WRC Birrigoolpa (30961) This bore was drilled for the New South Wales Department of Mines by the Water Resources Commission, about 40 km southwest of B M R Urisino 1 (lat. 29°44'S, long. 142°52'E). N o distinctive Toolebuc Formation or Urisino beds lithology can be distinguished (Zlotkowski, 1983), and if the lower part of the O o d n a d a t t a F o r m a t i o n is fully present, this section indicates southern margins of deposition of the Urisino beds (Fig. 4). A fine-grained glauconitic sandstone, which we identify as the Coorikiana Sandstone, was drilled below 114 m depth. O n e core was taken at 115.8-120.0 m to sample a level with strong gamma-ray a n o m a l y which had been detected in a previously drilled bore at the same site. Eight samples were taken for palynological examination, of which two, at 118.2 m and 120.0 m, are f r o m the core. T h e Coptospora paradoxa Zone was recognised between 116.9 m and 129.2 m depth. Examination of cuttings suggests that Phimopollenites pannosus first appears at 113.8 m, and is relatively c o m m o n in assemblages f r o m higher levels. The dinoflagellate percentage (Fig. 4) indicates the low abundance recovered. Diagnostic species which would allow identification of assemblages with Morgan's (1980#) zonal intervals were

not found, except for Muderongia tetracantha at 129.2 m, suggesting that this assemblage is not younger than the

Pseudoceratium turneri 'b' Subzone.

BMR Boulia 3 BMR Boulia 3, approximately 13.7 km east of Hamilton Hotel, near Boulia township (lat. 22°47 / 00"S, long. 140°43 '30"E), was drilled and fully logged by BMR (Burger, 1974&). Senior et al. (1975) designated the interval between 25.0 m and 36.0 m depth, which includes black mudstone and concretionary limestone, as a reference section of the Toolebuc Formation (Fig. 4). A weak gamma-ray anomaly is present over the interval. Cores were cut f r o m 20.7 m to 23.2 m, and f r o m 26.2 m to total depth. Spore and pollen assemblages f r o m 39.1 m and 43.1 m are identified as the Coptospora paradoxa Zone, and the lpwer limit of the Phimopollenites pannosus Zone is placed tentatively at 38.5 m. Abundant dinoflagellates recovered from the Wallumbilla F o r m a t i o n allowed samples f r o m 37.2 m to

43.1 m to be referred to the Pseudoceratium turneri 'c' Subzone. Endoceratium ludbrookiae occurs below the Toolebuc F o r m a t i o n at 36.3 m, a n d is a rare element in most overlying assemblages.

BMR Charleville 3-3A These boreholes were drilled and logged by BMR, about 12 km southeast of Charleville township (lat. 26°28'30"S, long. 146°20'00"E). Continuous cores were cut in 3A borehole f r o m 50.0 m to total depth, including the lower part of the Allaru M u d s t o n e and the upper part of the Coreena Member (Fig. 5). Ozimic (1986) f o u n d an u n c o n f o r m a b l e contact between these units at 76.6 m. This borehole was selected for palynological examination to investigate whether the absence of the Toolebuc F o r m a t i o n was due to erosion or to a lateral change of the depositional environment. Seven samples were taken f r o m 73.54 m to 84.80 m depth. The deepest samples yielded sufficient numbers of spores and pollen for the Coptospora paradoxa Z o n e to be identified between 80.70 m and 84.80 m. Assemblages f r o m the higher samples were impoverished a n d lacked the index species

TOOLEBUC FORMATION PALYNOLOGY

145

BMR AUGATHELLA 6 GAMMA RAY API UNITS — INCREASE

PERCENTAGE OF DINOFLAGELLATES

Phimopollenites pannosus

Endoceratium ludbrookiae a

Coptospora paradoxa

Pseudoceratium turneri

Crybe/osporites striatus ?

Total depth

1||P T

•

EPISODE 3B EPISODE 3A EPISODE 2 ? EPISODE 1 ?

91.1 n

WRC BIRRIGOOLPA(30961) OJ

I Total depth

EPISODE 3A

EPISODE 1 ?

134.1m

Stratigraphy see Zlotkowski,1 9 8 3 ; t h i s paper Superfical sediments

A / W W W V

BMR BOULIA3

^r^j&.rf:

3.6 10-

ALLARU MUDSTONE 2025.0

TOOLEBUC FORMATION 36.0

WALLUMBILLA FM

29.8

37 2 38.5 39.1 43.1

Phimopollenites pannosus

Endoceratium ludbrookiae a

Coptospora paradoxa

Pseudoceratium turneri c

EPISODE 2

Total depth 44.5m Stratigraphy see Burger, 1 9 7 4 b

Fig 4 Stratigraphy of two N E R D D C Project boreholes (Qld), one water bore (N.S.W.) (see also Ozimic, 1986), and positions of p e n o l o g i c a l zonal intervals. Lithological symbols explained in Fig. 3.

A. M c M I N N & D. B U R G E R

146

B M R C H A R L E V I L L E 3-3A GAMMA RAY API UNITS-^INCREASE

P. turner/

PERCENTAGE OF DINOFLAGELLATES

EPISODE 3B EPISODE 2 ?

C paradoxa T o t a l d e p t h 91.0m Superficial sediments

BMR CHARLEVILLE 5

TOOLEBUC F O R M A T I O N ~ - n _ T _ ~ - 69.24 73.05

\vy\rw\nj\r\ COREENA MBR T o t a l d e p t h 78.8m •B

- 73.05 - 75.70 N f 77,45 L 78.25

EPISODE 3A Coptospora paradoxa

B M R H A Y R I V E R 12

Phimopollenites pannosus

EPISODE 3B

Endoceratium ludbrookiae a

EPISODE 3A

Coptospora P. turneri

T o t a l d e p t h 86.18m S t r a t i g r a p h y s e e this p a p e r

B M R J E R I C H O 11

Endoceratium ludbrookiae

MBR T o t a l d e p t h 71.15m

EPISODE 4 a

EPISODE 3B

Coptospora paradoxa 60%

I I I I I

Fig. 5. Stratigraphy of four NERDDC Project boreholes (Qld and N.T.) (see also Ozimic, 1986), and positions of palynological zonal intervals. Lithologieal symbols explained in Fig. 3.

TOOLEBUC FORMATION PALYNOLOGY B M R T O O M P I N E GAMMA RAY API UNITS -•INCREASE 70 -

72.15 ALLARU MUDSTONE

147

PERCENTAGE OF DINOFLAGELLATES

Ph. pannosus

79.70 Endoceratium ludbrookiae a

Coptospora paradoxa

EPISODE 4

Pseudoceratium , turneri C

EPISODE 3B EPISODE 3 A

MEMBER Total depth 105.10m

Superficial sediments

k'Ji^'j I

90-92 94-96

§ 2 100 -

102.0

110 URISINO BEDS

120-

B IVI

R U R I S I N O

73.5

Phimopollenites pannosus Endoceratium ludbrookiae a

— I 110-112

122.0

130

132.0 COOR140 - IKIANA i SST 144.0 BULLDOG

116-118

- 122.0

123.7 126.2 129.9 135.4 137.2 139.4 142.5 146.8 148.2

EPISODE 3 A

Coptospora

SHALE Total depth 149.00m

P. turneri

b-c?

Pseudoceratium turneri b

EPISODE 1

60%

I I I I—I—1—I

20/A/38

Fig. 6. Stratigraphy of two NERDDC Project boreholes (Qld and N.S.W.) (see also Ozimic, 1986), and positions of palynological zonal intervals. Lithological symbols explained in Fig. 3.

was identified between 75.70 m and 78.25 m depth. Spores and pollen grains from the Toolebuc Formation were too scarce for an age determination. Dinoflagellates were extremely scarce to absent in the Coreena Member, and common to abundant in the Toolebuc Formation and Allaru Mudstone. The Endoceratium ludbrookiae Zone was identified at 73.05 m and 69.24 m. Spores, pollen, and dinoflagellates recovered from higher levels were extremely poorly preserved, and not suitable for biostratigraphic or BMR Charleville 5 This borehole was drilled and logged by BMR, statistical examination. approximately 11 km west of Charleville township (lat. 26 23'17"S, long. 146°02'04"E). Continuous coring from BMR Hay River 12 59.70 m to total depth (78.8 m) included the lower part of This borehole was drilled and wireline logged by BMR, in the Allaru Mudstone, the Toolebuc Formation, and the upper the channel of the Plenty River, approximately 80 km part of the Coreena Member. Of 39 samples taken from 59.70 southeast of Atula Station (lat. 23°52'05"S, long. m to 78.70 m for palynological examination, a selection of 136°43 '20 "E). The continuously cored interval from 50.0 m 5 productive samples is shown in Figure 5. to total depth includes the lower part of the Oodnadatta Recovery of spores and pollen grains from the Coreena Formation, Wooldridge Limestone Member correlative Member was satisfactory, and the Coptospora paradoxa Zone interval, and Coorikiana Sandstone (Fig. 5). Fourteen samples necessary to indicate zonal affinities. Abundant dinoflagellates were recovered from 78.45 m and higher levels, and the Endoceratium ludbrookiae Zone was recognised at 75.60 m and 73.54 m. The assemblages from 77.30 m and 78.45 m lack the index species E. ludbrookiae and are tentatively assigned to the preceding Pseudoceratium turneri 'c' Subzone (but see palynostatistical evaluation below).

148

A. McMINN & D. BURGER ALBIAN Early

Middle

Late

Crybelosporites \ Coptospora \ Phimopollsthatus

I paradoxa

, p™fsus

\ D E T T M A N N A N D PLAYFORD. 1 9 6 9 , D E T T M A N N A N D DOUGLAS, 1 9 7 6 — -I-

Asteropollis asteroides Clavatipollenites s s p * Coptospora paradoxa -p * Crybelosporites striatus Dictyotosporites speciosus —L Hoegisporis uniforma Microro veolatosporis canaliculars

PHosisporites grandis Rousea georgensis several tricolpate s p e c i e s Tricolpites variabilis Trilobosporites trioreticulosus

i c

Pseudoceratium turneri

Ifrcj Endoceratium ludbrookiae

* Endoceratium I Heslertonia striata Hystrichosphaeridium arundum Muderongia tetracantha Odontocnitina striatoperforata Oodnadattia tuberculata * Pseudoceratium turneri

MORGAN, 1978, 1980 a 20/A/39

Fig. 7. Stratigraphic ranges of selected zonal index palynomorph species, Albian, Great Artesian Basin.

collected for palynological examination yielded reasonably diverse spore and pollen assemblages. The Coptospora paradoxa Zone was identified in the Coorikiana Sandstone between 73.72 m and 86.12 m. The Phimopollenitespannosus Zone commences in the upper part of the Coorikiana Sandstone at 73.72 m, and was recognised also in the Oodnadatta Formation. The assemblages include small to moderate fractions of

dinoflagellates. The Endoceratium ludbrookiae Zone was

observed in the interval from 50.00 m to 80.00 m, and the assemblages from 83.35 m to 86.12 m, which lack E.

ludbrookiae, are tentatively assigned to the Pseudoceratium turneri Zone.

BMR Jericho II

This borehole was drilled approximately 25 km southeast of Barcaldine township (lat. 23°52'30"S, long. 145°3ri8"E), and fully logged by BMR. Continuous coring from 40.0 m to total depth covered the Allaru Mudstone, Toolebuc Formation, and upper part of the Coreena Member. Sixteen samples were collected for palynological examination, 9 of which are shown in Figure 5. Recovery of spores and pollen grains was satisfactory. The Coptospora paradoxa Zone was identified in the Coreena Member at 70.85 m and 68.15 m, and possibly also in the Toolebuc Formation at 65.15 m. The oldest specimens of Phimopollenites pannosus were observed at 59.15 m. Recovery of dinoflagellates from the Coreena Member and Toolebuc Formation was insufficient to establish zonal affinities. The assemblages from the Allaru Mudstone at 51.25 m and 53.25 m depth were identified as part of the Endoceratium

ludbrookiae 'a' Subzone.

Santos Oodnadatta 1

Santos Limited drilled this well in 1957 about 15 km northwest of Oodnadatta township (lat. 27°25'30"S, long. 135°19'20"E). It intersected a problematic and much discussed Lower Cretaceous marine sequence near the western margin of the Eromanga Basin (Fig. 3). The intervals between 6.5 m and 314.5 m, and between 396.0 m and total depth were

continuously cored. Unfortunately, no wireline logs were recorded. The well has provided valuable information on sequences of spores and pollen, dinoflagellates, and foraminifera, which have enabled detailed correlations to be made with the Lower Cretaceous of Queensland and New South Wales. The stratigraphy of the well has been described by Sprigg (1958), Ludbrook (1966), Freytag (1966), Wopfner et al. (1970), and Griffiths (1980). Their various stratigraphic interpretations illustrate the uncertainty as regards the nomenclature to be used for this part of the basin in the subsurface. The recorded lithology shows dense, dark, slightly sandy and silty mudstone above 128 m. A richly fossiliferous interval occurs between 96 m and 103 m depth which Freytag (1966) and Griffiths (1980) interpreted as the continuation in the subsurface of the Wooldridge Limestone Member. Below 128.5 m the lithology consists of dark grey, medium to fine-grained and steeply crossbedded sandstone, which several workers take to be an extension of the Coorikiana Sandstone in the subsurface. Brunnschweiler (1959) reported Albian ammonites

(Falciferella), bivalves (Aucellina, Inoceramus), and

gastropods from the Tambo Formation' (no depths were indicated by the author). Ludbrook (1966) studied the foraminifera in a comprehensive series of samples taken at 3 m (10 ft) intervals. She considered the 'upper Marree Formation' between 118.9 m and 64.0 m, which includes her Zone of Neobulimina australiana, to be of late Albian age. She recovered the planktonic species Hedbergella infracretacea, which Haig (1979) thought to be indicative of a late Albian age, between 131 m and approximately 67 m depth. Scheibnerova (1980), who also gave a detailed account of the foraminifera in this well, observed her Lingulogavelinella frankei Datum horizon within the Bulldog Shale at 198 m, and considered the Toolebuc event' as occurring at 100 m. The sequence of spores and pollen has been described by Dettmann (1963), Dettmann & Playford (1969), and Playford et al. (1975). As stated above, this well includes the reference

sections of the Crybelosporites striatus, Coptospora paradoxa, and Phimopollenites pannosus Zones. The sequence of dinoflagellates includes the Pseudoceratium turneri Zone between 221.9 m and 128.0 m, and the

Endoceratium ludbrookiae Zone between 124.1 m and 30.5 m (Morgan, 1978, 1980*7).

BMR Toompine 1

BMR Toompine 1 was drilled approximately 100 km northwest of Cunnamulla in southwestern Queensland (lat. 27°48 'S, long. 144°42'E) and logged by BMR (Fig. 6). The Urisino beds occur between 101.2 m and 102.2 m, overlain by the marine argillaceous Allaru Mudstone, and underlain by nonmarine, silty and argillaceous sandstone of the Coreena Member. Cores were cut at 10.0-12.0 m and 40.6-45.0 m depth, and continuously from 71.0 m to total depth. The recovery of spores and pollen grains was average to poor, and species diversity was low. Phimopollenites pannosus was found only in the assemblage from 72.15 m in the Allaru Mudstone; the interval between 79.70 m and 105.05 m is attributed provisionally to the Coptospora paradoxa Zone. The lower limit of the Endoceratium ludbrookiae Zone was observed at 102.15 m, immediately above the base of the Urisino beds, and the assemblage from 102.65 m probably

represents the Pseudoceratium turneri 'c' Subzone. No

dinoflagellates were recovered from 105.05 m.

TOOLEBUC FORMATION PALYNOLOGY

BMR Urisino 1

This bore was drilled and logged by BMR, in cooperation with the Geological Survey of New South Wales, approximately 80 km west of Wanaaring township (lat. 29°28 '27 "S, long. 143°17'52"E). Continuous cores were cut between 6 m and 77 m, and between 120 m and total depth (Fig. 6). Palynological samples were taken from both cores and cuttings. Ozimic (1986) recognised the Allaru Mudstone between 26 m and 102 m, the Urisino beds (Toolebuc Formation equivalent') to 122 m, and part of the Coreena Member of the Wallumbilla Formation to total depth at 149 m. We prefer to adopt the alternative interpretation shown in Figure 6, which Burger (1981) suggested while commenting on the similarity of this section to the Lower Cretaceous in South Australia, and we also adopt this interpretation for the WRC Birrigoolpa bore (Fig. 4). In the sequence of spores and pollen, Phimopollenites pannosus first appears at 120.0 m; the underlying assemblages are assigned to the Coptospora paradoxa Zone. Burger's (1981) previous identification of the Phimopollenites pannosus Zone at 126 m rested on the presence of the monolete spore

Mic'rofoveolatosporis canaliculars, which we know now to

range far into the Coptospora paradoxa Zone. Dinoflagellates are abundant in the Urisino beds, and scarce to absent in other intervals. Byrnes (1980) recognised a nonmarine emergent sequence characterised by desiccation cracks and plant roots between 124.3 m and 126.5 m. Endoceratium ludbrookiae first appears at the base of the Urisino beds at 122.0 m and occurs sporadically above this level. The highest occurrence of Muderongia tetracantha, which indicates the upper limit of the Pseudoceratium turneri 'b' Subzone, is at 139.4 m. No diagnostic species were recovered between that level and the Urisino beds, and assemblages from the 129.9-137.2 m interval are tentatively assigned to the Pseudoceratium turneri 'b-c' zonal interval.

SUMMARY OF BOREHOLE DATA

The above data demonstrate that environmental factors may place limitations on palynostratigraphic interpretations, and the following summary may help the reader in understanding the grounds on which we have based our conclusions. It is clear that the oldest stratigraphic appearance of

Phimopollenites pannosus in the sequence of spores and

pollen is a potentially valuable reference point for the NERDDC Project. At present this is the only criterion by which to distinguish the Phimopollenites pannosus Zone from

the preceding Coptospora paradoxa Zone. P. pannosus is

found readily in the Allaru Mudstone in Queensland. Studies in the Carpentaria Basin (in progress) suggest that the species is present also in the Toolebuc Formation, and the present study confirms this to be the case in BMR Boulia 3, BMR Jericho 11, and also in the Urisino beds of BMR Urisino 1, and the Wooldridge Limestone Member correlative of BMR Hay River 12. Isolated specimens of the species occur also in assemblages from the uppermost part of the Coreena Member in BMR Augathella 6, and the Coorikiana Sandstone in BMR Hay River 12. This log of data suggests that in central Queensland and Northern Territory the Phimopollenites pannosus Zone commences in the greensands of the upper Coreena Member and the Coorikiana Sandstone, and elsewhere in, or slightly below, the Toolebuc Formation, Urisino beds, and Wooldridge Limestone Member correlative. If this is the case, the top of

149

the greensands might be slightly diachronous, but further data are needed to verify this. We feel that the extension of the Coptospora paradoxa Zone into the Urisino beds and Allaru Mudstone in BMR Toompine 1 (Fig. 6) is anomalous and might flow from a high proportion of dinoflagellates in the palynological assemblages, which makes the presence of Phimopollenites pannosus very difficult to ascertain. We accept that the Phimopollenites pannosus Zone commences at, or slightly below, the base of the Toolebuc Formation (sensu lato) and that this Toolebuc event' represents by close approximation a time-concordant horizon in the sedimentary sequence of the Eromanga Basin. A potentially significant biostratigraphic horizon is marked by the oldest stratigraphic appearance of the dinoflagellate species Endoceratium ludbrookiae, which separates the zone of that name from the preceding Pseudoceratium turneri Zone. Burger (1982) identified the latter zone in the upper Wallumbilla Formation of the Hughenden region. An accurate determination of the first stratigraphic appearance of Endoceratium ludbrookiae is complicated by the fact that the oldest occurrence of the species in Queensland and New South Wales is facies-bound to a marine transgression connected with the Toolebuc event'. We recovered the species from the Toolebuc Formation and Urisino beds in several boreholes, and below the Toolebuc Formation (sensu lato) in BMR Boulia 3 and BMR Hay River 12. The observations from BMR Hay River 12 confirm earlier studies (Morgan, 1978) by indicating that the Endoceratium ludbrookiae Zone commences slightly earlier in time than the concurrent

Phimopollenites pannosus Zone.

GAMMA-RAY ANOMALIES

Weak to strong positive anomalies are shown on gammaray logs of the Toolebuc Formation in BMR Jericho 11, BMR Charleville 5, BMR Boulia 3, and BMR Augathella 6, in the latter boreholes within the interval of the Phimopollenites pannosus Zone. This anomaly occurs commonly in the Toolebuc Formation in borehole logs from central Queensland, and has been traced to concentrations of uranium in the sediment (Senior et al., 1978; Ramsden, 1980). Similar radioactive peaks have been observed also in petroleum wells in South Australia, and Moore & Pitt (1982) interpreted them as an extension of the Toolebuc Formation in the southwestern part of the basin. Ozimic (1986), who reviewed the log characteristics of the formation, found that the radioactive peaks occur wherever oil shales have been formed. However, anomalies due to the presence of uranium have been observed also in gamma-ray logs of many water bores in the Bulloo Embayment in northern New South Wales, where so far no oil shale has been observed (Ramsden et al., 1980; Herbert, 1980). Positive anomalies are registered also in gamma-ray logs of the Coorikiana Sandstone of BMR Urisino 1 and the WRC Birrigoolpa bore. Ramsden et al. (1982) attributed the anomaly in BMR Urisino 1 to the presence of potassium in a siltstone band. Burger (1981) and McMinn (1983) showed that the anomalies in both boreholes fall within the interval of the Coptospora paradoxa Zone, and are therefore older than the gamma-ray anomalies within the Toolebuc Formation.

EVALUATION OF PALYNOSTATISTICS

Statistical palynological studies of Recent sediments on the African continental shelf by Williams & Sarjeant (1967), Davey (1971), and Davey & Rogers (1975) have demonstrated

150

A. McMINN & D. BURGER

a broad relationship between the ratio of spores/pollen to dinoflagellate cysts in palynological assemblages, and the distance of sampled sites from the coast, in that dinoflagellates were found to increase proportionally towards the outer edge of the shelf. This concept was adopted in statistical palynological studies in the Lower Cretaceous of the Surat Basin (Burger, 1980), and the conclusions drawn on the basis that salinity was the major factor influencing dinoflagellate distribution, were generally in good agreement with other palaeoenvironmental evidence. Other factors, such as variations in water temperature, concentration of nutrients, and amount of available sunlight must have contributed, but their effects cannot be gauged from the statistical record. In many instances apparent statistical anomalies can be explained naturally by an excessive supply of certain spores and pollen species from nearby sources. In the Eromanga Basin, sample frequency has been much less favourable for this kind of study, but in recent years several boreholes, continuously cored by the Geological Survey of Queensland, have yielded outstanding results on palaeoenvironments in the Early Cretaceous. Statistical evidence from the northern Eromanga Basin (Burger, 1982) corresponded well with recorded events elsewhere in the basin. The NERDDC Project boreholes provided detailed statistical information from a narrow stratigraphic interval during the Albian, which could be integrated successfully into the broader picture. We feel, therefore, that the overall uniform sedimentation pattern was a good indication that events generally occurred on a basin-wide scale, and that statistical anomalies caused by local factors were unlikely to seriously distort the overall picture. We counted between 200 and 450 specimens at random in palynological assemblages from the boreholes reviewed above. Percentages of dinoflagellates are illustrated graphically for each borehole in Figures 3 to,6. We adopt the figures which Burger (1980) found for the Surat Basin as a guideline, interpreting 0-3% of dinoflagellates in palynological assemblages to indicate freshwater to marginally brackish marine environments, 3-8% of dinoflagellates brackish to coastal marine environments, and higher percentages to indicate open marine, i.e. epi-neritic to neritic conditions. On the basis of these percentage groups the graphs indicate recurrent changes from nonmarine to open marine environments at different locations before, during and after deposition of the Toolebuc Formation (sensu lato). Graphs of individual bore data are set out together in Figure 8, to compare dinoflagellate maxima and magnitudes in various parts of the basin, and to show how they interrelated in time. Sampling was incomplete below the Toolebuc Formation (sensu lato) but indicates that recovery of dinoflagellates was variable. This is interpreted in terms of overall marginal, mildly saline environments in the eastern and southern areas of the basin. We counted increased d i n o f l a g e l l a t e percentages in assemblages of the Pseudoceratium turneri 'b' Subzone in WRC Birrigoolpa (3%), BMR Urisino 1 (5%), and Santos Oodnadatta 1 (9%), and conclude that mildly saline conditions existed slightly below and within the Coorikiana Sandstone in the south and west of the basin. The temporally coincident fluctuations in the three drillholes suggest a brief marine incursion into the basin during the Pseudoceratium turneri 'b' Subzone, which we designate as Episode 1. This incursion may have been a basin-wide feature, since moderate to high dinoflagellate percentages have been found also in assemblages of the same subzone in the Coorikiana Sandstone of DFS Betoota 1

(Burger, 1981), and the upper Wallumbilla Formation of GSQ Hughenden 7 (Burger, 1982). High proportions of dinoflagellates were counted in assemblages from below the Toolebuc Formation in BMR Boulia 3 (maximum 73%). These counts fall within the Pseudoceratium turneri 'c' Subzone and thus might indicate a later marine incursion (Episode 2). Dinoflagellate percentages of 8% and 25% were counted for the Coreena Member of BMR Augathella 6. They fall within the interval of the Coptospora paradoxa spore-pollen Zone, and cannot be older than the Pseudoceratium turneri 'b' Subzone (see below), so that they probably indicate brief marine incursions of Episodes 1 and 2. We observed a moderate to substantial increase in the dinoflagellate percentage in assemblages from at or near the onset of the Endoceratium ludbrookiae Zone near the base of the Toolebuc Formation in BMR Augathella 6 and BMR Charleville 5 (over 70%), the Urisino beds in BMR Toompine 1 (48%), and below the Wooldridge Limestone Member correlative in Santos Oodnadatta 1 (31%). Also, marine shelly fossils are rare to absent in the upper part of the Coreena Member in the Tambo and Augathella areas, but reappear in the Toolebuc Formation (Day, 1969). This is taken to indicate that an important marine transgression occurred in the western and northern regions (here referred to as Episode 3A). Dinoflagellate assemblages are heavily dominated by the genus Diconodinium, and this is probably a sign of restricted conditions at the time of deposition. Moderate to low dinoflagellate percentages in BMR Jericho 11, BMR Boulia 3, and WRC Birrigoolpa (30961) probably indicate brackish to nonmarine horizons at the very edge of the Toolebuc sea'. The graphs of Santos Oodnadatta 1, BMR Hay River 12, BMR Boulia 3, and possibly also BMR Augathella 6, indicate marine conditions persisting in the west and north between Episodes 2 and 3A, slightly below the Toolebuc event'. Figure 8 does not include BMR Charleville 3-3A, which is shown in Figure 5. A sudden increase of dinoflagellates in the Coreena Member in this borehole (70-80%) indicates a marine transgression which could represent Episode 2 in the Pseudoceratium turneri 'c' Subzone. If this is the case, the unconformity at 76.6 m depth spans a sequence which would include Episode 3A, which means that locally the Toolebuc Formation was eroded. However, we feel that the dominance of the genus Diconodinium (up to 50%) in the dinoflagellate assemblages might conceivably have masked the presence of E. ludbrookiae. We do not exclude the possibility that the assemblages might represent marine Episode 3A in the Endoceratium ludbrookiae Zone. Moderate to high percentages of dinoflagellates have been found in the upper part of the Toolebuc Formation and lower part of the Allaru Mudstone of BMR Charleville 5 (not counted), BMR Hay River 12 (7.5%), BMR Jericho 11 (maximum 33%), BMR Augathella 6 (maximum 86%), BMR Charleville 3-3A (maximum 89%), and BMR Toompine 1 (maximum 45%), and also in and above the Wooldridge Limestone Member correlative in Santos Oodnadatta 1 (26%). These percentages are sufficiently high to signify a new marine incursion in the central and western regions (here indicated as Episode 3B). Reduced dinoflagellate percentages in BMR Urisino 1, BMR Boulia 3, and WRC Birrigoolpa indicate that regressive conditions prevailed in the western and southern regions Fig.

8. M a r i n e episodes, Albian, Great Artesian Basin, from dinoflagellate records in N E R D D C Project and other drillholes.

151

TOOLEBUC FORMATION PALYNOLOGY Pseudoceratium turneri

Depth

Endoceratium ludbrookiae

(metres)

| b-c

) )3 3 I SANTOS OODNADATTA 1

\ B M R H A Y RIVER 12

BMR BOULIA3

B M R J E R I C H O 11

BMR AUGATHELLA 6

B M R CHARLEVILLE 5

BMRTOOMPINE1

i BMR URISINO 1

WCR BIRRIGOOLPA (30961)

152

A . M c M I N N & D. B U R G E R

before the end of Toolebuc F o r m a t i o n deposition. A possible fresh incursion of the sea ( E p i s o d e 4) might be indicated by isolated c o u n t s of 4 % a n d 3 % above 55 m d e p t h in S a n t o s O o d n a d a t t a 1, a n d increasing d i n o f l a g e l l a t e p r o p o r t i o n s in assemblages f r o m the Allaru M u d s t o n e in B M R T o o m p i n e 1 ( 4 6 % ) a n d B M R J e r i c h o 11 (21%). M a r i n e E p i s o d e s 1 to 4 are d r a w n in Figure 9 as a curve of successive high a n d low sea levels in the E r o m a n g a Basin, a n d the a m p l i t u d e s of the m a x i m a roughly indicate the estimated regional effect of each high sea level phase. Episodes 1 a n d 3A-B clearly represent m a r i n e transgressions into the basin which M o r g a n (1980b) associated with world-wide phases of high eustatic sea level in the late early A l b i a n a n d the early late Albian. We agree with the eustatic interpretation

ALBIAN EARLY

Crybe/o- 1 sporites \ striatus |

LATE

MIDDLE

i Coptospora paradoxa

Phimopollenites pannosus 1

i | 1

of the Episodes, in view of the fact t h a t Exon & Burger (1981) a n d Burger (1986) f o u n d a definite r e l a t i o n s h i p between recurrent cyclic deposition in the Surat a n d E r o m a n g a Basins, a n d successive high a n d low eustatic sea levels, as have been discussed by Vail et al. (1977), C o o p e r (1977), a n d Vail & Todd (1981). We accept, therefore, t h a t t h e Toolebuc F o r m a t i o n , Urisino beds, a n d Wooldridge Limestone M e m b e r correlative ( f r o m their close a s s o c i a t i o n with eustatic E p i s o d e s 3A-B) represent a single time rock unit in the E r o m a n g a Basin. Other m a r i n e f l u c t u a t i o n s shown in Figure 9 may have been locally (tectonically?) i n d u c e d , or were t o o weak to register outside Australia.

ORGANIC MATURATION To obtain a m e a s u r e of the degree of m a t u r i t y of oil shale deposits in the n o r t h e r n p a r t of the E r o m a n g a Basin, we checked the c o l o u r of various organic particles f r o m the Toolebuc F o r m a t i o n in B M R Boulia 3, B M R Augathella 6, B M R J e r i c h o 11, a n d B M R Charleville 5. C o m p a r i s o n s were m a d e also with p a l y n o m o r p h s f r o m the u p p e r p a r t of the Coreena M e m b e r in B M R Augathella 6, B M R Jericho 11, and the C o o r i k i a n a S a n d s t o n e of B M R H a y River 12. We m a d e these checks d u r i n g l a b o r a t o r y processing of t h e rocks prior to o x i d a t i o n . T h e r e is n o evidence of deep weathering which w o u l d have s i g n i f i c a n t l y a l t e r e d t h e c o l o u r of the p a l y n o m o r p h s . We selected Cyathidites minor Couper, one of the most c o m m o n fern spores in Australian Mesozoic floras, a n d Diconodinium psilatum M o r g a n , which is a c o m m o n p e r i d i n i o i d species in A l b i a n d i n o f l a g e l l a t e assemblages in the basin. We also recorded the colour of structureless organic ('sapropelic') matter, which was recovered in great q u a n t i t i e s f r o m the t o p of the C o r e e n a Member, Toolebuc F o r m a t i o n , a n d basal A l l a r u M u d s t o n e , and a c c o r d i n g t o G l i k s o n (1982) m a y r e s u l t f r o m t h e b i o d e g r a d a t i o n of p r i m a r y organic m a t t e r in the sediment by cyanobacteria. Table 1 shows the c o l o u r ranges observed for different p a l y n o m o r p h s a n d the dispersed organic matter. This matter was usually f o u n d finely dispersed in the microscope p r e p a r a t i o n s f r o m the C o r e e n a M e m b e r a n d its colour varied f r o m grey to b r o w n . In the Toolebuc F o r m a t i o n it tended to f o r m b r o w n c l u m p s , which were d i f f i c u l t to break down in the laboratory.

TOOLEBUC FORMATION (sensu

fa to)

20/A/41

Fig. 9. Sea level movements, E r o m a n g a Basin (Episodes 1 to 4), and their association in time with Albian global eustatic sea level movements.

J u d g i n g by Staplin's (1977) c o l o u r ratings t h e Toolebuc F o r m a t i o n was f o u n d to be t h e r m a l l y i m m a t u r e , a n d a mean vitrinite reflectance of less t h a n 0.4 per cent stated by Ozimic (1986) c o n f i r m s this rating. T h e presence of a free-flowing kerosene-like fluid observed in s o m e b o r e h o l e s seems to indicate seepage f r o m a d i f f e r e n t source, as this fluid has the characteristics of a m a t u r e h y d r o c a r b o n of probably n o n m a r i n e origin. T h e 'mature' character of the p a l y n o m o r p h s in the Coreena M e m b e r a n d C o o r i k i a n a S a n d s t o n e is p r o b a b l y misleading.

COREENA MEMBER COORIKIANA SANDSTONE

TOOLEBUC FORMATION

spores and pollen grains

yellow to brown

colourless to light yellow

dinof/age/lates

colourless to yellow

colourless

sapropelic' detritus

grey or light brown

light to dark brown

20/A/44 TABU

1. C o l o u r of constituents in organic residue recovered f r o m selected boreholes.

TOOLEBUC FORMATION PALYNOLOGY Spores and pollen grains enclosed in sedimentary rocks exposed to deep weathering gradually change in chemical composition until all that remains are carbon residues. Bearing in mind the possible extent to which these rock units may have been exposed during marine regressive episodes, we believe that the spore colouration bears no relation to the thermal history of the greensands.

CONCLUSIONS

From the foregoing palynological review of the Albian rock sequence in the Eromanga Basin the following conclusions have been drawn. (1) The lower to middle Albian Coreena Member of the Wallumbilla Formation is palynologically correlated with the upper part of the Bulldog Shale, and the upper part of the member with the Coorikiana Sandstone. The Toolebuc Formation, Urisino beds, and Wooldridge Limestone Member correlative represent a lower upper Albian time rock unit which is recognised in Queensland, New South Wales, South Australia, and Northern Territory. The upper Albian Allaru Mudstone is correlated with the upper part of the Oodnadatta Formation which overlies the Wooldridge Limestone Member correlative. The associations found between the sedimentary units and the palynological zonal intervals are shown in Figure 2.

(2) Evidence from BMR Charleville 3-3A suggests that locally the eastern margin of the Toolebuc Formation in the

153

subsurface may be erosionally controlled. Towards the south, evidence from WRC Birrigoolpa suggests that the formation passes laterally into a non-calcareous marine mudstone. (3) Four successive episodes of marine deposition are detected in the interval studied. Episode 1 (late early Albian) and Episode 3 (early late Albian) are shown to be eustatically controlled. Episode 3 is associated with the Toolebuc Formation oil shales, and restricted conditions are reflected in the palynological record by a profusion of the dinoflagellate

genus Diconodinium.

(4) No relationship is found to exist between the thermal maturation of the Toolebuc Formation, as assessed by colouration of the organic matter, and the presence of a free kerosene-like fluid in the formation.

ACKNOWLEDGEMENTS

The authors are indebted to Dr G. Playford for his detailed comments and criticism. Messrs H. F. Doutch and S. Ozimic offered helpful suggestions and pointed out several omissions. The Exploration Manager of Santos Ltd in Adelaide very kindly provided a copy of the lithological and drilling logs of Santos Oodnadatta 1 in northern South Australia, on which Figure 3 is based. The Secretary of the Department of Mineral Resources of New South Wales and the Director of the Bureau of Mineral Resources in Canberra gave permission to publish this paper.

REFERENCES BRUNNSCHWEILER, R. O., 1959: New Aconoceratinae (Ammonoidea) from the Albian and Aptian of Australia. Aust. Bur. Miner. Resour. Geol. Geophys. Bull. 54. BURGER, D., 1968: Palynology of marine Lower Cretaceous strata in the northern and eastern Eromanga Basin, Queensland. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1 9 6 8 / 6 2 (unpubl.). BURGER, D., 1973: Palynological observations in the Carpentaria Basin, Queensland. Aust. Bur. Miner. Resour. Geol. Geophys. Bull. 140, 27-44. BURGER, D., 1974A: Palynology of subsurface Lower Cretaceous strata in the Surat Basin of Queensland. Aust. Bur. Miner. Resour. Geol. Geophys. Bull. 150, 27-42. BURGER, D., 19746: Stratigraphic drilling in the Boulia region, western Queensland. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1 9 7 4 / 3 9 (unpubl.). BURGER, D., 1980: Palynological studies in the Lower Cretaceous of the Surat Basin, Australia. Aust. Bur. Miner. Resour. Geo!. Geophys. Bull. 189. BURGER, D., 1981: Palynology of three NERDDC stratigraphic bores in the Eromanga Basin: preliminary report. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1 9 8 1 / 3 2 (unpubl.). BURGER, D., 1982: Palynological examination of Late Mesozoic sediments in GSQ Hughenden 7, and notes on geological events in the northern Eromanga Basin. Qld. Gov. Min. J. 83, 421-32. BURGER, D., 1986: Palynology, cyclic sedimentation, and palaeoenvironments in the Late Mesozoic of the Eromanga Basin; in This volume. BYRNES, J. G., 1980: BMR Urisino No. 1 bore and its bearing on prospects for oil shale in the Eromanga Basin, N.S.W. N.S.W. Geol. Surv. Rep. 1 9 8 0 / 1 7 7 (unpubl.). BYRNES, J. G., MORGAN, R. & SCHEIBNEROVA, V., 1975: Recent

evidence for the age of Great Australian Basin Cretaceous sediments in New South Wales. N.S.W. Geol. Surv. Q. Notes, 18, 2-13.

CLARKE, R. F. A . & VERDIER, J. P., 1967: A n investigation o f

microplankton assemblages from the Chalk of the Isle of Wight, England. Kon. Nederl. Akad. Wetensch. (Natuurk.) Verh. le Rks, 24. COOPER, M. R., 1977: Eustacy during the Cretaceous: its implications and importance. Palaeogeogr. Palaeoclimatol. Palaeoecol. 22, 1-160. DAVEY, R. J., 1971: Palynology and palaeo-environmental studies, with special reference to the continental shelf sediments of South Africa. Proc. II Planet. Conf Roma., 331-47. DAVEY, R. J. & ROGERS, J., 1975: Palynomorph distribution in recent

offshore sediments along two traverses off South West Africa. Marine Geol. 18, 213-25. DAY, R. W., 1969: The Lower Cretaceous in the Great Artesian Basin; in Campbell, K. S. W. (ed.), Stratigraphy and Palaeontology, Essays in honour of Dorothy Hill. ANU Press, Canberra, 140-73. DETTMANN, M. E., 1963: Upper Mesozoic microfloras from southeastern Australia. R. Soc. Vic. Proc. 77, 1-148. DETTMANN, M. E. & DOUGLAS, J. G., 1976: Palaeontology. Geol. Soc. Aust. Spec. Publ. 5, 164-9.

DETTMANN, M. E. & PLAYFORD, G., 1969: P a l y n o l o g y o f the

Australian Cretaceous: a review; in Campbell, K. S. W. (ed.), Stratigraphy and Palaeontology, Essays in honour of Dorothy Hill. ANU Press, Canberra, 174-210. EVANS, C. R. & STAPLIN, F. L., 1971: Regional facies of organic metamorphism. Can. Inst. Miner. Metall. Spec. Vol. 11, 517-20. EVANS, P. R. & HAWKINS, P. J., 1967: The Cretaceous below the Murray Basin. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1967/137 (unpubl.). EXON, N. F. & BURGER, D., 1981: Sedimentary cycles in the Surat Basin and global changes of sea level. BMR J. Aust. Geol. Geophys. 6, 153-9. FREYTAG, I. B., 1966: Proposed rock units for marine Lower Cretaceous sediments in the Oodnadatta region of the Great Artesian Basin. S. Aust. Geol. Surv. Q. geol. Notes, 18, 3-7.

A. McMINN & D. BURGER

154

GLIKSON, M., 1982: Cyanobacterial mats and other bacteria; major contributors to the formation of Lower Cretaceous Toolebuc oil shales; in Moore, P. S. & Mount, T. J. (compilers), Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 203. GRIFFITHS, M., 1980: Tooclla No. 1 Well Completion Dep. Mines Ener. Rep. 80/128 (unpubi).

Report. S. Aust.

HAIG, D. W., 1979: Cretaceous foraminiferal biostratigraphy of Queensland. Alcheringa, 3, 171-87. HERBERT, C., 1980: Prospects for oil shale in the Eromanga Basin, New South Wales. N.S.W Geol. Surv. Q. Notes, 40, 1-6. LUDBROOK, N. H., 1966: Cretaceous biostratigraphy of the Great Artesian Basin in South Australia. S. Aust. Geol. Surv. Bull. 40. MCMINN, A., 1980a: Preliminary palynology of BMR Urisino 1, Great Australian Basin, Part 1. N.S.W. Geol. Surv. Rep. 1980/115 (unpubi). MCMINN, A., 1980b: Preliminary palynology of BMR Urisino 1, Great Australian Basin, Part 2. N.S.W. Geol. Surv. Rep. 1980/270 (unpubi). MCMINN, A., 1983: The Toolebuc Formation in the Eromanga Basin of New South Wales. N.S.W Geol. Surv. Q. Notes, 50, 1-7. MIELIOUD, M. E., 1969: Dinoflagellates and acritarchs from some western European Lower Cretaceous type localities. First Int. Conf. Plankt. Microfoss., Geneva, 1967, Proc. 2, 420-34. MOORE, P. S. & PITT, G. M., 1982: C r e t a c e o u s of t h e s o u t h w e s t e r n

Eromanga Basin: stratigraphy, facies variations and petroleum potential; in Moore, P. S. & Mount, T. J. (compilers), Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 127-44. MORGAN, R., 1975: Some Early Cretaceous organic-walled microplankton from the Great Australian Basin. R.'Soc. N.S.W. J. Proc. 108, 157-67. MORGAN, R., 1978: Early and Middle Cretaceous of Australia. Ph.D. thesis, Univ. Adelaide

palynostratigraphy (unpubi).

MORGAN, R., 1980a: Palynostratigraphy of the Australian Early and Middle Cretaceous. N.S.W. Geol. Surv. Mem. Palaeontoi 18. MORGAN, R., 1980b: Eustacy in the Australian Early and Middle Cretaceous.N.S.W. Geol. Surv. Bull. 27. OZIMIC, S., 1982: Depositional environment of the oil shale-bearing Toolebuc Formation and its equivalents, Eromanga Basin, Australia; in Gray, J. H. (ed.), Proc. Fifteenth Oil Shale Symp. Colorado School of Mines, Golden, Colo., 137-48. OZIMIC, S., 1986: The geology and petrophysics of the Toolebuc Formation and its time equivalents, Eromanga and Carpentaria Basins; in This volume. PLAYFORD, G., H A I G ,

D. W . & D E T T M A N N , M . E . ,

1975: A

mid-

Cretaceous microfossil assemblage from the Great Artesian Basin, northwestern Queensland. Neues Jahrb. Geol. Palaont. Abh. 149, 333-62.

RAMSDEN,

R.,

DICKSON,

L.,

MEAKINS,

BENNETT,

A. J. R., 1980: Origin and significance of the gamma-ray anomaly, BMR Urisino 1 stratigraphic hole. Aust. CSIRO, Inst. Earth Resour. Restr. Invest. Rep. 1204R (unpubi). REYMENT, R. A., 1964: Albian ammonites from Fossil Creek, Oodnadatta, South Australia. R. Soc. S. Aust. Trans. 88, 21-36. SCHEIBNEROVA, V., 1980: Comparative foraminiferal biostratigraphy of the Santos Oodnadatta No. 1 section, South Australia. N.S.W. Geol. Surv. Rec. 19. SCHEIBNEROVA, V., 1981: Foraminifera and other microfossils in BMR Augathella 5, 6, and 7. N.S.W. Geol. Surv. Rep. 1981/15 (unpubi). SCHEIBNEROVA, V., 1983: Revised correlation of Cretaceous lithostratigraphic units in New South Wales. N.S.W. Geol. Surv. Q. Notes, 52. SCHEIBNEROVA, V. & BYRNES, J. G., 1977: C o r r e l a t i o n of Cretaceous

lithostratigraphic, lithological, palynological, and foraminiferal units in the Great Australian Basin. N.S.W. Geol. Surv. Q. Notes, 27, 1-11. S E N I O R , B. R „ E X O N , N . F. & B U R G E R , D . , 1 9 7 5 : T h e

Cadna-owie

and Toolebuc Formations in the Eromanga Basin, Queensland. Qld Gov. Min. J. 76, 445-55. S E N I O R , B. R . , M O N D , A . & H A R R I S O N , P. L . , 1 9 7 8 : G e o l o g y o f t h e

Eromanga Basin. Aust. Bull. 167.

Bur. Miner. Resour. Geol.

Geophys.

S M A R T , J . , G R I M E S , K . G . , D O U T C H , H . F. & P I N C H I N , J . , 1 9 8 0 : T h e

Mesozoic Carpentaria Basin and the Cainozoic Karumba Basin, north Queensland. Aust. Bur. Miner. Resour. Geol. Geophys. Bull. 202. SPRIGG, R. C., 1958: The Great Artesian Basin in South Australia. Geol. Soc. Aust. J. 5, 88-101. STAPLIN, F. L., 1977: Interpretation of thermal history from color of particulate organic matter—a review. Palynology, 1, 9-18. THIERSTEIN, H. R., 1974: Calcareous nannoplankton—Leg 26, Deep Sea Drilling Project. Deep Sea Drill. Proj. Initial Rep. 26, 619-67. THOMPSON, B. P., 1980: South Australia 1:1 000 000 geological map. S. Aust. Geol. Surv. VAIL, P. R. & TODD, R. G., 1981: Northern North Sea Jurassic unconformities, chronostratigraphy and sea-level changes from seismic stratigraphy; in: Cling, J. V. & Hobson, C. D. (eds), Petroleum geology of Continental Shelf of north-west Europe. Institute of Petroleum, London, 216-35. VAIL,

R.,

MITCHUM,

THOMSON,

S.,

1977:

Seismic

stratigraphy and global changes of sea level, Part 4: Global cycles of relative changes of sea level; in Payton, C. E. (ed.), Seismic stratigraphy—applications to hydrocarbon exploration. Am. Assoc. Pet. Geol. Mem. 26, 83-97. VERDIER, J. P., 1975: Les kystes de dinoflagelles de la section de Wissant et leur distribution stratigraphique au Cretace Moyen. Rev. Micropaleontol. 17, 191-7. WILLIAMS,

B., &

SARJEANT,

S.,

1967:

Organic-walled

microfossils as depth and shoreline indicators. Marine Geol. 5, 389-412.

RAMSDEN, A. R., 1980: Geochemistry, mineralogy & petrology of the type section of the Toolebuc Formation in BMR Boulia 3A stratigraphic drillhole. Aust. CSIRO, Inst. Earth Resour. Restr. Invest. Rep. 1197R (unpubi).

W O P F N E R , H . , F R E Y T A G , I. B. & H E A T H , G . R . , 1 9 7 0 : B a s a l J u r a s s i c -

R A M S D E N , A . R . , D I C K S O N , B. L . & M E A K I N S , R . L . , 1 9 8 2 : O r i g i n

ZI.OTKOWSKI, T. P., 1983: Description of cuttings from the exploratory drillhole WRC 30961 at Birrigoolpa Public Water Place. N.S.W. Geol. Surv. Pet. Rep. 1983/10 (unpubi).

and significance of the Toolebuc gamma-ray anomaly in parts of the Eromanga Basin. Geol. Soc. Aust. J. 29, 285-96.

Cretaceous rocks of western Great Artesian Basin, South Australia: stratigraphy and environment. Am. Assoc. Pet. Geol. Bull. 54, 383-416.

Geological Society of Australia Special Publication No. 12155-162

Basement structure and velocities under the central Eromanga Basin from seismic refraction studies J. Lock, C. D. N. Collins & D. M. Finlayson Bureau of Mineral Resources, Geology & Geophysics, P.O. Box 378, Canberra City, A.C.T. 2601.

ABSTRACT Seismic refraction recordings along a line extending from Mt Howitt 1 to Eromanga, over Eromanga and Cooper Basin sediments, revealed two intra-basement refractors not recorded by co-incident reflection profiling. These refractors were recorded at 3.5 and 5.0 km depth between the basal unconformity at 2.4 km depth and a low velocity zone at about 8.5 km depth. Faulting within the basin sediments extends into basement to at least 5.5 km, with increasing displacement with depth, causing lateral variation of structure and velocity. P-wave velocities increase gradationally with depth and average values along the traverse are 2.2 km s - 1 at the surface to 5.0 km s _ 1 at the basal unconformity; below that, the velocity increases to 6.0 km s _ 1 above a low velocity zone. A sequence of up to five multiples can be observed from 10 km beyond the shot, to at least 40 km and possibly 50 km. Travel-time characteristics indicate that they are waves refracted within basement and multiply reflected at the surface. Travel-time modelling of the first arrivals, multiple arrivals, and two way reflection times provide a tight constraint on the Velocity and structure to a depth of about 5.5 km.

INTRODUCTION In 1980, the Bureau of Mineral Resources, Geology and Geophysics (BMR) conducted seismic refraction surveys in the central Eromanga Basin as part of a large scale multidisciplinary study to investigate structure, stratigraphy, geological evolution and petroleum potential of the area (Pinchin & Senior, 1982). The program for the central Eromanga Basin Project, its objectives, and previous geological and geophysical investigations are outlined by Harrison et al. (1980). The planned geophysical program and its objectives are discussed in more detail by Moss (1980) and Pinchin (1980). One of the objectives of the seismic refraction program was to determine seismic velocities within the Eromanga Basin, underlying infra-basins and basement, which, combined with the vertical reflection data, could be used to interpret geological structure and rock type. There are few vertical reflections recorded from within the basement down to 8 seconds two-way vertical reflection time. The refraction program was designed to define the velocity structure to any sub-basement horizons or within any hidden basins. This detailed upper crustal velocity information would also be used to interpret vertical reflection and refraction data from the deep crust and upper mantle and to determine travel-time residuals. Figure 1 shows the location of the Eromanga Basin and infra-basins, outlines the study area and indicates the location of the seismic lines with respect to the concealed Adavale and Cooper Basins. Two reversed end-to-end 37.5 km traverses were recorded from 200 kg shots fired at Mt Howitt 1, Little Wonder Mine, and near Eromanga Township (Eromanga) (Fig 2). These two traverses were combined to give one

200 km

/V'

EROMANG^l^

ADAVALE BASIN

L.L.1

SOUTH AUSTRALIA ?

V/ c o o p e r * ^ - - - * ^

BASIN

ARROWIE BASIN

ARCKARINGA BASIN

PEDIRKA BASIN

SIMPSON DESERT BASIN

GEORGINA BASIN

GALILEE BASIN

BASiry ^QUEENSLAND NEW SOUTH WALES™

Fig. 1. Location diagram for E r o m a n g a Basin and c o n c e a l e d infra-basins. Study area s h o w n in detail in Fig. 2 is o u t l i n e d . S e i s m i c traverse a n d refraction shot point l o c a t i o n s are s h o w n .

156

J. LOCK, C. D. N. COLLINS & D. M. FINLAYSON sediments are generally flat-lying but are gently structured by broad low-amplitude folds. These folds were produced by movement on three vertical basement block-faults which has resulted in a combination of faulting and draping of the overlying sediments with differential compaction (Ingram, 1970). The most westerly fault is associated with the Mt Howitt Anticline and the most easterly with the Harkaway Anticline (Fig. 2). The Coonavalla Syncline lies between these anticlines. The adjoining geological provinces and tectonic setting are described in more detail by Finlayson et al. (1984).

MOUNT^TIT

HO WITT ANTICLINE —

FIRST-ARRIVAL DATA

KAWAY ANTICLINE Thargomindah Shelf SVNCLINE

100 km

Adavale

Basin

Concealed margin the Cooper Basin

Cooper

Basin

Concealed margin of the Adavale Basin

•

Shotpoint

- Seismic

•

Township

Fault

traverse

Fig. 2. Seismic traverse and shot point locations, central Eromanga Basin.

reversed 75 km line by firing 400 kg shots at Mt Howitt 1 and Eromanga. Recordings were made at 21 stations deployed at 1.875 km intervals between adjacent shot locations. Details of survey design and field operations have been described by Lock (1983). Co-incident 6-fold common depth point (CDP) seismic reflection data tied to Mt Howitt 1 in the west, were also recorded. In this paper these data are interpreted in terms of basin and basement velocity structure to a depth of 8.5 km.

Figure 3a shows the seismic refraction record section from Mt Howitt 1 to Eromanga and Figure 4a shows the reverse section. No strong wide-angle reflection branches are observed at short distances; therefore the velocity of the Eromanga and Cooper Basin sequences is interpreted as increasing continuously with depth. Arrivals between about 8 and 18 km distance have a lower apparent velocity than those recorded at greater distances. Data recorded on a reversed overlapping traverse extending east and west show that the clear impulsive first arrivals out to 75 km shown in Figures 3a and 4a extend out to about 100 km. Beyond this, the waveform changes and the first arrivals become emergent (Finlayson et al., 1984). Because the Permian-Cretaceous strata are generally flat lying and the structure is relatively simple (Wake-Dyster & Pinchin, 1981), preliminary models were developed from the first-arrival data assuming flat layering. Table 1 lists, and Figure 5a shows, the model derived from shots at Mt Howitt 1, Little Wonder Mine and Eromanga. In each model the velocity increases with depth and there are no first-order velocity discontinuities within either the basin strata or Ordovician basement. At Mt Howitt 1 (Table la, Figure 5a) the surface velocity of 2.3 km s increases through the sedimentary sequence to 3.2 km s at 800 m, 3.6 km s at 1.2 km and 5.0 km s at the 2.4 km deep basement unconformity. Between 2.4 km and 3.5 km depth the velocity increases relatively rapidly from 5.0 to 5.6 km s . Between 3.5 and 5.0 km depth the velocity increases from 5.6 to 5.9 km s . As the velocity gradient below 3.5 km decreases; the zone between the basement unconformity at 2.4 km and 3.5 km depth may, perhaps, represent the vertical extent of the Ordovician shales and metasediments. The amplitude decay of first arrivals beyond 100 km distance, noted by Finlayson et al. (1984), can be attributed to the presence of a low velocity zone at 8.5 km depth. This was demonstrated using a reflectivity-method synthetic seismogram program (Fuchs, 1968 (transl. Collins, 1979)), assuming a planar-layered model. The velocity below 8.5 km is not constrained by these data, but has been modelled as 5.6 km s . Comparison of the velocity/depth profiles at each shot point shows that while the velocity of the refractors varies little across the traverse, depths vary up to 500 m. The travel-times calculated using these velocity/depth models where flat layering is assumed are in good agreement with the observed times. In order to precisely model the travel-times, structure within the basins and particularly the basement must be taken into account. Refractors in the Eromanga Basin correlate with the Toolebuc Formation and the Wyandra Sandstone Member at the top of the Cadna-owie Formation. Another refractor is the unconformity at the base of the Cooper Basin in Mt Howitt 1, at the western end of the line. These horizons can be identified as strong reflectors on the vertical reflection section (Wake-Dyster & Pinchin, 1981) and are reflectors 2, _ 1

_ 1

- 1

GEOLOGY AND STRUCTURE

In the survey area the Early Jurassic to Late Cretaceous Eromanga Basin sequence is concealed by Tertiary and Quaternary sediments. The Permian to Triassic northeastern Cooper Basin unconformably underlies the Eromanga Basin. The Cooper Basin is thicker at the western end of the seismic traverse where it is drilled in Mt Howitt 1. It thins eastward and wedges out east of Eromanga and west of the Canaway Ridge. The detailed geology and structure of the basins have been inferred from well completion reports and previous seismic data. The combined average thickness of the Eromanga and Cooper Basins is 2.4 km. Eromanga Basin stratigraphy and structure are discussed in detail by Senior et al. (1978) and that of the Cooper Basin by Senior (1975) and Battersby (1976). Mt Howitt 1 penetrated 1500 m of Eromanga Basin sediments and 900 m of Cooper Basin sediments above the basin/basement unconformity at 2,400 m depth and bottomed in basement of tightly folded Ordovician shales and metasediments of the Thomson Fold Belt (Murray & Kirkegaard, 1978; Rumph, 1978). The vertical reflection section recorded along the refraction traverse (Wake-Dyster & Pinchin, 1981) shows that the

- 1

(a)

(b) I 4

Mt H o w i t t 1

D i s t a n c e (D) in

Eromanga

Fig. 3. (a) Seismic record section for shots at Mt Howitt l recorded eastwards to Eromanga. Trace amplitudes adjusted to constant gain, digitally band-pass filtered (2-15 Hz); adjusted proportional to distance 1 made to account for geometrical spreading, (b) Comparison of first-arrival data (X) and travel-time curve derived f r o m preferred velocity/depth model shown in Figure 5.

(a)

5 o

or r

(b) |4 O > CO 0

CO 8p

a -I

"O <D O

z r 25

_ x - x - x - x - x - x - x - x - x - x « x _ x _ x — x _ x _

. x , x _ x - x - x - x - x - x - x - x - x - x - x - * - x - x - x - x - X '

d> DC

- x - x - x —X"

/ k

I Eromanga

Distance

(D) in

26/G54/20

Mt Howitt 1

Fig. 4. (a) Seismic record section for shots at E r o m a n g a recorded westwards to Mt Howitt 1. Trace amplitudes adjusted to constant gain, digitally band-pass filtered (2-15 Hz); adjustment proportional to distance ~ 1 made to account for geometrical spreading, (b) Comparison of first-arrival data (X) and travel-time curve derived from preferred velocity/depth model shown in Figure 5.

SEISMIC REFRACTION, CENTRAL EROMANGA BASIN 3, and 7 of Pinchin & Senior (1982). The lithological units which give rise to vertical reflections must be thin, as the abrupt increases in velocity seen at these horizons in the Mt Howitt sonic log (Wake-Dyster & Pinchin, 1981) would otherwise produce wide-angle reflections which are not observed in the refraction data. For the purposes of developing a detailed velocity depth model, the basins were regarded as consisting of three refractors separated by the reflecting horizons. Velocity along each refractor was assumed to vary only by a small amount and the refractor to thicken and thin in response to depth changes to the reflecting horizons. The velocities and depths determined at each shot point were used as the basis for a continuous starting model. Ray tracing, using a computer program designed to treat laterally inhomogeneous models (Collins, 1980) was used to calculate travel-times through the model from each shot point. The model was refined, then retested until there was close agreement between

Velocity

~i 30

Wonder

Howitt

Eromanga

Mine

Howitt

Anticline

km/s

I Little

1 — x — r T 40 Distance in

Mt H o w i t t 1

159

the calculated and the observed travel-times. The model was further constrained by ensuring that calculated two-way vertical reflection times were in good agreement with the times shown on the vertical reflection section. In order to precisely model the first arrival times at distances beyond 20 km, it is necessary to invoke basement structure. Extending the faults identified on the seismic reflection section into basement, and increasing the displacement of these faults with depth, accomplishes this. Waves which propagate into the basement and arrive at the surface at distances greater than 30 km, have long path lengths and travel-times within the basement. As a result, arrivals beyond 30 km are quite sensitive to small changes in basement velocity gradient. This accounts for the small variations in velocity along the second basement refractor (Table 1, Fig. 5b). Departures of the observations from the travel-times calculated for a flat-layered model are satisfied by varying the thickness and velocity

Harkaway Coonavalla

Syncline

Distance

Anticline

Little Wonder

Mine

Fig. 5. (a) Velocity/depth profiles at each shot point (see Table 1). interface hatched, location of faults indicated by F.

(b) Schematic diagram of preferred model.

Eromanga

Basin/basement

ON o

10 Mt Howitt 1

40 Distance (D) in km

Eromanga

Fig. 6. Seismic record section for shots at Mt Howitt 1 recorded eastwards to Eromanga. Trace amplitudes adjusted as in Figure 3(a) and digitally band-pass filtered (3 to 7 Hz). Travel-times derived from preferred model for refracted arrivals multiply reflected at the surface are superimposed on observed secondary arrivals. Numbers 0 to 5 indicate number of times the arrival has been reflected at surface.

SEISMIC REFRACTION, CENTRAL EROMANGA BASIN

161

TABLE 1. P wave seismic velocity models for depths less than 10 km for (a) Mt Howitt 1 shooting east, (b) Little Wonder Mine shooting west, (c) Little Wonder Mine shooting east, (d) Eromanga shooting west. (a) (b) (c) (d) Depth Velocity Depth Velocity Depth Velocity Depth Velocity km km s km km s km km s km km s 0.0 2.3 0.0 2.2 0.0 2.2 0.0 2.2 0.8 3.2 1.2 3.0 1.1 3.0 1.1 3.0 1.2 3.6 1.5 3.6 1.4 3.6 1.3 3.4 2.4 5.0 2.5 5.0 2.4 5.0 2.4 5.0 3.5 5.6 2.5 5.6 3.5 5.6 3.5 5.5 5.0 5.9 4.5 5.7 4.5 5.7 5.0 5.8 8.0 6.0 8.0 5.8 8.0 5.8 8.0 5.9 8.5 5.6 8.5 5.6 8.5 5.6 8.5 5.6 _ 1

_ 1

gradient of the layers within the Eromanga and Cooper Basins, and basement, thus enabling the development of a continuous velocity/depth model along the traverse which incorporates structure. Figures 3b and 4b show arrival times read from the original records compared with travel-times calculated from the preferred velocity/depth model.

SECONDARY-ARRIVAL DATA

_ 1

- 1

time curves diverge slowly with time, and energy shifts progressively to larger distances as the number of multiples increases. He suggested that the amplitude along one multiple event depended on velocity in the region through which the ray travelled, local differences in the depth of the weathering layer and changes in the surface reflection coefficient. This probably explains the much lower amplitude of multiple 1 near 36 km (Fig. 5). The conditions necessary for the existence of these multiples are a high velocity gradient in the sedimentary sequences and a relatively high velocity gradient in the zone where the multiples turn (i.e. are refracted). This ensures relatively energetic propagation of the multiples to extended distances (McMechan & Mooney, 1980). A further requirement is a low surface velocity so that the energy returns to the surface at near vertical incidence. All three conditions are met by the preferred model (Fig. 5b) and ray tracing through the preferred model confirmed that the energy is returned to the surface near vertical incidence. Travel-times for rays refracted in the layer below the basins at 2.4 to 3.5 km depth and reflected at the free surface, were calculated using the preferred model. These times are shown on Figure 6 for five multiples. Curve 0 represents the primary wave, and curves 1, 2, 3, 4 and 5 are those multiples which have been reflected at the surface one, two, three, four and five times respectively. These travel-times are in good agreement with the observations indicating that the preferred model is well-constrained to at least 3.5 km depth (the base of the Ordovician?).

A feature of seismic refraction data in this region is a sequence of prominent later arrivals which, in general, consist of a wavelet of the same shape as the primary wave but often of larger amplitude. There are at least three such arrivals on any record section. These later arrivals are observed beyond 10 km to at least 40 km and possibly 50 km. Unambiguous identification of the secondary arrivals is difficult at larger distances. The dominant frequency in the initial and secondary arrivals is 5 Hz. Application of a 3 to 7 Hz band-pass filter removes the higher frequencies and accentuates five secondary arrivals in the record section from Mt Howitt 1 to Eromanga (Fig. 6). King & Falvey (1977) also recognised these multiples in refraction data recorded in the Cooper Basin in the vicinity of Thunda 1 to the northeast of this survey. Such multiples are relatively rare but have also been reported by McMechan & Mooney (1980) in the Imperial Valley, California. Multiples with purely reflected paths are characterised by travel-time curves which are concave upwards (Fig. 7a). The observations have travel-time curves which are concave downwards and are therefore refracted rays which are multiply reflected at the free surface (Fig. 7b). Meissner (1965), in a study of multiple events from flat layers in which the velocity increases with depth, showed five multiples with observable CONCLUSIONS Refractors in the Eromanga Basin correlate with the amplitudes were possible. He noted several features of this type of multiple, which can be seen in Figure 6. Multiple travel- Toolebuc Formation and the Wyandra Sandstone Member of the Cadna-owie Formation. There is also a refractor at the Permian-Ordovician basement unconformity, at the base of the Cooper Basin sequence. There are two refractors at 3.5 la) 12 3 (b) 1 2 3 and 5.0 km, recorded between the unconformity and a low velocity zone at about 8.5 km. The uppermost basement V, < V < v refractor between 2.4 and 3.5 km depth may indicate the vertical extent of the Ordovician shales and metasediments. Faulting within the basin sediments extends downward to at least 5.5. km, with increasing displacement at depth (up to 500 m) causing lateral variation in structure and velocity. The angle of faulting cannot be resolved and is assumed to be vertical. When structure is taken into account, velocity models developed assuming flat layering where structure is simple give 26/G54/18 a good fit to the first arrival data, and to the two-way travelFig. 7. Schematic ray paths and travel times for multiply times to identified reflecting horizons. The secondary arrivals reflected waves: (a) multiply reflected arrivals, (b) observed in the refraction record sections are multiples refracted within the basement and multiply reflected at the reflected-refracted arrivals (after Meissner, 1965). 2

162

J. LOCK, C. D. N. COLLINS & D. M. FINLAYSON

surface. The preferred velocity model for the traverse has velocity gradients which return the rays to the surface sufficiently focused, and at near vertical incidence, to give multiples with observable amplitudes for up to five surface reflections. The accordance of the multiple travel-times, firstarrival times and two-way reflection times with the observations, indicates that the model is well-constrained to 5.5 km depth.

ACKNOWLEDGEMENTS We wish to acknowledge the contribution of Chris Rochford, and the BMR seismic crew, to the field program. Dr B. J. Drummond critically reviewed the manuscript and his comments are gratefully acknowledged. This paper is published with the permission of the Director, Bureau of Mineral Resources, Geology and Geophysics.

REFERENCES

BATTERSBY, D. G., 1976: Cooper Basin gas and oil fields; in Leslie,

R. B., Evans, H. J. & Knight, C. L. (eds) Economic geology of Australia and Papua New Guinea—3. Petroleum. Australas. Inst. Min. Metall. Monogr. 7, 321-69. COLLINS C . D. N., 1979: Adaptation of the synthetic seismogram program "REFLEX" to the CSIRO CYBER 76 computer. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1979/7 (unpubl.). COLLINS, C . D. N., 1980: The crustal structure of the central Bowen Basin from deep seismic sounding. M.Sc. Thesis, Univ. Queensland (unpubl.). FINLAYSON, D. M . , COLLINS, C. D. N . & LOCK, J., 1984: P-wave velocity features of the lithosphere under the Eromanga Basin, eastern Australia, including a prominent mid-crustal (Conrad?) discontinuity. Tectonophysics, 101, 2 6 7 - 9 1 . FUCHS, K., 1968: Das Reflexions- und Transmissionsvermogen eines geschichteten Mediums mit beliebiger Tiefen-Verteilung der elastischen Moduln und der Dichte fur schragen Eingall ebener Wellen. Zeitschr. Geophys. 34, 389-413. HARRISON, P. L., MATHUR, S. P., Moss, F. J., PINCHIN, J. & SENIOR, B. R., 1980: Central Eromanga Basin Program proposals, 1980-1982. Aust. Bur. Miner. Resour. Geol. Geophys., Rec., 1980/32 (unpubl.). HAWKINS, L. V., 1961: The reciprocal method of routine shallow seismic refraction investigations. Geophysics, 26, 806-19. INGRAM, J. A., 1970: Eromanga, Qld, 1:250 000 geological map series. Bur. Miner. Resour. Geol. Geophys., explanatory notes SG/54-12. KING, D. & FALVEY, M . , 1977: A seismic survey in the Cooper Basin in Queensland. APEA J., 17(1), 78-84.

LOCK, J., 1983: Central Eromanga Basin seismic refraction surveys,

1980, 1981, Operations Report. Aust., Bur. Miner. Resour., Geol.

Geophys., Rec., 8 3 / 2 9 (unpubl.). MCMECHAN, G. A. & MOONEY, W. D., 1980: Asymptotic ray theory

and synthetic seismograms for laterally varying structures: theory and application to the Imperial Valley, California. Seism. Soc. Am., Bull., 70, 2021-35. MEISSNER, R., 1965: Multiple events in refraction shooting. Geophys. Prospect., 13, 617-58. MOSS, F. J. (coordinator), 1980: Central Eromanga Basin Project, progress report, January-June 1980. Aust., Bur. Miner. Resour., Geol. Geophys., Rec., 1980/60 (unpubl.). MURRAY, C. G. & KIRKEGAARD, A. G., 1978: The Thomson Orogen of the Tasman Orogenic Zone. Tectonophysics, 48, 299-325. PINCHIN, J., 1980: Central Eromanga Basin seismic survey, Qld, 1980—preview report. Aust., Bur. Miner. Resour., Geol. Geophys., Rec., 1 9 8 0 / 7 7 (unpubl.). PINCHIN, J. & SENIOR, B. R., 1982: The Warrabin Trough, western Adavale Basin, Queensland. Geol. Soc. Aust. J., 29, 413-44. RUMPH, B., 1978: Regional gravity and magnetic data of the central Eromanga Basin area: implications in crustal structure, regional geology and tectonic history. M.Sc. Thesis, Univ. Sydney {unpubl.). SENIOR, B. R., 1975: Notes on the Cooper Basin in Queensland. Qld Gov. Min. J., 76, 260-5. SENIOR, B. R., M O N D , A. & HARRISON, P. L., 1978: Geology of the Eromanga Basin. Aust., Bur. Miner. Resour., Geol. Geophys., Bull., 167. WAKE-DYSTER, K. D. & PINCHIN, j., 1981: Central Eromanga Basin seismic survey Queensland 1980. Aust., Bur. Miner. Resour., Geol. Geophys., Rec., 1982/22 (unpubl.).

Geological Society of Australia Special Publication No. 12, 163-173

The Canaway Fault and its effect on the Eromanga Basin J. Pinchin & Y. Anfiloff 1

'Flower Doery Buchan Pty Ltd, 77 Pacific Hwy, North Sydney, N.S.W. 2060. Bureau of Mineral Resources, P.O. Box 378, Canberra City, A.C.T. 2601.

ABSTRACT

The Canaway Fault is a 250 km long north-trending near-vertical fault within the Eromanga Basin, Queensland. It forms the eastern margin of the Canaway Ridge which separates the Cooper Basin from the Galilee Basin. Recent seismic and gravity data show that the fault is in places associated with granitic intrusions, and that a palaeo-high below the Canaway Ridge had some effect on early deposition within the Adavale Basin. Seismic data also show that movement of the fault began in the mid-Carboniferous, followed by almost continuous movement from the Permian to the late Middle Tertiary. The Canaway Fault is.unlikely to have acted as a total barrier to migration of hydrocarbons from depo-centres of the Eromanga Basin because fault displacement of Jurassic reservoir rocks was only 47 m in Late Cretaceous times, and the monoclinal appearance on seismic sections suggest no complete break in these rocks. However, anticlines adjacent to the fault have not been fully explored and could present prospective exploration targets.

INTRODUCTION

The Canaway Fault is a major structural feature in western Queensland (Fig. 1). It forms the eastern margin of the Canaway Ridge which separates the Cooper Basin from the Galilee Basin (Senior, 1975) and was previously thought to form the western boundary of the Adavale Basin. However, there are remnants of the Adavale Basin to the west of the Canaway Fault within the Warrabin and Barcoo Troughs (Pinchin & Senior, 1982). A combined seismic and gravity survey by the Bureau of Mineral Resources (BMR) in 1980 (Wake-Dyster & Pinchin, 1981) recorded several regional traverses across the central Eromanga Basin, two of which crossed the Canaway Fault. Detailed seismic and gravity results on these two traverses have enabled a new interpretation of the structure and timing of the Canaway Fault.

GEOLOGY Stratigraphy

The sedimentary sequence in western Queensland is composed of four overlapping sedimentary basins: the Adavale Basin of Devonian age, with its associated Cooladdi, Quilpie, Warrabin and Barcoo Troughs; the Cooper and Galilee Basins of Permian to Triassic age, and the Eromanga Basin of Jurassic to Cretaceous age (Fig. 1). Table 1 shows a simplified stratigraphy of these basins. Basement rocks in the area are poorly known. Along the Canaway Ridge, Budgerygar 1 encountered granodiorite, Yongala 1—schist, and Canaway 1—phyllite (see Fig. 2 for well locations). Most wells away from the ridge reached total depth in sedimentary rocks of Devonian age. The only wells near the Canaway Fault to penetrate Devonian rocks of the Adavale Basin were Yongala 1 and 2. The deeper Yongala 1 well penetrated 640 m of Buckabie Formation, 310 m of Etonvale Formation, 16 m of Cooladdi

Dolomite, and 26 m of Gumbardo Formation (Table 1). The Lissoy Sandstone, Log Creek Formation and Eastwood beds are not present here. The stratigraphy of the Permo-Triassic sedimentary sequence near the Canaway Ridge cannot be correlated exactly with the sequence within the centre of the Cooper and Galilee Basins. Permian sedimentary rocks on the western side of the 1I GALILEE / \ i r BARCOO SHEL£/ TROUGH S '

ADAVALE

BASIN

« I'QUILPIE 'TROUGH

BASIN

WARRABIN i ( — \ vV) TROUGH "IT A \ I ) VJ I - "/Area of Figu/e2\ ) iA-"' A

COOLADDI TROUGH

THARGOMINDAH SHELF

21/0/1 Boundary of Cooper and Galilee Basins Boundary of Adavale Basin. Quilpie and Cooladdi Troughs Boundary of Warrabin Trough as described in this paper

Fig. 1. Location map and geological setting.

J. P I N C H I N & V. A N F I L O F F

Fault, tick on downthrown side Where location of fault is approximate, line is broken; where inferred, queried; where concealed, shown by short dashes Anticline,

position

approximate

Syncline,

position

approximate

BMR 1980 seismic traverse Bouguer

anomaly

and

gravity

contour

Petroleum exploration dry, abandoned Petroleum exploration show of oil and gas,

(fxm.s

well, well with abandoned

Gas well

Budgerygar 1

East Windorah 1

Traversa 8 / -

A.Yongala 1

Thunda 1

jYongala 2

.^Canaway 1

21/0/2 Fig. 2. Structural trends, well locations, and regional gravity contours.

CANAWAY FAULT TABLL

1. Simplified stratigraphic nomenclature for Adavale, Cooper and Eromanga Basins. Nomenclature after Senior et al. (1978), and Paten (1977).

AGE

BASIN

GROUP

Quaternary Tertiary

STRATIGRAPHIC UNIT (sand, gravel, alluvium Eyre Fm e t c .

Cretaceous

Rolling Downs Group

Winton Fm Mackunda Fm Allaru Mudstone Toolebuc Fm Wallumbill, a Fm Cadna-owie Fm

c 03

Jurassic

Triassic Permian

w (3 ^ 8. o

Devonian

.5 (O$ >

1 Ordovician

Wyandra Sst Mbr

Hooray Sst Injune Creek Group

Westbourne Fm Adori Sst Birkhead Fm Hutton Sst Evergreen 1Fm unnamed pre Precipice Sst Hutton (Cooper Syncline area)

Nappame rri Fm

Moolayember Fm

•-I Clematis Sst to

Gidgealpa (unnamed coal 01 (upper Permian .2 absent?) Group facies) Merrirae1ia Fm

o Colinlea Sst

Buckabie Fm Etonvale Fm Cooladdi Dolom ite Lissoy Sst Log Creek 'Fm Eastwood Beds Gumbardo Fm Basement - granite, p h y l l i t e , schist e t c

ridge are a condensed sequence of coal measures of equivalent age to the Gidgealpa Group and Merrimelia Formation in the Cooper Basin (Senior, 1975); while on the east side of the ridge they are probably equivalent to the Colinlea Sandstone. Similarly, Triassic rocks are equivalent to the Nappamerri Formation on the west side of the ridge, and to Moolayember Formation plus Clematis Sandstone on the east side. The Eromanga Basin is continuous across the Canaway Fault, and although several conflicting nomenclatures have been used in the past, the one used here is from Senior et al. (1978).

Structure

165

GEOPHYSICS Seismic data

The network of seismic lines across the Canaway Fault is shown in Figure 3. Most of the seismic data prior to 1980 were single coverage. Two pre-1980 multiple-coverage lines, Welford 645 and 648, were reprocessed by the BMR to improve data quality, and some of the single coverage lines are also being processed. Seismic sections for BMR Traverses 6 and 8, which cross the Canaway Fault nearly at right-angles, reduced from WakeDyster & Pinchin (1981), are shown in Figures 4 and 5. Interpretation of most of the seismic reflectors within the Eromanga Basin and Cooper-Galilee Basin sequences is relatively easy, but since only a few exploration wells penetrate the Adavale Basin, identification of the deeper reflectors in that sequence is more difficult. Both seismic sections show that the Permian coal measures lap onto the Canaway Ridge. On Traverse 6 the Triassic Nappamerri Formation also laps onto the ridge, whereas on Traverse 8 the formation is continuous across the ridge and across the Canaway Fault. The locations of the depositional edges of both the Permian and Triassic rocks where they abut against the east side of the faulted Canaway Ridge are shown in Figure 3. Both sections also show Adavale Basin sedimentary rocks immediately east of the Canaway Fault, but none immediately west thereof. However, on Traverse 6 about 30 km west of the Canaway Fault, Devonian sedimentary rocks again appear to be present, dipping westwards into the Barcoo Trough (Fig. 4). The structure across the Canaway Fault is typical of many faults within the Eromanga and Cooper Basins. Within the shallowest part of the section the structure appears to be monoclinal with seismic reflectors continuing across it. This could be due to the fault movement being taken-up by slippage within lithic and feldspathic sandstones, siltstones and mudstones which contain abundant montmorillonite (B. R. Senior, pers. comm., 1981), and has the effect of broadening the fault zone towards the surface. Measurements of seismic reflection times and velocities show that the thickness of each unit increases slightly on the eastern side of the fault, indicating continuous fault movement during sediment deposition. Canaway Fault displacement is largest on Welford 645, near Budgerygar 1 well. This is shown by the structure contours on the top of the Wyandra Sandstone Member of the Cadnaowie Formation (Fig. 6) which were originally plotted from seismic data in two-way time (Everest, 1966; Hickey, 1971). The impedance contrast between the Wyandra Sandstone Member at the top of the Cadna-owie Formation and the overlying Wallumbilla Formation produces a strong seismic reflection that is easy to map. Here, the values have been converted to depth and to sea-level datum using an average velocity of 2350 m sec. as determined from seismic velocity analyses, and using a datum velocity of 2000 m s e c . T h e contours show the elongate Canaway Ridge west of the Canaway Fault, and the structure rising up to the fault on the east side to create the curved, almost monoclinal appearance.

The dominant structural trend is northerly (Fig. 2), but on the west side of the Canaway Fault there are some northwesttrending faults, and on the east side there are northeasttrending folds. The structure of the Eromanga, Cooper and Galilee Basins is characterised by broad, generally lowamplitude folds, and linear faults with throws of up to 300 m. Within the Adavale Basin there are large normal faults and 'deep-seated thrusts' (Auchincloss, 1976), and it is also possible Previous gravity information that large shear movements took place (Paten, 1977). Most In the region of the Canaway Fault, the quality of existing of this tectonism occurred during the Mid-Carboniferous gravity coverage varies considerably. In the WINDORAH Kanimblan Orogeny (Kirkegaard, 1974). 1:250 000 map sheet area, the Canaway Fault is revealed by -1

J. PINCHIN & V. ANFILOFF 143°45'

144°00' 25°00'

a network of detailed traverses, but in adjoining areas coverage is poor. There are areas where the absence of one or two stations in the 11 km network (Gibb, 1967) leaves holes in the coverage of between 20-40 km, and about one-third of the EROMANGA map sheet area has no coverage. In WINDORAH, a broad north-trending gravity ridge separates the Barcoo Trough from the main Adavale Basin to the east (Fig. 7). The eastern side of the gravity ridge is steep, and coincides with the Canaway Fault. The gravity expression of the fault is confined mainly to the WINDORAH sheet.

Detailed gravity information—BMR, J980

The detailed gravity data along Traverses 6 and 8 (Figs 4, 5) reveal gravity highs of 5-15 mGal amplitude associated with the Canaway Fault. As the gravity highs do not have a characteristic fault signature, the presence of other bodies close to the fault is suggested. On Traverse 6, there is a broad gravity low separating two highs, inferring a negative mass contrast caused by either a deep trough or a granitic body within basement. A less prominent gravity low in a corresponding position on Traverse 8 indicates that a granitic body or a trough could also occur here. It is conceivable that a line of granitic intrusions, or one elongate intrusion, lies parallel to the Canaway Fault.

;

Budgerygar 1 \

Gravity modelling

Gravity models based on information interpreted from the seismic sections along Traverses 6 and 8 (Figs 8, 9) did not reproduce the gravity features associated with the Canaway Fault, and it was necessary to introduce a low-density body into the basement about 5 km west of the fault. The lowdensity body is probably the weathered granodiorite intersected at the bottom of Budgerygar 1 which is situated between Traverses 6 & 8. To produce the required anomaly, the granodiorite would have to have a density 0.15 gm c m less than the basement density, and be located as high as possible in the section. It could therefore have been exposed, in the ancient topography at the time when Devonian sediments were being deposited. East of the Canaway Fault, the gravity data suggest a thick Devonian section underneath the lowest seismic reflector. This section cannot be evaluated accurately, as the fault anomaly is too near the end of the traverse, but probable thickness of Devonian sediments would be 1-2 km.

I Yongala 1 Yongala 2

-3

STRUCTURAL TIMING

Petroleum exploration dry, abandoned

m M 26°oo'

well—

Seismic traverse prior to 1980 BMR 1980 seismic and gravity traverse Seismic traverse reprocessed by No Triassic

BMR

Permian

Fig. 3. Location of seismic traverses.

The sequence of geological events in the vicinity of the Canaway Fault is shown in Figure 10. After the large fault movement during the Carboniferous, the elevated west side of the fault was gradually eroded, to leave only a gentle structural high in the Permian, when deposition in the region recommenced. Permian and Triassic strata gradually lapped onto and over the Canaway Ridge, further reducing the elevation change across the fault. During Jurassic and Early Cretaceous deposition in the Eromanga Basin there was continuous but small movement of the Canaway Fault so that by the Albian, when the Toolebuc Formation was deposited, vertical displacement at the level of the Hutton Sandstone and at the base of the Eromanga sequence was only 47 m (average of measurements made on seismic sections for BMR Traverses 6, 8, and Welford Line 645). By the late Mid-Tertiary a further 324 m displacement had occurred to produce an average of about 370 m at the base of the Jurassic. Seismic data provide evidence for early fault movement, but the latest stage of movement is determined by effects of

0 > z

c 5

1 WINTON COAL 7 TOP DEVONIAN

2 TOOLEBUC 8 COOLADDI

3 CADNA-OWIE 4 HUTTON 5 NAPPAMERRI 6 BASE PERMIAN 9 BASEMENT 10 COOLADDI ? (REFLECTORS M A R K STRATIGRAPHIC TOPS)

F i g . 4. S e i s m i c s e c t i o n a n d g r a v i t y p r o f i l e a l o n g Traverse 6.

10km _J

-500 E 1-550

I -600

WEST

EAST

o X

z fcp < >

•n

5 "A •n

1 WINTON COAL 2 TOOLEBUC 3 CADNA-OWIE 4 HUTTON 5 NAPPAMERRI 7 COOLADDI 8 LOG CREEK? (REFLECTORS M A R K STRATIGRAPHIC TOPS)

Fig. 5. Seismic section and gravity profile along Traverse 8.

6 TOP DEVONIAN

10km

CANAWAY FAULT 143°45'

144°00'

EFFECTS ON PETROLEUM MIGRATION

-<J>- Petroleum exploration well — dry. abandoned Datum: mean sea level t

Fig. 6. Structure contours, Wyandra Sandstone Member of Cadnaowie Formation. Contour interval 50 m.

169

the fault on surface geology. During the Mid-Tertiary, two episodes of severe weathering, at 60 Ma and 30 Ma ago, produced deeply weathered and duricrusted profiles (Idnurm & Senior, 1978). These deep-weathered layers are eroded from the west side of the Canaway Fault, resulting in topographic inversion and producing a fault scarp on the downthrown side to the east. Hence the latest fault movement post-dates the last severe weathering episode. From the onlap of Devonian sedimentary rocks onto basement on the west side of the Canaway Ridge it can be deduced that there was a palaeo-high here during MidDevonian time, although sediments of the Adavale Basin entirely covered the Canaway Ridge by the Late Devonian. The uplift in the Mid-Carboniferous seems to be asymmetrical, tilting the block west of the fault downwards to the west. It is possible that this westward tilt directed most of the erosion products from the Canaway Ridge towards the west, to be deposited in the Cooper Basin. Subsequent downwarping of the Cooper, Galilee and Eromanga Basins was accompanied by gradual fault displacements, which ceased in the late Mid-Tertiary. The timing and mechanism of petroleum migration within the Eromanga Basin is still being debated. Bowering (1982) considered that primary migration occurred during the Late Cretaceous to Early Tertiary in an essentially static groundwater regime, and that present artesian water flow is of insufficent strength to flush hydrocarbons from existing traps. He shows the areas of petroleum generation closest to the Canaway Ridge to be the Cooper and Thomson Synclines (overlying the Barcoo Trough, Fig. 1); hence most hydrocarbons would reach the Canaway Ridge from the west. Pitt (1982, this volume) studied the geothermal gradients in the Eromanga Basin and concluded that oil generation from the Permian sequence began during the Mid-Cretaceous and commenced from the Murta Member in the Early Miocene. He also stated that hydrodynamic flow in the Great Artesian Basin could be an important factor in the migration and trapping of petroleum. Senior & Habermehl (1980) suggested that westward artesian water movement had a major influence on petroleum migration and that likely petroleum accumulations would occur where structural or stratigraphic barriers blocked this westward flushing. The Canaway Fault, at right-angles to the flow of artesian water, could represent such a barrier. On the other hand, Moriarty & Williams (1982) conclude that a dip closure of 0.1° is sufficient to prevent flushing of oil from a structural trap by artesian water flow. The Canaway Ridge has closure well in excess of this figure and should not be flushed. It therefore seems unlikely that all hydrocarbons have been flushed from structures adjacent to the Canaway Fault, either because of insufficient hydraulic gradient or because the fault has blocked flow within the aquifers causing stagnation zones. The main factor controlling prospectivity of the area may be the timing of the structure with respect to hydrocarbon generation and primary migration. Although fault displacement during the Carboniferous was large, much of the elevation change across the fault was subsequently reduced by erosion, and by the Late Cretaceous, displacement at the Hutton Sandstone level was only 47 m. However, even this would have been enough to trap some petroleum, and since the fault grew continually throughout the Cretaceous and

170

J. P I N C H I N & V. A N F I L O F F

Fig. 7. Gravity profiles and contours over the Canaway Fault.

WEST

EAST

Basement 2.7 Fig. 8. Gravity model along Traverse 6 (density units g cm

0 1

10km I

CANAWAY FAULT WEST

171 EAST

-250

'Computed gravity anomaly

Bouguer gravity

°T

Sea level

2.1

Fig. 9. Gravity model along Traverse 8 (density units g cm ). 3

Tertiary, if most petroleum migration occurred during the section on the eastern side and a narrow granitic body in the Early to Middle Tertiary the potential traps would be larger. upper part of the basement on the western side. The gravity As previously mentioned, the seismic appearance of the feature occurs on other traverses (Fig. 7) demonstrating that fault within the Upper Cretaceous and Tertiary sections is that the gravimetric technique is effective in locating the fault. of a steep monocline rather than an abrupt break, and the The Canaway Fault has its strongest expression between sandstone reservoirs may not have been blocked sufficiently to completely prevent the migration of petroleum across the latitudes 25 °S and 26 °S (Fig. 3), but this is possibly because structure. However, the anticline on the west side of the fault of the amount of seismic and detailed gravity data here. Both could still be prospective, although probably not full to spill- geological mapping and regional gravity data suggest that the point. If this is the case, more exploration is warranted since fault extends for a total length of at least 250 km (Fig. 2). Budgerygar 1 and Canaway 1 were not drilled at the highest There is also no evidence for reversal of fault movement at points of the anticline (see Fig. 6) even on the present 5 km any time and it seems that the most likely cause of the fault seismic grid. Furthermore the onlap edges of the Triassic and is continuous passive vertical movements within the basement. Permian sequences (Fig. 3) could present stratigraphic traps The petroleum potential of this area has not yet been fully against the flanks of the Canaway Ridge. assessed and, considering the present controversy over the CONCLUSIONS migration of hydrocarbons in the Eromanga Basin, there is Between latitudes 25 °S and 26 °S, the Canaway Fault is no reason yet to downgrade the prospectivity of the area denoted by a gravity high, resulting from a deep sedimentary around the Canaway Fault.

J. P I N C H I N & V. A N F I L O F F

••.••Eroded •

Fault scarp /

PRESENT Further vertical fault movement of about 180m; followed by erosion

coal horizon ^ \ nto n_F orm atio n 1 Hutton Sandstone

/ + + + (+ + + / \ + + +/ EARLY TERTIARY 30 Ma Continued subsidence and deposition; fault movement (of 65m); deep weathering episodes at 60 Ma and 30 Ma Rivers d e p o s i t i n g H u t t o n S a n d s t o n e

EARLY JURASSIC 190

Continued subsidence and deposition; fault movement of 109m High , ground Coal s w a m p s

Coal s w a m p s ^COOPE^

- - —' ^

/ + +\ | +

[^GALILEE BASIN

+ 7

LATE PERMIAN 250 Ma Mid Carboniferous Kanimblan Orogeny; folding, uplift, granite intrusion; vertical fault displacement of at least 400m; followed by erosion of Devonian rocks Rivers d e p o s i t i n g B u c k a b i e F o r m a t i o n (red beds)

LATE DEVONIAN 360 Ma Subsidence and deposition of Devonian sediments Supratidal Cooladdi Dolomite

High ground

MID DEVONIAN 400 Ma Fig. 10. Development of the Canaway Fault Structure (Schematic— not to scale).

CANAWAY FAULT

173

REFERENCES

AUCHINCLOSS, G., 1976: Adavale Basin; in Knight, C. L. (ed.)

Economic Geology of Australia and Papua New Guinea—3. Petroleum. Australas. Inst. Min. Metall. Monogr. 5, 309-16. BOWERING, O. J. W., 1982: Hydrodynamics and hydrocarbon migration—a model for the Eromanga Basin. A PEA J. 22(1), 227-36. EVEREST, J. B., 1966: Petty Geophysical Co.—Panhandle area seismic survey, Qld. Aust. Bur. Miner. Resour. Pet. Search Subs. Acts, Rep. 66/11102 (unpubl.).

MORIARTY, K. C. ? & WILLIAMS, A. F., 1982: Hydrocarbon flushing

in the Eromanga Basin—fact or fallacy; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 311-28. PATEN, R. J., 1977: The Adavale Basin, Queensland; in Petroleum in Queensland, a stocktake for the future. Pet. Explor. Soc. Aust. Qld Branch, Symposium, Nov. 1977 (unpubl.). PINCHIN, J., & SENIOR, B. R., 1982: The Warrabin Trough, western Adavale Basin. Geol. Soc. Aust. J. 29, 413-24. PITT, G. M., 1982: Geothermal gradients in the Eromanga-Cooper Basin region; in Moore, P. S. & Mount, T. J. (compilers) GIBB, R. A., 1967: Western Queensland reconnaissance gravity Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. survey, 1957#61. Aust. Bur. Miner. Resour. Geol. Geophys. Rep. & Pet. Explor. Soc. Aust., Adelaide, 262-83. 131 (unpubl.). SENIOR, B. R., 1975: Notes on the Cooper Basin in Queensland. Qld Govt Min. J. 76, 260-5. HICKEY, F. L., 1971: Geophysical Associates Pty Ltd—Corrajah Seismic Survey, Qld. Aust. Bur. Miner. Resour. Petrol. Search SENIOR, B. R., & HABERMEHL, M . A., 1980: Structure, Subs. Acts, Rep. 71/1206 (unpubl.). hydrodynamics and hydrocarbon potential of the central Eromanga Basin, Queensland, Australia. B.M.R. J. Aust. Geol. IDNURM, M., & SENIOR, B. R., 1978: Palaeomagnetic ages of Late Geophys, 5(1), 47-56. Cretaceous and Tertiary weathered profiles in the Eromanga SENIOR, B. R., MOND, A., & HARRISON, P. L., 1978: Geology of the Basin, Queensland. Palaeogr. Palaeoclimatol. Palaeoecol. 24, Eromanga Basin. Aust. Bur. Miner. Resour. Geol. Geophys. 263-78. Bull. 167. KIRKEGARD, A. G., 1974: Structural elements of the northern part WAKE-DYSTER, K., & PINCHIN, J., 1981: Central Eromanga Basin of the Tasman Geosyncline; in Denmead, A. K. (ed.) The Tasman seismic survey Queensland 1980: operational report. Aust. Bur. Geosyncline—a symposium. Geol. Soc. Aust. Qld Div. 27-63. Miner. Resour. Geol. Geophys. Rec. 1981/22 (unpubl.).

Geological Society of Australia Special Publication No. 12, 175-182

Stratigraphy and tectonics of the Dalhousie Anticline, southwest Eromanga Basin

G. W. Krieg

South Australian Department of Mines and Energy, 191 Greenhill Road, Parkside, S.A. 5063.

ABSTRACT

Dalhousie Anticline is a roughly oval, regional structure which displays a considerable proportion of the Cretaceous Eromanga Basin strata. Folds and faults exposed in these strata together with the overlying assemblage of late Cainozoic deposits illustrate the tectonic development of the feature. The exposed anticlinal succession is a conformable, Neocomian to Cenomanian sequence of transgressive-regressive sandstone, shale, siltstone and minor carbonate, overlain disconformably by silicified, fluviatile sandstone and thin terrestrial sand/carbonate deposits of mainly Tertiary age. Within the breached anticline a concentrically disposed, Quaternary toposequence of gypsified gravel, lacustrine clay/dolomite and mound spring deposits has formed in response to anticlinal growth and breaching, and major artesian spring development. The shape of Dalhousie Anticline is controlled essentially by a polygonal pattern of fairly straight faults and monoclinal flexures. Anticlinal growth along these linear structures was in progress during Oligocene and early Miocene, but the major uplift did not occur until late Tertiary and Quaternary. Anticlinal growth is still active today.

INTRODUCTION

Although Dalhousie Anticline is only one of a number of breached, regional anticlines around the southwest margin of the Eromanga Basin (Fig. 1), the structure is of special interest because it illustrates more geology of this part of the basin than any other anticline, or area of comparable size in the region. This has resulted from the particular combination of geographic position and relatively large fold amplitude: the anticline is sufficiently far east, or basinward, to include the uppermost major unit, the Winton Formation, of the Eromanga Basin sequence; on the other hand marginal thinning coupled with major uplift has allowed erosion down to the top of the basal sandstones of the Cadna-owie Formation. Thus, the exposed Mesozoic interval extends from Cadna-owie Formation up to Winton Formation with many features of the sequence well displayed in certain areas. Faulting and local folding associated with anticlinal growth are also prominent features of the anticline. In particular a major fault zone through the core of the anticline has tapped the near-surface aquifer and produced a remarkable group of artesian springs and related deposits. These spring-related deposits together with other deposits derived from breaching of the anticline form a group of Cainozoic units which shed some light on the growth of the anticline. It is this variety of geological features displayed so close to each other that makes Dalhousie Anticline a particularly interesting example of Eromanga Basin geology. Hence, a brief description of these Mesozoic strata, and a general summary of anticlinal growth from examination of the exposed structure and related Cainozoic sequence are given below.

GENERAL PHYSIOGRAPHY

The surface expression of Dalhousie Anticline is marked by a discontinuous, oval rim of bold, high scarps composed

essentially of Tertiary silcrete. This rim is about 80 km by 35 km, oriented NNE, and overlooks an undulating terrain of low stony hills with occasional low gypseous scarps, and a few large tree-lined creeks. Offset slightly to the southwest from the geometrical centre of the anticline is an artesian spring complex with hot pools fringed by dense vegetation, mounds of mud and silt, saline flats and dissected dolomite plateaux. The springs complex itself is some 20 km by 8 km, again oriented NNE, and contains more than 60 active springs. Water from the springs usually evaporates or soaks away before escaping from the anticline, but during very wet seasons, run-off including that from the springs escapes via a very broad channel through the eastern rim of the anticline, into the Simpson Desert.

STRATIGRAPHY

The exposed anticlinal succession (Fig. 2) begins with the Cadna-owie Formation (Wopfner et al., 1970). The main lithology is a fine-medium, clean and well sorted, light grey sandstone. Coarser lenses, and small to medium scale current bedding occur in the lower parts and finely sandy siltstone and claystone, in places laminated and carbonaceous, occur in the upper parts. A prominent horizon of large water-worn boulders and cobbles, some with Devonian fish fossils, is diagnostic of the unit. Thin, strongly ferruginous layers are also characteristic and pyrite nodules are fairly common low in the sandstone interval. No fossils (apart from those in boulders) have been found in the Cadna-owie Formation from this area but it is regarded as a non-marine to marginal-marine unit transitional between the underlying fluvial Jurassic sandstone and the overlying marine Cretaceous shale and to be of Neocomian to Early Aptian age. Conformably overlying Cadna-owie Formation are formations of the Marree Subgroup (Thomson, 1980),

G. W. KRIEG

176

of Oodnadatta Formation is chosen at the base of the first khaki sandstone above the dark grey mudstone. An upper boundary to the sandy interval is difficult to map around the anticline because of poor exposure, but the'total sandy interval representing the lower part of Oodnadatta Formation would include both the Coorikiana and Wooldridge Limestone Members* The sandstone beds weather khaki-yellow, are very fine to medium grained, generally show small to medium scale, shallow angle cross-bedding and are very calcareous becoming

SIMPSON DESERT

Quaternary: mainly sand Undifferentiated Cainozoic: sand, alluvium, colluvium Early Tertiary: mainly silcreted s a n d s t o n e _ Mesozoic: mudstone, siltstone, sandstone. Pre Mesozoic deposits Fault Anticline

Drn. G.B.

Fig. 1. Regional geological setting of Dalhousie Anticline.

consisting of marine shale with sandstone interbeds. The lowest unit, Bulldog Shale (Freytag, 1966) is a dark grey, carbonaceous, pyritic and glauconitic shaley mudstone, with large concretions of hard, dark grey, fossiliferous marly limestone. At its base, Bulldog Shale is a dark coffee-brown colour and contains small lenses of fine to coarse, pale grey, friable sand and lenses of light khaki-yellow cone-in-cone limestone. Fossilized tree trunks and pieces of wood occur sporadically, and large water-worn boulders are not uncommon. The Bulldog Shale is Aptian, perhaps extending into Albian (Morgan, 1980), and a shallow marine environment is interpreted for the unit, with the basal portion deposited nearer shore. Bulldog Shale is overlain transitionally by Oodnadatta Formation (Freytag, 1966), a unit similar to Bulldog Shale but siltier and sandier especially in the lower half. The base

0 QUATERNARY

25 KILOMETRES

Undifferentiated sand and alluvium..

TERTIARY

EYRE FORMATION Silcreted quartzose sandstone.

CRETACEOUS-Eromanga Basin Units

WINTON FORMATION Altered clayey sandstone and claystone MARREE SUBGROUP OODNADATTA FORMATION Marine mudstone and siltstone. Interbeds of calcareous sandstone BULLDOG S H A L E Marine mudstone, marl lenses; lenticular sands and fossil wood at base; occasional boulders. C A D N A - O W I E FORMATION Sandstone with ferruginous layers, pyrite nodules. Prominent boulder horizon

Drn. G.B.

SADME 83-100

Fig. 2. Generalized geological map of Dalhousie Anticline showing distribution of Cretaceous Eromanga Basin deposits and early Tertiary Eyre Formation. * The Coorikiana Member of Freytag (1966) is the same lithostratigraphic unit as the Coorikiana Sandstone of Thomson (1980). Recent work (Moore et a!., this volume) indicates that the widespread outcrops can be correlated with a subsurface sandstone of regional extent and that the unit may be redefined formally as a formation, the Coorikiana Sandstone.

DALHOUSIE ANTICLINE 177 sandy limestone in places. Fine, dark green, pelletal glauconite chalcedonic limestone at the top (Cadelga Limestone occurs in some of the clay-rich interbeds. Thin-section equivalent). The age of Mount Willoughby Limestone is not examination reveals a mineralogy with abundant altered lithic known precisely but falls within the range Miocene to Pliofragments and up to 5 per cent fresh feldspar and mica. Pleistocene (Major, 1973; Wopfner, 1974; Firman, 1981). Macrofossils including Inoceramus, ammonites and other Within the breached Dalhousie Anticline three Quaternary molluscan remains have been found at several localities in this units rest unconformably on the folded Mesozoic sequence. sandy interval. The first unit is Alinerta Gravel (Krieg, in prep.), a gypsified, The upper part of Oodnadatta Formation consists of and locally carbonate-cemented bouldery to pebbly gravel with medium to dark grey, shaley siltstone, silty shale and claystone an ill-sorted sand/silt/clay matrix, the second unit, Dalhousie with very fine, variably micaceous sandstone intercalations. Formation (Krieg, in prep.), is a pale grey to cream Near the top, the unit passes up into a chemically altered zone. clay/dolomite sequence, and the third unit comprises unnamed Oodnadatta Formation is of Albian age and a shallow mound spring deposits that consist of clay and silt with a local marine environment with mild tectonic instability is envisaged variant of coarse to conglomeratic sand. The age of these units is only inferred to be Pleistocene to Holocene (Krieg, in prep.) for the unit. The youngest Mesozoic unit mapped in the Dalhousie as fossil evidence is lacking. Anticline is Winton Formation (Whitehouse, 1955; Exon & Senior, 1976) which conformably overlies Oodnadatta STRUCTURE Formation and is overlain disconformably by Early Tertiary The broad structure of Dalhousie Anticline is illustrated deposits. The unit occurs within a zone of profound chemical the *C seismic horizon (Fig. 3), a reflector originating from alteration so that exposures now consist of claystone, sandy by the base of Bulldog Shale (Freytag et al., 1967; Stadter, 1972). claystone and clayey sandstone, variably silicified and The contour plan derives from essentially reconnaissance variously coloured white, pale grey, mauve and mustard yellow. surveys to fit the geology exposed in the springs Fine, parallel and cross-lamination is preserved in the fine locality. modified Thus, this plan shows the structure as a simple, sandstone, and medium to large scale cross-bedding, including cut-and-fill sets with clay pebble layers along scour bases, occurs in the coarse and very coarse sandstone beds. Thin section examination shows an original lithic content including possible volcanic debris, feldspar and mica all now altered to clay. The lower part of Winton Formation as mapped probably includes the Mount Alexander Sandstone Member of Oodnadatta Formation, but the base chosen for the unit seems to be the most practical one for mapping. Thus, the uppermost Mesozoic map unit of Dalhousie Anticline probably correlates with Blanchewater Formation of the Marree area (Forbes, 1966, this volume). No fossil remains have been recovered from Winton Formation in this area, but its age is assumed to be late Albian to Cenomanian. A shallow or marginal marine to fluviatile environment of deposition is envisaged for the unit. Overlying Winton Formation and forming the uppermost widespread unit of the anticlinal sequence is Eyre Formation of Palaeocene-Eocene age (Wopfner et al., 1974). This unit is essentially a pale grey, mature quartz sandstone with a characteristic basal bed of, commonly, highly polished quartz, chert and silicified wood pebbles. It is typically coarse and cross-bedded in the lower part, with common large scour structures, and fine, well-sorted and massive in the upper part. Silicification increases upwards through the unit, being intense in the preserved top which is a very hard, brittle grey or cream silcrete, the Cordillo Silcrete of Wopfner (1978). Being so hard and resistant, it forms the ramparts that define the rim of the breached anticline. Also forming part of the anticlinal sequence is a sandstone/limestone unit occurring in shallow incisions cut into Cordillo Silcrete. This unit, Mount Willoughby Limestone of Nicol (1971) or Doonbara Formation plus Cadelga Limestone of Wopfner (1974), is located sporadically around the anticlinal rim high in the present landscape and also along the lower flanks. Thus, the unit is closely associated structurally with Cordillo Silcrete although disconformably overlying it. The sequence consists of a lower interval of red ferruginous sandstone of notably spherulitic habit and commonly with silcrete clasts at the base (Doonbara Formation equivalent) which passes up through Fig. 3. Structure contour map of the seismic ' C horizon over pink calcareous sandstone to a dense dolomitic and Dalhousie Anticline.

178

G. W. KRIEG

steadily curving surface with only the larger individual folds and faults superimposed on the general anticlinal form. A more reliable impression of structural pattern, however, may be represented by the smaller scale features commonly displayed within the exposed anticlinal section. Folds ranging from a few kilometres across down to a few hundred metres or less, and faults some few metres to tens of metres strike length, can readily be observed. Some faults may be directly related to local folds such as where bedding dips increase from shallow to vertical across a faulted limb. This poses questions of whether anticlinal growth occurred along intersecting faults and monoclinal flexures rather than by some uniform or nonspecific 'upbending' of strata, and whether smaller folds are, in general, fault or flexure-bounded. Thus, the distribution of faults and lineaments shown in Figure 4 may be a useful guide to the detailed anticlinal structure, suggesting a polygonal, lineament-controlled pattern of anticlinal growth.

TECTONIC DEVELOPMENT Early phase Deformation of the Mesozoic-early Tertiary stiata from their original planar form to their present anticlinal form may have begun as early as terminal Eocene immediately following deposition of Eyre Formation. Indeed one hypothesis for Cordillo Silcrete genesis (Wopfner, 1978) requires embryonic anticlinal growth of the waterlogged, newly deposited Eyre Formation sands as part of the mechanism for this silcrete formation. It would follow that incipient uplift must have occurred about the time the silcrete formed. The age of Cordillo Silcrete is controversial but I believe the regional rockunit relationships demonstrate it to pre-date lacustrine deposits in the region, such as Etadunna Formation, (Stirton, et al., 1961; Callen et al.y 1976) which are not known to be older than mid-Miocene. On this basis Dalhousie Anticline began to take shape in the Oligocene to early Miocene. Structural growth during Cretaceous-early Tertiary deposition is known for parts of the Eromanga Basin (Moore & Pitt, 1984) but from the rather general data available for Dalhousie Anticline such growth appears to have been minor for the Dalhousie region. The Oligocene to early Miocene uplift that formed the embryonic Dalhousie Anticline was very mild producing folds of low amplitude and an undulating landscape of low relief (Wopfner, 1974, p. 6). Thus, the occurrence of Doonbara Formation in broad shallow depressions rather than deep channels illustrates gentle erosion resulting from low topographic gradients. Such erosion, although mild, probably initiated the breaching of Dalhousie Anticline and provided detritus for deposition of Doonbara Formation in adjacent shallow synclines. The mid-Tertiary episode of uplift was followed by a period of tectonic stability which may have lasted well into the Pliocene. Relatively stable conditions are represented firstly by the intense reddening and rubbly habit of Doonbara Formation produced by essentially soil-forming processes (e.g. Wopfner, 1974, p. 6) and secondly by the presence of the overlying, chemically deposited carbonate and silica suggesting precipitation in shallow water, undisturbed for a considerable period. As the upper part of Mount Willoughby Limestone might be no older than late Tertiary (Major, 1973; Firman, 1981) it appears that stable conditions

Fig. 4. Distribution of major lineaments, folds of various scale, and local observed faults. Lineaments show suggestion of polygonal pattern.

J2 5 KILOMETRES QUATERNARY Undifferentiated sand, silt, gravel Undifferentiated units of the breached anticline TERTIARY EYRE FORMATION and associated deposits. CRETACEOUS Eromanga Basin deposits Observed fault Lineament: prominent physiographic alignment, air photo, landsat Anticline, syncline, plunge

Monocline showing lower line of dip change — 1 Dip of bedding 0

KILOMETRES

.• *

179

DALHOUSIE ANTICLINE lasted until this time. That Mount Willoughby Limestone has subsequently undergone much of the uplift which affected Cordillo Silcrete is demonstrated by its close structural association with that unit. Although anticlinal growth may have begun in the early to mid-Tertiary it seems that the main development has occurred only since the late Tertiary. This late TertiaryQuaternary uplift not only accentuated relief to a marked degree but its onset introduced a new phase of tectonic events which produced the singular association of Quaternary features in the breached Dalhousie Anticline.

Late phase

As a direct response to break-up of the anticlinal crest and to formation of a major fault-related system of springs, three groups of units—Alinerta Gravel, Dalhousie Formation and unnamed mound spring sediments—were deposited within the breached anticline (Fig. 5). In the gross view these units may be thought of as an intertonguing assemblage of deposits representing a general episode of uplift but the different lithofacies and physiographic character of each allows a more explicit interpretation of the neo-tectonics of the anticline. The-three units, of substantially differing lithologies, are distributed concentrically within the anticline. Alinerta Gravel is a strongly gypseous unit that occurs in a broad irregular zone around the perimeter of the breached anticline, inside the silcrete rim; Dalhousie Formation, is a clay/dolomite unit inside the zone of Alinerta Gravel and immediately surrounding the main body of springs; the unnamed mound spring deposits, consisting of clay, silt and, locally, coarse to conglomeratic sand, occur in the core of the anticline. Further to these lithological and lateral variations are the dissection and relative elevations of the units. Thus, the units, particularly Alinerta Gravel and Dalhousie Formation, are markedly dissected and now occur as remnants of once-moreextensive surfaces above the present base level of erosion. Significantly, these erosional remnants do not represent the remains of a single, concordant land surface, but rather a number of separate, discordant surfaces of different elevations. Moreover, each unit shows more than one topographic level. For example, two levels of Alinerta Gravel, the lower abutting the higher with an elevation difference of about 5 metres can be seen on the northeast margin of the anticline. Under stereoscopic air-photo examination two such levels can be traced over much of the anticline, and in places three or more levels can be distinguished. Similarly, at least two dolomite levels of Dalhousie Formation occur and the higher of these intertongues with the lower, second level of Alinerta Gravel. Multiple phases of mound spring construction are apparent, too, where newly deposited mound material occurring as damp or wet, unvegetated 'mud heaps', forms mud flows along gullies in the older mound material. The general elevation of these clay/silt mounds is slightly less than that of their local, coarse variant which caps isolated erosional remnants of similar elevation to the upper level Dalhousie Formation. These physiographic features are summarised in Figure 6 which shows the Cainozoic units as a topo-sequence stepped down from Cordillo Silcrete and Mount Willoughby Limestone with the highest elevations, to Alinerta Gravel, then Dalhousie Formation, and finally to the mound spring deposits at the lowest level, with substantial vertical overlap of the units (cf. Wopfner et al., 1967, p. 125). Fig. 5. Cainozoic units of Dalhousie Anticline showing the essentially concentric distribution of mound spring deposits (centre), Dalhousie Formation, and Alinerta Gravel (perimeter) inside rim of silicified Eyre Formation.

KILOMETRES QUATERNARY

Undifferentiated sand, silt, gravel

Mound spring deposits: clay, silt

DALHOUSIE BEDS Dolomite, clay —

?rTT

ALINERTA GRAVEL Gypcreted gravel ? PLIO—PLEISTOCENE MOUNT WILLOUGHBY LIMESTONE Red sandstone, carbonates, chert EARLY TERTIARY EYRE FORMATION Silcreted sandstone CRETACEOUS Undifferentiated marine shale, sandFault Spring drainage

KILOMETRES

lito 1

Oz (/><

SI <o

180

G. W. KRIEG thermoluminescence have not yet been tried. Whether, for example, (a) the pulses occurred more or less steadily throughout the Quaternary or most frequently during a part of the epoch, (b) the magnitude of net uplift has been constant or variable from pulse to pulse, and (c) the direction of movements on particular faults has always been in the same sense, are questions requiring more detailed study. However, the following summary regarding the development of Dalhousie Anticline may be made: (1) anticlinal growth occurred along faults or monoclines producing an essentially polygonal structural pattern, and the suggested model (Fig. 7) indicates initial breakup to have been simultaneous over the whole structure (stage 3); (2) growth occurred mainly during the Quaternary and was intermittent; (3) growth may still be active as indicated by (i) modern mound construction which may require new fracturing that would suggest some recent movement, and (ii) the spring area being one of erosion, not deposition (apart from mound construction itself) with dissection prominent, and slightly perched watercourses evident in places. Regarding (ii) however, the lowering of erosional base level by evaporation of Lake Eyre may be an important cause of dissection in Late Pleistocene Holocene times.

REGIONAL SPECULATION

Fig. 6. Diagrammatic section illustrating lateral and vertical separation of Cainozoic units within Dalhousie Anticline.

To explain the different lithologies, concentric distribution and vertical separation of the units, successive pulses of tectonic uplift alternating with periods of quiescence may be considered. During uplift, the rising areas are eroded and the local low areas receive erosional debris. In the following quiet period, gypsum and carbonate crystallisation associated with groundwater circulation or other surficial processes may affect the newly deposited debris, or fine clastics and chemical precipitates may deposit in pools of still water. When these processes are interrupted by the next pulse of uplift and the new assemblage of deposits raised, the cycle is repeated. Thus, in the general context of a rising and eroding anticline the higher the unit in a topo-sequence the older it will be. However, the specific effect in any locality for a particular time will depend on the details of landform, spring activity, and fault and fold movements. If, for example, spring run-off is blocked a lake may form but if subsequent movements disrupt the lake margin or tilt the floor, the lake may either drain, partly drain, or otherwise adjust to the new conditions. In summary, if particular processes such as scarp erosion or spring activity predominate at different times in different places then the resultant deposits must be separated both laterally and vertically. Such an arrangement of units occurs in Dalhousie Anticline caused by the stages of anticlinal growth as shown in Figure 7. The exact timing of particular pulses of uplift, and correlation of specific pulses with particular structures are beyond the resolution of available data. Fossil evidence from the topo-sequence is inadequate for this level of precision whilst such techniques as magneto-stratigraphy and

Whether or not the tectonic history of Dalhousie Anticline has any direct relevance for wider areas of the Eromanga Basin is open to question, as very little is known of structural development in the deeper sections. Although the structure is adjacent to a seismically active zone in the Simpson Desert to the east (Youngs & Wopfner, 1972) it may (possibly) belong to a different tectonic regime. Thus, on all seismic horizons (e.g. Youngs, 1975) the contour patterns over the anticline and west thereof are more complex and have steeper gradients than to the east. As interpreted depth to magnetic basement (possibly crystalline basement) beneath the anticline is deeper, not shallower than that to the east (according to gravity and magnetic interpretations of Milton & Moroney, 1975) the greater complexity of patterns cannot be attributed merely to shallow section. If basement block faulting is the mechanism for structural growth then the different seismic patterns represent different 'fault domains' in the deeper section. Thus structural development of Dalhousie Anticline may apply, for example, only to the exposed southwest margin, and there may be no direct correlation of tectonic detail with other parts of the basin. The detailed tectonic history of the southwest margin of Eromanga Basin is likely to remain unknown until a substantial body of data, such as that provided by drill core, can be obtained from the deeper sedimentary and crystalline rocks.

ACKNOWLEDGEMENTS

This paper is published with permission of the DirectorGeneral, Department of Mines and Energy, South Australia. The comments of referees H. Wopfner (University of Cologne) and V. A. Gostin (University of Adelaide) are gratefully acknowledged.

DALHOUSIE ANTICLINE

181

uplift. lakes drained, dissection (gypsum pptn)

lacustrine deposition

o o

g y p s u m ppln flow exit blocked 9

FLOW EXIT

•

111

dissection

gypsum (carbonate) pptn

crestal breakup

carbonate, chalcedony pptn _ CO

ujO^ < Q_ —i

mild uplift

minor local faulting

I < > 1H — GC QCLU <H TECTONIC ACTIVITY quiet

Silica accumulation

mild folding Columnar silcrete

faulting

R e d s a n d , s a n d y carbonate c h a l c e d o n y Erosional base level Gypsum Spring discharge

L a c u s t r i n e clay, carbonate

4 M o u n d spring deposits

/T\

Drn- G.Bl

Fie 7

Development o f Dalhousie Anticline showing stages o f growth related to tectonic activity.

SADME

83-105

G. W. KRIEG

182

REFERENCES CALLEN, R. A. & TEDFORD, R. H., 1976: New late Cainozoic rock

units and depositional environments, Lake Frome area, South Australia. R. Soc. S. Aust. Trans. 100, 125-68.

EXON, N. F. & SENIOR, B. R., 1976: The Cretaceous of the Eromanga

and Surat Basins. BMRJ. Aust. Geol. Geophys. 1, 33-50. FIRMAN, J. B., 1981: Regional stratigraphy of the regolith of the southwest margin of the Great Australian Basin Province. S. Aust. Dept Mines Ener. Rep. 81/40 (unpubl.). FORBES, B. G., 1966: The geology of the M A R R E E 1:250 000 map area. S. Aust. Geol. Surv. Rep. Invest. 28. FREYTAG, I. B., 1966: Proposed rock units for marine Lower Cretaceous sediments in the Oodnadatta region of the Great Artesian Basin. S. Aust. Geol. Surv. Q. geol. Notes, 18, 3-7. FREYTAG, I. B., HEATH, G. R. & WOPFNER, H., 1967:

Oodnadatta

1:250 000 geological map, SG53-15. S. Aust. Geol. Surv. KRIEG, G. W. (compiler), in press. DALHOUSIE, South Australia. Explanatory Notes, 1:250 000 geological series. Sheet SG53-11. S. Aust. Geol. Surv. MAJOR, R. B., 1973: The Mangatitja Limestone. S. Aust. Geol. Surv. Q. geol. Notes, 47, 4-9. MILTON, B. E. & MORONY, G. K., 1975: A regional interpretation of

1:100 000 gravity and aeromagnetic maps of the Great Artesian Basin in South Australia. S. Aust. Dept Mines Ener. Rep. Invest. 46. MOORE, P. S. & PITT, G. M., 1984: Cretaceous of the Eromanga Basin—implications for hydrocarbon exploration. APEA J. 24(1), 358-76. MORGAN, R., 1980: Eustacy in the Australian Early and Middle Cretaceous. N.S.W. Geol. Surv. Bull. 27. NICOL, D., 1971: The Mount Willoughby Limestone. S. Aust. Geol. Surv. Q. geol. Notes, 39, 1-2.

STADTER, M. H., 1972: Re-interpretation of structural contour plan of V horizon (top of Cadna-owie Formation) western Great Artesian Basin. S. Aust. Dept Mines Ener. Rep. 72/29 {unpubl.).

STIRTON, R. A., TEDFORD, R. H . & MILLER, A. H., 1961: Cenozoic

stratigraphy and vertebrate palaeontology of the Tirari Desert, South Australia. Rec. S. Aust. Mus. 14, 19-61. THOMSON, B. P. (compiler), 1980: South Australia 1:1 000 000 geological map. S. Aust. Geol. Surv. WHITEHOUSE, F. W., 1955: The geology of the Queensland portion of the Great Australian Artesian Basin. Appendix G, Artesian Water Supplies in Queensland. Dept Co-ord. Gen. Public Works. Qld Pari. Pap. A, 56-1955.

Post-Eocene history and stratigraphy of northeastern South Australia. R. Soc. S. Aust. Trans. 98, 1-12. WOPFNER, H., 1978: Silcretes of northern South Australia and adjacent regions; in Langford-Smith, T. (ed.) Silcrete in Australia. Univ. New England, Dep. Geogr., 93-141.

WOPFNER, H., 1974:

WOPFNER, H., CALLEN, R. A . & HARRIS, W. K., 1974: T h e Lower

Tertiary Eyre Formation of the southwestern Great Artesian Basin. Geol. Soc. Aust. J. 21, 17-51.

WOPFNER, H., FREYTAG, I. B. & HEATH, G. R., 1970: Basal Jurassic-

Cretaceous rocks of western Great Artesian Basin, South Australia. Stratigraphy and environment. Am. Assoc. Pet. Geol. Bull. 54, 383-415.

WOPFNER, H . & TWIDALE, C. R., 1967: Geomorphological history

of the Lake Eyre Basin; in Jennings, J. N. & Mabbutt, J. A. (eds) Landform Studies from Australia and New Guinea. Aust. Nat. Univ. Press, Canberra, 118-143. YOUNGS, B. C., 1975: The geology and hydrocarbon potential of the Pedirka Basin. S. Aust. Dept Mines Ener. Rep. Invest. 44. YOUNGS, B. C. & WOPFNER, H., 1972: Subsurface faults and recent

earthquakes in the Simpson Desert. S. Aust. Geol. Surv. Q. geol. Notes, 43, 8-11.

Geological Society of Australia Special Publication No. 12, 183-194

The Merrimelia Oil and Gas Field—a case history O. J. W. Bowering & D. M. Harrison 1

1 2

Delhi Petroleum Pty Ltd, 101 Grenfell St., Adelaide, S.A. 5000. Alliance Petroleum Australia NL, 35 Collins St., Melbourne, Vic. 3000; present address c/- Australian Hydrocarbons NL, 12 Creek St., Brisbane, Qld 4000.

ABSTRACT

The Merrimelia oil and gas field is hydrocarbon productive from reservoirs ranging in age from Early Permian to Late Jurassic. Gas and gas liquids are produced from Permian and Triassic reservoirs while oil is produced from Triassic and Jurassic reservoirs. The Merrimelia South high is bald of Permian sediments and the earliest drilling on the structure was designed to test Permian stratigraphic plays on its flanks. Of the five wells drilled during this stage, only Merrimelia 5 was completed as a potential gas producer. Later discoveries of oil and gas in several fields in sediments of the Eromanga Basin sequence led to the upgrading of the prospectivity of Jurassic and Early Cretaceous sandstones. The second phase of drilling on Merrimelia began eleven years after completion of the first phase and was aimed principally at Mesozoic targets. Discoveries of oil and gas were recorded from Jurassic, Triassic and Permian reservoirs, and the Merrimelia field thus became the first in the Cooper and Eromanga Basins to be productive from reservoirs of these ages. Like other fields of the Cooper and Eromanga Basins, Merrimelia contains structural accumulations with some stratigraphic control on trapping.

HISTORY OF EXPLORATION Geological and geophysical

Innamincka 1 drilled in 1959 to test Mesozoic and Cambrian sediments on a large closed anticlinal feature, was the first well to penetrate Permian sediments within the DelhiSantos licence areas and indicated the presence of a Permian basin below the Great Artesian Basin. Hydrocarbon 'shows' recorded in the Permian sediments in this well gave the first indication of their possibly prospective nature. Aeromagnetic and gravity surveys carried out between 1960 and 1964 outlined the extent of the Permian basin which was initially named the Coopers Creek Basin and subsequently the Cooper Basin (Wongela Geophysical Co. Pty Ltd, 1965). The first indication of the existence of a structural trend which includes the Merrimelia anticline (Fig. 1), resulted from a

Fig. 1. Location plan, Merrimelia oil and gas field.

reconnaissance aeromagnetic survey flown in 1961 (Innamincka-Betoota Aeromagnetic Survey by Aero Services Ltd). Although the broad reconnaissance nature of this survey was not designed to delineate individual anomalies, certain anomalies were suggested. Early in 1963 a refraction seismic line was shot northward from Dullingari 1 towards Innamincka and then westward to cross two of the aeromagnetic anomalies (United Geophysical Corp., 1963a). A strong reversal was indicated by this survey, which subsequent seismic surveying showed to be located across a saddle between the Gidgealpa and Merrimelia structures. This initial seismic survey was followed by two surveys during 1963 and 1964. These were respectively the Diamantina River—Gregory Range Seismic Survey and the Coopers Creek Seismic Survey. These were the first to define the general configuration of the Merrimelia structure (United Geophysical Corp., 19636, 1964). The early single-fold seismic mapping which was relatively primitive by modern standards, was nevertheless capable of mapping the C, P and Z horizons (see Fig. 2) and was able to delineate the attitude and thickness of Permian sediments below the Mesozoic section. It was a common practice of the licence operator during this period to record gravity data at each seismic shot point and much of the regional gravity data that aided in outlining the essential configuration of the Cooper Basin was obtained in this manner. In addition, Wongela Geophysical Co. Pty Ltd (1965) carried out a helicopter gravity survey in 1965 on behalf of Delhi and Santos (Strzelecki^Cooper Gravity Survey) and compiled a regional Bouguer gravity map. This map, part of which is shown in Figure 3, illustrates the major northeast trending ridge, now known as the Gidgealpa-MerrimeliaInnamincka Ridge, which separates the Patchawarra and Nappamerri Troughs. One notable feature observed during the initial mapping of individual structures was the marked thinning of the Permian sedimentary section over their crests. The larger

184

O. J. W. B O W E R I N G & D. M . H A R R I S O N

1 9 6 4

'C HORIZON

NO PERMIAN

1 9 6 4

'P' H O R I Z O N

Fig. 2. 1964 ' C and 'P' Horizon structure contour maps produced by single-fold seismic mapping.

BOUGUER ANOMALIES CONTOUR INTERVAL 1 MILLIGAL

Fig. 3. Bougucr gravity map showing Gidgealpa-Mcrrimclia-Innamincka Ridge.

MERRIMELIA FIELD structures with m o r e p r o m i n e n t vertical relief, such as Merrimelia a n d I n n a m i n c k a , are bald of Permian sediments at their crests, either as a result of non-deposition, o n l a p or of subsequent uplift a n d erosion, or a combination of these factors. In 1969 a f a r m o u t agreement was entered into between the Delhi-Santos-Vamgas c o n s o r t i u m on the one h a n d , a n d Alliance Petroleum Australia N.L. on the other whereby Alliance obtained the right to earn a 5 0 % working interest in a farmout area known as the Merrimelia-Innamincka Block by carrying out the Merrimelia Seismic Survey in 1970 a n d by drilling f o u r wells. T h e objectives of this survey were to m a p suitable closures on the f l a n k s of Merrimelia and to delineate the pinchout of Permian sediments. Attempts to delineate the Permian pinchout met with limited success due to the acoustic masking effect of the u p p e r Permian coals. Prior to the late 1960s, conventional dynamite techniques were primarily used with single-fold coverage a n d analogue recording a n d processing. T h e late 1960s and early 1970s saw the introduction of multi-fold coverage and digital processing and by 1972, all recording was digital. In 1973, 'Vibroseis' recording became the d o m i n a n t m e t h o d used for exploration within the licence areas, usually with twelve-fold coverage. T h e introduction of these techniques led to a significant improvement in data quality and greater accuracy in structural definition. Seismic surveys carried out in recent years over the Merrimelia structure include Oonabrinta, Moonlight Flat, Karawinnie a n d N a m o o k a (Delhi Petroleum Pty Ltd, 1978, 1979, 1980, 1981). These have enabled accurate determination of the crests of individual closures a n d provided drilling locations for wells aimed at testing the oil potential of the Mesozoic sequence. A c o m p a r i s o n between an early singlefold seismic section a n d a m o r e recent section is shown in Figures 4 and 5.

Exploration drilling T h e initial discovery of commercial gas at Gidgealpa established the Permian sedimentary section as the primary objective within the licence areas, and subsequent discoveries at Moomba, Daralingie and elsewhere confirmed this. Interest then centred u p o n the Permian sediments on neighbouring anticlinal structures, particularly those on the same structural trend as Gidgealpa. It was a logical step therefore, to extend exploratory drilling to the Merrimelia structure. Seismic m a p p i n g showed over 150 metres of closure at the P horizon (near t o p of Permian) a n d showed the Permian section to be either thin or absent at the crest of the structure, increasing in thickness to over 450 m at a distance of 8 km southeast of the crest. For this reason, Merrimelia 1 was located away f r o m the crest a n d sufficiently d o w n f l a n k to penetrate a suitable thickness of Permian sediments which, it was believed, would contain the two p o r o u s sandstone reservoirs in the u p p e r a n d lower parts of the section, that flowed gas at Gidgealpa (ref. Figs 2, 6). T h e objectives of this first well on the Merrimelia structure were: (i) the U p p e r and Lower Permian porous sandstone reservoirs, productive at the Gidgealpa Field; (ii) sandstones of Permo-Carboniferous age encountered below the Permian section at Gidgealpa. (Seismic data suggested that this unit, subsequently known as the Merrimelia Formation, wedged out updip from the well location); (iii) porous m a r i n e dolomites of C a m b r i a n age that yielded strongly gas-cut salt water in Gidgealpa 1. (The presence of these sediments was suggested by deeper seismic reflections).

185

Merrimelia 1 spudded in September, 1964 and was drilled to a total depth of 3150 m, bottoming in the PermoC a r b o n i f e r o u s Merrimelia Formation. Gas shows were recorded in Triassic and Permian sandstones which, when tested, flowed non-commercial gas to surface. Three subsequent wells were drilled on the Merrimelia structure during 1964 and 1965. T h e only results of any significance recorded during this stage of drilling were firstly, a small (0.6 m in pipe) recovery of free oil on top of m u d from the N a p p a m e r r i Formation in Merrimelia 2. Although this is the first recorded show of oil in Triassic sediments, it was concluded at the time that the lack of porosity and effective permeability precluded the production of commercial hydrocarbons. Secondly, a gas flow of 79 300 m Vday was recorded f r o m the Patchawarra Formation in Merrimelia 4. However, a low declining reservoir pressure made this recovery noncommercial and the well was a b a n d o n e d as a dry hole. Alliance Petroleum, after farming into the MerrimeliaInnamincka Block, drilled Merrimelia 5 in 1970. Flows of 283 000 mVday and 36 800 m V d a y were recorded from a lower Nappamerri sandstone and from a basal N a p p a m e r r i / upper Toolachee sandstone respectively. T h u s the first stage of exploration on the Merrimelia structure ended in 1970 with the completion of Merrimelia 5 as a gas producer from the lower part of the N a p p a m e r r i Formation and the upper part of the Toolachee Formation. T h e overall results of exploration to this stage which were directed toward the Permo-Triassic section, were somewhat disappointing in that the initial large potential of the Merrimelia structure, engendered by the results of Gidgealpa, was not realised. T h e reservoir sandstones of the Permian section, productive at the neighbouring Gidgealpa Field, proved to have p o o r reservoir characteristics at Merrimelia. In addition the Triassic sandstones appeared to lack continuity over much of the structure and the initial encouragement given to the possibility of oil production from this section did not materialise. Because of the need to develop additional gas reserves for the Sydney gas market, exploration activity after 1970 was focused upon other prospects. During the late 1960s and early 1970s discoveries were m a d e at Toolachee, Packsaddle, Tirrawarra, Delia and Strzelecki. Subsequent drilling activity during the early to late 1970s was directed largely to appraising and developing Permian gas discoveries with only minor wildcat drilling. In September, 1978, while drilling an appraisal well on the Strzelecki gas field (Strzelecki 3), oil was discovered in the N a m u r Sandstone Member of the M o o g a Formation and H u t t o n Sandstone of the overlying Mesozoic E r o m a n g a Basin. This discovery led ultimately to the second stage of exploration on the Merrimelia structure. Prior to the discovery of oil at Strzelecki, the E r o m a n g a Basin had been considered unprospective for hydrocarbons. The E r o m a n g a Basin forms part of the Great Artesian Basin which has an active artesian water drive, and the previous lack of success in this very extensive basin, led to the conclusion that the flushing action of the basin-wide artesian flow had removed any hydrocarbons that might otherwise have accumulated in it. Subsequent discoveries at Dullingari, Cuttapirrie and M a r a b o o k a , led to a reappraisal of the E r o m a n g a Basin sequence and an upgrading of its hydrocarbon potential. T h e second stage of exploration at Merrimelia began with the decision to drill a well at the crest of the structure which had previously been avoided because of the absence of Permian sediments. Eleven years after the completion of the fifth well, Merrimelia 6 was spudded in July, 1981 with the

186

TIME SECS

1963 LINE H

1 km

O. J. W, BOWERING & D. M. HARRISON

Fig. 4. 1963 seismic dip line across Merrimelia 'South High', using single-fold dynamite coverage, analogue recording and processing.

TIME SECS

SE NW

TIME SECS TIME SECS

I k m

LINE 8 0 - J W S

MERRIMELIA FIELD

Fig. 5. 1980 seismic dip line across Merrimelia 'South High', using multi-fold Vibroseis coverage, digital recording and processing.

187

188

O. J. W. BOWERING & D. M. HARRISON

Namur Sandstone Member of the Mooga Formation, the Hutton Sandstone and Nappamerri Formation as its primary objectives. A flow rate of 433 barrels of oil per day (BOPD) with 679 barrels of water per day (BWPD) was recorded from the top of the Namur Sandstone Member, while the Hutton Sandstone produced'a flow of 2738 BOPD with 162 BWPD. As in Merrimelia 5, the Nappamerri Formation was found to be gas productive with flow rates between 17 000 and 270 000 m 3 /day from drill stem tests (DSTs) in various zones. Merrimelia 6 is therefore the discovery well for commercial oil in the Namur Sandstone Member and Hutton Sandstone in the Merrimelia Field.

Merrimelia structure. The primary objectives were the Namur Sandstone Member and Hutton Sandstone. Oil was encountered in the Namur Sandstone Member and flowed to surface at 2534 BOPD. The Hutton Sandstone also tested oil, but at a rate which declined throughout the test and averaged 194 BOPD. A subsequent cased hole DST retested the Hutton Sandstone along with a sand in the basal portion of the Birkhead Formation and resulted in a stabilised flow of 576 BOPD. Two gas flows were recorded from sandstones in the middle part of the Nappamerri Formation. The higher sands flowed gas at 96 300 m 3 /day and the lower sands at 270 000 m 3 /day. The well was completed for oil production from the Namur Sandstone Member.

Appraisal and development drilling

Merrimelia 9 was programmed as an appraisal well and was located to intersect the gas-oil contact between Merrimelia 6 to the north which flowed gas from the Nappamerri Formation sandstones and Merrimelia 7 to the west which flowed oil from the same sands. However, these sands flowed gas in Merrimelia 9 at the rates of 113 300 m 3 /day from the lower sands and 2200 m 3 /day from the upper sands; in the latter case these sands are 10 m below highest known oil as mapped in the same sands in Merrimelia 7.

Following the Merrimelia 6 discovery, an appraisal drilling programme commenced with the principal objective of defining the Mesozoic oil reserves. The first well in this appraisal programme, Merrimelia 7, was located on the southwestern flank of the central culmination, downdip from Merrimelia 6. It was sited to test for a downdip extension of the gas accumulation discovered in a Nappamerri Formation sandstone reservoir in Merrimelia 6, and to investigate the possible existence of an oil leg within this reservoir. Jurassic sandstones presented secondary objectives. Oil was discovered in the primary objective and flowed to surface at 2850 BOPD. In addition, a gas flow of 69 400 m 3 /day was recorded from lower in the Nappamerri Formation, this being a different reservoir from that which flowed gas in Merrimelia 6. Merrimelia 7 is thus the discovery well for oil in the Nappamerri Formation. Merrimelia 8 was drilled to test the Mesozoic section on a separate subsidiary culmination at the crest of the

Merrimelia 10 is located east-southeast and slightly updip of Merrimelia 6, and is a development well for the Hutton Sandstone oil accumulation discovered in that well. A flow rate of 2326 BOPD was recorded and the well was completed for production from the Hutton Sandstone. Additional oil pay was mapped in the Namur Sandstone Member and the basal Hutton Sandstone and gas pay exists in the Nappamerri Formation. Merrimelia 11 located east-northeast of Merrimelia 6 is a development well for Hutton oil. The final clean-up flow was

MERRIMELIA FIELD 3200 BOPD from this unit. In addition, the well also recorded a flow of gas from the Nappamerri Formation. The No. 12 well, located southwest of No. 7 was programmed as a Unit gas development well to evaluate gas reserves in the basal Nappamerri and uppermost Toolachee sands. However, a drill stem test in a mid Nappamerri sand flowed oil to surface and because of the greater desirability of developing additional oil reserves for the Stony Point liquids scheme, it was decided to complete the well as an oil producer. A final clean-up flow of 1900 BOPD was recorded from this reservoir which appears from wireline log correlation to be equivalent to that which flowed oil in Merrimelia 7. The No. 13 well is located between the No. 7 and No. 12 wells and was drilled to test the continuity of the accumulation in the mid Nappamerri Formation sand and to locate the oil-water contact. A gas flow of 136 000 mVday was recorded from Toolachee Formation sands. In addition a gas flow of 215 000 m /day was recorded from the Patchawarra Formation, this being the first Early Permian discovery in the Merrimelia Field. 3

189

The full stratigraphic record occurs only in the basin depocentres and becomes less complete over basement ridges and towards basin margins. Thus the Merrimelia, Packsaddle and Innamincka highs are devoid of Cooper Basin (PermoTriassic) sediments. The loss of stratigraphic section over the ridges can be due to non-deposition, onlap or subsequent uplift and erosion. It is likely all three have occurred at some time over the Merrimelia High. The Cooper Basin, of PermoTriassic age, is one of a number of Palaeozoic basins which underlie the Jurassic-Cretaceous Eromanga Basin. The outlines of these infra-basins are shown in Figure 8. Because of the marked influence of the Gidgealpa-MerrimeliaInnamincka Ridge since Early Permian time on almost all subsequent sedimentation, an integrated description of stratigraphy, structure and geologic history is attempted and the stratigraphy of the Merrimelia High is presented from the authors' viewpoint. The following 'working stratigraphy' applied to the Merrimelia High, is currently in use by Delhi and its associates. It is summarised in Figure 9.

STRATIGRAPHY OF THE MERRIMELIA FIELD Pre-Permian Introduction The pre-Permian basement intersected in a number of the

The Merrimelia oil and gas field is located on and around the Merrimelia High which is one of many structures that together form part of an overall northeast trending but arcuate ridge termed the Gidgealpa-Merrimelia-Innamincka Ridge, which separates the sub-parallel Nappamerri and Patchawarra Troughs of the southern Cooper Basin (Fig. 7). Along this approximately 140 km long, sinuous, arcuate and narrow structural high trend are arranged from south to north in apparent en echelon display, the Gidgealpa, Merrimelia, Packsaddle and Innamincka anticlines.

Fig 7 Structural configuration of southern Cooper Basin.

Merrimelia wells consists of buff to white quartzite with interbedded red and green shale correlated with the 'Innamincka redbeds'.

Permian

As stated above, there are no Permian sediments on the crest of the Merrimelia High. Merrimelia 6 to 11 inclusive have Triassic sediments directly overlying 'Innamincka redbeds'. Merrimelia 1, 4, and 5 to the south of the crest, and

190

O. J. W. BOWERING & D. M. HARRISON

the Merrimelia High have contributed to the dramatic thinning of the Patchawarra Formation toward the crest. Throughout the Cooper Basin a disconformity has been recognised between the Upper and Lower Permian section which represents a weak to moderate tectonic reactivation (Kapel, 1972; Rumph, 1978; Thornton, 1978). On the Merrimelia High this is represented by an angular unconformity evident on processed dipmeter logs. The bounding faults that may have been present in pre-Permian times and were active during Patchawarra Formation deposition were reactivated, resulting in major uplift of the Merrimelia horst block, tilting of the Merrimelia Formation, Tirrawarra Sandstone and Patchawarra Formation away from the horst, and erosion. The Upper Permian Toolachee Formation, a fluviatile unit comprising channel and point-bar sandstones, floodplain shales and siltstones and backswamp coals, subsequently spread across the eroded land surface. The Toolachee Fig. 8. Infra-basins of western Eromanga Basin. Formation onlapped the Meifimelia High but did not completely cover the crest which remained an isolated topographic high or basement inlier during Toolachee Merrimelia 3 to the north of the crest penetrated Permian Formation deposition. Toolachee Formation sandstones strata. The Permian section in these wells is far from complete provide the reservoirs for the majority of current Cooper Basin and is limited to the Merrimelia, Tirrawarra, Patchawarra and gas reserves. As the better sandstones are concentrated in the Toolachee Formations (Gatehouse, 1972; Kapel, 1966). lower half of the formation and the upper half is dominated The basal Merrimelia Formation which has glacial by thick coals and shales, and given onlap onto the Merrimelia affinities, comprises diamictite, sandstone and shale. The High, the Toolachee Formation closer to the crest is therefore sediments are deficient in organic content and generally the poorer quality reservoir. Forty-eight metres of Toolachee lacking in porosity and permeability. The formation is thickest Formation were intersected in Merrimelia 1. In the troughs in Merrimelia 1 where it is 363 m thick. It thins updip by onlap the Toolachee Formation can attain 150 to 200 m thickness and is absent from the crest either due to non-deposition or (Battersby, 1976; Laws, 1977). more likely due to later erosion. Triassic The Merrimelia Formation is overlain unconformably by siltstones, and minor floodplain shales of the Lower the Tirrawarra Sandstone, which consists of a stacked to Fluvial Middle Triassic Nappamerri Formation (Papalia, 1969) sequence of braided-stream deposits. Thickness in the overlie Toolachee Formation conformably in the troughs Patchawarra Trough averages 50 m with a maximum of 76 m. but thintheonto highs both due to onlap and to erosion In Merrimelia wells, where present, it varies from 26 m to following majortheuplift Late Triassic time. Thus on the 38 m thick and so is seen to thin marginally over the Merrimelia High there during is probably a disconformity between Gidgealpa-Merrimelia-Innamincka Ridge. the Toolachee and Nappamerri Formations. To the north of The Patchawarra Formation overlies the Tirrawarra the high 160 m of Nappamerri Formation were intersected Sandstone on the Merrimelia High probably with minor in Merrimelia 2 and to the south, 143 m were intersected in internal unconformities, as both formations thin undip by Merrimelia 1. The formation thins to 79 m of predicted onlap. In the Nappamerri Trough, the Patchawarra Formation Triassic section from regional studies and seismic data in the is up to 480 m thick (Battersby, 1976). On the Merrimelia High Nappamerri Trough (Battersby, 1976). There is a transition the thickest intersection to date of Patchawarra Formation zone between the Toolachee and Nappamerri Formations is 159 m in Merrimelia 1. The Patchawarra Formation across which the carbon content of the sediments rapidly sediments are fluviatile and consist dominantly of floodplain decreases caused by a climatic change to considerably more shales and siltstones, backswamp coals and channel and point- arid conditions (Gausden, 1981). bar sandstones. On the Merrimelia High, Delhi and its associates recognise The upper Lower Permian formations, the Murteree Shale, an informal three-fold subdivision of the Nappamerri Epsilon Formation, Roseneath Shale and Daralingie beds, are Formation. A lower section of interbedded sandstone, shale deposited as a conformable sequence overlying the and siltstone is overlain by a dominantly sandy midPatchawarra Formation in the Patchawarra Trough. On the Nappamerri section which in turn is overlain by an upper Merrimelia High these formations are all absent. On the section of siltstone and shale. Oil and gas reserves exist in northwest flank of the high in Mudrangie 1 and 2 wells, the mid-Nappamerri Formation reservoirs and the lower Murteree Shale and Epsilon Formation were intersected. It Nappamerri Formation reservoirs contain gas reserves. is therefore likely that these formations were deposited over Sandstone unit correlation between wells is subjective due to the Merrimelia High but were later removed by the main pre- the low well density, but detailed facies and interconnection Toolachee Formation erosive episode. It is probable that the studies are in progress to elucidate this problem. The better depositional edges of the Roseneath Shale and Daralingie beds sandstone units are generally contained within the midlie within the troughs and did not exist on the Gidgealpa- Nappamerri Formation and despite loss of basal section due Merrimelia-Innamincka Ridge. Battersby (1976) noted that to onlap and loss of section from the top due to erosion, the during the deposition of the Patchawarra Formation over the mid-Nappamerri Formation extends over the high. This Gidgealpa High, there was significant thinning due to growth improves prospectivity for further discoveries within this faulting. It is most likely that growth faults on the flank of middle unit.

MERRIMELIA FIELD REMARKS Long period of deep w e a t h e r i n g E r o s i o n of b o s i n m a r g i n s o n d Tertiary structural highs. L i m i t e d d e p o s i t i o n of locustrine, fluvial ond loter a e o l i o n s e d i m e n t s . Fluvial * lacustrine Shallow marine Shallow marine, parotic

T r o n s g r e s s i v e sequence. Fluvial to s h a l l o w m a r i n e . O c c a s i o n a l development of bar ond b e o c h ? s a n d stones.

Fluviol - Lacustrine S o n d s t o n e s dominont

Fluvial - Locustrine Interbedded s h o l e s minor c o o l s .

ond

sondstones

Widespread erosion, climax of struc turol g r o w t h commenced in the Permian n o n - morine deposition West of the Cooper B o s i n . Fluvial- Locustrine

Fluviol - L o c u s t r i n e . ment. ( R e w o n

Arid

environ-

equivalents)

S h o l e s , c o o l s , fluviol

Locustrine

(fluvial)

restricted Basin

sonds

deposition

Southern

S h o l e s , c o o l s , fluviol

Cooper

sonds

Fluviol - periglaciol deposits. M a y overlie granites , metomorphics or u n m e t o m o r p h o s e d earlier poloeozoic sediments

Fig. 9. Stratigraphy of southern Cooper and western Eromanga Basins (after Gatehouse, 1972; Gausden, 1981; Moore & Pitt, 1982).

192

O. J. W. BOWERING & D. M. HARRISON

Jurassic During Middle to Late Triassic time, regional uplift and erosion coincided with local reactivation of Permian fault trends. Resultant erosional stripping was greater over the structurally high trends and an essentially peneplained surface developed. Major epeirogenic downwarping during the Early Jurassic initiated the Eromanga Basin. The Jurassic deposits that followed consist of massive fluviatile sandstone sequences with minor lacustrine shale and rare coal interbeds. Up to four gross cycles of deposition can be recognised each of which (with the exception of the basal cycle) commenced with braided stream deposits and graded upward into meandering stream deposits and finally into lacustrine shales, silts and minor floodplain coals. Each cycle is characterised by a gross fining upward sequence and decrease in depositional energy. No coal has been observed at the top of Cycle 4 possibly because of destruction by the subsequent Cretaceous marine transgression. Each cycle is probably initiated by basin-margin uplift with resultant rejuvenation of the sediment source areas (ref. Fig. 9). The four cycles are: Cycle 4 Namur Sandstone Member of the Mooga Formation Cycle 3 (Westbourne Formation (Adori Sandstone Cycle 2 (Birkhead Formation (Hutton Sandstone Cycle 1 'basal Jurassic' unit The Jurassic sandstone-rich formations have good porosity and permeability characteristics, and form part of the Great Artesian (or Eromanga) Basin aquifer system. The shale-rich formations completing each cycle form good seals-on the lower potential reservoir portions. Over the Merrimelia High, oil fields are defined in the Namur Sandstone Member and Hutton Sandstone and possibly the Birkhead Formation, with minor oil in the 'basal Jurassic5 unit. The Jurassic sequence above the Birkhead Formation is known as the Mooga Formation (Nugent, 1969) where it is not possible to discriminate the Westbourne Formation within a massive sandstone section. This is the case in many of the Merrimelia wells. The Mooga Formation includes the Murta Member, a sequence of lacustrine siltstones with thin interbedded lakemargin sandstones of Early Cretaceous age. Over the Merrimelia High the Jurassic section is remarkably constant in thickness, as it is over the Cooper Basin in general. It varies from 490 m to 536 m thick; much of the variation occurs within the Birkhead Formation which ranges from 43 m to 92 m thick in the Merrimelia wells.

Cretaceous Late Jurassic fluviatile sedimentation was followed by the development of an extensive lake system, the sediments of which are represented by the Murta Member of the Mooga Formation (Nugent, 1969). The five member subdivision of the Murta Member defined by Mount (1981) can be seen in the Merrimelia wells but nowhere on the Merrimelia High are the beach, bar or shoreface sands of unit M2 (Mount, 1981) developed. However, oil has been recovered in pipe on DST from 'tight' (low permeability) sands of the M4 unit in several Merrimelia wells. An eustatic rise in sea level during Early Cretaceous time, possibly accompanied by continued gentle subsidence, led to a widespread marine transgression across the entire Eromanga Basin. Sediments deposited during this transgressive phase are represented by the 'Transition beds'. The Cretaceous stratigraphy has recently been revised and the new units

defined by Moore & Pitt (1982) have been adopted in the stratigraphic column shown in Figure 9.

STRUCTURE OF THE MERRIMELIA FIELD Rumph (1978) hypothesised that 'Cambro-Ordovician movements of the Delamerian Orogeny within the Warburton Basin may have been responsible for the embryonic establishment of the NE structural trend which assumes major significance in the development of the Permian Cooper Basin'. The overall northeast trend of the Gidgealpa-MerrimeliaInnamincka Ridge was probably established in Early Palaeozoic time due to major tectonic stresses in the Australian craton. The ridge is an excellent example of the maxim 'once a structure always a structure, once a zone of weakness always a zone of weakness' which seems to apply to most intracratonic basins. It is likely that after the period of uplift and erosion of the pre-Permian 'Innamincka redbeds' and prior to Permo-Triassic deposition, a number of 'whalebacks' (terminology of Cameron (1961)) formed along the GidgealpaMerrimelia-Innamincka Ridge. It was on this surface that Permian deposition commenced. The principal mappable seismic events over the Merrimelia High are: (i) C horizon top 'Transition beds' (ii) DN horizon top Namur Sandstone Member (iii) DB horizon top Birkhead Formation (iv) H horizon top Hutton Sandstone (v) N horizon top Nappamerri Formation (vi) P horizon top Toolachee Formation (vii) V horizon top Patchawarra Formation (viii) Z horizon pre-Permian basement The density of seismic lines is such that the structure is now well mapped. The Merrimelia High is seen on all horizons to be about 40 km long by 5 km wide and separated by saddles from the Packsaddle High to the nor-theast and the Gidgealpa High to the south. The Merrimelia High itself has two major culminations separated by a low saddle. The Merrimelia 'South High' is a 16 km by 3 km regular, linear dome in contrast to the Merrimelia 'North High' which is somewhat irregular and splays east from the normal northeast trend, with overall dimensions of 11 km by 2.5 to 3.5 km. The Merrimelia oil and gas fields discovered to date are all on the Merrimelia 'South High'. One dry hole, Merrimelia 3, has been drilled on the Merrimelia 'North High'; however, there is structural relief updip from that well for the entrapment of hydrocarbon accumulations. From the evidence of Merrimelia 3, the 'North High' unlike the 'South High' is not devoid of Permian sections. The seismic data show that the faulting generally decreases in amplitude upward through the stratigraphic section and major older faults grade into monoclines or have only minor displacements. To illustrate this; vertical closure of the Merrimelia dome at C horizon is 90 m, compared with 280 m at P horizon (Fig. 2) and with the assistance of bounding faults, 490 m at the Z horizon. The Z horizon depth structurecontour map shows a complex pre-Permian fault system. Major northeast trending faults flank both sides of the Merrimelia High and numerous additional secondary, parallel faults add to the complexity of the structure. At the Z, P and N horizons, the Merrimelia High is best described as a horst. At Z horizon, displacement on the bounding faults to the horst is up to 300 m whereas on the P and N horizons, displacement is of the order of 50 m. This, together with the angular unconformity seen on the 'South High' mentioned earlier, suggests that major uplift occurred prior to deposition of the

193

MERRIMELIA FIELD Toolachee Formation. Thus horst uplift could have occurred either principally as an episode during the Early Permian to Late Permian hiatus, or concomitant and sympathetic with Early Permian deposition. It is difficult to decide which of these has been operative but it is likely to have been a combination of both. A second phase of horst uplift occurred during Late Triassic time as indicated by displacement at the P and N horizons. These bounding faults do not pass up into the Jurassic-Cretaceous section. Thus, the structural trends were established prior to Permian sedimentation. Structural control of sediment accumulation was quite strong during Early Permian time but became subdued by the Late Permian and Triassic. Folding during this period may have been restricted to compactional drape over the structural high. Jurassic-Cretaceous sedimentation was remarkably consistent both in lithofacies and thickness over much of the southern Cooper Basin, and the Gidgealpa-MerrimeliaInnamincka Ridge had little influence on deposition during these periods. Subsequently, weak structural rejuvenation may have begun in the Late Cretaceous but probably reached its peak in the Early to Mid-Tertiary, folding the Jurassic and younger sediments and tightening the pre-existing structure of the Permian and Triassic sediments. Mancktelow (1979) expressed the opinion that all faults presently outlined in the Cooper Basin appear to be normal gravity faults, generally with steeply dipping to near vertical fault planes. Vale (1980) however, has interpreted the Merrimelia northwest bounding fault to be variously a reverse fault and a normal fault with varying throw along its length. This, together with the en echelon arrangement of the NNE trending Tirrawarra and Coonatie anticlinal noses in the Patchawarra Trough, led Vale to propose a wrench hypothesis to explain the array of faults and folds. The wrench hypothesis was first proposed by Sprigg (1961) and has recently been extended by Blake (1982). In contrast, Mancktelow (1979) proposed a mechanism involving the interaction between two intersecting fold trends to explain the array of folds in the southern Cooper Basin. It is not the subject of this paper to argue either case.

HYDROCARBON SOURCE AND MATURATION Basinwide studies of thermal maturity and organic geochemistry have shown that the finer grained sediments contained within much of the Jurassic section and a significant portion of the Early Cretaceous section are capable of generating hydrocarbons (Kantsler & Cook, 1979). Vitrinite reflectance profiles from the Jurassic and Cretaceous sections from numerous wells in the Cooper Basin area show that the average reflectance gradient with depth is rather linear and generally low; less than 0.35% Ro km " 1 . These contrast with Permian gradients which are typically curved and generally exhibit somewhat higher average gradients. Highest gradients are in excess of 3.0% Ro k m - 1 in the Permian section of Burley 1 (Kantsler & Cook, 1979). A pronounced break in the gradient at the Triassic/Jurassic unconformity is common. The vitrinite reflectance gradient at Merrimelia 6 is shown in Figure 10. The gradient in this case is more curved than usual for the Mesozoic section of the Eromanga Basin. The reason for this is obscure but may possibly be due to the absence of Permian sediments over the crest of the Merrimelia structure. The increasing gradient toward the lower part of the section in this well may be due to the fact that Mesozoic sediments directly overlie pre-Permian rocks. The reflectance data indicate that the sediments at Merrimelia 6 enter the oil window at a depth of 1550 m (Cook, 1981) corresponding

MERRIMELIA No. 6 RV MAX AND RANGE PROBABLE CAVINGS REFLECTANCE PROFILE ALTERNATIVE REFLECTANCE PROFILE PRESENT WELL TEMPERATURE TOOLEBUC WALLUMBILLA TRANS

ZZZZZ MURTA NAMUR BIRKHEAD HUTTON BASAL J. •^ZZ NAPPAMERRI PRE-PERMIAN

T.D.

0.5

1.0

VITRINITE R E F L E C T A N C E Fig. 10. M e r r i m e l i a 6 vitrinite r e f l e c t a n c e p r o f i l e a n d t e m p e r a t u r e g r a d i e n t (extrap. b o t t o m h o l e t e m p . 107.7 °C at 2264 m).

to a vitrinite reflectance value of 0.5% Ro max. The zone of peak oil generation with regard to thermal maturity lies in the reflectance range 0.70% to 0.90% Ro max., corresponding to a depth interval of 2000 to 2200 m. The base of this interval lies below the top of the pre-Permian in Merrimelia 6 but may occur within Jurassic sediments down the flanks of the Merrimelia High. The reflectance gradients at Merrimelia are higher than average for the Eromanga Basin but are similar to those recorded in the deeper part of the Nappamerri Trough, which is recognised as one of the hotter domains of the basin (Cook, 1981). The geothermal gradient in Merrimelia 6 is 41 °C km 1 which is significantly above the world average for sedimentary basins, but is close to the norm for the Cooper Basin and the lower part of the Eromanga Basin. Temperatures corresponding to given reflectance values are higher than those for wells in the Nappamerri Trough (e.g. Burley 1). Since reflectance values are a function of time as well as temperature and pressure, the low reflectance/temperature relationship indicates a relatively recent rise in temperatures at Merrimelia. Reflectance values have yet to adjust to this temperature rise because of a delayed response. For this reason, reflectance values would have been significantly higher if current temperatures had existed since the Early Cretaceous, and it appears that a considerable rise in temperature occurred during the Late Tertiary. Oil generation in the Mesozoic sediments therefore, probably occurred soon after this late rise in temperature and the timing is favourable in relation to the development of structural traps. The organic content of the finer grained sediments throughout the Mesozoic section is variable in both type and quality. There is abundant organic matter which is predominantly of terrestrial, higher-plant origin. A number of horizons contain sufficient exinite to be regarded as potential sources of oil. The 'basal Jurassic' unit contains abundant exinite in association with coals which resemble those of the Poolowanna Formation of the Eromanga Basin to the west of the Birdsville Track Ridge, and are petrographically quite dissimilar from the coals of the Birkhead Formation. If coals are accepted as a potential source of oil, then the 'basal Jurassic' unit must be regarded as a good source rock. The overlying Hutton Sandstone contains little organic matter. Thin beds of siltstone within the formation contain abundant

194

O. J. W. B O W E R I N G & D. M . H A R R I S O N

exinite a n d may have m a d e a m i n o r c o n t r i b u t i o n to oil generation. C o a l s within the Birkhead F o r m a t i o n are rich in exinite which is also c o m m o n to a b u n d a n t in the finer grained clastics. At its level of maturation in Merrimelia 6 (0.5 to 0 . 7 % Ro m a x ) the B i r k h e a d F o r m a t i o n constitutes a g o o d to excellent source rock. Being d o m i n a n t l y a s a n d s t o n e facies, the organic c o n t e n t of the N a m u r Sandstone M e m b e r as a whole is low. However, this unit c o m m o n l y c o n t a i n s a considerable p r o p o r t i o n of siltstone a n d shale. T h e s e finer grained clastics within the N a m u r S a n d s t o n e M e m b e r c o n t a i n c o m m o n to a b u n d a n t exinite in Merrimelia 6 a n d they are 'early-mature'. T h e presence of fine micrinite in some of the vitrinite in one sample f r o m near the base of this unit is direct evidence of hydrocarbon generation, although the overall contribution of the N a m u r Sandstone Member to the volumes of oil generated is p r o b a b l y s u b o r d i n a t e to t h a t of the B i r k h e a d F o r m a t i o n . N o source rock data are yet available f r o m the M u r t a M e m b e r at Merrimelia. However, this unit is a proven source of oil elsewhere in the E r o m a n g a Basin, for example in the Dullingari Field ( M o u n t , 1981) a n d J a c k s o n Field ( A m b r o s e et al., 1982, this volume). The dispersed organic matter ( D O M )

of this unit is d o m i n a t e d by inertinite a n d exinite which includes sporinite, resinite a n d alginite. T h e level of m a t u r a t i o n of the M u r t a M e m b e r c o r r e s p o n d s to the t o p of the 'oil w i n d o w ' (i.e. the onset of oil g e n e r a t i o n ) a n d it is likely t h a t this unit has provided the source f o r at least s o m e of the oil within the N a m u r S a n d s t o n e M e m b e r at M e r r i m e l i a . T h e 'Transition beds' c o n t a i n a high p r o p o r t i o n of finer grained lithologies with c o m m o n to a b u n d a n t exinite derived f r o m b o t h higher plant material a n d p h y t o p l a n k t o n . T h e f o r m a t i o n is regarded as having g o o d oil potential a n d is marginally m a t u r e for oil at Merrimelia. It is likely to be more t h e r m a l l y m a t u r e d o w n the f l a n k s of the structure a n d p r o b a b l y occurs within the 'oil window'. T h e overlying Cretaceous m a r i n e section is i m m a t u r e for oil generation a n d its organic c o n t e n t is such that its potential for oil generation is m o d e r a t e to poor. O n e exception is the Toolebuc Formation which would be a good oil source rock if mature (Cook, 1981).

ACKNOWLEDGEMENTS T h i s p a p e r is published with the p e r m i s s i o n of Alliance Oil D e v e l o p m e n t N.L., Delhi P e t r o l e u m Pty Ltd, a n d other p a r t n e r s in the M e r r i m e l i a - I n n a m i n c k a F a r m o u t Block in P.E.L.s 5 & 6, S o u t h Australia.

REFERENCES AEROSERVICES LTD, 1961: Innamincka-Betoota Delhi Pet. Pty Ltd (unpubl.).

aeromagnetic survey.

BATTERSBY, D. G., 1976: Cooper Basin gas and oil fields; in Knight, C. L. (ed.) Economic geology of Australia and Papua New Guinea—3. Petroleum. Aust. Inst. Min. Metall. Monogr. Ser. 7,

321-68.

BL.AKE, R., 1982: Review of geology and hydrocarbon Merrimelia-Innamincka Block, PEL 5 & 6, South Alliance Pet. Aust. N.L. (unpubl.).

MANCKTELOW, N., 1979: Structure and tectonics of the Cooper Basin. Delhi Pet. Pty Ltd (unpubl.).. M O O R E , P. S. & P I T T , G . M . , 1 9 8 2 : C r e t a c e o u s o f t h e s o u t h w e s t e r n

prospects, Australia.

CAMERON, J. C., 1961: Some aspects of sub-surface geology. A PEA J., I, 7 - 1 2 . COOK A. C., 1981: Organic petrology of a suite of samples Merrimelia 6. Delhi Pet. Pty Ltd (unpubl.).

from

DELHI PETROLEUM PTY LTD, 1978: Final report Oonabrinta Seismic Survey PEL 5 and 6, SA. Delhi Pet. Pty Ltd {unpubl.). DELHI PETROLEUM PTY LTD, 1979: Final report Moonlight Seismic Survey. Delhi Pet. Pty Ltd {unpubl.).

Flat

DELHI PETROLEUM PTY LTD, 1980: Final report Karawinnie Seismic Survey. Delhi Pet. Pty Ltd {unpubl.). DELHI PETROLEUM PTY LTD, 1981: Final report Namooka Survey. Delhi Pet. Pty Ltd {unpubl.).

LAWS, R. A., 1977: The Cooper Basin in Queensland; Petroleum in Queensland: a stock take for the future. Pet. Explor. Soc. Aust., Brisbane.

Seismic

GAUSDEN, J., 1981: Prospect evaluations Merrimelia-Innamincka Block PEL 5 and 6 South Australia. Alliance Pet. Aust. N.L. (unpubl.). KANTSI LR, A. J. & COOK, A. C., 1979: Rank variation in the Cooper and Eromanga Basins central Australia. Delhi Pet. Pty Ltd (unpubl.). KAPEL, A. J., 1972: The geology of the Patchawarra area, Cooper Basin. A PEA J., 12(1), 53-7.

Eromanga Basin: stratigraphy, facies variations and petroleum potential; in Moore, P. S. & Mount; T. J., (compilers) Eromanga Basin Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide. MOUNT, T. J., 1981: Dullingari North 1: an oil discovery in the Murta Member of the Eromanga Basin. A PEA J., 21(1), 71-7. RUMPH, B., 1978: Regional gravity and magnetic data of the central Eromanga Basin area: implications on the crustal structure, regional geology and tectonic history. M.Sc. thesis, Univ. Sydney {unpubl.). SPRIGG, R. C., 1961: On the structural evolution of the Great Artesian Basin. A PEA J., 1, 37-56. THORNTON, R. C. N., 1978: Regional lithofacies and palaeogeography of the Gidgealpa Group. A PEA J., 18(1), 52-63. UNITED

GEOPHYSICAL

CORP.

1963a:

McGregor

Range

Seismic

Survey. Delhi Pet. Pty Ltd (unpubl.). UNITED GEOPHYSICAL C O R P .

1 9 6 3 b : Diamantina

River—Gregory

Range Seismic Survey. Delhi Pet. Pty Ltd (unpubl.). UNITED GEOPHYSICAL CORP. 1964: Coopers Creek Seismic Survey. Delhi Pet. Pty Ltd (unpubl.). VALE, K. R., 1980: Character of faulting—Merrimelia fault. Alliance Pet. Aust. N.L. (unpubl.). W O N G E L A GEOPHYSICAL C O . PTY L T D , 1965:

Gravity Survey. Delhi Pet. Pty Ltd

(unpubl.).

NW bounding Strzelecki—Cooper

Geological Society of Australia Special Publication No. 12, 195-201

The Birkhead Formation—a Jurassic petroleum reservoir I. M. Paton

SANTOS Ltd, 39 Grenfell Street, Adelaide, S.A. 5000.

ABSTRACT

The Birkhead Formation is a fluvial, lacustrine and coal-swamp deposit forming the lower part of the Middle to Late Jurassic Injune Creek Group which is flanked conformably above and below by higher energy braided-fluvial stream clastics. Overlying the Cooper Basin region the Birkhead Formation ranges up to 120 metres in thickness. The formation occurs over a significant portion of the Eromanga Basin although it is notably absent west of the Birdsville Track Ridge in South Australia. Recent major oil recoveries and flows in the Moorari, Strzelecki, Big Lake and Merrimelia Fields have shown the Birkhead Formation to include an important oil reservoir target in the Eromanga Basin. Although all these discoveries to date have been defined on structural highs, local stratigraphic development and distribution of reservoirs will play a dominant role in their field development. The Birkhead Formation overlying the South Australian sector of the Cooper Basin can be divided into three major units. Shoestring geometry, low-energy, fluviatile channels, and generally poor to fair reservoir quality of the sands, make exploration for Birkhead Formation structural-stratigraphic traps in their own right a high risk. However, in combination with the development of other Eromanga Basin reservoirs and/or Permian targets, the Birkhead Formation will remain an important secondary reservoir target.

INTRODUCTION

This paper outlines the results of a preliminary study undertaken on the regional geology and petroleum potential of the Birkhead Formation in that part of the Eromanga Basin overlying the southern Cooper Basin. The Jurassic Birkhead Formation was first formally defined by Exon (1966). The formation was described from the intersection in Amoseas Westbourne 1 as a lacustrine to fluviatile sequence of grey and brown siltstone, mudstone, fine brown labile sandstone with minor coal. The Birkhead Formation (Fig. 1) occurs over a significant portion of the Eromanga Basin although it is notably absent west of the Birdsville Track Ridge in South Australia. In the Cooper Basin region the Birkhead Formation often exceeds 60 m and ranges up to 120 m in thickness (Fig. 2). Historically the formation has been placed low in prospectivity as a Jurassic petroleum reservoir understandably due to the predominantly shaly nature of the sediments. However, recent drilling and testing has seen oil flows and significant oil recovery in drill-pipe in Moorari 3 and 4, Nappacoongee 2, Strzelecki 3 and 7, Merrimelia 9 and Big Lake 26. All these oil discoveries (Fig. 3) in the Birkhead Formation have been located over the Cooper Basin region.

STRATIGRAPHY

The Birkhead Formation is the lower, organic-rich shaly unit of the Middle Jurassic Injune Creek Group (Exon, 1966). This group of sediments forms the central cycle of Jurassic Eromanga Basin sedimentation. In the Surat Basin this lower unit of the Injune Creek Group is known as the Walloon Coal Measures. The Birkhead Formation is overlain conformably by the Adori Sandstone in the larger part of the Eromanga

Basin east of the Birdsville Track Ridge. However in that sector of the Eromanga Basin overlying the southern Cooper Basin, the Birkhead Formation conformably underlies the Namur Sandstone Member of the Mooga Formation. The upper contact with the Namur Sandstone Member may be sharp or transitional. The Adori Sandstone ranges in age from Middle to Late Jurassic (lower to upper J5-6 spore-pollen Zone). It is generally a thin sand deposited by a short-lived fluvial meandering stream phase, before lacustrine conditions resumed. The upper shale unit of the Injune Creek Group, known as the Westbourne Formation is assigned to the upper J5-6 spore-pollen Zone, of Late Jurassic age. The Birkhead Formation lies conformably on the massive Hutton Sandstone. The contact of the Birkhead with the Hutton Sandstone was commonly considered to be abrupt (Ambrose & Watts, 1980). However this study has shown that transitional contacts are not uncommon over the extent of the southern Cooper Basin region. The Hutton Sandstone was a higher energy, braided-stream deposit terminated upward by a blockage or restriction of drainage systems resulting in lower energy, coal swamp and fluvial deposition of the Birkhead Formation. The deposition of the Birkhead Formation marks the end of a major cycle of higher energy, fluvial sedimentary deposition in the Eromanga Basin (Wiltshire, 1979). In the Surat Basin the Bundamba Group is stratigraphically equivalent to the Hutton/Precipice sedimentary phase in the Eromanga Basin. The Birkhead Formation ranges in age from Middle to Late Jurassic (J4 to lower J5-6 spore-pollen Zones). Primarily the Birkhead Formation over the principal part of the study area is of Middle to Late Jurassic age (Price, 1979; Wiltshire, 1979).

I. M. PATON

196 AGE

PEDIRKA

BASIN

SOUTHERN

RECENT TO L.CRET.

COOPER

NORTHERN

BASIN

COOPER

BASIN

SUPERFICIAL DEPOSITS AND WINTON FORMATION MACKUNDA FORMATION MARREE SUBGROUP

ALLARU MUDSTONE TOOLEBUC FORMATION EARLY CRETACEOUS WALLUMBILLA FORMATION

ALLARU MUDSTONE OODNADATTA FORMATION

TOOLEBUC FORMATION

WALLUMBILLA

COORIKIANA SST.

FORMATION

BULLDOG SHALE CADNA-OWIE FORMATION

MOOGA FM.

TRANSITION BEDS

ALGEBUCKINA SANDSTONE JURASSIC

MURTA MEMBER NAMUR SANDSTONE MEMBER

WESTBOURNE

BIRKHEAD

FORMATION

HUTTON SANDSTONE

P00L0WANNA FORMATION PERMIAN-TR

FORMATION

ADORI S A N D S T O N E

BASAL

PEDIRKA BASIN SEDIMENTS

JURASSIC

COOPER BASIN S E D I M E N T S

Fig. 1. Regional stratigraphic nomenclature, Eromanga Basin (Pedirka, Cooper Basin regions). WACKETT 1 NAPPACOONGEE 2 STRZELECKI 3 MOORARI 3 BIG LAKE 26 MERRIMELIA 9

DST RESULTS YEAR 1978 # D.S.T. 3 (1664m-1694m) GTS -062 MMCMD. at 530PSI, 11mm TC. Rec. 152m GCM. 1979 • D.S.T. 5 (1631m-1638m) GTS in 12 mins. at 6PSI Rec. 60m Oil, 914m GCW. 1979 • D.S.T. 2 (1653m - 1679m) Rec. 64m O, 94m GCOM. L980 • D.S.T. 2 (2149m-2168 m) GTS at RTSTM Rec. 62-3 bbls Oil, 17 bbls Mud. 1982 • D.S.T. 1 (1950m - 1967m) GTS at RTSTM. Q 150 BOPD. L982 • D.S.T. 5 (1863m-1884m) GTS at bbls RTSTM Rec. 15 Oil, 51'API, 320m GCO.

Fig. 3. Major hydrocarbon discoveries in the Birkhead Formation.

80 100KM.

• WELL CONTOUR INTERVAL 15m.

Fig. 2. Southern Eromanga Basin, Birkhead Formation isopach map.

been misidentified as the Birkhead Formation within that area. There is no doubt that the Injune Creek Group and within it the Birkhead Formation have thinned in the southern Cooper Basin region, however wider correlation indicates that the Mooga Namur Sandstone is equivalent to the Adori Sandstone and Westbourne Formation which are developed in the northern Cooper Basin.

However the basal sector of the Birkhead Formation in RESERVOIR DEVELOPMENT WITHIN THE Wackett 1 proved to be of J4 age. Thus a limited hiatus to BIRKHEAD FORMATION Birkhead deposition has been noted from the palynology of The Birkhead Formation ranges up to 120 metres in Wackett 1. During the Late Jurassic the Birdsville Track Ridge thickness in the Cooper Basin region, generally thickening was still exerting a dominant effect on depositional patterns in a northeast direction (Fig. 2). The sand content varies within the Eromanga Basin, since the massive Middle to Late generally from 10 to 40 per cent over this area with the average Jurassic Algebuckina Sandstone interval correlates with an about 20 per cent. The main trend in the sand content alternating sand/shale sequence east of the Birdsville Track percentage is that it generally increases to the southwest, the Ridge. more distal Birkhead facies. The Birkhead Formation in the Wiltshire (1979) believes that the Injune Creek Group southern Cooper Basin is herein stratigraphically divided into section has thinned over the southern Cooper Basin and has three units (Unit 1, 2 and 3) (Figs 4-6).

197

BIRKHEAD FORMATION Unit 1 of the Birkhead Formation is characterised by finegrained sandstone 3-12 metres thick interspersed with dark coloured shale and siltstone. The equal sand-shale ratio typifies that of a fluvial, meandering stream environment. Unit 1 differs from Unit 3 in that point-bar, upward-fining log signatures can be identified more readily with cut-andfill deposition more evident. The channel sandstones of Unit 1 are characterised by a fining-upward sequence consisting of lateral-accretion deposits followed by overbank fines (vertical accretion). Lag deposits are overlain by trough crossbedded sandstones which are in turn overlain by ripple cross-

laminated fine-grained sandstone and/or horizontally laminated, fine-grained sandstone. The facies sequence continues with vertical accretion deposits introduced at flood stage. These overbank deposits are characterised by medium to fine-grained dark grey siltstones and shales with minor coal stringers and abundant root channels evident. A number of Birkhead Unit 1 cores, have been taken in the Strzelecki Field (Strzelecki 4 and 5) and the Moorari Field (Moorari 3 and 4) which together reveal the above generalised sequence. Unit 1 cores in Gidgealpa 1 and Moorari 4 reveal within the channel sequence intraformational shale clast horizons. These probably formed as a result of slumping of levee deposits, on the convex side of a meander, into active channels (Ambrose & Watts, 1980). Most of the major oil flows and recoveries from the Birkhead Formation to date have been from Unit 1 channel sandstones. Generally the field occurrences of these sandstones are within shoestring (Big Lake and Merrimelia) to several adjacent fluvial point-bar deposits (Strzelecki and Moorari). Examples of oilfields within such fluvial sandbodies are common. One such example is the Waltersburg Formation in Saline County, Illinois (Dickey, 1979). This is a channel sand about 0.6 kilometres wide and up to 12 metres thick. The width of the sandbody at Waltersburg as in Strzelecki and Moorari suggests that it is not a single channel but a series of adjacent point bars which piled against one another forming swells and swales (Figs 7, 8). The development of

Fig. 4. Birkhead Formation stratigraphic profile. C00NGIE

MOORARI

MERRIMELIA

BIG LAKE

STRZELECKI

MUNKARIE

MT. HOWITT

INGELLA

Fig. 5. Stratigraphic section A-A', Birkhead Formation, southern Eromanga Basin.

MULGA

KIDMAN

BURKE

WACKETT

TARTULLA

Fig. 6. Stratigraphic section A-A', Birkhead Formation, southern Eromanga Basin.

198

I. M. PATON STRZELECKI

Om L

55-8 =i 56-0 56-2 :

30m

DST

STRZELECKI

^ . ,8,

39m 3

REC.1109mMUD/QIL

STRZELECKI

40m

43m

EMULSION

DST

NGTS

REC.64m.0IL

.LOCALITY MAP

LOCATION

Fig. 7. Stratigraphic section A-A', Birkhead Formation, Strzelecki Field, southern Eromanga Basin.

MOORAR!

MOORARI

,70-9

LOCATION A

140° 10 I

•^MOORARI

2152m-2163m GTS at RTSTM in 51m Rec 13 bbls 51° APIO

D.S.T. 1. 2152m - 2163m GTS at RTSTM in 54mins. Rec. 31 bbls 0 11 bblsGCM.

D.S.T. 2. 2149m-2168m GTS at RTSTM in 20 mins. Rec 62 3 bbls 50 2°API0

D.S.T. 2. 2150m-2163m GTS at RTSTM in 80mlns. Rec.47 bbls 0 11 bbls OandGCM.

D.S.T. 3

D.S.T. 3. 2151m-2198m GTS at RTSTM in 53min Rec 80 bbls. 0 4 bbls OandGCM

2170m-2180m NGTS Rec 1 5m M

Fig. 8. Stratigraphic section A-A', Birkhead Formation, Moorari Field, southern Eromanga Basin.

the principal Birkhead Formation oil reservoir in the Big Lake Field (Fig. 9) reveals a channel complex between two paleohighs (Fig. 10). Petrographic analysis has revealed that sands within Unit 1 vary widely in reservoir quality. The main oil reservoir in Moorari 4 shows the sandstone to be medium-grained and slightly argillaceous. Quartz overgrowths welding the grains are common. The sandstone has occasional lithic, feldspathic and more common micaceous grains. The sandstone has welldeveloped intergranular pore space. Sandstones adjacent to the main oil reservoir are of much poorer quality with little primary pore space and with much argillaceous matrix and

quartz overgrowth development. Unit 1 sands from Big Lake 26 and Strzelecki 4 have been examined petrographically and are quite argillaceous fine- to medium-grained sandstones with poor primary void space development. It is clear that Unit 1 sandstones have generally better developed reservoir qualities than those in Units 2 or 3, which generally are tight with little to no void space. Argillaceous matrix filling pore space within the fine- to medium-grained sandstones is the principal reason for generally poor reservoir development. Unit 1 sandstones, although quite varied in quality, are generally cleaner at the base of each fluvial pointbar cycle.

BIRKHEAD FORMATION

199

Fig. 9: Stratigraphic section A-A\ Birkhead Formation, Big Lake Field, southern Eromanga Basin.

Unit 2 represents the lacustrine and coal-swamp deposition typically recognised as Birkhead Formation. It is characterised by thin, fine-grained sandstones and siltstones interlaminated with carbonaceous shales and thin coals. The sediments are characterised by low-energy, aqueous, fine-grained laminated sediments. It is to be noted however that no terminal hypersaline facies is evident in Unit 2, indicating that the

environment lacked aridity and the depositional environment was surrounded by coal-swamps and lush vegetation. The preservation of organic matter and absence of benthos, suggest that the deeper parts of the lacustrine environment were anaerobic and stagnant. Unit 2 is overlain and underlain by higher energy fluvial deposits in Units 3 and 1, and represents a period of lower energy, low clastic influx. Unit

200

I. M. PATON

2 is indicative of the regressive nature of the lacustrine and coal swamp environment. Thus fine-grained, laminated shales and siltstones interfinger with marginal fluvial and coalswamp deposits. Dispersed organic matter (DOM) is dominated by vitrinite indicating that Unit 2 is a typical microlithology of forest-swamp facies (Stach, 1975). Exinitepoor clarites of the Birkhead Formation were formed from forest litter (Mclntyre, 1980). The high proportion of vitrinite relative to inertinite indicates rapid deposition and a high groundwater level preventing oxidation by the atmosphere. Existence of anoxic conditions is supported by the common occurrence of disseminated sulphides. In summary, it is suggested that Unit 2 of the Birkhead Formation was deposited in a lacustrine environment with associated forest swamps, open marshes and exposed swamps. The land surface was probably very close to sea level with only slight tectonic movements required to create a mass of braided streams or quiet-water lacustrine conditions. It is Unit 2 that Nugent (1969) sees as typical Birkhead Formation, representing a period of subsidence resulting in swamping of the fluvial environment. Unit 3 is restricted in development in the southern Cooper Basin to the Patchawarra Central Block, and has also been recognised in the vicinity of the Kerna, Dullingari, Daralingie and Strzelecki Fields. Sands are generally 3-10 m thick and have a blocky to point-bar log signature. The sand-shale ratio in this unit is approximately equal and is indicative of a meandering-fluvial depositional environment. The sands are generally fine-grained and choked with clay matrix typical of low-energy meandering stream deposits. Channel-fill sequences are mostly fine-grained due to abrupt channel abandonment in meander-neck cutoff. A 3 metre core in Dullingari 1 consists in part of fine-grained rippled sandstone. Sedimentary structures include ripples, cut-and-fill features and minor slumps. The basal part of the core consists of medium-grained, tabular cross-bedded sandstone with channel-lag chert pebbles.

SEISMIC, SOURCE, SEAL AND TRAP NATURE OF THE BIRKHEAD FORMATION

Previously the Birkhead Formation has been of principal interest to oil explorers not as a potential reservoir in its own right but because of its mappable seismic character, seal to underlying Hutton Sandstone oil reservoirs and its fair to good source rock characteristics. The Birkhead Formation over much of the southern Cooper Basin area is marked by a sharp decrease in sonic velocity. This sharp decrease in velocity is mirrored as a marked peak in normal polarity seismic display (or a mappable trough in reverse-polarity display). This seismic character aids structural mapping of the thick homogeneous Hutton Sandstone sequence where seismic character definition is often very poor. However, perhaps the principal economic interest of the Birkhead Formation is its role as a seal to the underlying Hutton Sandstone oil reservoirs as seen in the Strzelecki, Merrimelia and Jackson Fields. The Birkhead Formation has been varyingly described by recent writers as of fair to good source rock character. In the southern Cooper Basin it contains both Type II and Type III kerogen. Type II kerogen consists of spores, cuticle and resin, while vitrinite is represented by Type III. Mudge (1980) and Mclntyre (1980) studied samples taken from selected wells in the southern Cooper Basin and consider the Birkhead Formation to have fair source-rock characteristics. Vitrinite

reflectivity taken by Mudge (1980) in Gidgealpa 15 and 16, Kudrieke 1, Tirrawarra 2 and Merrimelia 5 indicate the start of oil generation from Type I to Type II kerogen, but are considered too low to allow gas generation. The Dispersed Organic Matter (DOM) is dominated by vitrinite and exinite with inertinite being sporadic in occurrence. The coals are vitrite and clarite rich. Thomas (1982) is somewhat more optimistic about the oil potential of rocks dominated by terrestrial DOM and quotes the Gippsland Basin as an example. Substantial fluctuations in geothermal gradients in the southwest Eromanga Basin have been noted (see e.g. Pitt, this volume). Further, it has been considered that the Jurassic sediments are good potential source rocks for oil as well as gas (Poll, 1981). Although the depth and span of the hydrocarbon generation window varies over the southern Cooper Basin it cannot be assumed that all Birkhead Formation gas discoveries have been sourced in situ. Wackett 1 Birkhead gas represents a good case for migration from a deeper Permian source, as the structure is a sharp horst-like feature with faulting extending to the Jurassic. However, based on source-rock evidence and extensive common oil shows in well cuttings, the Birkhead Formation shales must be considered to be good potential source rocks for oil and gas. The Birkhead Formation with fair to good source characteristics will ensure that any development of reservoir quality sands in the formation will, with the proper structural setting and timing of migration provide all the ingredients for an oil or gas pool. Indeed the Birkhead Formation may also provide the oil source for the Namur and Hutton reservoirs lying conformably above and below it. All Birkhead Formation oil and gas discoveries to date have been within structural-stratigraphic traps. No discovery has been located outside local structural closure. Most Mesozoic structures are monoclinal or anticlinal drape features over Permian horsts or linear dislocations, with subsequent Tertiary structuring. Some element of growth and thinning in the Jurassic has been evident in most Jurassic discoveries over the southern Cooper Basin. This situation, coupled with the lack of oil or gas accumulations in very late Tertiary structures (Poll, 1981), suggests that generation and accumulation must have occurred in a relatively short interval after Jurassic deposition. This of course clashes with accepted maturation and generation thinking, but will not be pursued further here as it lies outside the scope of this discussion. The Moorari, Big Lake, Merrimelia and Strzelecki discoveries within the Birkhead Formation have emphasised that the shoestring or stacked channel sand nature of the reservoirs in Unit 1 necessitates stratigraphic modelling and detailed reservoir mapping to appraise and develop such oil pools. Development of more substantial oil and gas reservoirs above or below the Birkhead Formation can be planned to coincide with appraisal and development of the Birkhead reservoirs. Indeed, the channel nature of the sandstones makes the appraisal of these reservoirs in their own right a high risk. Successful further exploration for Birkhead Formation oil reservoirs can proceed in accord with exploration for the better-known and more substantial Hutton Sandstone and Namur Sandstone Member reservoirs. The structural prospect generation of all of these major reservoirs involves the same methodology. What may be applied to the ranking of the Birkhead Formation target, however, is a study of the stratigraphic development of Unit 1 and Unit 3 sandstones. Future development of Permian gas fields, for example, can be complemented with exploration for structurally high Unit 1 sandstones of the Birkhead Formation. Sand development in early appraisal wells on Permian Fields such as Daralingie

BIRKHEAD FORMATION

201

show similar wireline log character and stratigraphic sequence to known Birkhead Formation oil discoveries.

southern sector of the Cooper Basin the lowermost section (Unit 1) of the Birkhead Formation is a fluviatile sequence which generally contains the best reservoirs. Reservoir quality of the fluvial channel sandstones in the Birkhead Formation CONCLUSIONS varies from poor to fair with varying amounts of argillaceous Previously, the Birkhead Formation has been of principal matrix and siliceous overgrowth. Detailed stratigraphic interest to oil explorers because of its seismically mappable mapping of Birkhead channel sandstones is required to ensure acoustic-impedance character, its potential as a seal to that the appraisal and development of Birkhead Formation underlying Hutton Sandstone oil reservoirs, and its fair to oil reservoirs will be undertaken in the optimum manner. The excellent source-rock characteristics. Recent oil recoveries and channel-sand nature and generally poor to fair reservoir flows in such wells as Moorari 3, Nappacoongee 2, Merrimelia quality of the sandstones make exploration for Birkhead 9, Big Lake 26 and Strzelecki 3 have shown the Birkhead Formation structural-stratigraphic traps in their own right a Formation to be ranked as one of the important secondary high risk. However, in combination with the development and oil reservoirs in the Eromanga Basin sequence. Further, the exploration of other Eromanga Basin reservoirs or Permian Birkhead Formation Shales may provide the source for targets the Birkhead will remain an important secondary hydrocarbons trapped in the enclosed sandstones. Over the reservoir objective.

AMBROSE, G. J. & WATTS, T. R., 1980: Progress report

REFERENCES

Eromanga

NUGENT, O. W., 1969: Sedimentation and petroleum potential of the

of the Jurassic and Lower Cretaceous sediments from five wells in the Eromanga Basin, South Australia. B.Sc. (Hons) thesis, Univ. Adelaide {unpubl.). MUDGE, W. J., 1980: The hydrocarbon potential and environmental interpretations of the Jurassic-Lower Cretaceous section in five wells in the Eromanga Basin. B.Sc. (Hons) thesis, Univ. Adelaide {unpubl.).

POLL, J., 1981: The significance of the southwest Eromanga Basin oil and gas discoveries (central Australia). APEA J. 21(2), 33-8. PRICE, P. L., 1979: Palynological laboratory report No. 13/112; in Vegh, E. & Skerman, W. R. Wackett no. 1 well completion report. Delhi Petroleum Pty Ltd (unpubl.). STACH, E., 1975: Coal petrology. Gebruder Borntraeger, Berlin. THOMAS, B. M., 1982: Land-plant source rocks for oil and their significance in Australian basins. APEA J. 20. 164-76. WILTSHIRE, M. J., 1979: Eromanga Basin, Australia: petroleum prospects. Wiltshire Geological Services {unpubl.).

Basin study. S. Aust. Dep. Mines Ener. (unpubl.). DICKEY, P. A., 1979: Petroleum development geology. Pet. Publ. Co., Tulsa. EXON, N. E, 1966: Revised Jurassic to Lower Cretaceous stratigraphy in the southwest Eromanga Basin, Queensland. Qld Gov. Min. J. 67, 232-8. MCINTYRE, S., 1980: The hydrocarbon potential

Jurassic sequence in the south-western Great Artesian Basin.

A PEA J. 9(1), 97-107.

Geological Society of Australia Special Publication No. 12, 203-219

The nature and significance of the organic facies in the Eromanga Basin A. C. Cook

Department of Geology, University of Wollongong, P.O. Box 1144, Wollongong, N.S.W. 2500.

ABSTRACT

Organic matter in sedimentary rocks of the Eromanga Basin is dominated by coals and coal-related organic matter, but a significant part of the sequence contains a component of marinesourced organic matter. The coals and the coal-related organic matter show four major facies. These are: (1) Poolowanna: Coals are inertinite-rich but contain abundant exinite. Exinite contents are typically in the range 5-10% with cutinite equal to, or dominant over, sporinite and resinite. Associated sediments have a high content of inertinite but vitrinite is also abundant. (2) Walloon: The coals typically contain little or no inertinite and exceptionally large amounts of exinite. Suberinite is the most abundant exinite maceral but cutinite and resinite are also unusually abundant. Related sedimentary rocks also contain abundant vitrinite and exinite. This facies is characteristic of the Birkhead Formation but similar assemblages occur in some of the overlying Jurassic units. (3) Mooga: These coals are similar in many respects to the Walloon-type coals but contain significant amounts of inertinite and, in some cases, bituminite. (4) Winton: These coals contain abundant band vitrinite but are characterized by the presence of small amounts of spore-rich duroclarite. The Poolowanna/basal Jurassic suite of coals shows some similarities to coals in the Permian Toolachee Formation of the Cooper Basin but is distinguished by the abundance of resinite and cutinite and the paucity of Botryococcus-related alginite. Marine units in the lower part of the Cretaceous contain marine phytoplankton and comprise a fifth facies. Some of the organic matter in most of these units is of higher plant origin. The marine-sourced components occur as alginite (chiefly lamalginite) and as bituminite. The oil shales of the Toolebuc Formation form a 'type example' of this facies. Organic matter in the Jurassic part of the sequence is the probable source for at least some of the oil reservoired in the Mesozoic. The development of these five facies is significant in relation to: (1) the amount of organic matter present in the sequence; (2) the specific hydrocarbon yield of the organic matter and the type of hydrocarbons which can be generated; and (3) the level of maturation at which significant hydrocarbon generation occurs. In particular, the presence of bituminite, suberinite or some types of vitrinite results in significant oil generation commencing at about 0.45% vitrinite reflectance.

INTRODUCTION

The existence of oil-mature sections in parts of the Eromanga Basin section overlying the Cooper Basin was noted by Kantsler et al. in 1978 and in the same year oil was discovered in the Eromanga section during the drilling of a step-out well on a Permian gas reservoir. This discovery has been followed by a number of further oil discoveries and oil has flowed from five major horizons. The fields lie in a number of different geological settings. The oils are typically light, paraffinic and waxy. Identification of the sources of these oils awaits a fuller evaluation of a number of aspects including source-rock type, maturation level and geochemistry, oil chemistry, and hydrocarbon migration and trapping mechanisms. It is probable, however, that some of the oil is

generated from units within the Eromanga Basin sequence rather than from the underlying Cooper Basin or Pedirka Basin sequences (see Kantsler et al1982, 1983). A number of distinctive assemblages of organic matter (organic facies; Rogers, 1980) are characteristic of parts of the Eromanga Basin sequence. These form a basis for the systematic study of the source-potential of the sequences.

ORGANIC PETROLOGY

Organic petrology is the name given to the study, by penological methods, of organic matter occurring within rocks. The techniques largely derive from those of coal petrology (I.C.C.P., 1963, 1971, 1975) but have been extended to the study of organic matter in sedimentary and low grade

204

A. C. COOK

metamorphic rocks (Alpern, 1970; Teichmuller, 1971; Teichmuller & Teichmuller, 1981; Cook, 1982). Observation of polished sections in reflected white light using oil immersion lenses is supplemented by reflected light fluorescence-mode microscopy. Three major kinds of organic matter can be recognized in coals, and as dispersed organic matter in sedimentary rocks. These are set out in Table 1. Each maceral group can be divided into a number of macerals. Many parts of the sequence in the Eromanga Basin are within the brown coal rank range. In order to simplify descriptions, bituminous (hard coal) coal maceral names have generally been applied in this study to organic matter irrespective of rank. The presence and abundance of maceral groups or macerals, together with maceral associations, can be used to characterize the organic facies (Rogers, 1980) and thus the source potential of sequences in sedimentary basins. Cook (1975, 1981) has described some of the facies found in Australian coals and related these to floral successions, climate and tectonosedimentary controls and to the utilization behaviour of the coals and the hydrocarbon generation potential of the associated sequences. Retallack (1981) has indicated the importance of understanding palaeosols in developing interpretations of climates and floral assemblages. Smyth (1979, 1983) examined the relationship of the petrographic composition of coal to that of the dispersed organic matter (d.o.m.) in the interseam sediments. Smith (1981) suggested that three organic facies can be recognized in the coal measures of the Gippsland and Bass Basins and that the development of these facies exerts a control on hydrocarbon generation in these basins. Teichmuller (Teichmuller & Teichmuller, 1981; Teichmuller in Stach, 1982) has summarized a number of studies on organic petrology as they relate to source-rock evaluation. Thomas (1982) combined petrological information concerning some major coal-bearing sequences in Australia with pyrolysis data and a geochemical study of some Australian oils to conclude that coal-bearing sequences can have high source potential. Both Thomas (1982) and Cook (1981) suggested that this may be higher in Mesozoic and Tertiary sequences because of the higher exinite productivity of the younger floras. Smyth & Saxby (1981) examined the relationship of the geochemical properties of oils and potential source-rocks in the Pedirka Basin to the petrology of the potential source-rocks. TABLE J . Maceral groups. Origin Maceral Group Vitrinite Humified tissue, typically derived from lignin cellulose and tannins Exinite

Sporopollenin, cutin, suberin, resins, waxes, essential oils, algal and phytoplankton tests.

Inertinite

Thermally or biochemically altered tissues, primary inertinite.

EXPERIMENTAL Suites of samples have been examined from about 90 wells (Fig. 1). Most of the samples were examined in the course of maturation studies and data on organic matter type were obtained during these studies. Some sampling programmes were, however, specifically designed to evaluate the type and abundance of organic matter. Most samples are of cuttings, but conventional and sidewall cores were available for some wells. The cuttings samples are relatively free from contamination and the marked changes in the appearance of organic matter down-section make the detection of cavings or recirculated cuttings relatively easy. Samples were mounted 'as received', and concentration of organic matter by gravity separation, flotation or acid treatment was generally not employed. Samples were embedded in cold-setting resin and polished using slightly modified coal-polishing techniques. Examination of 'wholerock' samples, rather than concentrates, provides more reliable information on both the identity and the associations of the organic components. Samples were examined in reflected white light and reflected fluorescence-mode (Cook & Kantsler, 1980). The Leitz Orthoplan microscope used in the majority of the work permits rapid alternation of illumination modes. Fluorescencemode observations were usually made using a BG3 excitation filter, a TK400 dichroic mirror and a K490 barrier filter. Vitrinite reflectance measurements were made on all samples that contained vitrinite using a method based on draft SAA 2486 (1981), modified to take account of the problems associated with the measurement of reflectance on dispersed organic matter (d.o.m.). Organic matter has been classified according to the StopesHeerlen system as expanded by the l.C.C.P. (1963, 1971, 1975), modified where necessary to be suitable for studies on d.o.m. In particular, alginite terminology is based on that of Hutton et al. (1980) and Cook et al. (1981). Semi-quantitative methods have been used for the study of maceral abundance because the percentage of macerals present is generally too low to permit precise estimation using point-count methods.' An improved method of visual estimation of d.o.m. abundance has recently been developed by S. Padmasiri (pers. comm.). Properties Reflectance range small and lying between those of exinite and inertinite. Fluorescence weak to absent. Reflectance below that of vitrinite, except at high rank where reflectance becomes equal to or greater than that of vitrinite. Fluorescence distinct to intense. Reflectance above that of vitrinite except at extremely high rank. No fluorescence

Macerals (partial list only) Telocollinite (vitrinite A) Desmocollinite (vitrinite B) Vitrodetrinite Sporinite Cutinite Suberinite Resinite Fluorinite Alginite (telalginite, lamalginite) Exsudatinite (secondary resins) Liptodetrinite Fusinite Semifusinite Inertodetrinite Macrinite Micrinite Sclerotinite

N.T. QLD

_NH S.A.

• •

•

• •

• **

• 28°00j

• « •

•

.• *

•

20|

data

point

Kilometres

•

_QLD_ N.S.W.

29°0cy

29°1S'

A. C. COOK

206

ORGANIC MATTER FACIES

The major division of organic matter facies in the Eromanga Basin sequence is between assemblages that contain a marine-sourced element and those that have only terrestrially derived organic matter (Fig. 2, Tables 2, 3, 4). Tables 2 and 3 relate respectively to the maceral and microlithotype abundance in the coals typical of the facies distinguished, the marine Cretaceous facies containing no coals. Table 4 summarizes the maceral abundance of the dispersed organic matter in the non-coal sedimentary rocks. The four coalrelated facies have been named after the stratigraphic unit in which they are most prominently developed. They are not restricted to the unit named and it is possible that one of the other facies may be developed within the named unit but such development is likely to be minor and localized. It may be possible to sub-divide the facies further but such sub-division is not considered to be useful at this stage. The Walloon facies is named after the Walloon Coal Measures rather than the Birkhead Formation because coal seams are a much more prominent feature of the former unit and their very characteristic maceral assemblage is already well known and described (e.g. Cook, 1975). Marine sequences are developed in the Lower Cretaceous. Marine phytoplankton are found rarely at other horizons. Some of these last occurrences may, however, represent

reworked material. Facies dominated by non-marine organic matter facies are present in the Triassic (Peera Peera and Nappamerri Formations), the Jurassic and the Cretaceous (chiefly the Winton Formation). In the marine units, organic matter is rarely abundant, but is very widely distributed. The Toolebuc Formation is the major marine unit in which organic matter is abundant. The non-marine organic matter typically occurs as coal seams or in coal-associated organic-rich lithologies. Where organic matter occurs in epiclastic rocks it is typically more abundant in the finer clastics but some notable exceptions occur and the organic matter can be preferentially associated with siltstones and, rarely, sandstones. The amount of organic matter in the sequences is typically higher in association with the non-marine sequences as compared with the marine sequences. The vertical distribution of the organic facies is indicated in Figure 2.

COAL AND COAL-ASSOCIATED FACIES

The coal and coal-associated organic matter can be divided into four major facies. They form a succession of facies so that progression from one facies to another is dominantly time-related and little evidence is so far available to demonstrate that the different facies are, to any significant extent, lateral variants. More detailed work may, however, show that lateral variations are of greater significance than

TABLE 2. Characteristics of coals in organic facies of Eromanga Basin. Abundance terms: rare <0.1%; sparse 0.1-0.5%; common 0 . 5 - 2 . 0 % ;

abundant 2.0-15%; major >15%.

Stratigraphic Range Coal Vitrinite Band vitrinite Suberinite Sporinite Cutinite Resinite Alginite Bituminite Inertinite

Poolowanna Peera Peera Fm. and Poolowanna Fm.

Facies Walloon Westbourne Fm. Birkhead Fm. Hutton Sst.

Mooga Mooga Fm. Murta Mbr Namur Mbr

Sparse to common Sparse Major Abundant to Major Rare to sparse Rare to sparse Major to abundant Major to abundant Common Common to abundant Common to abundant Sparse Common, locally Sparse abundant Rare Botryococcus-type Rare Botryococcus- Not noted type Sparse to common Rare to absent Sparse to absent Rare to absent Common to Abundant to major typically inertoabundant, macrinite macrinite common detrinite only and inertodetrinite common

Abundant Major Common Common Common Common to abundant Locally common

Marine Cretaceous Allaru Mdst. Toolebuc Fm. Wallumbilla Fm. Cadna-owie Fm. (Mooga) Absent

Winton Winton Fm. and some minor non-marine intervals below the Winton Fm. Common Major Major Common Common Sparse to common Sparse to common Very rare Botryococcus- type Rare to absent Common, inertodetrinite and semifusinite

TABLE 3. Microlithotype associations of coals in coal-bearing facies. Refer Table 2 for abundance terms.

Vitrite Clarite Trimacerites Durite Inertite

Poolowanna Major Abundant but exinite content low Major, exinite and inertinite contents high Locally abundant, exinite content low to high Abundant, massive semifusinite and macrinite common

Walloon Typically sparse Major, exinite content very high Rare to absent Absent Absent

Facies Mooga Sparse to abundant Major, exinite content very high Abundant, exinite content high Common, exinite content high to very high ?Absent

Winton Major Sparse apart from resinite/ vitrinite associations' Abundant, exinite content high Rare to absent Rare semifusinitedominated fusite

ORGANIC FACIES

207

ORGANIC FACIES IN THE EROMANGA BASIN AGE

FORMATION

FACIES

WfNTON

FACIES

MARINE

FACIES

CO D

o LU o <

—

T o o l e b u c -

—

Allaru

•

Wallumbilla.-r

vS^.:

IUJ 0C

o —

Transition

B e d s

' M u r t a MOOGA

FACIES

WALLOON

FACIES

Westbourne CL or : :*

• M*^ — ' • • Adori

T V •. . • • • •

POOLOWANNA

Fig. 2. Stratigraphic section indicating stratigraphic range of the organic f a d e s .

FACIES

208

A. C. COOK

TABLE 4. Characteristics of d.o.m. in organic facies of Eromanga Basin. Refer Table 2 for abundance terms. Facies Marine Cretaceous Mooga Walloon Poolowanna Allaru Mdst. Mooga Fm. Westbourne Fm. Peera Peera Fm. Toolebuc Fm. Murta Mbr Birkhead Fm. and Poolowanna Stratigraphic Wallumbilla Fm. Namur Mbr Hutton Sst. Fm. Range Cadna-owie Fm. (Mooga Fm.) Rare to absent Sparse to common Common to abundant Rare Vitrinite Absent Rare to sparse Rare to absent Band vitrinite Rare to absent Absent Rare Rare Suberinite Rare Rare Sparse Sparse Common Sporinite Cutinite

Sparse to common

Sparse

Rare

Resinite

Rare

Rare but locally common

Rare

Alginite

Absent

Rare Botryococcus- type

Not noted

Rare to absent Rare to absent typically inertodetrinite only

Rare to common Sparse to common, macrinite and inertodetrinite

Sparse resin-like bodies not of higher plant origin Rare tasmanitids Common to abundant lamalginite Rare to abundant Sparse to common

Bituminite Inertinite

Rare to absent Abundant macrinite common

Winton Winton Fm. and some minor nonmarine intervals below the Winton Fm. Common Rare Sparse Sparse to common Sparse to common Rare Very rare

Botryococcus-

type Rare to absent Common, inertodetrinite and semifusinite

presently appears to be the case, e.g. Smyth & Cook (1984). in the Nappamerri Formation are rich in inertinite. The exinite Some lateral transitions in organic matter type occur but more characteristics are generally more similar to the coals found detailed sampling control would be necessary to demonstrate in the Upper Permian Toolachee Formation with sporinite that this variation is at the broad level used to distinguish being dominant over all other exinite macerals. More detailed the facies described in this paper. studies are necessary to establish more fully the nature and affinities of organic matter in the Nappamerri Formation.

Poolowanna facies

Coals typically form an important part of the sequence within the Poolowanna facies and are rich in both inertinite and exinite (Table 2; Figs 3a, b, c, d). Large semifusinite phytoclasts and large masses of macrinite are commonly prominent. The macrinite is typically homogeneous and some occurrences are texturally very similar to vitrinite. Cutinite is a prominent component. The cutinite sheets are relatively extensive and thick, and fluorescence intensity is commonly high. Resinite is commonly present but is rarely abundant. Exinite is preferentially associated with inertinite, rather than with vitrinite (Table 3). Rare telalginite derived from Botryococcus-type algae also occurs. D.o.m. is abundant in many of the finer-grained epiclastics. Vitrinite is typically rare and inertinite is the dominant maceral in the d.o.m., with large homogeneous grains of macrinite being a distinctive component (Table 4). Exinite ranges from rare to abundant, consisting chiefly of sporinite. Recently, coal seams and associated organic matter have been found in sequences assigned to the Triassic Nappamerri Formation. Like the coals of the Poolowanna facies, the coals

Walloon facies

Coals are typically present in the Walloon facies but are seldom as prominent or abundantin the Eromanga Basin as they are in the equivalent horizons in the Surat Basin. The coals are finely layered and characterized by an abundance of exinite (Figs 4a, b). Inertinite is commonly absent but, where present, typically consists of rare, small, grains of inertodetrinite. Sporinite is typically common to abundant but the abundance of suberinite, cutinite and fluorinite/resinite is a characteristic feature of the Walloon facies. Telalginite derived from Botryococcus-type algae is present in some samples, but is not typical or abundant. Telocollinite is normally much less abundant than desmocollinite and the desmocollinite contains large amounts of suberinite. Suberinite is the major component of some layers and typically exceeds 20 per cent. Some suberinite occurs with collinite cell fillings but in many occurrences the suberinite laminae are tightly packed with no other macerals being present over thicknesses typically up to 0.03 mm. Liptodetrinite is a common component of the exinite in most clarites. It has a wide range

Figures 3 to 7 are fluorescence-mode (Fl.) and reflected white light photomicrographs (R.L.) of organic matter in sedimentary rocks from the Eromanga Basin. All photographs were taken using oil immersion lenses. Most of the samples were sectioned approximately perpendicular to bedding. Exceptions are noted in the captions. Fig. 3. a. Poolowanna facies. Resin-rich durite. Attrital resinite is relatively common in the coals in this sample as in most coals from the Poolowanna Beds. However, hydrocarbon extracts from similar samples do not appear to contain any major contribution from resins. Walkandi Formation, Poolowanna Trough, 2368 m, cuttings sample. R max 0.81%. Fl., field width 0.46 mm. b. As for a, but in reflected white light, c. Poolowanna facies. Cutinite-rich coal flanked.by a thick layer of semifusinite. Exinite fluorescence is relatively strong and the vitrinite shows weak but distinct fluorescence. The presence of abundant strongly fluorescing cutinite distinguishes Poolowanna facies coals from the coals in the Permian. The presence of abundant inertinite and the intensity of the cutinite fluorescence partly distinguish this facies from the overlying Walloon facies. The blue colour of the inertinite is due to show-through of the primary or excitation beam and is not a fluorescence effect. Basal Jurassic, Windorah Trough, 2047 m, cuttings sample. R max 0.86%. FL, field width 0.34 mm. d. As for c, but in reflected white light. v

ORGANIC FACIES

209

210 A. C. COOK

ORGANIC FACIES of fluorescence intensity but the presence of strongly fluorescing types suggests that much of the liptodetrinite may have affinities to resinite. Most Walloon-type coals have a high content of adventitious mineral matter. Shaly coals are common and have petrographic characteristics similar to those of the coals. The abundance of exinite and the paucity of inertinite render the Walloon facies very distinctive. Some systematic lateral variations have been noted. For example, telocollinite is relatively abundant in and near the Windorah Trough, whereas pyrite is present in some sections flanking the southern margin of the Nappamerri Trough (Fig. 1). D.o.m. is most abundant in association with the claystones and is typically dominated by sporinite and vitrinite. Detrital resinite and cutinite are locally common. Vitrinite phytoclasts in the d.o.m. are typically small. Inertinite is not normally present.

Mooga facies

Coals developed within the upper part of the Jurassic are similar in many respects to the Walloon facies but differ in that they contain a significant content of inertinite (Figs 4c, d; 5a, b). The inertinite occurs in trimacerite microlithotypes and in durite (Table 3). Inertodetrinite and lenses of semifusinite may both be present. Material referable to bituminite (Figs 4c; 5a) is present in some of the inertiniterich lithologies. Suberinite is abundant and has a similar aspect to that in the coals of the Walloon facies. Some of the ?bituminite may represent degraded suberinite. Sporinite and cutinite are present in significant amounts, but as with the Walloon facies fluorescence colours and intensities are typically relatively subdued. Exinite is commonly the most abundant maceral in the d.o.m. in coal-related epiclastics (Fig. 5c). Vitrinite is typically more abundant than inertinite. The inertinite is typically inertodetrinite. Sporinite and liptodetrinite are the most abundant forms of exinite but phytoplankton (lamalginite) are present in some samples and this facies probably shows transitions to the marine facies of the Cretaceous. Inertinite may be more abundant in sections with marine affinities as compared with those which are coal associated.

Winton facies

Winton Formation coals have layers of 'band' vitrinite as their major component. They are typically of low rank and texto-ulminite is the most abundant vitrinite (huminite) maceral present. Gelinite and textinite are less common. Small inclusions of resinite are typically present. Some of this resinite has a low reflectance and gives intense fluorescence colours but most of the resinite has a reflectance only marginally below that of the vitrinite and displays weak fluorescence. Suberinite occurs in some layers but is a much less prominent component as compared with the Walloon facies coals. Cutinite is relatively rare in the coals and where it occurs tends to be associated with lenses of claystone forming shaly coal. Alginite (telalginite derived from Botryococcus-type algae) occurs rarely in the coals. The trimacerites are a distinctive feature and are vitrinite-rich with sporinite being the dominant exinite maceral (Figs 5d, e). The inertinite comprises small

211

fragments of semifusinite in the trimacerites and some larger layers of semifusinite and, more rarely, fusinite. The cell walls in the semifusinite show a wide range of thickness with thickwalled semifusinite being a distinctive component. D.o.m. ranges widely in composition. In coal-related sedimentary rocks vitrinite is typically the most abundant maceral and significant amounts of sporinite and cutinite are usually present (Figs 6a, b). Lithologies which appear to have less affinity with coal seams have inertinite as the dominant maceral, exinite being typically sparse to common and vitrinite being rare.

MARINE CRETACEOUS FACIES

The marine sequence lacks the coals of the other facies and typically has a significant component of marine phytoplankton. Some marine units are probably present in the Mooga Formation but core samples are not usually available and the marine elements found in cuttings samples could represent cavings. The interval from the Cadna-owie Formation (Transition beds) to the base of the Winton Formation is dominantly marine. D.o.m. is typically sparse and inertinite the most abundant maceral. Vitrinite is normally rare and where common may be associated with terrestrial coal-bearing sequences. Sporinite is widely distributed but cutinite is more restricted in its distribution. Lamalginite derived from phytoplankton (Fig. 6c) is the most characteristic component but it is rarely either abundant or the major organic constituent. Telalginite is rare and effectively confined to the presence of (some) tasmanitids in the Toolebuc Formation. Bituminite is rare in most units but is abundant in the Toolebuc Formation. The d.o.m. in the Toolebuc Formation has been described briefly by Hutton et al. (1980) and by Cook et al. (1981). These writers used the term 'mixed oil shale'. A more detailed description of the nature and origin of the organic matter in the Toolebuc Formation is presented by Sherwood and Cook in this volume. Organic matter referable to bituminite is the most abundant maceral in the Toolebuc Formation (Figs 6c, d). It occurs as extensive thin (up to 0.03 mm) layers. The layers interfinger so that mineral-rich lenses occur in a framework of bituminite. The bituminite typically has a reflectance close to 0.20% which is similar to the desmocollinite in the coals of the Winton Formation and it can be mistaken for vitrinite. In thin section, it has a reddish transmission colour and clearly shows significant affinities with vitrinite. Its fluorescence intensity is typically weak but some moderately to strongly fluorescing inclusions are present. Prolonged irradiation produces positive alteration (Fig, 7a) but the extent of this alteration ranges from very weak to strong. Some micrinitization is normally present (Sherwood & Cook, 1986). Many workers have considered that the bottom waters of the Toolebuc sea were anoxic. Kauffmann (1981) working on the ecology of the petrologically similar Posidonienschiefer, presents an alternative model that also appears to be appropriate for the Toolebuc Formation. The oxic/anoxic boundary is considered to lie typically at, or just above, the sediment/water interface. Kauffmann suggests that an algal-

Fig. 4. a. Walloon facies. Clarite with abundant resinite, suberinite and cutinite. The coal is traversed by discordant veins of weakly fluorescing exsudatinite. Walloon Coal Measures, Surat Basin, 1652.5 m, cuttings sample. R max 0.57%. Fl., field width 0.44 mm. b. As for a, but in reflected white light, c. Mooga facies. Exinite-rich duroclarite. The exinite consists of abundant cutinite with some sporinite and resinite; The vitrinite groundmass has an anomalously low reflectance (Fig. 4d) and distinct fluorescence, possibly due to the presence of dispersed bituminite. The inertinite is prominent and is a major distinguishing feature from the Walloon facies. Mooga Formation, Eromanga Basin, 1617 m, cuttings sample. R max 0.49%. Fl., field width 0.34 mm. d. As for c, but in reflected white light. v

212

A. C. COOK

213

ORGANIC FACIES fungal mat, similar to those found in the present day Santa Barbara Basin, formed on or above' the sediment interface and controlled the position of the oxic/anoxic boundary. The differences between the oxic and anoxic bottom water models are summarized by Sherwood & Cook (1985). An algal-fungal mat could in part be the source of the bituminite. Its anastomosing form is consistent with an origin from an algal-fungal mat or from some of the larger marine algae. The proximity of the mat to the anoxic layer may increase the proportion of the algal components which form humic compounds on biochemical coalification as compared with alginite derived from planktonic organisms. This could account for the higher reflectance and lower fluorescence overall as compared with alginite derived from planktonic forms. The algae in the mat evidently do not contain strongly fluorescing components but, during fluorescence alteration, thin brighter structures become apparent and these may be formed from lipid-rich algal strands. If the above interpretation of the origin of the bituminite in the Toolebuc Formation is correct, this type of bituminite is essentially derived from algae but the optical properties are very different from those of both telalginite and lamalginite. The colour of the bituminite in thin section is similar to that of vitrinite. However, the reflectance and fluorescence characteristics of bituminite clearly distinguish it from vitrinite.

This interpretation of the origin of the bituminite in the Toolebuc Formation suggests that the presence of the bituminite-forming algal-fungal mat is the key feature that distinguishes the organic-rich facies of the Toolebuc Formation from the majority of the other marine Cretaceous units. Bituminite is found in other units but typically as isolated ?clasts rather than as anastomosing layers. The development of the algal-fungal mat probably contributes to the organic matter content of the sediment in two ways. The mat itself contributes directly to the organic matter content and, by keeping the oxic/anoxic boundary close to or above the sediment/water interface, increases the proportion of planktonic material that is preserved. Thus, the presence or absence of bituminite may be of considerable significance in indicating the development of marine facies rich in organic matter.

DISCUSSION Possible causes of the development of different facies At the coarsest scale of distinction, the organic matter facies is determined by whether sedimentation occurred under marine or non-marine conditions. The non-marine sequence shows a succession of four facies. Some of the changes are probably associated with the evolution of new plant taxa.

TABLE 5. Summary of fluorescence properties of vitrinite. Desmocollinite and precursors

Vitrinite . Telocollinite and precursors Reflectance R v max <0.45%

Typically none but some wood structures show fluorescence related to botanical structure

Some fluorescence at very low ranks

0.45% to 0.70%

Typically none

0.70% to 1.1%

In some cases green fluorescence from globules and streams emitted frpm cracks, positive alteration common

Dull .brown fluorescence, development of a green haze due to oil emission and positive alteration

>1.1%

Typically little or no fluorescence, but oil emission occurs from vitrinite up to about 1.3% reflectance

Little or no fluorescence, oil emission may result in the development of a green haze for vitrinite (up to about 1.5% reflectance)

TABLE 6. Maturation ranges for oil generation by macerals. Reflectance of co-existing vitrinite (R v max) Maceral

Initial generation

Inertinite

0.4

Main oil generation range

Effective oil dead-line approx. 0.8%

Relative oil yield Low but inertinite may be significant in relation to migration of oil

Resinite

0.4-0.45

0.5-0.8

1.0

High

Suberinite

0.45

0.5-0.8

1.0

Moderate to high

Bituminite

0.4

0.5-0.8

0.9

Moderate to high Moderate

Vitrinite

0.45

0.5-1.0

1.3

Sporinite and Cutinite

0.6

0.7-0.9

1.1

High

Alginite

0.7

0.75-0.95

Very high

Fig. 5. a. Mooga facies. Clarite containing sparse to common inertinite and abundant exinite. The exinite consists of sporinite, cutinite and ?resinite set in a matrix of vitrinite and ?bituminite. The reflectance of the vitrinite-like material is significantly below that of the telocollinite measured to give the mean maximum vitrinite reflectance for the sample. Mooga Formation, Eromanga Basin, 1050 m, cuttings sample. R v max 0.52%. Fl., field width 0.22 mm. b. As for a, but in reflected white light, c. Mooga facies. Claystone with abundant liptodetrinite and common sporinite (orange) and inertinite (black). Cadna--owie Formation (Transition beds), Eromanga Basin, Merrimelia—Innamincka structural high, 1522 m, cuttings sample. R v max 0.47%. Fl., field width 0.45 mm. d. Winton facies. Duroclarite in a section nearly parallel with bedding. Inertinite is abundant and comprises both semifusinite and inertodetrinite. Sporinite is well preserved and shows moderate to strong fluorescence. Winton Formation, Eromanga Basin, 320 m, cuttings sample. R v max 0.38%. Fl., field width 0.45 mm. e. As for d, but in reflected white light.

214 A. C. COOK

ORGANIC FACIES Thus, the abundance of resinite and suberinite in the Walloon coals is probably an example of type changes associated with the development of new floristic elements. The ratio of inertinite to vitrinite may have been influenced by the nature of the flora and in particular by the habitats colonized by various floras. It seems probable, however, that climate was a major control in relation to the abundance of inertinite. Periods of relatively dry conditions must have been characteristic of the environments in which the Poolowanna facies organic matter was preserved (see e.g. Moore, this volume). The abundance of macrinite may have been caused by the drying of extensively gelified humic material. The abundance of thick-walled strongly fluorescing cutinite may also be associated with a need for the vegetation to withstand dry periods. The abundance of inertinite in the Mooga facies may be a function of climatic change but could also be associated with a sedimentary setting that was relatively close to the sea. Smith (1981) has argued that the abundance of inertinite in some of the coals in the Latrobe Group and its virtual absence in others is partly related to climate but is also related to proximity to a marine environment. More alkaline conditions and accelerated oxidation of humic material occur in the marginal marine setting. The inertinite in the Winton facies coals again probably indicates periodic dry seasons. Inertinite is typically much more abundant than vitrinite in marine environments. The production of inertinite is favoured by reworking. Some reworking is inherent in the process of the inclusion of higher plant material in marine environments and in many marine rocks all the humic matter is converted to inertinite. A uniformly moist climate must have been associated with the formation of the Walloon facies coals. The moist climate did not, however, inhibit the breakdown of woody tissue and thick bands of telocollinite are relatively uncommon in most occurrences of the Walloon facies. The felted masses of suberinite must have resulted from intensive degradation of woody tissue. Again an analogy exists with the inertinite-poor (vitrinite-rich) facies of the Latrobe Group where attrital vitrinite types are much more abundant than telocollinite or its precursors. A contrast can be drawn with the Winton facies where wood structures are well-preserved but significant alteration of humic matter to inertinite occurred at times. The weak fluorescence properties of most of the cutinite in the Walloon facies coals may be due to early diagenesis or may be a function of the composition of the plant cuticles. Most of the cuticles in the Walloon coals are thin and, if the climate was uniformly moist, thick waxy cuticle would not be required to reduce water loss.

Significance of the organic facies in relation to hydrocarbon generation The organic facies present in a sequence may exert a control over hydrocarbon generation in relation to, at least, three aspects: (1) the abundance and type of organic matter present in the sequence control the amount and type of hydrocarbons generated; (2) the balance of organic matter present as coal seams compared with that occurring as d.o.m. may determine the efficiency of migration; and

215

(3) organic matter type influences the rank level at which hydrocarbon generation occurs. Organic matter is generally more abundant in the nonmarine sequences as compared with the marine sequences. Except in the Poolowanna facies, the general ratio of (vitrinite + exinite) to inertinite is more favourable for oil generation in the non-marine sequences. The Toolebuc Formation represents a major exception to this generalization. Absolute concentrations of exinite are generally highest in the coals, intermediate in the coal-associated d.o.m. and lowest in the marine sequences. However, the ratio exinite to vitrinite has the reverse order. Alginite is a significant component only in the Toolebuc Formation and in a few other horizons in the marine Cretaceous sequence. It is conventional wisdom that coals cannot generate liquid hydrocarbons because of their high internal surface area and high sorptive capacity. Petrological examinations suggest that coals may not behave according to these theories. Exsudatinite is a bitumen-like substance and small veins of this maceral are commonly found in coals over the reflectance range 0.45% to 1.2%. However, no direct correlation appears to exist between the abundance of exsudatinite and the proximity of oil reservoirs. This lack of correlation could nevertheless be attributed to the complexity of the processes of primary and secondary migration. Many coals bleed oil from cracks during examination in fluorescence-mode (Figs 7c, d). Further, it is unrealistic to draw a sharp distinction between coals and d.o.m. to develop the assumption that migration from d.o.m. is possible but impossible from coals. In most coal measures, a continuum exists from barren epiclastics through carbonaceous variants to sh^ly coal and coal. Where should the line be drawn between generative and non-generative lithologies and is such a concept realistic? Except in prolific source-rocks, such as the Toolebuc Formation or the Poisidonienschiefer, inertinite and vitrinite are typically more abundant than exinite. Liquid hydrocarbons migrating out of exinite should be absorbed just as efficiently by inertinite and vitrinite in d.o.m. as by the same macerals in a coal. If it is considered that coals cannot generate liquid hydrocarbons, the four coal-containing facies must be assessed as poor source-sequences. Acceptance of some, or all, of the arguments above means that the coal-related facies should be assessed as fair to very good potential sources for hydrocarbon liquids depending upon the amount of coaly material present. Hydrocarbons are generated during the coalification process as the organic matter is disproportionated into a hydrogenand aliphatic-rich fluid phase, and an aromatic residuum that becomes more extensively condensed. Liquids generation and preservation are considered to occur over the rank range corresponding to vitrinite reflectances of about 0.5% to 1.3%. There is a general agreement on the high rank limit to oil occurrence (the oil deadline or oil death line). Heroux et al. (1979) report a vitrinite reflectance range for the onset of oil generation from 0.3% to 0.5%, but many workers consider that significant generation does not occur until a reflectance of 0.6% is reached. The range 0.7% to 0.9% vitrinite reflectance is commonly cited as being the principal zone of oil generation.

Fig. 6. a. Winton facies. Claystone with abundant vitrinite and cutinite and sparse resinite and sporinite. The cutinite fluorescence ranges from intense yellow to weak orange brown. Section oblique to bedding. Winton Formation, Eromanga Basin, 320 m, cuttings sample, R v max 0.38%. Fl., field width 0.45 mm. b. As for a, but in reflected white light, c. Bituminite showing weak (dull orange to brown) or no fluorescence occurring with phytoplankton-derived lamalginite (bright orange to yellow). Toolebuc Formation, Eromanga Basin, 1312 m, cuttings, sample. R v max 0.54%. Fl., field width 0.34 mm. d. As for c, but in reflected white light. Pyrite is abundant occurring as framboidal aggregates and fine specks. Some of the finely granular texture of the bituminite is due to the presence of micrinite.

216 A. C. COOK

ORGANIC FACIES The assessment of the rank level at which oil generation occurs is based inter alia upon: (1) the rank level found in presumed, source rocks and petroleum reservoirs (Teichmuller, 1982, p. 400); (2) correlation of the ratio of extractable hydrocarbons to organic carbon with vitrinite reflectance (Teichmiiller, 1974); and (3) changes in the textures and optical properties of macerals (Teichmuller, 1974). The possibility of vertical migration makes it difficult to set lower limits for the range over which oil generation occurs. Extraction studies face the problem that the material being examined is presumed to have sourced migrated petroleum as well as the retained hydrocarbons that can be extracted. Therefore the rise in extractable matter may relate as much to migration mechanisms as to the amounts of hydrocarbons that have been generated. The rise in extractable hydrocarbons corresponds with a change in vitrinite macerals from a hydrogel structure to a bituminogel structure (see Teichmuller in Stach et al. 1982 for a discussion of structural changes in vitrinite with increasing maturity). Over the rank range 0.7% to 0.9% vitrinite reflectance, much less water is expelled from organic matter as compared with the range 0.5% to 0.7% vitrinite reflectance. The changes in the organic matter structure and in the fluid expulsion mechanisms may result in an overestimation of the rank required for hydrocarbon generation if 'extractable hydrocarbons' data are used. Changes in the fluorescence characteristics of vitrinite with rank are very distinctive, and can be interpreted as providing evidence for changes in the extent to which hydrocarbons are able to migrate out of vitrinite. Teichmuller (1982) has d o c u m e n t e d t h e d e v e l o p m e n t of f l u o r e s c e n c e in desmocollinite typically at reflectances between 0.5% and 0.7% and the spectral shift to the red with further increases in maturity. The present author has observed similar changes for Australian vitrinites occurring both in coals and as d.o.m. Additionally it is clear that fluorescing desmocollinites (Fig. 7b) commonly cause the development of hazy green fluorescence within the immersion oil, presumably by the emission from the desmocollinite of petroleum-related compounds. The emission of globules and streams of green fluorescing oil commonly occurs from cracks in telocollinite (Figs 7c, d). These emissions are most commonly associated with vitrinite in the reflectance range of approximately 0.8% to 1.1% but also occur at both higher and lower rank. The higher maturity cut-off is probably related to an unfavourable balance between the rate of generation of new hydrocarbons and the rate of cracking of hydrocarbons. The lower maturity end of the range could relate to the commencement of hydrocarbon generation but, equally, could relate to changes in the mechanics associated with primary migration. The fluorescence behaviour of vitrinites is summarized in Table 5. Some changes in maceral composition, texture and optical properties are direct indicators of the extent of change in the solid phases and should indicate the extent of hydrocarbon generation. The presence of exsudatinite can be interpreted either as proof of hydrocarbon generation or as proof that

217

hydrocarbons generated have not been able to escape from the system. However, features such as the micrinitization of bituminite and changes in fluorescence intensity and colour of exinite are probably reliable indications of the expulsion of hydrocarbons. The range of rank over which liquid hydrocarbons are generated is considered to vary for each of the macerals. No consensus exists in relation to the way in which this variation occurs. Leythaeuser et al. (1980) imply that generation occurs first from alginite, then from sporinite and finally from vitrinite. Tissot and his co-workers conclude exactly the reverse. Tissot's conclusions (Tissot & Welte, 1978) accord best with the results of pyrolysis and hydrogenation experiments and with petrographic data. Recently Snowden & Powell (1982) have drawn attention to the development of oils from resiniterich source rocks having very low maturity. Smith & Cook (1980) have pointed out that, at low rank, inertinite undergoes the most rapid coalification. Taking their data together with published and unpublished data on fluorescence properties, estimates are presented in Table 6 of the rank ranges over which liquid hydrocarbons are generated by the various macerals or maceral groups. In assessing the hydrocarbon generation history of a sequence, it is necessary to take into account the type and abundance of the organic matter as well as its maturity (see e.g. Pitt, this volume). The specific yield of exinite macerals is much greater than that of vitrinite, but in terrestrial sequences the greater abundance of vitrinite makes it probable that vitrinite is equally important as a source of liquid hydrocarbons. At low ranks vitrinite, suberinite, resinite and bituminite are likely to be the most significant contributors to oil generation. At higher ranks sporinite and cutinite make a greater contribution. The contribution of alginite is delayed beyond that of the other macerals but the evidence suggests that hydrocarbon evolution from alginite is probably very rapid with a very high specific yield. This is likely to contribute to efficient primary and secondary migration. It is important to stress that the exinite macerals do not 'disappear' by dint of generating hydrocarbons as is sometimes claimed. Typically, they undergo some shrinkage and develop optical properties similar to those of vitrinite. They only 'disappear' to the extent that they become difficult to distinguish.

HYDROCARBON GENERATION EROMANGA BASIN

THE

Poolowanna fades In the central part of the Poolowanna Trough (Fig. 1) generation from all the macerals is likely to have occurred, but in most other areas, only the vitrinite, resinite and suberinite are likely to have generated hydrocarbons. Even in the high rank areas of the Poolowanna Trough, strong exinite fluorescence indicates that hydrocarbon generation is not complete. In areas flanking the Nappamerri Trough and in the Windorah Trough the units containing the Poolowanna facies are typically mature to late-mature and are likely to

Fig. 7. a. Positive alteration of the central part of the field. The fluorescence intensity of both bituminite and of diffuse mineral matter. Toolebuc Formation, Eromanga Basin, 762 m, cuttings sample. R v max 0.33% interpolated from samples in units above and below the Toolebuc Formation. Fl., field width 0.34 mm. b. Poolowanna facies. Duroclarite with abundant sporinite (yellow to orange) and distinctly fluorescing desmocollinite (brown). Vitrinite reflectance of this field 0.54%. Lower Poolowanna Formation, Poolowanna Trough, 2542 m, cuttings sample. R v max 0.80% interpolated from adjacent samples. Fl., field width 0.34 mm. c. Poolowanna facies. Green fluorescing oil cut emanating as streams and globules from a crack in telocollinite. Resinite (orange) is also present in the telocollinite. Poolowanna Formation, Poolowanna Trough, 2215 m, cuttings sample. R v max 0.74%. F.l., field width 0.45 mm. d. Poolowanna facies. A s for c, but in reflected white light. The crack is probably an artefact rather than a natural cleat fracture.

218

A. C. COOK

have been important sources of liquid hydrocarbons. Most of the generative organic matter occurs as coal seams. Exsudatinite is present in some coals. Oil emission from vitrinite is relatively common and probably indicates that oil generation has occurred.

Walloon fades

The abundance of resinite and suberinite in the organic matter of this facies suggests an ability to generate liquid hydrocarbons at relatively low rank. Table 6 indicates that partially generative ranks for the Walloon facies are reached over wide areas of the Eromanga Basin. In the central Nappamerri Trough, very high ranks are reached and in places the sequence containing the .Walloon facies is beyond the oil deadline. Exsudatinite is commonly developed and is found in coals with reflectances as low as 0.45%. Although much of the organic matter is coal-related, mineral matter is intimately mixed with much of the organic matter and the distinction between coals and d.o.m. is far from clear-cut in many parts of the sequence.

Mooga facies

The oil generating potential of the Mooga facies organic matter is very similar to that- of the Walloon facies. Maturation levels are generally lower but the presence of bituminite may partially compensate for this factor. The proportion of inertinite present is not sufficiently high to downgrade the source-potential.

Cretaceous marine facies

The alginite component of the organic matter in this facies has extremely high source-potential but is oil mature only in or flanking the Nappamerri Trough. The basal part of the marine sequence is probably mature enough over extensive areas to generate some oil from bituminite. However, bituminite is largely restricted to the Toolebuc Formation and this typically reaches only marginal maturity even for bituminite. If bituminite-rich lithologies occur further down in the marine sequence these could have much greater hydrocarbon potential as compared with the bituminite-poor lithologies.

Win ton facies

Sequences containing this facies are immature over most of the basin. Early maturity may be reached in the central parts of the Nappamerri Trough. Further data on this area may indicate the possibility of hydrocarbon generation.

CONCLUSIONS

The organic matter in the Eromanga Basin can be grouped into four distinct non-marine coal-dominated facies and one marine facies. Organic matter is generally more abundant in the non-marine sequences as compared with the marine sequence. The exinite content of the coals is relatively high, and exceptionally high in the case of the Walloon and Mooga facies. The coal-dominated facies are the result of differences in flora, climate and tectono-sedimentary setting. The marine sequence typically contains sparse dispersed organic matter, the major exception being the Toolebuc Formation where bituminite and lamalginite are abundant. An abundance of bituminite is characteristic of the Toolebuc Formation and may result from the growth and preservation of an algal-fungal mat on, or immediately above, the surface of the sediment. Assessment of the source-potential of the sequences in the Eromanga Basin is influenced strongly by the view taken of the location of the oil window for the various macerals and whether or not organic matter related to coal seams is considered as a potential source. Petrographic evidence suggests that significant generation of liquid hydrocarbons occurs at relatively low ranks from vitrinite, suberinite, resinite and bituminite. Coal and dispersed organic matter are part of a continuum in coal-bearing sequences. If organic matter in coal-related lithologies is accepted as being capable of yielding migrated liquid hydrocarbons during coalification, the coal-bearing facies have significant potential for sourcing oil. Sporinite and cutinite have generated hydrocarbons from more limited volumes of sedimentary rock as compared with the generative volume for vitrinite because of the relatively low rank levels in much of the Eromanga Basin. Alginite is largely confined to the Cretaceous marine sections where maturation levels are generally unfavourable for hydrocarbon generation from alginite.

ACKNOWLEDGEMENTS

Delhi Petroleum Pty Ltd and its associates kindly gave permission to publish information obtained from samples provided by them for petrographic examination, but the opinions expressed are those of the author. Maris Zwigulis of Delhi and Agu Kantsler have provided invaluable comment over a number of years and I am grateful to Anne Taylor and Sandradurage Padmasiri for valuable discussions. Assistance from Michelle Smyth, Peter Moore, David Gravestock and Jenny Jarman is acknowledged in preparing the final manuscript. Margaret Atkinson and Joan Cook typed the manuscript.

REFERENCES

ALPERN, B., 1970: Classification Petrographique des constituents

organiques fossiles des roches sedimentaires. Rev. Inst. Franc. Petrole et Ann. Combust, liquid. 25, 1233-66. COOK, A. C , 1975: The spatial and temporal variation of the type and rank of Australian coals; in Cook, A. C. (ed.) Australian black coal—its occurrence, mining, preparation and use. Australas. Inst. Min. Metall., Illawarra Branch, 63-84. COOK, A. C., 1981: Temporal variation of type in Australian coal seams. Bull. Centres Rech. Explor.-Prod. Elf-Aquitaine, 5, 443-59. COOK, A. C., 1982: The origin and petrology of organic matter in coals, oil shales and petroleum source-rocks. A. C. Cook with contributions from A. J. Kantsler, Univ. Wollongong, N.S.W.

COOK, A. C . & KANTSLER, A. J. (eds), 1980: Oil shale

petrology workshop, Wollongong, 1980. Keiraville Kopiers, Wollongong.

COOK, A . C . , HUTTON, A . C . & SHERWOOD, N . R., 1981:

Classification of oil shales. Bull. Centres Rech. Explor.-Prod.

Elf-Aquitaine, 5, 3 5 3 - 8 1 . HEROUX, Y., CHAGNON, A. & BERTRAND, R., 1979: Compilation and

correlation of major thermal maturation indicators. Am. Assoc. Pet. Geol. Bull. 63, 2128-44.

HUTTON, A . C., KANTSLER, A . J., COOK, A . C . & MCKIRDY, D. M.,

1980: Organic matter in oil shales. APEA J., 20(1), 44-67.

INTERNATIONAL COMMISSION FOR COAL PETROLOGY, 1963, 1971,

1975: International handbook for coal petrology. Centre National de la Recherche Scientifique, Paris, 2nd Edn and supplements.

219

ORGANIC FACIES KANTSLER, A . J., C O O K , A . C . & ZWIGULIS, M . , 1982: M a t u r a t i o n

patterns in the Eromanga Basin, in Moore, P. S. & Mount, T. J. (compilers), Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 284-95. KANTSLER, A . J., S M I T H , G . C . & C O O K , A . C . , 1978: L a t e r a l a n d

vertical rank variation: implications for hydrocarbon exploration. APEA J., 18(1), 143-56. KANTSLER, A . J., PRUDENCE, T. J. C . , COOK, A . C . & ZWIGULIS,

M., 1983: Hydrocarbon habitat of the Cooper/Eromanga Basin, Australia. APEA J., 23(1), 75-92. KAUFFMANN, E. G., 1981: Ecological reappraisal of the German Posidonienschiefer (Toarcian) and the stagnant basin model; in Gray, J., Boucot, A. J. & Berry, W. B. N. (eds), Communities of the Past. Hutchinson, 311-81. LEYTHAEUSER,

D.,

HAGEMANN,

W.,

HOLLERBACK,

SCHAEFER, R. G., 1980: Hydrocarbon generation in source beds as a function of type and maturation of their organic matter: a mass balance approach. World Pet. Congr. Proc. 2, 31-41, Heyden, London. RETALLACK, G. J., 1981: Two new approaches for reconstructing fossil vegetation with examples from the Triassic of eastern Australia; in Gray, J., Boucot, A. J. & Berry, W. B. N. (eds) Communities of the past. Hutchinson, 271-94. ROGERS, M. A., 1980: Application of the organic facies concepts to hydrocarbon source rock evaluation. World Pet. Congr. Proc. 2, 23-30, Heyden, London. SHERWOOD, N. R. & COOK, A. C., 1986: Organic matter in the

in Toolebuc Formation oil shales; in 'First Australian workshop on oil shale, 35-8. SHERWOOD, N. R. & COOK, A. C., 1985: Organic matter in the

Toolebuc Formation; in This volume. SMITH, G. C., 1981: Tertiary and Upper Cretaceous coals and coal measures sediments in the Bass and Gippsland Basins. Ph.D. Thesis, Univ. Wollongong, (unpubl.). SMITH, G. C. & COOK, A. C., 1980: The coalification paths of exinite, vitrinite and inertinite. Fuel, 59, 641-46. SMYTH, M., 1979: Hydrocarbon generation in the Fly Lake-Brolga area of the Cooper Basin. APEA J., 19(1), 108-14.

SMYTH, M., 1983: Nature of source material for hydrocarbons in the Cooper Basin, Australia. Am. Assoc. Pet. Geol. Bull. 67, 1422-28. SMYTH, M . , COOK, A . C . & P H I L P , R. P., 1984: B i r k h e a d r e v i s i t e d :

petrological and geochemical studies of the Birkhead Formation, Eromanga Basin. APEA J. 24(1), 196-216. SMYTH, M. & SAXBY, J. D., 1981: Organic petrology and geochemistry of source rocks in Pedirka-Simpsoh Desert Basin,- central Australia. APEA J. 21(1), 187-99. SNOWDON,

POWELL,

G.,

1982:

Immature

oil

and

condensate—modification of hydrocarbon generation model for terrestrial organic matter. Am. Assoc. Pet. Geol. Bull. 66, 775-88. STANDARDS

ASSOCIATION

AUSTRALIA,

1981:

Microscopical

determination of the reflectance of coal macerals. AS 2486-1981, 12p. TEICHMULLER, M., 1971: Anwendung kohlenpetrographischer Methoden bei der Erdol- und Erdgasprospektion. Erdolu. Kohle, 24, 69-76. TEICHMULLER, M., 1974: Enstehung und Veranderung bituminoser Substanzen in Kohlen in Beziehung zur Enstehung und Unwandlung des Erdols. Fortschr. Geol. Rheinland. u. Westf., 24, 65-112. TEICHMULLER, M., 1982: Fluorescenzmikroskopische Anderung von Liptiniten und Vitriniten mit Zunehmendem Inkohlungsgrad und ihre Beziehungen zu Bitumenbildung und Verkokungsverhalten. Geol. Land. N-W, Krefeld. TEICHMULLER, M., 1982: Stach's Textbook of Coal Petrology, 3rd Edn, Gebruder Borntraeger. TEICHMULLER, M . & TEICHMULLER, R . , 1981: T h e s i g n i f i c a n c e o f

coalification studies to geology—a review. Bull. Centres Rech. Explor-Prod. Elf-Aquitaine, 5, 491-534. THOMAS, B. M., 1982: Land-plant source rocks for oil and their significance in Australian basins. APEA J. 22(1), 164-78. TISSOT, B. P. & WELTE, D. H., 1978: Petroleum

occurrence. Springer-Verlag, Berlin.

formation

and

Geological Society of Australia Special Publication No. 12, 221-240

Source rock evaluation and maturation history of the central Eromanga Basin V. L. Passmore & C. J. Boreham Bureau of Mineral Resources, P.O. Box 378, Canberra City, A.C.T. 2601.

ABSTRACT The central Eromanga Basin contains a thick sequence of Jurassic and Cretaceous clastics that are prospective for oil and gas. Cores and cuttings from 41 wells were analysed by chemical and microscopic methods to provide a quantitative assessment of their source potential, the areal extent of the source interval, and the possible timing of hydrocarbon generation. Abundant organic source matter is present in a number of formations throughout much of the study area. Richness and oil yield are highly variable within the four intervals examined, reflecting differences in kerogen type and maturity of individual formations. Locally, the Wallumbilla and Birkhead Formations and the Evergreen Formation correlative, containing dominantly mixed Type I/II kerogen, are the main oil-prospective units. The Westbourne Formation, generally rich in gas-prone Type III organic matter, has some minor oil potential. Minor oil potential can also be expected from the thin shales within the Hutton Sandstone which contain predominantly mixed Type II/III kerogen. Despite the problems of combining maturation profiles determined from wet gas content, Tmax, and vitrinite reflectance, the data obtained from these different techniques generally agree that present-day potential is restricted to the western part of the central Eromanga Basin where geothermal gradients and depth of burial were sufficient for prospective source rocks to reach the main oil zone. None of the source rocks is sufficiently mature to have generated significant quantities of gas. A minimum burial depth of 1200 m is required for the onset of significant hydrocarbon generation to occur given the present geothermal gradients. Sufficient depth of burial for source rocks to have reached the main oil zone was not attained until the Late Cretaceous. If the proposed lower geothermal gradients were operative during the Mesozoic, hydrocarbon generation is unlikely to have commenced until the Tertiary.

INTRODUCTION The Eromanga Basin is the largest of the four Phanerozoic sedimentary basins covering the southwest corner of Queensland. That part of the Eromanga Basin, outlined on Figure 1, has been designated the central Eromanga Basin (Senior et al., 1978). The name central Eromanga Basin region is used in this paper when referring to the area geographically without restricting it to the Eromanga Basin sequence. The gas potential of this region was first recognised by the discovery of the Gilmore Field in 1964 in the Devonian Adavale Basin. This was followed in succeeding years by gas discoveries in the Permian and Triassic of the Cooper Basin and the Jurassic of the Eromanga Basin (Fig. 1). The prophecy for oil potential in the region was not confirmed until the discovery in 1981 of commercial quantities of oil in Jackson 1. This paper presents results of recent investigations of the oil and gas potential of possible source rocks in the central Eromanga ^asin. The study uses new data that greatly expand on earlier investigations in the central Eromanga Basin, and attempts to delineate those sediments with the greatest oil potential, their areal extent, and possible timing of hydrocarbon generation. Results from organic geochemical analyses were used to determine source rock richness (quantity and quality of organic matter), kerogen type, and maturity.

Data used in this study were derived from analyses by the Bureau of Mineral Resources Petroleum Technology Laboratory (BMR), Australian Mineral Development Laboratories (AMDEL), CSIRO Mineral Research Laboratories Fuel Geoscience Unit (CSIRO), Robertson Research International Limited (Robertson Research), and Shell Development Pty Ltd (Shell). Most of the samples were obtained from wells subsidised under the Commonwealth Government's Petroleum Search Subsidy Acts (which operated between 1958 and 1974). Canned cuttings samples from more recently drilled wells were provided by Delhi Petroleum Pty Ltd (Delhi), Crusader Oil N.L., and Lennard Oil N.L. Stratigraphic nomenclature (Fig. 2) follows that of Exon & Senior (1976) and Senior et al. (1978) particularly for the subsidised and older well data. Company nomenclature, summarised in Armstrong & Barr (1982), was adopted for the newer wells. Formation ages correspond to those of Burger (1982).

PETROLEUM EXPLORATION Sprigg (1982) has traced the history of petroleum exploration in the Eromanga Basin from its earliest days, and details of discoveries are reviewed by Armstrong & Barr (1982). Until the late 1970s primary exploration targets were in basins

V. L. PASSMORE & C. J. BOREHAM

222

150° 1 200 I km

SYM

FORMATION

BOL

AGE

Winton Formation Mackunda Formation

Coorikiana Sandstone Bulldog Shale

Limit Cooper/Galilee Basins Limit Adavale Basin Discoveries

• Oil #EGas & Gas #AGas

Eromanga Basin Eromanga Basin Cooper Basin Adavale Basin

Fig. 1. Hydrocarbon discoveries in central Eromanga Basin region.

underlying the Eromanga Basin. The Eromanga Basin sandstones were commonly considered to have been water flushed and generally unprospective (Laing, 1969; Vine, 1976) despite hydrocarbon shows and porous reservoir rocks. Over 100 exploration wells have been drilled in the central Eromanga Basin region since the 1950s and the area remains a site of active exploration. The first Eromanga Basin discovery in the region occurred in 1978 when Wackett 1 discovered gas in the Birkhead Formation. This was followed in 1981 by the discovery of oil in Jackson 1 in the Hutton Sandstone and Westbourne Formation, and the Murta Member of the Mooga Formation. Since 1981 several additional oil and gas discoveries have been made in the same area in these units.

PREVIOUS GEOCHEMICAL INVESTIGATIONS

A number of organic geochemical studies have been carried out on Eromanga Basin rocks, using coal and disseminated organic matter in the fine-grained rocks as the organic source material. Published results, such as those of Kantsler et al. (1978, 1982), Smyth & Saxby (1981), Cook (1982), and McKirdy (1982), concentrated mainly on the South Australian part of the Eromanga Basin. The only published results from the central Eromanga Basin are those of Senior & Habermehl (1980), the preliminary results of Passmore & Boreham (1982) on maturation and source richness, and a recent paper on the Cooper and Eromanga Basins by Kantsler et al. (1983) which covers the western central Eromanga Basin region. Company investigations carried out within the central Eromanga Basin have remained largely unpublished and confidential.

Wallumbilla CD Formation LU

Wyandra Transition Beds Sandstone Member Murta Cadna-owie too Member Formation Hooray § 2 Namur ^ E Sandstone Sandstone £ Member Westbourne Formation

LLI v Q_ Z £ 3 D lu O

Adori Sandstone

z o o

Birkhead Formation

cc cc

CO D

Allaru Mudstone Toolebuc Formation

Oodnadatta Formation

Evergreen Formation Precipice Sandstone

0) +-» <0 0 "U -O

(/)C/>< CC D

Hutton Sandstone Basal Jurassic

LLI

Fig. 2. Stratigraphic nomenclature of central Eromanga Basin and time interval groupings of geochemical data.

GEOLOGY Stratigraphy

Within the study area, the Eromanga Basin sequence reaches a maximum thickness of about 3000 m; the thickest accumulations occur in the Thomson and Cooper Synclines and in the Wilson Depression (Senior & Habermehl, 1980) (Fig. 3). The basin sediments range in age from Early Jurassic to Late Cretaceous (Exon & Senior, 1976; Senior et al., 1978). The Jurassic sequence is mainly composed of continental quartzose sandstone interbedded with carbonaceous Siltstone, mudstone, and minor coal. The overlying section consists of an Early Cretaceous marine sequence of labile sandstone interbedded with siltstone, mudstone, and claystone, and a lithologically similar Late Cretaceous sequence deposited under paralic, lacustrine, and fluvial conditions. The cyclic nature of the sequence is believed to reflect global eustatic sea-level changes (Burger, 1982).

Structure

Rocks in the central Eromanga Basin are only mildly deforme^. Folds are usually broad and of low amplitude, and

SOURCE ROCKS AND MATURATION

223

Geochemical results are presented on the subsequent figures using these four intervals.

Source richness

Fig. 3. Well location, central Eromanga Basin (refer Table 1 for well names).

faults have throws of up to 300 m. West of the Canaway Fault the structures have a predominantly northerly trend, but east of the Canaway Fault they trend northeast (Fig. 3). The faults and folds reflect greater deformation in the underlying basins or in basement rocks (Senior et al., 1978). Major synclines are closely aligned to the axes of the infra-basins.

Sedimentation

Total organic carbon (TOC) content was determined for the majority of the core and cuttings samples used in the study. Fifty-two core samples were analysed for total extractable organic matter (EOM), and thirty-five of these samples were separated by liquid chromatography into saturated hydrocarbons (SATS), aromatic hydrocarbons (AROM), and polar (N-, S-, O- containing) organic compounds (POLAR). Gas chromatograms were recorded for the saturated hydrocarbon fractions only. A source rock richness plot adopted by Jackson et al. (1980) is used to rate source rock potential (Fig. 4). All but four samples fulfil the minimum TOC content of 0.5% (Dickey & Hunt, 1972) required for a clastic rock to have any hydrocarbon source potential. Rocks of Early Cretaceous to Early Jurassic age have a fair to very good source potential. The greatest potential is found in the carbonaceous siltstone and shale of the Middle Jurassic Westbourne and Birkhead Formations, and the Early Jurassic Evergreen Formation correlative. While Figure 4 suggests that several samples are possible hydrocarbon sources, it is necessary to integrate the chemical data with the results of organic microscopy to determine maturation levels, kerogen type, and oil yield before oil or gas potential can be determined. Twenty-four of the 64 cores sampled provide these data.

The Eromanga Basin sequence in the study area shows no evidence of widespread disruption or non-deposition between the Early Jurassic and the Late Cretaceous except for one widespread unconformity in the Late Jurassic (Burger, 1982). The main depositional axis is the elongate, northerly trending Cooper Syncline which generally coincides with the Windorah Trough in the underlying Cooper Basin, while a shallower, less well developed depocentre formed east of the Canaway Fault. During the Jurassic, the main depositional trend was easterly, but with the initiation of the Cooper Syncline in the Cretaceous the trend was reoriented to the north. * Although sediment thickness within the study area varies laterally, vertically the ratio of the thicknesses of the Jurassic, Lower Cretaceous, and Upper Cretaceous are roughly similar except for the eastern edge where the Upper Cretaceous is absent. The rate of sedimentation significantly increased through time, having been most rapid during the early Late Cretaceous. The calculated sedimentation rates are 5 to 10 times higher in the Late Cretaceous than in the Jurassic, the largest difference being in the Cooper Syncline where Late Cretaceous sediments are thickest.

SOURCE ROCK ANALYSES

Up to six geochemical methods were used in the investigation to analyse 364 core and cuttings samples from 41 wells. The wells sampled and types of analyses are shown in Table 1; basic data are presented in Appendix 1. The locations of the wells sampled are shown in Figure 3; each reference number (ref. no.) in Table 1 corresponds to a well • Early-Late Cretaceous • Middle Jurassic • Early Cretaceous/ * Early Jurassic number in Figure 3. For ease of data handling, formations Late Jurassic in the Eromanga Basin sequence are grouped into four time intervals designated Early Jurassic, Middle Jurassic, Late Fig. 4. Source rock richness of Jurassic and Cretaceous core samples, Jurassic-Early Cretaceous, and Early-Late Cretaceous (Fig. 2). central Eromanga Basin.

224

V. L. PASSMORE & C. J. BOREHAM

T A B L E 1. Geochemical analyses performed on source rocks from each exploration well sampled in study area.

Well Name Alkina 1 Arrabury 1 Balfour 1 Barcoo Junction 1 Barrolka East 1 Belah 1 Betoota 1 Bodalla 1 Boree 1 Buckabie 1 Budgerygar 1 Bury 1 Canaway 1 Carlow 1 Chandos 1 Chandos South 1 Cumbroo 1 Dilchee 1 Durham Downs 1 East Windorah 1 Etonvale 1 Fairlea 1 Galway 1 Gilmore 1 Gilpeppee 1 Gilpeppee 2 Hume. 1 Ingella 1 Jackson 1 Kyabra 1 Log Creek 1 Morney 1 Mt Howitt 1 Orientos 1 Quilberry 1 Roseneath 1 Stafford 1 Tartulla 1 Thunda 1 Westbourne 1 Yongala 1 TOC analysis only

Type of Geochemical Analysis Head Vitrinite Kerogen Space Gas Reflectance Type X X X X X X X X X

Extractable Organic Matter X X

X X X X X X X X X

X X X X X

X X X X

X X X

X X X X X

X X X

X X

X X X

X X

Rock Eval

X X X X X X X X X

X X

X X X X X

X X

Gas Chromatography

X X X X

X X X X X X

Reference Number 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

Maturation

Source rocks are commonly classified as immature, mature or overmature depending on their stage of thermal alteration. Maturity ratings used in this study were obtained principally from the percentage wet gas content (iC -C /^C -C x 100) derived from headspace gas analysis of canned cuttings, and from vitrinite reflectance (VR) values determined from coals and dispersed organic matter. In some wells, these ratings are supplemented by maturity data obtained from gas chromatograms of the SATS fraction and the Rock-Eval temperature maximum; Tmax values of 435 °C-465 °C outline the oil zone (the lower limit of which varies with kerogen type (Tissot & Welte, 1978)). Maturation levels for 13 recently drilled wells (Table 1, Fig. 3) in the region west of the Canaway Fault were determined from headspace gas analyses. Following Powell (1978), maturity levels as defined by percentage wet gas content are: immature less than 30%; above 30% represents the beginning of the mature stage; the main oil zone occurs where wet gas content remains above 60%; the transition froAi the mature to the overmature stage is indicated by a decrease in wet gas content below 30%. The overmature stage is 2

differentiated from the immature stage by its higher vitrinite reflectance values. Vitrinite reflectance measurements were carried out on core samples from 25 wells (Table 1, Fig. 3). Source rocks having vitrinite reflectance values of less than 0.5% are immature, those with values between 0.5% and 1.3% are mature (Fig. 5) (Tissot & Welte, 1978). Significant oil generation commences between 0.5% and 0.7%, and beyond 1.3% (the oil deadline) little oil is produced. Gas generation does not become important until a reflectance value of 1.0% is attained. The main oil zone in the central Eromanga Basin is considered to be between 0.6% and 1.1% (Passmore & Boreham, 1982). Vitrinite reflectance values are plotted against depth to produce a maturation profile for the central Eromanga Basin (Fig. 5). The Late Jurassic-Early Cretaceous and Middle and Early Jurassic samples range between the immature zone and the highly desirable main oil zone. The Cretaceous samples are mostly immature. The main oil zone occurs at depths in excess of 1200 m (Passmore & Boreham, 1982). This limit appears to be a minimum burial depth for oil generation although, as Figure 5 shows, source rocks at far greater depths of burial may still be immature.

SOURCE ROCKS AND MATURATION

225

paraffin maxima to a lower carbon number. The odd-to-even preference of immature Early Jurassic samples in the eastern edge of the study area can be contrasted with the mature Early Jurassic sample in the southwest, from Orientos 1 (ref. no. 34), which shows no odd-to-even preference. The limits used for the onset of the mature stage of oil generation are generally less rigid than those given above for vitrinite reflectance, wet gas, and Rock-Eval Tmax values. To varying degrees these limits are affected by source type and migration. To illustrate the difficulties encountered when combining maturation profiles obtained by independent techniques, two wells, Morney 1 (ref. no. 32) and Jackson 1 (ref. no. 29), are discussed. i a Using the commonly accepted values, the top of the mature stage for both Morney 1 and Jackson 1, based on wet gas content, occurs within the Early Cretaceous Allaru Mudstone, whereas the top of the main oil zone occurs above 1200 m within the Early Cretaceous Wallumbilla Formation. Using J*r? / A A vitrinite reflectance data, the mature zone begins within the i m s f i Early Cretaceous Wallumbilla Formation in Morney 1 and m P W in the Late Jurassic Namur Sandstone Member in Jackson g p P p s r ' 1, while the onset of the main oil zone, now below the 1200 m m i / / / limit, commences in the Late Jurassic Namur Sandstone Member in Morney 1 and the Early Jurassic Hutton Sandstone in Jackson 1. In a study of the hydrocarbon potential of the Canadian Arctic Islands, Powell (1978) concluded that the fa//I onset of the mature stage, determined by wet gas content, corresponds very well with that determined by vitrinite reflectance. That this is not the case for Morney 1 and Jackson 1 is probably a result of a combination of two factors, the 0-5 1 1-5 variation in kerogen type which can redefine maturation limits, Vitrinite reflection ( % VR) and the difficulty encountered in determining maturation zones where sandstone units are present. Sandstones can give • Early-Late Cretaceous anomalously low wet gas content irrespective of the level of • Early Cretaceous/Late Jurassic maturation (Powell, 1978), and upward migration of • Middle Jurassic hydrocarbons through a thick sandstone unit will result in $ Early Jurassic the upward displacement of the apparent top of the mature stage. In Morney 1 and Jackson 1 thick sandstones are E3 12/Q/29 interbedded with shales, expressed in the wet gas content plot (Fig. 7) as a 'sawtooth' profile. Fig. 5. Thermal maturation profile of the Jurassic and Cretaceous Based on Tmax data for both wells the predominantly sediments, central Eromanga Basin, determined from vitrinite marine Early Cretaceous section is marginally mature, whereas reflectance measurements. the predominantly terrestrial Jurassic sequence is mature. Notwithstanding the difficulties encountered in pinpointing Cross-sections through the study area (Figs 6, 7, 8) illustrate the present level of maturity. The most significant lateral the maturity limits, all techniques concur that the Jurassic variation in maturity occurs from east to west (Fig. 6). East is mature in both Morney 1 and Jackson 1. of the Canaway Fault, the Eromanga Basin sequence is mostly immature to initially mature, whereas in some areas west of the Canaway Fault the entire Jurassic and part of the Early Kerogen type and oil yield Cretaceous sequence fall within the main oil zone. On the Kerogen type was partly determined by petrographic cross-section in Figure 7, which obliquely crosses the Cooper identification of three maceral families: exinite, vitrinite, and Syncline, and Figure 8, which runs parallel to the axis of the inertinite. Type I and Type II kerogens are rich in the exinite Cooper Syncline, Jurassic and Early Cretaceous sediments macerals, derived from continental and marine debris and are within the mature zone. Where the underlying Cooper higher plants. In Type II kerogen the exinite is usually derived Basin sediments are thin or absent, at the ends of the cross- from marine organic matter deposited in a reducing section, the limits of the mature zone occur at shallower environment, or the kerogen may have a moderate vitrinite depths; this effect is not observed in the cross-section in component (Tissot & Welte, 1978; Durand, 1980). Depending Figure 8. on the individual maceral that constitutes the exinite group, This significant increase in maturity from east to west can the oil-prone exinite will give relatively high to very high oil be observed from other analyses. The gas chromatograms of yields (Cook, 1982). Vitrinite, the main component of Type the SATS fraction (Fig. 9, Table 1), although source dependent, III kerogen (Tissot & Welte, 1978), originates from terrestrial show an increase in maturation both with depth and in a matter and results in relatively moderate oil yields. On the westerly direction across the basin, as is indicated by a loss other hand, inertinite has little or no oil yield as a result of of the 'naphthene hump', a decrease in the odd-to-even the alteration of organic matter by oxidation, action of micropreference in the C -C normal paraffins, and a shift in organisms, biochemical or thermal processes. IMMATURE

MATURE

OVERMATURE

V. L. PASSMORE & C. J. BOREHAM

w Betoota 1

Westbourne 1 »

Gilpeppee 1

Yongala 1

Elmore 1

— n r o e a Level Z Sea Jf

Datum

IMMATURE

12/Q/22 4K Cretaceous J Jurassic Tr Triassic

P Permian C Carboniferous D Devonian

I I I

Mature zone

Main oil zone

Fig. 6. East-west cross-section with the main maturation stages and oil generating zone superimposed. Maturity determined by vitrinite reflectance.

Twenty-five core samples from 16 wells (Table 1, Appendix 1) were analysed for kerogen type, and the maceral families plotted on Figure 10. Although a moderate to high inertinite content is present in many of the samples, more than half have an exinite content between 5% and 62%, implying that fair to excellent oil yields can be expected from those that are within the main oil zone. All of the intervals have at least one sample (Fig. 10) with an exinite content greater than 5%.

Pyrolysis Although 54 core samples were analysed, their coverage of the Eromanga Basin sequence is sporadic and little or no data are available for some formations. In order to provide a more complete coverage of the Eromanga Basin sediments, in the western part of the study area 270 cuttings samples were taken

at regular intervals from 6 wells (Table 1, Rock-Eval). These cuttings samples were analysed for TOC and were then pyrolysed using the Rock-Eval method of Espitalie et al. (1977). Parameters for the Rock-Eval method have been defined for source rocks by Espitalie et al. (1977) and Clementz et al. (1979). Rocks with 'petroleum potential' values of less than 2, from 2 to 6, and greater than 6 have poor, moderate, and good source richness, respectively. Kerogen type and hence source type (oil or gas) is characterised by the 'Hydrogen Index' and the 'Oxygen Index' which are cross-plotted in a van Krevelen-type plot. High hydrogen indices relative to oxygen indices indicate oil-prone source or exinite-rich Type I kerogen, whereas low hydrogen and high oxygen indices typify gas-prone source or Type III kerogen, and a mixed source,

SOURCE ROCKS A N D MATURATION Gilpeppee 2

% Wet Gas

Total Gas (ppm)

Tartulla 1 Total Gas (ppm)

Bi Morney 1 % Wet Gas Total Gas (ppm) Organic Carbon % 20 40 60 80 i o 2

Barrolka East 1 % Wet Gas Total Gas (ppm)

" % Wet Gas 0 20 40 60 80 ! 0

0I ' I I ' I

Jackson 1 Total Gas (ppm) 10 10

Organic Carbon % 10 1 2 3 4 5

I I I I II 5

Fig. 7. North-south cross-section with the main maturation stages and oil generating zone superimposed. Maturity determined by wet gas content, results shown above cross-section. (Refer Table 2 for abbreviations of formation names).

V. L. PASSMORE & C. J. B O R E H A M

% W e t Gas

sw Dilchee

Total Gas (ppm)

% W e t Gas

Total Gas (ppm)

% W e t Gas

Kyabra

Belah 1

Total Gas (ppm)

Hume

Barrolka East

Alkina

Ingella

% W e t Gas

Total Gas (ppm)

Organic Carbon %

East Windorah

IMMATURE

Fig. 8. Northeast-southwest cross-section with the main maturation stages and oil generating zone superimposed. Maturity determined by wet gas content, results shown above cross-section. (Refer Table 2 for abbreviations of formation names).

229

SOURCE ROCKS AND MATURATION

</> D

OLU >

<(0 ^UJ LU

cc o

I be

jJ-MlliWUi

"jsjbiilillm^

o C/>

</<)

CC_Q) 34 .,b,

DTJ 2

17 , b x i j j j u i i l i i i 15

33 „ b

C »

jiliju^ ^JjtLujJilil^

^LiJlJULJ^

I.H

V) .

wiit

OCLU <

40 a b (

JuLXju^

iiuili 34

JLLU

39 JJiiill

Fig. 9. Gas chromatographs of the saturated hydrocarbon (SATS) fraction of several units in selected wells.

Type II kerogen, shows intermediate values. As maturation increases, the hydrogen and oxygen index values progressively decrease along broad pathways for the three kerogen types (Fig. 11). Tmax was used to indicate the level of organic maturation.

TOC results from the cuttings analysed by the Rock-Eval method confirm the core results that each of the four intervals is rich in organic carbon, but there are extreme variations in source richness and kerogen type within each interval, as illustrated by the pyrolysis plots in Figure 11. The observed

230

V. L. PASSMORE & C. J. BOREHAM

TABLE 2. Abbreviations of formation names used in Figures 7 and 8. Tra—Transition Beds Wi—Winton Fm Mu—Murta Mbr Ma—Mackunda Fm Na—Namur Ss Mbr Al—Allaru Mdst We—Westbourne Fm To—Toolebuc Fm Ad—Adori Ss Wa—Wallumbilla Fm Bi—Birkhead Fm Oo—Oodnadatta Fm Hu—Hutton Ss Co—Coorikiana Ss Ev—Evergreen Fm Bu—Bulldog Sh Ba—Basal Jurassic Wy—Wyandra Ss Mbr Tri—Triassic Ca—Cadna-owie Fm P—Permian Ho—Hooray Ss PP—Pre-Permian

VITRINITE

100

variability within each interval reflects the difference in the dominant kerogen type characterising each of the formations in the interval. For example, in the Middle Jurassic plot the Westbourne Formation is mainly characterised by gas-prone Type III kerogen, whereas the Birkhead Formation contains oil-prone mixed Type I/II kerogen. In the Early Jurassic Hutton Sandstone, mixed Type II/III kerogen predominates, and in the Cretaceous Wallumbilla Formation Type I/II kerogen occurs in the zones showing the greatest source richness.

OIL POTENTIAL INTERPRETATION

In the source rock analyses section, the data relevant to each of the four factors affecting oil potential are discussed separately for the four broad time intervals. In this section oil potential ratings of individual formations and specific samples are discussed with reference to the source rock richness, maturity, kerogen type, and oil yield results. Twentyfour core samples from 15 wells provided sufficient data for source potential analysis. Because of data limitations our Fig. 10. Triangular plot of the maceral contents of selected cores. interpretation from core results are mainly confined to (Refer Fig. 2 for key to time intervals). formations in the Early and Middle Jurassic, whereas cuttings results encompass all the Eromanga Basin sequence.

Fig. 11. Kerogen Type determinations for Jurassic and Cretaceous formations in the western central Eromanga Basin. Pyrolysis values determined by the Rock-Eval method. (Refer Fig. 2 for key to time intervals).

SOURCE ROCKS AND MATURATION The oil potential of the Cretaceous and Late Jurassic rocks is rated as poor from the limited core data. Rock-Eval analysis on cuttings however indicates that moderate oil potential exists in the source rocks of the Early Cretaceous Wallumbilla Formation. The source richness determined from cuttings samples for the Early Cretaceous is excellent in Galway 1 (ref. no. 23) and very good in Kyabra 1 (ref. no. 30). At best these units are marginally mature, requiring a greater maturity to fully realise their oil potential. The 14% exinite in samples from Boree 1 and Fairlea 1 (9 and 22 respectively on Fig. 10) suggests a moderate oil source, but their immaturity implies a present low capacity for generation which downgrades their oil potential rating. Middle Jurassic sediments are rated as having fair to very good source richness (Fig. 4). The marginally mature exiniterich core samples from Westbourne 1 and Canaway 1 (40 and 13 respectively on Figure 10) indicate only moderate hydrocarbon generation can be expected. These samples are further downgraded to a fair source for oil due to a fair to moderate source richness rating (Fig. 4). Betoota 1 (ref. no. 7) which has a very good source rating (Fig. 4) has only a fair oil generative potential because of its low exinite content. Pyrolysis data indicate a good source richness rating and a mixed Type I/II kerogen for the Birkhead Formation, principally in Durham Downs 1 (ref. no. 19) and Kyabra 1 (ref. no. 30). Temperature measurements imply that the Birkhead Formation in these wells is presently within the main oil zone and can be considered a good source for oil. Fair oil potential exists in the vitrinite-rich (Type III) shales of the Middle Jurassic Westbourne Formation. Within the marginally mature to mature Early Jurassic sediments, source richness ranges from poor to very good (Fig. 4). Exinite-rich organic matter from the Evergreen Formation correlative in Budgerygar 1 and Yongala 1 (11 and 41 respectively on Figure 10), as well as the highly vitrinitic organic-rich shales of the Evergreen Formation correlative in Chandos South 1 (16 on Figure 10) have a moderate oil generative potential. The last is the most mature of the three wells and thus has the best present-day hydrocarbon generation potential. Additional minor oil potential occurs in the shales of the Hutton Sandstone which contains predominantly mixed Type II/III kerogen. Eromanga Basin coal seams and shales showing good gas potential (low Hydrogen Index, high Oxygen Index) are all immature in relation to significant gas generation.

231

The oil from the Jackson and Jackson South Fields, if locally sourced, provides proof of a favourable burial and thermal history for oil generation within the basin. The geological history of most of the central Eromanga Basin implies that the duration of burial is similar for the same source rock sequence across the study area. However, large variations exist in the depth of burial of sediments of similar age, as illustrated by the structural cross-section in Figure 12. Vitrinite reflectance and wet gas content measurements suggest that only source rocks in parts of the western central Eromanga Basin have had a sufficient depth and duration of burial to reach the main oil zone. The apparent lack of maturity of source rocks in sediments of similar age and at a similar or greater depth in other parts of the central Eromanga Basin must, therefore, be due to other factors. Present geothermal gradients over much of the study area are above average by world standards for intracratonic basins. Examination of uncorrected present-day geothermal gradients from exploration wells in the central Eromanga Basin (Nicholas et al., 1980) shows lateral changes in gradient from 30°C/km to 50°C/km. Additional data from water bores (Senior & Habermehl, 1980) suggest that the lateral variation could be as high as 50°C/km (30°C/km to 80°C/km). High values are often coincident with areas of shallow basement

FACTORS AFFECTING GENERATION OF HYDROCARBONS

The initiation of significant hydrocarbon generation is a function of the maceral types, burial history (depth and duration of burial) of the source rocks, and the thermal history of the area. There are problems in determining the onset of generation. Cook (1982) suggests that, depending on maceral type, vitrinite reflectances of between 0.4% and 0.7% may indicate the lower limit of oil generation, whereas Saxby (1982) maintains that a much higher value of approximately 1.0% is required before oil can be generated. At the beginning of the oil zone the lower Tmax limit is higher for Type I and II kerogen than Type III kerogen. Cook (1982), summarising previous literature, considers that the oil generation threshold occurs at a lower maturation level in vitrinite and the resinite maceral of exinite, than in the exinite macerals sporinite and cutinite. Initial generation from the alginite maceral of exinite starts at a higher maturation level than that for the other macerals but its hydrocarbon evolution is probably very rapid.

Fig. 12. Changes in maturity across the central Eromanga Basin relative to sediment thickness and depth of burial.

V. L. P A S S M O R E & C. J. B O R E H A M

232

a n d occur a d j a c e n t to m a j o r faults. S o m e high gradients are a t t r i b u t e d by P o l a k & Ramsey (1977) to u p w a r d m i g r a t i o n of h o t artesian water a l o n g fractures. Recent studies o n g e o t h e r m a l gradients a n d organic m a t u r a t i o n levels in t h e C o o p e r Basin region (Kantsler et ai, 1978, 1982, 1983; Pitt, 1982; Schwebel et ai, 1980) suggest t h a t the present high

g r a d i e n t s are a relatively recent p h e n o m e n o n . F r o m p a l e o g r a d i e n t calculations, P i t t (1982) c o n c l u d e d t h a t the g r a d i e n t s existing p r i o r t o t h e mid-Tertiary were 1 0 ° C / k m to 2 0 ° C / k m lower t h a n t h e present t e m p e r a t u r e gradients. By p l o t t i n g t h e level of m a t u r i t y f o r t h e Jurassic and Cretaceous units above the structural cross-section, the relative

(a) EARLY CRETACEOUS

IMMATURE

Depth (km)

JURASSIC

(b)

IMMATURE

JURASSIC

Betoota 1

Gilpeppee 1,2

Aikina 1

Chandos South 1

Yongala 1

Gilmore 1

Westbourne 1

Depth (km)

LATE CRETACEOUS

JURASSIC

Mature zone Main oil zone Cooper/Galilee Basins

Fig. 13. Maturation history of the central Eromanga Basin, depicted by the changes in maturity of selected wells through time.

SOURCE ROCKS AND MATURATION 233 effects of geothermal gradient and depth of burial on maturity CONCLUSIONS can be observed (Fig. 12). Areas of high source maturity This study has established the presence of mature source roughly coincide with areas of thick Late Cretaceous rocks rocks with significant oil generative potential in the central and greatest burial, suggesting that burial depth is a significant Eromanga Both Jurassic and Cretaceous sediments in factor in determining the level of maturity of source rocks most of theBasin. area contain sufficient quantities of organic in the central Eromanga Basin. There is a wide divergence carbon to study be classed as good source rocks. Locally, a in maturity between rocks at similar depths in the eastern and of mixed Type I/II, exinite-rich kerogen, western parts of the area. The variations in maturation level predominance characterises prime oil source sequences in the Early noted on the cross-sections in Figures 7 and 8 are attributed Cretaceous Wallumbilla Formation and the Jurassic Birkhead to variations in present geothermal gradients. A detailed study Formation and the Evergreen Formation correlative. Although of the burial and thermal histories of the Cooper and the Jurassic Westbourne Formation contains exinite Eromanga Basins is presented elsewhere (see Pitt, this volume). (mixed Type 11/111 kerogen) most of the organic some matter is gasprone Type III. Source rock maturity data indicate that the prospective TIMING OF HYDROCARBON GENERATION source rocks are probably restricted to the western half of To graphically depict timing of hydrocarbon generation the the central Eromanga Basin. The thin sedimentary pile and level of maturity of source rocks in 7 wells has been plotted low geothermal gradients east of the Canaway Fault suggest for 3 time intervals in the central Eromanga Basin's history, that any source rocks in this region are unlikely to have reached (Fig. 13). A line at 1200 m has been superimposed on each the main oil zone. Throughout the region, Jurassic source of the sections in Figure 13 to mark the minimal burial depth rocks are the most prospective in terms of both oil-generative considered necessary for the onset of significant oil generation potential and level of maturity. In the regions with the highest in the basin. geothermal gradients the Early Cretaceous Wallumbilla The Jurassic sequence was immature and incapable of Formation also offers moderate oil-generative potential where generating hydrocarbons at the start of the Early Cretaceous it enters the main oil zone. However, the central Eromanga due to insufficient depth of burial. Even the more deeply Basin sediments appear not to have reached sufficient buried Cooper Basin sediments had yet to reach the main oil maturity to generate significant quantities of gas. At preserttzone (Fig. 13a). However, by the start of the Late Cretaceous day geothermal gradients, a minimum depth of burial of (Fig. 13b) the entire Cooper Basin sequence could have entered 1200 m is required for source rocks to reach the main the main oil zone, whereas the bulk of the source rocks in threshold of oil generation. Oil generation in most, if not all, the Eromanga Basin sediments remained immature. Early of the source rock units could have occurred no earlier than Jurassic source rocks in the deeper parts of the basin lay just the Late Cretaceous, even at the present high geothermal inside the main oil zone. It is highly speculative whether gradients. Initial oil generation is most likely to have generation had commenced.at the start of the Late Cretaceous commenced in the Tertiary with peak generation occurring in the hottest, deepest sites in the basin. The Early Jurassic at present for the organic-rich Jurassic sediments. source rocks were well below the minimum depth of burial required at present geothermal gradients. If, however, the lower ACKNOWLEDGEMENTS palaeogradient modelled by Pitt (1982) for the Cretaceous is Samples and the formation tops for wells evaluated by accepted, Early Jurassic source rocks may not have reached headspace gas techniques were supplied through the courtesy the oil generative threshold. In shallower Jurassic sediments of Delhi Petroleum Pty Ltd, Crusader Oil N.L., and Lennard and Jurassic sediments in the cooler parts of the basin, Oil N.L. We are also indebted to Maris Zwigulis of Delhi for generation could not have been initiated until after deposition providing vitrinite reflectance values for Jackson 1 and of the Late Cretaceous Winton Formation (Fig. 13c). Morney 1. Dr M. R. Walter, B. A. McKay and E. Nicholas of the Pitt (1982) and Kantsler et al. (1983) suggest that oil generation for most of the Eromanga unit commenced in the Bureau of Mineral Resources provided useful criticism. Tertiary. Investigations of Passmore & Boreham (1982) have Thanks are also due to Drs A. Kantsler, D. Gravestock, and shown that on the western side of the Canaway Fault source G. Woodhouse for their critical comments. Dr D. Burger rocks as young as Early Cretaceous have reached the initial provided age controls for the formations. This paper is stages of hydrocarbon generation, and peak oil generation published with the permission of the Director, Bureau of Mineral Resources, Geology & Geophysics. is occurring now.

REFERENCES ARMSTONG, J. D. & BARR, T. M., 1982: The Eromanga Basin; in

Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 2 0 - 4 2 . BURGER, D., 1982: Palynology of the Eromanga Basin and its application; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc Aust. & Pet. Explor. Soc. Aust., Adelaide, 174-83.

CLEMENTZ, D. M., DEMAISON, G. J. & DALY, A. R., 1979: New

pyrolysis device speeds on-site source bed evaluations. Oil Gas

7. 77, 142-6.

COOK, A. C., 1982: Organic facies in the Eromanga Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium,

summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 234-57.

DICKEY, P. A. & HUNT, J. M., 1972: Geochemical and hydrogeologic

methods of prospecting for stratigraphic traps. Am. /4ssoc. Pet. Geol. Mem. 16, 136-67. DURAND, B. (ed.), 1980: Kerogen: insoluble organic matter from sedimentary rocks. Editions Technip, Paris.

ESPITALIE, J., MADEC, M., TISSOT, B. P., MENNIG, J. J. & LEPLAT,

P., 1977: Source rock characterisation method for petroleum exploration. Offshore Technol. Conf, Houston, Pap. 2935,

439-44. EXON, N. F. & SENIOR, B. R., 1976: The Cretaceous of the Eromanga and Surat Basins. BMR J. Aust. Geol. Geophys., 1, 33-50.

V. L. PASSMORE & C. J. BOREHAM

234

M., 1983: Hydrocarbon habitat of the Cooper/Eromanga Basin, Australia. APEA J. 23(1), 75-92.

Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 262-95. POLAK, E. J. & RAMSAY, D. C., 1977: Canaway Ridge, Queensland, geophysical survey, 1973. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1977/29 (unpubl.). POWELL, T. G., 1978: An assessment of the hydrocarbon source rock potential of the Canadian Arctic Islands. Can. Geol. Surv. Pap. 78-12, 1-80. SAXBY, J. D., 1982: A reassessment of the range of kerogen maturities in which hydrocarbons are generated. J. Pet. Geol. 5(2), 117-28.

KANTSLER, A . J., SMITH, G . C . & COOK, A . C., 1978: L a t e r a l a n d

SCHWEBEL, D. A., DEVINE, S. B. & RILEY, M . , 1980: Source, maturity

JACKSON, K. S., HAWKINS, P. J. & BENNETT, A . J. R., 1980: R e g i o n a l

facies and geochemical evaluation of the southern Denison Trough, Queensland. APEA J. 20(1), 143-58. KANTSLER, A . J., COOK, A . C . & ZWIGULIS, M . , 1982: M a t u r a t i o n

vertical rank variation: implications for hydrocarbon exploration. APEA J. 18(1), 143-56. LAING, A. C. M., 1969: Review of geology and case history of petroleum exploration in central Eromanga sub-basin. APEA J. 9, 88-96. MCKIRDY, D. M., 1982: Aspects of the source rock and petroleum geochemistry of the Eromanga Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 258-9. NICHOLAS, E., LOCKWOOD, K. L., MARTIN, A . R . & JACKSON, K. S.,

1981: Petroleum potential of the Bass Basin. BMR J. Aust. Geol. Geophys. 6, 199-212. NICHOLAS,

E.,

RIXON,

HAUPT,

A.,

1980:

Uncorrected

geothermal map of Australia. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1980/66 (unpubl.). PASSMORE, V. L. & BOREHAM, C. J., 1982: Source rock chemistry and maturation of the central Eromanga Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 296-306. PITT, G. M., 1982: Geothermal gradients in the Eromanga-Cooper Basin region; in Moore, P. S. & Mount, T. J. (compilers)

and gas composition relationships in the southern Cooper Basin. APEA J. 29(1), 191-200. SENIOR, B. R. & HABERMEHL, M . A., 1980: Structure, hydrodynamics

and hydrocarbon potential of the central Eromanga Basin, Queensland, Australia. BMR J. Aust. Geol. Geophys. 5, 47-56. SENIOR, B. R., MOND, A . & HARRISON, P. L., 1978: G e o l o g y of the

Eromanga Basin. Aust. Bur. Miner. Resour. Geol. Geophys., Bull. 167. SMYTH, M. & SAXBY, J. D., 1981: Organic petrology and geochemistry of source rocks in Pedirka-Simpson Desert Basin, central Australia. APEA J. 21(1), 187-99. SPRIGG, R. C., 1982: The long lead-up to commercial oil discovery in the Mesozoic Eromanga Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 2-19. TISSOT, B. P. & WELTE, D. H . , 1978: Petroleum

formation

VINE, R. R., 1976: Eromanga Basin; in Knight, C. L. (ed.) Economic geology of Australia and Papua New Guinea—3. Australas. Inst. Min. Metall., Monogr. 7, 306-9.

APPENDIX 1. Central Eromanga Basin source rock data base (refer Fig. 2 for key to time intervals).

Source rock chemistrya Well Interval Depth (m) Arrabury 1 Balfour 1 Betoota 1

Boree 1 Buckabie 1

Budgerygar 1! Bury 1 Canaway 1

Carlow 1 Chandos 1

• • * • • • * • * • A • * • • • • • • * * * • • • • A • • • •

TOC %

800. l b 1933.9b 936.6 1.28 1283.2 2.00 1460.0 0.20 1591.1 4.30 1746.8 0.15 163.6 1.23 919.0 1.20 344.0 2.42 995.0 0.78 1207.0 0.68 1508.4 3.60 697.0 0.69 334.4 20.4 500.2 1.45 921.1 0.25 1091.2 0.65 1212.8 1.15 1330.1 0.45 1416.1 6.15 1217.7 1.25 123.4b 531,9b 1427.7 1.15 1542.2b 3.00 1615.4 5.70 1805.9 68.1 1805.9 16.2 1806.0 16.2 1807.4b 13.4

EOM ppm

715 1892 533 3281 105 610 548 1123 1085 1310 2947 1058 5357 212 142 190 1058 429 6118 578

505

133356 12355 6153

SATS

ppm

AROM ppm

POLAR ppm

85 555 41 330 24 46 75

60 625 33 1404 10 34 35

210 20 266 43 164 423

743

312

846

316 57 18 22 69 60 118 67

330 26 12 6 125 49 1645 180

654 88 82 31 276 153 1700 179

166

137

183

699 1594

and

occurrence—a new approach to oil and gas exploration. Springer-Verlag, Berlin.

55620 1915

ASPH

ppm

351

42301 3830

279

619

3336

VR °7o 0.46 0.45 0.52 0.63 0.76 0.95 0.99 0.38 0.42 0.45 0.55 0.57 0.55 0.55 0.47 0.46 0.50 0.69 0.49 0.48 0.52 0.45 0.36 0.38 0.54 0.67 0.70 0.62 0.71 0.71

VITR EXIN INEF °7o °7o °7o

43 94

3 3

49 3

53 76

14 8

33 16

1 1 1

41 10 52 94

14 1

85 32 4

SOURCE ROCKS AND MATURATION Chandos South 1 Cumbroo 1

Etonvale 1

Fairlea 1 Galway 1 Gilmore 1 Gilpeppee 1

• •* • • • • • •

Orientos 1

Quilberry 1

1.30

1320

2246.0

2.74

3170

1701.4

0.75

363

• • A A

• *

•

• A

• •

• • • • •

• A A A A A A A

* *

•* *

Westbourne 1

Yongala 1

1473 4230 6673

Stafford 1

0.54 25.2 3.13

650.4

• • •*

Thunda 1

1762.0 1767.0 1935.0

•

* Roseneath 1

41410

* Mt Howitt 1

39.6

195.0 243.8 b 381.0 b 381.0 b 457.2 b 746.7 b 1569.7 b

• •*

Log Creek 1

2072.0

•

•* A A

• •* *

818.3 b 1716.0 b 1898.9 b 2081.7 b 2135.l b 2136.6 b 2424.4

3147

7123

10518

19753

235 0.71

3 3 3

0.45 0.65 0.64

4 4 4 4 4 4 4

3.30 0.47 0.48 0.48 0.47

10.0 9.9 1.90 3.00 159

346

132

326

640

0.37

110

0.93

1 4 4 4 1 4 1

371

0.50 0.49 0.51

1 1 1

0.37 0.73 0.88

1.70 1.30 2.70

0.89 1.00 0.65

1145

492.2 b 1290.8b 1730.6

3.35

1284.0 1420.0

0.39 1.60

1200 4173

0.45 0.54

1440.0 1584.0 1584.4

0.26 1.76 1.85

11958 1623 1386

0.51 0.56 0.56

192.0 b 210.3 b 655.3 b 655.3 b 719.3 b 1020.0

1.10 7.40 12.7 12.6 3.70 1.16

1239.0 b 1266.4 b 1293.8 b 1321.3 b 1348.7 b 1376.l b 1412.7 b 1476.7 b 1522.4 b 1586.4 b 1607.8 b 1623.0 b 1638.3 b 1679.4 b 1696.2 b

2.04 1.63 1.57 0.93

2058

602

446

574

402

0.72 0.47

560

3 3

4 4 4 4 4 3

0.57

903

3 3 1

3 3 3 3 3 3 3 3 3 3 3 3 3 3 3

0.65 0.90 0.98 1.52 1.25 1.10 27.7 1.64 2.42 7.51 4.56 286

570

791

1383.7

4.06

3395

1879.0 2108.0 2115.0

1.15 10.4 3.24

1753 1533 7795

171.6 617.8 930.8

0.90 2.40 5.45

206 2207 4357

12 331 270

15 190 344

59 541 745

1355.4 1390.8 1561.2 1723.3 1825.7 1914.7

0.50 0.10 0.40 0.50 1.20 1.05

591 42 380 414 590 753

47 8 77 61 107 123

12 10 74 96 287 374

306 40 141 122 128 187

1518

0.46

109 746 2666

0.47 0.46 0.58 0.44 0.49 0.56 0.56 0.51

2 3 3 3

0.40 0.71 0.71 32 7

30 12 17

2 2 2

40 84

60 13

1 1 1 1 1 1

70 56 76

a. Data reproduced from BMR Source Rock Data base b. Cuttings TOC Total organic carbon EOM Extractable organic matter SATS Saturated hydrocarbons AROM Aromatic hydrocarbons POLAR N,S,0-polars (pentane soluble) A S P H N,S,0-polars (pentane insoluble) VR Vitrinite reflectance V1TR Vitrinite EXIN Exinite INER Fusinite + Semi-fusinite + Inertodeterinite 1. CSIRO 2. A M D E L 3. Robertson Research 4. Shell • Early-Late Cretaceous A E . Cretaceous/L. Jurassic • Middle Jurassic Early Jurassic

V. L. PASSMORE & C. J. BOREHAM

236 Rock-Eval pyrolysis data Well Name Bodalla l

Durham Downs l e

Formation

Depth 3 (m)

TOC %

Tmax °C

HI b

OI c

Petroleum d Potential 0.3 0.4 2.0 0.6 0.5 0.8

Cadna-owie Fm

1219.2-1228.3 1234.9-1240.5 1249.7-1255.8 1264.9-1274.1 1280.2-1289.3 1295.4-1304.5

1.99 1.26 4.48 2.26 2.17 2.05

442 440 444 440 444 445

16 35 44 26 23 37

19 14 39 27 29 26

Hooray Ss

1310.6-1319.8 1325.9-1341.1 1341.1-1350.3 1356.4-1362.5 1371.6-1386.8 1386.8-1396.0 1402.1-1414.3 1423.4-1432.6

3.99 2.03 4.05 3.70 1.17 1.72 1.56 1.20

447 449 443 450 445 450 441 443

117 48 37 51 50 46 22 18

11 12 38 19 29 20 30 38

0.5 1.0 1.5 1.9 0.4 0.8 0.4 0.2

Westbourne Fm

1432.6-1447.8 1447.8-1453.0 1453.9-1463.0 1472.2-1475.2 1478.3-1484.4

1.53 1.77 2.26 4.51 3.55

446 445 441 444 445

15 37 41 25 11

45 38 30 47 64

0.2 0.7 0.9 1.1 0.4

Adori Ss

1496.6-1508.8 1508.8-1521.0 1524.0-1530.1 1539.2-1545.3 1554.5-1560.6 1569.7-1572.8

1.45 1.21 1.09 2.18 4.34 1.12

447 441

29 21 27 74 106 25

11 20 27 10 12 23

0.4 0.3 0.3 1.6 4.6 0.3

Birkhead Fm

1585.1-1594.1 1606.3-1615.4 1615.4-1624.6 1630.7-1636.8 1645.9-1658.1

3.38 2.51 1.60 1.89 3.82

450 450 451 442

77 126 47 64 188

30 18 50 25 19

2.6 3.2 0.8 1.2 7.2

Hutton Ss

1661.2-1670.3 1676.4-1691.6 1694.7-1697.7 1706.9-1722.1 1725.2-1734.3 1737.4-1743.5

5.79 6.02 1.18 2.18 4.81 1.06

440 443 440 445 440 440

282 75 33 58 127 59

11 10 37 19 46 28

16.3 4.5 0.4 1.3 6.1 0.6

Evergreen Fm

1752.6-1761.7

1.38

442

0.5

Precipice Ss

1783.1-1792.2

1.25

442

0.4

Wallumbilla Fm

1219.2-1222.3 1228.3-1237.4 1246.6-1255.8 1280.2-1289.3 1289.3-1298.5 1307.6-1316.7 1324.4-1333.8 1342.9-1352.1 1352.1-1361.5 1371.3-1380.7 1380.7-1390.5 1402.1-1405.1 1417.3-1420.4

2.22 2.49 2.85 1.08 1.19 0.71 0.95 0.81 0.81 0.86 1.11 1.38 1.38

441 444 443 439 437 440 439 440 442 446 443 443

39 42 46 24 37 25 22 22 23 33 32 36 45

34 39 33 27 38 13 8 8 16 11 11 14 11

0.9 l.i 1.3 0.3 0.4 0.2 0.2 0.2 0.2 0.3 0.4 0.5 0.6

Cadna-owie Fm

1432.6-1435.6 1447.8-1450.8 1463.0-1466.1 1478.3-1481.3 1493.5-1496.6 1508.8-1511.8

1.27 1.11 2.05 1.14 1.56 0.99

443 444 446 442 444 445

47 43 39 48 51 60

16 16 26 24 38 14

0.6 0.5 0.8 0.6 0.8 0.6

Hooray Ss

1524.0-1533.1 1539.2-1542.3 1563.6-1566.7 1569.7-1572.8 1585.0-1588.0 1600.2-1603.2 1615.4-1618.5 1630.7-1636.8 1645.9-1649.0 1658.1-1667.3 1676.4-1679.4

1.81 0.67 0.71 1.96 1.34 2.40 1.54 9.87 3.06 2.77 0.47

443 445 444 443 445 444 443 444 445 447 449

54 45 227 63 55 128 140 236 130 161 105

35 49 143 35 41 36 45 9 21 21 44

1.0 0.3 1.6 1.2 0.7 3.1 1.6 23.2 4.0 4.5 0.5

444 445 437

SOURCE ROCKS AND MATURATION Well Name

Gal way l e

Formation

Depth 3 (m)

TOC

237 HI b

OI c

Petroleum' Potential

87 83 83 151 185 312 115 93

31 79 49 108 65 42 38 36

0.7 1.0 0.5 1.0 1.0 11.7 1.6 0.9

Tmax °C

Westbourne Fm

1691.6-1694.7 1706.9-1713.0 1722.1-1737.4 1737.4-1740.4 1752.6-1755.6 1767.8-1770.9 1783.1-1786.1 1798.3-1801.4

0.75 1.16 0.61 0.68 0.55 3.75 1.41 0.97

448 447 448 445 445 442 445 448

Adori Ss

1813.6-1816.6

0.36

446

166

0.2

Birkhead Fm

1828.8-1837.9 1844.0-1847.1 1859.3-1862.3 1874.5-1877.6 1886.7-1889.8

2.71 9.74 1.01 0.87 1.78

451 450

333 405 241 90 246

19 7 23 20 24

9.0 39.5 2.4 0.8 4.4

Hutton Ss

1917.2-1920.2 1938.5-1941.6 1950.7-1953.8 1966.0-1969.0 1981.2-1984.2 1996.4-2008.6 2011.7-2014.7 2026.9-2030.0 2042.2-2045.2 2057.4-2060.4 2078.7-2081.8 2106.2-2109.2 2121.4-2124.5

0.58 0.39 0.15 14.30 3.79 1.39 0.82 2.66 0.60 0.88 0.66 1.26 0.79

453 447 450 451 445 451 452 449 449 450 450 450

160 122 110 93 271 207 116 244 113 97 95 124 127

62 63 51 4 16 26 31 19 44 38 46 35 44

0.9 0.5 0.2 13.3 10.3 2.9 1.0 6.5 0.7 0.9 0.6 1.6 1.0

Mackunda Fm

1219.2-1222.3 1234.4-1237.5 1249.7-1252.7 1264.9-1268.0 1271.0-1274.1

1.14 1.38 1.17 5.69 6.72

440 435 435 434

96 156 92 145 605

28 30 22 13 11

1.1 2.2 1.1 0.3 40.7

1280.2-1283.2 1295.4-1298.4 1310.6-1313.7 1325.9-1328.9 1341.1-1344.2 1356.4-1359.4 1374.6-1377.7 1389.7-1392.9 1402.1-1405.1 1417.3-1420.3 1432.6-1435.6 1447.8-1450.8 1463.0-1466.1 1478.3-1481.3 1493.5-1496.6

2.89 2.33 7.26 3.81 3.62 1.13 4.04 1.03 0.78 0.93 0.93 0.99 1.90 1.26 1.18

263 545 591 408 420 75 483 38 31 55 64 41 151 60 70

9 8 17 17 11 19 18 15 13 13 11 7 7 15 7

7.6 12.7 42.9 15.5 15.2 0.9 19.5 0.4 0.3 0.5 0.6 0.4 0.3 0.8 0.8

Cadna-owie Fm

1508.8-1511.8 1524.0-1527.0 1539.2-1542.3 1554.5-1557.5 1572.8-1575.8 1585.0-1588.0

1.25 1.37 1.77 1.35 1.31 1.32

436 434 440 435 437 439

72 81 59 45 61 65

15 9 7 8 10 8

0.9 1.1 1.1 0.6 0.8 0.9

Hooray Ss

1600.2-1603.2 1615.4-1618.5 1630.7-1633.7 1645.0-1649.0 1661.2-1664.2 1676.4-1679.4 1691.6-1694.7 1706.9-1709.9 1722.1-1725.2

1.57 1.21 0.96 1.18 1.17 1.25 1.29 1.18 1.20

438 441 439 440 440 441 440 440 440

51 41 41 45 61 59 45 47 50

9 11 17 13 10 9 8 11 9

0.8 0.5 0.4 0.5 0.7 0.7 0.6 0.6 0.6

Westbourne Fm

1737.4-1752.6 1752.6-1755.6 1767.8-1770.9 1783.1-1786.1 1798.3-1801.4 1813.6-1816.6 1828.9-1831.8 1844.0-1847.1

1.29 1.06 1.09 1.21 1.16 1.22 2.07 1.58

441 442 442 441 442 441 444 444

71 45 40 43 76 42 75 64

9 14 11 13 13 15 11 8

0.9 0.5 0.4 0.5 0.9 0.5 0.2 0.1

Wallumbilla Fm

451

435 433 433 431 431 435 433 436 435

436 438 438

V. L. PASSMORE & C. J. BOREHAM

238

Jackson l f

Adori Ss

1859.3-1862.3 1871.5-1874.5

1.25 0.90

442 442

60 70

10 9

0.1 0.1

Birkhead Fm

1889.8-1892.8 1917.2-1920.2 1935.5-1938.5 1950.7-1953.8 1966.0-1969.0

1.85 1.91 2.63 1.97 1.45

447 446 446 444 443

87 128 135 70 39

18 8 6 9 13

0.1 0.1 0.3 0.1 0.05

Hutton Ss

1981.2-1984.2 1996.4-1999.5 2011.7-2014.7 2026.9-2030.0 2045.2-2048.3 2057.4-2060.4 2072.6-2075.7 2087.9-2090.9 2103.1-2106.2 2121.4-2124.5 2133.6-2136.6 2148.8-2150.4 2164.1-2165.6 2182.4-2183.9 2194.6-2196.1

1.21 1.27 1.10 1.65 1.33 1.38 1.24 0.96 1.44 1.39 1.64 1.44 1.28 1.07 1.32

443 441 443 442 443 442 442 443 442 442 441 442 439 441

49 68 51 61 61 80 49 42 52 54 98 61 38 30 37

9 6 11 7 10 7 10 7 8 10 4 5 6 17 8

0.1 0.1 0.05 9.1 0.1 1.1 0.6 0.4 0.8 0.8 1.6 0.9 0.5 0.3 0.5

Evergreen Fm

2209.8-2211.3 2225.0-2226.6 2240.3-2241.8 2255.5-2257.0

1.04 2.93 1.24 1.17

442 456 446 453

67 61 65 58

9 8 18 12

0.7 1.8 0.8 0.7

Precipice Ss

2270.8-2272.3 2284.5-2286.0 2302.7-2304.3 2315.0-2316.5

1.52 14.6 1.09 1.22

453 450 435 438

51 214 57 91

8 4 14 9

0.8 31.2 0.6 0.6

Winton Fm

45.7- 64.0 91.4- 109.7 137.2- 155.4 182.9- 201.2 256.0- 274.3 274.3- 292.6 347.5- 365.8

1.39 1.20 2.20 1.34 14.30 6.90 7.60

444 448 447 446 433 438 433

33 25 26 35 50 58 86

50 150 172 131 54 55 54

0.5 0.3 0.6 0.5 7.3 4.0 6.6

Mackunda Fm

374.9- 393.2 411.5- 429.8 457.2- 475.5

1.55 3.52 2.80

441 442 443

35 60 36

160 61 86

0.5 2.1 1.0

Allaru Mdst

502.9- 521.2 548.6- 566.9 594.4- 611.7 640.1- 658.4

1.39 1.19 1.39 1.49

440 435 436 436

38 34 28 40

95 62 42 59

0.5 0.4 0.4 0.6

Wallumbilla Fm

685.8- 704.1 731.5- 749.8 780.3- 798.6 823.0- 868.7 868.7- 887.0 914.4- 932.7 960.1- 975.4 1005.8-1021.1

1.74 1.77 3.12 1.06 1.51 1.00 1.13 1.15

435 438 441 431 437 436 438 441

58 53 40 45 33 29 22 54

40 49 48 42 46 29 15 29

1.0 0.9 1.3 0.5 0.5 0.3 0.4 0.6

Transition Beds

1051.6-1066.8

0.51

442

209

1.1

Murta Mbr

1097.3-1112.5 1143.0-1158.2

0.73 0.99

440 441

49 102

58 57

0.4 1.0

Namur Ss Mbr

1188.7-1204.0 1234.4-1252.7

0.87 0.32

428 437

117 31

137 146

1.6 0.2

Westbourne Fm

1280.2-1298.4 1325.9-1344.2

0.82 1.38

430 426

90 96

70 79

2.9 3.6

Adori Ss

1371.6-1389.9

0.98

434

108

100

2.1

Birkhead Fm

1417.3-1432.6

1.83

437

118

3.0

Hutton Ss

1463.0-1478.3

1.43

436

1.5

SOURCE ROCKS AND MATURATION Well Name Kyabra l f

Morney l f

Formation

Depth 3 (m)

239

TOC <7o

Tmax °C

HI b

Petroleum d Potential

Winton Fm

125-140 170-185 215-230 260-275 305-320 350-365 395-410 440-455 485-500 530-545 575-590 625-640 675-685 695-710 715-730 760-775 805-820 850-866

1.10 5.40 0.90 10.90 1.50 1.10 1.50 2.30 3.80 2.80 11.50 21.60 4.40 1.30 0.80 2.10 1.90 1.90

442 458 450 433 442 441 436 441 434 441 425 424 435 434 431 437 438 432

33 21 34 75 38 30 34 58 41 25 104 106 66 43 48 52 42 46

88 85 85 63 74 60 48 48 91 25 50 40 66 53 178 127 58 81

0.4 1.2 0.3 8.2 0.6 0.3 0.5 1.4 1.6 0.7 12.0 23.0 2.9 0.6 0.4 1.1 0.8 0.9

Mackunda Fm

940-955 1030-1046 1075-1090

1.30 1.50 10.30

437 433 419

50 56 440

30 59 13

0.7 0.8 45.3

Wallumbilla Fm

1120-1135 1200-1215 1245-1260 1290-1305 1335-1350 1380-1390

2.25 1.60 1.10 1.10 1.15 1.75

447 448 452 453 450 455

568 495 65 75 137 184

57 110 35 37 38 43

12.9 8.1 0.8 0.9 1.6 3.3

Cadna-owie Fm

1425-1440 1470-1480 1515-1530

1.40 0.85 1.10

450 459 457

178 260 196

46 180 103

2.6 2.3 2.2

Hooray Ss

1605-1620

2.05

459

254

169

5.3

Westbourne Fm

1650-1665 1680-1695 1725-1740

1.40 1.25 1.15

457 452 455

467 234 259

300 312 329

6.6 3.0 3.1

Adori Ss

1770-1780

1.25

456

148

127

1.9

Birkhead Fm

1815-1830 1860-1870

5.40 1.25

453 460

779 545

85 137

42.9 16.3

Hutton Ss

1905-1920 1940-1955 1980-1995 2025-2040 2055-2070

1.50 1.10 1.20 4.05 2.65

459 460 455 456 459

504 555 596 284 343

155 309 290 189 178

7.7 6.2 7.3 9.3 6.8

Evergreen F m / Precipice Ss

2100-2115 2140-2155

1.25 1.45

454 456

527 513

211 824

7.7 45.6

Winton Fm

243.8- 259.1 289.6- 304.8 335.3- 350.5 381.0- 396.2

1.30 6.95 2.05 1.30

440 432 440 438

70 127 84 93

139 63 88 80

1.0 9.1 1.9 1.3

Mackunda Fm

426.7- 449.6 472.4- 487.7 518.2- 533.4 563.9- 579.1 609.6- 624.8 655.3- 670.6 701.0- 716.3

1.00 1.20 1.25 1.20 1.05 0.95 1.10

437 435 419 436 431 433 437

71 36 68 60 47 60 69

82 36 128 55 109 167 49

0.6 0.8 1.0 0.9 0.6 0.7 0.9

Wallumbilla Fm

746.8- 762.0 792.5- 807.7 838.2- 853.4

3.00 1.25 0.90

424 429 431

346 156 133

14 26 47

10.8 2.1 1.4

V. L. PASSMORE & C. J. BOREHAM

240 Well Name

Formation

Depth 3 (m)

Tmax °C

HIb

OIc

Petroleumd Potential 1.5 0.7

TOC

883.9- 899.2 929.6- 944.9

1.40 0.90

432 434

96 65

36 24

Transition Beds

975.4- 990.6

1.20

437

Murta Mbr

1021:1-1036.3

1.30

441

119

2.1

Namur Ss Mbr

1066.8-1082.0 1112.5-1127.8

0.80 2.50

435 438

82 131

37 32

0.9 4.0

Westbourne Fm

1158.2-1173.5 1204.0-1219.2

0.70 0.90

437 435

82 88

64 72

0.7 1.1

Birkhead Fm

1341.1-1356.4

1.05

440

163

2.3

Hutton Ss

1386.8-1402.1 1432.6-1447.8 1524.0-1539.2 1569.7-1585.0 1615.4-1630.7 1661.2-1676.4 1706.9-1722.1 1752.6-1767.8

0.20 0.30 0.30 5.80 1.85 1.10 2.35 2.05

438 437 436 443 428 441 439 445

170 116 126 283 185 156 170 168

70 70 50 5 21 15 17 12

0.5 0.5 0.6 19.3 4.1 2.1 4.6 5.4

Depth from K.B. b H I Hydrogen Index: S 2 /TOC (mg hydrocarbon/g TOC) O I Oxygen Index: S 3 /TOC (mg hydrocarbon/g TOC) d Petroleum Potential: S, +S 2 (mg hydrocarbon/g rock) TOC and Rock-Eval by Robertson Research f TOC by A M D E L , Rock-Eval by B M R

1.3

Geological Society of Australia Special Publication No. 12, 241-254

Geochemistry of oil shale in the eastern Eromanga Basin J. D. Saxby

CSIRO Division of Fossil Fuels, P.O. Box 136, North Ryde, NSW 2113.

ABSTRACT

Organic geochemical data are presented for 154 core samples from 15 shallow holes (usually less than 100 m) drilled along the eastern margin of the Eromanga Basin from Longreach to Charleville. Oil assay data are given for 46 of these samples. Correlations between geochemical parameters (organic carbon, carbonate carbon, pyrite and oil yield), geophysical logs (e.g. neutron, gamma ray) and lithostratigraphic units are shown diagrammatically. Organic contents and oil yields tend to decrease towards the south. The atomic H/C ratio of Toolebuc Formation kerogen (1.1 ± 0.2) is relatively constant and indicates more than 50% aromaticity. On average, 27% of organic matter is converted to oil on pyrolysis, while 53% remains as char and 20% is evolved as gas. The nature of organic matter present in small amounts in the time-equivalent 'Urisino beds'- in the south is unclear but the data are consistent with more aromatic, terrestrially-derived material. Partial or extensive weathering to depths of 50 m or more has altered kerogen, as well as the distribution of inorganic minerals, resulting in decreased oil yields.

INTRODUCTION is described in other papers in this volume, as well as by Senior The largest and most widespread oil shale deposit in et al. (1975, 1978) and Exon & Senior (1976). In this paper the geochemistry of oil shale near the eastern Australia occurs in the Cretaceous Toolebuc Formation of the Eromanga Basin (Ozimic, 1982). An inferred in situ resource margin of the Eromanga Basin is investigated. Samples have of 3 8 3 8 x 10 1 of shale yielding an average 60 L t has been taken from 15 shallow holes drilled between Longreach been calculated (Gibson, 1981). The most intensive exploration and Charleville, a distance of over 300 km (Table 1, Fig. 1). has been centred near Julia Creek where the Toolebuc The holes were located far enough into the basin away from Formation oil shale outcrops on the St Elmo Structure. Data outcrops, for the Toolebuc Formation to be encountered on Toolebuc Formation oil shale outcropping on the rims of beyond the extensively-weathered surface zone. The Toolebuc the Eromanga Basin in the east, south and west are much Formation was continuously cored and samples were chosen more sparse, in the centre of the basin, where the Toolebuc in order to reflect all significant vertical lithological variations. Formation is intersected at 1000 m or more, information is Cores and cuttings were examined megascopically and their even more restricted, coming only indirectly from petroleum lithologies were correlated with petrophysical logs (Ozimic, exploration activities. The geology of the Eromanga Basin 1981a, b\ Stephenson, 1982). The drilling, logging and 9

TABLE 1. Location of studied stratigraphic holes in the eastern Eromanga Basin and depth of weathered zone in relation to Toolebuc

Formation. All depths below ground level.

Base of extensive weathering Total depth (m) Longitude (m) Latitude Well 145°19'16"E 12.0 Longreach 6 41.0 23 °37 '35 "S 71.2 9.7 145 °31 '18" Jericho 11 23 °51 '30" 18.0 24 °15'00" 145 °39 '07" 65.6 Tambo 38 145 °47 '04" 70.2 15.0 24 °32 '58" Tambo 40 47.9 15.0 24 °29'56" 145 °56 '46" Tambo 41 15.0 36.3 146°06'19" 24 °51 '21" Tambo 42 33.2 10.3 146°09'25" 24 °58 '22" Tambo 44 31.7 5.0 146 °11 '10" 25 °03 '55" Augathella 5 99.4 12.0 146 °05 '35" 25 °36'13" Augathella 6 99.3 23.0 146 °42 '02" Augathella 7 25 °01 '45" 85.7 20.0 146°20'00" 26 °28'30" Charleville 3A ^35.0 32.1 146 °21 '42" 26 °11'18" Charleville 4A 50.0 78.8 146 °02 '04" 26 °23'17" Charleville 5 85.2 14.0 146 °09' 26 °50' Charleville 6 151.8 10.0 146°09' 26 °50' Charleville 7 * Refers to Urisino beds (time-equivalent of the Toolebuc Formation) (Ozimic, 1982) t Not present § Undifferentiated

Top of Toolebuc Formation (m) 25.0 55.9 58.0

Thickness of Toolebuc Formation (m) 6.0 9.3 5.4

35.9 16.7 15.2 14.6 54.5 72.5 t t 60.5 64.9* 130.5*

8.3 2.8 5.8 1.2 10.5 11.5 t t 12.6 15.5* 9.7*

242

J. D. SAXBY

H<°

147°

scanning. Wide fluctuations in parameters are possible, particularly where depth increments are large. Organic carbon is taken as the difference between total and carbonate carbon; pyrite is calculated as 1.875 x pyritic sulphur. Individual holes are not discussed in detail but some broad implications are apparent. A close correlation between petrophysical and geochemical logs has not been achieved, and, if this were desired in any future study, much closer geochemical sampling, together with precise, calibrated wireline logs, would be required.

Carbon analyses

Fig. 1. Location of stratigraphic holes in the Eromanga Basin. (L=Longreach; J=Jericho; T=Tambo; A=Augathella; C = Charleville).

analytical work form part of a Commonwealth Scientific and Industrial Research Organization-Bureau of Mineral Resources (CSIRO-BMR) joint research project on exploration methodology applicable to a widespread oil shale deposit such as that within the Eromanga Basin (Ozimic & Saxby, 1983).

ANALYTICAL METHODS

All geochemical analyses were carried out on core samples which were visually free from contamination. Cores were crushed and ground to less than 0.2 mm and air-dried under laboratory conditions before determining moisture, ash, total carbon, carbonate carbon, hydrogen, nitrogen, total sulphur, pyritic sulphur and sulphate sulphur. In each case Australian or British standard methods developed for coals were used. Because sample sizes were limited, oil assays were carried out by a new method developed at CSIRO (Watson, 1984). This procedure uses a much smaller sample than the modified Fischer assay but is otherwise reasonably comparable. All samples containing more than 5 per cent organic carbon (and selected lower grade samples) were analysed in this way for oil, water, char and gas + loss yields. Results for all samples are given in Tables 2 and 3.

RESULTS AND DISCUSSION

Chemical parameters in Table 2 can be related to stratigraphic units, lithology and various petrophysical logs (spontaneous potential, resistivity, density, neutron and gamma-ray). Results for all holes are given in Figures 2-16. Solid trend lines have been used to join data points for visual

On the whole, organic carbon contents are fairly low. The highest value obtained from the 154 samples tested is 13 per cent in Tambo 38. Even in this hole, which has the greatest carbonaceous content of the 15 examined, an organic carbon content of more than 10 per cent is maintained over less than 4 m. Excluding samples from the overlying Allaru Mudstone and underlying Coreena Member, average organic carbon values for the Toolebuc Formation are as follows: Longreach 6 (5.9 per cent), Jericho 11 (5.2 per cent), Tambo 38 (9.8 per cent), Tambo 41 (8.3 per cent), Tambo 44 (6.2 per cent), Augathella 5 (4.0 per cent), Augathella 6 (2.8 per cent), Augathella 7 (3.1 per cent), Charleville 5 (2.9 per cent), Charleville 6 (2.1 per cent), Charleville 7 (1.2 per cent). Values for Charleville 6 and 7 refer to the Urisino beds, which consists of interbedded siltstone and sandstone and is time equivalent to the Toolebuc Formation. Carbonate carbon values are highly variable, ranging from 0 up to a value of 8 per cent for a Toolebuc Formation sample in Longreach 6. The latter value corresponds to 67 per cent CaC0 assuming calcite is the mineral present. Even within 'limestone' of the Toolebuc Formation, carbonate contents may be low and a significant drop does not always occur at the boundaries of the Toolebuc Formation with the Allaru Mudstone and the Coreena Member of the Wallumbilla Formation. Similarly, no clear correlation between organic and inorganic carbon exists. Overall, carbonate carbon contents decrease to the south. Average Allaru Mudstone, Toolebuc Formation (Urisino beds) and Coreena Member values for the most southerly holes at Charleville are 0.1, 0.4 and 0.1 per cent respectively. 3

Nitrogen analyses

Although nitrogen levels are low, precise analytical work enables a clear relationship between organic carbon (and oil yield) and nitrogen to be observed. This implies the presence of mainly organic nitrogen in these samples, i.e. a dearth of inorganic nitrogen-containing minerals. In fact, for the Toolebuc Formation as a whole, nitrogen determination is probably the best single elemental analysis to use as a general indicator of oil yield. Organic carbon cannot of course be measured directly in the presence of inorganic carbon. For the 32 samples containing more than 5 per cent organic carbon, a linear relationship is observed between nitrogen and organic carbon. The slope of this line (0.033) corresponds to a N/C atomic ratio of 0.029. The N/C ratio for seven unweathered Toolebuc Formation kerogens from Julia Creek, Boulia and Mayneside in the Eromanga Basin and from Burketown and Dobbyn in the Carpentaria Basin is 0.028 ± 0.04 (Riley & Saxby, 1982). This indicates that the richest oil shale samples in the eastern Eromanga Basin contain kerogen which has a similar nitrogen content to Toolebuc Formation kerogens up to 700 km to the west and northwest. Below 5 per cent organic carbon, N/C values tend

OIL SHALE GEOCHEMISTRY

243

TABLE 2. Analytical results on core samples from the Eromanga Basin. Depth (m) and formation*

Moisture

Total carbon

Analyses (%, air-dried basis) Carbonate Total carbon Hydrogen Ash Nitrogen sulphur

Pyritic sulphur

Sulphate sulphur

Augathella 6

51.40-52.40 A 54.80-55.10 T 55.60-55.80 T 56.03-56.15 T 56.25-56.40 T 56.80-57.00 T 57.80-58.00 T 58.45-58.64 T 59.25-59.40 T 59.75-59.90 T 60.30-60.40 T 60.80-60.96 T 61.70-71.96 T 62.40-62.60 T 63.20-63.40 T 63.80-64.08 T 64.95-65.15 C 65.80-66.00 C

7.0 7.4 6.0 5.8 4.7 5.2 6.3 5.6 6.7 5.6 6.1 6.3 6.2 6.9 8.1 6.8 7.7 7.4

1.8 2.7 2.9 4.3 8.0 6.1 3.0 2.8 2.5 2.2 2.2 2.0 1.7 1.8 2.0 3.0 3.1 2.1

0.0 0.0 0.0 0.4 1.1 0.2 0.0 0.0 0.0 0.1 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

1.4 1.6 1.4 1.5 1.6 1.6 1.4 1.2 1.4 1.2 1.3 1.4 1.3 1.4 1.5 1.4 1.5 1.4

86.1 84.6 85.7 84.1 79.2 82.7 85.2 86.4 85.2 86.8 85.6 86.5 86.6 86.0 84.8 84.7 83.8 86.1

0.09 0.13 0.15 0.19 0.29 0.26 0.14 0.13 0.12 0.12 0.10 0.11 0.12 0.10 0.09 0.15 0.14 0.09

0.84 0.91 0.82 0.87 1.04 1.15 1.17 1.47 1.17 1.23 1.26 1.04 0.87 1.01 1.05 1.09 1.33 1.35

0.60 0.82 0.76 0.77 0.77 0.84 0.96 1.24 1.04 1.02 0.99 0.84 0.69 0.82 0.91 0.83 1.19 1.19

0.09 0.07 0.05 0.04 0.07 0.10 0.14 0.17 0.14 0.22 0.15 0.15 0.16 0.12 0.11 0.16 0.11 0.15

Augathella 7

71.45-71.60 A 72.90-73.10 T 74.15-74.30 T 75.15-75.30 T 75.70-75.86 T 76.70-76.88 T 78.15-78.30 T 78.90-79.10 T 79.95-80.14 T 81.15-81.30 T 82.00-82.21 T 83.10-83.27 T

7.1 6.8 7.2 6.7 7.0 7.3 6.4 6.2 7.2 6.5 7.4 7.0

3.2 5.5 2.9 6.2 2.8 2.3 2.1 2.1 2.3 3.0 3.4 2.2

0.2 0.2 0.1 0.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0

1.6 1.7 1.5 1.8 1.5 1.5 1.4 1.4 1.5 1.4 1.6 1.4

84.4 81.8 84.7 81.0 84.9 85.3 86.3 86.6 85.3 85.4 84.1 86.3

0.15 0.25 0.14 0.27 0.14 0.11 0.11 0.11 0.11 0.14 0.16 0.10

0.88 1.01 1.00 1.05 1.25 1.22 1.05 0.94 1.03 1.09 1.32 1.33

0.82 0.89 0.93 0.89 1.08 0.74 0.72 0.67 0.73 0.49 0.92 0.85

0.08 0.07 0.06 0.07 0.08 0.07 0.08 0.06 0.07 0.10 0.07 0.11

Charleville 3A

50.59-50.65 A 52.55-52.65 A 55.55-55.68 A 58.90-59.00 A 61.60-61.70 A 64.52-64.62 A 67.15-67.24 A 69.65-69.70 A 72.35-72.45 A 74.25-74.36 A 76.95-77.05 C 77.95-78.10 C 78.95-79.10 C 79.60-79.70 C

3.8 3.9 3.3 3.1 3.2 2.9 3.4 2.7 4.4 4.7 3.7 3.8 5.0 3.1

0.1 1.4 1.5 1.4 2.0 1.9 1.7 1.1 1.1 1.4 1.2 1.2 1.1 1.4

0.0 0.0 0.0 0.0 0.4 0.3 0.4 0.3 0.2 0.1 0.2 0.1 0.1 0.4

1.1 1.1 1.1 1.0 1.0 1.0 1.0 0.9 1.0 1.1 1.0 1.0 1.1 0.8

90.0 89.5 89.7 89.9 88.5 88.9 88.8 90.8 89.7 88.9 89.9 90.2 89.7 91.3

n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.

0.03 1.00 1.11 0.82 0.93 0.93 0.80 0.69 0.73 0.92 1.02 1.09 1.06 1.08

n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.

Charleville 4A

19.00-19.10 A 24.30-24.40 A 29.50-29.60 A 31.60-31.70 A

2.0 3.6 7.8 4.6

0.1 0.1 0.1 0.1

0.0 0.0 0.0 0.0

1.2 1.2 1.5 1.1

89.7 89.5 86.4 89.6

n.d. n.d. n.d. n.d.

0.03 0.02 0.02 0.02

n.d. n.d. n.d. n.d.

Charleville 5

59.70-59.80 A 60.66-60.76 T 60.95-61.05 T 61.66-61.76 T 62.05-62.15 T 62.60-62.70 T 63.22-63.32 T 64.25-64.35 T 65.44-65.54 T 66.60-66.70 T 68.70-68.76 T 69.62-69.72 T 70.40-70.50 T 70.60-70.70 T 70.80-70.90 T 72.80-73.00 T 73.00-73.10 C

3.5 3.0 2.9 2.7 4.2 4.2 4.2 2.4 4.2 4.3 3.5 2.8 3.5 3.9 3.4 3.7 3.0

4.6 4.8 5.5 5.1 4.0 3.5 3.7 2.7 2.2 1.8 1.9 2.5 1.9 1.9 3.4 1.5 1.1

0.0 0;0 0.0 0.0 0.0 0.0 0.3 0.5 0.5 0.2 0.3 1.3 0.3 0.2 0.7 0.1 0.0

1.3 1.3 1.4 1.3 1.4 1.3 1.3 1.0 1.2 1.2 1.1 0.8 1.1 1.1 1.1 1.0 0.8

85.2 85.4 84.4 85.2 84.9 85.8 85.4 88.2 85.5 87.7 88.9 87.4 88.6 88.4 86.0 90.0 91.6

0.22 0.24 0.25 0.25 0.21 0.16 0.16 0.12 n.d. n.d. n.d. n.d. n.d. n.d. 0.15 n.d. n.d.

1.09 0.99 1.12 1.03 0.99 1.10 1.06 1.14 1.19 1.21 1.25 0.88 0.97 0.94 1.62 1.48 1.30

0.79 0.70 0.75 0.71 0.75 0.86 0.82 0.94 n.d. n.d. n.d. n.d. n.d. n.d. 1.33 n.d. n.d.

0.10 0.09 0.09 0.10 0.08 0.09 0.09 0.09 n.d. n.d. n.d. n.d. n.d. n.d. 0.15 n.d. n.d.

Well

244

J. D. SAXBY Depth (m) and formation*

Moisture

Total carbon

Pyritic sulphur

Sulphal sulphu

Longreach 6

27.16-27.30 T 28.08-28.23 T 29.50-29.65 T

2.3 1.4 4.0

10.3 13.9 6.6

5.0 8.0 0.0

1.0 0.9 1.4

70.7 61.1 84.6

0.20 0.23 n.d.

1.07 1.17 1.23

0.67 0.69 1.00

0.20 0.26 0.17

Jericho 11

43.05-43.38 A 44.37-44.70 A 45.49-45.81 A 46.70-47.02 A 48.10-48.25 A 48.80-49.10 A 50.24-50.53 A 50.95-51.25 A 51.64-51.93 A 52.22-52.51 A 53.56-53.89 A 54.52-54.87 A 57.08-57.15 T 58.71-58.98 T 59.78-60.11 T 61.93-62.15 T 62.40-62.57 T 64.10-64.15 T 64.78-64.82 T 66.35-66.41 C

5.9 5.5 4.5 4.3 5.0 4.8 4.3 4.4 3.0 3.5 3.8 5.8 5.6 4.2 1.9 3.2 3.2 3.4 4.0 3.3

1.5 1.8 1.4 2.0 1.9 1.6 1.8 1.8 2.4 1.9 1.8 1.9 2.1 2.4 9.0 14.7 13.3 10.2 1.3 1.4

0.1 0.1 0.2 0.4 ,0.3 0.1 0.8 0.5 1.1 0.5 0.2 0.4 0.1 0.8 6.6 3.0 2.6 3.6 0.0 0.1

1.1 1.2 1.0 1.0 1.1 1.1 0.9 1.0 0.8 0.8 0.9 1.1 1.2 1.0 0.7 1.7 1.6 1.2 0.9 0.8

88.1 88.3 88.7 88.7 88.4 89.2 88.5 88.8 88.3 89.7 90.3 87.6 87.5 87.5 67.7 67.9 72.7 73.5 91.0 91.9

n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. 0.43 0.41 0.27 n.d. n.d.

0.59 0.82 0.75 0.92 0.91 0.85 0.65 0.81 0.77 1.02 1.02 1.10 1.44 1.33 0.79 2.29 1.98 1.39 1.52 1.64

0.50 0.65 0.62 0.75 0.84 0.76 0.54 0.77 0.64 0.88 0.94 1.02 1.25 1.21 0.64 1.57 1.34 1.02 1.50 1.64

0.07 0.11 0.12 0.11 0.09 0.09 0.11 0.05 0.11 0.14 0.09 0.14 0.12 0.17 0.11 0.29 0.24 0.18 0.08 0.06

Tambo 38

56.15-56.20 A 56.80-56.89 A 57.15-57.20 A 57.55-57.92 A 58.90-59.00 T 59.28-59.55 T 60.24-60.50 T 60.90-61.30 T 61.60-62.00 T 62.22-62.40 T

4.3 4.5 4.5 2.8 2.5 2.8 3.9 4.2 4.4 4.8

6.3 3.9 2.7 11.1 16.7 16.5 16.3 14.0 9.3 5.4

2.2 1.1 0.3 5.2 6.1 4.6 3.3 3.0 2.1 0.2

1.2 1.1 1.1 1.1 1.4 1.4 1.9 1.8 1.5 1.4

80.0 84.7 87.8 68.1 62.1 68.7 67.6 70.4 76.7 85.1

n.d. n.d. n.d. 0.25 0.40 0.43 0.51 0.44 0.33 0.27

1.10 1.11 1.42 1.05 1.38 5.01 1.86 1.99 1.18 1.08

0.88 0.99 1.16 0.78 0.76 4.19 1.13 1.24 0.88 0.87

0.13 0.08 0.09 0.11 0.17 0.28 0.19 0.35 0.10 0.11

Tambo 40

65.40-65.5X) W 69.20-69.32 W 70.05-70.20 W

5.0 5.9 5.2

2.6 2.5 3.4

0.4 0.2 0.3

1.1 1.2 1.2

87.7 87.4 86.7

n.d. n.d. n.d.

1.06 1.40 1.21

0.91 1.13 0.90

0.11 0.25 0.25

Tambo 41

33.92-34.18 A 34.44-34.75 A 35.33-35.64 A 36.58-36.62 T 37.65-37.77 T 38.55-38.68 T 39.80-39.93 T 40.95-41.11 T 41.85-42.00 T 42.75-42.90 T 43.65-43.78 T

6.0 5.4 5.4 3.5 4.0 3.9 4.4 3.5 3.8 5.4 6.0

2.1 2.6 5.6 7.8 7.4 9.5 11.5 14.4 12.2 7.3 5.3

0.1 0.3 1.8 1.8 1.1 1.7 2.0 3.5 1.6 1.5 0.2

1.2 1.2 1.2 1.2 1.4 1.4 1.7 1.6 1.7 1.5 1.5

87.4 87.1 80.8 80.5 80.8 78.5 75.1 69.1 75.4 79.4 83.8

n.d. n.d. n.d. 0.26 0.31 0.32 0.40 0.41 0.43 0.26 n.d.

1.00 1.00 1.39 1.40 1.37 1.72 1.73 1.82 1.90 1.29 0.99

0.85 0.91 1.10 1.07 1.00 1.22 1.25 1.11 1.33 1.02 0.84

0.09 0.10 0.22 0.17 0.22 0.23 0.15 0.25 0.21 0.10 0.07

Tambo 42

19.75-19.80 C 20.72-20.98 C 23.30-23.46 C 24.15-24.30 C 24.80-24.93 C 25.90-26.10 C 27.15-27.30 C 29.96-29.99 C 30.70-30.93 C

5.5 5.4 5.4 5.1 5.4 5.7 5.8 6.8 6.6

1.0 1.3 1.4 0.8 1.0 0.8 0.7 0.9 1.1

0.4 0.1 0.1 0.0 0.0 0.2 0.0 0.1 0.0

1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.1 1.1

89.3 89.1 88.9 90.2 89.6 89.6 89.9 88.6 88.4

n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.

0.28 0.52 0.82 0.28 0.34 0.33 0.35 0.66 0.89

0.23 0.47 0.83 0.25 0.36 0.29 0.34 0.61 0.81

0.05 0.05 0.09 0.04 0.05 0.06 0.07 0.06 0.06

Tambo 44

15.15-15.28 T 16.10-16.30 T 17.25-17.50 T 17.90-18.11 T 19.30-19.48 T

5.9 6.6 6.0 6.2 8.2

7.0 6.0 9.9 4.8 5.0

0.5 0.0 1.2 0.1 0.1

1.6 1.7 1.8 1.5 1.7

81.2 81.7 76.0 83.4 81.5

n.d. n.d. n.d. n.d. n.d.

2.04 1.70 1.56 1.61 1.98

1.50 1.37 1.16 1.37 1.67

0.28 0.15 0.10 0.10 0.16

Augathella 5

12.25-15.40 T 16.50-16.65 C 16.90-17.02 C

6.5 3.3 7.8

4.0 6.4 1.3

0.0 4.4 0.0

1.5 0.7 1.3

84.2 77.0 87.1

0.19 0.11 0.07

1.25 0.84 0.87

0.77 0.54 0.73

0.36 0.19 0.07

Well

Analyses (%, air-dried basis) Carbonate Total carbon Hydrogen Ash. Nitrogen sulphur

OIL SHALE GEOCHEMISTRY Depth (m) and formation*

245

Analyses (%, air-dried basis) Carbonate carbon Hydrogen Ash Nitrogen

Moisture

Total carbon

Total sulphur

Pyritic sulphur

Sulphate sulphur

Charleville 6

63.10-64.10 A 64.10-64.90 A 64.90-66.00 U 66.00-67.10 U 67.10-68.10 U 68.10-69.10 U 69.10-70.10 U 70.10-71.10 U 71.10-72.00 U 72.00-72.95 U 72.95-74.20 U 74.20-75.10 U 75.10-76.10 U 76.10-77.10 U 77.10-78.10 U 78.10-78.95 U 78.95-80.00 U

5.0 3.7 6.5 8.2 6.8 8.8 6.5 8.7 5.5 7.8 3.8 6.9 7.3 7.6 9.9 6.2 7.7

0.8 0.3 0.8 2.5 1.9 1.5 1.5 1.1 1.3 6.9 1.1 0.9 1.1 1.9 6.4 1.5 6.9

0.0 0.0 0.1 0.8 1.0 0.5 0.5 0.2 0.2 0.1 0.0 0.0 0.2 0.9 0.4 0.1 0.5

1.1 0.8 1.2 1.4 1.3 1.5 1.3 1.5 1.2 1.7 0.9 1.3 1.3 1.3 1.8 1.3 1.7

89.7 92.6 88.6 82.4 85.9 84.9 86.9 85.9 88.8 79.2 91.4 88.3 87.6 85.0 77.8 88.0 79.1

n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. 0.11 n.d. n.d. n.d. n.d. 0.10 n.d. 0.12

0.48 0.41 0.42 2.41 0.61 0.60 0.40 0.43 0.63 0.78 0.37 0.53 0.15 0.05 0.12 0.07 0.23

0.42 0.33 0.35 1.02 0.58 0.56 0.38 0.42 0.60 0.69 0.34 0.54 0.18 0.08 0.08 0.11 0.16

0.08 0.07 0.06 1.06 0.05 0.06 0.04 0.04 0.04 0.06 0.03 0.03 0.02 0.02 0.03 0.03 0.04

Charleville 7

137.60-138.60 U 138.60-139.60 U 139.60-140.60 U 140.60-141.60 C 141.60-142.60 C 142.60-143.50 C 143.50-144.50 C 144.50-145.90 C

5.1 5.4 4.0 4.9 4.8 4.8 6.1 8.5

2.2 1.9 2.2 0.4 0.2 0.2 0.9 1.5

0.1 0.1 1.7 0.1 0.1 0.1 0.1 0.3

1.2 1.2 0.9 1.0 1.0 1.0 1.2 1.4

88.0 88.2 85.9 90.9 91.3 91.0 89.0 85.7

n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d.

0.54 1.36 0.01 0.02 0.03 0.06 0.07 0.12

0.51 1.32 0.10

0.06 0.06 0.01 0.02 0.02 0.02 0.02 0.02

Well

0.11 0.10 0.06 0.07

0.11

* A—Allaru Mudstone C—Coreena Member of Wallumbilla Formation T—Toolebuc Formation i . U - U r i s i n o beds } t , m e eqU1Valent W—Undi f ferentiated n.d.—not determined

to be higher and more scattered. Ratios up to 0.07 may indicate weathering, cf. N/C of 0.075 for extensively weathered Julia Creek kerogen (Riley & Saxby, 1982).

Sulphur analyses Pyritic sulphur contents are, with one exception, below 2 per cent. Sulphate sulphur values are often less than 0.1 per cent; higher values over a series of adjacent samples probably indicate oxidation of pyrite during Recent weathering (e.g. most of the Toolebuc Formation section intersected in Tambo 38). Organic sulphur results, being calculated by difference (organic sulphur = total sulphur - pyritic sulphur sulphate sulphur), are subject to considerable uncertainty. Not surprisingly, when organic sulphur is plotted against organic carbon, no clear relationship emerges. The richest shales tend to have an atomic S/C ratio of ~0.016, but lower values, tending to zero for organic-poor samples, are also obtained. Augathella 7 is exceptional in that S/C ratios for the upper four Toolebuc Formation samples average 0.005, in contrast to 0.050 for the lower seven samples. Evidently, based on sulphur content, two types of kerogen are present. However, source homogeneity is suggested by the constant N/C ratios, and similar degrees of Recent weathering can be inferred from the levels of sulphate sulphur. The seven kerogens from Julia Creek and other areas referred to earlier are high in organic sulphur (5 to 8 per cent daf; atomic S/C = 0.031 ± 0.006) but weathered Julia Creek kerogen is much lower (S/C = 0.011) (Riley & Saxby, 1982). Most samples used in the present study are below the depth of obvious Recent weathering (Fig. 2) but more subtle bacterial or non-biological alteration of minerals and organic matter cannot be excluded.

Ash analyses The experimentally determined ash values in Table 2 do not represent mineral matter in the original samples. In general, true mineral matter in any sample can be calculated: mineral matter (%) = 100 - organic matter (%) - moisture (1) below 105 °C (%) Organic matter can be approximated as 1.33 x organic carbon, since Toolebuc Formation kerogens contain ^75 per cent carbon (Riley & Saxby, 1982); values of moisture released below 105 °C are given in Table 2. The largest difference between true mineral matter and measured ash content is 22 per cent for the Tambo 38 sample at 60.24 to 60.50 m.

Hydrogen analyses Total hydrogen values in Table 2 are composed of three contributions: total hydrogen (%)=organic hydrogen (%)+moisture below 105°C (%)/9+mineral water above 105°C (<7o)/9 (2) Calculation of true organic hydrogen contents in each sample depends on estimating the hydrogen present in clays and other minerals, and which is given off (as water) above 105 °C during ash determinations. Two independent procedures are possible: (i) The difference between mineral matter and ash can be attributed to mineral water and known ashing reactions (CaC0 3 - CaO; FeS2 - Fe 2 0 3 ): mineral matter (%) - ash (°7o) = 0.440 CaC0 3 (%) + 0.333 FeS2 (°7o) + mineral water above 105 °C (%) (3) Mineral matter is calculated using equation (1), mineral water is calculated using equation (3) and thence organic hydrogen is calculated from equation (2). When this calculated value

J. D. SAXBY

246 TABLE 3. Oil assay results on Eromanga Basin samples.

Gas + loss

Well

Depth (m)

Formation*

Oil

Char Water (%, air-dried basis)

Tambo 38

57.55-57.92 58.90-59.00 59.28-59.55 60.24-60.50 60.90-61.30 61.60-62.00 62.22-62.40

A T T T T T T

2.51 4.81 4.84 5.82 4.78 3.24 1.77

5.37 4.45 5.05 6.81 6.83 7.14 8.00

90.6 88.3 86.7 84.6 85.9 87.9 89.1

1.5 2.4 3.4 2.8 2.5 1.7 1.1

Tambo 41

36.58-36.62 37.65-37.77 38.55-38.68 39.80-39.93 40.95-41.11 41.85-42.00 42.75-42.90

T T T T T T T

2.19 1.43 2.45 3.44 4.20 4.29 2.17

6.98 8.26 6.98 7.38 6.48 7.72 8.51

89.4 88.2 87.9 86.2 86.2 85.7 8.8.0

1.4 2.1 2.7 3.0 3.1 2.3 1.3

Tambo 44

15.15-15.28 16.10-16.30 17.25-17.50

T T T

1.85 1.72 3.25

9.71 10.38 9.32

86.9 86.6 85.3

1.5 1.3 2.1

Augathella 6

56.25-56.40 56.80-57.00

T T

2.53 1.92

8.85 9.56

86.7 87.0

1.9 1.5

Augathella 7

72.90-73.10 75.15-75.30

T T

1.34 1.94

10.54 10.54

86.5 86.0

1.6 1.5

Charleville 6

63.10-64.10 64.10-64.90 64.90-66.00 66.00-67.10 67.10-68.10 68.10-69.10 69.10-70.10 70.10-71.10 71.10-72.00 72.00-72.95 72.95-74.20 74.20-75.10 75.10-76.10 76.10-77.10 77.10-78.10 78.10-78.95 78.95-80.00

A A U U U U U U U U U U U U U U U

0.58 0.30 0.40 0.52 1.33 0.72 0.69 1.30 0.91 0.92 1.60 0.89 0.82 1.02 0.84 1.09 0.88

8.67 6.54 8.75 10.61 8.05 9.89 9.34 9.54 8.51 11.08 6.11 8.93 9.26 8.79 11.05 8.86 10.51

90.5 92.9 90.6 84.2 86.9 87.2 88.0 88.4 89.9 86.2 92.3 90.0 89.4 86.7 85.5 89.5 85.5

0.3 0.3 0.3 4.7 3.7 2.2 2.0 0.8 0.7 1.8 0.0 0.2 0.5 3.5 2.6 0.6 3.1

Charleville 7

137.60-138.60 138.60-139.60 139.60-140.60 140.60-141.60 141.60-142.60, 142.60-143.50 143.50-144.50 144.50-145.90

U U U U C C C C

0.88 0.91 1.07 0.55 1.06 0.89 0.80 1.13

8.34 6.96 7.36 8.01 7.06 6.57 8.44 8.99

90.3 91.4 91.4 91.4 91.9 91.3 90.5 89.4

0.5 0.7 0.2 0.0 0.0 1.2 0.3 0.5

* A—Allaru Mudstone G—Coreena Member of Wallumbilla Formation T—Toolebuc Formation ^ . , U-Urisino beds } time e1ulvalent of organic hydrogen is plotted against organic carbon, a reasonably linear relationship is obtained. The slope of this line implies an average atomic H / C ratio of 1.36. (ii) If water obtained in the oil assay (Table 3) represents the sum of moisture below 105 °C and mineral water above 105 °C, then organic hydrogen can be calculated from equation (4): total hydrogen (%) = organic hydrogen (%) +• assay water (%)/9 (4) The resulting atomic H / C ratios vary from 0.80 to 1.24 with a mean of 1.03.

The mean H / C for Toolebuc Formation kerogen from Riley .& Saxby (1982) is 1.22. Differences between this value and those from procedures (i) and (ii) are relatively small but may reflect genuine variations in organic matter type. Alternatively, the differences may be due to analytical errors or to approximations made in the calculations. For example, method (ii), in ignoring possible water generation from organic matter, will probably result in low H / C values. It should be remembered that both methods relate to the total organic matter present and not just to the insoluble kerogen.

OIL SHALE GEOCHEMISTRY Organic carbon ( % ) 2 4 6

Carbonate carbon ( % ) 2 4 6

Pyrite (%) 0-5 1-0 1-5

- 10

EX 1 n Mudstone

Siltstone

Sandstone

Limestone

Fig. 2. Correlation between geochemical and petrophysical log parameters for Longreach 6.

Fig. 3. Correlation between geochemical and petrophysical log parameters for Jericho 11. For key to lithologies see Fig. 2.

J. D. SAXBY

248

Carbonate carbon (%)

o 40 -

Pyrit* (%)

Oil 1%)

T 0 73-25iii

Fig. 4. Correlation between geochemical and petrophysical log parameters for Tambo 38. For key to lithologies see Fig. 2. Organic carbon (•/.) 1 2 3 1—I—T

Carbonate carbon (%) 0 1 0-2 0-3

Pyrite (%i 10 2 0

Fig. 5. Correlation between geochemical and petrophysical log parameters for Tambo 40. For key to lithologies see Fig. 2.

249

OIL SHALE GEOCHEMISTRY Neutron 0 increase

Gamma ray API units increase

Density decrease

Stratigraphic units

Organic c a r b o n {•/•)

Carbonate carbon ( % )

Pyrite

Oil (%)

(%)

Allaru Mudstone

-o\

[Neutron

\-j-I-

% S

S I—U

31=

J Toolebuc

( Zrlr

\ 1

. —

formation

S :•':'•:•: Coreena Member 1 I t d 47-80J

Fig. 6. Correlation between geochemical and petrophysical log parameters for Tambo 41. For key to lithologies see Fig. 2.

Density decrease

>, | Neutron 0 increase

Water table

f J

Organic carbon ( % ) 0.4 Q.8 1>2

Gamma r Q y A P I

un,r$

increase

Stratigraphic uniti

Carbonate carbon ( % ) ^ 0 . 2 0 . 3 1

Q.5

Pyrite ( % ) ^

1 ~Ji"<C

Allaru Mudstone (mainly weathered)

-/'19-5-

Density

Toolebuc Formation Neutron

Coreena Member

Fig. 7. Correlation between geochemical and petrophysical log parameters for Tambo 42. For key to lithologies see Fig. 2. Organic

Carbonate

Pyrite

Oil

Fig. 8. Correlation between geochemical and petrophysical log parameters for Tambo 44. For key to lithologies see Fig. 2.

J. D. SAXBY

250

>2 Gamma .2 ray Stratigraphic -c API units units ~ increase

Organic carbon ( % )

Carbonate carbon (%)

1020 30

4-0

Pyrite (%)

0-5 10 1-5 15

Allaru Mudstone

- \-15'8 Toolebuc Formation Coreena Member

Fig. 9. Correlation between geochemical and petrophysical log parameters for Augathella 5. For key to lithologies see Fig. 2.

Spontaneous potential - ..... Resistivity

Density decrease

Organic carbon (•/•)

Carbonate carbon (•/•)

Pyrite

Oil

Fig. 10. Correlation between geochemical and petrophysical log parameters for Augathella 6. For key to lithologies see Fig. 2.

Oil and gas yields Values of oil/organic carbon can be calculated for the 20 Toolebuc Formation samples in Table 3 and range from 0.23 to 0.45 with a mean of 0.36. Hence, on average, 27 per cent of the organic material is converted to oil, assuming organic matter is 1.33 x organic carbon. The richest zone encountered (58.9 to 61.3 m in Tambo 38) averages 53 L t~' (assuming an oil density of 0.95 g cm - 3 ) and has a high oil/organic carbon ratio (0.44). This may point to the presence of differing types of kerogen, with oil-prone material becoming relatively more dominant in richer samples. In the southernmost holes (Charleville 6, 7) oil yields are low and difficult to measure

accurately due to the much higher percentage of water. As a result, the linear relationship between organic carbon and oil yield in such holes is not as clear cut as when oil yields are higher (Figs 4, 6, 8, 10, 11, 15, 16). Gas yields are neither as accurate nor as precise as oil yields, since they are obtained by difference and include losses. However, gas/organic carbon ratios for the Toolebuc Formation samples span a fairly small range (0.21 to 0.34) and the average value (0.26) indicates that 20 per cent of the organic material is converted to gas. The mean amount of organic matter remaining in the char is thus 53 per cent, although this decreases to 49 per cent for the rich 2.4 m zone in Tambo 38.

OIL SHALE GEOCHEMISTRY

Fig. 11. Correlation between geochemical and petrophysical log parameters for Augathella 7. For key to lithologies see Fig. 2.

Fig. 12. Correlation between geochemical and petrophysical log parameters for Charleville 3A. For key to lithologies see Fig. 2.

J. D . S A X B Y

252

i/i Caliper decrease Of Density o Neufron increase

>N CJl o JZ -3

Gamma ray Lithostratigraphic AP! units units increase

Organic carbon ( % )

Carbonate carbon ( % )

Pyrite {%)

Superficial deposits

AUaru (m)

BGL ( m )

Mudstone

32.-1m

Depth

Charleville 4A

Coreena Member

Gamma ray log from Charleville 4 Erosional surface Charleville 4 TD 77 0 m

Caliper Density Neutron

decrease increase

Spontaneous potential Resistivity

Lithology

Cores

Fig. 13. C o r r e l a t i o n b e t w e e n g e o c h e m i c a l a n d p e t r o p h y s i c a l l o g p a r a m e t e r s f o r C h a r l e v i l l e 4 A . F o r key t o l i t h o l o g i e s see F i g . 2.

Gamma ray API units L irnosTrarigrapnic increase units

Organic carbon ( % )

Carbonate carbon ( % )

Pyrite (%)

Spontaneous potential

Caliper Density

Superficial deposits

Toolebuc Formation

Washout Erosional surfoce Resistivity

Coretna Member Sandstone

Fig. 14. C o r r e l a t i o n b e t w e e n g e o c h e m i c a l a n d p e t r o p h y s i c a l l o g p a r a m e t e r s f o r C h a r l e v i l l e 5. F o r key t o l i t h o l o g i e s s e e F i g . 2.

(m) Depth

Mudstone

auoz pajag4Da/vy

Allaru

BQL

Neutron

OIL SHALE GEOCHEMISTRY Neutron ^increase

Organic carbon (%* 20 4 0 6 0

Gamma r a y Stratigraphic API units units increase

253

Carbonate carbon (%) 0-5 1 0 1-5

i i r

Pyrite (%) 1 0 2 0 3-0 i

Oil (%) 0-5 1-0 1-5 I

Superficial s a n d and silcrete 14-0 —

Cainozoic

clays 40

72 f 49-4 H 76

Allaru Mudstone

64 9 - | Urisino beds 80

80-4 T D 85-2 m- Coreena Member

Fig. 15. Correlation between geochemical and petrophysical log parameters for Charleville 6. For key to lithologies see Fig. 2. Neutron ^increase

Gamma r a y Stratigraphic API units units increase

Organic carbon (%) 1 0 20 30 " i — i — r

Carbonate carbon (•/.) 0-5 1 0

1 5

i—i—r

Pyrite (%) 1 0 2 0 30

n—i—r

Oil (•/.) 0-5

1 0 1-5

i—I—r

Superficial sand and gravel - H0 0 -

Water table^

134

20 Cainozoic 136

clays 40 48 0 Transitional zone (Tweathered) Allaru Mudstone 72-0-

138

140

100 142

120

140

^TD151-8 m

Fig. 16. Correlation between geochemical and petrophysical log parameters for Charleville 7. For key to lithologies see Fig. 2.

CONCLUSIONS

(i) Varying thicknesses of Toolebuc Formation oil shale along the eastern margin of the Eromanga Basin from Longreach to Charleville have been analysed geochemically. Only low-grade oil shale has been intersected. The most

significant zone was one located near Tambo, which was 2 m yielded 53 L t In general organic contents and y decrease towards the south. (ii) Toolebuc Formation organic matter from the sampled holes does not vary greatly in composition. Atomic H/C

t h i c k

a n d

o i l

i e l d s

254

J. D. SAX BY

values fall in the range 0.9 to 1.3, significantly lower than most oil shales, including Rundle and Green River (1.5 - 1.6) (Saxby, 1980). If it is assumed that completely aromatic and aliphatic kerogens have H/C ratios of 0.6 and 1.8 respectively, then Toolebuc Formation organic matter appears to be- 60 per cent aromatic. Nitrogen content of the carbonaceous material is-2.5 per cent, while organic sulphur varies up to 10 per cent with the most organic-rich samples containing -3.2 per cent. (iii) On average, 27 per cent of the Toolebuc Formation organic matter is converted to oil on pyrolysis, 20 per cent gas is produced and 53 per cent remains as char in the spent shale. Evidently, this low conversion to oil is a reflection of a high content in the unheated shale of aromatics, which are non-volatile, insoluble in organic solvents and, for the most part, thermally stable. The proportion of kerogen converted to oil (rather than gas or char) increases with organic carbon content. (iv) A precise determination of total nitrogen can be used as an indicator of organic carbon (and thence indirectly of 011 yield) in any particular part of the Toolebuc Formation. (v) The sulphur content of Toolebuc Formation shale oils (commonly ~5°7o) results mainly from the higher than average organic sulphur content of the kerogen. This sulphur is probably present as sulphide and disulphide bonds, as well as in heterocyclic aromatic rings.

(vi) Metal carbonate contents in limestone' ot the Toolebuc Formation vary from 0 to 67 per cent and are not directly related to oil yields. (vii) Moisture contents in ground, air-dried samples vary from 1.4 to 9.9 per cent. As expected, carbonate-rich, claypoor samples are lowest in water lost up to 105 °C. Water obtained on retorting ranges up to 11.1 per cent of the airdried shale and can be considered as the sum of moisture, chemically-bound water released from clays, and a small amount produced from decomposition of oxygen-functional groups in kerogen. Water released above 400 °C during retorting may assist in hydrogenation of shale oil. (viii) Although most of the wells in this study were positioned to intersect and recover Toolebuc Formation samples that were not extensively weathered, some changes in mineralogy, redistribution of metals (such as vanadium) and bacterial alteration of kerogen cannot be excluded.

ACKNOWLEDGEMENTS

Support for this research was provided under the National Energy Research, Development and Demonstration Programme administered by the Commonwealth Department of Resources and Energy. NERDDC Project 78/2616 on oil shale methodology is a joint CSIRO/BMR investigation. The analytical assistance of N. Watson and R. Judd is gratefully acknowledged.

REFERENCES EXON, N. F. & SENIOR, B. R., 1976: The Cretaceous of the Eromanga

and Surat Basins. Aust. Bur. Miner. Resour. GeolrGeophys. J. 1, 33-50. GIBSON, D. L., 1981: Oil shale in Australia—its occurrence and resources. Aust. Min. Ind. Q. 33, 105-13. OZIMIC, S., 1981C: Stratigraphic drilling in the Cretaceous Toolebuc Formation in the southern and eastern Eromanga Basin, 1980. A contribution to BMR/CSIRO. NERDDC Project, 78/2616. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1981/12 (unpubl.). OZIMIC, S., 19816: Stratigraphic drilling in the Cretaceous Toolebuc Formation in the Charleville district, southern Queensland, June 1981. A contribution to BMR/CSIRO, NERDDC Project, 78/2616. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1981/3.8 (unpubl. I. OZIMIC, S., 1982: Depositional environment of the oil shale-bearing Cretaceous Toolebuc Formation and its equivalents, Eromanga Basin, Australia. Proceedings of the 15th Oil Shale Symposium, Colorado School of Mines, 137-48. OZIMIC, S. & SAXBY, J. D., 1983: Oil shale methodology, Final Report of Project 78/2616, National Energy Research

Development and Demonstration Program. Department of Resources and Energy, Canberra, NERDDP/EG/83/187. RILEY, K. W. & SAXBY, J. D., 1982: Association of organic matter and vanadium in oil shale from the Toolebuc Formation of the Eromanga Basin, Australia. Chem. Geol. 37, 265-75. SAXBY, J. D., 1980: Oil shale in Australia. Energy Resour. Technol. 2, 30-4. SENIOR, B. R., EXON, N. F. & BURGER, D., 1975: The Cadna-owie and Toolebuc Formations in the Eromanga Basin, Queensland.

Qld Gov. Min. J. 76, 444-55. SENIOR, B. R., MOND, A. & HARRISON, P. L., 1978: Geology of the

Eromanga Basin. Aust. Bur. Miner. Resour. Geol Geophys. Bull. 167. STEPHENSON, A. E., 1982: Stratigraphic drilling in the Cretaceous Toolebuc Formation in the southern Eromanga Basin, 1981. A contribution to BMR/CSIRO, NERDDC Project, 78/2616. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1982/25 (unpubl.). WATSON, N., 1984: A small-scale method for determining oil yields from oil shales. Fuel, 63, 1455-8.

Geological Society of Australia Special Publication No. 12, 255-265

Organic matter in the Toolebuc Formation N. R. Sherwood & A. C. Cook Department of Geology, University of Wollongong, P.O. Box 1144, Wollongong, N.S.W. 2500.

ABSTRACT Organic matter in Toolebuc Formation mixed oil shales comprises mainly bituminite and micrinite with lesser amounts of liptodetrinite, lamalginite, telalginite and inertodetrinite. Sporinite, resinite/bitumen, vitrinite and organic matter of unknown affinity occur in trace amounts. Because of the different specific oil yields of various macerals and variable maceral composition of mixed oil shales, a knowledge of organic petrology coupled with total organic carbon data is required to enable interpretations on total hydrocarbon yields. Bituminite is probably derived from planktonic or benthonic algae; it differs from alginite by virtue of different precursor algae, syndepositional or postdepositional effects or a combination of the three. Lamalginite is derived mainly from dinoflagellate and acritarch cysts; telalginite is derived from tasmanitid cysts. Liptodetrinite probably originates from comminution of lamalginite, fine disseminations of bitumen/resinite and fragmentation of other exinites. Allochthonous higher plant matter is the source of most inertodetrinite, vitrinite and sporinite. Resinite/bitumen probably originates from waxes, oils, fats and resins exuded from organic matter. Although mean maximum vitrinite reflectance data indicate that the Toolebuc Formation is immature to marginally mature, it could possibly have generated significant amounts of oil during maturation. Toolebuc Formation oil shales were deposited mainly in a distal, offshore marine shelf environment. The sediments were largely anaerobic but overlying waters could have been aerobic at times. The main postdepositional effect is micrinitization of bituminite. Absorbed radioactive mineral matter could contribute to coalification of bituminite and micrinitization.

INTRODUCTION Toolebuc Formation oil shales have an areal extent of greater than 600,000 km2 (Fig. 1) and the present study indicates total in situ possible resources of more than 2,000 x 109 barrels of shale oil. The formation outcrops in parts of northern Queensland and occurs at depths of up to 1.5 km in parts of South Australia. Disagreement and confusion have beset studies related to the classification and genesis of organic matter in Toolebuc Formation oil shales. Misleading information has arisen because of inherent methodology problems. Employing chemical methods to analyse total organic matter can lead to misleading data because average values are obtained for a mixed maceral assemblage. During sample preparation used by palynologists, a large part of the fine-grained material is commonly discarded or classified as 'amorphous kerogen' because of its lack of obvious palynomorphic structures. In this study, reflected white and incident ultraviolet/violet light (fluorescence-mode) microscopy was used to study organic matter in 10 drill core samples (supplied by Pacific Coal Pty Ltd) and miscellaneous oil exploration well cuttings samples (supplied by Delhi Petroleum Pty Ltd) of Toolebuc Formation oil shales (Fig. 1). The samples were mounted in cold-setting polyester resin, ground and polished. Examination of organic matter in 'whole-rock' samples allows all macerals to be observed and spatial relationships studied. Additional information gathered from examination of petrologically similar rocks has also proved valuable for the study.

Organic matter in Toolebuc Formation oil shales mainly comprises bituminite/micrinite (see Teichmuller, 1974 for definitions) with disseminated lamalginite, telalginite, liptodetrinite and inertodetrinite. Sporinite, resinite/bitumen, vitrinite and organic matter of unknown affinity occur in trace amounts. Significant vertical variations in organic matter abundances and compositions occur within Toolebuc Formation oil shales. Some changes also occur laterally but the lateral control available is insufficient to make definitive statements concerning this variation. This mixed maceral assemblage led Hutton et al. (1980) and Cook et al. (1981) to classify these and other similar rocks as mixed oil shales. Other mixed oil shales include Toarcian oil shales of France (Paris Basin) and Germany (Posidonienschiefer), Kimmeridgian oil shales of Britain, and some Antrim (Michigan Basin) and Chattanooga (Kansas) oil shales of U.S.A. A suite of organic-rich rocks, including one Toolebuc Formation oil shale sample, was analysed by RockEval pyrolysis techniques (see Tissot & Welte, 1978). Crushed whole rock samples were pyrolysed in an inert gas atmosphere, followed by detection of produced gaseous hydrocarbons and carbon dioxide.

TYPES OF ORGANIC MATTER A summary of organic matter types, morphologies, abundances and origins in Toolebuc Formation oil shales is presented in Table 1.

N. R. SHERWOOD & A. C. COOK

256

30 S

SYDNEY ^CANBERRA MELBOURNE

Oil shale known or likely within Toolebuc Formation city town drill hole extent of Eromanga Basin Euroka Arch

% tu o o

HOBART DMM/GD615

Fig. 1. Map of Australia showing probable areal extent of Toolebuc Formation oil shales.

Bituminite/micrinite Bituminite and micrinite are ubiquitous throughout Toolebuc Formation oil shales, in places constituting up to 40 per cent by volume of the rocks (Fig. 2). The bituminite has a massive to fine-grained texture (Fig. 2a) and a mean maximum reflectance (in oil immersion) from about 0.10 to 0.35 per cent, averaging 0.20 per cent (measured according to S.A.A., 1981). It commonly fluoresces dull and rarely medium orange and brown upon ultraviolet/violet light excitation (Fig. 2b). Varieties that do not fluoresce are also present. Fluorescence alteration (change in fluorescence emission after short-term or prolonged ultraviolet/violet light irradiation) varies from weakly to strongly positive (i.e. enhanced emission) (Fig. 2c). Some bituminite is associated with fish remains (Fig. 2b). Micrinitization of bituminite is a prominent feature and transitional matter is common (Fig. 2d). Micrinite typically occurs within lenses or layers of bituminite and is a prominent textural feature in much of the bituminite. It also occurs more

rarely as massive lenses having a relatively high reflectance and no fluorescence. Intermediate forms have characteristics between micrinite and bituminite and are probably best classified as micrinitized bituminite. Where observed in polished sections, bituminite exhibits predominantly vitrinitic optical and textural characteristics with some exinite attributes (cf. resinous vitrinite and fluorescing desmocollinite). If extensive micrinitization has occurred, affinities are with inertinite. In thin section (transmitted light), bituminite is red and has some characteristics similar to those of vitrinite. Bituminite is a characteristic maceral of mixed oil shales.

Lamalginite Lamalginite (alginite B of Hutton et al., 1980; Cook et al., 1981; i.e. lamellar alginite) is also ubiquitous throughout Toolebuc Formation oil shales; however, at most it only constitutes about 10 per cent of any sample studied. In perpendicular section it commonly appears as very thin

257

TOOLEBUC FORMATION ORGANIC PETROLOGY TABLE 1. S u m m a r y of organic matter in Toolebuc Formation oil shales. Key: sparse—0.1-0.5%; rare— <0.1%. Organic matter type

Average range in abundance (Z of total rock)

bituminite/ micrinite

liptodetrinite

1-10

Probable main progenitor

dull orange and brown or no fluorescence

fine-grained to massive lenses and laminae

planktonic or benthonic algae, ?radioactive alteration

bright yellow to dull orange

fragments less than 0.002 mm in diameter

comminuted dinoflagellate and acritarch cysts dinoflagellate and acritarch cysts and possibly other algae tasmanitid cysts

absent-1

bright to medium yellow and greenish yellow

ellipsoidal to discshaped

medium yellow and orange

disc-shaped

spore and pollen

medium to bright yellow

homogeneous and structureless pods

waxes, fats and oils

no fluorescence

fragments from 0.0020.030 mm in diameter

higher plant matter

lamellae spirally or concentrically arranged

sporangia or algae

sporinite

absent-rare

resinite/bitumen

rare

inertodetrinite and vitrinite

rare-2

organic matter of unknown affinity

Form

bright yellow to dull orange

lamalginite

telalginite

Common fluorescence intensity and colour

medium to bright yellow

strands less than about 0.02 mm long (Fig. 2c) but longer, thicker-walled forms are also present. In parallel section the lamalginite commonly exhibits a spherical shape and in places ornamentation (Fig. 3a, b). It has a very low reflectance and typically fluoresces from bright yellow to dull orange. Partial replacement of lamalginite by highly reflecting mineral matter is a common feature. Lamalginite occurs disseminated throughout the oil shales, but is concentrated in some laminations and is commonly associated with bituminite. Ramsden (1983, p. 18) apparently referred to lamalginite as filamentous organic matter'.

Telalginite Telalginite (alginite A of Hutton et al., 1980; Cook et al., 1981) is defined as discrete colonial or unicellular algal bodies that are spherical to disc shaped. In samples studied to date, telalginite is rare but it is possible that telalginite-rich facies exist within Toolebuc Formation oil shales. Most mixed oil shale occurrences contain a type of telalginite similar to the major maceral of tasmanite oil shales (e.g. of Tasmania and Alaska). Telalginite in the Toolebuc Formation oil shales has a low mean maximum reflectance (0.02 to 0.049o) and commonly fluoresces bright and medium yellow (Figs 2b, c; 3d). It occurs disseminated throughout the oil shales and is concentrated in some beds. Ramsden (1983, p. 18>*eferred to telalginite as 'spore-like bodies'.

Other macerals Because of the small size of some lamalginite, the term liptodetrinite is reserved for exinite fragments less than about 0.002 mm in diameter. Liptodetrinite is ubiquitous in Toolebuc Formation mixed oil shales, generally constituting between 1 and 10 per cent by volume of the rocks (Figs 2b, c; 3a, c). The fluorescence and reflectance properties are variable because of its mixed genesis but commonly resemble those of lamalginite. Liptodetrinite is disseminated throughout the oil shales.

Inertodetrinite occurs disseminated throughout the oil shales but at most constitutes 1 to 2 per cent by volume of any sample studied. It has a high reflectance and does not fluoresce. Sporinite, resinite/bitumen, vitrinite and organic matter of unknown affinity all occur in trace amounts dispersed throughout the rocks. In the samples studied, sporinite has a low reflectance and commonly a medium yellow and orange fluorescence. Resinite/bitumen refers to homogeneous, structureless exinite. It commonly occurs as small (about 0.01 mm diameter) pods and has a low reflectance and typically medium to bright yellow fluorescence. Vitrinite has a mean maximum reflectance between 0.40 and 0.60 per cent and does not fluoresce. Organic matter of unknown affinity has a low reflectance and medium to bright yellow fluorescence, but its morphology is unusual in that it is composed of thin lamellae arranged concentrically or spirally. As well as discrete macerals, most samples exhibit background fluorescence. This fluorescence is pervasive throughout the mineral matrix and shows positive alteration after extended ultraviolet/violet light excitation (Fig. 2c).

Specific hydrocarbon yields for various organic matter types As shown in Figure 4, specific organic matter hydrocarbon yields vary with organic matter type. In Figure 4, organicrich rocks have been divided into oil shale types, according to dominant organic matter type (after Huttort et al., 1980 and Cook et al., 1981), perhydrous coals and subhydrous coals. Hydrocarbon yields per unit weight of organic carbon, as determined by Rock-Eval analysis (i.e. S, + S 2 /TOC; see Tissot & Welte, 1978), varies with rock type and consequently with organic matter type. In general, alginite-dominated rocks have the highest specific yields and vitrinite- or inertinite-dominated rocks (mostly coals) have the lowest specific yields. If the samples of alginite-rich Antrim oil shales (Michigan Basin, U.S.A.) and Toolebuc oil shales in Fig. 4 had similar organic carbon values, they would have substantially different oil yields

Fig. 2. Fluorescence m o d e (Fl.) a n d white reflected light (R.l.) photomicrographs of Toolebuc Formation mixed oil shales (sections perpendicular to bedding), (a) Sample rich in bituminite/micrinite. Bituminite (grey), micrinitized bituminite (grey with white flecks), highly reflecting mineral matter (white) a n d clay/silt-sized mineral matter (brown matrix) present (latter out of focus d u e to polishing relief of organic matter). Exinite macerals not visible in white reflected light d u e to very low reflectances. R.l., field width = 0.28 m m . (b) Fish remains (orange elongate f r a g m e n t s ) associated with dull greenish brown fluorescing bituminite. Also present: small and large tasmanitid cysts (bright yellow), lamalginite ( m e d i u m greenish yellow), liptodetrinite (yellow flecks) a n d n o n fluorescing clay/silt-sized mineral m a t t e r matrix. Fl., field width = 0.15 m m . (c) Positive alteration of b a c k g r o u n d a n d bituminite fluorescence emission. S a m p l e irradiated for a b o u t 30 m i n u t e s with a higher powered objective ( x 125) causing e n h a n c e d fluorescence emission f r o m that part of the field. Bituminite (dull brown lamellae), micrinitized bituminite (black lamellae), lamalginite (yellow lamellae), telalgmite (bright yellow cyst in right-hand bottom corner), liptodetrinite (yellow flecks), highly reflecting mineral matter (black spherical particles) and clay/siltsized mineral matrix (with some background fluorescence) present. Fl., field width = 0.28 mm ( x 5 0 objective), (d) Micrinitized bituminite (grey with white flecks), bituminite (dark grey), and

highly reflecting mineral m a t t e r

( w h i t e ) in c l a y / s i l t - s i z e d

mineral matter

matrix

(largely o u t

f o c u s ) . R.l., f i e l d w i d t h

0.15

mm.

Fig. 3. Fluorescence-mode photomicrographs of Toolebuc Formation mixed oil shales, (a) Section parallel to bedding, showing the acritarch Veryhachium (centre) and liptodetrinite in clay/silt-sized mineral matter matrix. Fl., field width = 0.18 mm. (b) Demineralized strew mount showing dinoflagellate/acritarch cyst. Fl., field width = 0.18 mm. (c) Parallel section showing liptodetrinite possibly derived from ?Nostocopsis algae, associated with dull fluorescing bituminite. Fl., field width = 0.28 mm. (d) Perpendicular section showing tasmanitid telalginite with characteristic radial canals. Fl., field width = 0.18 mm.

N. R. SHERWOOD & A. C. COOK

mixed oil shales

Toolebuc Formation ( ~ 5 0 % alginite), South Australia

(bituminite/ micrinite and alginite)

subhydrous coal (vitrinite)

a >

Antrim ( ^ 8 0 % alginite), Michigan, U.S.A.

Callide, Queensland

Norwich Park, Queensland

perhydrous coal (vitrinite)

Muswellbrook, N.S.W. I

cannel coals (vitrinite with higher plant exinite)

Mae Teep Basin, Thailand I Yallourn, Victoria

Fang Basin, Thailand

L-.

E (0 O)

£ o

canneloid shales (vitrinite and higher plant exinite) kukersite (telalginite: Gloeocapsomorpha)

Ban Pa Ka Li Basin, Thailand Eucla Basin, South Australia I

Kukersite, Estonian S.S.R.

Q» > KUl

torbanites (telalginite: Botryococcus

-related)

Joadja, N.S.W. Middle River seam. N.S.W.

Condor, Queensland Rundle Formation, Queensland lamosites (lamalginite)

Jae Horn Basin,Thailand Green River Formation, U.S.A. Mae Sot Basin, Thailand

—I— 200

400

—I 600

1 800

1 1000

— I 1200

1 « 1 1400 1600

PYROLYSIS HYDROCARBON YIELD OF ORGANIC MATTER (mg hydrocarbon per g organic carbon) Fig. 4. Organic matter hydrocarbon yields from pyrolysis for various organic-rich rock types.

DMM/GD744

TOOLEBUC FORMATION ORGANIC PETROLOGY 261 upon pyrolysis. Because mixed oil shales have a maceral suggested that bituminite could be derived from Phaeophyta assemblage of predominant bituminite/micrinite (having (brown algae; seaweeds, kelp and related forms) or affinities to vitrinite) and alginite, both of which vary in Rhodophyta (red algae; seaweeds). Cell size and cell contents abundance, the maceral composition must be known to enable of modern kelp are consistent with indistinct internal interpretations of hydrocarbon yield from total organic carbon morphology that some bituminite exhibits upon prolonged data. ultraviolet/violet light irradiation. These types of source materials could explain the humic characteristics of bituminite compared with alginite derived from Tasmanites- or ORIGINS OF ORGANIC MATTER Botryococcus-related genera. Bituminite Ramsden (1983, p. 23) suggested that CoccolithoBituminite is the most abundant and characteristic maceral phoridaceae could be a source material for much of the of mixed oil shales. It has been observed in Posidonien- 'nondescript lamellar organic matter' in Toolebuc Formation schiefer (Teichmuller, 1974) (and stratigraphically equivalent oil shales at Julia Creek. The main basis for this interpretation Toarcian Shales of Paris Basin, France), Kimmeridgian Shale was identification of abundant fragmented coccoliths in the (U.K.), Antrim Shale (Michigan), Sharon Springs, Heebner, mineral matter matrix. However, in view of their size, Eudora, Stark, Little Osage and Chattanooga Shales (all from Coccolithophoridaceae are unlikely to have been the source Kansas), as well as Toolebuc Formation mixed oil shales. of the relatively massiyely structured occurrences of By extension, they were probably not a major Recognition and classification of bituminite has been a bituminite. of any of the bituminite; there being little difference problem for organic petrologists over the years. For Toolebuc source in the optical Formation oil shales Hutton et al. (1980) grouped it with the occurrences. properties of the massive and the more diffuse matrix, not specifying it as a maceral. In a recent International Bituminite, at least in part, probably arises from Committee for Coal Petrology (I.C.C.P.) 'round robin* set of analyses of Toarcian Paris Basin mixed oil shales it was Cyanophyta (blue-green algae or cyanobacteria). Glikson classified by various workers as: 'impregnation of bituminous (1982) has identified types resembling Nostoc and Anabaena matter', 'bitumen or bituminous/humic ?matter , 'organic in Toolebuc Formation oil shales, using scanning and matter of unknown affinity' and 'bituminite' (I.C.C.P., 1982). transmission electron microscopy. These types may not, The term bituminite has not become widely accepted even however, be the sole or even the main contributors to though it has been extensively described by Teichmuller in bituminite in the Toolebuc Formation oil shales. A cyanophyte a number of original papers and in Stach's Textbook of Coal mat is a possible precursor for some bituminite, notably that Petrology (Stach et al., 1975). Some authors probably do not which forms a matrix for mineral matter. Gebelein (1969) has use 'bituminite' because of the connotations of bitumen in noted that mucilaginous sheaths of Cyanophyta can act as sediment traps. Williams & Reimers (1983) suggested that mats the term; bituminite is not a bitumen. to these are precursors for much of the organic matter Ramsden et al. (1982, p. 295) (after Swanson et al., 1960) similar in the Miocene Monterey Formation of California. have divided organic matter in the oil shale fades of the Progenitor algae could have been planktonic (e.g. Toolebuc Formation into 'sapropelic (lamellar alginite B)' and 'humic, derived from the woody tissues of plants'. This Sargassum at or near the surface as suggested by Johnson approach is misleading because sapropelic is a term for an (1981) for a petrologically similar oil shale in Ohio) or environment and is not equivalent to alginite B (lamalginite) benthonic (e.g. as suggested by Kauffman (1981) for the and humic matter does not necessarily originate from wood. petrologically similar Posidonienschiefer). A benthonic as Ramsden et al. (1982, p. 295) have also written, 'The only opposed to a planktonic algal source has also been a topic material of possible humic origin consists of finely divided of disagreement for many other oil shales. particulate matter somewhat resembling huminite in thin Most alginite identified to genus or species level has a section'. Cook et al. (1981) have classified the bituminite in planktonic algal source. Botryococcus-related telalginite, Toolebuc Formation rocks as 'vitrinite-like matter'. Ramsden occurring in torbanite oil shales is derived from a planktonic (1983, p. 18) referred to bituminite as 'finely divided organic green alga, and Tasmanites telalginite, occurring in tasmanite matter . . . similar in appearance to huminite although not oil shales, is derived from a planktonic marine alga. demonstrably of humic origin' and as 'discrete laminae'. He Lamalginite identified as Pediastrum, occurring in some apparently referred to micrinite (and possibly some bituminite) Tertiary lamosite oil shales from Queensland (Hutton, 1982) as 'micron-sized granulation' (p. 18). and from Thailand (Sherwood et al.> 1984) is derived from freshwater planktonic alga; most dinoflagellates and Some of this classificatory confusion and inconsistency is aacritarchs marine and planktonic. All these identifiable because of exinite maceral names referring to specific plant forms are are markedly different in optical properties from tissues (such as sporinite), or secretions (such as resinite), bituminte, however. Some lamalginite occurring as thick layers, whereas bituminite has uncertain origins and has properties continuous in bedding planes, appears to be derived intermediate between those of vitrinite and the strongly more from benthonic algae. fluorescing exinite macerals. The major precursor matter of bituminite remains uncertain Most workers including Teichmuller (1974), Stach et al. though some cyanobacteria have been identified. Until (1975), Sherwood & Cook (1982), Jacob (1982), Cook (1982) evenprogenitor material is definitely proven to be algal, it is and Hutton, A. C. (pers. comm., 1982) agreed that bituminite the best to classify this component of the organic matter is probably, at least in part, derived from algae. The nature probably in Toolebuc Formation oil shales as bituminite but not to of this progenitor algal material is, however, uncertain. The optical properties and mode of occurrence of bituminite differ regard it as exinite or vitrinite. from those of organic matter traditionally referred to as Lamalginite alginite. These differences are probably due to dissimilar Definite algal affinities of lamalginite do not become clear precursor algae, syndepositional or postdepositional effects or, a combination of the three. Sherwood & Cook (1982) have until samples are studied in parallel section. It is then apparent ,

262

N. R. SHERWOOD & A. C. COOK

that most lamalginite is derived from cysts of dinoflagellates and acritarchs with small contributions from indeterminate algal forms. Proximate (devoid of processes), chorate (with processes) (Fig. 3a, b) and peridinioid (star-shaped) forms are pfesent. Dinoflagellates are unicellular, biflagellate algae, which range in size from 0.005 to 2 mm in diameter (commonly 0.025 to 0.25 mm). They are mostly planktonic and marine, but freshwater and benthonic forms also exist. Their life cycle comprises a vegetative stage, a zygotic stage and a resting stage. During the resting stage encystment occurs by formation of 'sporopollenin' walls within the cellulose walls of the vegetative stage. They then sink to the bottom, the outer walls decompose and resistant parts are preserved as dinocysts. Widespread encystment occurs primarily during blooming periods. In general, thick-walled varieties are characteristic of littoral environments whereas thin-walled types are found in offshore sediments. Acritarchs are very similar to dinoflagellates in morphology, but have uncertain origins. Many authors believe that acritarchs can be dinoflagellates lacking preserved definitive dinoflagellate features and most agree that acritarchs originate mostly from unicellular algae (Evitt, 1963; Downie et al., 1963; Sarjeant, 1969, 1970; Lister, 1970). Williams (1978) stated that if Holocene phytoplankton were preserved some would probably be classified as acritarchs. Acritarchs are commonly unicellular but colonial forms also exist. They can be preserved as ornamented or smooth-walled cysts, and like dinoflagellates, were dominantly marine organisms. In quiescent offshore environments, forms with long processes are commonly found, whereas in nearshore turbulent conditions naked or short-processed forms are prevalent. Some acritarchs identified in the Toolebuc Formation oil shales include Veryhachium (Fig. 3a), and Micrhystridium. Minor to trace amounts of lamalginite could have arisen from other thin-walled algae (e.g. ?Nostocopsis as in Fig. 3c). Ramsden (1983, p. 23) suggested that lamalginite could originate from 'original filamentous algae', presumably referring to Coccolithophoridaceae. This seems unlikely to be the case by virtue of the size and form of lamalginite in relation to those of Coccolithophoridaceae.

Telalginite

Telalginite in Toolebuc Formation oil shales originates from tasmanitids (unicellular, thick-walled algae, probably with chlorophyte affinities). Tasmanitid cysts generally range in size from about 0.05 to 0.6 mm in diameter with cell walls from about 0.005-0.02 mm thick. They are characterized by numerous, fine radial canals (Fig. 3d), with pits on wall surfaces. The arrangement and morphology of these canals are used by palynologists for species identification. Ramsden (1983) claimed that telalginite could originate from zoospores of Coccolithophoridaceae. Tasmanitids are planktonic marine algae and are commonly found in shales formed in reducing environments.

Other macerals

Most liptodetrinite probably originates from comminution of dinoflagellate/acritarch cysts, in particular from detached processes. Lesser amounts are probably tiny pods of resinite/bitumen or fragmented remains of other exinites. Micrinite occurs as an alteration product of bituminite (as discussed below). Inertodetrinite is generally derived from oxidized higher plant matter; however, in Toolebuc Formation oil shales some is possibly derived from faunal remains.

Vitrinite is also derived from higher plant tissue. Sporopollenin contained in the walls of spores and pollen is a major precursor of sporinite. Resinite/bitumen is probably derived from waxes, fats, oils and resins exuded from organic matter or in some cases possibly from migrated hydrocarbons. Pods of resinite-like material occur within lenses of bituminite and may be representative of original botanical structures. Organic matter of unknown affinity is possibly derived either from sporangia of water plants, other algae or could represent faunal remains. Background fluorescence is probably caused by exinite below the polished surface ('show-through'), finely comminuted biodegraded exinite, faunal oils, dispersed bacteria, fluorescing mineral matter and bitumen absorbed on mineral grains.

LEVEL OF MATURATION

As indicated by vitrinite reflectance data, the Toolebuc Formation has a low maturation level over its entire area. Vitrinite is rare within the Toolebuc Formation, but it is present within overlying and underlying units. Using drillhole data, interpolations can be made to obtain the reflectance value that vitrinite would have if it occurred within the Toolebuc Formation and if there were no intermaceral effects. The pattern of reflectance distribution for the Toolebuc Formation is very similar to that illustrated by Kantsler et al. (this volume) for the top Transition beds/Cadna-owie Formation horizon. The Toolebuc Formation vitrinite reflectance values are, however, about 0.2 per cent less in high rank areas and about 0.1 per cent less in low rank areas than those of the top Transition beds/Cadna-owie Formation horizon. In outcrop areas, estimated vitrinite reflectances for the Toolebuc Formation range from 0.35 per cent to 0.45 per cent, but the values are typically 0.40 per cent to 0.55 per cent in the subsurface. The Toolebuc Formation would normally be considered immature or at best marginally mature for oil generation. The extent of micrinitization within the bituminite may be evidence that some oil generation has occurred from this unit. Bituminite can probably generate significant amounts of oil at lower ranks as compared with most other macerals (Cook, this volume). Evans et at. (1984) concluded that alkane production, by the loss of functional groups, is largely complete at 0.65 per cent vitrinite reflectance. Moore & Pitt (1984) reported oil shows from a number of intersections of the Toolebuc Formation. Some oil generation has therefore probably occurred from organic matter in the Toolebuc Formation. The yield is likely to be small in relation to the ultimate potential yield of the formation because of its low level of maturity. Migration of oil from the unit may not be efficient, reservoir rocks are not well developed at the Toolebuc Formation horizon and structuring is generally less developed in the upper parts of the Eromanga Basin section, as compared with deeper parts. The oil shale facies represents, however, an unusually rich source rock and this may, in part, compensate for the combination of other unfavourable factors.

DEPOSITIONAL HISTORY

Syndeposition

From stratigraphy, organic and inorganic petrology, palaeontology (including palynology) and geochemistry of

TOOLEBUC FORMATION ORGANIC PETROLOGY the Toolebuc Formation, it is generally agreed that the oil shales were deposited in a marine environment (see Moore & Mount, 1982). Ozimic (1982a, this volume) proposed a quiet, anaerobic, deep-water environment in a transgressive sea. He also suggested that the sea was most likely stratified with a permanent halocline below a layer of fresh water and that these conditions would lead to enhanced prolific productivity in the euphotic zone, a permanent or intermittent oxygen depletion in the lower part of the water column of deeper parts of the basin, and limited establishment of benthonic fauna. As a general model Ozimic (1982b) has suggested a silled basin with a fresh marine water input introducing nutrients and soluble radioactive matter. It should be noted, however, that the Toolebuc oil shale facies is developed on both sides of the Euroka Arch, which Ozimic (1982b) proposed as the barrier to water circulation.

263

chemistry, sea level, circulation pattern and algal growth) would cause the boundary to move up or down. Cook (1982) has proposed that a model similar to this could apply to Toolebuc Formation oil shales. Cyanophyta (bacterial.mats) have been found in deep marine basins (Soutar & Crill, 1977) as well as in tidal flats, and they can be characteristic of anaerobic sediments (Williams & Reimers, 1983). Williams & Reimers (1983) also mentioned that other workers have found that some bacteria prefer to live in oxygen-depleted

A model similar to this has been generally accepted for the petrologically similar Posidonienschiefer (see Kauffman, 1981). The Posidonienschiefer has been set as the 'type' example for stagnant, anaerobic depositional conditions because of the lack of bacterial decay and scavenging (inferred from high percentages of organic carbon and sulphur, and presumed high original hydrogen sulphide), abundant nodular and disseminated pyrite, lack of bioturbation and current flow (as evidenced by continuous laminations), and excellent preservation of fossils (Kauffman, 1981). For similar reasons, Toolebuc Formation oil shales have been presumed to be a sapropelic-environment deposit. From work on oceanic sediments, Ryan & Cita (1977) and Miiller & Suess (1979) claimed that organic matter preservation can be enhanced by high sedimentation rates. Arthur & Schlanger (1979) attributed Cretaceous, oceanic, organic-rich sediments to 'oceanic anoxic events' either caused by a restricted basin (Ryan & Cita, 1977) or by an expanded oxygen minimum layer (Thiede & van Andel, 1977). Arthur (1979) and Jenkyns (1980) stressed the importance of palaeoclimates for anoxic conditions and organic-rich sedimentary deposition; Jenkyns (1980) added that in the Cretaceous, some settings had anoxic-oxic boundaries at or near sediment-water interfaces. Dean & Gardner (1982) claimed that anoxic bottom conditions are commonly the result of high organic matter input rather than the cause of organic-rich strata. Habib (1982), from studies on north Atlantic, Lower Cretaceous sedimentary rocks, found that organic richness is primarily a function of types and rates of organic matter supply. In the present study, organic matter productivity is considered to be a governing factor for oil shale deposition. After detailed paleocommunity and biofabric analyses of the Posidonienschiefer, Kauffman (1981) found benthic fauna and paleocurrent indicators. He therefore favoured a modified gyttja model (oxygenated, nutrient-rich waters overlying anaerobic sediments) (Fig. 5a, b, c) rather than the previously accepted sapropelic model (oxygen-depleted waters overlying anaerobic sediments). Morris (1980) noted ubiquitous epifauna as well as low oxygen tolerant infauna in Kimmeridgian (U.K.) mixed oil shales and suggested that anoxic conditions extended only up to sediment surface. Paleocommunities alternating over short periods of time indicate fluctuating environmental conditions which may be mainly controlled by movement of an anaerobic-aerobic boundary at or near the sediment-water interface. From biofabric evidence it appears that almost continuous bottom current flow has occurred during Posidonienschiefer sediment deposition and consequently Kauffman (1981) has suggested that an algal-fungal mat grew a few centimetres above the bottom, entrapping anaerobic waters (Fig. 5a). Changes in the environment (e.g. water

limited currents algal fungal mat

•02 -o 2

sediment

2 2

—

algal fungal mat sediment

—

Igal fungal mat

°2 02

sediment

Fig. 5. Variations of gyttja-type models for Toolebuc Formation oil shale deposition (after Kauffman, 1981). (a) algal-fungal mat and oxic-anoxic b o u n d a r y above water-sediment interface, (b) algal-fungal mat at water-sediment interface and oxicanoxic boundary at water-sediment interface, (c) algal-fungal mat at water-sediment interface and oxic-anoxic b o u n d a r y below water-sediment interface.

264

N. R. SHERWOOD & A. C. COOK

regimes and that this suggests that they are environmentally specific to zones near the aerobic-anaerobic boundary. As suggested above, a mat could aid in forming the boundary. Studying organic matter in oil shales and making comparisons with petrologically similar deposits allow an insight into depositional history. In the gyttja-type model, bituminite would be derived partly from the benthic algalfungal mat and in the stagnant basin model it would be derived from phytoplankton. Most dinoflagellate/acritarch cysts in Toolebuc Formation oil shales are thin-walled and they commonly have delicate processes (Fig. 3a, b). They were therefore probably deposited in a low-energy, distal, offshore environment. The presence of abundant liptodetrinite (probably largely from comminution of detached processes), thick-walled, naked lamalginite, and allochthonous inertinite and spore/pollen grains shows that there were probably times of at least limited current activity. From the work done to date on organic petrology of Toolebuc Formation oil shales, they were evidently deposited offshore, as anaerobic sediments. Limited current activity possibly occurred and, therefore, extending this inference and assuming a stagnant basin is probably misleading. It is uncertain whether the main organic matter is derived from planktonic, or planktonic and benthonic sources, but both are possibilities.

lamalginite to bituminite proximal to radioactive mineral grains. A marked gamma-ray anomaly is associated with Toolebuc Formation oil shales and a correlation has been found between uranium concentration and percentage of organic carbon (Ramsden et al., 1982). Humic acids have been found in Toolebuc Formation oil shale organic matter (Glikson, 1982) (probably deriVed-from bituminite) and some types of humic acids are capable of absorbing large amounts of uranium (Ramsden et al1, 1982). It is therefore a possibility that soluble radioactive matter in Toolebuc sea waters precipitated as submicroscopic uraninite and was absorbed by humic acids in bituminite (or algal precursors). Radioactive bombardment from uraninite would cause the reflectance to increase and fluorescence to decrease (i.e. coalification) giving rise to bituminite and subsequently to micrinite. Radiation haloes are not present and any radiation effects are pervasive rather than localized (cf. Stach et al1975, p. 60, 168).

CONCLUSIONS

The organic matter within Toolebuc Formation oil shale comprises essentially a two component system: dominant bituminite/micrinite and lamalginite (with associated liptodetrinite). The specific yield of shale oil appears to be markedly different for these two macerals (Fig. 4). A knowledge of the petrographic composition as well as the total organic carbon content is necessary to permit prediction of oil yields. Effects of regional rank are small; although the Postdeposition oil shales occur in an immature (soft brown coal to hard brown Although the exact origins of bituminite are uncertain, its coal) part of the stratigraphic section some oil may have been preservation and associations give an insight into generated. The uncertainties regarding the history of postdepositional effects. As mentioned above, bituminite deposition of Toolebuc Formation oil shales indicate that more probably has algal affinities; however, the optical properties work is required on stratigraphy, sedimentology and differ widely from those of material assigned to alginite. paleobiology to unravel the details. As the Toolebuc Formation Micrinite is commonly associated with bituminite and may be marginally mature for oil generation in some parts bituminite-micrinite transitions are common. If original algal of the Eromanga Basin, such a study has potential economic material was altered to bituminite, it could be due to the same as well as scientific interest. processes that have caused micrinitization. These processes could include decomposition (involving redistribution of oxygen present in original molecular structures during or after ACKNOWLEDGEMENTS deposition), biodegradation, gelification and coalification. This work was in part supported by a grant from the Aerobic phases may also have been of significance as National Energy Research Development and Demonstration evidenced by the presence of layers of benthonic shells. Council for a project on 'Petrology of Low Rank Oil Shales'. Micrinite can be a coalification product of bituminite (Stach David Martin did the drafting and Margaret Atkinson, et al, 1975) and it seems possible that bituminite can be a Margaret Voorwinden, Yvonne Watkin, Jenni Thew, Jenny coalification product of alginite. In a sample of Chattanooga Jarman and A. Moose did the typing. Analabs (Bentley, W.A.) mixed oil shale, the authors have observed transitions from carried out the Rock-Eval analyses.

REFERENCES

ARTHUR, M. A., 1979: Paleoceanographic events—recognition, resolution and reconsideration. Rev. Geophys. Space Phys. 17, 1474-94. ARTHUR, M. A. & SCHLANGER, S. O., 1979: Cretaceous 'oceanic anoxic events' as causal factors in development of reef-reservoired giant oil fields. Am. Assoc. Pet. Geol. Bull. 63, 870-85. COOK, A. C., HUTTON, A. C. & SHERWOOD, N. R., 1981:

Classification of oil shales. Centres Rech. Explore Prod. ElfAquitaine Bull. 5, 353-81. COOK, A. C., 1982: Organic facies in the Eromanga Basin; in Moore, P. S. & Mount, T. J. (compilers). Eromanga Basin symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 234-57. COOK, A. C., 1986: The nature and significance of the organic facies in the Eromanga Basin; in This volume.

DEAN, W. E. & GARDNER, J. V., 1982: Origin and geochemistry of

redox cycles of Jurassic to Eocene age, Cape Verde Basin (DSDP site 367), continental margin of northwest Africa; in Schlanger, S. O. & Cita, M. B. (eds). Nature and origin of Cretaceous carbon-rich facies. Academic Press, London, 69-77. DENSON, N. M. & GILL, J. R., 1965: Uranium-bearing lignite and carbonaceous shale in the southwestern part of the Williston basin—a regional study. U.S. Geol. Surv. Prof Pap. 463. DOWNIE, C., EVITT, W. R. & SARJEANT, W. A. S., 1963: Dinoflagellates, hystrichospheres and the classification of the acritarchs. Geol. Sci. Stanford Univ. Publ. 7, 1-16. EVANS, E. J., BATTS, B. D. & SMITH, J. W., 1984: Determination of the hydrocarbon prospectivity of sediments by hydrogenation. A PEA J. 24, 222-9. EVITT, W. R., 1963: A discussion and proposals concerning fossil dinoflagellates, hystrichospheres and acritarchs, I and II. Natl. Acad. Sci. U.S.A. Proc. 49, 158-64, 298-302.

TOOLEBUC FORMATION ORGANIC PETROLOGY GEBELEIN, C. D., 1969: Distribution, morphology and accretion rate

265

environment of the Toolebuc of recent algal stromatolites: Bermuda. J. Sediment. Petrol. 39, Formation, a Cretaceous oil shale in the Eromanga Basin-, 49-69. Symposium on research on the origins of black shales and oil shales in Australia. Baas Becking Laboratory, Canberra GLIKSON, M., 1982: Cyanobacterial mats and other bacteria: major {unpubl.). contributors to the formation of the Lower Cretaceous Toolebuc oil shales; in Moore, P. S. & Mount, T. J. (compilers) Eromanga OZIMIC, S., 1986: The geology and petrophysics of the Toolebuc Basin symposium, summary papers. Geol. Soc. Aust. & Pet. Formation and its time equivalents, Eromanga and Carpentaria Explor. Soc. Aust., Adelaide, 203. Basins; in This volume. HABIB, D., 1982: Sedimentary supply origin of Cretaceous black RAMSDEN, A . R., DICKSON, B. L. & MEAKINS, R. L., 1982: Origin shales; in Schlanger, S. O. & Cita, M. B. (eds) Nature and origin and significance of the Toolebuc gamma-ray anomaly in parts of Cretaceous carbon-rich fades. Academic Press, London, of the Eromanga Basin. Geol. Soc. Aust. J. 29, 285-96. 113-27. RAMSDEN, A. R., 1983: Microscopic petrography of oil shales at Julia Creek, northwestern Queensland. Geol. Soc. Aust. J. 30, 17-23. HUTTON, A . C., 1982: Sources of organic matter in Tertiary oil shales in Australia; Symposium on research on the origins of black RYAN, W. B. F. & CITA, M. B., 1977: Ignorance concerning episodes shales and oil shales in Australia. Baas Becking Laboratory, of ocean-wide stagnation. Mar. Geol. 23, 197-215. Canberra (unpubl.). SARJEANT, W. A. S., 1969: Microfossils other than pollen and spores HUTTON, A . C., KANTSLER, A . J., COOK, A . C . & MCKIRDY, D. M., in palynological preparation; in Erdtman, G. (ed.). Handbook 1980: Organic matter in oil shales. A PEA J. 20, 44-67. of palynology: morphology, taxonomy, ecology. Munksgaard, Copenhagen, 165-208. INTERNATIONAL COMMITTEE FOR COAL PETROLOGY, 1982: Reports by working group on unfigured organic matter (unpubl.). SARJEANT, W. A. S., 1970: Xanthidia, palinospheres and 'hystrix': review of the study of fossil unicellular microplankton with JACOB, H 1982: International Committee for Coal Petrology ring organic cell walls. Microscopy, 31, 221-56. analyses report for the working group on unfigured organic matter (unpubl.). SHERWOOD, N. R., COOK, A . C., GIBLING, M . & TANTISUKRIT, C . 1984: Organic petrology of a suite of sedimentary rocks JENKYNS, H . C., 1980: Cretaceous anoxic events: from continents to associated with some coal-bearing basins in northwestern oceans. J. Geol. Soc. London, 37, 171-88. Thailand. Int. J. Coal Geol. 4, 45-71. JOHNSON, G. O., 1981: Overview of oil shale development in Ohio; in Proceedings, 1981 eastern oil shale symposium. University SHERWOOD, N. R. & COOK, A. C., 1982: The organic petrology of a suite of oil shale samples from Kansas: a report prepared for of Kentucky Institute for Mining and Minerals Research, Kansas University. Department of Geology, University of Lexington, 169-72. Wollongong {unpubl.). KANTSLER, A. J., COOK, A. C . & ZWIGULIS, M., 1986: Organic SOUTAR, A. & CRILL, P. A., 1977: Sedimentation and climatic maturation in the Eromanga Basin; in This volume. patterns in the Santa Barbara Basin during the 19th and 20th KAUFFMAN, E. G., 1981: Ecological reappraisal of the German centuries. Geol. Soc. Am. Bull. 88, 1161-72. Posidonienschiefer (Toarcian) and the stagnant basin model; in Gray, J., Boucot, A. J., & Berry, W. B. N. (eds). Communities STACH, E., MACKOWSKI, M.-TH., TEICHMULLER, M., TAYLOR, G. H., CHANDRA, D. & TEICHMULLER, R., 1975: Stach's textbook of of the past. Hutchinson, Stroudsberg, 311-81. coal petrology. Borntraeger, Berlin. LISTER, T. R., 1970: A monograph of the acritarchs and chitinozoa from the Wenlock and Ludlow Series of the Ludlow and STANDARDS ASSOCIATION O F AUSTRALIA, 1981: Methods for microscopical determinations of the reflectance of coal Millichope areas, Shropshire. Palaeontogr. Soc. Monogr. 1, macerals. Standards Association of Australia, Sydney. 1-100. SWANSON, V. E., 1960: Oil yield and uranium content of black shales. MOORE, P. S. & MOUNT, T. J. (compilers), 1982: Eromanga Basin U.S. Geol. Surv. Prof. Pap. 365-A. symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. TEICHMULLER, M., 1974: Uber neue macerale der liptinit-gruppe und Soc. Aust., Adelaide. die entstehung von micrinit. Fortsch. Geol. Rheinld. u. Westf MOORE, P. S. & PITT, G. M., 1984: Cretaceous of the Eromanga 24, 37-64 (English translation by Wheelhouse, G. H.). Basin—implications for hydrocarbon exploration. APE A J. 24, THIEDE, J. & VAN ANDEL, T. H., 1977: The paleoenvironment of 358-76. anaerobic sediments in the Late Mesozoic south Atlantic Ocean. MORRIS, K. A., 1980: Comparison of major sequences of organicEarth Planet. Sci. Lett. 33, 301-9. rich mud deposition in the British Jurassic J. Geol. Soc. London, TISSOT, B. P. & 'WELTE, D. H., 1978: Petroleum formation and 37, 157-70. occurrence: a new approach to oil and gas exploration. SpringerMULLER, P. J. & SUESS, E., 1979: Productivity, sedimentation rate, Verlag, Berlin. and sedimentary organic matter in the oceans—I. Organic WILLIAMS, G. L., 1978: Dinoflagellates, acritarchs and tasmanitids; carbon preservation. Deep Sea Res. 26A, 1347-62. in Haq, B. U. & Boersma, A. (eds) Introduction to OZIMIC, S., 1982C: Depositional environment of the Toolebuc micropaleontology. Elsevier, New York, 292-326. Formation and its equivalents, Eromanga Basin, Australia; in Moore, P. S. & Mount, T. J. (compilers). Eromanga Basin WILLIAMS, L. A . & REIMERS, C . 1983: Role of bacterial mats in oxygen-deficient marine basins and coastal upwelling regimes: symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. preliminary report. Geology, 11, 267-69. Soc Aust., Adelaide, 2 0 0 - 2 . M

OZIMIC, S., 1982b: Depositional

Geological Society of Australia Special Publication No. 12. 267-272

Organic matter and vanadium in the Toolebuc Formation, northern Eromanga Basin and southern Carpentaria Basin K. W. Riley & J. D. Saxby

CSIRO Division of Fossil Fuels, P.O. Box 136, North Ryde, NSW 2113.

ABSTRACT

Organic matter and vanadium are closely associated in oil shales of the Toolebuc Formation in the northern Eromanga and southern Carpentaria Basins. Vanadium occurs in a number of chemical forms, including vanadates, hydrated oxides, silicates and within porphyrin-type organic complexes. The concentration of vanadium varies from 250 to 3200/^g g in a selection of organic-rich samples from near Toolebuc Formation outcrops in widely separated areas of the basins. The organic carbon content of the same samples varies from 5.5 to 16.3%. The origin of the vanadium-rich organic matter is not clear. It may arise from particular organisms incorporated in the sediment or by diagenetic replacement by vanadium of magnesium in chlorophyll pigments of algae or bacteria. Another possibility is a contribution from a crude oil residue, resulting either from a large oil seep into the extensive Toolebuc sea' or from migration into an initially porous Toolebuc,Formation limestone reservoir followed by bacterial degradation. Recent deep weathering has redistributed the vanadium in most samples. _l

INTRODUCTION

The Julia Creek oil shale deposit, discovered in 1966, has a high vanadium content. The aim of this paper is to give an overview of vanadium distribution and geochemistry, not just at Julia Creek, but also in other areas, particularly the northern Eromanga Basin. Samples of organic-rich sediments and weathered surface material from widely separated sites in the Early Cretaceous Toolebuc Formation and two similar samples from the southern Carpentaria Basin are investigated. Specific objectives have been to examine the distribution of vanadium, association of vanadium with carbonaceous material, and composition of the organic matter. The Eromanga Basin is one of the most extensive basins in Australia and in general terms covers an area of one million square kilometres from latitudes 20 °S to 34 °S and longitudes 132 °E to 147 °E. The Carpentaria Basin is contiguous with the Eromanga Basin across the Euroka Arch and extends northward, underlying most of the Gulf of Carpentaria. The heterogeneous Toolebuc Formation consists of limestone, coquinite, labile sandstone, siltstone, mudstone, calcareous shale and oil shale, and varies in thickness up to 75 m (Senior et al., 1975, 1978; Smart et al, 1980). The oil shales are believed to have been deposited in reducing conditions in sheltered areas of a shallow epicontinental sea which covered the basins in the late Albian (Exon & Senior, 1976). Vast reserves of low-grade oil shale occur within the Toolebuc Formation (Swarbrick, 1974), and occupy an area of almost 0.5 million square kilometres, north of an approximate line joining Bedourie and Charleville (Ozimic & Saxby, 1983). However, most investigations have centred on an area near Julia Creek where the formation outcrops

and substantial reserves are sufficiently near the surface for open-cut mining to be feasible. Petrographic studies of oil shale from this region indicate that it is well-laminated with carbonate interbeds and consists of a mixed assemblage of organic matter within a matrix of micrite and clay minerals (Hutton et al., 1980). The occurrence of vanadium in both weathered and deeper oil shale samples has been investigated by Norrish & Patterson (1976). It was concluded that most of the vanadium in oil shale from Julia Creek is associated with interstratified micaceous montmorillonite clay, with only minor amounts chemically-bound within the organic matter. Vanadium bonded to organic matter, including vanadyl porphyrins, has recently been detected in selected oil shale samples from the Toolebuc Formation (Riley & Saxby, 1982). Ekstrom et al. (1982), using mass spectrometry, and Fookes & Loeh (1983), using nuclear magnetic resonance techniques, have undertaken chemical studies of the porphyrins in Julia Creek oil shale.

EXPERIMENTAL Samples

Oil shale samples were obtained from 14 stratigraphic drill holes and petroleum exploration wells (Table 1). Samples were selected from core recovered from each well using high organic carbon content as the main criterion. Extensively weathered oil shale was obtained from the Toolebuc Formation outcrop on the St Elmo Structure near Julia Creek. The distribution of samples sites is shown in Figure 1. Samples were crushed to less than 210 /tm and air dried before analysis.

268

K. W. RILEY & J. Df SAXBY

Determination of carbon and vanadium in oil shales Organic carbon contents were determined by difference, after analysis for total carbon (Leco Analyser) and carbonate carbon (Knott & Belcher, 1975). Total vanadium was determined by atomic absorption spectrometry (AAS) (Varian Techtron AA5/IM6 Spectrophotometer) after complete dissolution of the samples using hydrofluoric, nitric and perchloric acids. Vanadium soluble in hydrochloric acid, and vanadium soluble in hydrochloric acid/hydrofluoric acid were

determined by AAS on solutions obtained by sequential digestion with 5 M hydrochloric acid and 2.5 M hydrochloric acid/15 M hydrofluoric acid. Results are given in Tables 1 and 2.

Chloroform extraction Oil shale samples (5 g) were extracted for 8 hours with distilled A.R. grade chloroform in a soxhlet. Extracts were adjusted to a predetermined volume with chloroform and

TABLE 1. Sample details, carbon and vanadium analyses. Sample Location

Organic carbon (%)

Carbonate carbon (%)

Total vanadium (jig g~')

5.9 5.9 13.1 9.6 6.9 5.5 16.3 11.7 13.7 < 0.1 6.6 11.4 13.7 10.6 8.7

0.3 2.0 6.2 7.3 8.0 <0.1 6.4 3.0 4.5 9.3 <0.1 7.5 6.8 3.3 1.2

250 750 1900 1650 1000 250 3200 1600 1900 2600 300 1800 2500 1850 300

Augathella 7; 146°16'12"E, 25°42'16"S; 75.15-75.3 m Bedourie 1; 139°20'30"E, 24°15'40"S; 107.36-108.36 m Boulia 3A; 140°43'30"E, 22°47'00"S; 32.05-32.29 m. Boulia 10B; 140°46'05"E, 22°42'25"S; 42.97-44.00 m Burketown 1; 139°32'E, 18°04'S; Core 5; 394.70 m Charleville 5; 146°02'04"E, 26°23'17"S; 60,95-61.05 m Dobbyn 2; 140°13'E, 19 0 ll'S; Core 6; 114.30 m Jericho 11; 145°3ri8"E, 23°51'30"S; 61.93-62.15 m Julia Creek, exploration drilling (1980); 30.00-32.00 m Weathered coquinite from outcrop, St Elmo Structure, Julia Creek Longreach 6; 145°19'16WE, 23°37'35"S; 29.50-29.65 m Mayneside 1; 142°31'E, 23°35'S; Core 1; 905.60 m Springvale 9; 140°37'50"E, 23°08'10"S; 134.00-135.11 m Tambo 38; 145°39'07"E, 24°15'00"S; 58.90-59.00 m Tambo 44; 146°09'25"E, 24°58'22"S; 17.25-17.50 m

TABLE 2. Distribution of vanadium in Toolebuc Formation oil shale (expressed as percentage of total vanadium). Sample Augathella 7 Bedourie 1 Boulia 3A Boulia 10B Burketown 1 Charleville 5 Dobbyn 2 Jericho 11 Julia Creek—exploration Julia Creek—weathered outcrop Longreach 6 Mayneside 1 Springvale 9 Tambo 38 Tambo 44 n.d. Not determined

HCl-soluble

HCl/HF-soluble

CHCl 3 -extractable

Kerogen-bound

63 74 n.d. 81 81 58 94 56 34 87 60 37 82 69 62

37 17 n.d. 11 16 41 4 36 n.d. 12 39 14 10 25 31

<0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 <0.1 0.7 <0.1 <0.1 0.8 <0.1 <0.1 <0.1

n.d. n.d. 1 n.d. 3 <1 2 n.d. 22 <1 <1 48 n.d. n.d. n.d.

TABLE 3. Chloroform-extractable material in Toolebuc Formation samples. Total extract

Sample

Augathella 7 Bedourie 1 Boulia 3A Boulia 10B Burketown 1 Charleville 5 Dobbyn 2 Jericho 11 Julia Creek—exploration Julia Creek—weathered outcrop Longreach 6 Mayneside 1 Springvale 9 Tambo 38 Tambo 44

(«gg"')

(% of total organic matter)

Metallated porphyrin («g g " V

Vanadium (ug g' 1 )

1500 1400 7500 2800 2700 1100 4000 4900 16 000 n.d. 2800 16 000 4700 6000 3000

1.9 1.8 4;3 2.2 2.9 1.5 1.9 3.2 8.8 n.d. 3.1 11.0 2.6 4.3 2.9

<2 <2 < 2 10 <2 <2 4 6 350 <0.005 <2 730 12 5 <2

<1 <1 <1 2 <1 <1 1 2 110 n.d. <1 150 3 1 <1

* Calculated as vanadyl deoxophylloerythroetioporphyrin (M.W. = 542) n.d. Not determined

TOOLEBUC FORMATION ORGANIC MATTER AND VANADIUM

269

Fig. 1. Location of sampled drillholes and outcrops of Toolebuc Formation.

ultraviolet-visible spectra obtained in the 350-700 nm wavelength region (Varian Techtron Model 635 UV-VIS Spectrophotometer). Concentrations of extractable metal porphyrins were calculated from the spectra using the absorbance values at the «-peak (575 nm, e = 2 x 104 1 mol-'cm 1 ) and the Soret peak (412 nm, e = 3.3 x 105 1 mol 'cm" 1 ) (Hodgson et al.y 1968). Chloroform was

removed from the extracts under a stream of nitrogen and the residues were digested with nitric and perchloric acids prior to analysis for vanadium. Results are given in Table 3.

Composition of kerogen Kerogen has been isolated by demoralization from a number of the selected samples (Riley & Saxby, 1982).

270

K. W. RILEY & J. D. SAX BY

4. Elemental compositions and vanadium content of kerogens. Nitrogen Hydrogen Sulphur Carbon Sample (%, dry, mineral-free basis) 2.7 7.1 5.5 71.2 Boulia 3A 2.4 6.2 7.7 Burketown 1 72.8 2.6 7.1 6.5 Dobbyn 2 69.8 2.3 5.5 74.5 7.0 Mayneside 2.5 5.5 . 77.1 7.8 Julia Creek—exploration 5.1 1.7 2.7 58.2 Julia Creek—weathered outcrop * By difference n.d. Not determined

TABLE

Elemental compositions and vanadium contents of these kerogens are given in Table 4.

RESULTS AND DISCUSSION

Oxygen* 13.6 10.9 14.1 '10.8 7.1 32.2

Atomic ratios O/C H/C 0.143 1.20 0.112 1.27 0.155 1.22 0.109 1.13 0.069 1.21 0.410 0.56

Ash Vanadium W 0<g g ) 10.2 150 12.1 300 8.7 300 3.0 3700 13.7 1900 2.4 n.d. _ l

0-i,

The concentration of vanadium in the samples varies from 0-3 / 250 to 3200 /xg g~'. Within the Toolebuc Formation high levels of vanadium are widely distributed over the Eromanga Basin and, in all probability, over the Carpentaria Basin. /o / Lower concentrations of vanadium tend to be found in samples from the southeast towards Augathella and Charleville, where the sediments generally contain less organic 0-2 / matter and also less carbonate (Fig. 1, Table 1). Two samples ° As o / o from this southeastern area are higher in carbonate (Jericho / 11, Tambo 38) and contain more organic matter and vanadium / than samples from neighbouring locations. Ozimic (1982), on the basis of relative abundances of coquinite (limestone) and 0-1 calcareous shale, defines two kerogenous facies in the Toolebuc o / Formation. Thus the varying amounts of carbonate, organic / matter and also vanadium may be a reflection of a change / in facies. o-o Alternatively,, mobile organic compounds rich in vanadium 0 5 10 15 (such as those found in crude oils) may have migrated into the limestone-rich sediment or even have seeped into the basin Organic carbon (wt '/•) during active sedimentation. Diagenesis would have altered between total vanadium and organic carbon for this migrated material during and after its incorporation into Fig. 2. Relationship selected organic-rich oil shale samples from the Toolebuc organic matter already in the sediments. This would explain Formation (Table 1). the relationship between vanadium and organic carbon (Table 1, Fig. 2), where a common source for the vanadium and most of the organic matter is indicated. Extrapolation vanadium (~60 to 90%) occurs in a hydrochloric acid-soluble of the line in Figure 2 seems to indicate that approximately form. The only exceptions are the oil shale from Julia Creek4 per cent organic carbon, which is not associated with exploration and Mayneside 1, where 20 to 50 per cent of the vanadium, is present in these samples. vanadium is bound in the kerogen and approximately 35 per Although the vanadium is initially related to the organic cent is present in a hydrochloric acid-soluble form. matter, it occurs at all locations in a variety of forms, The UV-visible spectra of the chloroform extracts indicate including: that vanadyl porphyrins are present in most samples. Vanadyl (1) species soluble in hydrochloric acid (probably hydrated porphyrins are most dominant for those samples (Julia oxides and vanadates adsorbed on clays or precipitated Creek—exploration and Mayneside 1), in which high in the limestone), concentrations of kerogen-bound vanadium occur. Spectra (2) species soluble in hydrofluoric acid (probably vanadium with maxima occurring at about 412, 537 and 575 nm are silicates within clays or micas), typical of vanadyl deoxophylloerythroetioporphyrins or (3) organically-bound vanadium. etioporphyrins (Fig. 3a, b) (Hodgson & Baker, 1967; Hodgson This final form includes extractable porphyrin complexes et al., 1967). There is also a peak at approximately 600 nm such as are shown in Figure 3a-c and higher molecular weight in spectra of the porphyrin-rich extracts. This is possibly due vanadium complexes within the kerogen, including possible to the presence of rhodoporphyrins,' i.e. porphyrins having highly-fused aromatic porphyrin structures of the type shown a benzene ring fused to, and conjugated with, the main in Figure 3d (Yen, 1975). Evidence for the existence of high porphyrin nucleus (Baker et al1967). The presence of a molecular weight porphyrins in a Triassic oil shale has been mixture of porphyrins (including rhodoporphyrins), whose reported by Blumer & Snyder (1967), who suggested that absorbance maxima do not exactly coincide, would help polymeric porphyrins are incorporated into the structure of explain the discrepancy between the molar amounts of kerogen. Oxidation of the exterior of a relatively small vanadium and porphyrins extracted. There are approximately porphyrin ring system may also render the resulting vanadium two moles of vanadium extracted for each mole of porphyrin complex relatively insoluble in organic solvents. Most of the determined by considering the oi and Soret peaks to be due

TOOLEBUC FORMATION ORGANIC MATTER AND VANADIUM

(a)

(b)

271

vanadium from the organic matter. If this has occurred, it is clear that the release of organically-bound vanadium cannot be directly related to depth below the present surface. The Julia Creek sample was obtained from only 30 m, whereas other samples from greater depths contain little or no organically-bound vanadium. Since the Toolebuc Formation is a confining bed between aquifers in the Great Artesian Basin (Habermehl, 1980), the movement of water (and bacteria) through these aquifers may result in the release of vanadium from organic matter at varying depths. The generally uniform composition of the kerogens indicates that the release of organically-bound vanadium occurs at an early stage in alteration of the oil shale. Vanadium liberated as oxides or vanadates could be transported by percolating waters to sites of adsorption on clays or be precipitated as calcium vanadates in the limestone (Evans, 1978). Redistribution of vanadium within the inorganic matrix of the oil shale would be dependent on factors such as Eh and pH of the ground water (and hence the oxidation state of vanadium) and relative abundances of clay and carbonate minerals. Recent weathering is more likely to result in the formation of hydrated oxides of vanadium and calcium vanadates, rather than the incorporation of vanadium into clay structures. Thus the observed relative abundances of hydrochloric acid-soluble and hydrochloric/hydrofluoric acid-soluble vanadium may indicate that most of the vanadium release has occurred during the present-day weathering episode.

(d) Fig. 3. Typical structures of vanadyl porphyrins: (a) deoxophylloerythroetioporphyrin, (b) etioporphyrin III, (c) rhodo-type benzporphyrin, (d) possible structure of a highly fused aromatic porphyrin molecule (from Yen, 1975).

to a single porphyrin type (Fig. 3a or 3b). Concentrations of vanadium in both original shale and extracts from Julia Creekexploration and Mayneside 1 are not unlike those found in the Dictyonema oil shale of Estonia. The latter contains 0.09 per cent vanadium and gives an extract containing 0.32 per cent vanadium (Urov & Klesment, 1979). Elemental compositions of all kerogens isolated so far from Toolebuc Formation samples indicate that the organic matter is fairly uniform throughout a large area. The kerogens are not particularly rich in hydrogen ( H / C = 1.2) compared with most other oil shales ( H / C ~1.5) (Robinson & Dinneen, 1967). Organic sulphur and nitrogen contents are high and both occur predominantly in heterocyclic aromatic structures. Kerogen isolated from the weathered coquinite is much lower in hydrogen ( H / C = 0.56), lower in sulphur and higher in nitrogen. This kerogen, which comprises the largest aromatic polycyclic nuclei of the original organic matter, has survived to the present day because of its resistance to complete weathering and bacterial attack. Partial oxidation of this kerogen has reduced its original H / C value and increased its initial O / C ratio. Petrographic examination of the oil shales indicates that the major component of organic matter in the Toolebuc Formation is bituminite (64-84% by vol.), with lesser amounts of lamalginite (6-25%) and micrinite (4-13%) (M. Smyth, personal communication). Although organic matter is reasonably uniform in composition over a wide area of the Toolebuc Formation, there are large variations in the distribution of vanadium between organic and inorganic phases. Reasons for this are not clear. It has been suggested (Riley & Saxby, 1982) that weathering processes have resulted in the destruction of vanadyl porphyrin structures within the kerogen and hence in the release of

CONCLUSIONS (1) Concentration of vanadium in 14 widely separated oil shale samples from the Toolebuc Formation varies from 250 to 3200 fig g" 1 . In a sample of extensively weathered Julia Creek shale 2600 Mg g _ I are present. These contents are all much higher than typical average levels for shales (130 ng g" 1 ), sandstones (20 /*g g - 1 ), limestones (20 ng g" 1 ) and igneous rocks (135 iig g _ l ) (Bowen, 1966). Vanadium concentrations tend to be higher in the Carpentaria Basin and in northern parts of the Eromanga Basin and lower in the southeast of the study area. (2) Vanadium and some of the organic matter in the oil shales have a common source. Vanadium-rich organic matter of algal or bacterial origin may have been deposited from the Toolebuc sea' or, after deposition, vanadium in percolating waters may have replaced other metals in chlorophyll-type pigments from algae or bacteria. Alternatively, vanadium-rich organic matter could conceivably result from widespread seepage of crude oil into the basin or from degradation of petroleum reservoired in an initially porous Toolebuc Formation limestone. Diagenesis and limited maturation during subsequent burial have produced the resultant mixed kerogenous material. (3) Although vanadium is related to organic matter, it occurs in a number of chemical forms in the Toolebuc Formation. Most is present as relatively labile adsorbed or precipitated species, possibly hydrated oxides or metal vanadates. Vanadium also occurs in the silicate structure of micas or clays and, in some areas, significant amounts of organically-bound vanadium are present. Particularly high concentrations of organically-bound vanadium have been found at Julia Creek and the sample from 906 m in Mayneside 1 has about half its vanadium chemically bound within the kerogen. These samples contain the lowest proportion of labile vanadium ( - 3 5 % ) . (4) Vanadium is probably mobilized from the organic matter by weathering and is concentrated in adjacent inorganic

272

K. W. RILEY & J. D. SAXBY

host rocks. This mobilization occurs at varying depths due to oxidation and biological activity associated with water movement through aquifers of the Great Artesian Basin, in which the Toolebuc Formation acts as one of the confining beds. (5) Mobilization of vanadium appears to occur before there is significant alteration of the average chemical composition of organic material. Elemental analyses of kerogen from the Toolebuc Formation are quite similar over wide areas where the oil shale facies persists.

ACKNOWLEDGEMENTS

The collaboration of CSR Limited and the Bureau of Mineral Resources in the provision of. samples and in discussions is gratefully acknowledged. Some of the samples used in this research were obtained during a joint CSIRO/BMR project supported under the National Energy Research, Development and Demonstration programme administered by the Commonwealth Department of Resources and Energy.

REFERENCES BAKER, E. W., YEN, T. F., DICKIE, J. P., RHODES, R. E. & CLARK,

L. F., 1967: Mass spectrometry of porphyrins. II. Characterization of petroporphyrins. Am. Chem. Soc. J. 89,

3631-9. BLUMER, M. & SNYDER, W. D., 1967: Porphyrins of high molecular

KNOTT, A. C. & BELCHER, C. B., 1975: Determination of carbon dioxide in coal and minerals. Talanta, 22, 751-3. NORRISH, K. & PATTERSON, J. H., 1976: Characterisation of

vanadiferous clays, Julia Creek, Queensland. 25th International Geological Congress, Sydney, Symposium 104.2, Abstracts, 3, weight in a THassic oil shale: evidence by gel permeation 756-7. chromatography. Chem. Geol. 2, 35-45. OZIMIC, S., 1982: Depositional environment of the oil shale-bearing BOWEN, H. J. M., 1966: Trace elements in biochemistry. Academic Cretaceous Toolebuc Formation and its equivalents, Eromanga Press, London. Basin, Australia. 15th Oil Shale Symposium Proc., Colorado School of Mines, 137-48. EKSTROM, A., LOEH, H. & DALE, L., 1982: The petroporphyrins in the oil shale from the Julia Creek deposit; in Moore, P. S., & OZIMIC, S. & SAXBY, J. D., 1983: Oil shale methodology, final report Mount, T. J. (compilers) Eromanga Basin Symposium summary of project 78/2616, National Energy Research Development and papers, Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, Demonstration Program, Department of Resources and Energy, 215-6. Canberra, NERDDP/EG/83/187. EVANS, H. T., Jr, 1978: Vanadium; in Wedepohl, K. H. (ed.) RILEY, K. W. & SAXBY, J. D., 1982: Association of organic matter Handbook of geochemistry. Springer-Verlag, Berlin. and vanadium in oil shale from the Toolebuc Formation of the Eromanga Basin, Australia. Chem. Geol. 37, 265-75. EXON, N. F. & SENIOR, B. R., 1976: The Cretaceous of the Eromanga and Surat Basins. Aust. Bur. Miner. Resour. Geol. Geophys. J. ROBINSON, W. E. & DINNEEN, G. U., 1967: Constitutional aspects 1, 33-50. of oil-shale kerogen; in Proceedings of the 7th world petroleum congress. Elsevier, Amsterdam, 3, 669-80. FOOKES, C. J. R. & LOEH, H. J., 1983: Porphyrins of Australian oil shales: a nuclear magnetic resonance study. Proc. First Aust. SENIOR, B. R., EXON, N. F. & BURGER, D., 1975: The Cadna-owie Workshop on Oil Shale, Lucas Heights, 65-8. and Toolebuc Formations in the Eromanga Basin, Queensland. Qld Govt Min. J. 76, 444-55. HABERMEHL, M. A., 1980: The Great Artesian Basin, Australia. Aust. Bur. Miner. Resour. Geol. Geophys. J. 5, 9-38. SENIOR, B. R., MOND, A. & HARRISON; P. L., 1978: Geology of the HODGSON, G. W. & BAKER, B. L., 1967: Spectra of selected Eromanga Basin. Aust. Bur. Miner, Resour. Geol. Geophys. Bull. 167. geochemically significant porphyrins and chlorins. Chem. Geol. 2, 187-98. SMART, J., GRIMES, K. G., DOUTCH, H. F. & PINCHIN, J., 1980: The Carpentaria and Karumba Basins, North Queensland. Aust. Bur. HODGSON, G. W., BAKER, B. L. & PEAKE, E., 1967: Geochemistry Miner. Resour. Geol. Geophys. Bull. 202. of porphyrins; in Nagy, B. & Colombo, U. (eds) Fundamental aspects of petroleum geochemistry. Elsevier, Amsterdam, SWARBRICK, C. F. J., 1974: Oil shale resources of Queensland. Qld 177-251. Geol. Surv. Rep. 83, 22-9. HODGSON, G. W., HITCHON, B., TAGUCHI, K., BAKER, B. L. & UROV, K. E. & KLESMENT, I. R., 1979: Geochemical characterization PEAKE, E., 1968: Geochemistry of porphyrins, chlorins and of the organic matter in Precambrian and Early Palaeozoic schists. Geochem. Int. 16, 50-6. polycyclic aromatics in soils, sediments and sedimentary rocks. Geochim. Cosmochim. Acta, 32, 737-72. YEN, T. F. (ed.), 1975: Chemical aspects of metals in native petroleum; HUTTON, A. C., KANTSLER, A. J., COOK, A. C. & MCKIRDY, D. M., in The role of trace metals in petroleum. Ann Arbor Science, 1980: Organic matter in oil shales. APEA J. 20, 44-67. Ann Arbor, Michigan, 1-30.

Geological Society of Australia Special Publication No. 12, 273-286

Cyanobacterial mats: major contributors to the organic matter in Toolebuc Formation oil shales M. Glikson & G. H. Taylor

Centre for Resource and Environmental Studies; Australian National University, GPO Box 4, Canberra, A.C.T. 2601.

ABSTRACT

Oil shales of the Early Cretaceous Toolebuc Formation in the Eromanga Basin, Queensland, consist of irregularly spaced calcareous and organic-rich laminae suggestive of cyanobacterial mats. Observations of the predominant organic matter in reflected light microscopy reveal it to be of low reflectance (R = 0.1-0.3%) with a vitrinite-like nature. Vitrinite reflectance is about 0.5%. The low maturity of the Toolebuc Formation is also confirmed by the relatively high H/C vs Q/C atomic ratios. Scanning electron microscopy of whole rock fragments, and transmission electron microscopy of organic matter concentrates reveals filamentous organisms resembling Oscillatoria. Transmission electron microscopy of organic matter also reveals humic acids and other microbial remains; the latter were most likely responsible for the degradation of primary organic matter, mainly in an oxygenated water layer. The occurrence of pyrite within the calcite indicates that calcite was deposited in the anoxic zone. The irregular extent and thickness of the calcareous laminae are inconsistent with seasonal precipitation. They are more likely to have resulted from microenvironmental changes caused by high population densities attained by cyanobacteria, leading to their eventual death. Ammonia released from cyanobacterial decomposition was responsible for a rise in pH which resulted in calcite precipitation. This is supported by the carbon isotopic composition of crystalline calcite associated with organic matter, as compared to carbon isotopic compositions of Inoceramus shells. The isotopic composition of most organic matter in the oil shales (5 C = -28 to -29ppt) may indicate C0 -enriched environment as well as contributions from isotopically lighter microbial lipids. The abundance of planktonic foraminifera and coccolith remains in the oil shales suggests a productive oxygenated upper water layer in the environment of deposition. Anoxic conditions above or at the sedimentwater interface are evident from the preservation of high concentrations of organic matter and an abundance of pyrite. 0

INTRODUCTION

Hutton et al. (1980) applied optical microscopy to the study of organic matter in oil shales and observed that the organic matter in the Toolebuc Formation oil shales is of a very finegrained nature, and constitutes a mixture of alginite B and other components, the former being more abundant. These oil shales were assigned by the above authors to 'mixed type'. Under reflected light the oil shale appears to be laminated with 'carbonate interbeds'. The constituents of the organic matter were described as: Tasmanites, palynomorphs, acritarchs, dinoflagellates, fragmented cutinite and liptodetrinite. Hutton et al. (1980, p. 58) also mention 'lamellar alginite, with yellow to orange auto-fluorescence'. Cook et al. (1981) pointed out the close association of organic matter in the Toolebuc Formation with calcite, as is evident from polished rock surfaces observed in reflected light. The predominance of vitrinite-like organic matter in lamellar arrangement, its weak fluorescence after irradiation and resemblance to the vitrinite-like organic matter in the Camooweal oil shale were also pointed out by these authors. Ramsden et al. (1982) carried out a detailed microscopic study of the Toolebuc oil shales, emphasising the abundance of

coccolith remains, and proposed the term 'coccolithic micrite' to some horizons where the remains of this planktonic organism seem to be predominant. Ramsden et al. (1982) used scanning electron microscopy (SEM) in addition to optical microscopy. Extensive microscopical studies, applying SEM, were carried out by Scheibnerova (pers. comm., 1982) who noted unusual structures which are interpreted in the present study (Fig. 1C, D, E, G) as being the remains of cyanobacteria. In the present study examination of polished surfaces of whole rock samples revealed the organic matter to be of a lamellar appearance, closely associated with calcareous matter (Fig. 1A). The organic matter is of vitrinite-like nature but of a lower reflectance than vitrinite (Figs 8, 9). The organicrich laminae interbedded with calcite are morphologically suggestive of 'algal mats'. Pyrite abounds in massive and framboidal form within the organic matter as well as within calcareous remains of organisms (Fig. IB) and within the crystalline calcite. A wealth of planktonic foraminifera and coccoliths is apparent from polished rock surfaces in reflected light as well as thin sections in transmitted light. It seems likely from these observations that the planktonic organisms were deposited into a soft mat and appear to have displaced and

274

M. GLIKSON & G. H. TAYLOR

CYANOBACTERIAL MATS IN O I L S H A L E S

275

TABLE 1. Stable c a r b o n - i s o t o p e c o m p o s i t i o n o f organic matter from selected samples, Toolebuc Formation. O M = organic matter; H A = humic acids. Name o f Well and Sample Number Tambo 3 8 / 1 0 - 2 Tambo 38/11-1 Tambo 3 8 / 1 1 - 2 Tambo 38/ 1-3 Tambo 3 8 / 1 1 - 3 Tambo 3 8 / 1 1 / 3 Tambo 3 8 / 6 - 3 Tambo 3 8 / 6 - 3 Tambo 3 8 / 2 - 3 Tambo 3 8 / 9 - 3 Boulia Boulia Boulia Boulia Boulia Boulia

Fraction

<5 , 3 C PPT Relative to P D B

61.0 63.0

OM OM OM OM OM-HA HA OM HA OM OM

-27.75 -26.35 -27.61 -27.60 -28.40 -28.40 -28.86 -25.60 -23.00 -27.99

30.0 30.5 31.0 31.4 32.0 32.3

OM OM OM OM OM OM

-27.27 -28.40 -28.14 -29.00 -27.80 -28.15

Depth in m 54.0 56.0 57.0 58.0-59.0

3/B3-2 3/B3-3 3/B3-5 3/B3-1 3/B3-4 3/B3-6

Jericho 1 1 / J 1 1 - 1 1

54.0

-27.12

Longreach 6 / L 6 - 2 Longreach 6 / L 6 - 1 2

25.3 28.0

OM OM

-27.90 -27.19

Augathella 6 / A U 6 - 4 0 Augathella 6 / A U 6 - 4 0

32.0 32.0

Julia Ck (weathered z o n e ) Julia Ck (weathered z o n e ) Urella D o w n s / G S - 6 Urella D o w n s / G S - 4 'Petita B o r e V G S - 5 Birrigoolpa/GS-8

warped the mats as indicated f r o m their position, which is discordant with the laminations. T h e estimated proportion of organic to mineral matter varies from about 10 to 30 per cent. The main mineral component is calcite which is either associated with the organic matter or occurs in the form of remains of organisms such as foraminifera and coccoliths. Mineralogy of the Toolebuc Formation oil shales as shown by X-ray diffraction reveals that the minerals comprising the majority of samples are calcite and clay (montmorillonite, illite and chlorite). Sample 6 - 3 (Tambo 38, Table 1) proved to have a significant quartz component. An apparent feature of the calcite-organic matter association is the irregularity in extent and thickness of laminae and their often intergrown nature. This is best observed when viewing polished rock surfaces in reflected light. When using ultraviolet (UV) light with blue filter excitation, the only fluorescence response comes from occasional occurrences of land plant spores or pollen (dark yellow fluorescence) and from occasional dinoflagellates (greenish-yellow fluorescence). The predominant vitrinite-like organic matter lacks fluorescing properties. Organic matter concentrates observed in transmitted light are predominantly amorphous, lacking any structure attributable to an organism. Observations were carried further by using transmission electron microscopy (TEM) to examine ultra thin sections of the organic matter. Microbial components of the a m o r p h o u s organic matter are beyond the resolving power of the optical microscope and can only be detected with T E M .

32.0 120.0 301.0 127.0

OM-HA HA

-25.10 -22.90

OM-HA HA

-25.35 -25.56

OM OM OM OM

-22.00 -23.60 -23.31 -25.70

METHODS OF STUDY Optical microscopy has been carried out on polished rock surfaces with the aim of observing relationships between the organic matter and the mineral matrix. Ultraviolet (UV) mode with blue filter excitation was also applied. Whole rock fragments were examined by scanning electron microscopy (SEM). Transmission electron microscopy (TEM) observations were made on ultra-thin sections of organic matter concentrates. After demineralisation of the samples by HC1 and HF, the organic matter concentrates were embedded in polyester resin and sectioned on an ultramicrotome to a thickness of 0.03 to 0.05 fim. Sections were then lifted on to formvar-coated copper grids for observation in a J E O L 100 C electron microscope. Humic acids were extracted by sodium hydroxide treatment and acid precipitation. The dry humic acids were then embedded in polyester resin, ultra-thin sections prepared and observed in T E M . Lead citrate and colloidal iron staining were used on duplicate sections for comparative observations. The carbon isotopic composition of the organic matter was determined by the method of Kaplan et al. (1970). C 0 2 from calcium carbonate samples was collected following phosphoric acid treatment. Isotope ratios were determined on a Micromass 602D mass spectrometer. In selected samples, humic acids were analysed separately. M a j o r element analysis of C , H , 0 , N and S was carried out on organic matter concentrates. Yeh's (1963) method was used

Fig. 1. A , vitrinite-like organic matter in association with calcite (reflected light), x 200. B, pyrite filling foraminiferal chamber (reflected light), x 200. C, inferred cyanobacterial remains (TEM), x 30 000. D, as for C, showing joint segmentation, cf. Oscillatoria. E, calcareous replacements o f inferred cyanobacteria resembling Oscillatoria (SEM), x 11 000. F, vitrinite-like organic matter in T E M ; loosely packed, low electron-dense c o m p o n e n t and dispersed humic acids (globular structures associated with electron-dense particles), x 8300. G, as for E, showing greater detail (SEM), x 15 0 0 0 (micrograph courtesy V. Scheibnerova).

276

M. G L I K S O N & G. H . TAYLOR

C B

E D

CYANOBACTERIAL MATS IN OIL SHALES 277 for C and H determinations, while 0,N and S were analysed On occasions calcareous outlines or replicas of micro on a Carlo Erba model 1106 elemental analyser. organisms closely resembling cyanobacteria of Oscillatoria type were observed in SEM examination of whole rock samples (Fig. IE, G). This mode of preservation has been RESULTS AND DISCUSSION. encountered and discussed in studies of the cyanobacterial Vitrinite-like organic matter; light microscopy and communities of Solar Lake, Red Sea, both in their habitat and in the laboratory (Krumbein et al., 1977). The rarity of TEM type of preservation is explained in terms of very high TEM observations of ultra-thin sections of organic matter this rates of precipitation of the calcite (Berner, 1968). In most concentrates revealed, for the most part, organic matter of cases Ca salts of fatty acids are formed prior to calcite, and low electron density (i.e. light grey appearance), where transformation one to the other, involves a considerable occasionally filamentous structure could be distinguished density increasefromcausing the destruction of external (Figs 1C, D; 2B), particularly when enhanced by metal staining morphology. The close association of vitrinite-like matter and (Fig. 3A, B). Light microscopy combined with TEM calcite is a significant feature of Toolebuc Formation oil shales. (Glikson, 1984 a, b) has shown that algae (chlorophytes), The irregularity of the calcareous lenses and laminations in chrisophytes) in sediments are represented by highly terms of thickness and lateral extension cannot be explained fluorescing organic matter in UV light (blue filter excitation), as a result of seasonal changes and are more probably due i.e. Botryococcus, Pediastrum and dinoflagellates. On the to localised micro-environmental changes in the cyanobacterial other hand, bacterial remains (including cyanobacteria), their mats. This conclusion is supported by the isotopic degradation products and metabolites do not exhibit compositions of organic matter-associated calcite as compared fluorescing properties. The vitrinite-like organic matter to the isotopic composition of benthic shells (Table 2). predominant in Toolebuc Formation oil shales does not fluoresce. When viewed as dispersed matter in transmitted light it appears amorphous, which is characteristic of degraded T A B L E 2. Carbon-isotope compositions of diagenetic calcite and organic matter. The association of the low reflectance vitrinitemollusc shells (Aucellina, Inoceramus), Toolebuc like matter with microbial remains and degradation products Formation. was first suggested by Taylor (1966 in Stach et al., 1975, 5 C of shells 5C of calcite pp. 193,194). Framboids of pyrite within the organic matter Sample (ppt) relative (ppt) relative are common (Fig. IB). Pyrite also appears as specks Identification to PDB to PDB disseminated throughout the organic matter in association B3-5 + 0.20 -0.60 with humic acids. The humic acid content is small and usually B3-6 + 1.25 -0.21 less than 1%. Following alkali separation from remaining Unspecified -2.20 -6.20 organic matter, embedding, ultra-thin sectioning and Unspecified + 0.41 -6.50 observing in TEM, the humates display the form of very small Unspecified -0.10 -6.00 spheres or globules, usually in aggregates (Fig. IF; Fig. 2A). B3-6 + 1.80 -0.20 J11—17 -2.40 These structures are seen to be irregularly dispersed in small L6-8 + 0.10 concentrations throughout the rest of the organic matter (Fig. L6-8 -0.60 IF). Most Toolebuc Formation samples studied in TEM, with J11-24 -5.20 the exception of sample 6-3, have revealed an irregular mode L6-1 + 1.17 of humic acid distribution, in small clusters throughout the 1-2 -0.59 rest of the vitrinite-like organic matter. However sample 6-3 L6-1 + 1.19 -0.88 (Fig. 2A) shows a higher concentration of humic matter and B3-6 + 0.39 filling of vacant spaces within the remaining organic matrix, L6-1 + 0.84 -1.66 rather than irregular distribution. Electron dense particles are L6-1 -1.01 1-2 -0.53 occasionally associated with the humic fraction or are B3-7 + 0.27 scattered throughout the rest of the organic matrix. These, L6-3 + 0.26 when analysed by electron probe proved to contain pyrite. L6-2 +0.28 Another metal detected is copper in electron dense coccoid forms (Fig. 2D). The predominant light grey (low electron density) organic matter displays a loosely packed, almost sponge-like appearance. Filamentous organisms were Original filaments of inferred cyanobacteria as described recognised in several instances (Fig. 1C, D; Fig. 2B). Single above are not uncommon in the organic matter concentrates. filaments are rarely distinguishable, since the organisms mostly However, detailed structure of the filaments is only rarely merge into a mass of loosely compacted matter. Most visible (Fig. ID). Filamentous structure may be enhanced by filaments do not exhibit any internal structure, although metal staining (Fig. 3A, B); it has been shown that N-fixing occasionally cross walls can be recognised (Fig. ID). The cyanobacteria possess specific glycolipids, members of which filaments are non-branching and show a straight or curved may contain numerous hydroxyl groups, which have been character. Individual filaments are about 0.25 /tm wide and shown to cause hydrogen bonding of lead when subject to 1-2 ftm long. On one occasion a bundle of smaller filaments lead organic stains (Nichols Wood, 1968; Lewis & Knight, was observed (Fig. 2B) possibly belonging to another type 1977). Lead citrate stain used in& the present study is responsible of organism, the disintegration of which is clearly visible. for the coarse granular precipitate (Fig. 3B). Glycogen 13

Fig. 2. A, humic acids concentrated in cavities within predominant organic component (TEM), x 20 000. B, bacterial remains as part of main component of vitrinite-like organic matter (TEM), x 33 000. C, microbial remains resembling Flectobacillus marinus (TEM), x 33 000. D, E, microbial remains resembling Caulobacter sp. In D, cf. Caulobacter occurs in association with coccoid bacteria (TEM), x 33 000.

278

M. GLIKSON & G. H. TAYLOR

Fig. 3. A, B, stain-enhanced (lead citrate) filamentous structures within predominant component of vitrinite-like organic matter (TEM), X 40 000. C\ D, coccolith remains (SEM), x 4400.

CYANOBACTERIAL MATS IN OIL SHALES granules, which are found in cyanobacteria as well as in other microbial organisms, have a special affinity for lead (Wolk, 1973). Among the filamentous structures, two main types can be recognised, namely individual relatively large and thick filaments (0.5 jim wide, up to 5 n m long) and smaller filaments (less than 0.2 /an wide, less than 2/xm long) usually appearing in bundles. The former type are compressed into a more or less homogeneous substance where borderlines of filaments are no longer discernible without staining. It is therefore likely that a significant part of the vitrinite-like organic matter, where no structure is visible in TEM, has been derived from compressed cyanobacteria and possibly sulphate-reducing bacteria. Compression or homogenisation of individual filaments may occur due to both dewatering and probably also the breakdown of a major part of cell walls. It has been noted that cyanobacterial cell walls consist of a high proportion of carbohydrates and amino acids (Wolk, 1973). The latter two compounds may react in a melanoidin-type reaction as has been demonstrated (Hedges, 1978) to form humic acids. Humic acids irregularly dispersed throughout the remaining organic matter in Toolebuc Formation oil shales, suggest a local source for their formation. The in situ formation of the humic acid/alkali-soluble fraction in these samples is supported by C-isotope studies (Table 1). The second type of filamentous structures, somewhat smaller than the first, is characterised by their appearance in bundles. These may represent the remains of sulphatereducing bacteria. The abundance of pyrite, both as framboidal and massive forms, throughout the organiccalcareous complex and within the foraminiferal shells indicates anoxic conditions at the water-sediment interface, and possibly above it. The anaerobic environment would have enhanced growth of the cyanobacterial community, as well as preservation of the organic matter following death of the organisms. Some cyanobacteria are able to carry out anoxygenic photosynthesis using only photosystem I (Cohen et at., 1975). This type of photosynthesis requires anaerobic conditions as well as the presence of a reducing component such as H 2 or H 2 S (Brock, 1979). Another significant aspect of this phenomenon is that the process occurs only when light intensities are low and there is insufficient energy to activate photosystem II (definition; Brock, 1979). At higher light intensities even under anaerobic conditions photosystem II would be activated and H 2 0 dissociated, leading to 0 2 production. Accumulations of oxygen have been proved to be toxic to these organisms (Pardue et al., 1976). The high concentrations of H 2 S inhibit photosystem II and hence close associations of sulphate-reducing bacteria and benthic cyanobacteria are common. The low intensity light requirements enabled the cyanobacterial mats to photosynthesise in spite of a rich planktonic layer blocking out most of the light. The ability of Oscillatoria-type organisms to thrive under the above conditions eliminates competition from other benthic bacterial forms and enables the one adapted organism to form thick accumulations of organic matter. It has been shown that in recent cyanobacterial mats sulphate reduction and calcite precipitation are stoichiometrically related (Jorgensen & Cohen 1977; Krumbein et al. 1977). Precipitation is related to an increase in pH or a shift in the carbonate-bicarbonate equilibrium as a result of photosynthesis. Calcite in the anoxic zone is further induced by various anaerobic bacteria (Krumbein et al., 1977). It was observed that calcite precipitation occurred upon the death of the organisms as a result of excretion of amino acids by decomposing cyanobacteria (Trichet, 1968). It has been

279

demonstrated that bacterial decomposition of amino acids under anaerobic conditions in sediments leads to an accumulation of ammonia in interstitial waters, raising the pH and bringing about the precipitation of Ca + + . Therefore, sulphate reduction and ammonia formation are the two major processes responsible for C a + + precipitation (Bernert, 1968, 1971). Difficulties in assigning fossil organisms to living taxa are well known in palaeontology or palaeobotany. Identification has to be based principally on the observation of morphological features which, depending on modes of preservation, are often not clearly outlined. The filamentous micro organisms encountered as calcareous replicas as well as organic forms resemble Oscillatoria type cyanobacteria and sulphate-reducing filamentous microbial organisms. The main difference between the fossil forms encountered in the Cretaceous oil shales and their living counterparts is in their size. Modern cyanobacteria can be resolved in optical microscopy, whereas the Cretaceous forms are only recognised by electron microscopy. The question arises, why are the (possibly) cyanobacterial remains in the Lower Cretaceous Toolebuc Formation diminished in size? Preparation techniques cannot be regarded as responsible since the smaller sizes are apparent from SEM observations of untreated specimens as well as in the demineralised organic remains. The question remains whether the decrease in dimensions is pre- or post-depositional. Considering the first possibility, it has been known from studies of phytoplankton (Malone et al., 1975) that populations decrease in size with progressive nitrate depletion. The same outcome has been reported from experiments conducted with cyanobacteria (Pardue et al., 1976), where it became evident that as the population increased, the N-sources decreased and the size of individual organisms diminished. The other explanation may be in postdepositional changes such as dewatering and compaction (J. Bauld, pers. comm., 1983) affecting the organic remains. However, whereas this process may explain the small size of the organic remains, it does not account for the small size of the calcareous replicas which form rapidly and are not generally subject to secondary size reduction upon shallow burial. Other organisms. Other microbial remains are rare. However, in sample 1-3 (Tambo 38, Table 1), clumps of a stalked organism (Fig. 2D, E) resembling Caulobacter were encountered. This organism forms rosettes which in turn aggregate into clumps of up to 2 /Am across (Fig. 2E). The stalked organism appears in association with a highly electron dense coccoid form closely resembling microbial remains encountered in the organic matter of the Cretaceous Messel oil shales (Degens & Ittekkot, 1982). The electron density was attributed to metal bonding by the cell wall. Electron probe analysis of the coccoid forms in the Toolebuc Formation displayed mainly copper peaks. An organism resembling the living form Flectobacillus marinus was observed occasionally. The micro organism appears in chains or clusters, individual forms being of a doughnut shape with one convex side, as is characteristic of the living form (Fig. 2C). The Flectobacillus-like organism and the electron dense coccoid forms may have been part of the aerobic microbial assemblage responsible for degradation of primary organic matter while in the oxygenated water column. It is noteworthy that the above mentioned microbial remains were only observed in isolated cases, and do not compose a significant portion of the organic matter. The latter is dominated by filamentous or loosely compacted cryptogranular organic matter of low

280

M. GLIKSON & G. H. TAYLOR

electron density. Caulobacter has been described by Sieburth (1979) and is known to attach itself to algae and other planktonic organisms. Certain strains of living Caulobacter are known to be iron depositing (Sieburth, 1979). The living Flectobacillus marinus is found today on various marine algae and has also been isolated from decomposing mangrove leaves in the Florida Keys (Sieburth, 1979). The organic parts of coccoliths may also have contributed to the vitrinite-like organic matter as has been previously suggested (Ramsden et al., 1982). It is evident from various studies that there are two forms in the coccolith life cyclenamely a bare-cell form and a coccolith-clad form (Sikes & Wilbur, 1982; Eppley et al1967). The bare-cell form is buoyant, and would therefore comprise part of the phytoplankton which would have been degraded in the oxygenated water zone, and mostly consumed by aerobic bacteria. However, the coccolith-clad form is not buoyant and, as is suggested by their numerous remains, organic parts of coccoliths may have contributed to the organic matter in these deposits.

Carbon-isotope composition of organic matter No previous data are available on the carbon-isotope composition of the Toolebuc Formation. In the present study, isotope analyses have been carried out mainly on organic matter as well as on selected calcareous matter, shown in Tables 1 and 2 respectively. Where possible, organic matter has been fractionated into humic acids and analysed separately from remaining material in order to determine the source of the humic acids. The analysed carbonates were taken from calcite associated with the organic matter as well as from benthic shells. Depletion of 5 ,3 C from oil shale bearing horizons ranges between - 2 7 ppt to - 2 9 ppt, representing the most ,3C depleted samples in the Toolebuc Formation. In interpreting these results the following factors need to be considered. Low $13C values may indicate concentration of the more degradation-resistant lipidic matter. It has been noted after examining particulate organic matter from different water depths in the Gulf of Mexico that all fractions, including the kerogen-like residue, became increasingly depleted in l3C with depth due to bacterial reworking (Eadie et al., 1978). The same study has shown an increase in overall lipid content with depth, explained as a result of concentration of more stable organic compounds. In carrying out isotope analysis of different fractions of several plants and algae, it has been found that lipids were depleted in 13C by about 4 ppt relative to the total plant (Volger & Hayes, 1979). It has been demonstrated that phytol is relatively enriched in 12C, (Bogacheva et al., 1979) a finding with possible bearings as to the isotope composition of isoprenoids whose precursor is believed to be phytol. A correlation may be expected between overall content of isoprenoids and the carbon isotope value of organic matter in sediments. Pronounced peaks of pristane and phytane as observed from GC-MS analysis of the Toolebuc Formation oil shale organic matter (R. P. Philp, CSIRO, pers. comm., 1983) may be a factor in obtaining isotopically lighter organic matter in those samples. Another important factor influencing the high ,3C depletion in these samples may be the amount of C0 2 available in the immediate environment of cyanobacterial mats. It has been demonstrated that an increase in C0 2 supply to cultures of cyanobacteria was followed by increase in fractionation coinciding with maximum metabolism (Pardue et al.y 1976). T. Sharkey (pers. comm., 1982) obtained values of 5 , 3 c

around - 30 ppt and lower for cyanobacteria cultured in C0 2 -enriched atmosphere. In view of the above it seems reasonable to assume that a high C0 2 concentration in the mats was an important factor influencing low carbon isotope values in the analysed samples. Recent cyanobacterial mats are considerably heavier isotopically than Toolebuc Formation samples. 613C values of - 1 3 to -17 ppt were reported for cultures of cyanobacteria (Calder & Parker, 1973). Carbon isotope compositions of recent cyanobacterial mats from a lagoon in Baffin Bay, Texas, were -14.4 to -16.7 ppt (Behrens & Frishman, 1971). However, with onset of degradation, depletion of the organic matter in respect to 13C should increase. It has been noted (Kaplan & Rittenberg, 1964) that the common sulphatereducing bacterium Desulfovibrio desulfuricans preferentially metabolises I2C, an important observation which would account for further depletion in 13C of the cyanobacterial mats* The latter are always associated with sulphate-reducing bacteria. The dark, anaerobic metabolism of certain matforming cyanobacteria, such as Oscillatoria may bring about further depletion in 13C (Degens, 1969). For samples where the humic acid fraction has been analysed separately from remaining organic matter, similar results or slightly 'heavier' organic matter values have been determined, with the exception of sample 6-3 (Table 1). Values obtained for the Toolebuc humic acid fraction mostly indicate an autochthonous source or in situ formation, probably from amino acids and carbohydrates released by decomposition of cyanobacterial mats and possibly other benthic organisms to form marine humates; such reactions have been suggested as a possible origin for marine humic acids (Hedges, 1978). Extensive studies conducted on recent sediments showed 613C of marine humic substances to be slightly heavier than the rest of the organic matter but in close correspondence with the isotope composition of plankton in the environment studied (Nissenbaum & Kaplan, 1972; Brown et al., 1972). Sample 6-3 revealed humic acid fraction values 'lighter' than for the rest of the organic matter by about 3 ppt. It is suggested here that the humic acids in this sample are allochthonous and haye.been derived from a different, possibly terrigenous, source. TEM observations of organic matter from sample 6-3 reveafed a different organisation of the organic components. The humic acid fraction in this sample is considerably larger than in the other oil shale samples, and is not dispersed irregularly throughout the rest of the organic matter, but is confined to pockets. The latter occurrence is characteristic of oil shale samples whose humic acid fraction yielded different isotopic values from the remaining organic matter (Glikson, 1984a). Sample Au6-40 from Augathella 6 well yielded values closer to those of terrigenous source material^ rather than organic matter from open marine domains. 613C of Julia Creek organic matter is in agreement with values obtained for oxidised matter (Irwin et al., 1977). The Augathella 6 sample had different isotope compositions from the oil shale-bearing sequence, namely -25.1 ppt, which is close to values also obtained from Birrigoolpa. These values mark an increase in terrigenous material towards the south of the Eromanga Basin. Augathella is situated south of the oil shale-bearing deposits and within the regime which received input from rivers in the southern part of the Eromanga Basin. Sample GS-8 from Birrigoolpa consists of a coaly band within an arenaceous sequence. The isotope value of -25.7 ppt is the * Bacterial metabolites are incorporated in the organic matter preserved.

281 CYANOBACTERIAL MATS IN OIL SHALES common value obtained for coals (Rigby et al. 1981). In other between the isotopic values of the two carbonates is sufficient localities in New South Wales where samples of equivalent to postulate two carbon sources (Table 2). age to the Toolebuc Formation oil shale sequence have been The carbon isotope composition of the CaC0 of analysed the values obtained were in the range of -22.0 to Inoceramus is within the range obtained for hypersaline -23.6 ppt. These samples proved to have a significant environments (Degens & Epstein, 1964). The 'Toolebuc sea' component of phytoplankton, which may account for the was not a closed basin in the true sense, rather it was partly isotope values. Studies of organic carbon-isotope divided from the open sea in the north by the St Elmo rise compositions of recent continental-derived clastic sediments which may have been the reason for the heavier, more saline in the Gulf of Mexico had S^C values of -21.7 ppt to bottom waters, making it a stratified sea. -22.6 ppt for planktonic organisms (Sackett & Thomson, 1963). y

Elemental composition of organic matter in the Toolebuc Formation

Carbon-isotope composition of carbonates

Carbon content of the Toolebuc Formation oil shale facies is between 10 and 63 per cent in the organic matter concentrate (Fig. 4) and hydrogen varies between 2 and 6.5 per cent. The carbon and hydrogen content decrease with phasing out of the oil shale facies towards the south of the Eromanga Basin, i.e. the equivalents of the Toolebuc Formation in New South Wales (see Ozimic, this volume). Some of the southern (GS) samples are high in carbon content but are low in hydrogen. The organic matter is mostly composed of dinoflagellates and land plant remains. Fluviatile influence in that part of the basin was more pronounced than marine influence. Some GS samples plot on a Van Krevelen diagram in the kerogen Type III zone. The oil shales proper fall within Type II (Fig. 5). However, these oil shales are not homogeneous, as can be seen from Fig. 5 and from fluctuations in carbon and hydrogen contents within vertical sections. These variations represent environmental changes associated with changes in the biota and fluctuations in the extent of the oxic/anoxic zone, rather than from the burial effect with depth. It is clear from previous studies as well as the present study that biodegraded organic matter and cyanobacterial remains tend to produce lower H/C ratios than algal accumulations dominated by Botryococcus (torbanite) or Pediastrum (Rundle). Cyanobacteria are

Carbon-isotope values of analysed benthic shells differ significantly from values obtained from calcareous matter interbedded with the organic matter. These findings point to two different sources of carbon, namely: (1) carbon from C0 released by cyanobacteria and already depleted in C which provided carbon for the CaC0 precipitating in the mat; and (2) bicarbonate of sea water from which the Inoceramus shells derived their carbon. The latter is therefore 'heavier* than the carbon incorporated in the CaC0 precipitate. Evidence for diagenetic carbonates forming from organic-rich sediments shows that C0 released from decomposing organic matter dissolves readily in pore water to be incorporated in CaC0 during precipitation, 'lighter' C0 in these carbonates would be anticipated. This is confirmed by analysis of coccolith-rich carbonates and diagenetic calcite nodules (Irwin et al., 1977). The values obtained by their studies showed a slightly more pronounced difference between the two carbonates than results obtained from the Toolebuc Formation samples. The reason for the less pronounced differences between shells and diagenetic calcite obtained in the prfesent study lies mainly in the difficulty of separating the CaCOj precipitate from coccolith shells and planktonic foraminifera. However, the difference 13

10 Y = 113+ 0 - 0 9 X r = 0 - 9 9

t I

C (wt%)

Fig. 4. Plot of hydrogen v5 organic carbon for samples from Toolebuc Formation. Refer Fig. 5 for sample key.

M. GLIKSON & G. H. TAYLOR

282 2 5 r

D o 1-5

£o

D 0 1-0 \ 1

T42

•

Au6

•

0-2

0-3

J0-4

J0-5

-L 0-6

J0-7

O / C ( atomic ratio )

0-8

J11

•

0-5

T38

0-9

L6 Au5

Fig. 5. Van Krevelen diagram of H/C vs O/C atomic ratios for samples from Toolebuc Formation.

relatively low in lipid content (4-9% dry weight) and high in carbohydrates and proteins (Wolk, 1973) in contrast to green algae such as Botryococcus with an oil content of 30-40 per cent dry weight (Wake & Hillen, 1981). Nitrogen in organic matter originates from proteinaceous compounds in organisms, which are decomposed into amino acids, some of which are resistant to biodegradation and have been recovered from sediments as old as Precambrian. However, in modern marine sediments humic acids are believed to be forming in situ by melanoidin-type reactions of amino acids and carbohydrates (Hedges, 1978). High nitrogen concentrations have been obtained in the humic acid fraction of organic matter derived from cyanobacterial mats (Stuermer et al., 1978). Nitrogen-fixing cyanobacteria may be responsible for the relatively high nitrogen content in Toolebuc Formation oil shales, in view of their very low humic acid content. It has been shown that carbonate-poor kerogens have low N/C ratios as do algal kerogens (Saxby, 1976), whereas carbonate-rich cyanobacterial remains are high in nitrogen. Nitrogen concentrations in the Toolebuc Formation show a close association with organic,carbon content (Fig. 6). Sulphur content in the Toolebuc Formation oil shale facies varies from 1 per cent (Jericho 11) to 23 per cent (Boulia 3) and in the non-oil shale-bearing sequence (samples GS) from 0.6 to 7.2 per cent. Most of the sulphur is contained within pyrite. The amount of H S and consequently pyrite present in sediments results from the amount of available sulphate, which in sea water is unrestricted, and of metabolisable organic matter for bacterial sulphate reduction. Sulphatereducing bacteria derive energy from oxidation of organic carbon anaerobically and a direct relationship has been found between sulphate reduction and concentrations of metabolisable organic matter (Lew, 1981), expressed as the atomic ratio of C/N to total sulphur. A similar relationship 2

can be seen between C/N and S in Toolebuc Formation samples (Fig. 7). Samples of low S and high C/N may reflect fluviatile influence (samples GS) in the southern part of the Eromanga Basin, and insufficient sulphate for sulphate reduction. Oxygen fluctuates between 5.6 and 16 per cent in organic matter concentrates of samples from the Toolebuc Formation oil shale facies. Higher values are probably associated with higher humic acid content. In samples where the humic acid content was sufficiently high to be separated from remaining organic matter, higher oxygen contents of 13 per cent and more were determined. Samples with a high humic acid content such as sample 6-3, some samples from Augathella, Tambo 42 and GS, also indicate incorporated terrigenous organic matter. GS samples from the non-oil shale facies in the southern Eromanga Basin generally contain much higher oxygen levels than oil shale-bearing samples, up to 23 per cent. Incorporation in the organic matter of carboxyl groups and aromatic compounds may be responsible for the high oxygen content in these samples. Whereas hydrogen is directly proportional to carbon in concentration (Fig. 4) the oxygen content is not obviously correlated with carbon content. However, organic matter with a high oxygen content is relatively lower in hydrogen, which no doubt reflects the type of organic matter and especially the oxidation potential of the depositional environment.

DEPOSITIONAL ENVIRONMENT

The wealth of planktonic organisms, such as coccoliths and foraminifera, whose calcareous remains were incorporated into the mat of benthic organisms on the sea floor (Fig. IB), indicate an oxygenated top water layer with access to the open marine environment. The great abundance of coccoliths (Fig. 3C, D) indicates optimal conditions for those organisms.

CYANOBACTERIAL MATS IN OIL SHALES

283

2-5 r-

Y = 0-08+ 0 - 0 3 X

• • •

r= 0 - 8 5

• •

' •

•

C (wt%) Fig. 6. Plot of nitrogen vs organic carbon for samples from Toolebuc Formation. Refer Fig. 5 for sample key.

60 o o L.

E £ o

40 0

\ o

• A

A "

0 O

S%wt Fig. 7. Plot of C / N V5 sulphur content for samples from Toolebuc Formation. Refer Fig. 5 for sample key.

284

M. GLIKSON & G. H. TAYLOR

Optimum conditions for the planktonic coccolith forms are salinities exceeding 34 ppt (Hulbert & Rodman, 1963). Optimum temperatures for the growth of a common marine coccolith, Coccolithus huxleyii is in the 18 to 24 °C range, salinity of 35 ppt and pH around 8. Abnormalities in form were found to occur at temperatures below 18 and above 25 °C (Watabe & Wilbur, 1966). Furthermore, it was demonstrated that the above mentioned salinities were optimal in maximum coccolith calcification or shell formation (Sikes & Wilbur, 1982). The vast accumulation of calcified coccoliths in Toolebuc Formation oil shales points to optimal conditions for shell formation. Calcification may also be in response to higher pH, as tolerance of calcified organisms to increases in pH is greater than for uncalcified forms. No direct evidence is available regarding the temperatures and salinities of the bottom waters. Only indirect assumptions may be made from the-existence of cyanobacterial mats, which point to relatively warm waters such as exist today in Spencer Gulf (South Australia), Shark Bay (Western Australia), Laguna Guerrero Negro (Mexico) and Solar Lake (Red Sea). The precipitation of carbonate within the cyanobacterial mat indicates high pH (above 7). Salinities associated with such mats at present are higher than those of average sea water, i.e. hypersaline in Solar Lake (Krumbein et al., 1977), 50 ppt in Shark Bay and 40-45 ppt in Spencer Gulf (Burne et al., 1980). The water temperature reported for Spencer Gulf is between 12 and 25 °C and for Shark Bay 18 to 28 °C. The temperatures assumed for the surface waters of the Toolebuc sea, between 18 and 24 °C, are in the range of today's habitat of cyanobacterial mats. Assuming salinities of about 35 ppt for surface waters of the Toolebuc sea', the denser bottom waters would be expected to have higher salinities, perhaps 40-50 ppt. The abundance of pyrite, throughout the organic-calcitic complex as well as within foraminiferal shells indicates anoxic conditions, probably at the water-sediment interface and possibly above it: anoxic conditions were necessary for growth and existence of the microbial association of cyanobacterial forms with sulphate reducers and ultimately made possible the accumulation of their remains. It has been shown experimentally that sulphide production is directly dependent on the concentration of both sulphate and dissolved organic carbon (Ramm & Bella, 1974). The latter is directly related to the organic matter available in the immediate environment. It has been shown (Bauld et al., 1980,1981), that cyanobacteria provide the energy for sulphate-reducing bacteria, the highest sulphate reduction rates coinciding with the metabolically active part of the cyanobacterial mat. The environment of deposition proposed for Toolebuc Formation oil shales is a marine basin with an anoxic zone close to or above the sediment-water interface, and an aerated upper water level accessible to the open sea. The latter conditions sustained a rich planktonic microflora and microfauna, whereas the former provided optimum conditions for the development of cyanobacterial mats and associated sulphate-reducing bacteria, preservation of organic matter and production of considerable quantities of pyrite. Undisturbed quiet waters are postulated for deposition of the oil shalebearing sediments. This model is supported by the presence of unbroken shells of Inoceramus and Aucellina preserved and embedded in the organic and mineral matter. Fine clay is the principal detrital component, with the exception of sample 6-3 which displayed quartz in XRD. The unusually large concentrations of coccoliths point to specific environmental conditions in the upper water zone. It has been demonstrated that blooms of organisms such as coccoliths

indicate the rise of oxygen-depleted waters (Ryan & Cita, 1977). The above observations support an interpretation of the depositional environment of Toolebuc Formation oil shales in terms of a stagnant basin whereby accumulation and preservation of these organic-rich deposits were possible.

SUMMARY AND CONCLUSIONS

The predominant mineral in the Toolebuc Formation oil shales is calcite in close association with organic matter, in the form of shells such as the benthic pelecypod Inoceramus and shells of planktonic organisms such as foraminifera and coccoliths. The organic matter is of a vitrinite-like, low reflectance nature when observed in reflected light microscopy, and of an amorphous appearance in transmitted light microscopy. TEM observations of the organic matter allow three types to be distinguished: (1) predominantly low electrondense, loosely packed matter of a pronounced granularity, occasionally resolved into the second type; (2) filamentous organisms. The dissolution and/or compaction of these can commonly be traced or enhanced by metal staining. Internal structure of the filamentous micro organisms is seldom observed in the organic remains, possibly due to cell wall disintegration, dewatering and subsequent shrinkage of the protoplast; (3) humic acid, or alkali-soluble fraction. Components (1) and (2) of the vitrinite-like organic matter were compared and found identical to cyanobacterial mat material subjected to laboratory simulated degradation and diagenesis*. True vitrinite is rare, and when observed displays reflectances characteristic of low-maturation organic matter (cf. Figs 8, 9). Carbon-isotope compositions of organic matter in the Toolebuc Formation oil shales are characterised by marked

Fig. 8. Vitrinite and fusinite reflectance (Ro) for samples from Toolebuc Formation oil shale facies. 1

Kindly supplied by B. Bubela, Baas-Becking Geobiological Laboratory (CSIRO-BMR), Canberra.

285 CYANOBACTERIAL MATS IN OIL SHALES phase had to be in the range of 18° to 24 °C. Postulated salinities were 35 ppt, and pH around 8. These conditions were optimal for the growth of coccoliths whose remains are found in abnormally high concentrations throughout the Toolebuc Formation oil shale facies. No direct evidence is available regarding the temperatures and salinities of the bottom waters. Oxygen-isotope analysis carried out on Inoceramus and Aucellina shells gave values which may not be representative of the original sea water. Only indirect assumptions may be made from the existence of cyanobacterial mats as to bottom water temperatures. These point to relatively warm waters such as exist today in Spencer Gulf (South Australia), Shark Bay (Western Australia), Laguna Guerrero Negro (Mexico) and Solar Lake (Red Sea). Undisturbed shells abundant in some horizons of the Toolebuc Formation indicate a quiet water environment. Stagnant conditions close to the sediment-water interface or above it are suggested by high pyrite and high organic contents.

zUJ I 10 oUJ if)

CO LL

o or UJ CO 1 5 D Z

ACKNOWLEDGEMENTS

We wish to thank M. J. Rickard and K. S. W. Campbell, Department of Geology, and George West, Research School of Biological Sciences, ANU for providing facilities for sample preparation and TEM studies. Thanks are due to J. Preston, Dept Forestry (ANU) for use of SEM. Elemental analysis was done by the Analytical Unit, JCSMR, ANU. We are indebted to J. W. Smith (CSIRO, North Ryde) for use of isotope laboratory facilities and wish to thank D. Rigby and G. Hart 05 1-0 for their assistance, and D. Goodchild (CSIRO, Plant Industry) for TEM—probe analysis. Ro % Samples were supplied by the BMR, CSIRO and the Fig. 9. Reflectance (Ro) of vitrinite-like organic matter for samples Geological Survey of NSW. Thanks are due to M. W. from Toolebuc Formation oil shale facies. Sandstrom (AIMS), D. M. McKirdy and M. Walter for reviewing the manuscript and to A. Y. Glikson (BMR) for depletion in C. Several possible explanations are advanced: major elements data processing. (1) if the lighter C 0 is the result of lipid matter concentration only, similar values would be expected from other Toolebuc samples. However, this is not the case, as only REFERENCES the oil shale samples yield low 5 C values; (2) cyanobacteria thriving in a carbon dioxide-rich environment may be responsible for the higher fractionation; and (3) dark, BAULD, J., 1980: Blue-green algal mats: activities of geobiological significance. Baas-Becking Geobiol. Lab. Ann. Rep. 1980, 9. anaerobic metabolism of the Oscillatoria-like cyanobacteria yields similar results. A likely factor influencing enrichment BAULD, J., 1981: Geobiological role of cyanobacterial mats in sedimentary environments: production and preservation of in C could be the activity of sulphate-reducing bacteria organic matter: BMR J. Aust. Geol. Geophys. 6, 307-17. known to preferentially metabolise C. Sulphate-reducing bacteria are a significant component of cyanobacterial mats. BEHRENS, E. W. & FRISHMAN, S. A., 1971: Stable carbon isotopes in blue-green algal mats. J. Geol. 79, 94-100. Organic matter from samples taken from the southern parts of the Eromanga Basin, where no oil shale deposits have BERNER, R. A., 1968: Calcium carbonate concretions formed by the decomposition of organic matter. Science, 59, 195-7. formed, yields isotopic values characteristic of micro-plankton and terrigenous material. The humic fraction of most BERNER, R. A., 1971: Principles of Chemical Sedimentology. McGraw-Hill, New York. Toolebuc oil shale organic matter is isotopically similar to, or slightly heavier than, remaining organic matter, which BOGACHEVA, M. P., KODINA, L. A. & GALIMOV, E. M., 1979: Intramolecular carbon isotope distributions in chlorophyll and points to an autochthonous origin, probably by decomposition its geochemical derivatives; in Douglas, A. G. & Maxwell, J. R. of cyanobacteria and benthic shells. It is believed that (eds) Advances in organic chemistry, 12, 679-87. planktonic organisms were degraded in the oxygenated water column whereby little if any organic remains would have BROCK, T. D., 1979: Biology and microorganisms. Prentice Hall Inc. New Jersey. reached the bottom waters. Isotopic compositions of CaC0 of benthic shells compared to CaC0 closely associated with BROWN, F. S., BAEDECKER, A., NISSENBAUM, A. & KAPLAN, I. R., 1972: Early diagenesis in Saanich Inlet, a reducing fiord. organic matter, point to two different carbon sources. The Geochim. Cosmochim. Acta, 36, 1185-203. organic matter-associated CaC0 derived its C 0 from decomposing cyanobacteria already depleted in C, whereas BURNE, R. V., COLWELL, J. B., PAIN, L. & TRATT, M. M., 1980: Sedimentological studies in the Spencer Gulf. Baas-Becking Inoceramus has derived its carbon from the bicarbonate of Geobiol. Lab. ann. Rep. 1980: 11-6. the sea-water reservoir. CALDER, J. A. & PARKER, P. L., 1973: Geochemical implications of induced changes in C fractionation by blue-green algae. As noted earlier in this study, the temperature of surface Geochim. Cosmochim. Acta, 37, 133-40. and near-surface waters during the Toolebuc depositional ,3

M. GLIKSON & G. H. TAYLOR

286

COHEN, Y., JORGENSEN, B. B., PADAN, E. & SHILO, M., 1975:

Sulphide-dependent anoxygenic photosynthesis in the cyanobacterium Oscillatoria limnetica. Nature, 257, 489-91. COOK, A. C., 1982: Organic fades in the Eromanga Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 234-55. COOK, A. C., HUTTON, A. C. T. & SHERWOOD, N. R., 1981:

Classification of oil shales: in Geology of coal, oil shales and kerogens. Centres Rech. Explor. Prodc. ElfAquitaine, Bull. 5,

269-721.

DEGENS, E. T., 1969: Biogeochemistry of stable carbon isotopes: in Eglinton, G. & Murphy, M. T. J., (eds) Organic geochemistry. Springer Verlag, Berlin. DEGENS, E. T. & EPSTEIN, S., 1964: Oxygen and carbon isotope ratios in coexisting calcites and dolomites. Geochim. Cosmochim. Acta, 28, 23-44. DEGENS, E. T. & ITTEKKOT, V., 1982: In situ metal-staining of biological membranes in sediments. Nature, 298, 262-4. EADIE, B. J., JEFFREY, L. M. & SACKETT, W. M., 1978: Some observations on the stable carbon isotope composition of dissolved and particulate organic carbon in the marine environment. Geochim. Cosmochim. Acta, 42, 1265-9. EPPLEY, R. W., HOLMES, R. W. & STRICKLAND, J. D. H., 1967:

Sinking rates of marine phytoplankton measured with a fluorometer. J. Expl. Marine Biol. Ecol. 1191-208. GLIKSON, M., 1984a: Transmission electron microscopy and Cisotope composition of oil shales, with particular reference to Australian deposits. Ph.D. thesis, A.N.U., Canberra (unpubl.). GLIKSON, M., 1984Z?: Further studies on torbanites and coorongite using transmission electron microscopy and C-isotope analysis. Org. Geochem. HEDGES, J. I., 1978: The formation and clay mineral reactions of malanoidins. Geochim. Cosmochim. Acta, 42, 69-76. HULBERT, E. M. & RODMAN, J., 1963: Distribution of phytoplankton

NICHOLS, B. W. & WOOD, J. B., 1968: New glycolipid specific to

nitrogen-fixing blue-green algae. Nature, 217, 767-8.

NISSENBAUM, A. & KAPLAN, I. R., 1972: Chemical and isotopic

evidence for the in situ origin of marine humic substances. Limnol. Oceanogr., 17, 570-80.

PARDUE, J. W., SCALAN, R. S., BAALEN, C. VAN 8C PARKER, P. L.,

1976: Maximum carbon isotope fractionation in photosynthesis, by blue-green algae and a green alga. Geochim. Cosmochim. Acta, 40, 309-12.

RAMM, A. E. & BELLA, D. A., 1974: Sulphide production in

anaerobic microcosms. Limnol. Oceanogr., 19, 110-18. RAMSDEN, A. R., 1983: Microscopic petrography of oil shales at Julia Creek, Northwestern Queensland. J. Geol. Soc. Aust. 30, 17-23. RIGBY, D., BATTS, B. D. & SMITH, J. W., 1981: The effect of maturation on the isotopic composition of fossil fuels. Org. Geochem. 3, 29-36. RYAN, W. B. F. & CITA, M. B., 1977: Ignorance concerning episodes of ocean-wide stagnation. Marine Geol. 23, 197-215. SACKETT, W. M. & THOMSON, R. R., 1963: Isotopic organic carbon composition of recent continental derived clastic sediments of eastern gulf coast, Gulf of Mexico, Am. Assoc. Pet. Geol. Bull. 47, 525-31.

SAXBY, J. D., 1976: Chemical separation and characterization of kerogen -from oil shale; in Yen, T. F. & Chiligarian, G. V. (eds) Oil shale. Devel. Petrol. Sci 5, 103-28. Elsevier, Amsterdam. SIEBURTH, J. McN. 1979: Sea microbes, Oxford University Press, U.K. SIKES, C. S. & WILBUR, K. M., 1982: Functions of coccoliths formation. Limnol. Oceanogr. 27, 18-26. SPIRO, B. & AIZENSHTAT, Z., 1977: Bacterial sulphate reduction and calcite precipitation in hypersaline deposition of bituminous shales. Nature, 269, 235-7.

species with respect to salinity between the coast of southern New England and Bermuda. Limnol. Oceanogr. 8, 263-9.

STACH, E., MACKOWSKY, M-th., TEICHMULLER, M., TAYLOR, G. H., CHANDRA, D. & TEICHMULLER R., 1975: Coal petrology.

1980: Organic matter in oil shales. APEA J. 20(1), 44-67. IRWIN, H., CURTIS, C., & COLEMAN, M., 1977: Isotopic evidence for source of diagenetic carbonates formed during burial of organicrich sediments. Nature, 269, 209-13. JORGENSEN, B. B. & COHEN, Y., 1977: Solar Lake (Sinai) 5. The sulphur cycle of the benthic cyanobacterial mats. Limnol. Oceanogr. 22, 657-66. KAPLAN, I. R., SMITH, J. W. & RUTH, E., 1970: Carbon and sulphur concentration and isotopic composition in Apollo 11 lunar samples. Proc. Apollo 11 Lunar Sci. Conf. 2, 1317-29. KAPLAN, I. R. & RITTENBERG, S. C., 1964: Microbiological fractionation of sulphur isotopes. J. Gen. Microbiol. 34, 195-212. KRUMBEIN, W. E., COHEN, Y & SHILO, M., 1977: Solar Lake (Sinai) 4. Stromatolitic cyanobacterial mats. Limnol. Oceanogr. 22, 635-56. LEW, M., 1981: The distribution of some major and trace elements in sediments of the Atlantic Ocean (DSDP Samples). 1. The distributions of sulphur, sulphur isotopes and Mn, Fe, Zn and Cu. Chem. Geol. 33, 205-24. LEWIS, P. R. & KNIGHT, D. P., 1977: Staining methods for sectioned material, in Glanert, A. M. (ed.) Practical Methods in Electron Microscopy, North Holland.

STUERMER, D. H., PETERS, K. E. & KAPLAN, I. R., 1978: Source

HUTTON, A. C., KANTSLER, A. J., COOK, A. C. & MCKIRDY, D. M.,

MALONE, T. C., GARSIDE, C., HAINES, K. C. & ROELS, O. A., 1975:

Nitrate uptake and growth of Chaetoceros sp. in large outdoor continuous cultures. Limnol. Oceanogr., 20, 9-19.

Gebruder Borntraeger, Berlin.

indicators of humic substances and proto-kerogen. Geochim. Cosmochim. Acta, 42, 989-98. TAYLOR, G. H., 1966: The electron microscopy of vitrinites. Advances in chemistry, Am. Chem. Soc. 55, 274-83. TRICHET, J., 1968: Etude de la composition de la fraction organique des oolites: comparation avec celles des membrances des bacteries et des cyanophycees. C. R. Hebd. Seanc. Acad. Sci. 267. VOLGER, E. A. & HAYES, J. M., 1980: Carbon isotopic compositions of carboxyl groups of biosynthesized fatty acids; in Douglas, A. G. & Maxwell, J. R. (eds) Adv. Org. Geochem. Pergamon Press, 697-704. WAKE, L. V. & HILLEN, L. W., 1981: Nature and hydro-

carbon content of blooms of the alga Botryococcus braumi occurring in Australian freshwater lakes. Aust. J. Marine, Freshwater Res. 32, 353-67.

WATABE, N. & WILBUR, K. M., 1966: Effects of temperature on

growth, calcification and coccolith form in Coccolithus huxleyii, (Coccolithineae). Limnol. Oceanogr. 11, 567-75. WOLK, C. P., 1973: Physiology and cytological chemistry of bluegreen algae. Bacter. Rev. 37, 32-101. YEH, C. S., 1973: Rapid microdetermination of carbon and hydrogen. A new universal combustion tube filling. Microchem. J. 7, 303-10.

Geological Society of Australia Special Publication No. 12, 287-304

Organic geochemical fades of the Cretaceous Bulldog Shale, western Eromanga Basin, South Australia D. M. McKirdy , J. K. Emmett , B. A. Mooney - , R. E. Cox & B. L. Watson 1

1 3

Australian Mineral Development Laboratories, Frewville, S.A. 5063. ESSO Australia Limited, Sydney, N.S.W. 2001. Present address, Ampol Research Laboratory, Wynnum, Qld 4178.

ABSTRACT

Recent discoveries of oil in the Murta Member (Mooga Formation) and Cadna-owie Formation, and of gas in the Coorikiana Sandstone, have focussed attention on the hydrocarbonsource potential of Early Cretaceous marine and lacustrine shales throughout the southern Eromanga Basin. One such unit is the Bulldog Shale, comprising dark grey shale and siltstone deposited during a major marine transgression. Detailed analysis of 24 core samples from SADME Toodla 1, a stratigraphic well located on the Muloorina Ridge near the western edge of the Eromanga Basin, has permitted the construction of an organic geochemical profile of the lower two-thirds (114 m) of the formation. The profile reveals significant stratigraphic variation in the composition of the geolipid and kerogen fractions of the dispersed organic matter superimposed on a down-hole trend of increasing total organic carbon (TOC = 0.8-2.0%). The causes of this variation were changes in sedimentation rate, in situ microbial activity, and the balance of autochthonous marine and allochthonous terrestrial organic inputs. At least three distinct organic geochemical facies can be recognised, each with characteristic Rock-Eval parameters, alkane pattern and kerogen composition. Kerogen changes from Type III (H/C = 0.75-1.1) in Facies A at the base of the formation, through Type III (H/C = 0.7-0.75) in Facies B, to Type IV (H/C = 0.6-0.7) in Facies C at the top of the section examined. Although exinite (mostly phytoplankton, sporinite and liptodetrinite) comprises 20-75 per cent of its organic matter, the Bulldog Shale is essentially gas-prone throughout. The reason for this surprising lack of oil-source potential is the oxidised state of the highly comminuted exinites. Vitrinite is absent, or only a minor component of the dispersed organic matter. The Bulldog Shale at Toodla 1 is thermally immature (VR ~ 0.37%). The organic geochemistry of a slightly more mature Bulldog Shale sequence in the Delhi Strzelecki 3 exploration well (VR = 0.48-0.55%) confirms the widespread lateral continuity of its predominantly gas-prone organic matter, but also provides a partial analogue for the origin of light, paraffinic Murta-type crude oils.

INTRODUCTION

Most of the established hydrocarbon reserves of the Eromanga Basin occur in Jurassic strata (Armstrong & Barr, 1982; Kantsler et al., 1983). However, the Cretaceous section also is highly prospective as evidenced by oil flows from the Murta Member (Mooga Formation) at Dullingari North 1 (450 barrels of oil per day-BOPD) and Jackson 1 (338 BOPD), and from the Cadna-owie Formation (Transition beds) at Tintaburra 1 (85 BOPD); and gas flows from the Coorikiana Sandstone in Strzelecki 8 (10 600 m per day) and Marabooka 3 (18 400 m per day). These discoveries focus attention on the source-rock potential of Early Cretaceous marine and lacustrine sediments throughout the southern Eromanga Basin (Fig. 1). The Bulldog Shale (Fig. 2), the subject of the present study, is one of several organic-rich Early Cretaceous units in the Eromanga Basin sequence that, given appropriate dispersed organic matter (DOM) and adequate thermal maturity, could be effective source rocks for petroleum hydrocarbons (Moore & Pitt, 1982; McKirdy, 1982; Kantsler et al., 1983). Although it has a significant algal/bacterial component, organic matter 3

preserved in the marine Bulldog Shale, the laterally equivalent Wallumbilla Formation, and in the underlying Cadna-owie Formation is predominantly of land-plant origin (Cook, 1982; McKirdy, 1982). Gas generation from such humic organic matter (Type III kerogen) commences at a vitrinite reflectance of VR ~ 0.6 per cent (Monnier et al., 1983), whereas the rank threshold for significant oil generation is VR ~ 0.7 per cent (Powell & Snowdon, 1980). The depth-reflectance studies of Kantsler et al. (1983, this volume) show that only within the central Nappamerri Trough and the Cooper Syncline (Cooper Basin region) do Early Cretaceous sediments reach the maturation levels required for oil generation from woody-herbaceous DOM. Elsewhere in the Eromanga Basin, less mature (VR = 0.45-0.7%) Cretaceous sediments may be effective sources for hydrocarbons, but only if they contain significant amounts of thermally labile exinite (viz. resinite, fluorinite, suberinite, bituminite) and/or lipid-enriched desmocollinite (bacteriallydegraded cellulosic plant tissue) (cf. Cook, 1982; Snowdon & Powell, 1982; Powell & Snowdon, 1983). Therefore, when evaluating the hydrocarbon-source potential of the mostly

D. M. McKIRDY ET AL.

288

immature to marginally mature Cretaceous marine (and lacustrine) sediments of the Eromanga Basin, it is essential to know the nature and distribution (vertical and lateral) of their respective organic facies. SADME Toodla 1, a stratigraphic well drilled near the southwestern margin of the Eromanga Basin, intersected 164.5 m of Bulldog Shale (Griffiths, 1980). This well was designed to test the stratigraphy and source-rock potential of the Mesozoic section across the Muloorina Ridge, an uplifted block of Proterozoic basement rock which separates two Permian infra-basins, the Pedirka Basin to the northeast and the Arckaringa Basin to the southwest (Fig. 1). Detailed analysis of 24 core samples from Toodla 1 has permitted the construction of an organic geocheipical profile of the lower 114 m (68 per cent) of Bulldog Shale (and the top of the underlying Cadna-owie Formation). The profile incorporates data from Rock-Eval pyrolysis; the analysis of solventextractable organic matter (EOM, nominally C 1 5 + ) by liquid chromatography, gas chromatography (GC), and gas chromatography-mass spectrometry (GC-MS); and elemental, isotopic, and pyrolysis-GC analyses of kerogen. The geochemical information was supplemented by organic petrological data obtained from examination of the DOM in polished sections of shale core using reflected light microscopy (white light and fluorescence mode), and of kerogen strewn mounts using transmitted light microscopy.

Late J

AGE

c/> 3 O LU O <

1—

LU <r o

i. cu LU

PROXIMAL SEQUENCE WINTON

FORMATION

MACKUNDA

FORMATION

OODNADATTA

FORMATION

COORIKIANA

SANDSTONE

BULLDOG

FORMATION MURTA

MEMBER

MOOGA

<cc

MACKUNDA

Early

—>

FORMATION

ALLARU

MUDSTONE

TOOLEBUC

FORMATION

WALLUMBILLA

^ ^

NAMUR

SANDSTONE

MEMBER

FORMATION MURTA

HOORAY

FORMATION

CADNA-OWIE

MEMBER

SANDSTONE

WESTBOURNE ADORI

Mid

o </)

FORMATION

SHALE

CADNA-OWIE

Q) CO —1

DISTAL SEQUENCE WINTON

FORMATION

SANDSTONE

BIRKHEAD FORMATION

BIRKHEAD

FORMATION

HUTTON

SANDSTONE

SANDSTONE basal J U R A S S I C '

Fig. 2. Generalised stratigraphy of the Eromanga Basin, South Australia (after Moore & Pitt, 1984).

BULLDOG SHALE ORGANIC GEOCHEMICAL FACIES 289 This profile of the Bulldog Shale reveals appreciable to account for the occurrence of fossiliferous Lower Devonian stratigraphic variation in the composition of the geolipid and quartzite boulders within conglomeratic sediments at the base kerogen fractions of its DOM, superimposed on a downhole of the Bulldog Shale along the southwestern margin of the trend of increasing total organic carbon (TOC). The nature Eromanga Basin. and likely causes of this variation are considered. The Bulldog Strzelecki 3 (Fig. 1) penetrated a thicker and deeper section Shale kerogen is of mixed algal-higher plant origin. Misleading of Bulldog Shale (1020.5-1285.5 m depth; total thickness = assessments of its oil versus gas-generating potential may result 265 m). The constituent shale and silty mudstone apparently if only optical methods of source-rock analysis are used. The were deposited in a similar proximal marine setting to that implications of this finding for routine source-rock studies which characterised the Toodla 1 locality during Aptian time, in the Eromanga Basin are discussed. Finally, the Toodla 1 although lack of core data makes interpretation of the profile is compared with the organic geochemistry of a more Strzelecki 3 sequence difficult. deeply buried section of Bulldog Shale in the Delhi Strzelecki 3 exploration well located some 450 km to the east in the Tennapera Trough of the Eromanga/Cooper Basin (Fig. 1). ANALYTICAL METHODS Rock samples used in this study were conventional core This comparison throws light on the origin of Murta-type paraffinic oils, one of two major oil families found in the (Toodla 1); and a combination of sidewall cores and canned Eromanga Basin (McKirdy, 1982, 1984c, b\ Kantsler et al., cuttings (Strzelecki 3). Total organic carbon (TOC) was determined by acid digestion in 5N HC1 followed by 1983). combustion and measurement of the resultant C 0 in a Leco IR-12 carbon analyser. Whole-rock samples (100 mg) were BULLDOG SHALE analysed by the Rock-Eval pyrolysis technique (Espitalie et The electrofacies, lithology, age, palaeontology, distribution al., 1977). The S parameter (mg C0 /g TOC) was measured and environment of deposition of the Bulldog Shale have been in a separate analytical run using a second, acid-treated aliquot discussed by Moore & Pitt (1982, 1984, 1985). The formation of each rock sample. is of Aptian to early Albian age and comprises grey shale and Powdered rock (~ 100 g; less in case of sidewall core siltstone, with minor thin interbeds of fine-grained sandstone, samples) was solvent-extracted with azeotropic that were laid down during a widespread marine transgression. benzene/methanol in Soxhlet apparatus for 24 hours. Removal The lower one-third of the unit is particularly dark and shaly. of the solvent gave the extractable organic matter (EOM, It rests conformably on non-marine to paralic siltstone and nominally C ) . Asphaltenes were precipitated from the sandstone of the Cadna-owie Formation, and is overlain by EOM with petroleum ether (IP method 143/57) and the the Coorikiana Sandstone (Fig. 2),' a regressive marine asphaltene-free fraction separated into saturated hydrocarbons shoreface deposit. (alkanes), aromatic hydrocarbons and ONS-bearing The Bulldog Shale is a shallow marine deposit which attains compounds (resins) by column chromatography on 80 parts a maximum thickness of about 320 metres in the Moomba activated silica gel over 20 parts activated alumina. Alkanes area (Moore & Pitt, 1982). Its lateral equivalent in northeastern were eluted with petroleum ether, aromatic hydrocarbons with South Australia and Queensland is the lithologically similar petroleum ether/benzene (85:15), and resins with Wallumbilla Formation (Fig. 2). A distinctive feature of both methanol/benzene (90:10). Subsequent analysis revealed that formations is a basal organic-rich shale unit up to 25 m thick, in some Toodla 1 samples monoaromatic tricyclic diterpenoid which is characterised on wireline logs by a high gamma-ray hydrocarbons had co-eluted with the alkanes. response and low sonic velocity. The petroleum ether eluate was analysed by capillary gas In Toodla 1, the Bulldog Shale occurs between 99.5 and chromatography (GC) using a Perkin Elmer Sigma 2B gas 264 m depth (total thickness = 164.5 m). Griffiths (1980) chromatograph fitted with a Grob-type spitless injector and divided the formation into two informal sub-units. The upper a 25 m x 0.2 mm i.d. vitreous silica column (QC2/BP1, SGE sub-unit (99.5-166.5 m) is a regularly interlaminated sequence Australia). A temperature programme of 60 to 275°C at 5°C of pale grey shale and argillaceous siltstone containing m i n , and injector and detector temperatures of 280°C, glauconite, pyrite and abundant well-preserved land plant were employed. Helium at a linear velocity of 32 cm sec (lignitic) material and shelly fossils. The lower sub-unit was the carrier gas. Normal alkanes ( C - C ) and the (166.5-264 m) comprises dark grey to black shale with minor major isoprenoid alkanes, pristane (C ) and phytane (C ), siltstone and sandstone interbeds, the latter increasing in were identified on the basis of their relative retention times. abundance towards the base of the unit. The shale is GC analysis of the Strzelecki 3 alkane fractions was considerably more bioturbated than that in the upper sub- undertaken with a 45 m x 0.5 mm i.d., OV-lOl coated, glass unit. Burrows are variously infilled by sand, glauconite and SCOT column. pyrite. The glauconite-pyrite association is indicative of The identity of pristane, phytane, dehydroabietane and alternating oxic-anoxic conditions (Berner, 1981). At Toodla dehydroabietin in the C hydrocarbons of several Toodla 1, and in nearby outcrop (Moore & Pitt, 1985), fossil wood 1 core extracts was established by comparison of their mass fragments, commonly impregnated with pyrite, and possible spectra and GC retention indices with literature data (Haug pyritised root casts, are notable features of the basal part of & Curry, 1974; Wakeham et al,1980; Tan & Heit, 1981). Gas the Bulldog Shale. This suggests close proximity to a non- chromatography-mass spectrometry (GC-MS) was performed marine sediment source, or the transport of terrigenous using a Hewlett Packard 5992B instrument and data system. detritus to deep water e.g. by turbidity currents. The Bulldog The vitreous silica capillary column (12 m x 0.2 mm i.d., Shale appears to have been deposited, probably below wave QC2/BP1, SGE Australia) was temperature-programmed from base, in a shallow marine environment. However, the presence 100 to 260°C at 5°C m i n . The carrier gas was helium. in the lower sub-unit of silt and sand lenses, some of which Kerogen was isolated essentially as described in Powell et al. display mesoscale tabular and trough cross bedding (Griffiths, (1975), except that no attempt was made to remove pyrite by 1980), suggests that this otherwise low-energy environment chemical means. The kerogen concentrate was analysed for was subject to intermittent strong current activity. Flint et al. carbon, hydrogen, nitrogen, sulphur, oxygen and ash by (1980) proposed episodic debris flows over low-angle slopes standard microanalytical techniques. Pyritic iron was 2

15+

-1

15+

-1

D. M. McKIRDY ET AL.

290

determined by atomic absorption spectrometry and appropriate corrections made for pyritic sulphur and the conversion of pyrite to haematite during ashing. Stable carbon isotope ratios (reported in the 8 C per mil notation relative to the PDB limestone standard) were measured by the Chemistry Department, Western Australian Institute of Technology. Pyrolysis-gas chromatography (PGC) of kerogens was performed using a Chemical Data Systems Pyroprobe 120 solids pyrolyser (incorporating an extended temperature programming facility), in tandem with a Perkin Elmer Sigma 3 gas chromatograph. The sample (0.5-2.0 mg) was pyrolysed under helium at 700°C for 1 minute and the pyrolysate swept directly onto the front end of the GC capillary column (25 m x 0.3 mm i.d. vitreous silica, QC3/BP1, SGE Australia) where it was trapped at -40°C. The column temperature programme was as follows: 10°C for 3 min; 10 to 275°C at 6°C min" ; isothermal at 275°C until all peaks eluted. Pyrolysate trapped by bubbling splitter effluent into a vial of methylene chloride was analysed separately by GC-MS (as above) in order to identify individual aromatic hydrocarbons and phenolic compounds. DOM was examined in situ in polished sections of core cut perpendicular to bedding using reflected light and fluorescence mode microscopy; and as kerogen concentrates mounted in glycerine jelly using transmitted light microscopy. The reflectance of vitrinite phytoclasts was measured with a Leitz MPV1.1 microphotometer fitted to a Leitz Ortholux microscope and calibrated against synthetic standards. Measurements were made in oil immersion (n = 1.518) with monochromatic light (wavelength 546 nm) at a temperature of 24 ± 1°C. Fluorescence observations employed a 3 mm BG3 excitation filter, a TK400 dichroic mirror and a K510 suppression filter. 13

TABLE

J 5 +

TOODLA 1, MULOORINA RIDGE Maturity Vitrinite reflectance (VR = 0.34-0.37%) and thermal alteration index (TAI = 2.1) measurements indicate that the Bulldog Shale at this well locality is too immature for catagenic release of hydrocarbons from its kerogen to have begun. Tmax values are mostly less than 430°C and therefore consistent with this lack of maturity. However, the most striking feature of the Tmax profile (Fig. 3) is its segmented character. Offsets or discontinuities in the profile at approximately 175, 190, 220 and 233 m depth correspond to subtle changes in kerogen type and mark the boundaries of the organic facies discussed below. The unusual trend of decreasing production index ( S J / S J + S ) with increasing depth (Fig. 3) is attributable to the combined effects of low maturity and an up-hole increase in the inertinitic character of the DOM (Table 6), i.e. S decreases while Sj remains essentially constant. Very low hydrocarbon yields ( ^ 6 mg/g TOC; <25% of EOM: Table 3), high pristane/Az-heptadecane ratios 2

1. Rock-Eval pyrolysis data, Bulldog Shale and Cadna-owie Formation (lowermost sample), SADME Toodla 1. TOC PC PI S /S S,+S Tmax s s, s

DEPTH (m)

151.7-151.8 157.4-157.6 160.0-160.5 165.1-165.9 170.3-170.9 179.0-179.3 183.1-183.4 186.5-186.7 194.5-194.8 195.5-195.8 203.0-203.3 209.0-209.3 214.7-215.0 217.3-217.6 222.8-223.0 225.2-225.5 230.6-230.9 236.1-236.4 241.0-241.4 246.5-246.8 254.6-255.0 258.8-259.4 260.1-260.4 265.3-266.0 KEY Tmax

S, s, s;

RESULTS TOC and Rock-Eval pyrolysis data for samples of Early Cretaceous sediments from Toodla 1 (core) and Strzelecki 3 (canned cuttings) are listed in Tables 1 and 2. Pertinent C extract yield and compositional data for these two wells are summarised in Tables 3 and 4. The elemental and carbon isotopic compositions of kerogens isolated from selected Toodla 1 core samples are given in Table 5. Tables 6 and 7 summarise the organic petrology and vitrinite reflectance (VR) of the in situ DOM in the Toodla 1 core and Strzelecki 3 sidewall cores. Table 8 is a compilation of thermal alteration index (TAI) data and of the particulate organic matter types present in the Toodla 1 kerogen concentrates.

s. + s,

0.03 0.03 0.03 0.03 0.02 0.05 0.02 0.04 0.04 0.03 0.02 0.03 0.03 0.03 0.03 0.02 0.03 0.04 0.03 0.03 0.02 0.03 0.02 0.07

425 429 429 429 435 429 430 430 427 425 425 425 423 424 429 429 428 431 429 429 429 436 433 436

0.15 0.15 0.14 0.13 0.11 0.28 0.23 0.18 0.27 0.38 0.18 0.39 0.36 0.26 0.31 0.30 0.29 0.44 0.40 0.44 0.42 0.92 0.79 2.50

0.69 0.73 0.75 0.74 0.65 0.86 0.77 0.88 0.93 0.81 0.70 0.79 0.83 0.84 1.19 1.16 1.33 1.39 1.31 1.30 1.39 1.58 1.67 1.70

= position of S peak in temperature program (°C) = kg free hydrocarbons/tonne rock = kg hydrocarbons (kerogen pyrolysate)/tonne rock = kg CO, (organic)/tonne rock = Potential Yield 2

0.22 0.21 0.19 0.18 0.17 0.33 0.30 0.20 0.29 0.47 0.26 0.49 0.43 0.31 0.26 0.26 0.22 0.32 0.31 0.34 0.30 0.58 0.47 1.47

0.18 0.18 0.17 0.16 0.13 0.33 0.25 0.22 0.31 0.41 0.20 0.42 0.39 0.29 0.34 0.32 0.32 0.48 0.43 0.47 0.44 0.95 0.81 2.57

PI PC TOC HI OI

0.17 0.17 0.18 0.19 0.15 0.15 0.08 0.18 0.13 0.07 0.10 0.07 0.08 0.10 0.09 0.06 0.09 0.08 0.07 0.06 0.05 0.03 0.02 0.03

0.85 0.82 0.91 0.94 0.77 1.08 1.06 0.96 0.97 1.10 0.71 1.00 1.10 0.90 1.20 1.12 1.09 1.39 1.46 1.44 1.29 1.85 1.63 1.99

0.01 0.01 0.01 0.01 0.01 0.03 0.02 0.02 0.03 0.03 0.02 0.03 0.03 0.02 0.03 0.03 0.03 0.04 0.04 0.04 0.04 0.08 0.07 0.21

Production Index ( S / S , +S ) Pyrolysable Carbon (wt. <7o) Total Organic Carbon (wt. %) Hydrogen Index (mg h'c, S / g TOC) Oxygen Index (mg C 0 , S / g TOC) 2

17 18 15 13 14 25 21 18 27 34 25 39 32 28 25 26 26 31 27 30 32 49 48 125

81 89 82 78 84 79 72 91 95 73 98 79 75 93 99 103 122 100 89 90 107 85 102 85

B U L L D O G S H A L E O R G A N I C G E O C H E M I C A L FACIES

291

TABLE 2. Rock-Eval pyrolysis data, Bulldog Shale, C a d n a - o w i e F o r m a t i o n (1310.6-1356.3 m) a n d M u r t a M e m b e r (1365.5-1411.2 m), Delhi Strzelecki 3. See Table 1 for key. D E P T H (m)

Tmax

s,+s2

S2/S3

TOC

1018.0-1027.1 1036.3-1045.4 1054.6-1063.7 1072.9-1082.0 1091.2-1100.3 1109.5-1118.6 1127.8-1136.9 1146.0-1155.1 1164.3-1173.4 1182.6-1191.7 1200.9-1210.0 1219.2-1228.3 1237.5-1246.6 1255.8-1264.9 1274.1-1283.2 1310.6-1319.7 1328.9-1338.0 1347.2-1356.3 1365.5-1374.6 1383.8-1392.9 1402.1-1411.2

440 442 442 440 442 442 437 447 434 438 438 437 437 441 439 440 438 441 444 442 441

0.03 0.04 0.06 0.04 0.03 0.03 0.03 0.01 0.05 0.04 0.03 0.05 0.07 0.07 0.07 0.09 0.08 0.15 0.42

0.61 0.77 1.10 1.08 0.75 0.84 0.41 0.38 0.43 0.53 0.34 0.37 0.71 0.67 0.67 0.59 0.50 0.67 2.21

0.85 0.95 1.42 1.80 1.29 0.88 0.56 0.86 1.48 0.89 0.41 0.50 0.61 0.29 0.71 0.44 0.34 0.42 0.47

0.64 0.81 1.16 1.12 0.78 0.87 0.44 0.39 0.48 0.57 0.37 0.42 0.78 0.74 0.74 0.68 0.58 0.82 2.63

1.93 1.52 3.00 1.32 2.16 1.30 1.02 1.04 1.15 1.41 0.90 0.86 1.33 1.23 2.36 1.06 0.86 0.83 1.88

0.83

0.12

0.78

0.79

0.65 0.90

0.05 0.07

0.78

31 50 36 81 34 64 40 36 37 37 37 43 53 54 28 55 58 80 117 89 100

44 62 47 136 59 67 54 82 128 63 45 58 45 23 30 41 39 50

0.52

0.05 0.05 0.05 0.04 0.04 0.03 0.07 0.03 0.10 0.07 0.08 0.12 0.09 0.09 0.09 0.13 0.14 0.18 0.16 0.20

0.05 0.06 0.09 0.09 0.06 0.07 0.03 0.03 0.04 0.04 0.03 0.03 0.06 0.06 0.06 0.05 0.04 0.06 0.22

0.13

0.71 0.81 0.77 0.60 0.58 0.95 0.73 0.44 0.29 0.59 0.82 0.74 1.16 2.31 0.94 1.34 1.47 1.59 4.70 0.62 0.98

(pr/«-C 17 >1: Table 3) and the strong predominance of odd over even-carbon-numbered tf-alkanes (Figs 14, 16) testify to the immaturity of the extractable geolipids. Organic geochemical fades When plotted against depth in the Toodla 1 well section, selected Rock-Eval (Figs 3-5), kerogen (Figs 6, 7) and C 1 5 + hydrocarbon (Fig. 8) parameters reveal systematic trends in the concentration and composition of the organic matter preserved in the Bulldog Shale. On the basis of this vertical variation, three discrete organic geochemical facies (and one sub-facies) have been identified. Each facies has a characteristic kerogen composition (Figs 9-12), organic petrology (Fig. 13) and C , 5 + hydrocarbon distribution (Figs 14, 16).

0.13

0.58

143 101

Facies A covers the depth range 233-264 m in the Bulldog Shale and extends downwards into the upper Cadna-owie Formation. Organic carbon content exceeds one per cent throughout (TOC = 1.29-1.99%), although potential hydrocarbon yields are generally low (Sj + S 2 = 0.43-0.95 kg/tonne) (Fig. 4). Only the Cadna-owie Formation sample displays moderate source potential for oil (Sj + S 2 = 2.6 kg/tonne). Likewise, pyrolysable carbon values are low (PC = 0.04-0.21%). In parallel with TOC, hydrogen index and kerogen atomic H / C values increase steadily with depth (HI = 27-125; H / C = 0.72-1.03) (Figs 5, 6). In terms of its elemental and carbon isotopic composition, this Type III kerogen is quite distinct from that preserved in the other facies (Figs 9, 10).

150-

175-

CO OQ

200-

£j

t/5 C/5

225]

250-

275400

420

440

460

Tmax(°C)

480

500

0.0

0.1

0.2

0.3

0.4

0.5

PRODUCTION INDEX 84-527

SADME

Fig. 3. Tmax a n d p r o d u c t i o n index (S,/S, + S 2 ) profiles, Bulldog Shale and C a d n a - o w i e F o r m a t i o n (lowermost sample), Toodla 1.

TABLE 3. C J 5 + extract yield and composition, SADME Toodla 1 core. Depth m

Formation

151.70 160.00

Bulldog Sh.

TOC W%

EOM yield ppm

H'C yield mg/gC

Arom. °7o

ONS <Vo

1.2 2.1

0.85 0.91 0.77 1.06 0.97 0.71 1.00 1.10

208 149

6 1

170.35 183.10 194.47 203.00 209.00 214.70

133 157 308 219 265 302

3 2 3 6 2 3

24.7 6.8 10.8 9.8 6.6 18.3 4.9 7.7

4.6 3.3 1.3 2.4 3.8 3.5

35.9 64.6 47.7 58.8 52.8 42.7 35.8 43.5

222.80 230.62 241.00 254.65 258.85 260.14

1.20 1.09 1.46 1.29 1.85 1.63

352 271 307 251 342 366

3 4 3 2 4 4

7.6 8.6 9.3 10.1 13.8 10.2

3.8 4.9 5.0 1.6 5.4 5.8

1.99

455

13.9

6.4

265.28

Cadna-owie Fm.

Hydrocarbon Parameters

EOM Composition Sat. °7o

Asph %

Pr/n-C 17

Ph/n-Cjg

Pr/Ph

DHA/n-C 2 0 0.29 0.09 0.10 1.4 1.9 0.66 0.72 2.4

38.2 26.5

0.31 0.41

0.33 0.32

0.77 1.2

0.64 1.0 1.0 0.79 3.6 6.5 1.5 1.3 1.6 0.68 1.4 0.93

0.34 0.64 0.65 0.56 14.9 27.5 1.0 0.82 0.68 0.39 0.95 0.55

1.6 1.5 1.7 1.2 0.24 0.27

61.0 48.3 67.5 62.0 63.7 61.7

36.9 28.1 39.3 36.6 55.5 45.3 27.6 38.2 18.2 26.3 17.1 22.3

60.5

19.2

1.6

1.1

1.4 1.6 2.2 1.9 1.7 1.9 1.7

3.3 2.6 2.0 0.76 0.10 0.19

0.11

DHA = dehydroabietane 2 o

TABLE 4. C J 5 + extract yield and composition, Delhi Strzelecki 3. VR

TOC

Wt°7o

EOM yield ppm

0.48-0.55

1.07

601

1054.6-1063.7 1094.2 1127.8 1158.8 1189.3 1253.9

3.00* 0.90 0.97 0.71

872 461 276

351 10

552 1136

1274.1 -1283.2 1283.2 1298.4 1316.1 1344.2 1362.4

2.36 1.07 1.44 0.63

Depth

Sample

swc Cuttings SWC

Cuttings SWC

Cuttings

1042.4

1365.5-1374.6

* Probable cavings t Stained

Formation

Bulldog Sh.

0.82 1.45

Cadna-owie Fm. Murta Member (Mooga Fm.)

0.55-0.58

0.58-0.60

0.80 1.26

220

H'C yield mg/gC

EOM Composition

Alkane Parameters

Sat. °/o

Arom. %

ONS °7o

Asph %

Pr/n-C, 7

Ph/n-C 18

Pr/Ph

7.7

6.1

53.1

33.1

0.85

0.14

4.7

13.5 53.1 24.8

2.3 14.7 10.8

62.1 10.5 38.7

22.1 21.7 25.7

1.93* 0.61 0.56

0.38 0.10 0.10

4.6 5.7 4.5

19.8

11.0

50.2

19.0

0.76

0.13

5.5

8.0 16.1 13.3

46.6 37.8 40.8

27.1 24.4 26.5

1.00

0.24

30 36

18.4 21.7 19.4

0.60 0.98

4.5 4.8 6.7

0.76 0.37

0.07

4.5

0.37

0.09

4.3

499 851 1581 423 510 1258

25.7

17.7

35.6

38.5

9.5

32.8

21.0 19.2

884

38.6

8.7

33.2

19.5

t*i ^ t>

6.1

TABLE 5. Kerogen elemental and isotopic composition, S A D M E Toodla 1. Depth m

Formation

151.70 160.00 170.35 183.10 194.47 209.00 214.70 222.80 230.62 236.12 241.04 258.85 265.28

Bulldog Sh.

Cadna-owie F m .

70.07 70.99 69.79 70.24 69.15 (69.72) 69.81 72.51 71.12 71.46 69.25 69.44 72.74

(1) S organic = S total - S pyritic (2) By difference (3) By direct determination using Unterzaucher method * Ash assumed to be entirely F e 2 0 3 (ex pyrite) d . m . m . f . Dry, mineral-matter free ( ) Dry, ash-free value n.d. Not determined

3.99 3.45 3.51 3.95 4.09 (4.18) 4.38 4.15 4.21 4.33 4.38 5.16 6.23

N d.m.m.f.

1.61 1.56 1.61 1.53 1.52 (1.52) 1.58 1.43 1.51 1.58 1.59 2.01 1.97

S (1)

O (2)

O (3)

Fe pyritic

°7o

2.2 1.9 1.6 2.1 3.6 (4.6) 2.8 1.7 2.2 1.0 2.0 2.5 2.0

22.1 22.1 23.5 22.1 21.6 (20.0) 21.4 20.2 20.0 21.6 22.7 20.8 17.0

n.d. n.d. 21.5 19.5 21.3 n.d. 18.7 20.3 n.d. n.d. 19.7 19.1 16.2

3.2 2.8 6.8 6.1 4.7 3.0 5.1 2.6 2.1 2.2 7.7 4.5 4.3

2.0 2.9 *

n.d. *

1.8 1.0 *

5.3 1.0 0.6

Ash

H/C

O/C atomic

—

C/N

6,3CPDB oo

-24.21 -24.18 -23.68 -23.19 - 24.26 n.d. -25.46 -23.98 n.d. n.d. n.d. -28.02 -28.17

0//

0.68 0.58 0.60 0.67 0.71 0.71 0.75 0.68 0.70 0.72 0.76 0.89 1.03

0.24 0.23 0.24 0.22 0.23 0.21 0.22 0.21 0.21 0.23 0.23 0.21 0.17

51 53 51 54 53 53 52 59 55 53 51 40 43

X >

O >

O tn O o

X m o >

r ^

o 5 GO

294

D. M. McKIRDY ET AL.

T A B L E 6. Dispersed organic matter and vitrinite reflectance, SADME Toodla 1.

Depth m

Formation

VR °7o

151.70

Bulldog Sh.

0.34 (n = 13)

160.00 170.35 183.10

0.37 (n = 16) 0.36

194.47

(n-8)

203.30 209.35 214.70 222.80 230.62 241.00 254.65 258.85 260.14 265.28

0.34 (n = 22)

Cadna-owie Fm.

Percentage of DOM V I E

p h y t o / S p

E x i n i t e

M a c e r a l s

Ratio

65:35

phyto, sp (trace ?cut/lipto)

< 5 < 5 10

65 70 60

30 25 30

70:30 55:45 80:20

phyto, sp, lipto phyto, sp, lipto phyto, sp, cut

55:45

phyto, sp, cut, ?res

— < 5 — — 20

45 55 60 70 50

55 40 40 30 30

65:35 70:30 60:40 70:30 60:40

sp, phyto, lipto, ?res phyto, lipto, ?sp, tela phyto, sp, lipto, cut, ?res phyto, sp, tela phyto, sp, lipto, tela

<5 <5 5 — —

50 35 35 40 25

45 60 60 60 75

30:70 45:55 60:40 55:45 55:45

sp, phyto (trace tela, lama) sp, phyto, lipto phyto, sp, cut, lipto phyto, sp, lipto phyto, sp, lipto, lama, cut

KEY V Vitrinite phyto phytoplankton res I Inertinite lipto liptodetrinite cut E Exinite sp sporinite tela *Much of exinite appears oxidised (muted fluorescence colour and intensity)

resinite cutinite telalginite

lama bmite

lamalginite bituminite

T A B L E 7. Dispersed organic matter and vitrinite reflectance, Delhi Strzelecki 3. Refer Table 6 for key.

Depth m 1042.4 1094.2 1127.8 1158.8 1189.3 1253.9 1283.2 1298.4 1316.1 1344.2 1362.4

Formation Bulldog Sh.

VR °Io 0.48-0.55

Cadna-owie Fm.

0.55-0.58

Murta Member (Mooga Fm.)

0.58-0.60

Percentage of DOM V I E — 30 70 10 85 5 85 10 <5 — 85 10 90 5, <5 75 20 5 75 20 5 45 5 55 — 30 70. — 30 70 — 60 40

Both DOM and kerogen descriptions indicate a higher proportion of ostensibly oil-prone constituents (viz. phytoplankton, sporinite, liptodetrinite, Table 6; cf. amorphous, biodegraded terrestrial, spore/pollen, Table 8), and this is reflected in the more aliphatic character of the corresponding kerogen pyrolysis-GC traces (Fig. II). The abundance of aromatic and phenolic species relative to w-alkyl moeities is low, although this abundance increases towards the top of the Facies A interval (Fig. 7). C alkane distributions (Fig. 14) are dominated by waxy C fl-alkanes (Fig. 16) indicative of a major land plant input. Pristane/phytane ratios are low (pr/ph = 1.7-2.2) because of the thermal immaturity of the organic matter. The relative concentration of dehydroabietane shows a steady upward increase through Facies A into the basal part of the overlying Facies B (Figs 8, 14). Likely biological precursors of this monoaromatic hydrocarbon (Fig. 15) are higher-plant diterpenoid resin acids like abietic acid (Simoneit, 1977). A similar origin was proposed for the dehydroabietane found in a solvent extract of the Rundle oil shale (Regtop et al., 1983). This basal facies, dated as early Aptian by Foraminospora asymmetricus (N. Alley, pers. comm.), was deposited during 15+

23+

p h y t o / S p

Ratio 40:60 70:30 65:35 60:40 70:30 45:55 30:70 60:40 — '

—

0:100

E x i n i t e

Macerals*

lipto, sp, bmite, phyto lipto, phyto, sp, bmite lipto, sp, phyto, bmite, ?lama lipto, bmite, phyto, sp lipto, bmite, phyto, sp, cut bmite, lipto, sp, lama, phyto, tela bmite, lipto, sp, lama, cut, phyto bmite, lipto, lama, phyto, sp, cut, tela lipto, bmite lama, lipto, tela lama, sp, tela, lipto

the initial phase of a major transgressive-regressive cycle in the Eromanga Basin (Morgan, 1980). It appears to contain the only oil-prone organic matter (Fig. 12) in the Bulldog Shale/Cadna-owie Formation section examined. Facies B occurs between 175 and 233 m depth in Toodla 1. A sub-facies, Facies B', occupies the middle portion of the unit (approx. 190-220 m). Total organic carbon (TOC = 0.71-1.20%), pyrolysable carbon (PC = 0.02-0.03%) and potential hydrocarbon yield (S + S = 0.20-0.42 kg/tonne) are all less than in Facies A (Fig. 4). The hydrogen index and kerogen atomic H/C values of Facies B'(HI = 25-39; H/C = 0.71-0.75) are slightly higher than those of Facies B (HI = 18-26; H/C = 0.67-0.70) (Figs 5, 10). This is considered due to intense anaerobic bacterial reworking of mainly allochthonous woody Type III organic matter (cf. increased proportions of amorphous plus biodegraded terrestrial types of particulate organic matter in kerogen: Table 7). Other evidence of in situ bacterial activity is provided by the very high relative abundance of the isoprenoid alkanes, pristane and phytane, in the C alkanes (Figs 8, 14). Another notable feature of the Facies B' geolipids is a sharp decrease in the abundance of dehydroabietane (Fig. t

15+

295

BULLDOG SHALE ORGANIC GEOCHEMICAL FACIES

POTENTIAL YIELD (S1+S2) 84-528

SADME

Fig. 4. Total organic carbon and potential hydrocarbon yield (S, + S ) profiles, Bulldog Shale and Cadna-owie Formation (lowermost sample), Toodla 1. 2

H/C FACIES, 0 1 *™ atomic

£ LU

C/N atomic

813C

OIL PRONE %

GQ OQ

200-

0/C atomic

22584-530 SAOME

250-

Fig. 6. Variation of kerogen composition with depth in Bulldog Shale and Cadna-owie Formation (lowermost sample), Toodla 1. Percent oil prone = sum of particulate organic matter types 1-4 inc., Table 8.

275"

100

150

200

250

300

84-529

SADME

HYDROGEN INDEX

FACIES DEPTH

Cresols

n-Cii:!

C2~Biphenyl Prist-1-ene

n-Ci&i

n-Ci7:i

Fig. 5. Hydrogen index profile, Bulldog Shale and Cadna-owie Formation (lowermost sample). Toodla 1.

8). The pyrolysates of Fades B' kerogens are distinguished by their high C -biphenyl/n-C ,. alkene ratios (Fig. 7), although the significance of this parameter is not yet understood. Facies B was deposited under low-energy conditions during a high stand of sea level (including cycle 8 , Morgan, 1980). The base of the Crybelosporites striatus spore/pollen unit, which coincides with the onset of a regressive pulse elsewhere in the Eromanga Basin, occurs at 207 m depth in Toodla 1 (N. Alley, pers. comm.). The /2-alkane profiles of these sediments (Fig. 16), supported by the optical recognition of 2

84-531 SADME

Fig. 7. Variation of kerogen pyrolysis-GC parameters with depth in Bulldog Shale and Cadna-owie Formation (lowermost sample), Toodla 1.

D. M. McKlRDY ET AL.

296 facies D E £ ™

Pr/Ph

Pr/n-C 1 7

1.1

DHA/n-C ,

Ph/n-Cis

FACIES A

o \

X 0.9 h

O OSL o

14.9 27.5

84—532 SADME

Fig. 8. Variation of C hydrocarbon composition with depth in Bulldog Shale and Cadna-owie Formation (lowermost sample), Toodla 1.

11 * *

% 1

0.5

^Nf 1

A C I E S

F A C , E S

[ / \

FACIES C

0.10

0.20

ATOMIC RATIO 0 / C

0.30 84-533

SADME

Fig. 9. Variation of kerogen type with organic facies, Bulldog Shale and top Cadna-owie Formation, Toodla 1. Toolebuc Formation kerogen [*] included for comparison.

phytoplankton in their DOM (Table 6), suggest periodic high inputs of algal lipids (e.g. at 230.62 and 203.00 m depth). However, the high proportions of aromatics and phenols in the kerogen pyrolysate (Figs 7, 11, 12) demonstrate quite conclusively that the organic matter is gas-prone. Facies C continues the upward trends of decreasing organic richness (TOC = 0.82-0.94%) and hydrocarbon generative potential (Sj + S = 0.13-0.18 kg/tonne), and of deteriorating kerogen quality (PC = 0.01%; HI = 13-18; atomic H/C = 0.58-0.68), although there is some evidence of a reversal in these trends near the top of the Toodla 1 section studied (Fig. 6). The kerogen in Facies C is hydrogen-poor, oxygen-rich Type IV organic matter (Fig. 9). Accordingly, the kerogen pyrolysate is very aromatic in character and gas-prone (Figs 7, 11, 12). 2

FACIES FACIES

-22

I FACIES C

-24

-26 8 1 3 C

PDB

-28 SADME 84—534

Fig. 10. Variation of kerogen elemental and carbon isotopic composition with organic facies, Bulldog Shale and top Cadna-owie Formation, Toodla 1.

Alkane distributions in Facies C (Figs 14, 16) are similar to those of Facies B, except in the uppermost sample analysed (151.70 m depth) where a major algal input is apparent. The concentration of dehydroabietane is low and comparable with that in lower Facies A (Fig. 8). The up-hole decrease in TOC within the Bulldog Shale at Toodla 1 reflects a steady decrease in sedimentation rate and, consequently, a slower rate of passage of detrital organic matter through the near-surface zone of intense bacterial and fungal decay (Ibach, 1982). The correlation between w-alkane profiles (Fig. 16) and phytoplankton to sporinite ratios (Table 6) for Toodla 1 generally is poor. This is to be expected in an oxic marine environment where planktonic (fatty) lipids are more susceptible to bacterial oxidation than are land-plant waxes (Goldhaber & Kaplan, 1975).

1.5

\O X o h< en l.o o

0.7-

STRZELECKI 3, TENNAPERA TROUGH Maturity The Bulldog Shale in this exploration well is marginally mature (VR = 0.48-0.55%). As at Toodla 1, Tmax values (434-447°C) are considerably higher than normally would be expected for Type III kerogen of such maturity (viz. ca 430-435°C; EspitaliS et al., 1983). A trend of increasing maturity with depth is clearly evident from the production index profile of Early Cretaceous sediments (Bulldog Shale, Cadna-owie Formation, Murta Member) in Strzelecki 3 (Fig. 17). Organic geochemical facies Although its TOC values fall within the same range as at Toodla 1, the lower two-thirds of the Bulldog Shale in Strzelecki 3 do not display the same regular down-hole increase in organic carbon content (Fig. 18). This may be because cuttings, not core, were analysed. There is some evidence from C extract data (Table 4) that organically .very rich cuttings from the upper one-third of the Bulldog Shale (TOC ^ 2 % ) may be contaminated by cavings from overlying Cretaceous sediments. Profiles of potential hydrocarbon yield and hydrogen index (Figs 18, 19) demonstrate the presence of the same poor quality, apparently gas-prone kerogen that characterises most of the Bulldog Shale section examined at Toodla 1. This is despite exinite contents of 20-45 per cent in the DOM near the base of the formation and in the underlying Cadna-owie Formation (Table 7). The exinite comprises bituminite and/or 15+

BULLDOG SHALE ORGANIC GEOCHEMICAL FACIES

297

UX2

Fig. 11. Pyrolysis-GC traces of representative kerogens, Bulldog Shale and top Cadna-owie Formation, Toodla 1. Key: (A = aromatic hydrocarbon, P = phenol, I = isoprenoid alkane (or alkene); numbers refer to carbon numbers of ;;-ulkenc//i-ulkunc doublets.

D. M. McKIRDY ET AL.

298

TOODLA-l 170.35m BULLDOG SHALE FACIES C

TOLUENE/n-C 7:1 84-536

SADME

Fig. 12. Variation of kerogen pyrolysate with organic facies, Bulldog Shale and top Cadna-owie Formation, Toodla 1.

INERTINITE

Ph 19

VITRINITE

Fig. 13. Variation of organic petrology with organic facies, Bulldog Shale and top Cadna-owie Formation, Toodla 1.

'structureless organic matter' (cf. Kantsler et al., 1983), liptodetrinite, sporinite and discrete phytoplankton (Table 7). However, as at Toodla 1, the muted fluorescence of much of the exinite (phytoplankton, sporinite, liptodetrinite: moderate to dull orange) attests to its partial oxidation and consequent loss of oil-generating potential. The sidewall core «-alkane profiles shown in Figure 20 suggest a strong algal/bacterial affinity for the EOM, although the corresponding pristane/phytane ratios (pr/ph = 4.5-5.7) are more indicative of terrigenous organic matter. This is characteristic of most of the marine Cretaceous sequence in the Eromanga Basin (P. S. Moore, pers. comm.). Sampling of the Bulldog Shale in Strzelecki 3 is too sparse to permit recognition of the individual organic geochemical facies identified in Toodla 1. Nevertheless, a feature of the DOM preserved in the Bulldog Shale at both well localities is the high concentration of inertinite (50-90%, except in lower part of Facies A at Toodla 1) and the general paucity of vitrinite (commonly 5%).

I [ |

TOODLA-l 265.28m CADNA-OWIE FORMATION FACIES A

ML/U

IJujjUA* ^ ^

Fig. 14. Gas chromatograms of C hydrocarbons (alkanes, plus dehydroabietane*) in Bulldog Shale and top Cadna-owie Formation, Toodla 1.

IMPLICATIONS FOR EROMANGA BASIN SOURCE-ROCK ANALYSIS The Bulldog Shale lies above the gas and oil-generation window for resinite-poor terrigenous organic matter in Toodla 1 and Strzelecki 3 (VR < 0.6% at both localities). Nevertheless, many of the findings of the present study are pertinent to the hydrocarbon-source potential of similar Cretaceous marine units throughout the Eromanga Basin. Maturity In shales and siltstones containing inertinite-rich DOM, maturation levels deduced from Rock-Eval Tmax measurements are consistently higher than indicated by measured VR values. For Type III kerogen (Fig. 9) Tmax is

BULLDOG SHALE ORGANIC GEOCHEMICAL FACIES

299

a function of not only maturity but also DOM composition, particularly in the pre-peak oil generation phase of catagenesis (VR < 0.8%). The major controlling factor appears to be the relative abundance of vitrinite and inertinite, as illustrated by the following data: DOM*

VR °7o

Tmax °C

Reference

Vitrinite-rich (V; minor I, E) Inertinite-rich ( I > E ; minor V)

0.5 0.5

430 440

Espitali£ et al., 1983 This study; unpubl. data

•Both have bulk elemental composition of Type III kerogen.

rf^

Likewise, in thermally immature sediments (VR <0.5%), DOM composition may exert a major influence on another Rock-Eval maturation parameter, the production index (Figs 3, 17). In Toodla 1, the inverted production index profile is attributable largely to a down-hole decrease in the relative abundance of inertinite and a concomitant increase in pyrolysable carbon (PC) and hence also in kerogen pyrolysate (S2).

COOH

Source input The proximal marine Bulldog Shale contains a mixture of allochthonous land-plant detritus and autochthonous aquatic organic remains. This DOM comprises mostly inertinite and exinite; vitrinite is rare to absent. Discrete phytoplankton (including colonial Botryococcus-Wke forms, listed as telalginite in Tables 6 and 7), and sporinite in association with liptodetrinite (probably derived from both phytoplankton and sporinite), are the major exinite components. Although not observed at Toodla 1, bituminite is present in the Bulldog Shale at Strzelecki 3, particularly near the base of the formation where it is the dominant exinite. FACIES C

ABIETIC ACID C20H30O2

^s^As^

DEHYDROABIETIN C19H28

DEHYDROABIETANE C20H30 84-539

FACIES B Pr/Ph

DEPTH (metres)

SADME

Fig. 15. Structure and probable derivation of two diterpenoid hydrocarbons identified in the Bulldog Shale, Toodla 1.

FACIES A

DEPTH (metres)

Pr/Ph

DEPTH (metres)

194.47

1.7

241.00

203.00

1.2

Pr/Ph

183.10

0.77

</)

0.27

254.65 LU

—!

258.85

260.14

160.00

222.80

170.35

230.62

265.28

CARBON NUMBER

CARBON NUMBER 84-540

SADME

Fig. 16. Normal alkane profiles and corresponding pristane/phytane ratios, Bulldog Shale and top Cadna-owie Formation, Toodla 1.

D. M. McKIRDY ET AL.

300

This mixed-source, inertinite-rich organic matter assemblage appears to be characteristic of other Early Cretaceous shallow marine mudstones and siltstones in the Eromanga Basin, including those of the underlying Cadna-owie Formation and the Wallumbilla Formation (basin centre equivalent of the Bulldog Shale) (Cook, 1982). The distinctive marine organic facies of the Toolebuc Formation is atypical in that it contains abundant bituminite (Hutton et al.y 1980), the origin and palaeoenvironmental significance of which are discussed by Cook (1982) and Sherwood & Cook (this volume).

0.0

0.1

0.2

0.3

PRODUCTION INDEX

0.4 84-541

SADME

Fig. 17. Production index ( S , / S , + S 2 ) profile of Early Cretaceous units (Bulldog Shale, Cadna-owie Formation, Murta Member), Strzelecki 3.

Accordingly, the Toolebuc 'oil shale' kerogen is richer in hydrogen (atomic H / C = 1.19-1.26) and has a Type II elemental composition (Fig. 9). Geochemical signatures of the Aptian mixed organic matter originally deposited in a generally low-energy marine environment at the Toodla 1 locality include the following: (1) low ratios of pristane to phytane (pr/ph H 2); (2) the obvious presence of dehydroabietane (derived from higher plant resins); and (3) Type III kerogen with a moderately high sulphur content (2-4%). Oxic conditions in the water column resulted in extensive oxidation of planktonic remains and land-plant exinites (cf. Demaison & Moore, 1980; Jenkyns, 1980; Tissot et al., 1980). The bottom waters were sufficiently oxygenated to allow bioturbation of the underlying sediment (at least at Toodla 1) and, ultimately, the preferential decay of the cellulose and lignin-based precursors of vitrinite by aerobic bacteria and fungi. Anomalously high phytane concentrations in some of the Bulldog Shale (Facies B) extracts (ph/rt-C, 8 pr/ph-< 1) correlate with the highest asphaltene contents (45-55% of EOM) and probably reflect intense reworking of detrital organic matter (terrigenous and aquatic) by anaerobic bacteria below the oxic-anoxic boundary in the sediment column. Comparison of the Toodla 1 extracts with those from the more deeply buried Bulldog Shale sequence at Strzelecki 3 demonstrates how the extractable hydrocarbons change dramatically with increasing thermal maturity over the range VR = 0.35-0.55%. The rt-alkane profiles of the Bulldog Shale at Toodla 1 are (with only three exceptions) dominated by land-plant-derived C 2 3 + waxy hydrocarbons displaying a marked odd/even predominance indicative of their thermal immaturity (Fig. 16). In marked contrast, the Strzelecki 3 (Bulldog Shale) nalkane distributions are similar to those of a mature non-waxy oil and suggestive of an algal or bacterial origin (Fig. 20). Paradoxically, however, the associated pristane/phytane ratios are high (pr/ph = 4.5-5.7) and more indicative of terrigenous

1000-

POTENTIAL YIELD (S1+S2) 84-542

Fig. 18. Total organic carbon and potential hydrocarbon yield (S, + S 2 ) profiles of Early Cretaceous section, Strzelecki 3.

SADME

BULLDOG SHALE ORGANIC GEOCHEMICAL FACIES

301

pristane from woody-herbaceous organic matter commences (Powell et al, 1978). This, in conjunction with the high concentrations of discrete pristane (relative to phytane) in certain aerobic bacteria (Han et al., 1968), probably accounts for the observed high pristane/phytane ratios. Differences in the respective alkane distributions of the Bulldog Shale at Toodla 1 and Strzelecki 3 are unlikely to reflect only differences in source material. It is true that bituminite is present at Strzelecki 3 but not at Toodla 1. On the other hand, recognisable algal remains (phytoplankton) are proportionally more abundant in the DOM at Toodla 1 (Tables 6, 7), contrary to what might reasonably be inferred from a comparison of the corresponding rt-alkane patterns (Figs 16, 20). Hydrocarbon generating potential It is of interest that the C normal and isoprenoid alkane distributions of the Bulldog Shale at Strzelecki 3 are very similar to those of the light, low-wax, paraffinic oils found in the Murta Member of the Mooga Formation at Dullingari and Jackson. The origin of these oils from marginally mature (VR - 0.5-0.6%) Early Cretaceous source beds containing bacterially reworked terrigenous organic matter is confirmed by the composition of their gasoline-range and biomarker (sterane, triterpane) hydrocarbons (McKirdy, 1984a, b). Direct evidence that some liquid hydrocarbons have in fact been generated within the Bulldog Shale at Strzelecki 3 is provided by the oil, visible (using fluorescence-mode microscopy) as thin films of gold fluorescence coating quartz grains in polished sections of several sidewall cores (Kantsler, 1979). These particular samples are shales with moderately high C hydrocarbon yields (expressed as mg/g TOC, Table 4). Nevertheless, the Rock-Eval pyrolysis (whole-rock) and pyrolysis-GC (kerogen) results clearly demonstrate that the Bulldog Shale at Toodla 1 and Strzelecki 3 is essentially gas prone. The reason for this serious discrepancy is simple: exinite preserved in the Bulldog Shale is highly comminuted and therefore, not surprisingly, also oxidised. Its partial oxidation can be detected by comparing the observed fluorescence colour and intensity of any individual exinite component (e.g. phytoplankton, sporinite) with that expected from its maturation level as determined independently by vitrinite reflectance. Oxidation has been found by us and by C. B. Foster (unpubl. data) to preferentially destroy the hydrogenrich lipid components (fats, waxes) which give rise to both the fluorescence and oil-source potential of exinite. In marked contrast, previous studies of Cretaceous marine and lacustrine units in the Eromanga Basin commonly have attributed definite oil-generating potential to these rocks (where they have attained adequate thermal maturity) because of the high exinite content of their DOM (e.g. Moore & Pitt, 1984). Throughout the Bulldog Shale at Toodla 1, and within its basal part at Strzelecki 3, exinite comprises 20-75 per cent of the DOM visible in polished section (Tables 6, 7; Fig. 13). Ostensibly oil-prone constituents account for a similar proportion of the identifiable particulate organic matter in kerogen concentrates of the same rock samples (Table 8; Fig. 6). It is clear that extreme caution is necessary when using organic petrological and/or palynological data alone to assess source quality. Information on not only the abundance of exinite, but also its state of preservation (including fluorescence characteristics), is essential. This is true particularly when dealing with rocks containing inertiniterich organic matter of mixed algal/land plant derivation. 15+

200

250

HYDROGEN INDEX

100

150

SADME

Fig. 19. Hydrogen index profile of Early Cretaceous section, Strzelecki 3. DEPTH (metres)

Pr/Ph

1042.4

4.7

15+

5.7

1094.2 Zj

1127.8 1253.9 1274.1 6.7

J.4 "

15 20 25 30 35

CARBON NUMBER

84-544

SADME

Fig. 20. Normal alkane profiles and corresponding pristane/phytane ratios, Bulldog Shale and Cadnaowie Formation, Strzelecki 3.

source material. This apparent anomaly arises from the marginal maturity of the DOM (VR = 0.48-0.55%) and its mixed bacterial-higher plant origin. At this maturation level generation of hydrocarbons occurs mainly from the amorphous algal/bacterial component of the kerogen e.g. bituminite (Powell & Snowdon, 1980, 1983). Thus, early catagenetic A7-alkanes of algal/bacterial origin will rapidly dilute the primary biolipid input of terrigenous /7-alkanes. Almost simultaneously, the preferential catagenetic release of

302

D. M. McKIRDY ET AL.

TABLE 8. Visual kerogen and thermal alteration index, SADME Toodla 1. Depth

Formation

151.70 160.00 170.35 183.10 194.47 203.00 209.00 214.70 217.60 222.80 230.62 236.12 241.00 258.85 265.28

""Particulate Organic Matter Types (%)

TAI 1

Bulldog Sh.

Cadna-owie Fm.

2.1 2.1 2.1 2.0 2.1 2.1 2.1 2.1 2.1 2.2 2.1 2.1 2.1 2.1 2.1

5 5 15 10 5 55 10 10 10 5 5 5 20 45 40

20 10 5 10 15 5 20 10 10 10 10 10 10 10 30

15 15 5 10 10 tr 15 15 15 15 10 10 10 5 5

60 70 75 60 65 40 55 60 60 65 70 70 55 40 25

—

— — — — —

tr — — — — — — — —

KEY 1 = Amorphous 2 = Structured aqueous 3 = Biodegraded terrestrial 4 = Spore/Pollen 5 = Structured terrestrial 6 = Inertinite

CONCLUSIONS This study has identified a vertical succession of three distinct organic geochemical facies within the Bulldog Shale, an Early Cretaceous marine unit in the southwestern Eromanga Basin. These facies were defined using the geochemical and microscopic techniques of hydrocarbon source-rock analysis. They are a manifestation of progressive changes in sedimentation rate, and in the balance of autochthonous (algal) and allochthonous (higher plant) organic inputs which survived in situ microbial activity, during a major marine transgression. Total organic carbon,, kerogen H/C atomic ratio, and exinite content are highest in the basal facies of the Bulldog Shale and decrease upwards through succeeding phases of the transgression. This vertical deterioration of source richness and quality, a function of decreasing sedimentation rate, has important implications for predictive mapping of marine source rocks in the Eromanga Basin. Maximum total organic carbon values in the range 1.5-2.5%, and oil and gas-prone Type III kerogen (H/C = 0.75-1.1, S ~ 2%) characterise the transgressive shallow marine environment represented by the basal unit of the Bulldog Shale. The rest of the formation is gas prone. From the viewpoint of oil source-bed genesis, the key factor limiting organic richness and source quality (kerogen type) in the case of the Bulldog Shale appears to have been the oxidation of marine phytoplankton and land-plant exinites that occurred during sedimentation through, and early diagenesis below, an oxic water column. Pyrolytic assessments of kerogen quality using either RockEval (whole rock) or pyrolysis-GC (kerogen) techniques may underestimate the oil-proneness of organic matter which comprises a mixture of inertinite and exinite, and has the bulk composition of a Type III kerogen. Standard optical microscopic techniques, on the other hand, commonly are incapable of adequately characterising a large proportion of the organic matter present in marine shales and siltstones. In the Bulldog Shale, for example, so-called Indeterminate fines' (probably gas prone) comprise up to 70 per cent of the

— —

10 5 — —

5 5 5 5 5 5 — —

Oil Prone %

Indeterminate Fines <7o

40 30 25 30 30 60 45 35 35 30 25 25 40 60 75

5 5 25 5 5 70 5 10 10 5 5 5 30 50 50

* After Masran & Pocock (1981) Oil prone = Sum of Types 1-4 inclusive TAI = Thermal alteration index

total particulate organic matter. This incomplete optical characterisation of the kerogen or DOM may lead to overestimation of its oil-source potential, particularly when the state of preservation and fluorescence characteristics of the exinite are ignored. The composite hydrocarbon generation model recently proposed by Powell & Snowdon (1983) includes two major kinds of oil-prone Type III kerogen. Our studies of the Bulldog Shale have led to the recognition of another variety of Type III kerogen which, although exinite-rich, is gas prone. This particular kerogen has a mixed aquatic-terrestrial origin and comprises mostly oxidised exinite (phytoplankton, sporinite) and inertinite (I = E; minor V). Its anomalously high Tmax values distinguish it from normal vitrinite-rich Type III kerogen of equivalent maturity. Such organic matter almost certainly is the source of gas reservoired in the overlying Coorikiana Sandstone at Strzelecki 8 and Marabooka 3. Within the basal Bulldog Shale at Strzelecki 3, bituminite (i.e. algal and/or vascular plant remains reworked by anaerobic bacteria) is the major exinite in the inertinite-rich DOM. Similar mixed (microbial-higher plant) Type III kerogen in marginally mature (VR ~ 0.5-0.6%) Early Cretaceous mudstones is the likely source of the light Murta-type crude oils found in the Dullingari and Jackson fields of the Eromanga Basin. Further work is required to establish the minimum exinite/inertinite ratio necessary to ensure initial saturation of the inertinitic kerogen matrix with catagenetic hydrocarbons, and subsequently to permit pressure-driven expulsion of excess oil from the source rock during primary migration. ACKNOWLEDGEMENTS An earlier version of this paper was presented at the 53rd ANZAAS Congress, Perth, May 1983. Rock samples and background geological information for this study were provided by the South Australian Department of Mines and Energy (Toodla 1) and Delhi Petroleum Pty Ltd (Strzelecki 3). The extraction work and initial gas chromatography were performed by H. W. Sears (AMDEL).

B U L L D O G S H A L E O R G A N I C G E O C H E M I C A L FACIES G. W. Woodhouse (formerly of WAIT) undertook most of the Rock-Eval analyses. Visual kerogen descriptions and TAI measurements were the work of Sally Wise and Jill Stevens (Esso Australia). N. Alley (SADME) provided palynological data on Toodla 1. The Queen Elizabeth Hospital, Woodville, S.A. allowed one of the authors (R.E.C.) access to their GC-MS facilities. The figures were drafted by T. Sullivan (SADME). A n n Hards typed several versions of the manuscript. The contributions of all these individuals and organisations are gratefully acknowledged. M. Griffiths (SADME) and M. Zwigulis (Delhi) are thanked for their assistance during the course of the study. Discussions with C. B. Foster (WMC) led to our recognition of the link

303

between the oxidised condition of the exinite and its otherwise inexplicable lack of oil-source potential. The paper benefited from the critical reviews of P. S. Moore (Delhi) and L. A. Frakes (Monash University). Finally, the authors wish to thank Delhi Petroleum Pty Ltd and their partners in PELs 5 and 6, and the Director-General of Mines and Energy, S.A., for permission to publish. The research project of which this study is a part was supported by the National Energy Research Development and D e m o n s t r a t i o n P r o g r a m m e a d m i n i s t e r e d by the Commonwealth Department of National Development and Energy.

REFERENCES

ARMSTRONG, J. D. & BARR, T. M . , 1982: The Eromanga Basin; in

Moore, P. S. and Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 2 0 - 4 2 . BERNER R. A., 1981: A new geochemical classification of sedimentary environments. J. Sediment. Petrol., 51, 3 5 9 - 6 5 . COOK, A. C . 1982: Organic facies in the Eromanga Basin; in Moore, P. S. and Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 2 3 4 - 5 7 . DEMAISON, G. J. & M O O R E , G. T., 1980: Anoxic environments and oil source bed genesis. Am. Assoc. Pet. Geol., Bull., 64, 1179-209. ESPITALIE, J., LAPORTE, J. L., M A D E C , M . , MARQUIS, F., LEPLAT, P., PAULET, J. & BOUTEFEU, A., 1977: M&hode rapide de

caracterisation des roches meres, de leur potentiel petrolier et de leur degre devolution. Res. Inst. Franc. Petrole, 32, 23-42. Geochemical logging. Seminar, Institute Francais du Petrole S.E. Asia, Singapore, Dec. 1983 (unpubl.).

ESPITALIE, J., MARQUIS, F. & BARSONAY, I., 1983:

FLINT, R. B., AMBROSE, G . J. & CAMPBELL, K. S. W., 1980:

Fossiliferous Lower Devonian boulders in Cretaceous sediments of the Great Australian Basin. Trans. R. Soc. S. Aust., 104(3), 57-65. GOLDHABER, M . B. & K A P L A N , I. R., 1975: Controls and consequences of sulfate reduction rates in recent marine sediments. Soil Sci. 119, 4 2 - 5 5 . GRIFFITHS, M . , 1980: Toodla No 1 well completion report. S. Aust. Dept. Mines Ener. Rep. 80/128 (unpubl.). ;

HAN, J., MCCARTHY, E . D., VAN HOEVEN, W., CALVIN, M . & BRADLEY, W. H . , 1968: Organic geochemical studies II. A

preliminary report on the distribution of aliphatic hydrocarbons in algae, in bacteria, and in a recent lake sediment. Proc. Natl Acad. Sci. U.S. 59, 2 9 - 3 3 . HAUG, P. & CURRY, D. J., 1974: Isoprenoids in a Costa Rican seep oil. Geochim. Cosmochim. Acta, 38, 601-10. HUTTON, A . C., KANTSLER, A . J., COOK, A . C . & MCKIRDY, D. M . ,

1980: Organic matter in oil shales. A PEA J. 20(1), 44-67. IBACH, L. E. J., 1982: Relationship between sedimentation rate and total organic carbon content in ancient marine sediments. Am. Assoc. Pet. Geol. Bull. 66, 170-88. JENKYNS, H . C., 1980: Cretaceous anoxic events: from continents to oceans. J. Geol. Soc. London, 137, 1 7 7 - 8 8 . KANTSLER, A. J., 1979: Strzelecki No. 3, petrographic descriptions of organic matter in sidewall cores. Rep. for S. Aust. Dept. Mines and Energy and Delhi Petroleum Pty Ltd (unpubl.).

KANTSLER, A . J., PRUDENCE, T. J. C . , COOK, A . C . & ZWIGULIS,

M., 1983: Hydrocarbon habitat of the Cooper/Eromanga Basin, Australia. APEA J. 23(1), 75-92.

KANTSLER, A. J., COOK, A. C . & ZWIGULIS, M., 1986: Maturation

patterns in the Eromanga Basin: in This volume.

MASRAN, Th.C. & POCOCK, S. A. J., 1981: The classification of plant-

derived particulate organic matter in sedimentary rocks; in Brooks, J. (ed.) Organic Maturation Studies and Fossil Fuel Exploration. Academic Press, London, 1 4 5 - 7 5 . MCKIRDY, D. M . , 1982: Petroleum geochemistry and source-rock potential of the Cooper Basin and superjacent Eromanga Basin. Rep. for Delhi Petroleum Ltd (unpubl.).

MCKIRDY, D. M . , 1984a: Notes on the source affinity

and thermal maturity of Eromanga Basin crude oils. AMDEL Rep. F7243/84 for Delhi Petroleum Pty Ltd (unpubl.). MCKIRDY, D. M . , 1984b: Source rock and petroleum geochemistry, Naccowlah Block, southwest Queensland: conclusions and recommendations. AMDEL Rep. F6752/84 (Addendum) for Delhi Petroleum Pty Ltd (unpubl.). MONNIER, F., POWELL, T. G . & SNOWDON, L. R., 1983: Qualitative and quantitative aspects of gas generation during maturation of sedimentary organic matter. Examples from Canadian frontier basins; in Bjoroy, M. et al. (eds) Advances in Organic Geochemistry 1981. Wiley, Chichester, 487-95. MOORE, P. S. & PITT, G. M . , 1982: Cretaceous of the southwestern Eromanga Basin: stratigraphy, facies variations and petroleum potential; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 127-44. MOORE, P. S. & PITT, G. M . , 1984: Cretaceous of the Eromanga Basin—implications for hydrocarbon exploration. APEA J., 24(1), 358-76. M O O R E , P. S. & PITT, G. M . , 1985: Cretaceous subsurface stratigraphy of the southwestern Eromanga Basin: a review: in Linsday, J. M. (ed.) Stratigraphy, palaeontology, malacology. Papers in honour of Dr Nell Ludbrook. S. Aust. Dept. Mines Ener. Spec. Publ. 5, 269-86. MORGAN, R., 1980: Eustasy in the Australian Early and Middle Cretaceous. NSW Geol. Surv., Bull., 27. POWELL, T. G. & SNOWDON, L. R., 1980: Geochemical controls of hydrocarbon generation in Canadian sedimentary basins; in Miall, A. D. (ed.) Facts and Principles of World Petroleum Occurrence. Can. Soc. Pet. Geol., Mem. 6, 421-46. POWELL, T. G . & SNOWDON, L. R., 1983: A composite hydrocarbon generation model—implications for evaluation of basins for oil and gas. Erdol und Kohle, 36(4), 163-70. POWELL, T. G., COOK, P. J. & MCKIRDY, D. M . , 1975: Organic geochemistry of phosphorites: relevance to petroleum genesis. Am. Assoc. Pet. Geol., Bull., 59, 6 1 8 - 3 2 . POWELL, T. G., FOSCOLOS, A . E., GUNTHER, P. R. & SNOWDON,

L. R., 1978: Diagenesis of organic matter and fine clay minerals: a comparative study. Geochim. Cosmochim. Acta, 42, 1181-97.

REGTOP, R., CRISP, P. T. & ELLIS, J., 1983: Molecular markers and

other extractable components from Rundle oil shale. Proc. First Aust. Workshop on Oil Shale, 7 3 - 7 5 . 1986: Organic matter in the Toolebuc Formation: in This volume. SIMONEIT, B. R. T., 1977: Diterpenoid compounds and other lipids in deep-sea sediments and their geochemical significance. Geochim. Cosmochim. Acta, 41, 463-76. SNOWDON, L. R. & POWELL, T. G., 1982: Immature oil and condensate—modification of hydrocarbon generation model for terrestrial organic matter. Am. /Issoo. Pet. Geol., Bull., 66, 775-88. TAN, T. L. & HE IT, M., 1981: Biogenic and abiogenic polynuclear aromatic hydrocarbons in sediments from two remote Adirondack lakes. Geochim. Cosmochim. Acta, 45, 2 2 6 7 - 7 9 . SHERWOOD, N. R. & COOK, A. C.,

304

D. M. McKIRDY ET AL.

TISSOT, B., DEMAISON, G., MASSON, P., DELTEIL, J. R . & COMBAZ,

WAKEHAM, S. G . , SCHAFFNER, C . & GIGER, W M 1980: Polycyclic

A., 1980: Paleoenvironment and petroleum potential of Middle Cretaceous black shales in Atlantic Basins. Am. Assoc. Pet. Geol., Bull. 64, 2051-63.

aromatic hydrocarbons in Recent lake sediments—II. Compounds derived from biogenic precursors during early diagenesis. Geochim. Cosmochim. Acta, •44, 415-29.

Geological Society of Australia Special Publication No. 12, 305-322

Organic maturation in the Eromanga Basin

A. J. Kantsler , A. C. Cook & M. Zwigulis 1

1 2 3

Shell Development Company of Nigeria Ltd, EXPW, PMB2418, Lagos. Nigeria. Department of Geology, University of Wollongong, P.O. Box 1144, Wollongong, NSW, 2500. Delhi Petroleum Pty Ltd, 101 Grenfell Street, Adelaide, SA, 5000.

ABSTRACT

Vitrinite reflectance data are available from more than 90 wells which penetrate up to 2500 m of Jurassic-Cretaceous sediments in the southwestern Eromanga Basin. Reflectance ranges from less than 0.3% in parts of the Cretaceous Winton Formation to more than 1.7% in some intersections of the Jurassic Hutton Sandstone. Isoreflectance surfaces show varying relationships to regional structure and present-day subsurface temperature isotherms. Maturation modelling studies indicate a variable thermal history which can generally be accommodated within the context of an attenuating thermal regime. However, high regional heat flow is related primarily to high levels of crustal heat production. Coal beds and land-plant related organic detritus are present throughout the Eromanga section which includes at least three variably developed, potential source rock intervals. The onset of significant hydrocarbon generation from these intervals generally followed rapid deposition of the Cenomanian Winton Formation but in some areas overlying the central Nappamerri Trough of the Cooper Basin, intial oil generation may have accompanied this phase of regional subsidence. Host rocks for most oil and gas discoveries in the Eromanga Basin are associated with low to moderate levels of organic maturity, implying a source either from underlying, mature, coal-bearing Permian sediments or from more mature Jurassic section downdip. Maps of maturity at the top of several potential Jurassic-Cretaceous source intervals indicate that most future Eromanga Basin oil discoveries will be found flanking mature Jurassic (and Permian) sequences such as those in the Nappamerri Trough and Cooper and Thomson Synclines.

INTRODUCTION

The southwestern Eromanga Basin (Fig. 1) overlies the Carboniferous-Triassic Cooper Basin, Carboniferous-Permian Pedirka Basin, Triassic Simpson Desert Basin and Carboniferous-Permian Arckaringa Basin each of which is considered herein as a separate geographic sub-province. Some 12 000 megalitres (75 million barrels) of recoverable oil and 1.84 x 10 m (65 BCF) of recoverable gas have been found in sandstones of Early Jurassic to Early Cretaceous age, particularly in the vicinity of the southern Cooper Basin. At end of July 1982 in-place reserves totalled 18100 megalitres (114 million barrels) of oil and 2.24 x 10 m (79 BCF) of gas (Armstrong & Barr, 1982). Further discoveries have since been made. The origin and distribution of these hydrocarbons have been the subject of much conjecture (Poll, 1981; Bowering, 1982; McKirdy, 1982a, b\ Armstrong & Barr, 1982; Kantsler et al, 1983). Thin, rich, mainly humic source rock horizons which are locally developed within the non-marine Jurassic section and the marginal marine Early Cretaceous section (Thomas, 1982; McKirdy, 1982a, b\ Kantsler et al., 1983; Passmore & Boreham, this volume) are thought to be the source for much of the oil. However, a Permian origin or contribution is equally probable for some discoveries. In the absence of detailed data on the occurrence and lateral extent of these source rock horizons, this paper attempts to delineate prospective areas of the basin as follows: 9

(i) by definition of hydrocarbon generative areas at each of the possible source rock horizons; (ii) by establishing the timing of hydrocarbon generation with respect to the timing of structuration.

TECTONIC HISTORY

The Eromanga Basin, which forms part of the 1.7 million km area of the Great Artesian Basin, is a broad intracratonic downwarp covering an area of approximately 1 million km . The basin comprises a broad saucer-shaped depression in the west (where it overlies the Pedirka and Simpson Desert Basins) and a series of north- to northeasttrending depressions in the east which reflect the structure of the underlying Cooper Basin (Fig. 1). Major epeirogenic downwarp of most of eastern Australia in the Late Triassic-Early Jurassic initiated Eromanga Basin deposition on an irregular erosion surface. The cause of this downwarping is difficult to reconcile with simple models of thermal contraction (e.g. Sleep & Snell, 1976) or crustal extension (e.g. McKenzie, 1978) as subsidence of the underlying Carboniferous-Triassic sequences occurs much earlier than that of the overlying sequence. Subsidence of the Eromanga Basin sequence also merges into the subsidence pattern of a much larger area of continental Australia and, as Bally & Snelson (1980) have previously surmised, some thought needs to be given to the nature of lithospheric flexuring required to generate such supra-regional basins. 2

2 d W n

m | •?

Ui o ON

AMADEUS BASIN (Pl-C)

GALILEE BASIN ^ (C--&) / /

PEDIRKA BASIN

3 g «

DWANNA

MUSGRAVE BLOCK OFFICER BASIN (BI-6 , D - C ) \

/-t \ » H/

PANDIEBUR

> Z H r rn >

POOLOWANNA TROUGH

WARBURTON

C/3

COOPER BASIN ( P-T* )

BASIN

PEAKE-DENISON < INLIER

> 5"-

SIMPSON DESERT BASIN

CO % P D2

£ 2=

ADAVALE BASIN (D-C)

p o O

j^Jfc strzeleoT^-®^

o *

ARCKARINGA BASIN (P)

NSW APPROX. LIMIT ER0MAN6A BASIN

GAWLER BLOCK STUART SHELF

ADELAIDE FOLD BELT

ARROWIE BASIN

. LACHLAN \ FOLD \ BELT

(Bl-O)

S A P NSW

5 c r

ORGANIC MATURATION Middleton (1980) has suggested that the initial (PermianTriassic) phase of subsidence may have accompanied deep crustal metamorphism during the latter part of a subcrustal heating event. This was followed immediately by subsidence due to thermal contraction during the cooling period. In this two-phase history, each phase of subsidence exhibits exponential-like decrease with time as exemplified by the Nappamerri Trough region (Fig. 2). Sleep (1976) has explained similar two-phase burial history as superposition of eustatic variation upon the exponential thermal contraction subsidence curve. This hypothesis has some support in the Late Triassic erosion evident throughout much of eastern Australia. McKenzie's (1978) model of two-phase subsidence, which entails a rapid initial phase due to extension of the lithosphere followed by a slower, thermally contracting phase, is discounted by Middleton (1980) in view of its assumption of instantaneous stretching. However, Sclater & Christie (1980) maintain that such a model is capable of explaining the widespread post-rifting subsidence in the intra-cratonic North Sea Basin, with which the Cooper sub-province shows some affinity. Structural development within much of the Eromanga Basin sequence is initimately related to the structural configuration of underlying, structurally conformable Carboniferous-Triassic basins. Incipient pre-Permian rifting created a series of grabens, half-grabens and uplifted fault blocks related to high-angle basement faults, which remained active throughout Permian deposition. The regional structure of the Eromanga Basin is characterised by numerous broad, low amplitude folds, major faults and fault-bounded anticlines (e.g. Harkaway, Mt Howitt, Erabena), and several prominent anticlinal trends with surface expression (e.g. Curalle, Innamincka, Morney). Most structural development during the Mesozoic is related to drape and differential compaction although some structures (e.g. Merrimelia, Mt Crispe-Witcherrie Ridge) show evidence of concomitant growth through continued movement of basement faults. A major phase of apparent left-lateral, wrench-induced folding occurred in the Early Tertiary and created many of the larger anticlines in the northern Cooper, Simpson Desert and Pedirka Basin sub-provinces while enhancing structures elsewhere (Wopfner, 1960; Sprigg, 1961; Moore & Pitt, 1984). This phase of tectonism continued until

307

at least the late Oligocene (Wopfner et al., 1974; Senior et al.y 1978) and was followed by minor post-Miocene downwarping (e.g. Poolowanna Trough).

STRATIGRAPHY AND BASIN HISTORY The stratigraphy adopted in this paper (Fig. 3) is illustrated and described by Moore & Pitt (1982), Gravestock (1982), Wiltshire (1982) and Moore (1982). In summary, up to 1000 m of Jurassic fluvial, lacustrine and coal swamp sediments were deposited in a broad, gently subsiding, cratonic basin. Marine transgression in the Neocomian-Aptian was accompanied by paralic-shallow marine sedimentation, which persisted throughout the Albian. Subsidence culminated with rapid deposition of the Cenomanian Winton Formation, a paralic, lacustrine and fluvial sequence up to 1100 m thick. The total Cretaceous sequence is more than 1600 m thick in several of the major depressions (Nappamerri, Patchawarra, Poolowanna and Windorah Troughs, Fig. 1).

OCCURRENCE OF HYDROCARBONS Oil and gas have been found in most of the units within and below the Coorikiana Sandstone at depths between 975 m and 2400 m (Fig. 3). Major producing horizons are the Hutton Sandstone (Early-Middle Jurassic) and Namur Sandstone Member (Late Jurassic-Early Cretaceous) of the Mooga Formation. Most accumulations discovered to date occur in simple anticlinal structures (e.g. Barr & Youngs, 1981; Smith, 1983) or fault-bounded anticlines (e.g. Porter, 1978), although the Murta Member oil accumulation at Dullingari occurs in a combination trap formed by drape of a laterally restricted shoreface sand over a basement horst (Mount, 1982). The majority of the commercial discoveries (described by Armstrong & Barr, 1982) are located over the southern Cooper Basin sub-province in structures which flank the major depositonal troughs, in particular the Nappamerri Trough. In part, this geographic imbalance in the distribution of discoveries reflects historically higher levels of exploration activity in the South Australian part of the Cooper subprovince. However, good oil shows have also been observed in the Simpson Desert sub-province (Poolowanna 1) and a substantial oil discovery has recently been made in the northern Cooper sub-province (Tintaburra 1).

SOURCE ROCKS

Fig. 2. Subsidence history Kirby 1, central Nappamerri Trough, Cooper $nd Eromanga Basins.

Several good source rock intervals containing mainly humic (Tissot Type II/III) to humic (Tissot Type III) organic matter are developed locally within the Eromanga Basin (see Cook, 1982, this volume) above good to excellent, mixed to mainly humic source sequences in the underlying Permian strata of the Pedirka, Cooper and Arckaringa Basins. Early Palaeozoic basement sediments appear organically lean (Moore 1982; Youngs & Moorcroft, 1982; Kantsler et al., 1983) but data are sparse and this conclusion may, in some areas, prove premature. In general, dispersed organic matter and coals in the Permian section are exinite-poor, and vitrinite and inertinite predominate (Fig. 4). Although such land plant-dominated source rocks are generally considered to be a source primarily of gas and condensate, local concentrations of exinite (up to 25% of some coals and 5% of some accompanying sediments) and bacterially-degraded organic matter enhance oil source potential. Smyth (1979) and Kantsler et a/.(1983) have suggested that the oil found in the Patchawarra Trough of the Cooper Basin (Tirrawarra Sandstone) is sourced from overlying coal measures sediments. In view of their widespread

A. J. KANTSLER, A. C. COOK & M. ZWIGULIS

308

Triassic sequences such as the Nappamerri and Walkandi Formation are generally of red-bed affinity and have little source potential. However, Upper Triassic sediments present in the Simpson Desert and western Cooper Basins

distribution, Permian coal measures constitute a vast potential hydrocarbon source. Wherever generation and migration pathways allow, they must be considered a possible source of Eromanga Basin hydrocarbons. EROMANGA

BASIN

WINTON

FM.

MACKUNDA

FM.

(IIOO)

EROMANGA BASIN (200) OODNADATTA

FM.

ALLARU (270)

(300) 175)

TOOLEBUC

BULLDOG SHALE

(120)

MUDSTONE

FM. ^(700) >

(460) TRANSITION

COORIKIANA S S T W A L L U M B I L L A FM

-(20)

BEDS

( C A D N A - OWIE

FM)

- SUB GROUP

WYANDRA S S T MBR MURTA

MBR ^

(80)

(60) NAMUR S S T

ALGEBUCKINA SST

[MAREE

:E5T

(800)

MBR • .

ADORI S S T BIRKHEAD

ALGEBUCKINA

SST

(150)

(110)

(95) #

HUTTON S S T

(230) •

'BASAL

POOLOWANNAFM.

JURASSIC (132)

SIMPSON DESERT BASIN

COOPER BASIN

MT. TOONDINA FM,

^(400)

BOORTHANNA

FM.

OFFICER BASIN

PURNI FM

\ >< m

v^jjssP-? .

CROWN POINT FM.

AMADEUS BASIN

* oc a

LU CL

WARBURTON BASIN

ARROWIE BASIN

Fig. 3. Generalised stratigraphy of southern Eromanga Basin and underlying sedimentary basins showing stratigraphic occurrence of significant hydrocarbons and source rocks (flag symbol).

21 OQ *

Wlnton Fm.

Allaru

Mdst.

Wallumbilla

Fm.

Cadna-owie

Fm.

B l r k h e a d Fm.

M o o g a Fm.

100-,

80-

3 a. 60-

3 40-

11:

Coal

20-

!-» o (T) - l D- C.

1 1

1 E

75 100

0-20

0-10

95-100

0-5

80-97

N =2 1

0 - 15

1 1

1i 1

*T E

. 1

1 E

P 1

1 E

8 0 - 100

o- 1 0

0-10

75- 100

5-20

0-12

60-97

0-2

1-45

N= 3

N == 3

N= 5

ri E 2-3

N = 15

N == 9

Epsilon Fm.

P a t c h a w a r r a Fm.

o o > o

# Hutton S a n d s t o n e

' b a s a l Jurassic 1

100-

T o o l a c h e e Fm. •

R o s e n e a t h Sh.

60-

40-

20-

iI 1

i, E

r1 E

rl E

i f E

1 f E

i1 E

65 -99

0-30

0-10

25-90

3-70

3-10

15-90

7-82

1-8

25-52

43-72

2-5

15-55

33-82

3-10

o- 100

0-90

0-20

N = 12

N=3

25 o z

80-

N := 2 7

N == 3

N --6

N =1 5 5

310

A. J. KANTSLER, A. C. COOK & M. ZWIGULIS (Peera Peera Formation and equivalents) commonly contain are enriched in radioactive elements and that heat flow is an upper, organic-rich interval (e.g. Thomas, 1982; fig. 11). typically about 2.6 Heat Flow Units (HFU) as compared with Organic matter is dominated by inertinite and vitrinite average continental crust heat flow of 1.4 HFU (Gretener, although exinite contents are typically high (10-20%). Where 1981). Lower gradients (3-4°C/100 m) occur in the developed, this Upper Triassic coal swamp facies has fair to Patchawarra Trough, Pedirka, Simpson Desert and Arckaringa good source potential and could charge overlying basal Basins and imply regional heat flow of about 1.6-2.1 HFU which is still appreciably higher than average. Sass & Jurassic sandstones (e.g. Smyth & Saxby, 1981). Jurassic sediments contain a number of thin, rich source- Lachenbruch (1979) relate this to a stable continental crust rock intervals, particularly within the Birkhead Formation, with high average radioactivity and little contemporary but also at base Hutton Sandstone and unnamed basal tectonic activity of thermal origin. Seismicity in the Simpson Jurassic levels (including Poolowanna Formation) and locally Desert region of the Eromanga Basin is the result of strikewithin the Westbourne and Mooga Formations (Cook, 1982; slip motion along old fracture zones (Stewart & Mount, 1972). McKirdy, 1982a, b; Kantsler et al., 1983; Passmore & Boreham, Most other areas (e.g. northern Cooper Basin sub-province) this volume). Jurassic coals and organic matter are dominated have intermediate temperature gradients, which can be related by vitrinite (frequently perhydrous) and exinite (suberinite- to the depth and nature of underlying basement lithologies rich), although the basal Jurassic sequence is commonly (e.g. Pitt, 1982, this volume). inertinite-rich (Fig. 4; Cook, this volume). Differences in bulk thermal conductivity between the The overlying marine Cretaceous section is commonly Permian coal measures sequence, the Jurassic artesian aquifer organically lean with poor pyrolysis yields although fair to system and the Cretaceous mudstone sequence (i.e. sector good (occasionally excellent) source intervals containing mixed temperature gradients) are also likely to have a profound effect (marine and terrigenous) Type II/III organic matter are on organic maturation. Locally, thermal anomalies around developed locally within the Murta Member, Cadna-owie major basement faults (e.g. Polak & Ramsay, 1977) may also Formation, Wallumbilla Formation and Toolebuc Formation complicate the simple pattern of temperature gradient (McKirdy, 1982a, b\ Kantsler et al., 1983). Bituminite and variation described above. alginite commonly dominate the exinite content of these uppermost Jurassic and Early Cretaceous source rocks. ORGANIC MATURITY Previous studies (Tissot & Welte, 1978; Powell & Snowdon, The abundance of coal and land-plant organic matter 1980; Radke et al. 1980; Thomas, 1982) supported by throughout Jurassic and Upper Cretaceous sections of more recent work (Stainforth, 1984; Durand & Paratte, the EromangatheBasin allows accurate determination of organic 1984) conclude that the threshold of significant oil maturity by vitrinite reflectance. shallow marine to paralic generation from humic source rocks similar to those Lower Cretaceous sequence is lessThe amenable to VR maturation described above, occurs at maturation levels between 0.7% studies, but sufficient first-cycle vitrinite is generally present and 0.8% vitrinite reflectance (VR) with peak generation to permit reliable estimates of organic maturity. Data are occurring around 0.9% VR. However, some authors suggest for more than 90 wells in the southwestern Eromanga that rocks of mixed (Type II/Type III) or humic (Type III) available Basin and provide good control in the southern Cooper affinity are capable of generating significant amounts of light sub-province but a lesser degree of control elsewhere.Basin oil and condensate at relatively low levels of maturity (Lane Sample control generally extends from the Cretaceous down & Jackson, 1980; G. Demaison, pers. comm. 1980; Cook, 1982; Snowdon & Powell, 1982). Although the latter authors to the Permian, but is limited to the Cretaceous to basal have recently described initial naphthenic oil generation from Jurassic section over some structural highs which are bald resinite-rich source rocks at maturation level as low as 0.4% of Cooper Basin sediments. The number of samples examined VR, such oils and source rocks are not typical of the Ero- from the Eromanga part of the section is typically in the range manga Basin. Nonetheless, the widespread occurrence of five to fifteen for each well. Reflectance/depth profiles hydrocarbons or hydrocarbon indications within the lower constructed for each well, in conjunction with maturation part of the Early Cretaceous sequence (e.g. Murta Member, modelling techniques (Lopatin, 1971; Waples, 1980), have been Wyandra Sandstone Member, Caorikiana Sandstone) does used to develop generalisations concerning rank variation suggest the possibility of a low-maturity source rock play. within the basin. Further study is required to either mature or discount this Cooper Basin sub-province play concept. Rank variations in the Cooper Basin sub-province (Figs 5, GEOTHERMAL GRADIENTS 6) has been reviewed by Kantsler & Cook (1979) and Kantsler The present-day geothermal gradient range in both the et al. (1983). Areas of high maturity in the Eromanga Basin Eromanga Basin and underlying Carboniferous-Triassic basins correspond with the development of very high levels of coal is generally 3.0-6.0°C/100 m although Pitt (1982, this volume) rank within the underlying Cooper Basin (Fig. 7). Isoreflecrecords some extreme values outside these limits. Temperature tance surfaces are inversely related to basement in the data are derived mostly from bottom-hole temperatures which Nappamerri Trough in response to the combined effects of have been corrected for cooling by circulation of drilling high basement heat flow and an early high heat flow history. fluids. Temperature gradient variation is illustrated and A more positive relationship exists elsewhere (Kantsler & discussed by Senior & Habermehl (1980), Schwebel et al. Cook, 1979). (1980), Pitt (1982, this volume) and Kantsler et al. (1983) and The Winton Formation is immature (<0.6% VR) and, is summarised as follows: despite the presence of exinite-rich coals and mudstones (Fig. High gradients (greater than 5°C/100 m) occur on the 4), is considered unlikely to be a source of commercial shallower flanks of the Cooper sub-province (Senior & quantities of hydrocarbons. Locally developed source intervals Habermehl, 1980) and in and around the Nappamerri and containing humic (Type III) to mixed (Type II/III) or Tenappera Troughs (particularly in areas underlain by granitic sapropelic (Type II) organic matter in the Lower Cretaceous basement). Middleton (1979) has shown that these granites sequence are immature (VR<0.6%) to initially mature y

ORGANIC MATURATION

Fig. 5. Vitrinite reflectance, top Lower Cretaceous Cadna-owie Formation

R^, max).

311

312

R^ max).

A. J. KANTSLER, A. C. COOK & M. ZWIGULIS

Fig. 6. Vitrinite reflectance, top 'basal Jurassic' or base Jurassic unconformity

D KIDMAN 2 PELKETA 1 KIRBY 1 NAPPACOONGEE 2 I I riLPAREE PACKSADDLE | |^ f f f ^ I 1 0 ? * ? " ! A-1 MULGA 1 BURLEV »

NORTH

o o

o O > o

c ^

$ o z

731

KOPPERAMANNAWARBRECCAN HIGH

PATCHAWARRA TROUGH

GM I TREND

NAPPAMERRI TROUGH

MN TENNAPERRA TREND TROUGH

TOOIACHEE TREND

314

A. J. KANTSLER, A. C. COOK & M. ZWIGULIS

(VR=0.7%) over much of the basin (Fig. 5). Source rocks in the Murta Member-Cadna-owie Formation interval reach the 0.6-0.7% VR threshold for significant oil generation in the central Nappamerri Trough (Figs 5, 7) and may have sourced the condensate-like Murta Member oils found at many localities above the Cooper Basin. However, in the absence of substantial geochemical data, an origin from deeper, more mature source rocks via vertical migration and natural fractionation cannot be discounted. Middle Jurassic Birkhead Formation and Lower Jurassic or basal Hutton Sandstone source rocks reach maturation levels suitable for peak oil generation (VR=0.9%) in the Nappamerri Trough and the main axial depressions (Arrabury Trough, Ullenbury Depression, Cooper and Thomson Synclines) of the northern Cooper Basin (Figs 6, 7). Both intervals are sufficiently mature (VR>0.7%) over a large enough area to have sourced much of the oil and gas found in Eromanga Basin reservoirs around the margins of the Nappamerri Trough. Nonetheless, some doubt remains as to whether sufficient Jurassic source rock thickness is available to provide the necessary charge for some major discoveries (e.g. Strzelecki, Jackson), and more detailed work on potential hydrocarbon yield from these relatively thin source rock intervals is required. In the Patchawarra Trough and on the neighbouring Gidgealpa-Merrimelia-Innamincka (GMI) anticlinal trend, oils reservoired in the Birkhead Formation and Hutton Sandstone or basal Jurassic sandstones may have been generated locally but some could possibly be derived from underlying more mature Permian section or Permian section truncated down flank. In the case of the large anticlinal structures migration is facilitated locally by well developed crestal faulting (e.g. Merrimelia). No discussion of the origin of Eromanga Basin hydrocarbons in the Cooper Basin sub-province would be complete without some consideration of the possible role of underlying Permian source rocks. The Permian section is presently mature for oil generation throughout much of the Patchawarra and Tennapera Troughs, around the flanks of the Nappamerri Trough (Fig. 7) and over much of the northern Cooper Basin. Permian sediments in the northern Tenappera Trough, central Nappamerri Trough and central Cooper Syncline are currently mature for gas generation and have undoubtedly sourced most of the gas discovered to date in these areas.

Pedirka and Simpson Desert Basin sub-province The pattern of rank variation in the poorly drilled Pedirka and Simpson Desert sub-province (Fig. 8) appears relatively simple and to be controlled primarily by burial depth. Unlike the Cooper sub-province the range of temperature gradient variation is limited (3.1-4.3 °C/100 m). A plot of VR versus depth for all wells (Fig. 9) shows a relatively tight trend with little spread of data between wells. Maturation at top Cadna-owie Formation and top Poolowanna Formation (the latter is a major Eromanga Basin source interval in this region) shows an apparently simple concentric pattern (Figs 5, 6) which follows the saucer-like structure of the basin. The only significant departure from this pattern occurs over the McDills-Mt Crispe anticlinal trend where at least 400 m of uplift is indicated (Fig. 8). The nature of the Pedirka sub-province source rocks is described elsewhere (Smyth & Saxby, 1981; Smyth & Cameron, 1982; Thomas, 1982; McKirdy, 1982a, b\ Cook, 1982) but both major source bed intervals (i.e. Early Jurassic Poolowanna Formation and the Middle Triassic upper Peera Peera

Formation) contain predominantly humic organic matter. The Poolowanna Formation reaches, or exceeds, the maturation threshold of intense oil generation (0.7% VR) over the central part of the Simpson Desert sub-province where a potentially large generative zone exists (Figs 6, 8). Maturation at the base of the Mesozoic is slightly higher, indicating that the underlying Peera Peera Formation and the uppermost unit of the Early Permian Purni Formation (the latter containing abundant humic organic matter) are well placed to source oil or condensate. However, Permian source potential is restricted by the distribution of mature Purni Formation, which is limited to the western flank of the Poolowanna Trough. The margins of the basin are not generative due to low levels of maturity. Prospects in these areas must rely on updip migration from active generation zones in the trough axis but may also suffer from poor structural focus of migration paths, which is typical of such interior sag basins.

Arckaringa and Arrowie Basin sub-provinces Limited maturation data published by Moore (19826) from the Arckaringa sub-province and Youngs & Moorcroft (1982) for the Arrowie sub-province suggest that maturation levels along the shallow southern margin of the Eromanga Basin are rarely likely to rise above 0.5% VR. The thin veneer of Eromanga Basin section in this area is wholly immature. However, the basal 300 m of Permian section in the Boorthanna Trough of the Arckaringa Basin is mature and appears to have some source potential (Moore, 19826). Eromanga Basin plays in this area would, of necessity, rely on a Permian source. Similarly, exploration for Eromanga Basin objectives above the Arrowie Basin would rely on charge being derived from the underlying Cambrian sequence.

Discussion Representative plots of vitrinite reflectance (VR) versus depth (Fig. 10) show substantial differences in maturation trends between the various sub-provinces of the Eromanga Basin. These differences are particularly marked at Lower Jurassic levels where they reflect trends established in the underlying infra-basins or troughs. Comparison of average, present-day temperature gradients between these areas implies a strong positive correlation of rank variation with crustal heat flow. Changes in gradient or offsets of VR vs depth profiles occur locally, either at or near the base Jurassic unconformity level or near the base of the Cadna-owie Formation. Such changes in gradient are not uncommon (Dow, 1977; Robert, 1980) and can be related to any or all of the following factors: Unconformity. The extent of the offset at the base Jurassic unconformity reflects the extent of Mid-Late Triassic uplift and erosion (maximum 500 m). Differing thermal history. The lower sequence has a longer history of exposure to high temperatures and is possibly associated with high palaeo-heat flow, either at the initiation of basin subsidence or during subsequent orogenic events. Robert (1980) interprets offsets within a continuous depositional sequence (e.g. Tanbar North 1) as being caused by thermal events. Differing thermal conductivity.Temperature data available from drillstem tests and intermediate and final logging runs indicate dog-leg temperature gradients throughout the Cooper sub-province. High gradients (e.g up to 8°C/100 m in Burley 1) commonly occur within the coal-rich (multiple seams up to 20 m thick) Permian sequence and reflect very low thermal conductivities (i.e. Permian coals act as a thermal blanket).

M T CRISPE I

WrrCHERRE I

BIRDSVILLE TRACK RIDG F

POOUOWANNA TROUGH

MT CRISPE TRFND PURNI I

MOKARI I

MACUMBA I

WALKANDI I

POOLDWANNA I

KUNCHERINNA

PANDIEBURRA I

WEST

EAST CRETACEOUS ORGANIC MATURATION

Fig. 8. West-east cross-section, Pedirka and Simpson Desert sub-province showing isoreflectanco lines

MC D I L L S

Lower Palaeozoic

Kilometres

R , max). 315

A. J. KANTSLER, A. C. COOK & M. ZWIGULIS

316

Conductivity contrasts between the Cretaceous marine mudstone and siltstone sequence, Jurassic sandstone aquifer sequence and Permian coal measures sequence may give rise to at least some of the variations seen on maturation profiles. Differing vitrinite type. At least some of the variation observed in many maturation profiles can be related to variations in vitrinite type either through floral evolution (e.g Cook, 1981; Thomas, 1982) or through the quality (and

quantity) of vitrinite available for measurement. The latter problem is particularly severe within the Early Cretaceous marine sequence but is unlikely to be wholly responsible for offsets in maturation profiles such as occur at Tanbar North 1 (Fig. 10). Local thermal perturbations. Thermal anomalies associated with fluid migration around deep-seated basement faults (Buntebarth & Schopper, 1976; Polak & Ramsay, 1977) or

1000

u. Q CD in Q> L. a>

E 2000 Q. Q> Q

3000

COLSON MACUMBA MOKARI PANDIEBURRA P00L0WANNA POONARUNNA PURNI PUTAMURDIE WITCHERRIE

0-2

0-3

0-4

0-5

J_L

0 - 6 0-7 0 - 8 0 - 9 10

2'0

Vitrinite Reflectance, % R 0 max

Fig. 9. Vitrinite reflectance (°/o R 0 max) versus depth below derrick floor (BDF) for wells in Pedirka and Simpson Desert sub-province.

317 ORGANIC MATURATION within an active aquifer system (Gretener, 1981) may also give a maturation 'high' persisting into the Lower Cretaceous rise to local perturbations on maturation profiles. Such section (Kantsler et al., 1983). The common presence of offsets anomalies are inferred on some structurally high trends in or changes in gradient of plots of VR vs depth at or near the Cooper sub-province (Kantsler & Cook, 1979). the base of the Eromanga sequence (Fig. 10) in this region suggests that this Permian-Triassic phase of high heat flow was laterally widespread, although its influence appears to THERMAL HISTORY diminish towards the flanks of the major depocentres (e.g. High palaeo-heat flow in the Nappamerri Trough of the Cuttapirrie 1, Fig. 10). Likely causes of this high initial heat Cooper sub-province (e.g. Burley 1, Figs 5, 7, 10), has created flow are upwellng of hot asthenosphere or elevation of the

Top Transition Beds —

Base Jurassic Unconformity

1000

2000

3000

0-4

0-5

0-7

1-0

1-5

2-0

Vitrinite Reflectance, % R Q max

Fig. 10 Representative vitrinite reflectance versus depth profiles for various sub-provinces of the Eromanga Basin. Number in brackets is present-day geothermal gradient (°C/100 m).

318

A. J. KANTSLER, A. C. COOK & M. ZWIGULIS

geotherm following crustal extension (Sclater & Christie, 1980; Middleton, 1980) superimposed upon high levels of crustal heat production (Sass & Lachenbruch, 1979; Middleton, 1979). In the Nappamerri Trough, early high heat flow was probably enhanced by initially high levels of radiogenic decay of ?Carboniferous basement granites. Most Eromanga Basin maturation profiles are characterised by a relatively uniform increase of reflectance with depth in the Cretaceous mudstone/siltstone sequence. This suggests stable heat flow conditions over much of the basin throughout the Cretaceous and Tertiary. Profiles of maturation within the Jurassic strata are more variable and this may reflect the combined effects of an attenuating thermal regime, changes in bulk thermal conductivity and possible artesian effects. However, the apparently large offsets evident in the Mid-Upper Jurassic sections in some wells (e.g. Tanbar North, Fig. 10) are difficult to explain adequately in these terms as they imply a prolonged state of thermal disequilibrium between successive stratigraphic horizons. A syndepositional thermal event is also unlikely to be responsible since dissipation of heat so close to the surface would not allow adequate build-up of formation temperatures within the time available. Furthermore, none of the Cooper or Eromanga coals examined shows any sign of incipient mesophase (microcoke) development, thus offering no support for the concept of a short-lived thermal event associated with igneous intrusion. As the non-erosional offsets generally occur in wells with relatively limited sample control within the Jurassic-Cretaceous sequence, they are considered, in part, to be possible artefacts of vitrinite type and sample distribution. They are not discussed further in this review although the problem warrants a more thorough investigation through more detailed sampling.

In the absence of firmly established porosity reduction functions for major lithologies within the Eromanga Basin, and of a clear understanding of heat-flow history, thermal histories were approximated using an empirically derived, attenuating thermal regime. In view of the protracted quiescence of the basin, present-day temperature gradients were considered to be an expression of possible minimum heatflow. Temperature gradients within active present-day rifts (e.g. Rhine Graben) range up to 8°C/100 m and reflect, by analogy, possible maximum heat flow (approx. 3.5 HFU) during an incipient rift phase. A first pass at modelling measured VR data from individual wells was made using a constant (present-day) geothermal gradient throughout the burial history. Results from the Patchawarra and Poolowanna Troughs (Figs 11, 12) suggest

MATURATION MODELLING Although the evolution of intracratonic basins remains conjectural (Fischer, 1975; Bally & Snelson, 1980), subsidence linked with cooling of sub-crustal thermal anomalies is widely accepted as a principal mechanism in their formation (Sleep & Snell, 1976; Turcotte & Ahern, 1977; McKenzie, 1978; Sclater & Christie, 1980; Middleton, 1980). The data above illustrate the possible association of Eromanga Basin subsidence with an attenuating thermal regime (cf. Pitt 1982, this volume). Deposition of the underlying Permian-Triassic sequences, particularly the Cooper Basin sequence, may represent faultcontrolled initial subsidence during an incipient (but subsequently aborted) rift phase, which was associated with high heat flow. Integrated burial and heat flow histories are now widely used to estimate levels of organic maturation within sedimentary basins via the technique described by Lopatin (1971), Waples (1980), Wright (1980) and Falvey & Deighton (1982). Most such models use simple back-stripped well or seismic data linked to a constant temperature gradient history. However, recent developments (Sclater & Christie, 1980; Falvey & Deighton, 1982; Stainforth, 1984) suggest that the use of decompacted sediment thicknesses, together with time-variant heat flow models and porosity-dependent thermal conductivities permit a far more rigorous evaluation of thermal history and, therefore, of timing of hydrocarbon generation. In general, oil and gas generation are predicted to occur earlier when all such effects are fully accounted for. This is because decompacted burial paths follow a deeper (higher temperature) route, and because the thermal conductivity of porous rocks increases with compaction (resulting in higher near-surface temperatures).

Fig. 11. Vitrinite reflectance vs depth plot, modelled trends, and subsidence/maturation history Cuttapirrie 1 (Patchawarra Trough).

319

ORGANIC MATURATION Vitrinite Reflectance, % R 0 max

Fig. 12. Vitrinite reflectance vs depth plot, modelled trends, and subsidence/maturation history Poolowanna 1 (Poolowanna Trough).

that maturation of the entire sedimentary section hese relatively 'cool' areas can generally be modelled within the constraints of the present-day thermal regime. Results from the margin of the 'hot' Nappamerri Trough show that temperature gradients considerably greater than present-day (approximately 5°C/100 m) are required to satisfactorily model maturation (Fig. 13). The VR data can also be modelled using a thermal history which declines from a maximum temperature gradient of 10°C/100 m during a postulated Late Carboniferous-Early Permian thermal event. Alternatively, satisfactory results may be obtained by including a Late Cretaceous-Tertiary thermal event in the maturation history (temperature gradients up to 6.8°C/100 m). However, these modelled results should be considered subjectively given the inherent limitations of the Lopatin technique (cf. Tissot, 1984). Differences between modelled and measured data in the

Fig. 13. Vitrinite reflectance vs depth plot, modelled trends, and subsidence/maturation history Kirby 1 (Nappamerri Trough).

Eromanga Basin section (Fig. 13) may reflect higher thermal conductivities and a lower temperature gradient within this part of the sedimentary package. Each of the wells modelled is representative of a major Eromanga Basin depositional trough, and the results therefore suggest a possible latest time for the onset of hydrocarbon generation in these areas (Figs 11-13). When the model constraints outlined previously are also considered (e.g. Stainforth, 1984) these onset times will be brought forward (e.g. Poolowanna 1, Fig. 12). However, modelling studies suggest that, irrespective of the thermal history used, most Eromanga Basin hydrocarbon generation post-dates the final phase of Cenomanian subsidence. In the central Nappamerri Trough (Burley-Kirby area), maturation modelling suggests that basal Hutton (Early

320

A. J. KANTSLER, A. C. COOK & M. ZWIGULIS

Jurassic) source rocks passed through the oil window (0.6-1.2'% VR) at latest in the Late Cretaceous-Mid Tertiary (Fig. 13). Birkhead Formation source rocks remained within this window for much of the Tertiary and are now nearing the final phase of active oil generation. Murta Member/ Transition beds (Cadna-owie Formation) source rocks appear to have remained initially oil-mature since the Early Tertiary. In the central Patchawarra and Poolowanna Troughs significant oil generation (0.8-1.2% VR) from basal Hutton or basal Jurassic source rocks occurred in Late Tertiary to Recent times and generation is continuing at present. Birkhead Formation source rocks apparently reached initial oil maturity in the Mid-Tertiary but are yet to enter the zone of significant oil generation. Upper Jurassic/Lower Cretaceous source rocks are immature. In the central Poolowanna Trough, Upper Triassic and Lower Jurassic source rocks reached initial maturity, at latest, in the Early Tertiary and entered the zone of significant oil generation, at latest, in the Late Tertiary. Maturation levels decrease rapidly away from the centre of the trough resulting in progressive (marginward) delay in the onset of oil generation.

DISCUSSION

Many anticlinal structures in the Eromanga Basin developed as drapes over older pre-Permian and Permian-Triassic fault blocks, some of which have a history of syndepositional growth. A major phase of left-lateral, wrench-induced folding took place in the Early Tertiary (Late Paleocene-Late Eocene) and created many of the very large anticlines found in the northern Cooper sub-province while enhancing structures elsewhere. A subsequent phase of generally gentle warping took place in the Mid Eocene-Mid Oligocene (Wopfner et al., Senior et al., 1978; Moore & Pitt, 1984). Minimal Late Cretaceous-Tertiary subsidence, and in some cases uplift, has led to a Mid-Late Tertiary slowing of the maturation process with a probable subsequent decrease in the rate of hydrocarbon generation. In view of their pre-oil generation development, most earlyformed structures are well placed in time to receive charge from a Jurassic source. However, relatively late-formed structures on the margins of the major troughs, in particular the mature-overmature Nappamerri Trough, have an oil charge risk because their development may post-date significant phases of oil generation and migration. All structures in the less mature parts of the Eromanga Basin have an inherent element of risk due to insufficient charge having been generated from erratically developed Jurassic source rocks. Few Jurassic reservoirs are filled to spill-point suggesting that generation is incomplete or that significant generation predated Early Tertiary structural enhancement. However, other factors such as inadequate source-rock volume, dispersal of hydrocarbons during migration, gas flushing, poor reservoir, or ineffective cap rocks may also have influenced charge volume (Kantsler et al., 1983). Play concepts which rely on Late Tertiary-Recent charge may be of higher risk if the required migration paths are unsealed and/or opposed to regional artesian flow patterns, which were established following Early to Mid Tertiary uplift (see Habermehl, this volume). Eromanga Basin oils are light (39-56° API), apparently mature, paraffinic to naphthenic crudes, some are condensatelike, others have a high wax content (McKirdy, 1982a, b\ Kantsler et al., 1983). A lack of gasoline range hydrocarbons in some oils is evidence of water washing either during

migration or following entrapment. Most host rocks are sufficiently mature (VR=0.6 to 0.8%) for some in situ generation but the geochemical maturity of many oils suggests generation and migration from more mature source rocks (VR=0.8%) in the trough axes. Oils found in the immature to early mature Murta Member are geochemically distinct from the more waxy crudes found lower in the sequence (McKirdy, 1982a, b\ Kantsler et al., 1983). Despite their ostensible maturity, the Murta crudes may be sourced locally from initially mature source rocks containing labile exinite, including bacterially-degraded organic matter and a minor marine component (Cook, 1982; McKirdy, 1982a, b\ Kantsler et al., 1983). As such, they may represent examples of early light oil and condensate generation and therefore constitute a separate play to that of the underlying Early and Middle Jurassic source/reservoir pairs. Alternatively, they may be products of natural fractionation occurring during vertical migration (cf. Tissot, 1984) which is terminated beneath the regional Bulldog Shale-Wallumbilla Formation seal sequence. Some reservoirs such as the Nappamerri Formation and Hutton Sandstone at Merrimelia and the Hutton Sandstone at Jackson, directly overlie, or are connected via faults, stratigraphic convergence or truncation, to Permian reservoir or carrier beds. In the absence of definitive oil to source correlations, a Permian origin, or contribution, cannot be ignored for some Eromanga oil accumulations. Gas found in Jurassic reservoirs is moderately wet to dry. Compositional data and isotopic character (Rigby & Smith, 1981) imply a Permian source at localities such as Namur and Wackett (McKirdy, 1982; Kantsler et al., 1983) but elsewhere the data are more equivocal and either a Jurassic or Permian source is possible. A Permian source is generally favoured because of the more extensive and mature source sequence available.

CONCLUSIONS

1. Mature Jurassic-Cretaceous source sequences are found in major depocentres of the Eromanga Basin above the Nappamerri Trough, Cooper Syncline and Poolowanna Trough. 2. Modelling of maturation data suggests a history of; oil generation from mature Lower-Middle Jurassic source rocks throughout the Late Cretaceous and Tertiary. Maturity considerations favour most generation from the more mature Nappamerri Trough and Cooper Syncline. An apparent decrease in the rate of maturation and generation during the Mid-Late Tertiary may influence charge volumes for lateformed structures but insufficient data are presently available to further evaluate this concept. 3. As the timing of maturation may be critical in establishing the validity of some plays, it is important that regional maturation/generation models be refined in order to permit a proper rating of prospects. 4. A need exists for definitive oil to source correlation studies, together with regional source rock screening surveys, to elucidate the role of Jurassic and Early Cretaceous source rocks.

ACKNOWLEDGEMENTS

The authors are indebted to Delhi Petroleum Pty Ltd and Shell Development (Australia) Pty Ltd for their assistance with this review and to Delhi's joint venture partners for permission to publish. Peter Moore, David Gravestock and David McKirdy are thanked for their generous and constructive editorial comment.

ORGANIC MATURATION

321

REFERENCES

ARMSTRONG, J. D. & BARR, T. M., 1982: The Eromanga Basin; in

Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, Summary Papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 2 0 - 4 2 . BALLY, A. W. & SNELSON, S., 1980: Realms of subsidence; in Miall, A. D. (ed.) Facts and Principles of World Petroleum Occurrence, Can. Soc. Pet. Geol.. Mem. 6, Calgary, 9-94. BARR, T. M. & YOUNGS, B. C., 1981: Cuttapirrie-1 an oil discovery in the Early Jurassic of the Eromanga Basin. A PEA J. 21(1), 60-70 BOWERING, O. J. W,, 1981: Hydrodynamics and hydrocarbon migration—a model for the Eromanga Basin. A PEA J. 22(1), 227-36. BUNTEBARTH, G . & SCHOPPER, J. R., 1976: Heat flow caused by water migration along faults in dependence on petrophysical parameters. Proc. Int. Congr. Thermal Waters, Geothermal Energy and Volcanism of the Mediterranean Area, Athens, 1976, 2, 41-9. COOK, A. C., 1982: Organic facies in the Eromanga Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, Summary Papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 2 3 4 - 5 7 . Dow, W. G., 1977: Kerogen studies and geological interpretations. J. Geochem. Explor. 7, 77-99. DURAND, B. & PARATTE, M., 1984: Oil potential of coals: a geochemical approach in Brooks, J. (ed.) Petroleum Geochemistry and Exploration of Europe, Geol. Soc. (Lond.) Spec. Publ. 12, Blackwell, Oxford, 255-65. FALVEY, D. A . & DEIGHTON, I., 1982: Recent advances in burial and thermal history analysis. A PEA J. 22(1), 6 5 - 8 1 . FISCHER, A. G., 1975: Origin and growth of basins, in Fischer, A. G. & Judson, S. (eds) Petroleum and Global Tectonics, Princeton Univ. Press, Princeton, 47-79. GRAVESTOCK, D. I., 1982: Jurassic to Lower Cretaceous stratigraphy of the Eromanga Basin, South Australia—problems and progress in sub-surface correlation, in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, Summary Papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 79-91. GRETENER, P. E., 1981: Geothermics: using temperature in hydrocarbon exploration. Am. Assoc. Petrol. Geol., Education Course Note Series, 17. Tulsa. KANTSLER, A . J. & COOK, A . C., 1979: Rank variation in the Cooper and Eromanga Basins, central Australia. Delhi Petroleum Pty Ltd (unpubl.). KANTSLER, A . J., PRUDENCE, T. J. C., COOK, A . C . & ZWIGULIS,

M., 1983: Hydrocarbon habitat of the Cooper/Eromanga Basin, A PEA J. 23(1), 75-92. LANE, F. H. & JACKSON, K. S., 1980: Controls on occurrence of oil and gas in the Beaufort-Mackenzie Basin; in Miall, A. D. (ed.) Facts and Principles of World Petroleum Occurrence; Canadian Soc. Pet. Geol. Mem. 6, Calgary, 4 8 9 - 5 0 7 . LOPATIN, N . V., 1971: Temperature and geologic time as factors in coalification. Akad. Nauk SSSR Izv. Ser. Geol., 3, 9 5 - 1 0 6 . (Russian). English Translation, N. H. Bostick, Illinois State. Geol. Surv. 1972. MCKENZIE, D., 1978: Some remarks on the development of sedimentary basins. Earth Planet. Sci. Lett. 40, 25-32. MCKIRDY, D, M„ 1982a: Aspects of the source rock and petroleum geochemistry of the Eromanga Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, Summary Papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 258-59. MCKIRDY, D. M . , 19826: Petroleum geochemistry and source-rock potential of the Cooper Basin and superjacent Eromanga Basin. Delhi Petroleum Pty Ltd (unpubl.).

MIDDLETON, M . F., 1979: Heat flow in the Moomba, Big Lake and

Toolachee gas fields of the Cooper Basin and implications for hydrocarbon maturation. Aust. Soc. Explor. Geophys. Bull. 10, 149-55. MIDDLETON, M . F., 1980: A model of intracratonic basin formation, entailing deep crustal metamorphism. Geophys. J. R. Astronom. Soc. 62, 1-14. MOORE, P. S., 1982A: Mesozoic geology of the Simpson Desert region, northern South Australia; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, Summary Papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 46-57. MOORE, P. S., 19826: Hydrocarbon potential of the Arckaringa Region, central South Australia. A PEA J. 22(1), 237-53. MOORE, P. S. & PITT, G. M., 1982: Cretaceous of the southwestern Eromanga Basin: stratigraphy, facies variations and petroleum potential; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, Summary Papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 127-44. MOORE, P. S. & PITT, G. M . , 1984: Cretaceous of the Eromanga Basin—implications for hydrocarbon exploration. A PEA J. 24(1), 358-76. MOUNT, T. J., 1982: Geology of the Dullingari Murta oilfield; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, Summary Papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 356-74. PASSMORE, V. L. & BOREHAM, C. J., 1986: Source rock evaluation and maturation history of the central Eromanga Basin; in This volume. PITT, G. M., 1982: Geothermal gradients in the Eromanga/Cooper Basins; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, Summary Papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide. 262-83. POLAK, E. J. & RAMSAY, D. C., 1977: Canaway Ridge, Queensland, geophysical survey, 1973. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1977/29 (unpubl.). POLL, J. J. K., 1981: The significance of the southwest Eromanga Basin oil and gas discoveries (central Australia). A PEA J. 21(2), 33-8. PORTER, C. R., 1978: The Pedirka Basin—a preliminary exploration review; in Onshore South Australia, A PEA Seminar, Australian Mineral Foundation (unpubl.). POWELL, T. G. & SNOWDON, L. R., 1980: Geochemical controls on hydrocarbon generation in Canadian sedimentary basins; in Miall, A. D. (ed.) Facts and Principles of World Oil Occurrence. Can. Soc. Pet. Geol. Mem. 6, Calgary. RADKE, M . , SCHAEFER, R. G., LEYTHAEUSER, D. & TEICHMULLER,

M., 1980: Composition of soluble organic matter in coals: relation to rank and liptinite fluorescence. Geochim. Cosmochim. Acta, 44, 1787-860. ROBERT, P., 1980: The optical evolution of kerogen and geothermal histories applied to oil and gas exploration; in Durand, B. (ed.) Kerogen—insoluble organic matter from sedimentary rocks, Editions Technip, Paris, 385-414. SASS, J. H. & LACHENBRUCH, A. H., 1979: Thermal regime of the Australian continental crust; in McElhinny, M. W. (ed.) The Earth: its origin, structure and evolution, Academic Press, London, 301-51. SCHWEBEL, D. A., DEVINE, S. B. & RILEY, M., 1980: Source maturity and gas composition relationships in the southern Cooper Basin: A PEA J. 20(1), 191-200. SCLATER, J. G. & CHRISTIE, P. A. F., 1980: Continental stretching: an explanation of the post-Mid-Cretaceous subsidence of the central North Sea Basin. J. Geophys. Res. 85, 3711-39. SENIOR, B. R., M O N D , A. & HARRISON, P. L., 1978: Geology of the Eromanga Basin. Aust. Bur. Miner. Resour. Geol. Geophys. Bull. 167.

322

A. J. KANTSLER, A. C. COOK & M. ZWIGULIS

SENIOR, B. R. & HABERMEHL, M. A., 1980: Structure, hydrodynamics

and hydrocarbon potential, central Eromanga Basin, Queensland, Australia. BMR J. Aust. Geol. Geophys. 5, 47-55. SLEEP, N. H., 1976: Platform subsidence mechanism and 'eustatic' sea-level changes. Tectonophysics, 36, 45-56. SLEEP, N. H. & SNELL, N. S., 1976: Thermal contraction and flexure of mid-continent and Atlantic marginal basins. Geophys. J. R. Astronom. Soc. 24, 125-54. SMITH, B. L., 1983: Seismic investigation of the Merrimelia Field. A PEA J. 23(1), 192-202. SMYTH, M., 1979: Hydrocarbon generation in the Fly Lake—Brolga area of the Cooper Basin. APEA J. 19(1), 108-14. SMYTH, M. & SAXBY, J. D., 1981: Organic petrology and geochemistry of source rocks in Pedirka-Simpson Desert Basins, central Australia; APEA J. 21(1), 187-99. SMYTH, M. & CAMERON, M., 1982: Organic petrology and source rock potential of sediments in the Eromanga Basin, South Australia. Int. J. Coal. Geol. 1, 263-81. SNOWDON, L. R. & POWELL, T. G., 1982: Immature oil and condensate, modification of hydrocarbon generation model for terrestrial organic matter. Am. Assoc. Pet. Geol. Bull. 66, 775-88. SPRIGG, R. C., 1961: On the structural evolution of the Great Artesian Basin. APEA J. 1(1), 37-56. STAINFORTH, J. G., 1984: Gippsland hydrocarbons—a perspective from the basin edge. APEA J. 24(1), 91-100. STEWART, I. C. F. & MOUNT, T. J., 1972: Earthquake mechanisms

in South Australia in relation to plate tectonics. J. Geol. Soc. Aust. 19, 41-52.

THOMAS, B. M., 1982: Land-plant source rocks for oil and their

significance in Australian basins. APEA J. 22(1), 164-78.

TISSOT, B. P., 1984: Recent advances in petroleum geochemistry

applied to hydrocarbon exploration. Am. Assoc. Pet. Geol. Bull. 68, 545-63. and Occurrence. Springer-Verlag, Berlin-Heidelberg. TURCOTTE, D. L. & AHERN, J. L., 1977: On the thermal and subsidence history of sedimentary basins. J. Geophys. Res. 82, TISSOT, B. P. & WELTE, D. H., 1978: Petroleum Formation

3762-66. WAPLES, D. W., 1980: Time and temperature in petroleum formation:

Application of Lopatin's method to petroleum exploration. Am. Assoc. Pet. Geol. Bull. 64, 916-26. M. J., 1982: Late Triassic and Early Jurassic sedimentation in the Great Artesian Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, Summary Papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 59-67. WOPFNER, H., 1960: On some structural developments in the central part of the Great Australian Artesian Basin. R. Soc. S. Aust. Trans. 83, 179-93. WOPFNER, H., CALLEN, R. & HARRIS, W. K., 1974: The Lower Tertiary Eyre Formation of the southwestern Great Artesian Basin. J. Geol. Soc. Aust. 21(1), 17-51. W R I G H T , N . J. R., 1980: Time, temperature and organic maturation—the evolution of rank within a sedimentary pile. J. Pet. Geol. 2, 411-25. YOUNGS, B. C. & MOORCROFT, E., 1982: The petroleum potential of the eastern Arrowie Basin and Frome Embayment., APEA J. 22(1), 80-101. WILTSHIRE,

Geological Society of Australia Special Publication No. 12, 323-351

Geothermal gradients, geothermal histories and the timing of thermal maturation in the Eromanga-Cooper Basins G. M. Pitt

Western Mining Corporation Ltd, 168 Greenhill Rd, Parkside, S.A. 5063.

ABSTRACT

The geothermal gradient, a measure of temperature increase or decrease through the stratigraphic section, may vary considerably within that section due to factors such as lithology, degree of compaction, porosity and nature of the pore-fluid. Data from different sources should therefore be integrated with caution, or recalculated on a standard basis. Gradients from temperature measurements in different stratigraphic units should not be integrated. EromangaCooper Basin data are available from wireline logging runs and drill stem tests. Regional maps have been compiled showing gradients from Cadna-owie Formation to surface, 'basement' to Cadna-owie Formation and 'basement' to surface. Six zones recognised on the latter map compilation show a clear relationship to regional tectonic structure. Hydrocarbon maturation modelling has been carried out on four drilled sections, adjusting the thermal history to obtain a close match with the observed vitrinite reflectance curve. Timing of progressive maturation in combination with structural timing, is a vital aspect in the assessment of prospectivity of undrilled structures in the Eromanga-Cooper Basin region. Average palaeogradients of 2.5 °C/100 m (Beanbush 1) to 4.5-5.5 °C/100 m at Moomba 3 are required to model current vitrinite curves. Generation thresholds for a particular unit are reached in significantly different epochs. Variations in predominant maceral types within stratigraphic units alter levels of maturity required by various generation thresholds, thus maceral type is incorporated into the maturation model. In areas with higher palaeogradients, e.g. at Moomba 3 and Curalle 1, significant oil generation preceded Tertiary structuring. However, in 'cooler' areas, exemplified by Beanbush 1 and Tartulla 1, significant oil generation post-dated Tertiary deformation. Thus, in such areas, Tertiary structures should be regarded as prospective.

INTRODUCTION

In the Cooper and Eromanga Basins of central Australia (Fig. 1), oil and gas are generated and reservoired at stratigraphic levels ranging from the Early Permian to the Early Cretaceous. As recently discussed by Kantsler et al. (1983, this volume), the area has undergone a complex thermal history, with a resultant complexity of organic maturity trends. The highest levels of organic maturation are attained in the Nappamerri Trough, such as in the Moomba and Big Lake Fields, where dry gas is now generated from, and reservoired in, Early Permian formations (Schwebel et al., 1980). There, even the Neocomian Cadna-owie Formation is generative, with vitrinite reflectances of 0.60 to 0.70% Ro. By contrast, a different thermal history clearly pertained in the adjacent Patchawarra Trough. In the Tirrawarra Field, for example, wet gas and oil are generated from, and reservoired within, the same Early Permian units as those at Moomba and Big Lake. The youngest producible hydrocarbon accumulation currently recorded occurs in the Strzelecki Field, within the Early Cretaceous Coorikiana Sandstone (Moore & Pitt, 1982). Gas flowed from this formation at the rate of 9910 m per day in Strzelecki 8, at a depth of 1006 m. 3

Hydrocarbon shows have commonly been recorded in stratigraphic and petroleum exploration wells in the Eromanga Basin from still higher stratigraphic levels. For example, Ozimic (1981) reported a 'kerosene-like fluid' flowing from the Albian Toolebuc Formation in a number of shallow wells, drilled by the Bureau of Mineral Resources, Geology and Geophysics (BMR). The origin of such shows in the late Early Cretaceous sequence is contentious, for, by conventional measure, the section is organically immature. The stratigraphic range of the oil and gas 'window', is now reasonably fully appreciated. Consequently, the importance of further vitrinite reflectance and thermal studies in determining the degree of local or regional maturation (or lack thereof) is diminished, with respect to the major areas of exploration in the Cooper-Eromanga Basins. However, there is an increasing awareness of the critical nature of the relationship between timing of structuring and that of generation. Such a relationship can be elucidated from an incorporation of maturation modelling and observed reflectance data. Vitrinite reflectance distribution and its significance have been considered in detail by Cook, Kantsler and others (see

324

G. M. PITT

Fig. 1. Location of study area.

Kantsler et al., 1983, and references therein). Temperature studies, other than for production engineering purposes, have been less adequately covered: thermal gradient maps of the central southern Cooper Basin are presented in Kantsler et al. (1978), Middleton (1979a) and Schwebel et al. (1980). A simplified, more regional plan also appeared in Kantsler et al. (1983). Maps discussed in the present paper were originally presented by Pitt (1982). The purposes in writing this paper are twofold: to present a comprehensive regional survey of geothermal gradients; and to model the maturation profiles of four Eromanga-Cooper Basin petroleum wells in order to infer their respective thermal and generative histories. A detailed mathematical treatment such as that of Falvey & Deighton (1982) is avoided. However various practical aspects, such as estimation of formation temperature, compilation of burial curves and Lopatin modelling are discussed in detail in order to stimulate consideration and application of these concepts by other explorationists in the Cooper-Eromanga Basin region. The reader is referred to Gretener (1981) for a useful coverage of the subject, in which the theory, instrumentation and applications of geothermics are discussed in depth.

NATURE AND MEASUREMENT OF THE GEOTHERMAL GRADIENT

In simplest terms, the geothermal gradient is the rate of temperature increase or decrease, up or down through the crust. It is influenced by many factors, most notably the degree of heat input at the base of the rock column under study, the amount of radiogenic heat within that rock column and

the thermal conductivity of the section. The latter factor varies inter alia according to lithology, temperature, degree of compaction, and porosity. The gradient is proportional to the heat flow and inversely proportional to the thermal conductivity; for instance, water is a poor conductor relative to the rock mass, thus a porous, friable, aquifer sand will have a higher gradient than a tight, silicified sandstone—though not, of course, a necessarily higher temperature. The rock column is not homogeneous; it contains strata which vary by way of lithotype, degree of compaction and porosity. Thus, like the integrated sonic log or velocity survey profile, the thermal gradient varies considerably up the drilled section. In fact, sonic velocity maps of the Cooper Basin region show regional variations sympathetic with those of thermal gradients (K. R. Seedsman, pers. comm., 1983). The reason apparently lies in the similar physical nature of heat and sound transmission as a consequence of which a good conductor of heat is likely to be a good conductor of sonic waves as well. Rocks with higher inter- and intracrystalline strength are likely to be both better thermal and sonic conductors. This trend is apparent in Figure 2, redrawn from Gretener (1981). Basement rocks are typically better thermal conductors than sedimentary rocks of overlying basins. Porosity has the effect of decreasing conductivity; pore-filling fluids are very poor conductors. Within a basinal sequence, and given a constant heat input, the geothermal gradient will be greater in soft shale, coal, or porous, poorly lithified sandstone for instance, and less in indurated siltstone, quartzite or carbonate. Conductivity differences are also the partial cause of the marked difference between typical basinal geothermal gradients (2.0-4.5 °C/100 m) and those of shield or cratonic areas (0.8-1.5 °C/100 m) (Gretener, 1981). However, the influence of varying basin thicknesses on geothermal gradients should also be appreciated. Consider a thick, homogeneous basinal sequence, overlying crystalline basement (Fig. 3). O

CONDUCTIVITY ( m c a l / c m s e c °C) S 10 15

DUNITE GRANITE

SANDSTONE DOLOMITE LIMESTONE SHALE SLATE

COAL CLAY DEEP-SEA SEDIMENTS (PISTON CORES)

i—i

•

ANHYDRITE HALITE

•

GYPSUM

ICE

• •

WATER OIL GAS COPPER

~900

meal /cm •<

PK 299

Fig. 2. Typical conductivities for range of common rock types. Note low values for coal and pore-fluids (oil, water, gas); these materials are effective insulators. Redrawn from Gretener (1981).

325

MATURATION TIMING AND GEOTHERMAL GRADIENTS REGIONAL

BASIN

GRADIENT 15

/ -

10 5

VERTICAL GRADIENT

Fig. 3. Isotherm behaviour over progressively shallowing, homogeneous basinal sequence, with 'basement' outcrop beyond basin margin. In this example, 'basement' conductivity is considerably greater than that of basinal sequence.

Conductivity differences will result in higher basin and lower basement gradients. As the basement shallows towards the basin margin, the accompanying lower gradients will also be at shallower depths. However, the gradients within the thinner basinal sequence will be even greater than those of the equivalent thicker basinward sequence. The reason for this is shown geometrically in Figure 3. It may be rationalised by observing that, on the basin margin, the more thermally conductive basement rocks conduct more heat to shallower depths. Thereafter a steeper gradient must pertain in the overlying, thin, basin sequence, due to the rapid temperature drop to ambient surface temperatures. Consequently, the geothermal gradient within a homogeneous basinal infill will increase as the underlying basement shallows, although the conductivities and other factors remain constant. Beyond the basin edge, in shield areas, gradients drop sharply as they derive from measurements taken in basement which does not have the insulating cover, provided elsewhere by a basinal sequence. In such shield areas heat flow may be high, but because it is unimpeded by any insulating cover and because conductivity is high, gradients are low and moderate temperatures are encountered at considerable depths. Significant anomalies in basinal areas are defined by a departure from the behaviour described above. Thus the increase of gradients towards the southeastern margin of the Cooper Basin, in the Jackson region (Fig. 4) are not anomalous in this regard.

Inhomogeneous basement adds a further complicating dimension. In general, less conductive basment will result in a lower heat input at the base of the basinal succession, resulting in lowered thermal gradients of the overlying sediments. More-highly conductive basement (granite) at Moomba may be a factor in the Moomba-Big Lake thermal high, whether the heat source within the granite be deepseated, or radiogenic, or both.

Calculation of the thermal gradient

Geothermal gradients are calculated quite simply by dividing the temperature change over a vertical section by the length of that section: T, - T Grad = —

°C/100 m

where T, is the formation temperature at depth D, and T is the formation temperature at depth D . The application of this equation is unambiguous at depth, for 'interval' gradients. For 'overall' gradients, from ground surface to the base of the sequence under consideration, T,is the formation temperature at D,, the total depth of the borehole; while T is ambient temperature at ground surface where, of course, D, = 0. 2

K> ON 135°00l 2 3 ° 0 0'

138°00'

1 4 5 ° 0 0* 2 3° 0 0'

26°00'

2 6° 0 0

^ H H

n o

2 9° 15' 1 3 5 ° 0 0'

145

MATURATION TIMING AND GEOTHERMAL GRADIENTS As simple as this relationship may seem, published lists of gradients may be (and usually are) incompatible for two significant reasons, ignoring instrumental errors. Firstly, different studies have applied different surface temperatures for calculating the overall gradient: for instance, both Nicholas et al. (1980) and Polak & Horsfall (1979) used meteorological maps of surface (air) temperatures, while the latter also added 1.5 °C to the air temperature to approximate ground temperature. Middleton (1979a), on the other hand, assumed a ground surface temperature of 25 °C. For the present study, I apply yet another value: 20 °C. It is argued that the 'baseline' temperature to which the most meaningful overall thermal gradient is drawn is not the temperature at ground surface, where it is subject to diurnal and seasonal variations, but the rock temperature some metres or tens of metres below the ground. Such a temperature is probably 18-20°C in the region under discussion. Cull & Denham (1979) recognise the problem of defining a baseline temperature and point out the ideal situation of having access to a temperature measurement from some 100 m depth, well away from diurnal and seasonal variations. However, they also recognise the rarity of such data and greater availability of borehole-related measurements. Thus the problem of selecting a standard surface temperature is reiterated. Having established (or at least settled upon!) such a surface temperature, formation temperatures must be determined, to provide values for T 2 a n d / o r T,. This is the second significant source of error in gradient calculation and is discussed in detail below.

Sources of subsurface

temperature

data

Subsurface temperature observations may be made in a variety of ways: recording from mines or tunnels, purposedrilled and -equipped temperature observation wells and ocean floor probes. In the context of the study area, data are available only from water and petroleum wells. The following techniques are available for temperature measurement in petroleum wells. Some comments are made on their quality and availability. Wireline logging. Maximum temperatures are recorded during all runs of petrophysical wireline logs. The recorded temperature is not the true formation temperature, but must be corrected to an estimated formation temperature due to cooling of the formation by mud circulation prior to logging. Such data are, however, available from all petroleum wells, other than those few abandoned due to major engineering problems (such as collapsed hole or those with equipment stuck irretrievably in the hole). Even in these cases, some information is available from 'top-hole' logs. Drill stem tests. A maximum temperature is generally recorded near the top packer of the test tool. On the basis that gauges are checked prior to running a test, the instrumental accuracy is high; if the test is technically successful and flows a quantity of formation fluid (gas, water or oil) a reasonable estimate of the formation temperature is obtained. In gas tests, adiabatic cooling around the test tool is probably not a problem, as the gas most likely remains compressed at or near formation pressures until some distance up the drill string. However this aspect should be considered further. On rank wildcats, the lack of shows may warrant, at the extreme, no drill stem tests at all. Thus no temperature data would be available from this source. On the other hand, a dozen or more drill stem tests may be run on a prolific well.

327

The advantage of these over wireline log measurements is that many more data points may be available through the drilled section. Static pressure surveys. Temperature recordings are an integral part of static pressure surveys. Typically such surveys are run only on production-completed wells, after such time as temperature and pressure stabilisation are assured. Consequently a high vertical density of stabilised measurements may be available, but from fewer wells than the techniques discussed above. Moreover, although instrumental precision is of the order of 0.01 °C, convective circulation or turnover of the fluids in the borehole is of a greater magnitude. Temperature surveys. Temperature surveys are rarely run in petroleum wells in the Cooper-Eromanga Basin region, primarily due to the expense and rig-time involved, and the writer is not aware of any attempts to run a full-scale temperature survey subsequent to pressure-temperature stabilisation. Again, instrumental accuracy is high, but prestabilisation surveys require correction for mud circulation, while post-stabilisation surveys could be conducted only on suspended or production-completed wells, and are subject to convective turnover. Despite this, some well-planned temperature and differential temperature surveys would be of immense value, as shown by Gretener (1981). The most productive approach is therefore the use of corrected wireline log temperatures, with drill stem test temperatures providing added resolution and a check on the accuracy of extrapolation of wireline log temperatures to estimated formation temperatures. In a typical petroleum well, three or four wireline logs are run between 4 and 30 hours after cessation of drilling and mud circulation. During this period the bore wall and the fluid in the borehole are gradually returning to formation temperature. Some 30 hours after cessation of circulation, the measured temperature probably approaches by a few per cent the true formation temperature (Pitt, 1982, fig. lb). Equilibrium is probably attained after 50-100 days. Two groups of methods are used for deriving an equilibrium, or stabilised, temperature from readings taken in a borehole that has been cooled by mud circulation. Group one purely comprises mathematical constructions, which extrapolate the temperature trend to 100 days or more after cessation of circulation. The studies of Nicholas et al. (1980) and Ball (1982) avoided even this by adopting a still simpler approach of using uncorrected values from the last logging run while Middleton (1979a) added a constant 20 °C to 'raw' bottom-hole temperatures, to approximate the formation temperature. These days, the more common methods recognise the approximately inversely exponential rate of return to formation temperature. If the duration of post-drilling circulation is not available, a semi-log plot of temperature against time since cessation of circulation will extrapolate the temperature to a stabilised value at, say, 10 or 100 days. If circulation times are known, Horner plots, analogous to those constructed for pressure buildup during drill stem tests, may be constructed. The application of Horner plots is described in Fertl & Wichmann (1977): for successive logging runs log (t/(T+t)) is plotted against bottom-hole temperature, where t is the time since cessation of circulation and T is the duration of circulation after drilling. The intercept of a best-fit line on the temperature axis corresponds to log (t/(T-i-t))=0, that is, where t / ( T + t ) = l. This condition is approached when t is very much greater than T, thus the borehole has had sufficient time to stabilise.

328

G. M. PITT

Methods in group two for estimating formation temperatures involve comprehension of the dynamics of heat flow within a borehole, after cessation of circulation. Basic equations of heat flow, temperature behaviour and thermal diffusivity are applied to establish a set of master curves, against which data sets are. compared. Middleton (1979&) published an early variant of the method, using a square borehole model. Leblanc et al (1982) developed the model further, using conduction equations describing cylindrical boreholes, and compared their approach with the t/(T+t) method. Subsequently, Middleton (1982) published a modified description of the thermal behaviour of a borehole cooled by mud circulation. He pointed out that thermal stabilization is often slower than that predicted by simple thermal conduction models, and suggested that mud within the borehole may continue circulating after pumping has stopped. On this basis, he compiled a series of new time-temperature curves. Formation temperatures were then estimated by matching time-sequential bottom-hole temperatures to the appropriate member of the family of curves. In practice, I have found the Horner and 'semi-log' methods to be the most useful. Leblanc et al (1982) have shown that the Horner method, which ignores the physical nature of the system, agrees to within 1-2 per cent of the curve-fitting procedures described by them and by Middleton. However, while bottom-hole temperature and time since cessation of mud circulation are recorded, duration of mud circulation (necessary for Horner-plotting) is generally available only from drilling tour sheets. These are often difficult to access in a retrospective study. Thus the simple semi-log approach is most commonly used. It is worth noting that semi-log and Horner estimates, from the same data, commonly match to within 2 per cent. Unless otherwise possible, estimated formation temperatures reported here are derived from semi log plots and drill stem tests.

REGIONAL GEOTHERMAL GRADIENT MAPS

The geothermal gradient of an inhomogeneous sedimentary basin will vary areally, vertically through the stratigraphic section, and with geological time. Gradient maps and tabulations should therefore clearly indicate to what stratigraphic interval they refer. Drilling programmes in deeper parts of the Cooper Basin commonly call for intermediate wireline logging from, and casing to the Cadna-owie Formation. Thereafter, drilling proceeds to total-depth, with further logging. Consequently, three regional maps may be compiled showing: (1) 'basement to-surface gradients; (2) Cadna-owie Formation-to-surface gradients; and (3) basement-to-Cadna-owie Formation gradients. These are described and discussed below. A fourth map showing the numerical difference between the latter two has also been compiled. 'Basement', for the purposes of this paper, comprises a heterogeneous assortment of pre-Permian rocks which underlie the Cooper Basin sequence, often with angular unconformity. They consist of arenites, shales, carbonates and intrusives, at low to medium grades of metamorphism. 1

'Basement'-to-surface gradients

Figures 4 and 7 depict the distribution of gradients based on corrected temperatures recorded during final logging runs, at total depth. The gradient thus reflects the total temperature change from pre-Permian basement, or from basal Permian,

to near-surface. The isograds of these figures clearly show a zonation, (Roman numerals in Fig. 5 described below). Well locations are shown in Figures 6 and 8. Zone I. This zone covers the area from Daralingie and Moomba, to Wackett and Jackson. Within the zone, gradients exceed 4°C/100 m. A high of 5 ° to 5.7 °C/100 m lies over the Moomba area, while a similar high is present over the Strzelecki area. The maximum thermal gradient over the interval from basement to surface was recorded at Strzelecki Field, with a value of some 6°C/100 m. It is probable that individual wells will exceed this figure. The Moomba 'thermal high' was previously recognised by Kantsler et al (1978), Middleton (1979c) and Schwebel et al (1980). Calculations by Middleton (1979a) showed that radiogenic heat generation from a ?Carboniferous granite intrusion underlying the Moomba-Big Lake area was sufficient to account for the high heat flow implied by the gradients. Schwebel et al (1980, fig. 7) show an interpretation of granite basement distribution; its easterly extension coincides with continued high gradients. Some comments on the causal relationships will be made below. Zone II. The area of the Patchawarra Trough of the Cooper Basin and the Birdsville Track ridge comprise this zone. However, although there are abundant data along the southeastern side of the zone, information is particularly sparse elsewhere. Thus definition of the western limit of the zone is difficult. The contact between Zones I and II corresponds with the so-called Gidgealpa-Merrimelia-Innamincka (GMI) trend, an anticlinal trend separating the Nappamerri and Patchawarra Troughs of the Cooper Basin. The GMI trend is a 'thermal hinge', across which the gradient drops rapidly from 4-5.5°C/100 m, characteristic of Zone I, to 3.2-4°C/100 m characteristic of Zone II. The regional thermal structure of the area, as it relates to the GMI trend is less easily perceived on Figure 7 where, due to scale and well density, local structural variations are more apparent. Zone III Most of the area of the northern Cooper Basin is included in this zone. Wells are fewer than in the southern Cooper Basin and detailed gradient contouring is not possible. Nevertheless, it is clear that gradients are typically above 4.25°C/100 m. Well-to-well trends suggest that gradients probably reach 5°C/100 m in the Tanbar Trough, east of Curalle. At the southeastern end of the zone, gradients again increase to 5°C/100 m or more, coincident with a thinning and shallowing Permian section. Zone IV. In the Thunda-Yongala area a zone of 'relatively' low gradients, of 2.4 to 4.0°C/100 m, extends into Zone III from the east. There is no clear structural reason for this, but the Toowoomba-Charleville lineament of O'Driscoll & Keenihan (1980) appears to be located coincidentally (Fig. 5) and this should be considered in future structural studies. Zones V, VI. These zones occupy the area of the northeastern and southwestern Pedirka Basin respectively. Gradients in the former zone are greater than 4°C/100 m. In Zone VI, Southwest of a line through Walkandi 1 and Colson 1, they are less than 4°C/100 m. Data are sparse, but there is the suggestion of a thermal hinge, analogous to that over the GMI trend, between Zones V and VI.

Cadna-owie Formation-to-surface gradients

These gradients are derived from intermediate logging runs, which provide a measure of the temperature in the Cadnaowie Formation. The gradients are not fully compatible with those of the BMR water well survey (Polak & Horsfall, 1979),

135°00 2 3° 00'

1 3 8 ° 0 0'

1 42°00'

1 4 5° 0 0 ' 2 3 ° 0 0'

2 25

25 o z H

2 6° 0 0

2 6 ° 00*

O > z a o m X m 70

r O > D H co

2 9 ° 15 1 35°00'

29°15* 1 45°00'

UJ UJ O

1 35°00'

1 40°00'

1 38 00 T

1 36 0 0

142°00'

1 4 4°00'

F e r m o y 1* Maranda Mayneside

Newlands •Marduroo

2 4°00' \ Brothers

River

Stormhill

Adria

• Thomas Colson

26°00l\

•Erabena

Betoota

Downs 1

Mokari 1 Purni 1 • Macumba 1 .poolowanna 1 *Witcherrie 1 • Walkandi 1 Mt. C r i s p e 1 Kuncherinna 1 • Pandieburra

•Oodnadatta

• Belmore

1* ' B a n

Ban

Budgerygar Thunda 1

1•

£uralI.

Ingella

•

1* 2 . .

Yongala

Leopardwood 1 Chandos 1*# C o t h a l o w Tanbar N o r t h 1< South 1 Tanbar 1* C h a n d o s Alkina 1• • Putamurdie 1 1 Buckabie Cumbroo G i l p e p p e e 11 • Bodalla 1 - Mt. H o w i t t 1 Barrolka East 1 Barrolka 1* * -Wareena 1 Paning 1. 'Durham Downs 1 •Tartulla 1

P 2 H H

•Naccowlah 1 • Jackson 1

28°00'

2 8° 00'}

Weena

29° 15' 1 35°00'

12 4°00'

1 •

1 Morney

• McDills

1•

Galway •Hale

1 4 5°00' 12 3°00'

• Penrith

1 36°00'

1 38°00'

1•

.Kumbarie 1 •Gurra 1 1

_L ' C h e r r i 1 4 0 00

• Binera

Downs

I 14 2

144°00'

J 29° 15' 145o00'

1 39°30' 10

1 40°00' I

1 40°30l

1 4 1°30'

2 7°1 O 1

27°30'

2 7°30'

28°00'

28°30'

28°45'. 1 3 9°30'

1 42°00'

1 40°00'

140°30'

_1_ 1 4 1°00'

1 4 1°30'

28°45' 1 42°00'

140°30'

1 40°00'

1 3 9 >30' 2 7 ° 1 O1

1 4 1°00' •Arrabury

Coongie 1 «

1 4 1°30'

142 °00l

2 7 ° 1 O1

• Cuttapirrie 1 Karmona

•Beanbush 1

Kudrieke 1 •

27°30'

Wimma 1•

•Yanpurra 1

Coonatie 1

• Innamincka 1 •3 •Packsaddle 1 '2

•Moorari 1 •1

1 • _2 7°30!

• Merrimelia 9

H• 5

Kanowana 1 C o o p e r s C r e e k• 1 •

Wackett

Merrimelia 3 •1 * -Wantana 1 Mudrangie 2

B r o l g a 1* Fly L a k e 3 . 1 -4 T i r r a w a r r a 2 •• i 5*

•Tallalia 1

•Burley 1 O

T i n d i l p i e 1* Jack Lake 1 •

28°00'

4*J#g* G i d g e a l p a M o o m b a 6 • 5 . *26 »8

• Lake Hope 1 S p e n c e r 2 • 1*

7• •3 .1 - 1 3

Moomba\\

4D e l i a 2 » *5*1 6• Namur 1

O r i e n t o s N o r t h 1* .Orientos 1 Dullingari 3 B e l a h 1« Y a n k o 1.Burke 1 ^ Hume 12 2* • E P s , l o n 1 R o s e n e a t h 1* •Wolgolla 1 •Coochilara 1 'Kidman 1 -Ashby 1

\ ^ * . ^ * Big L a k e ••• 10* S t r z e l e c k i 2. Wooloo 1 Mudlalee 1* •3• D a r a l i n g i e 1 0 „ A Pando North 1 ,2 M u r t e r e e 1. M u r t e r e e C 1* T o o l a c h e e 1P a n d o 1*#2

-7

Wirrarie 1 •

28°30'

B o x w o o d 1*

M u r t e r e e A 1* "Wancoocha 1

2 8°00'

i.#.4

•Wills 1 Toolachee East 1 "Brumby 1

•Tickalara 1

N a r y i l c o 1*

T i l p a r e e A 1*

28°30'

•Mulga 1 28°45'

28°45 1 3 9 ° 3 01

1 40°00l

1 40°30'

1 4 1°30'

1 42°00'

h H

MATURATION TIMING AND GEOTHERMAL GRADIENTS which are strongly influenced by the temperature and conductivity of the porous Jurassic aquifer (Hutton and associated sandstones) into which most wells were drilled. The mapped distribution of the gradients (Fig. 9) is grossly influenced by The Brothers 1 in the central far north. This well has a gradient of 8.5 °C/100 m, due to the fact that Mesozoic sediments rest directly on high conductivity (low-gradient) prePermian basement. All other wells mapped in Figure 9 have a substantial Permo-Triassic section and are compatible with each other. The removal of a top-most low sonic velocity or low thermal conductivity layer by erosion can strongly influence sonic velocity or temperature gradient values. This is probably a significant component in the marginal areas. In the Patchawarra Trough a thermal low of about 3.2°C/100 m is coincidental with the structural low. This is the normal thermal structure for a trough or basin. However, the structural lows of the Nappamerri and Tennapera Troughs comprise thermal highs. The preserved sedimentary section of the three troughs is very similar. Thus the thermal difference in the Patchawarra and Nappamerri-Tennapera area is more fundamentally significant and probably lies with the nature of the pre-Permian basement.

Basement-to-Cadna-owie Formation gradients Basement to Cadna-owie Formation gradients (Fig. 10) are derived from the temperature and depth differential between intermediate and total-depth logging runs, usually to Cadnaowie Formation and pre-Permian basement respectively. They consequently span the Permian to Jurassic (inclusive) interval. In the study area, Permo-Triassic rocks are present between the Jurassic and basement sequences in some portions while elsewhere they are absent. The great variation of the stratigraphic section represented by this interval will clearly strongly influence the gradient value, and mapped gradients must be read with this in mind.

Difference between interval gradients The difference between the Permo-Jurassic and post-Jurassic gradients (Fig. 11) is a crude measure of the vertical variation of the geothermal gradient in the drilled section. Since the depth of the temperature measurements is fairly constant stratigraphically (intermediate casing generally set in Cadnaowie Formation and total depth is at base Permian or top pre-Permian) one may expect the relationship to be constant and determined by the ratio of bulk thermal conductivities of the two sections. In fact, this is not observed. The zero contour in Figure 11 marks equality between the interval gradients. Negative values reflect higher PermoJurassic gradients and correspond with the deeper troughs of the southern Cooper Basin and possibly (from a single data point: Tanbar North 1) within the central northern Cooper Basin. Structural high trends in the Merrimelia and Dullingari area give positive values. Indeed, there is a good correlation with regional gravity maps which also reflect gross basin structure.

DISCUSSION f

Basement-to-surface gradients

It is not surprising that 'basement-to-surface gradients closely reflect regional tectonic structure, based as they are, on temperature observations from the base of the Permian section and/or top of the pre-Permian basement. The major

333

thermal features identifiable with the regional tectonic and basin structure are listed below. (1) The Patchawarra Trough, Nappamerri-Tennapera area and northern Cooper Basin each have distinctive thermal regimes as described previously. (2) The Gidgealpa-Merrimelia-Innamincka (GMI) anticlinal trend is associated with a thermal 'hinge'. The hinge is sharply truncated at its northeastern end, near Innamincka 1. (3) The structural junction between the northern and southern Cooper Basins, known, among other names, as the Karmona-Jackson trend, is clearly shown by a strong change in pattern and trend of the thermal gradients. The northwestern portion of this trend is co-linear with the 4°C/100 m isograd; extrapolated further, it aligns with the Toomba Fault system (Fig. 5) of the Toko Syncline (Georgina Basin). Traced southwards, the same 4°C/100 m isograd swings southwest, marking the GMI trend. (4) A westward extrapolation of the Toowoomba-Charleville Lineament of O'Driscoll & Keenihan (1980) coincides with the geothermal low of Zone IV. Wells in this zone have gradients below 4°C/100 m (e.g. Thunda 1: 3.4°C/100 m; Yongala 1: 3.3°C/100 m; Ingella 1: 3.9°C/100 m), compared with 4-5°C/100 m elsewhere in the northern Cooper Basin. The reason for the relationship is still unclear. Many major lineaments are described in terms of their (often subtle) manifestations on remotely-sensed imagery while their physical nature remains undefined. This case is no different, but it would seem reasonable to suggest that the relationship can be traced to the underlying basement structure. (5) The relatively rapid drop of gradients normal to a Walkandi-Colson trend has yet to be identified in geological terms. Zone VI corresponds broadly to the distribution of the Permian in the Pedirka Basin (sensu stricto). Thus the thermal hinge between Zones V and VI may, like that over the GMI trend, have structural significance, in that there are Triassic and Jurassic, but no Permian rocks in Zone V. Data points in Zones V and VI are few and inferences from them should be tentative. With this caution, it would seem that the Zone V-VI boundary aligns with a change of gradient trend, south of Moomba, in the Kumbarie-Gurra area. In Figure 5, I have hypothesised that these are segments of a large-scale thermal lineament which would parallel the Toomba-Karmona-Jackson lineament discussed above. Two other independent features are generally coicident with the possible Walkandi-Gurra 'lineament' and would support the probability of an underlying basement tectonic relationship. Firstly, three of four recent earthquake epicentres documented by Wopfner & Youngs (1972) lie close to the lineament in the Simpson Desert. Secondly, the southeastern extension of the lineament marks the approximate southwestern limit of the major portion of Cooper Basin subcrop. Just as basement-involved tectonics are considered to be responsible for the regional structure of the overlying basins, so too, the origin of the regional thermal variations of the Cooper Basin area is considered to lie in the nature of the 'basement'. Specifically, two characteristics are involved: regional variation of the amount of heat transmitted into the upper crust from deep-seated sources; and the thermal characteristics of different basement lithologies. The areas of the northern and southern Cooper Basins have

UJ OJ 1 35°00 2 3 ° 00* OO

1 45°00' 2 3°00'

o I' o' ^ = = n o 3 3

2 6° 00' —

—

26°00'

H H

2 9 ° 15 135°00

138°00

142°00'

2 9 ° 1 51 14 5°00'

1 38°00'

1 35°00' 2 3° 0 0 '

1 45°00' I 23°00'

o ~ — o

25 o >

26°00'

2 6° 00'

o tn

m po X

O >73

a 5 z H in 200

=d KILOMETRES

2 9° 1 5 1 35°00'

1 38°00'

1 4 2° 0 0 '

29°15' 14 5 ° 0 0 1 IM

336

G. M. PITT

Fig. 11. Contour plan showing difference between Cadna-owie Formation-to-surface gradients and 'basement-to-Cadna-owie Formation gradients for specific wells, where full suite of data available. Positive values recorded where Cadna-owie Formation-to-surface gradients are greater. C.I. 1.0°C/100 m.

MATURATION TIMING AND GEOTHERMAL GRADIENTS

337

significantly different basement types. The southern Cooper Basin is underlain by scarcely metamorphosed CambroDevonian sediments of the Warburton Basin (Gatehouse, 1983). In the writer's opinion, these sediments have been structured along northeastern-southwestern trends inherited from earlier structural trends of underlying Adelaidean and crystalline basement.

Consequently the Moomba high may be related to the presence of granite intrusives through processes unrelated to Middleton's proposal. Alternatively the association could be purely fortuitous and the high gradients due, perhaps, to a crustal 'hot spot' instead.

In contrast, the northern Cooper Basin is underlain by low to medium grade metamorphosed Palaeozoic sediments with meridional Tasmanide trends. Permian and Mesozoic structural trends in the overlying basinal sequences are therefore also meridional.

An explanation of the distribution of these gradients is less clear, but some comments may be made. It is likely that the increasingly high gradient values of the few wells north of 26 °S latitude and west of 142 °E longitude are due to the absence of a Permo-Triassic section. In most such wells, the Eromanga Basin sequence rests on crystalline basement or folded Palaeozoic strata. 'Basement' sequences typically have a high thermal conductivity and consequently low thermal gradients. For the reasons shown in Figure 3, a shallow, overlying sedimentary section will display a greater geothermal gradient than a thicker adjacent sequence, under otherwise identical conditions. A 'low' over the Patchawarra Trough area and 'high' over the southern Nappamerri Trough area are the two other major thermal features to be noted at Cadna-owie Formation level. While the depths of the Cadna-owie Formation and Early Permian formations are respectively similar over the two troughs, the gradient values are very different. This is evidence that there are more fundamental influences at work, with regard to gradient distribution, than merely basin depth.

Conductivity differences in these basement types will clearly exert a control on the gradient characteristics of the overlying basinal sections, as discussed above. Rocks of lower porosity and greater intercrystalline strength will be better thermal conductors. Thus igneous and metamorphic rocks will be typically better thermal conductors than sedimentary rocks. A higher metamorphic grade of the basement to the northern Cooper Basin will result in an enhanced thermal flux (due to increased conductivity) to the base of the Permian section and thence a higher gradient from base-of-Permian to surface. Such relationships may go part of the way to explaining the difference between the northern Cooper (Zone III) and Patchawarra Trough (Zone II). However the MoombaStrzelecki thermal high is clearly anomalous in this regard and other explanations must be considered. Information presently available (Schwebel et al., 1980, fig. 7; Gatehouse, 1983) indicates that three granitic intrusions, of possible Carboniferous age, subcrop at the basal Permian unconformity. The first lies in the Wooloo—southwestern Moomba area; the second, over central-northern Moomba field; and the third, in the Wolgolla-Epsilon area. Middleton (1979a) computed the radiogenic heat productivities of Ordovician and granitic basement, and Permian core samples from Moomba 1 to 8, Big Lake 1 and Brumby 1, from uranium, thorium and potassium analyses. Heat generative capacity of the Permian samples was higher than Ordovician (sedimentary) basement and some four times higher than typical heat production in sediments. Middleton attributed this to 'derivation from . . . basement rocks high in concentration of these elements'. The heat generative capacity of the granite samples was considerably higher again, and Middleton concluded that 'Surface heatflow in the Moomba-Big Lake region is fully consistent with static crustal heat flow models entailing a heat production of 20 HFU in a granite layer between 7 to 10 km thick; analyses of basement granitic cores for U, Th and K concentrations confirm the order of magnitude of this heat production'. Since this work, the Moomba thermal high has been commonly attributed to radiogenic heat generation from the Moomba granites. However, the picture may not be so simple. Granite subcrop distributions and gradient highs are not fully coincident. In areas such as Strzelecki there is no granite subcrop but there is a thermal gradient high; this could be simply explained by invoking a shallow (but not drilled or seismically defined) granite intrusion. The two are coincident in part of Moomba Field and in the Wolgolla-Epsilon area. However, in southwestern Moomba-Wooloo there is an area of granite characterised by lower gradients. Igneous crystalline bodies such as the Moomba granites are more thermally conductive than sediments or metasediments. They consequently may act as conductive 'pipes' resulting in an enhanced heat flux into the basal Permian and a higher gradient from the basal Permian to the present-day surface.

Cadna-owie Formation-to-surface gradients

During this work, there has also been some speculation as to the role of the Jurassic aquifer on thermal gradients in the Cretaceous. The structurally deepest and hottest portion of the Eromanga Basin overlies the Cooper Basin, in particular, the Nappamerri Trough. Water within the Great Artesian Basin system originates in eastern Queensland and moves in a southwesterly to westerly direction, to emerge in springs on the southwestern margin of the basin in the Lake Eyre area. As the artesian waters traverse the cooler northern Cooper Basin and/or Cunnamulla Shelf and then the NappamerriTennapera Trough areas, they will increase in temperature due both to the greater gradients and greater depth. By doing so, a portion of the heat flux will be absorbed. The consequence of this is that gradients in the overlying Cretaceous sequence may be lowered. Conversely, passing on to the cooler Patchawarra Trough and structurally shallower areas, the heated waters may actually input thermal energy to the system and thereby increase overlying geothermal gradients. No figures have been calculated for such a model of 'lateral convective transfer' of the heat flux from the Moomba high and it is not known whether such effects are significant and detectable in practice, considering the relative rates of vertical heat conduction and aquifer flow. (

Basement-to-Cadna-owie Formation gradients

The interval represented by these gradients is less stratigraphically homogeneous than those discussed above. In the Cooper Basin region, the gradient interval covers Jurassic, Triassic and thick Permian. Elsewhere, Triassic and/or Permian may be absent. Such factors must strongly influence the isograd distribution. An area of particularly high interval gradients is present in the southern Cooper Basin and may lend credence to Middleton's {\919a) inference of radiogenic heating within the Permian sediments. Despite a poor spread of data points, there is also some suggestion of thermal effects related to the Karmona-Jackson lineament.

338

G. M. PITT

MODELLING OF ORGANIC MATURATION Since recent papers by Wright (1980) and Waples (1980), there has been much interest in the applications of the organic maturation modelling technique devised by Lopatin (1971, 1976). This is essentially a simple mathematical technique employing the Arrhenius equation. The resulting 'maturation index' value is related to the 'real world' via a vitrinite reflectance correlation compiled by Waples (1980). Subsequent to Waples' paper, other workers have applied, commented upon and refined his methods (Cohen, 1981; Waples, 1981, 1982; Katz et al., 1982; Falvey & Deighton, 1982; Pitt, 1982; Kantsler et al., 1983). Gretener & Curtis (1982) described a modified Lopatin method, which produced a socalled 'oleum scale' of organic maturation. Their work is particularly valuable in underscoring the different effects of time and temperature on organic maturation. While the effect of temperature is exponential on reaction rates in organic metamorphism, the effect of time is linear. Consequently, at low temperatures, time (even geologically long spans) has little effect on organic maturation. At high temperatures, the reaction rates are sufficiently high that time again plays no important role. These aspects are also inherent in the mathematics of Waples' Lopatin method. Basis of the Lopatin

technique

The reader is referred to Waples' papers and Gretener & Curtis (1982) for clear and detailed descriptions of the Lopatin and 'oleum' techniques. The Arrhenius equation suggests that the reaction rate is approximately doubled for each 10 °C rise in temperature. A 'reaction factor' R (-2) may be defined. Lopatin chose the temperature interval 100-110 °C as a datum interval, assigning it an index value of n = zero. The temperature window of 90-100 °C has an index value of n = -1, and the 110-120 °C window, an index of n = +1. Ignoring, for the moment, the time factor, the relative reaction rate of the 90-100 °C window would be Rn (or approximately 2~') - 0.5 of that of the 100-110°C window. For the 110-120 °C and 120-130 °C windows, reaction rates proceed at Rn =• 2 and 4 times, respectively, the rate of that of the 100-110 °C window. The exponential effect of temperature can be clearly seen here. On the other hand, effect of time is linear: doubling the time doubles the maturation. Consequently, the increment of maturity added during a particular temperature window is: Maturityj = ( a T J ( R n i ) , where Tj is the length of time spent by the sediment in the temperature interval i. The total maturity (TTI, or Time-Temperature Index) of a given sample is given by: TTI = 2 ( a T J (Rn) whereby the incremental maturities, contributed from all temperature windows through which the sediment passed, are summed. The TTI is not an absolute value, but relative to the base value of n, chosen as n=0 for 100-110°C. A TTI/Ro correlation, established by Waples (1980) relates TTI values to vitrinite reflectances. Maturation levels are interpreted using this correlation. Input

requirements

Maturation modelling techniques rely primarily on the determination of a time-temperature curve for each stratum

under consideration. The knowledge of the present-day thermal gradient, the subject of the first part of this paper, is an important parameter, defining present-day temperatures. However, to determine the temperature history of strata, a burial history and thermal gradient history for the particular stratigraphic section must be supplied. Each has its own difficulties. The compilation of 'best guess' burial curves relies upon the following factors: a consideration of the current correlation between absolute ages and palaeontological zonations; the section as drilled; and an estimate of the amount of section deposited, then eroded, before further (ultimately preserved) section was deposited. The latter estimate is gained from such items as a knowledge of regional thickness variations and, in particular, breaks in vitrinite reflectance curves, and variations from typical sonic velocitydepth relationships. In the Cooper Basin region, there are three significant unconformities: Early-Late Permian, ?Late Triassic-Early Jurassic and Late Cretaceous-Early Tertiary. The Early to Late Permian break is of less importance to maturation studies, as the Early Permian sediments were immature during the Late Permian. However, the size of the ?Late Triassic-Early Jurassic break (in terms of metres of section eroded) will influence estimates of timing of onset of early maturity of the Early Permian. Similarly, estimates of the amount of Cretaceous section eroded will influence maturation modelling of the Permian, Triassic and Jurassic. The degree of ?Late Triassic erosion is apparent on vitrinite reflectance curves of many wells. An example of this is provided by Kantsler et al., (1983, fig. 15). Breaks in the trend of the profile provide evidence of lost section; reconstitution of the curve can indicate the magnitude of the section lost. In the case of the Cooper Basin area, breaks of zero to .500 m are inferred for individual drilled sections by this means. Since the burial curve is more accurately known than the thermal history, maturation indices are modelled, converted to vitrinite reflectance-equivalent values and compared with those observed. Thereafter the thermal history is iteratively adjusted to reduce major discrepancies with the observed reflectance curve, until a unique solution is obtained for the former. It is important to note that if the section under consideration were of vertically constant thermal gradient and if the thermal conductivity did not change with time, then only a single, unique, gradient history would match the observed vitrinite curve. In reality, the thermal conductivity of a stratum will change (generally increase) as it undergoes burial, compaction and diagenesis; thus, at any moment the thermal conductivity and gradient of a drilled section will vary vertically due to increasing compaction and lithology change. Recent work (e.g. Falvey & Deighton, 1982), incorporates a sediment compaction factor. If this is done, it is equally arguable that the increasing thermal conductivity (decreasing thermal gradient) with progressive sediment compaction should also be considered. Sediment decompaction will elevate model maturities somewhat. On the other hand, elsewhere in this paper it is suggested that low-, not high-conductivity sequences are more important in enhancing maturity due to their insulating effect on underlying sediments. Thus I tend to agree with the philosophy implied by Gretener & Curtis (1982) that such refinements may overstate our present understanding of the subject; more field data on thermal transmissivity and the underlying principles of maturation are needed.

MATURATION TIMING AND GEOTHERMAL GRADIENTS

Computer-processed modelling In his original paper, Waples (1980) remarked on the feasibility of computer-processing of Lopatin-type modelling. The modelling procedure, itself, is a simple but rigorous mathematical construction, whereby the connection to the

339

'real world' is provided by carefully considered inputs of burial and thermal histories. The technique is best applied when it may be run numerous times, refining in particular, thermal histories to produce vitrinite-equivalence maturation curves which match observed curves as closely as possible. Such manual, iterative calculation is tedious, and computer-

ARo

-0.4

-0.3

RUN 3 RUN~5

RUN 4

-0.2

-0.1

-0.2

-0.1

0.1

0.2

0.1

0.2

0.4

0.6 ,

-i—i—i—r—i -

O 3.5 5\ 23.0h .5-

ARo -0.3

0.3,

3.5h -RUNS 6 , 8 , 9 £ Q.

Ma. B.P.

LU Q

RUN 6 -0.3 I

-0.2 1—

-0.1

ARo 0

RUN 8 0.1

0.2

0.3,

E a

O 3.5 -

E 2.8. T 2.6 =

RUN 9

— RUN 1 1 — Q.

Ma. B.P.

RUN 10

RUN 12

Fig. 12. Successive modelling runs of Beanbush 1. A Ro = Rmod - Ro. Gradient histories are shown; Arrhenius 'R' values: Run 1, 2, 3: 2.0; Run 4: 2.4; Run 5: 2.1; Run 6: 2.1; Run 8: 2.4; Run 9, 10, 11, 12: 2.3. Note compressed scale for positive A R o values for Runs 1-5.

MOOMBA

340

Run

10 -

12 -

MODELLED

DEPTH

(km)

OBSERVED

MODELLED

Ro %

CURALLE

OBSERVED

MODELLED (km)

(km)

TARTULLA

G. M. PITT

MODELLED

DEPTH

OBSERVED DEPTH

Fig. 13. Comparison of modelled and observed vitrinite reflectance (Ro) curves. Note compressed Ro scale for Moomba 3.

BEANBUSH

Ro %

299

GEOTHERMAL GRADIENT (®C/100m) GEOTHERMAL GRADIENT (®C/100m)

Fig. 14. Summary of geoihermal gradient histories resulting in 'successful' modelling of maturation curves, details given in Table 1. Gradient value at any point in time reflects average gradient of existing basinal section.

GEOTHERMAL GRADIENT (®C/100m)

MATURATION TIMING AND GEOTHERMAL GRADIENTS

B.P. YEARS MILLION B.P.

1 TARTULLA

CURALLE MILLION

YEARS MILLION

B.P.

YEARS

MOOMBA 1 BEANBUSH

341

G. M. PITT

342

processing is advisable. In-house computer-processed methods of maturation modelling have been developed, and results incorporated in, for example, Kantsler et al. (1983), Falvey & Deighton (1982), Pitt (1982) and Thomas & Brown (1983). The method applied in this study assumes that the thermal gradient at any moment is constant through the stratigraphic % (ARo/Ro x 1 0 0 )

section (though this is recognised as a first approximation) while the gradient can vary with time. Horizons for which an analysis is required are identified by their present-day depth. Output comprises the following: a printout for each horizon, tabulating time, depth, gradient, formation temperature, timing of 10 °C temperature windows, interval maturation index and summed maturation index; a plot for each horizon graphing time-depth, time-thermal gradient, time-formation temperature and time-maturation index; and a composite plot showing the burial history and maturation of all horizons (e.g. Figs 16-19).

Reporting of results Given the current first-generation state of development of the Lopatin technique, it is misleading to describe a well or stratigraphic section as 'successfully modelled', as has occurred in some recent instances. Success, in this context, is relative. For instance, thermal gradients applied in such 'successful' modelling are often assumed to be vertically constant within the stratigraphic section, and to be temporally constant within a particular stratigraphic unit. Lithology change and compaction are two reasons why this may not be so. Thus 'success' may be obtained from some very generalised, even incorrect, assumptions, indeed.

PK 299

Fig. 15. Deviation of modelled vitrinite reflectance curves from those observed in Beanbush 1, Curalle 1 and Tartulla 1, 'hung' on top of Cadna-owie Formation.

LOPATIN

MATURATION

MODELLING

TRIASSIC

JURASSIC

PERMIAN

Moreover, what degree of agreement between modelled and observed maturation indices constitutes 'success'? Of the recent papers publishing results of modelling, only Thomas & Brown (1982, fig. 7) and Pitt (1982, table 1) give any data on the degree to which modelled and observed maturation indices (e.g. vitrinite reflectances) match. In my opinion, published modelling results should always be accompanied by figures or tables showing the closeness of the match throughout the section under consideration.

CRETACEOUS

TERTIARY STRATIGRAPHIC

Tc—

CO LU

Tr — Te 1500

. Pa .

2.20

BEANBUSH

Q_ UJ

M A X I M U M TT1 ATTAINED - 242

(C)

1983 WMC

Petroleum

Ex.

Dlv.

Compiled

GRAHAM

04-AUG-83

Fig. 16. Computer-generated geohistory plot for Beanbush 1. Abbreviations for horizons shown on Figures 16-19 are: Tc, Toolebuc Formation; K, Coorikiana Sandstone; Mu, Murta Member; Bh, Birkhead Formation; BJ, 'basal' Jurassic; Tr, top of Triassic; eTr, intra-Nappamerri Formation; Te, intra-Toolachee Formation; Pa, intra-Patchawarra Formation.

343

MATURATION TIMING AND GEOTHERMAL GRADIENTS

LOPATIN

MATURATION

MODELLING

TRIASSIC

JURASSIC

PERMIAN

CRETACEOUS

TERTIARY STRATIGRAPHIC

K Mu

Tr — Te Pa

MOOMBA

M A X I M U M TT1 A T T A I N E D - 2 7 6 7

(C)

1 9 8 3 UMC P e t r o l e u m

Ex.

Dlv

Fig. 17. Computer-generated geohistory plot for Moomba 3.

Fig. 18. Computer-generated geohistory plot for Tartulla 1.

Complied

GRAHAM

04-AUG-83

344

G. M. PITT

Fig. 19. Computer-generated geohistory plot for Curalle 1.

An example of modelling

In a following section, the maturation history of Beanbush 1 and three other wells is modelled and discussed. Here, the progressive modelling of Beanbush 1 is detailed to show the sensitivity of maturation curves (expressed as reflectance equivalent) to factors such as changing thermal histories and variation of the Arrhenius factor R, and to show the logic processes involved in compiling an acceptable model curve. Figure 12 depicts, for 250 m intervals, the difference ( A RO) between modelled vitrinite values (Rmod) and observed values (Ro) for different runs. This relies upon comparison with the best fit curve of observed vitrinite reflectances. The average deviation of observed individual values from the best-fit curve defines the logical limit to matching modelled and observed reflectances. The simplest approach to modelling is to assume a constant gradient (temporally and vertically), equal to that at the present day. Run 1 indicates the degree of maturation, in terms of equivalent vitrinite reflectance, attained by the stratigraphic section at Beanbush 1 by a 'best guess' burial history and assumed constant gradient of 3.5°C/100 m, the present-day basement-to-surface value. The result is not surprising in view of Karweil-method results published by Kantsler and coworkers (e.g. Kantsler et al., 1978): if operative since the Permian, vitrinite reflectances would be 125 to 175% of those observed throughout the drilled section. It is therefore inferred that the present gradient at Beanbush 1 is a recent phenomenon unless other unrecognised factors are operative. Subsequent model runs 2 to 5 evaluate maturation profiles, applying 2.5 and 3.5°C/100 m until 2 Ma B.P. and thence the current 3.5°C/100 m from 2 Ma B.P. to the present day. Run 2(3.0°C/100 m) is clearly still too high to account for

the observed maturation profile. Runs 3, 4 and particularly run 5 approach the observed profile more closely with a gradient of 2.5 °C/100 m until 2 Ma B.P. However, all fail to model the degree of maturation pertaining below 3000 m. Runs 3 to 5 differ in that R is set at 2, 2.4 and 2.1 respectively. The influence of varying R (both in direction and degree) can thereby be seen: increasing R rotates the curve anticlockwise, while decreasing R rotates the curve clockwise. Runs 6, 8 and 9 attempt to correct for the under-modelling of maturation beneath about 2750 m. To do so, the relatively recent increase from 2.5 to 3.5°C/100 m, previously set at 2 Ma B.P., is now set at 10 Ma B.P. Due to the exponential nature of temperature, maturation within the higher temperature windows will be influenced to a greater extent. The value of R was set at 2.1, 2.4 and 2.3 respectively. Curves derived from Runs 1 to 9 demonstrate magnitude of changes caused by specified changes in thermal gradient and the Arrhenius reaction factor. The gradient history is further tailored in run 10, to allow for a gradual rather than instantaneous rise from 2.5 to 3.5 °C/100 m at about 10 Ma B.P. An R value of 2.3 appears from previous runs to be appropriate to most of the section and is applied in runs 10, 11 and 12. Run 11 increases the pre-10 Ma B.P. gradient to 2.8°C/100 m but over-models the entire section. Finally, run 12 applies 2.6°C/100 m instead and decreases the under-modelling of run 10 in top and bottomhole sections. Runs 10 and 12 thus most successfully model the whole section, on 250 m depth intervals, with about 70 per cent of data points within 0.02 per cent Ro of the observed vitrinite reflectance curve. Expressed as a percentage departure from observed values, most are within - 5 and +4 per cent with a major deviation (at total depth) of 13 per cent. Plots such as those shown in Figure 12 provide a very

345

MATURATION T I M I N G A N D G E O T H E R M A L G R A D I E N T S

TABLE 1. Modelled (R mod), observed (Ro), and deviations from (ARo) vitrinite reflectance values for studied wells at 250 m depth increments.

(1) = average per cent deviation, sign ignored. (2) = average per cent deviation taking sign into account. Ro % Observed Modelled (100-ARo/Ro) ARO Ro TTI Rmod

Average

Depth (m)

Beanbush 1 Run 10 1250 1500 1750 2000 2250 2500 2750 3000 3250 3500 3600

0.96 1.71 2.81 4.89 9.60 15.42 26.78 48.20 88.58 162.29 201.58

0.399 0.435 0.491 0.524 0.594 0.655 0.745 0.890 1.055 1.305 1.393

0.42 0.46 0.49 0.54 0.59 0.65 0.73 0.90 1.15 1.45 1.60

-0.021 -0.025 + 0.001 -0.016 + 0.004 + 0.005 + 0.015 -0.010 -0.095 -0.145 -0.207

Beanbush 1 Run 12 1250 1500 1750 2000 2250 2500 2750 3000 3250 3500 3600

1.03 1.75 3.06 5.59 9.88 17.53 32.68 59.34 104.19 192.39 242.20

0.401 0.438 0.500 0.530 0.596 0.675 0.790 0.938 1.120 1.380 1.445

0.42 0.46 0.49 0.54 0.59 0.65 0.73 0.90 1.15 1.45 1.60

-0.019 -0.022 + 0.010 -0.010 + 0.006 + 0.025 + 0.060 + 0.038 -0.030 -0.070 -0.155

-4.5% -4.8% + 2.0%

Moomba 3 1500 1750 2000 2250 2500 2750 2880

8.20 22.46 63.99 185.54 522.8 166.7 2767

0.570 0.718 0.958 1.366 1.770 2.260 2.510

0.4 0.56 0.90 1.30 1.90 2.40 3.0

+ 0.170 + 0.158 + 0.058 + 0.066 -0.130 -0.140 - 0.490

+ 17% + 28% + 6.4% + 5.1% -5.8% - 16.3%

0.22

-5.0% -5.4% + 0.2% -2.9% + 0.6% + 0.7% -2.0%

-1.1%

-8.3% - 10.0% -12.9%

(1) 4.4% (2) - 4 . 2 %

-1.8%

+ 1.0% + 3.8% + 8.2% + 4.2%

(1) 4.3% (2) - 0 . 8 %

-2.6%

-4.8% -9.7%

(1) 12.2% (2) +3.9%

-6.8%

Tartulla 1 250 500 750 1000 1250 1500 1750 2000 2250 2575

0.40 0.65 1.49 3.00 5.77 9.07 19.11 42.97 102.06

0.21 0.24 0.34 0.425 0.500 0.533 0.585 0.690 0.863 1.109

0.21 0.24 0.34 0.39 0.44 0.49 0.55 0.65 0.78 1.09

0 0 0 + 0.035 + 0.060 + 0.043 + 0.035 + 0.040 + 0.083 + 0.019

0% 0% 0% + 8.9% + 13.6% + 8.7% + 6.4% + 6.1% + 10.6% + 1.7%

(1) 5.6% (2) +5.6%

Curalle 1 250 500 750 1000 1250 1500 1750 1833

0.81 1.92 4.53 10.70 25.24 59.48 140.0 162.8

0.38 0.445 0.518 0.607 0.737 0.943 1.260 1.305

0.39 0.41 0.51 0.61 0.74 0.94 1.165 1.34

+ 0.005 + 0.008 -0.003 -0.003 + 0.003 + 0.095 -0.035

-0.01

-2.5% + 1.1% + 1.6% - 0.49% -0.41% + 0.32% + 8.1%

(1) 2.1% (2) +0.6%

sensitive display of the accuracy of successive iterative runs and it is recommended that data should always be presented in a similar form when reporting modelling results. The sensitivity of such plots is appreciated compared with simple plots of R m o d and Ro (Fig. 13). In the following examples, the assumed size of various erosional intervals is specified; they were estimated using factors outlined previously, but it is not possible to discuss them individually here.

- 2.6%

BEANBUSH 1 Beanbush 1, drilled in the Patchawarra Trough, provides an excellent opportunity to test maturation modelling, due to the fact that vitrinite studies indicate a relatively simple geothermal history and that, being centrally situated within the Patchawarra Trough, it has a section least modified by erosion. There is little evidence, from displacements of the vitrinite or sonic velocity curves, of major erosional breaks. Consequently, a burial curve is compiled on the assumption

346

G. M. PITT

of erosion of a nominal 30 m of Early Permian, of 50 m of Late Triassic and of negligible Winton Formation. The overlying Tertiary is relatively thick (240 m), probably comprising Eocene to Miocene deposits. Post-Miocene erosion is not considered to be significant. It may be argued that the amount of erosion was greater than that assumed here. If so, the degree of maturation may be underestimated, although I would argue strongly against major erosional intervals at Beanbush 1. Details of modelling runs of the Beanbush 1 section were discussed previously in order to show the degree of change of modelled maturation effected by a change of input data. Runs 10 and 12 (Figs 12, 13) most closely model the observed reflectance curve. They invoke a temperature gradient which averaged 2.5 to 2.6°C/100 m through the stratigraphic section until ca 10 Ma B.P. Thereafter, a slow increase from 2.5 or 2.6 to the present-day 3.5°C/100 m took place (Fig. 14). An immediate rise to 3.5°C/100 m at 10 Ma B.P. over-models the younger section. High gradients (of whatever origin), pre-170 Ma B.P., have little effect either on modelling results or on maturation of the Permo-Triassic in nature and may be unrecognised. An Arrhenius R value of 2 does not appear appropriate (Fig. 14); substitution of approximately 2.3 results in better modelling of the whole section (Table 1).

MOOMBA 3 In the Moomba-Big Lake area a relatively thick and largely complete Early Permian section is preserved. An original thickness of 500 m is assumed, with erosional thinning to the present 411 m. The Late Permian-Triassic section is now 285 m thick and 55 m are assumed eroded prior to Jurassic deposition. Likewise, the Winton Formation, currently 695 m thick, is assumed here to have had 200 m eroded between 92 Ma B.P. and Eocene-Miocene deposition. Some 270 m of Eocene and Miocene sediments overlie the Cretaceous at Moomba. Post-Miocene erosion is considered minimal. Attempts to model maturation at Moomba 3 were the least satisfactory of the four sections reported here. Observed reflectances of 0.4-0.5 per cent Ro are not reached until 1500-1750 m depth rising to 3 per cent before 3000 m. Modelling runs using the present gradient result in strong overmodelling of the younger section (above 2750 m) and moderate over-modelling of the remainder. To remedy this, high pre-mid-Jurassic gradients (5.50 C/100 m) were applied with high Arrhenius R values (2.6). Alternatively, particularly high gradients and R values may be invoked but these produced excessive mis-correlation in one part or another of the vitrinite curve. The best correlation to the observed vitrinite curve constituted an average 16 per cent positive or negative fit to observed vitrinite values. It requires an average gradient of 5.5 °C/100 m until 170 Ma B.P. (base Jurassic), decreasing to 4.0 thereafter, and 4.0 rising to the present 5.6°C/100 m over the past 5 Ma (Fig. 14). The sensitivity of the technique is low at pre-to early-maturity levels in the Permian, however, and further work is needed. It is notable that in order to obtain reasonable geothermal scenarios an Arrhenius R factor of 2.3 to 2.6 was required.

CURALLE 1 The Curalle 1 section differs from that at Beanbush 1 and Moomba 3 in that any pre-Nappamerri section is absent due to erosion and onlap and that the structure today has surface expression due to post-Late Cretaceous arching and postOligocene or Miocene erosion.

The Triassic-Jurassic break is considered to be relatively small due to the identification of Late Triassic strata. Consequently, compilation of the Curalle burial curve is simplified, since only the post-Winton Formation unconformity is significant. It is assumed that 700 m of Winton Formation were originally deposited, of which 550 m have been eroded. Such thicknesses are derived from the reconstruction of Winton Formation stratigraphy across the Tanbar Trough (northern Cooper Basin); they allow for possible thinning on the flanks of the trough. Erosion is interpreted to have occurred both before and after OligoceneMiocene silcrete development (Moore & Pitt, 1984). The complexity of Tertiary weathering phases, sedimentation and silcrete genesis (see, e.g. Wopfner, 1974; Pitt, 1978), is such that a further refinement of the Tertiary structural history of the Curalle Dome is not yet possible. Maturation modelling, using a burial history based on the above precepts, the present-day gradient of 4.5 °C/100 m, and R = 2.0, results in strong over-modelling of the younger section and under-modelling of the older section. However, applying R = 2.3 and average gradient of 4.2°C/100 m until 2.0 Ma B.P. (afterwards rising to 4.5°C/100 m from 2.0 Ma B.P.) (Fig. 14) produces a remarkable agreement of 2.06% average (positive or negative) with the observed vitrinite curve (Table 1). Modelling with the latter thermal history and R = 2 results in an average deviation of 7.3%, reiterating the fact that R = 2.3 produces a more successful model.

TARTULLA 1 The following factors were taken into account for the compilation of a burial curve for Tartulla 1: a strongly eroded Early Permian section consisting now only of Patchawarra Formation; relatively thick Triassic, implying reduced erosion; and a thick Winton Formation from which a maximum of some 200 m have probably been eroded. The current basement-to-surface gradient in Tartulla 1 is about 4.0°C/100 m. If this value is applied in maturation modelling, equivalent vitrinite reflectances are obtained considerably in excess of those observed. Close agreement is obtained applying a 3.0 to 3.1 °C/100 m average for the section, until 5 Ma B.P, after which the gradient rises to 4.0°C/i00 m at the present day (Fig. 14). An R value of 2.2 to 2.3 appears to model the reflectance curve more satisfactorily than R = 2.0 (Fig. 13).

DISCUSSION Thermal histories The most successful model gradient histories for the four wells are summarised in Figure 14. Modelling of these wells seems to confirm the view expressed by organic petrologists that the current high thermal gradients are a relatively recent phenomenon: if those high gradients were applied throughout the history of the particular section under study, most of that section will be strongly over-modelled. The timing of the recent rise in gradient has been taken at 5 to 10 Ma B.P. in the present work. If the rise is immediate at 10 Ma B.P., or assumed to be earlier still, the post-Jurassic section is over-modelled. Consequently it is considered that the rise to current gradients took place within the past 5 Ma or so and may have been very recent. A corollary of this is that the recent high gradients have not yet had a significant effect on hydrocarbon generation: this can be shown by calculating some example interval maturation indices.

MATURATION TIMING AND GEOTHERMAL GRADIENTS

347

predominant maceral types within Eromanga Basin source rocks. Such data must be considered when interpreting TTI values and the equivalent reflectances. Table 2 gives details of modelled generation timing for seven source-reservoir horizons in the four wells studied here. Moomba 3 is consistently the most advanced well with respect Value of the Arrhenius R Factor to maturation; it is often nearly matched by Curalle 1. This During modelling of the four wells, the value of R was is, at first sight, a surprising result but is explained by the varied between 1.6 and 2.6. Conventionally, R is considered high palaeogradients which pertained at Curalle. Beanbush to approximate 2.0. However, in the present work, a value 1 and Tartulla 1 had similar 'threshold histories' to each other of 2.3 was found to produce more satisfactory results and the until the Early Tertiary and thereafter deviated markedly. The use of R = 2.0 produced high Cretaceous and low Permian latter effect is attributed to thick Tertiary deposition (250 m) the Patchawarra Trough (Beanbush area), followed by maturation estimates that could not be simultaneously over negligible erosion. Consequently a 5-10 °C formation corrected by any reasonable geothermal history. temperature increase was caused and maintained by the Two possible explanations are suggested for such a value Tertiary deposits. However, at Tartulla active erosion of of R. Firstly, that it has a genuine physical basis in that Cretaceous Winton Formation is presently proceeding, and reaction rates for organic matter under geological conditions only thin Tertiary Eyre Formation is believed to have been proceed at a rate of 2.3 times faster for each 10°C interval. deposited and/or preserved for a significant amount of time. Thus it would constitute a valid refinement of the Lopatin According to Cook (1982, this volume) common to technique for, at least, the study area. Secondly, it may be an artificial correction for factors which are either unknown abundant macerals in the Bulldog Shale and Wallumbilla or not understood. For instance, it may result from the fact Formation are typically of the early-generative type, with that the basinal section does not have a vertically constant initial generation at 0.4-0.5 per cent Ro. My results suggest thermal conductivity: the application of R = 2.3 may merely that at Curalle 1 and Moomba 3, the Toolebuc Formation be correcting for effects of insulating shaley and coally and Coorikiana Sandstone reached 0.4 per cent Ro at about sections, which influence vertical heat flow and maturation. early-mid Tertiary (ca 30-50 Ma B.P.). A value of 0.5 per cent Ro was reached at 7.5-5.0 Ma B.P. Generation of significant This is discussed in another context below. volumes of hydrocarbons is demonstrated at Strzelecki 8, near Moomba 3, which has flowed gas from the Coorikiana Influence of low-conductivity sequences Sandstone. (Recent studies by McKirdy et al. (1985) suggest Figure 15 depicts the discrepancy of modelled curves against that the Bulldog Shale, the probable source of hydrocarbons observed vitrinite curves, expressed as a percentage deviation in the Coorikiana Sandstone, is gas-prone). from Ro, for Beanbush 1, Curalle 1 and Tartulla 1. All are Common macerals in the Early Cretaceous Murta Member 'hung' relative to a nominal datum, the top of the Cadnathe Mooga Formation are vitrinite, bituminite and inertinite owie Formation. It is clear that for each well there is a mis- of which, to Cook (1982), commence initial generation modelling which is systematic with respect to stratigraphy and at Ro =according 0.4 to 0.45 per cent, with significant oil generation is similar from well to well (unit thicknesses are similar in at Ro = 0.5 per cent (TTI = 3). This threshold was reached the three wells). at 80-57 Ma B.P., with the exception of Beanbush 1 where, The consistent mis-modelling may be due to an unspecified due to the low gradient, TTI = 3 was reached only at 25 Ma failure of the Lopatin theory, to replicate natural processes. B.P. (early Miocene). Since entering the interpreted main phase However, I feel it is more likely due to the thermal of generation (Ro = 0.5 per cent, TTI = 3) maturation has characteristics of the stratigraphic section. Two over-modelling increased very slowly and only at Moomba 3 has it exceeded peaks occur, within the Bulldog Shale/Wallumbilla Formation Ro = 0.65 per cent or TTI = 15. If the 'most active' phase and approximately at the level of the Toolachee Formation of generation is regarded as commencing at Ro = 0.6 per coals. Both are low-conductivity, insulating sequences. cent, this phase was entered at 20 and 9 Ma B.P. in Moomba Undermodelling, where Ro (observed) is greater than Rmod 3 and Curalle 1 respectively and has not yet been reached at (modelled), occurs in the Jurassic and Early Permian beneath Beanbush 1 and Tartulla 1. these layers. It is suggested that this is evidence of obstruction The importance of maceral type is well shown in the case of heat flow by the low-conductivity shale and coal sequences, leading to a greater thermal maturation—a 'pressure-cooker' of the Murta Member. If macerals such as alginite, sporinite effect—in the underlying Murta Member and Early Permian, and cutinite had predominated, generation would not yet have respectively, than would otherwise be the case. Oil generation commenced in most areas. The Birkhead Formation is characterised by earlyfrom the marginally mature Murta Member source beds (often from a depth of some 1200 m) may be largely due to generating vitrinite (the major component) and lesser resinite, favourable maceral type and the thermal insulation provided but also later-generating sporinite and cutinite giving an by the Bulldog Shale/Wallumbilla Formation, as described extended oil window, using Cook's (1982) figures, from Ro = 0.5-1.0 per cent. Earliest onset of generation from earliestabove. generating macerals took place about 90 Ma B.P, while the main oil generation phase was entered at 85-75 Ma B.P. MostTIMING OF GENERATION active generation, defined as Ro 0.6 per cent, commenced at Interpretation of interval and aggregate maturation indices 82 and 70 Ma B.P. in Moomba 3 and Curalle 1, while from the Lopatin modelling (Table 2) can provide information Beanbush 1 and Tartulla 1 have lagged behind due to low on the timing of generation. It is now generally accepted that formation temperatures. Only in Moomba 3 is the main oil different maceral types have different generation thresholds generating phase currently nearly complete. In all wells, oil and yields. Cook (1982, this volume) has recently published generation was most active during the Late Tertiary-present most valuable data tabulating these thresholds and the day.

Of the four studied wells, only Moomba 3 required a pulse of high gradients in the Permian to Jurassic. Beanbush 1, Curalle 1 and Tartulla 1 all were satisfactorily modelled with constant gradients prior to 5-10 Ma B.P. Thus the pulse may be confined to the Nappamerri Trough.

G. M. P I T T

348

TABLE 2. Maturation timing (Ma B.P.) for selected stratigraphic units in studied wells. Thresholds

TTI Ro

1 0.4

3 0.50

10 0.60

15 0.65

TOOLEBUC FORMATION Beanbush 1 30 Moomba 3 Tartulla 1 35 Curalle 1 80

7.5

COORIKIANA SANDSTONE 53 Moomba 3

MURTA MEMBER Beanbush 1 Moomba 3 Tartulla 1 Curalle 1

25 74 57 80

66 86 80 89

75 1.00

160 1.30

1000 2.0

1500 2.2

Current Maturation TTI Ro(mod)

1.69

0.435

1.54 3.05

0.42 0.50

absent *

3.93

0.51

6.21 18.23 5.77 10.31

0.540 0.680 0.535 0.603

12.74 116 14.64 26.71

0.627 1.175 0.645 0.748

24.76 421 19.68 69.48

0.733 1.65 0.695 0.980

100 1462 45.71 137

1.10 2.19 0.88 1.25

218 2767 76.1

1.41 2.51 1.005

1.2

BIRKHEAD FORMATION — Beanbush 1 Moomba 3 94 Tartulla 1 88 94 Curalle 1

90 75 87

4.8 82 39 70

78 0 60

2.7

'BASAL' JURASSIC Beanbush 1 Moomba 3 Tartulla 1 Curalle 1

90 99 92 99

80 94 82 92

40 90 45 85

15 88 20 78

TOOLACHEE FORMATION Beanbush 1 103 127 Moomba 3 Tartulla 1 97.5 Curalle 1* 103

92 99 92 96

75 93 74 91

64 91 63 89

3.5 84

1.0

—

•Note: Toolachee Formation absent in Curalle 1; Early Triassic at 1737 m depth was modelled. PATCHAWARRA FORMATION Beanbush 1 129 Moomba 3 156 Tartulla 1 101 Curalle 1

98 104 92

88 96 85

82 92 80 absent

In the 'basal Jurassic', late-generating sporinite and cutinite and early-generating, but low-yield, inertinite are the significant macerals in dispersed organic matter (d.o.m.). Vitrinite is a major component of the typical 'basal Jurassic' coals, therefore generation thresholds for the 'basal Jurassic' are significantly higher than for the Birkhead Formation or Murta Member. Thus, although Ro = 0.4 per cent (TTI = 1) was reached at 90-J00 Ma B.P., there was no generation. The main phase of generation (Ro = 0.7?) commenced at about 80-70 Ma B.P. in Moomba 3 and Curalle 1 but only some 15 Ma B.P. in Beanbush 1 and Tartulla 1, if the interpretations of generation thresholds for the 'basal Jurassic? are correct. If so, moreover, the effect of different maceral types in the Birkhead and 'basal Jurassic' can be seen to counterbalance differences in depth and formation temperature. Since the early Miocene, the 'basal Jurassic' in Moomba 3 has been predominantly gas-generative. The mid-Toolachee Formation reached an Ro of 0.5 per cent at about 99-92 Ma B.P. Thereafter, maturation proceeded rapidly at Moomba 3 and Curalle 1, to reach an Ro of 1.0 per cent by 84 and 63 Ma B.P. respectively, due to rapid Winton Formation deposition and high thermal palaeogradients. In Moomba 3, the Toolachee Formation was probably generating significant volumes of oil from ca 93 Ma B.P. until the start of the Tertiary. Wet gas was generated through the Tertiary and Quaternary and the section has been within the dry gas window only over the past million years or so. Increasing maturation at Curalle 1 was arrested thereafter by mid-Tertiary uplift and the Early Triassic is still now late-mature for oil and mature for wet gas. At Beanbush

30 88 0.5

3 82

1.8

1 and Tartulla 1, early maturation timing was similar, but thereafter, due to low palaeogradients, the Toolachee Formation has remained oil- and wet gas-generative. Significant oil generation from the Patchawarra Formation is interpreted to have commenced between 96 and 85 Ma B.P. In Moomba 3, by 82 Ma B.P. Patchawarra Formation .'had become purely wet-gas-generative and this pertained through the Tertiary until about 10 to perhaps 5 Ma B.P. (late Miocene), since which time the unit has been generating dry gas. At Beanbush 1, oil was generated until the mid-late Tertiary; thereafter it has generated wet gas. At Tartulla 1 the unit has been oil- and gas-generative to the present day. As a concluding comment it should be noted how critical the generative characteristics of different maceral types are to generation timing. As the data presented in this paper show, a difference of 0.1 or 0.2% Ro may mean a difference of 15 to 50 Ma in the timing of initial generation. Equally, ignoring the maceral types, the individual's own opinion and prejudices of Ro generation thresholds are important. For instance if one contends that inception of oil generation in the Patchawarra Formation in Moomba 3 took place at 0.4% Ro, it will correspond to a formation depth of 890 m and temperature of 65 °C at 156 Ma B.P. (mid-Jurassic Birkhead Formation time) under the burial and thermal histories postulated here. Instead, if 0.5% Ro is substituted, generation commenced at a formation depth of 1700 m and temperature of 88 °C at 104 Ma B.P. (mid-Albian Toolebuc Formation time). This has important implications for generationstructuring interrelationships in the context of Eromanga and Cooper Basin tectonics.

MATURATION TIMING AND GEOTHERMAL GRADIENTS

THE FUTURE OF MATURATION MODELLING Lopatin-type modelling techniques are regarded with considerable reserve by many workers and caution is certainly justified at present. However the potential of the method, as shown in this paper, is such that it deserves more study and development to determine if that potential can be realised. Currently there are five sources of error in modelling maturation: the burial curve, the accuracy of vitrinite reflectance determinations (particularly when made on nonvitrinite macerals) and compilation of reflectance curves, the nature of the geothermal history, the degree of variation of thermal conductivity in the vertical section and the correlation of TTI to Ro. Moreover, we know little of organic reaction rates or the degree to which the reaction system is open, and of possible reaction catalysts, in the geological environment. To improve the future application of the technique, our knowledge of these aspects must be refined where possible. In this paper some preliminary sensitivity analyses have been made showing the degree to which variation of R and of hitherto constant thermal gradients influence maturation patterns (Fig. 14). The facility of computer-processing of Lopatin modelling should allow more complete sensitivity analyses of all related aspects. This will indicate areas where the method can usefully be improved and, conversely, where variation of input data has little effect. The Lopatin processing programme used by the writer currently assumes an average thermal gradient which is constant through the stratigraphic section, but which may be varied with time. R may also be varied. Lopatin modelling by other authors use a constant R, and temporally and vertically constant, or dog-leg, thermal gradients. Future, 'second generation' Lopatin modelling should take into account the thermally inhomogeneous stratigraphic section to provide a complete profile of the temperature history of all horizons, and should allow for rapid interactive or automatic iterative re-running of the model under specified constraints, to approach the observed Ro curve. A major gap in our knowledge concerns the detailed thermal behaviour of the Eromanga-Cooper Basin section, which contains lithologies of greatly varying thermal conductivities. This situation would be resolved by modern temperature logging from surface to basement of one or more stabilised wells.

THE BOTTOM LINE In the context of the Eromanga-Cooper Basin region, the practical importance of the Lopatin technique lies in the relationship between generation and maturation timing, and that of structuring. This, so to speak, is the bottom line. Pre-Cretaceous structural growth is commonly held to be 'safe'. A structural trap formed at such time would clearly have existed before generation and migration. Tertiary structuring is more contentious. Dry wells on the Tertiary Innamincka, Curalle and Morney structures have reinforced general opinion that they were formed too recently. Stratigraphic considerations suggest that Tertiary structuring was predominantly Oligocene in age (22-36 Ma B.P.) (Pitt, 1978; Wopfner, 1974), although incipient structural growth may have taken place through the Eocene or earlier. The estimated maturation timings given in Table 2 show that, for example, in Moomba 3 and Curalle 1, significant oil generation in pre-Murta Member units occurred prior to such structuring. The failure of Curalle 1 to intersect any

349

substantial oil pools may be attributed to the fact that all preMurta Member units are estimated to have reached 0.65 % Ro before 60 Ma B.P. On the other hand, in Beanbush 1 and Tartulla 1, units down to the Toolachee Formation are, due to their 'cooler' thermal histories, interpreted to be entirely or predominantly oil generative after Tertiary structuring. At Tartulla, a Toolachee (Late Permian) oil and gas discovery, much of the structuring is of Tertiary age (Middleton & Boardman, 1983). Consequently, it is suggested that the Tertiary component of structuring is not prospective where a high thermal history has pertained (considering only oil generation and primary migration)—such as in the Nappamerri Trough. However, in cooler areas, exemplified by Beanbush 1 and Tartulla 1, significant oil-generation post-dated Tertiary deformation. Thus, in such areas, Tertiary structures should be regarded as prospective.

CONCLUSIONS Geothermal gradients Geothermal gradients are particularly sensitive to data quality. Borehole cooling and thermal conductivity variation are important influences and published listings should be critically examined and recalculated on compatible bases before subsequent use. The geothermal gradient normally varies vertically, as well as areally, due to thermal conductivity differences in the stratigraphic column. In the Cooper-Eromanga Basin region, systematic data are available to map gradients of specific stratigraphic intervals. Such a map of 'basement'-to-surface gradients correlates with regional tectonic structure. The influence of the Cretaceous shales as an insulating layer, and of the Jurassic aquifer in lateral heat distribution may be important, and deserve further consideration.

Maturation modelling This paper has shown that first-generation Lopatin modelling of four wells in different structural and thermal domains of the Cooper and Eromanga Basins can produce geothermal gradient histories which result in vitrinite curves closely approximating those measured from samples. The accuracy of the modelling must be judged from future work, but the order of difference from well to well is believed to be realistic. The present work appears to support the views of organic petrologists that there has been a geologically recent rise in geothermal gradients which as yet has had little effect on maturation levels. Generation threshold timings depend on the interaction between source-bed maceral types and temperature histories. Thermal gradients, which broadly describe such temperature histories, vary from 2.5°C/100 m (Beanbush 1) to 4.6-5.5 °C/100 m (Moomba 3) in the four wells modelled here. The Murta Member of the Mooga Formation reached a maturity of 0.5% Ro (nominal inception of significant oil generation) at 25 Ma B.P. in Beanbush 1 and at 74 and 80 Ma B.P. in Moomba 3 and Curalle 1. The Birkhead Formation reached 0.6% Ro at 15 Ma B.P. (mid-Miocene) in Beanbush 1, to 80 Ma B.P. (Mid-Late Cretaceous) in Moomba 3. Dry gas has been generated in the Early Permian of the Moomba area only over the past 2-10 Ma. These figures exemplify the different generation histories pertaining to different parts of the Cooper-Eromanga region. The Arrhenius 'reaction factor' is normally considered to equal 2.0, but in the present study 2.3 or more was required

350

G. M. PITT

for adequate modelling. This may be an artifact of the technique, or due to the method correcting for a thermally inhomogeneous basinal sequence. Alternatively, under geological conditions and geological periods of time, organic reactions may indeed proceed at an accelerated rate of ca 2.3 for every 10 °C increase in temperature. Consistent over- and under-modelling of certain parts of the stratigraphic section is ascribed to enhanced maturation beneath major insulating sequences such as Early Cretaceous shales and Permian coals. Favourable maceral type and insulation provided by the overlying Wallumbilla Formation are regarded as the two critical factors in organic maturation of the Murta Member source-beds. The interplay of Jurassic, Cretaceous, and particularly Tertiary structuring and generation timing is commonly, and correctly, now held as a critical factor in successful exploration for oil- or gas-charged structures in the Cooper-Eromanga Basins. The maturation vs time profiles presented here emphasise the critical nature of (1) differing average

palaeogradients at different locations within the region, and (2) predominant maceral type within the particular source rock, to the geographic variation of generation timing. In areas of higher palaeogradients, such as at Moomba 3 and Curalle 1, significant oil generation preceded Tertiary structuring. However, in 'cooler' areas, exemplified by Beanbush 1 and Tartulla 1, significant oil generation postdated Tertiary deformation. Thus, in such areas Tertiary structures should be regarded as prospective.

ACKNOWLEDGEMENTS

The writer gratefully acknowledges release of data by joint venture Partners in PEL 5 & 6 and ATP 259P, which has made this work possible. Bob Schnell (W.M.C.) compiled the software for the Lopatin maturation programme. The assistance and patience of Helen Brownie and Pam Walker is acknowledged, in the compilation of papers with which the writer has been associated.

REFERENCES BALL, S. M., 1982: Exploration applications of temperatures recorded on log headings— an up-the-odds method of hydrocarbonchanged porosity prediction. Am. Assoc. Pet. Geol. Bull. 66, 1108-23. COHEN, C. R., 1981: Time and temperature in petroleum formation: application of Lopatin's method to petroleum exploration: discussion. Am. Assoc. Pet. Geol. Bull. 65, 1647-48. COOK, A. C., 1982: Organic facies in the Eromanga Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 234-57. CULL, J. P. & DENHAM, D., 1979: Regional variations in Australian heat flow. B.M.R. J. Aust. Geol. Geophys. 4, 1-13. FALVEY, D. A. & DEIGHTON, I., 1982: Recent advances in burial and thermal geohistory analysis. APEA J. 22(1), 65-81. FERTL, W. H. & WICHMANN, P. A., 1977: How to determine static BHT from well loj data. World Oil, Jan. 1977, 105-6. GATEHOUSE, C. G., 1983: Geology of the Warburton Basin in South Australia. S. Aust. Dept Mines Ener. Rep. 255/81 (unpubl.). GRETENER, P. E., 1981: Geothermics; using temperature in hydrocarbon exploration. Am. Assoc. Pet. Geol. Education Course Notes 17. • GRETENER, P. E. & CURTIS, C. D 1982: Role of temperature and time on organic metamorphism. Am. Assoc. Pet. Geol. Bull. M

66, 1124-29. KANTSLER, A. J., SMITH, G. C. & COOK, A. C., 1978: Lateral and

vertical rank variation: implication for hydrocarbon exploration. APEA J. 18(1), 143-56.

KANTSLER, A. J., PRUDENCE, T. J. C., COOK, A. C. & ZWIGULIS,

M., 1983: Hydrocarbon habitat of the Cooper/Eromanga Basin, Australia. APEA J. 23(1), 75-92. KATZ, B. J., LIRO, L. M., LACEY, J. E. & WHITE, H. W., 1982: Time and temperature in petroleum formation: application of Lopatin's method to petroleum exploration: discussion. Am. Assoc. Pet. Geol. Bull. 66, 1150-2. LEBLANC, Y., LAM, H-L., PASCOE, L. J. & JONES, F. W., 1982: A

comparison of two methods of estimating static formation

temperature from well logs. Geophys. Prosp. 30, 348-57. LOPATIN, N. V., 1971: Temperature and geologic time as factors in coalification. A.N. SSSR Izvestiya Ser. Geol. 3, 95-106. (Russian, reviewed in Wright, 1980). LOPATIN, N. V., 1976: Historico-genetic analysis of petroleum generation: application of a model of uniform continuous subsidence of the oil-source bed. A.N. SSR Izvestiya Ser. Geol. 8, 93-101. (Russian, reviewed in Wright, 1980).

MCKIRDY, D. M., EMMETT, J. K., MOONEY, B. A., C o x , R. E. &

WATSON, B. L., 1985: Organic geochemical facies of the Cretaceous Bulldog Shale, western Eromanga Basin, South Australia; in This volume. MIDDLETON, M. F., 1979a: Heat flow in the Moomba, Big Lake and Toolachee Gas Fields of the Cooper Basin and implications for hydrocarbon maturation. Aust. Soc. Explor. Geophys. Bull. 10, 149-155. MIDDLETON, M. F., 19796: A model for bottom-hole temperature stabilisation. Geophysics, 44, 1458-62. MIDDLETON, M. F., 1982: Bottom-hole temperature stabilisation with continued circulation of drilling mud. Geophysics, 47, 1716-23. MIDDLETON, M. P. & BOARDMAN, M. H., 1983: Tartulla—a case history of a Permian discovery; in Permian Geology of Queensland. Geol. Soc. Aust. Brisbane, 335-342. MOORE, P. S. & PITT, G. M., 1982: Cretaceous of the southwestern Eromanga Basin: stratigraphy, facies variations and petroleum potential; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust. Adelaide, 127-44. MOORE, P. S. & PITT, G. M., 1984: Cretaceous of the Eromanga Basin—implications for hydrocarbon exploration. APEA J. 24(1), 358-76. NICHOLAS, E., RIXON, K. & HAUPT, A., 1980: Uncorrected geothermal map of Australia. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1980/66 (unpubl.). O'DRISCOLL, E. S. T. & KEENIHAN, S. L., 1980: The Toowoomba Charleville Lineament in southern Queensland. APEA J. 20(1), 16-24. OZIMIC, S., 1981: Stratigraphic drilling in the Cretaceous Toolebuc

Formation in the southern and eastern Eromanga Basin, 1980. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1981/12 (unpubl.). Notes, Murloocoppie 1:250 000 geological map, SH53-2. S. Aust. Geol. Surv. PITT, G. M., 1982: Geothermal gradients in the Eromanga-Cooper Basin region; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust. Adelaide, 262-83. PITT, G. M., 1978: Explanatory

POLAK, E. J. & HORSFALL, C. L., 1979: Geothermal gradients in the

Great Artesian Basin, Australia. Aust. Soc. Explor. Geophys. Bull. 10, 144-8.

SCHWEBEL, D. A., DEVINE, S. B. & RILEY, M., 1980: Source, maturity

and gas composition relationships in the southern Cooper Basin. APEA J. 29(1), 191-200.

MATURATION TIMING AND GEOTHERMAL GRADIENTS SPICER, H. C., 1942: Observed temperatures in the earth's crust. Geol. Soc. Am. Spec. Pap. 36, 281-92.

THOMAS, B. M. & BROWN, S. A., 1982: Hydrocarbon generation in

the northern Perth Basin. APEA J. 23(1), 64-74. WAPLES, D. W., 1980: Time and temperature in petroleum formation: application of Lopatin's method to petroleum exploration. Am. Assoc. Pet. Geol. Bull. 64, 916-26. WAPLES, D. W., 1981: Time and temperature in petroleum formation: application of Lopatin's method to petroleum exploration: reply. Am. Assoc. Pet. Geol. Bull. 65, 1649.

351

WAPLES, D. W., 1982: Time and temperature in petroleum formation: application of Lopatin's method to petroleum exploration: reply. Am. Assoc. Pet. Geol. Bull. 66, 1152. WRIGHT, N. R. J., 1980: Time, temperature and organic maturation—the evolution or rank within a sedimentary pile. J. Pet. Geol. 2, 414-25. WOPFNER, H., 1974: Post-Eocene history and stratigraphy of northeastern South Australia. R. Soc. S. Aust. Trans. 98, 1-12. WOPFNER, H. & YOUNGS, B. C., 1972: Subsurface faults and recent

earthquakes in the Simpson Desert. S. Aust. Geol. Surv. Q. geol. Notes, 43, 8-11.

Geological Society of Australia Special Publication No. 12, 353-376

Regional groundwater movement, hydrochemistry and hydrocarbon migration in the Eromanga Basin M. A. Habermehl

Division of Continental Geology, Bureau of Mineral Resources, Geology and Geophysics, P.O. Box 378, Canberra City, A.C.T. 2601.

ABSTRACT

In the Eromanga Basin, part of the hydrogeological Great Artesian Basin, confined aquifers occur in continental quartzose sandstones of Jurassic and Cretaceous age. Confining beds consist of siltstone and mudstone, the main one being a marine mudstone of Cretaceous age. Largescale groundwater movement is generally directed to the west, southwest and south in the main part of the basin. Groundwater movement is slow, and residence times of the water are large, but groundwater salinities are low. Hydrochemical differences characterize different aquifers and regional groundwater flow patterns, particularly in the southwestern part. Groundwater derived from the eastern recharge area is characterized by Na-HC0 -Cl and water from the western recharge area by Na-Cl-S0 . Many of the Jurassic-Lower Cretaceous fine-grained sediments are sufficiently mature source rocks to have generated hydrocarbons, and most of the Jurassic-Lower Cretaceous sandstones are reservoir rocks in which several commercial and sub-commercial oil and gas discoveries have been made. Migration of these hydrocarbons has probably been influenced by the artesian groundwater flow. Gas samples taken at the surface from flowing artesian water wells contain variable amounts of light hydrocarbons. Methane concentrations range up to 510 000 microlitres per litre gas, and the sum of the hydrocarbons ethane to heptane range up to about 14 000 microlitres per litre. The dissolved hydrocarbons transported in the artesian groundwater are probably derived from adjoining source rocks or from existing hydrocarbon accumulations. 3

INTRODUCTION

This paper reviews the regional groundwater hydrology and hydrochemistry of the Eromanga Basin, and reports the results of a pilot study on hydrocarbons dissolved in artesian groundwater. Groundwater hydrodynamics and hydrochemistry are significant factors in hydrocarbon exploration, and this paper is an attempt to contribute to a better understanding of the influence of groundwater movement on the migration and entrapment of hydrocarbons, and on the preservation and destruction of oil and gas accumulations in the Eromanga Basin. The Eromanga Basin consists of Lower Jurassic to Upper Cretaceous sedimentary rocks and underlies parts of Queensland, New South Wales, South Australia and the Northern Territory (Fig. 1). It is a constituent sedimentary basin of the hydrogeological Great Artesian Basin (G.A.B.), and comprises about two-thirds of the area and most of the stratigraphic sequence of the G.A.B., which occupies about one-fifth of the continent (Habermehl, 1980). The G.A.B. is located in arid and semi-arid parts of Australia, and stretches from the Great Divide across lowlying interior plains to the main topographic depression near Lake Eyre. Flowing artesian water supplies are of sufficient quantity and quality to provide for pastoral, domestic, and town water needs. Flowing artesian water wells occur mainly in the near-marginal areas of the basin and are nearly absent in the central part, where the main Lower Cretaceous-Jurassic aquifers occur at great depth. In this area, where the drilling

of deep water wells is not economic, many non-flowing wells obtain water from the shallower, but more saline, Upper Cretaceous aquifers. The aquifer system is multi-layered and consists of Jurassic and Cretaceous quartzose sandstones of continental origin (Habermehl, 1980). Intervening confining beds are siltstones and mudstones; a thick, marine, argillaceous sequence of Cretaceous age forms the main confining unit. The basin is uplifted and exposed along its eastern margin and tilted southwest. Recharge occurs mainly in the eastern marginal zone, and minor recharge takes place at the western margin. Natural discharge occurs from springs at the southwestern, western and southern margins; most springs are associated with structural features (Habermehl, 19826).

Geology

The Eromanga Basin consists of a conformable sequence of Lower Jurassic to Upper Cretaceous sedimentary rocks (Parkin, 1969; Wopfner et al. 1970; Exon & Senior, 1976; Vine, 1976; Senior et al., 1978; Burger & Senior, 1979). The basin is mainly a broad downwarp trending northeast. It is contiguous with the Surat and Carpentaria Basins across shallow ridges and platforms of older sedimentary, metamorphic or igneous rocks, represented by the Euroka Arch, Nebine Ridge and Cunnamulla Shelf (Fig. 1). The Mesozoic sedimentary sequence in the central part of the Eromanga Basin reaches a maximum total thickness of about 3000 m. y

M. A. HABERMEHL

354 136°

146°

in places wedges out against shallow, buried basement structures (Figs 1, 4). The Eromanga Basin is partly concealed by widespread Cainozoic continental sediments which range from relatively thin veneers to strata several hundreds of metres in thickness, forming shallow basins. They rest on the deeply weathered erosional surface of the Cretaceous sediments; the Tertiary sediments are themselves weathered and silicified; Quaternary sediments are mostly unconsolidated. Tertiary basalts cover some areas in the northeastern part of the Eromanga Basin.

Structure

Fig. 1. Location map of Eromanga Basin, showing intermediate ridges and underlying basins (after Habermehl, 1980, 1982/7, c).

Early geological and hydrological investigations have been listed by many authors including Senior et al. (1978) and Habermehl (1980, 1982a). Geological surface mapping by the Bureau of Mineral Resources, Geology and Geophysics (BMR) and State Geological Surveys during the 1950s to 1970s, provides basin-wide coverage. Subsurface information is available from several hundred petroleum exploration wells drilled by private companies, geophysical surveys carried out by companies and government authorities, stratigraphic holes drilled by BMR and State Geological Surveys, and from water wells. Wire-line logs from water wells provide additional sources of data (Habermehl & Morrissey, 1980, 1983).

Stratigraphy

The stratigraphic succession in the Eromanga Basin, excluding local marginal facies equivalents and minor units, is summarized in Figures 2 and 3. Correlation of these rock units with those of the Bowen, Galilee, Surat and Carpentaria Basins, which together constitute the Great Artesian Basin, is shown in Figure 3 and Habermehl (1980, table 1). The Jurassic sequence comprises continental quartzose sandstone, with lesser siltstone and mudstone. Siltstone, mudstone and lithic sandstone were laid down in shallowmarine environments during Early Cretaceous times. During the Late Cretaceous sandier sediments were deposited in lacustrine and fluviatile environments. Most units crop out in the eroded eastern margin of the Basin which was uplifted during Cainozoic times, and some units are exposed along the western margin. Cretaceous outcrops in the central part are markedly weathered. The Jurassic rock sequence thins and

The Eromanga Basin is asymmetrical, elongated northeastsouthwest, and tilted towards the southwest. The southern and northeastern margins have dips of about 2°, but in the west and southwest the regional dip increases to about 5°. Cainozoic and earlier uplifts along the eastern margin, and subsidence in the central and^. southwestern parts, led to the basin's asymmetry. The broad structures in the basin are shown in Figure 4. Many of the near-surface folds, particularly monoclines grade downwards into faults and are the products of draping and differential compaction of sediments over fault-bounded basement blocks. Major fault and fold systems extend across the basin, in places forming en echelon structures. Displacements of up to 300 m along major faults, affect major aquifers in the Eromanga Basin. Displacement of Jurassic-Cretaceous sediments along normal faults is usually much less, but there is major faulting of some PermoTriassic and older sediments. Hydrodynamic conditions in the Eromanga Basin, considered on a basin-wide scale, will not generally be influenced by folding of the sedimentary sequence. Faults, however, are locally significant, in that they disconnect aquifers and elsewhere create hydraulic connections between different aquifers. They can act as pervious or impervious structures to horizontal or vertical groundwater movement. In general they do not significantly affect the broad regional flow patterns.

GROUNDWATER HYDROLOGY Hydrogeological units

Aquifers in the Eromanga Basin are present in the Precipice, Hutton, Adori and Hooray Sandstones, in the Cadna-owie Formation and their equivalents, and in the Mackunda and Winton Formations (Figs 2, 3). The major confining beds are the Evergreen, Birkhead, Westbourne, Wallumbilla and Toolebuc Formations and their equivalents, the Allaru Mudstone, and parts of the Mackunda and Winton Formations (Figs 2, 3). The aquifers are generally well separated from high salinity water in older units, though in some areas the low-permeability Triassic Moolayember Formation and the low-salinity aquifers in the Clematis Sandstone underlie the Jurassic sequence. Elsewhere, hydrogeological 'basement' comprises impervious sedimentary, metamorphic or igneous rock, forming an aquiclude or aquifuge. The Jurassic and Triassic aquifers form part of the Great Artesian Basin, as the Great Artesian Basin consists of confined aquifers and confining beds throughout the Middle Triassic to Late Cretaceous sedimentary sequence in the constituent sedimentary Bowen, Galilee, Eromanga, Surat and Carpentaria Basins (Figs 2, 3; Habermehl, 1980). Most of the aquifers are continuous and hydraulically connected across the constituent geological basins. A simplified aquifer model

355

GROUNDWATER AND HYDROCARBON MIGRATION STRATIGRAPHIC UNIT

LITHOLOGY

AGE

METRES

AQUIFERS

AQUIFER MODEL

VWN/WWWWWWW*

• Potentiometric surface J aquifer — Ground surface • Potentiometric surface ~ K aquifer Regional watertable

Winton Fm K aquifer Mackunda Fm

Toolebuc Fm Coreena Mbr

t.r

Doncaster Mbr Cadna-owie Fm Hooray Sst Westbourne Fm Adori Sst J aquifer

Birkhead Fm

Hutton Sst Evergreen Fm Precipice Sst Moolayember Fm Clematis Sst

Sandstone

Conglomerate

Chemically

SUtstone

Calcareous

Conformable v/\/\/\/\/v

altered

Unconformable

Fig. 2. Schematic lithostratigraphic c o l u m n , E r o m a n g a Basin, s h o w i n g s i m p l i f i e d aquifer groups and hydraulic potentials o f main aquifers (after H a b e r m e h l , 1980, 1982c).

(Figs 2, 5) shows the geometry of the main 'J' and 'K' aquifer groups, i.e. the aquifers in the Lower Cretaceous-Jurassic and Cretaceous sequence, respectively. As a result of the onlapping nature of the Jurassic and Lower Cretaceous sequence, migration of water occurs from aquifers in the lower part to aquifers higher in the sequence, where these units subcrop against impervious basement, as along the northern margin of the Thargomindah and Cunnamulla Shelves. Most of the lithostratigraphic units have relatively uniform gross geometry, lithology, structure, texture and depositional origin throughout the basin, though in detail large variations occur. Similar generalities apply to the hydraulic characteristics of the aquifers and confining beds if groups of aquifers are considered, which correspond broadly to aggregates of individual aquifers occurring in lithostratigraphic units. Separate aquifers within individual sandstone units in detail show marked differences laterally and vertically in hydraulic properties, especially the permeability and porosity of the rocks.

Hydraulic characteristics of aquifers and confining beds Transmissivity values determined from periodic systematic tests carried out by the State water authorities on flowing artesian water wells in Queensland and New South Wales since the early development of the G.A.B., generally range from 1 to 2000 m2 day - 1 . Low values, usually less than 20 m2 day -1 , predominate in the south-central, higher values of 10s to 100s m2 day are present in the northern and southern parts (Habermehl, 1980, table 1). Hydraulic conductivity values range from 10 1 to 10 m day - 1 , the majority being in the lower part of that range. Almost all wells are completed in aquifers in the Cadna-owie Formation and Hooray Sandstone and their equivalents, which therefore provide most of the available data. Storage coefficient values of the aquifers could not be determined from the available water well data. The relationship between aquifer porosity and storage coefficient was used to derive the latter from sonic logs run in petroleum

M. A. H A B E R M E H L

356 EROMANGA BASIN EROMANGA BASIN SW (SA) W,NlQLD,NT,SA)

CARPENTARIA BASIN (QLD)

EROMANGA BASIN Central, E£(QLD, NSW)

SURAT BASIN (QLD,NSW)

COONAMBLE EMBAYMENT (NSW)

.; I Confined t . 1 aquifer j Confining

Aquifers tapped generally produce non-flowing artesian wells

Winton Fm'

:Mt'.Alexond*r. •••••Sst-Mbr.v.-

Sst

Bulldog Sh . • . : • :MtArindB6rMx Cadna-owie Fm

MhL

Mbr

1 1 1 Don- JoQesV coster Mbr^Qd

Cadna-owie Fm

Griman Creek .Fm.

Aliaru Mdst

Toolebuc Fm

WookJridqe Lst Mbr Coorikiana

- ?

Normonton Fm Allaru Mdst

Toolebuc Fm Wallumbilla Fm

-.Gilbert- River'. Fm';.

l i i Minmi Mbr

: IWVondrd Sst'-MfrT C a d n a - o w i e Fm

'. Burigil 'Fm '•

Kinqull Mbr

• ; M o o g a Sst

.'• M o o g a S s t . Algebuckina Sst '. . ;

•. -.'

Orallo Fm

V. Hooray. Sst;

'• Hooray .Sst;.' •

Aquifers rapped generally produce flowing artesian wells

; Gubbei'omunda . Sst: • Westbourne Fm Lth • Ado ri Sst • Birkhead Fm |HSst u

11111111,1ITTTTT

Westbourne Fm

Westbourne

.Ajgri • $$t :

Hompsteod I :Hompsteod:Sst."'.

Evergreen^Boi^e-Sst'.Mbf • Precipice Sst

Moolayember

'•*•'.'.;' Pilliga.Sst

Walloon Coal Measures

•Btrkhead-Fm ••

. Hutton". Sst! ;.';

XO/sAAsiAAA/V/

••Spnrioboh-.Ssafy?

•j

Evergreen Fm

Moolayember Fm

Purlawaugh Fm

ibafi

Moolayember Fm

Garrawilla Voles j \ j \ / \ / \ / \ r ^ / \ r \ j \ r \

Napperby

Bds

(Wallingarah Cr Fm) (subsurface)

Clematis Sst'.; ///jDunda"Bds l / /

/ / / 7 / / A

. . . . .Dioby.-.Bds V . : •'; •.•.•. ( W o n a r - S s t ) - '

• '. Clematis; Sst-

Rewan Gp

GALILEE BASIN (QLD)

GALI LEE BOWEN BASIN (QLD)

BOWEN BASIN (QLD/NSW)

Black J a c k

Aquifers tapped generally produce flowing artesian wells

GUNNEDAH BASIN (NSW)

/ / / / / Hydrogeological basement

Fig. 3. Correlation of hydrogeological units in Eromanga Basin and other constituent basins of the Great Artesian Basin (after Habermehl, 1980, 1982a, b, c).

exploration wells in the basin. Storage coefficient values average about 10" 5 . Intrinsic permeability ranges from several tens to several thousands of millidarcys. Porosity values range from about 10 to 30% (Habermehl, 1980, table 1); both decrease with depth. Average vertical hydraulic conductivities of the leaky, very low permeability, confining beds range from 10 1 to 10 4 m day" 1 .

Recharge Outcrop areas of aquifers provide the major recharge to the aquifer units and occur mainly along the eastern margins of the G.A.B. (Figs 6, 7). Outcrops of aquifers and subcrops of aquifers in the Eromanga Basin sequence, overlain by sandy sediments along the western margins of the basin, provide the possibility of recharge in those areas (Fig. 8.) Recharge has been interpreted from the distribution of potentials. Habermehl (1980) showed that the G.A.B. recharge computed by the G A B H Y D mathematical model represent only about 2% of the amount of water available for recharge from rainfall

(Habermehl & Seidel, 1979; Seidel, 1980). Williams & Coventry (1981) in their recharge studies in the northeastern margin of the Eromanga Basin found that deep drainage through soils in the area could contribute a significant part of the estimated annual recharge. Studies of environmental isotopes (include D / H , 1 3 C/ , 2 C, 14C, 1 8 0 / 1 6 0 , and 36C1) and hydrochemical characteristics of the groundwater in areas basinwards of the recharge areas support the delineation of recharge areas, and the assumption of continuing recharge from geological to modern times (Airey efal. 1979, 1983; Calf & Habermehl, 1984; Habermehl, 1983). Stable i s o t o p e ( D / H a n d 1 8 0 / 1 6 0 ) results show unequivocally that the artesian water is of meteoric origin. Observed systematic variations in chloride, sodium and bicarbonate levels of water samples with distance from the recharge area probably reflect variations in the mean annual rainfall and the rate of infiltration of recycled salt throughout the Late Quaternary; the minimum and maximum of the chloride values correlate with the last glacial and interglacial period respectively (Airey et al., 1979).

GROUNDWATER AND HYDROCARBON MIGRATION

•www

Boundary between Codna-owie Formation/Hooray Sandstone and Mooga Sandstone

Fauif

2Q0—

Contours (m), datum M S L, contour interval

357

200m

Fig. 4. Structure contour map, base of Rolling Downs Group (Wallumbilla Formation) and top of upper main aquifer in Lower CretaceousJurassic sequence (after Habermehl, 1980, 1982c).

Discharge

Discharge from the G.A.B. takes place naturally in the form of concentrated outflow from springs, vertical leakage towards the regional watertable, subsurface outflow into neighbouring basins, and artificially by means of free artesian flow and pumped abstraction from wells , drilled into the aquifers. Springs and areas of seepage are abundant at the margins of the Eromanga Basin (Fig. 7). About 600 springs in the G.A.B. are concentrated in eleven groups, and nearly all are located in the Eromanga Basin (Habermehl, 1982b). Artesian springs are generally associated with (1) faults along which the water flows upwards, (2) abutment of aquifers in the Jurassic and Cretaceous sequence against impervious bedrock, or (3) pressurized water breaking through thin confining beds near the discharge margins of the basin. Many artesian springs have built conical mounds by mechanical deposition of particles from pressured aquifers and confining beds, and by chemical precipitation of solids dissolved in artesian groundwater. Artesian springs and their deposits in the Lake Eyre region show a range from topographically high springs to younger, topographically lower springs as a result of the lowering of the landsurface and

spring outlet levels in Quaternary times (Habermehl, 1980, 19826). Lowering of spring outlet levels has resulted from stepwise lowering of the land surface by erosion and denudation, and the breakthrough of water at a lower level. The latter causes a progressive lowering of the pressure heads in the spring areas, and reduced flow from higher springs. Clogging of the upper outlets of springs by sediments and (secondary) carbonate leads to breakthrough and discharge of water at a lower level, with a similar effect on the hydraulic conditions. Fossil spring deposits, of possible Pleistocene age, rise several tens of metres above the present land surface where active springs now occur (Habermehl, 19826). The hydraulic gradient had steepened as a result of lowering of outlet levels in Late Quaternary times, before man caused even more significant changes to the hydraulic gradient. Rate of movement and throughflow of groundwater in the basin as a result of steeper hydraulic gradients is therefore higher than at any previous time, which contrasts with some of the conclusions presented by Bowering (1982). Discharges from springs generally are small and most springs produce much less than 10 L s~'; few have larger discharges, which range up to 85 L s . The accumulated discharge of about 600 - 1

M. A . H A B E R M E H L

-3o°

Lake Frome

S A

Lake Torrens

\j^Lake Gairdner

•Broken Hill

NSW

A' MS L

Confining bed Cretaceous aquifer Confining bed Jurassic aquifer Basement

Boundary between confined aquifers and confining beds Watertable

Potentiometric surface Jurassic aquifer, 1880

Fault

Petroleum exploration well

Potentiometric surface Jurassic aquifer, 1970

Topographic contour (m) above MSL

Fig. 5. Map and cross-section showing lateral extent of simplified hydrogeological units (after Habermehl, 1980).

AUS1/908

GROUNDWATER A N D H Y D R O C A R B O N M I G R A T I O N

Fig. 6. Potentiometric maps, main aquifers in Lower Cretaceous-Jurassic sequence, 1880 and 1970 (after Habermehl, 1980, 1982c).

360

M. A. HABERMEHL

•

G r e a t A r t e s i a n Basin spring

Springs from Tertiary sediments and

basalts

Fig. 7. Recharge and natural discharge (springs) areas, and directions of regional groundwater flow (Habermehl, 1980, 19826, c).

springs is estimated at about 1500 L s _ l . Discharges have declined since water well development started in the basin. Vertical leakage from aquifers upwards through semipervious confining beds occurs throughout most of the basin and, despite the low percolation rate, involves a considerable volume of water. High evaporation rates and a deep phreatic surface (generally several tens of metres below the ground surface), largely conceal vertical leakage. Geological evidence suggests that subsurface outflow into any of the surrounding or underlying basins is negligible, with the exception of outflow directed northwards across the Euroka Arch, into the southern part of the Carpentaria Basin. Most groundwater from the Eromanga Basin has been exploited from flowing artesian water wells since artesian water was discovered in 1878 and development commenced (Habermehl, 1980, 1982tf). Wells are up to 2000 m deep, but average about 500 m. Flows from individual wells exceed 10 000 m 3 day - 1 (more than 100 L s~ l ), but the majority have much smaller flows. About 3100 of the 4700 flowing artesian wells drilled in the G.A.B., of which a large part are located in the Eromanga Basin, remained flowing during the 1970s and early 1980s. The accumulated discharge of these wells was about 1.5 x 106 m 3 day 1 compared to the

maximum flow rate of about 2 x 106 m 3 day - 1 from about 1500 flowing artesian wells around 1918. Flowing artesian water wells obtain their water from aquifers in the Lower Cretacous and Jurassic sequence (mainly the upper aquifers); the original non-flowing artesian water wells generally tap aquifers in the Winton and Mackunda Formations. These (about 20 000) non-flowing wells are generally shallow, and several tens to hundreds of metres deep. It is estimated that these generally windmill-operated pumped wells supply, on average, 10 m 3 day" 1 . High initial flow rates and well pressure have diminished as a result of the release at high pressure of water from 'elastic storage' in the groundwater reservoir. Exploitation of the basin's aquifers has caused significant changes in the rate of various discharges in time (Habermehl & Seidel, 1979; Habermehl, 1980; Seidel, 1980). Prior to development, the basin was in a natural steady-state condition, with equilibrium between recharge and natural discharge. Following development, natural discharge diminished. A visible effect has been the diminution in flow from springs in the south-central and southwestern parts of the basin. Abstraction by wells caused a steepening of the hydraulic gradient and allowed more recharge water to enter the system. At present a new steady-state condition has been

21 133°30 onn

Potentiometric surface contour (metres above msl) of the mam aquifers in the Lower Cretaceous-Jurassic sequence Regional direction

groundwater

flow

Flowing

artesian

spring

Flowing

artesian

waterwell O

Petroleum

exploration

Basement

Group

Hydrochemistry

well

c 2: a 1

m po >

l U l

Na- HCO3-CI

Z O

f|||

Na-Cl-S0 4

X •

ill! ||| Mill'

Na-CI-HCO3

Na-CI-HCO3-SO4 Na-Ca-Cl-S0 4

o o > yO CO O Z

6 g r o u p s w i t h different || 11| 11| chemistry

o 73

I I I I I I I I il

Na-Mg-CI-S0 4 -HC0 3

362

M. A. HABERMEHL

reached in which total recharge and discharge are approaching equilibrium again (Habermehl & Seidel, 1979; Habermehl, 1980; Seidel, 1980).

Groundwater levels The potentiometric surface of the aquifer in the Lower Cretaceous and Jurassic sequence was above ground surface over the whole of the basin before exploitation began. Since then, the regional potentiometric surface of the exploited aquifers has dropped by several tens of metres in many heavily developed areas (Habermehl, 1980, fig. 7). It is still above ground level throughout most of the basin, through in other areas flows from artesian wells ha^e ceased and water has to be pumped. The potentiometric surface of aquifers in the upper part of the Cretaceous sequence has always been below ground surface, consequently wells tapping these aquifers are non-flowing artesian, and have to be pumped. Potentiometric maps derived from the GABHYD computer simulation model (Seidel, 1980; Habermehl, 1980), show the conditions during early years of development and the 1970s for the main aquifers in the Lower Cretaceous-Jurassic sequence (Fig. 6). Water wells are irregularly distributed, with dense concentrations in some areas and few or no wells elsewhere. The model results provide consistent, simplified and regularly spaced sets of data points across the basin, including extrapolated data for areas where no wells are present. Development has led to considerable changes in the patterns of isopotential contours. Habermehl & Seidel (1979) and Seidel (1980) show predicted drawdowns, changes in discharges and potentiometric maps as a result of possible future developments. Hydraulic gradients of aquifers in the Lower Cretaceous and Jurassic sequence changed significantly in some areas and are about 1:3 000 in the southwestern-central part of the basin. Changes in aquifer potentials in the upper part of the Cretaceous sequence are much less as withdrawals are relatively small. Hydraulic gradients of aquifers in the Lower Cretaceous and Jurassic sequence changed in some areas from 1:2150 to.1:1750, and from 1:2300 to 1:1600. Hydraulic gradients of aquifers in the upper part of the Cretaceous sequence are about 1:1800.

GROUNDWATER MOVEMENT Regional groundwater movement has been interpreted from the potentiometric surface contour maps of the main aquifers in the Jurassic and Lower-Cretaceous sequences. Flow directions in the central part of the Eromanga Basin are generally towards the west, southwest and south. In the western part of the basin regional groundwater movement is towards the southeast and south (Fig. 7). Groundwater movement is slow, and based on hydraulic data probably around 1 m year "1, as hydraulic gradients are low and porosities high. The long residence time of the water has been confirmed by environmental isotope studies (Airey et al., 1979, 1983; Calf & Habermehl, 1984). New techniques using chlorine-36 for the dating of this very old groundwater are being evaluated, as the ages (several 100 000 years) are outside the range of conventional carbon-14 techniques. Residence times between 10 000 and 30 000 years have been determined for groundwater downgradient of the main eastern recharge area (Calf & Habermehl, 1984). Further basinward, ages gradually increase, and an age of about 1.4 x 106 years was established for groundwater from a well near the QueenslandSouth Australian border. Chlorine-36 isochrons were found to show good agreement with groundwater residence times

calculated from hydraulic data (Airey et al., 1983). Hydraulic data suggest that artesian groundwater in the southwestern part of the basin near the discharge areas, could have entered the eastern recharge areas of the aquifer system about two million years ago.

GROUNDWATER CHEMISTRY Most of the basin's artesian groundwater in the main aquifers in Jurassic and Lower Cretaceous sequences contains usually about 500 to 1500 mg L" 1 total dissolved solids, with some wells showing salinities of more than 2000 mg L~'. The water is chemically of the sodium-bicarbonate-chloride type, and these ions contribute more than 90% of the total ionic strength of solutes in the main basin area. Near the recharge areas calcium, m a g n e s i u m and sulphate concentrations are proportionately higher, but these decrease basinward. Variations of the major ion concentrations and ratios occur along the flowlines. Sodium and bicarbonate concentrations generally increase along the flowlines in most parts of the basin. Chloride and sulphate concentrations also increase in most areas, though an initial decrease occurs basinward of the northeastern marginal area. Along the southerly directed flowpaths from the northeastern recharge area to the New South Wales border, concentrations of all four ions decrease, and then increase or remain approximately constant. In the centre of the basin some deep wells with high chloride concentrations occur, and very low to semi-stagnant flows along deep and long flowpaths might account for these high chloride concentrations and high salinity values. In the northwestern part sodium, bicarbonate and chloride increase along the flowlines, but sulphate concentrations are constant. The increase of bicarbonate and decrease in sulphate concentrations in the basin might result from biochemical reduction of sulphate which produces carbon dioxide and hydrogen sulphide. Groundwater in the basin has low sulphate concentrations, and hydrogen sulphide and carbon dioxide are common gases in many wells. Gradations in total dissolved solids values occur within the aquifers with a general increase along the flowlines of the groundwater (Habermehl, 1980, fig 17; Habermehl, 1983). Groundwater from flowing artesian water wells located approximately along a flow line (on the Birdsville Track, between about 2 7 ° a n d 30°S, and 138°and 139°E) show this distinct increase in salinity, and also an increase in chloride component, towards the discharge zone near the basin margin. In the southwestern part of the basin, groundwater is characterised by sodium-chloride-sulphate. Carbon-14 and the stable isotope ratio I 3 C/ , 2 C indicate the recharge area, and, basinwards, show trends similar to those downgradient of the main, northeastern recharge area (Calf & Habermehl, 1984). The groundwater, which before recharge flowed as surface water over outcropping evaporitic and gypsiferous rocks near the western recharge area, meets and mixes with west and southwest moving water derived from the eastern recharge area of the basin (Fig. 7). Regional groundwater flow patterns interpreted from potentiometric surface contour maps are in good agreement with observed changes to the hydrochemistry of springs and flowing artesian water wells in the southwestern part of the basin (Fig. 8). The two major regional groundwater flow directions show different hydrochemical characteristics, with westward flowing water being of the sodium-bicarbonatechloride type and eastward flowing water being of the sodiumchloride-sulphate type. These flows, within the same aquifer system meet and mix, and are directed towards the main discharge area near the basin's southwestern margin. This

CL 3 * SL

0 ?>o c

c z

o 3 3 fo sa 3

U ^

OQ C

1 m

DO q T

AUSTRALIA f"

a • t> rmtmS^}

O W ri o x r v 5 3 3 3 a. <T> £

70 O o > 70 CD O

o 70

% O z

Direction Fault

of regional

or LANDSAT

Anticline Syncline

groundwater interpreted

Area

flow linear

feature

of possible

Flowing

artesian

groundwater waterweU

stagnation

364

M. A. HABERMEHL

combined groundwater flow is distinguished by a mixed hydrochemistry, dominated by sodium-chloride-bicarbonate and sodium-chloride-bicarbonate-sulphate type water. Poor quality water of much higher salinity occurs in aquifers in the Winton and Mackunda Formations throughout the basin, and is chemically of sodium-chloride-bicarbonate type. The flowing artesian waterwells sampled for this study are shown in Figure 10 (compare Fig. 2 for aquifer identification). Most flowing artesian water wells in the area obtain their water from aquifers in the Cadna-owie Formation and Hooray Sandstone. The chemistry of major elements in groundwater of the flowing artesian water weHs sampled during 1980 is shown in Figure 11. Generally the groundwater chemistry of individual aquifers is rather uniform, though slight differences occur in the hydrochemistry within a single aquifer. Groundwater from different aquifers in the Cretaceous and Jurassic sedimentary sequence can be identified by their particular hydrochemical composition. Groundwater from the Cretaceous aquifers (Winton and Mackunda Formations) exhibit sodium-chloride (-bicarbonate) type chemistry with relatively high total dissolved solids values. The Lower Cretaceous-Jurassic aquifers produce groundwater which is characterized by sodium-bicarbonate (-chloride) type water. Total dissolved solids values of the latter are relatively low and range from about 260 to 3917 mg L" , which generally represents fresh to brackish water (Collins, 1975; Matthess, 1982), and pH values are between 7.5 and 8.6 (Fig. 11). A larger sulphate component occurs in the northern part of the area (Fig. 11), possibly caused by the subcrop of evaporites in the Devonian Etonvale Formation of the Adavale Basin, which underlies part of the lowest aquifer in the Jurassic sequence. Elsewhere, a thick sequence of tight, cemented rocks of Devonian to Triassic age usually separates the evaporites from the bottom aquifer in the Jurassic sedimentary sequence. Chemical data were analysed by a multi-variate statistical computing technique, which allowed grouping of samples (water wells) into subsets that are related by the dissimilarity of variables measured. Seven groups were differentiated from the most significant variables (Fig. 12). The majority of sampled wells are contained in the first two groups in Figure 12. Five other groups are present, four of which consist of only one well, and one of three wells. The basis for separation of the two main groups is a difference in the ions sulphate, silica and bicarbonate in these samples (Fig. 13). The Lower Cretaceous-Jurassic aquifers show generally increasing amounts of total dissolved solids along their flowpaths. Several wells completed in Cretaceous aquifers produce water with the type of hydrochemistry expected (e.g. RN 357 and 7311), but some water wells completed in Cretaceous aquifers (RN 3945, 3947, 3950 and 15671) produce sodiumbicarbonate-chloride dominant water, which is normally associated with Jurassic-Lower Cretaceous aquifers (Figs 11, 12). These wells are located a short distance west of the Canaway Fault, and the 'anomalous' hydrochemistry suggests that upward leakage occurs within the fault zone from the Cadna-owie Formation and Hooray Sandstone east of the fault into the Winton Formation west of the fault. Polak & Ramsay (1977) differentiated the quality of the groundwater in these aquifers by resistivity methods. Geochemical changes in the groundwater probably result from interaction with the aquifer mineralogy, particularly through ion exchange with clay minerals, which specifically affects sodium ratios. Chloride concentrations are generally little-affected except by input variations in the recharge area. Bicarbonate levels are similarly largely determined by processes 1

in the recharge area. Following the initial chemical evolution of the groundwater within and close to the recharge area, few variations affect the more distant groundwater chemistry within the aquifers though residence time's of this water in the predominantly quartzose, lithic and clayey sandstones are very long, and contact has taken place at elevated temperatures. The minor basinward changes to the hydrochemistry possibly reflect relative chemical inertness of the aquifer rocks. Few data are available on the mineralogy and geochemistry of aquifers and confining beds, and further work is required to determine their interaction and controls on water chemistry. The major chemical data have been plotted in a modified Durov diagram (Fig. 14), which despite some disadvantages, is considered adequate for preliminary grouping of water types (Lloyd, 1965; Matthess, 1982). Figure 14 emphasizes the uniformity of groundwater chemistry within Lower CretaceousJUrassic aquifers, and the apparent lack of chemical evolution during the long movement of groundwater through the aquifers (apart from a small increase in salinity). High salinity and chloride values are mainly related to Cretaceous aquifers, and the high sodium-chloride values in these aquifers probably reflect non-flushing characteristics and the marine origin of some adjoining Cretaceous sediments. Some wells in the Lower Cretaceous-Jurassic aquifers show high salinity and chloride values, and this could probably be attributed to restricted circulation and the result of metasomatism of groundwater.

Groundwater temperatures

Water temperatures in wells tapping aquifers in the Early Cretaceous and Jurassic sequence generally range from about 30° to 100 °C at the surface. Spring ground surface temperatures range from about 20° to 45 °C. Geothermal gradients calculated from temperature logs in water wells in the basin range from about 15 °C k m to 100°C k m with a mean of 48°C km" (Polak & Horsfall, 1979); Cull & Conley (1983) give a mean gradient of 38.6 ± 0.1°C k m . Nicholas et al (1980), Senior & Habermehl (1980) and others give geothermal gradient values from some petroleum exploration wells. Pitt (1982) and Cull & Conley (1983) consider from petroleum well data that water well values, though internally consistent, are too high. High gradients occur where shallow basement rocks are found, adjacent to some major faults, and near some discharge areas. -1

-1

HYDROCARBON OCCURRENCE, MIGRATION AND ENTRAPMENT

Hydrocarbon source rocks are abundant in the fine-grained clastics of the Eromanga Basin sequence. These rocks are sufficiently mature in some parts of the basin, because of their depth of burial and the past and present geothermal gradients, to have generated hydrocarbons (Senior & Habermehl, 1980; Cook, 1982; Kantsler et al., 1982; McKirdy, 1982; Passmore & Boreham, 1982; Pitt, 1982). Commercial and sub-commercial discoveries of oil and gas reservoirs have been made in several Jurassic and Cretaceous sandstones in recent years (Moore & Mount, 1982), contradicting earlier beliefs that the basin-wide groundwater flow had flushed hydrocarbons out of the sequence. Natural seepages of hydrocarbons have not been recorded in the basin, though Bowering (1982) reports on hydrocarbons obtained from the Coward Springs borehole in a spring deposit near the southwestern margin of the basin. Most flowing artesian water wells in the basin yield water which

Regional direction

Flowing

groundwater

artesian

waterwell

Cretaceous

aquifers

flow

tapping:

Lower aquifers

Cretaceous-Jurassic

Triassic

aquifers

Registered number (Queensland Water Resources Commission)

GROUNDWATER AND HYDROCARBON MIGRATION 365

Fie 10 Potentiometric surface contours, showing regional groundwater flow directions in the main aquifers in the Lower Cretaceous-Jurassic sequence, and flowing artesian water wells in t h e central Eromanga Basin sampled in 1980.

Potentiometric surface contour (metres above msl) of the mam aquifers in the Lower Cretaceous-Jurassic sequence

366

% total equivalents per million (epm)

Regional groundwater flow direction Flowing

artesian

Registered (Queensland Resources

SOUTH QUEENSLAND AUSTRALIA

waterwell

number Water Commission)

M. A. HABERMEHL

Fig. n. Chemical composition of groundwater from flowing artesian water wells shown in Fig. 10.

Total sum of above ions in mg/i fppmj

Flowing

Regional groundwater direction

flow

artesian waterwell

tapping:

Cretaceous

Lower aquifers

Cretaceous-Jurassic

aquifers

•

Triassic

aquifers

Registered number (Queens4497 land Water Resources O Commission) 73 0 Group Hydrochemistry z Na- HCO3-CI

0 0

Na- HCO3-CI-SO4

•

Na-Ca-CI- HCO3-SO4

m >

z a

Na-Ca-CI- HCO3

Na-HCO3-CI (Ca-S0 4 )

a 7* o o > & CO

O 1— 1 1 Na- HCO3-CI L

O z Na-Ca- Mg-CI- HCO3-SO4 T o t a l Dissolved Solids 0 - 8 0 0 mg/l 800-1 600 m g / l 1600-3000 mg/l 3000-10000 mg/l > 10000 mg/l

o 70

368

M. A. HABERMEHL

contains gases, and several produce noticeable amounts of hydrocarbon (largely methane) and carbon dioxide gases. Hydrocarbon indications were frequently discovered during the drilling of many artesian water wells in the first half of this century (Interstate Conference on Artesian Water, 1913, 1914, 1925; Mott, 1952; Geological Survey of Queensland, 1960), but few led to further, larger, hydrocarbon discoveries. A model of hydrodynamic-influenced migration of hydrocarbons, and their possible accumulation and entrapment in suitable structural zones or near barriers of stratigraphic or diagenetic origin was proposed by Senior & Habermehl (1980). To assess the occurrence and composition of dissolved (gaseous) hydrocarb6ns, some flowing artesian water wells in the central part of the Eromanga Basin were sampled during 1980 and 1982. Gas samples were obtained from about 60 long-established (several decades) flowing artesian water wells, which were 'randomly' drilled with

respect to geological structures, in contrast to most petroleum exploration wells. Most water wells tap the upper, main Lower Cretaceous-Jurassic aquifer (Cadna-owie Formation/Hooray Sandstone). Gas samples were collected in 500 ml glass bottles, which had been filled with groundwater. Water was flushed through the sample container from a hose inserted in the wellhead, and after some time exsolved gas, coming out of the water stream, was allowed to accumulate above the flowing water. Several weeks elapsed between collection and analysis by gas liquid c h r o m a t o g r a p h y . N o dissolved gas concentrations were determined because of the sampling method and sample types obtained. It is appreciated that the ratios of gas components might be (substantially) different in the groundwater under aquifer conditions, as different gases preferentially exsolve in the upward waterflow in the wells. Different sampling times and water volumes were required for different wells depending on the amount of gas produced. Methane concentrations in the gas samples range up to about 510 000 microlitres per litre, and the sum of the hydrocarbons ethane to heptane ranges up to about 14 000 microlitres per litre. Several samples contain relatively high concentrations of ethane to heptane, which suggests that the groundwater acts as a carrier for dissolved hydrocarbons derived from adjoining source rocks or from reservoirs. Additional gas samples were collected in 500 ml glass gas sampling tubes with stopcocks at the top and bottom. Gas was allowed to accumulate above the water, and preserved for hydrocarbon and carbon isotope analyses (Habermehl, 1983; Airey et al, 1983). Water samples were collected in one litre polyethylene bottles for chemical analysis. The composition of the gases analysed by gas liquid chromatography is given in Tables 1 and 2. Oxygen plus argon, carbon dioxide and nitrogen represent the main constituents of the gases sampled in 1980. The qualitative results show that the hydrocarbon components consist mainly of methane and lesser amounts of ethane to heptane (Fig. 15). Methane values are relatively high in some wells which tap Cretaceous aquifers in immature source rocks (Passmore & Boreham, 1982). Several wells completed in Lower Cretaceous-Jurassic aquifers, in a sequence considered by Passmore & Boreham (1982) to contain largely mature source rocks in most of the area, also show higher methane values, though the values for total dissolved hydrocarbons are generally low. Samples obtained by relatively better techniques in 1982 show higher hydrocarbon values for the repeatedly sampled wells (Table 2, Fig. 16). Hydrogen and helium are other major constituents of the gas (Table 2). Some samples derived from water coming from aquifers in the Lower Cretaceous-Jurassic sequence, mainly in the Cadna-owie Formation and Hooray Sandstone, produce relatively high values for methane to heptane. Water from wells on the Eulo Ridge and Cunnamulla Shelf, where the Eromanga Basin sequence is thinner, relatively shallow and subsurface temperatures are lower, contain only small ratios of light hydrocarbons. Methane values of wells in these areas are much lower than from wells in the deeper part of the basin, and several other ratios (e.g. propane/ethane) are likewise different (Tables 1, 2; Figs 15, 16).

19/Q/7 Fig. 13. Dendrogram showing differentiation of chemical data. For key see Fig. 12.

Relatively high ratios of helium and carbon dioxide occur in samples obtained from several wells (RN 3771, 4022, 358) in the Canaway Fault area. Concentration of crustal-produced helium could occur in slow-moving groundwater in deep aquifers, but enrichment through decay of radioactive elements in the aquifers or underlying rocks are other possible sources (Torgersen & Clarke, 1985). A granitic intrusion in the area of the Canaway Fault, postulated by Pinchin & Anfiloff (1982) could be such a source.

GROUNDWATER AND HYDROCARBON MIGRATION

Fig. 14. Chemical composition of water, central Eromanga Basin, plotted in modified Durov diagram. For key see Fig. 12.

19. a s Total dissolved solids (mg/l) Percentage m i l l i g r a m equivalents

369

370

M. A. HABERMEHL

TABLE 1. Gas chromatograph analyses of gases collected from flowing artesian water wells in central Eromanga Basin (Great Artesian Basin), September 1980. Registered Number Waterwell Water Resources Commission of Queensland RN 14050 RN 3822 RN 12312 RN 14486 RN 16768 Innamincka-1 RN 2424 RN 15239 RN 5092 RN 358 357 RN RN 7311 RN 4022 RN 390 RN 148 RN 1728 RN 4782 RN 1475 RN 1473 RN 1474 RN 4910 RN 4270 RN 4165 RN 3033 RN 1298 RN 3489 RN 1551 RN 17223 RN 4497 RN 17263 RN 1057 RN 3945 RN 3950 RN 3947 RN 305 RN 3771 RN 3770 RN 15671 RN 155 RN 2079 RN 2080 RN 1262 RN 5 RN 4902 RN 401 RN 403 RN 2431 RN 5330

O +2

C02

16500 41900 48500

C3H8

CH 4

C4H,O

C 5 H 12

6HI4

80500 9200 7800

29500 7700 60

—

69400 19700 35300

65200 38600 9300

36300 108200 14600

7H16 +

higher

microlitres per litre (gas)

150 20 15

140 30 10

130 40 10

140 50 10

110 130 5

—

230 800 80

20 110 15

10 40 25

<5 30 10

<5 <5 15

<5 <5 20

balance balance balance

400 95 15

balance balance balance

—

104500 29600 20900 21700 28500 32700 28400 22000 15700 24100 34000 22000 143900 42000 20700 27200 25900

25500 38600 3800 8600 9100 3800 9400 15500 27900 25100 15700 26000 9100 21500 21600 12600 12600

150 34600 160 66200 15200 17900 30 170 28950 8600 64300 125000 46100 47500 43700 18400 5100

60 400 20 95 140 120 5 85 150 55 730 1400 380 450 360 240 120

140 45 5 5 35 5 <5 20 35 10 200 550 150 120 70 100 25

440 35 5 <5 20 5 <5 5 15 10 80 270 100 65 40 85 25

750 15 5 <5 5 15 <5 5 15 5 20 45 20 15 15 55 15

510 <5 5 <5 10 15 <5 15 5 5 10 20 10 35 5 15 10

200 <5 5 <5 20 45 <5 30 10 10 15 5 20 55 15 20 10

balance balance balance balance balance balance balance balance balance balance balance balance balance balance balance balance balance

—

28300

12100

—

23600

12000

19700

1600

10 —

5 —

—

balance

—

17000

3700

216500

—

23000 6400 3000 3400 2000

18000 278100 69500 26000 1900 10000 43600 28200

310 90 65 55 40 50 430 400

200 15 5 <5 1 <5 120 300

410 15 5 <5 15 <5 30 500

530 20 10 <5 15 <5 5 400

460 10 40 5 <5 5 5 50

340 20 50 5 5 5 25 80

balance balance balance balance balance

—

5 <5 <5

10 <5 <5

20 5 5

balance balance balance

—

36100 32600

6600 —

—

37500 26800 26800

2000 7500 2500

6900 60 110

20 <5 15

5 <5 <5

Groundwater in contact with source rocks or oil and gas traps in reservoir rocks will accumulate soluble hydrocarbons and organic compounds. Buckley et al. (1958) in their studies of gas compositions of dissolved hydrocarbons in groundwater in the southern USA found regional and local variations in concentrations of dissolved hydrocarbons. They observed local enrichment in dissolved hydrocarbons in close proximity to some, but not all oil fields, when water was sampled beyond the edge of the oil accumulation. Enrichment was apparently restricted to a very short lateral distance, and might suggest a localized origin of the oil or gas. Dickey & Hunt (1972) and Dickey (1979) concluded that the content of methane in water

—

— —

balance

16800 23600 28300 41500 160500

—

120

balance

—

balance balance

cannot be used as an indicator of proximity to a gas field, though the presence of gas in formation water throughout a large area may be used as a regional indicator that gas or oil fields may be present. Price (1976) and McAuliffe (1980) considered the solubility of hydrocarbons in water, and ruled out migration of hydrocarbons in water in significant quantities. McAuliffe (1979) ruled out solution as a principal mechanism for oil or gas migration, and concluded that water appears to disperse soluble oil and gas constituents instead of concentrating them in reservoir traps. Roberts (1980), Schowalter (1979), Berg (1975), Collins (1975), Baker (1967) and Smith (1966) dealt with aspect of hydrocarbon migration

GROUNDWATER A N D H Y D R O C A R B O N MIGRATION

371

TABLE 2. Gas chromatograph analyses of gases collected from flowing artesian water wells in central Eromanga Basin (Great Artesian Basin), May/June 1982. Registered Number Waterwell Water Resources Commission of Queensland

1996 305 305 1184 3771 3771 358 RN 4022

co2

mol. per cent

0.01 0.02 0.04

CH 4

C2H6

vol. (gas)

C3H8

C 4 H 10

C5HI2

6HI4

7H16 +

higher

microlitres per litre (gas)

< 0.01 < 0.01 <0.01

1.75 9.95 1.80

0.13 1.55 1.50

0.05 25.35 42.70

10 1930 2820

25 730 560

40 1050 310

35 840 130

35 470 110

20 100 50

—

0.02 0.11 0.15 0.33

<0.01 0.11 0.28 0.21

11.60 1.40 0.96 1.75

2.95 3.75 14.85 0.92

14.00 43.70 51.10 13.70

1020 2470 4370 1460

310 930 1400 520

210 960 1180 420

90 680 480 200

40 340 210 95

15 90 65 40

balance balance balance balance

0.06

<0.01

5.10

6.25

31.00 51.35

2470 10300

900 1870

600 750

240 350

65 290

Innamincka-1

110

balance balance

RN 15239 RN 401 RN 2431 RN 2271 RN 1338 RN 67 RN 2049 RN 13698 RN 1754

—

0.07 0.15 0.12 0.09 0.09 0.01 <0.01

<0.01 <0.01 <0.01 <0.01 <0.01 <0.01 <0.01

4.75 3.75 2.25 0.50 0.84 2.10 0.03 0.60

210 110 840 30 150 10 30 670

35 25 4310 170 810

25 20 3890 140 850

15 20 1790 100 480

10 310 50 110

—

0.29 0,55 0.28 0.10 0.06 0.20 0.33 0.92

25 20 1940 110 320

—

7.30 2.00 1.80 5.10 1.60 2.05 3.35 3.15

RN RN RN RN RN RN RN DM KIN

balance balance balance

71J 1 1I 11

RN 5092

and entrapment, and l o t h (1980) described some hydraulic aspects of the transport and accumulation of petroleum. Other authors who studied petroleum migration problems include Hitchon (1984), C h a p m a n (1983), Dahlberg (1982), Roberts & Cordell (1980), Neglia (1979), Tissot & Welte (1978), Chiarelli (1978), Cordell (1972), Toth (1970), Hubbert (1953), and many others. Hydrodynamics and hydrocarbon occurrences in the adjoining Surat Basin were discussed by Conybeare (1970) and Hitchon & Hays (1971). It is generally recognized that primary migration of hydrocarbons in the fine-grained source rocks can be influenced by groundwater movement. However, groundwater flow has a far greater influence on the secondary migration of hydrocarbons through more permeable rocks, where movement is mainly governed by buoyancy, capillary pressure and hydrodynamics (Chapman, 1983; Schowalter, 1979; Tissot & Welte, 1978; Berg, 1975). Hydrodynamic conditions in a basin where hydrocarbons are being mobilized might determine the direction of secondary migration, and change the buoyancy movement. The remobilization of trapped hydrocarbons is generally thought to involve small quantities, transported over limited distances. In the E r o m a n g a Basin, hydrocarbons have been encountered within the basin-wide moving groundwater in water wells which penetrate the Lower Cretaceous-Jurassic aquifers, particularly in those wells located in the deeper part of the basin. Most of the porous and permeable quartzose sandstones which form the aquifers, also constitute reservoir rocks for hydrocarbon accumulations in the basin. The hydrocarbons present in the groundwater probably originate largely from diffusion and solution by flowing groundwater from adjoining source beds and existing traps. Water washing of trapped hydrocarbons in the Eromanga B&sin has been shown to occur by Smyth & Saxby (1981), and McKirdy (1982) considers all the Jurassic and Cretaceous oils to be water washed. Water washing effects, which include the removal of

—

120 1320

180 2190

140 1550

100 700

50 170

balance balance balance balance balance balance balance balance

lighter hydrocarbons, are the result of migration, and the longer the secondary migration paths, the greater the effects (Chapman, 1982). Hydrocarbons will not be significantly flushed out of existing traps under present geological and hydrological conditions as shown by Chapman (1983) and Moriarty & Williams (1982). However, any changes to these conditions, including faulting, changes to geological dips, hydrodynamic conditions (e.g. hydraulic gradients), density of hydrocarbons or groundwater and trap seals that are critical to entrapment will lead to re-migration. Dissolved hydrocarbons could become trapped and accumulate in suitable structural, stratigraphic or diagenetic traps. In some areas, faults locally displace or disconnect aquifers, and obstruct part or all of the groundwater flow in the main Early Cretaceous and Jurassic aquifers, which is directed normally at these structures (Fig. 9; after Senior & Habermehl, 1980; Senior, 1982). Little is known about the hydraulic character of these faults, which could act as impermeable barriers or as preferential permeable zones. Where they act as barriers, flow directions will have changed and salinity of semi-stagnant water could increase and hydrocarbons accumulate. Alternatively the faults might act as hydrodynamic traps for hydrocarbons while the water enters adjacent rocks (Chapman, 1983). Migration of hydrocarbons from within either the Eromanga Basin sequence or from the underlying Cooper, Galilee or Adavale Basins, will have been affected by groundwater hydrodynamics. Hydrocarbons could have accumulated and been trapped near such faults in stagnant or semi-stagnant groundwater zones, and in differential permeability barriers of structural, stratigraphic or diagenetic origin (Senior & Habermehl, 1980). Some hydrocarbons might have been re-transported by the groundwater. Data on the concentration of hydrocarbons in the groundwater are lacking, but the contribution to the formation of accumulations in traps by solution transport of hydrocarbons derived from existing traps is probably small.

372 combined mole percent

Regional groundwater

Flowing artesian waterwell Cretaceous

flow

tapping:

aquifers

Lower aquifers

Cretaceous-Jurassic

Triassic

aquifers

Registered number (Queensland Water Resources Commission)

SOUTH QUEENSLAND AUSTRALIA

19/0/9

M. A. HABERMEHL

Fig. 15. Composition of light hydrocarbons (methane to heptane) in gases collected from flowing artesian water wells, central Eromanga Basin, 1980.

CH^ mole percent

o o 3

mole percent

combined

mole percent O 70 0

Regional

groundwater

flow

c z

1 m D

Flowing

artesian

Cretaceous

•

Lower aquifers Triassic

Registered (Queensland Resources

2271

waterwell

to tapping: . >

aquifers Cretaceous-Jurassic

aquifers

a x

U TO O o >

number Water Commission)

TO CO O

z o

WfSTiBN

SnN \ A

L, WAtES f i J ^ /

70 $ O

374

M. A. HABERMEHL

Hydrocarbon generation took place in Jurassic and some of the more deeply buried Cretaceous source rocks since the Late Cretaceous and Early Tertiary. Migration of hydrocarbons from fine-grained source rocks into reservoir rocks and migration in reservoir rocks has occurred since those times, and has been affected by water movement caused by compaction and by artesian groundwater flow. Confined groundwater conditions were present in most aquifers in the basin since Jurassic and Early Cretaceous times, but hydrodynamic conditions approached present-day flow patterns only following uplift of the eastern marginal area during the Late Cretaceous to Early Tertiary. Potential traps for hydrocarbons were largely formed by folding and faulting during the Tertiary (Senior & Habermehl, 1980) and migration into suitable structural traps will have taken place since then, whereas stratigraphic traps were already in existence. Groundwater flow patterns during the Tertiary and Quaternary, which had an influence on migrating hydrocarbons, were broadly similar to the present groundwater flow directions, though hydraulic gradients were initially smaller. Hydrogeological, hydrochemical and isotope hydrology data have delineated the present groundwater flow patterns and areas of possible groundwater stagnation, i.e. possible hydrocarbon traps have been identified. Hydrogeological and hydrochemical results indicate that not all possible traps are low-permeability or sealed zones. The origin of hydrocarbons in the artesian groundwater, their migration routes and relation to possible existing accumulations require further study.

CONCLUSIONS

Gases from the flowing artesian water wells in the Eromanga Basin are generally dry (C,/C,_ > 0.98), with few exceptions. The light hydrocarbons methane to heptane are relatively soluble in water, and as the most mobile components of petroleum, are generally regarded as useful petroleum exploration indicators (Tissot & Welte, 1978). The volume of hydrocarbons dissolved in Eromanga Basin groundwater is not known, and a quantitative analysis would require sampling at aquifer depths and pressures. Further analyses are also required to interpret their origin. A more detailed palaeohydrological reconstruction of the basin, in conjunction with a history of its structural development and better knowledge of the generation, origin and distribution of hydrocarbons, could improve understanding of the migration and entrapment of hydrocarbons in the Eromanga Basin. It is clear that hydrocarbons are related in significant ways to the hydrological regime, but the present model is rather simple. A deeper understanding of the hydrodynamics and hydrochemistry of the basin will enhance prospects for hydrocarbon exploration. 7

ACKNOWLEDGEMENTS

I thank Drs C. J. Boreham, P. J. Cook, C. D. Oilier, J. D. Waterhouse, A. F. Williams, and T. Torgensen for critically reading the manuscript. This paper is published with the permission of the Director, Bureau of Mineral Resources, Geology and Geophysics, Canberra.

REFERENCES AIREY, P. L., CALF, G. E., CAMPBELL, B. L., HABERMEHL, M . A . , HARTLEY, P. E. & ROMAN, D., 1979: Aspects of the isotope

hydrology of the Great Artesian Basin, Australia; in Isotope Hydrology 1978, 1, 2 0 5 - 1 9 . Proceedings International Symposium on Isotope Hydrology, Neuherberg, Fed. Rep. Germany,; Int. Atomic Energy Agency, Vienna.

AIREY, P. L., BENTLEY, H . W., CALF, G. E., DAVIS, S. N . , ELMORE, D., GOVE, H . , HABERMEHL, M . A . , PHILLIPS, F. M . , SMITH, J., & TORGERSEN, T., 1983: Isotope hydrology of the Great

Artesian Basin, Australia. Aust. Water Resour. Coune. Conf.

Ser. 8 (I), 1-11. BAKER, E. G., 1967: A geochemical evaluation of petroleum migration

and accumulation; in Nagy, B. & Colombo, U. (eds) Fundamental aspects of petroleum geochemistry. Elsevier, Amsterdam, 299-329. BERG, R. R., 1975: Capillary pressures in stratigraphic traps. Am. Assoc. Pet. Geol Bull. 59, 9 3 9 - 5 6 . BOWERING, O. J. W., 1982: Hydrodynamics and hydrocarbon migration—a model for the Eromanga Basin. APEA J. 21 (1), 227-36. BUCKLEY, S. E., HOCOTT, C . R. & TAGGART, M . S.,

1958:

Distribution of dissolved hydrocarbons in subsurface waters; in Weeks, L. G. (ed.) Habitat of oil. Am. Assoc. Pet. Geol., Tulsa, 850-82.

BURGER, D. & SENIOR, B. R., 1979: A revision of the sedimentary

and palynological history of the northeastern Eromanga Basin, Queensland. Geol. Soc. Aust. J. 26, 1 2 1 - 3 2 .

CALF, G. E. & HABERMEHL, M. A., 1984: Isotope hydrology and

hydrochemistry of the Great Artesian Basin, Australia; in Isotope Hydrology 1983, 3 9 7 - 4 1 3 . Int. Atomic Energy Agency, Vienna.

C H A P M A N , R. E., 1982: Effects of oil and gas accumulation on water

movement. Am. /4550c\ Pet. Geol. Bull. 66, 368-78. geology. Elsevier, Amsterdam.

C H A P M A N , R. E., 1983: Petroleum

CHIARELLI, A., 1978: Hydrodynamic framework of eastern Algerian

Sahara—influence on hydrocarbon occurrence. Am. Assoc. Pet. Geol. Bull. 64, 667-85. of oilfield waters. Elsevier, Amsterdam. CONYBEARE, C . E. B., 1970: Solubility and mobility of petroleum under hydrodynamic conditions, Surat Basin, Queensland. Geol. Soc. Aust. J. 16, 667-81. COOK, A. C., 1982: Organic facies in the Eromanga Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 234-57. CORDELL, R. J. 1972: Depths of oil origin and primary migration: a review and critique. Am. Assoc. Pet. Geol. Bull. 56, 2029-67. CULL, J. P. & CONLEY, D., 1983: Geothermal gradients and heat flow in Australian sedimentary basins. BMR J. Aust. Geol. Geophys. 8, 329-37. DAHLBERG, E. C., 1982: Applied hydrodynamics in petroleum exploration. Springer-Verlag, New York. DICKEY, P. A., 1979: Petroleum development geology. Petroleum Publ. Co., Tulsa. DICKEY, P A. & H U N T , J. M., 1972: Geochemical and hydrogeologic methods of prospecting for stratigraphic traps; in King, R. E. (ed.) Stratigraphic oil and gas fields—classification, exploration methods, and case histories. Am. Assoc. Pet. Geol. Mem. 16, 136-67. EXON, N. F. & SENIOR, B. R., 1976: The Cretaceous of the Eromanga and Surat Basins, BMR J. Aust. Geol. Geophys. 1, 33-50. GEOLOGICAL SURVEY OF QUEENSLAND, 1960: Occurrence of petroleum and natural gas in Queensland. Geol. Surv. Qld Publ. 229. HABERMEHL, M. A., 1980: The Great Artesian Basin, Australia. BMR J. Aust. Geol. Geophys. 5, 9-38.

COLLINS, A. G., 1975: Geochemistry

375

GROUNDWATER AND HYDROCARBON MIGRATION HABERMEHL, M. A., 1982cr: Investigations of the geology and hydrology of the Great Artesian Basin 1878-1980. Aust. Bur. Miner. Resour. Geol. Geophys. Rep. 234. HABERMEHL, M. A., 1982b: Springs in the Great Artesian Basin, Australia—their origin and nature. Aust., Bur. Miner. Resour., Geol. Geophys., Rep., 235. HABERMEHL, M. A., 1982c: The Eromanga Basin within the Great Artesian Basin: in Moore, R S. & Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 384-97. HABERMEHL, M. A., 1983: Hydrogeology and hydrochemistry of the Great Artesian Basin, Australia. Aust. Water Resour. Counc. Conf. Ser. 8 (3), 83-98. HABERMEHL, M. A., (in press): Hydrochemistry and regional groundwater flow in the southwestern part of the Great Artesian Basin, Australia. HABERMEHL, M. A. & MORISSEY, J. A., 1980: Great Artesian Basin wire-line logged waterwells, 1:2 500 000. Aust. Bur. Miner. Resour. Geol. Geophys. HABERMEHL, M. A. & MORISSEY, J. A., 1983: Index of geophysical

well logs acquired by BMR from waterwells in the Great Artesian Basin, 1960-1975. Aust. Bur. Miner. Resour. Geol. Geophys. Rep. 241. HABERMEHL, M. A. & SEIDEL, G. E., 1979: Groundwater resources

of the Great Artesian Basin; in Hallsworth, E. G. & Woodcock, J. T. (eds) Proceedings of the symposium Land and Water Resources of Australia. Australian Academy of Technological Sciences, Sydney 30 October-1 November 1978. Aust. Acad, of Technol. Sci., Melbourne, 71-93. HITCHON, B., 1984: Geothermal gradients, hydrodynamics, and hydrocarbon occurrences, Alberta, Canada. Am. Assoc. Petrol. Geol. Bull. 68, 713-43. HITCHON, B. & HAYS, J., 1971: Hydrodynamics and hydrocarbon

occurrences Surat Basin, Queensland, Australia. Water Resour. Res. 7, 658-76. HUBBERT, M. K., 1953: Entrapment of petroleum under hydrodynamic conditions. Am. Assoc. Pet. Geol. Bull. 37, 1954-2026.

MORIARTY, K. C. & WILLIAMS, A. F., 1982: H y d r o c a r b o n flushing

in the Eromanga Basin—fact or fallacy? in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 313-28. M O T T , W . D., 1952: O i l in Q u e e n s l a n d . Qld Gov. Min.

J. 53,

848-61.

NEGLIA, A., 1979: Migration of fluids in sedimentary basins. Am. Assoc. Pet. Geol. Bull. 63, 573-97; Discussion, 1543-7. NICHOLAS,

E.,

RIXON,

HAUPT,

A.,

1980:

Uncorrected

geothermal map of Australia. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1980/66 (unpubl.). PARKIN, L. W. (ed.), 1965: Handbook of South Australian geology. S. Aust. Geol. Surv., Adelaide. PASSMORE, V. L. & BOREHAM, C. J., 1982: Subsidence and associated

source rock maturation for the central Eromanga Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 296-306. PINCHIN, J. & ANFILOFF, V., 1982: The Canaway Fault and its effects on the Eromanga Basin, in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 161-9. PITT, G. M., 1982: Geothermal gradients in the Eromanga—Cooper Basin region; in Moore, P. S. & Mount, T. J. (compilers), Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 262-83. POLAK, E. J. & HORSFALL, C. L., 1979: Geothermal gradients in the

Great Artesian Basin, Australia. Aust. Soc. Explor. Geophys. Bull. 10, 144-8. POLAK, E. J., & RAMSAY, D. C., 1977: Canaway Ridge, Queensland, geophysical survey, 1973. Aust. Bur. Miner. Resour. Geol. Geophys. Rec. 1977/29 (unpubl.) PRICE, L. C., 1976: Aqueous solubility of petroleum as applied to its origin and primary migration. Am. Assoc. Petrol. Geol. Bull. 60, 213-44. ROBERTS, W. H., 1980: Design and function of oil and gas traps; in Roberts, W. H. & Cordell, R. J. (eds) Problems of petroleum migration. Am. Assoc. Petrol. Geol. Stud. Geol. 10, 217-40.

the

ROBERTS, W. H . & CORDELL, R. J., 1980: P r o b l e m s of petroleum

interstate conference on artesian water, Sydney, 1912. Govt Printer, Sydney.

migration. Am. Assoc. Petrol. Geol. Stud. Geol. 10. SCHOWALTER, T. T., 1979: Mechanics of secondary hydrocarbon migration and entrapment. Am. Assoc. Petrol. Geol. Bull. 63, 723-60. SEIDEL, G. E., 1980: Application of the GABHYD groundwater model of the Great Artesian Basin, Australia. BMR J. Aust. Geol. Geophys. 5, 39-45. SENIOR, B. R., 1982: LANDSAT interpreted structure and groundwater flow within the Eromanga Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 329-42.

INTERSTATE CONFERENCE O N ARTESIAN WATER, 1913: Report

INTERSTATE CONFERENCE O N ARTESIAN WATER, 1914: Report

the

second interstate conference on artesian water, Brisbane, 1914. Govt Printer, Brisbane. INTERSTATE CONFERENCE O N ARTESIAN WATER, 1925: Report

the

fourth interstate conference on artesian water, Perth, 1924. Govt Printer, Sydney. KANTSLER, A . J., C O O K , A . C . , & ZWIGULIS, M . , 1982: M a t u r a t i o n

patterns in the Eromanga Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 284-95. LLOYD, J. W., 1965: The hydrochemistry of the aquifers of northeastern Jordan. J. Hydrol. 3, 319-30. MATTHESS, G., 1982: The properties of groundwater. John Wiley, New York. MCAULIFFE, C. D., 1979: Oil and gas migration—chemical and

physical constraints. Am. Assoc. Pet. Geol. Bull. 63, 761-81. MCAULIFFE, C. D., 1980: Oil and gas migration: chemical and physical constraints; in Roberts, W. H. & Cordell, R. J. (eds) Problems of petroleum migration. Am. Assoc. Pet. Geol. Stud. Geol.

10, 89-107.

MCKIRDY, D. M., 1982: Aspects of the source rock and petroleum geochemistry of the Eromanga Basin; in Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide, 258-9. MOORE, P. S. & MOUNT, T. J. (compilers), 1982: Eromanga Basin Symposium, summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide.

SENIOR,

R.,

HABERMEHL,

A.,

1980:

Structure,

hydrodynamics and hydrocarbon potential, central Eromanga Basin, Queensland, Australia. BMR J. Aust. Geol. Geophys. 5, 47-55. SENIOR, B. R . , M O N D , A . & H A R R I S O N , P. L . , 1978: G e o l o g y o f t h e

Eromanga Basin. Aust. Bur. Miner. Resour. Geol. Geophys. Bull. 167. SMITH, D. A., 1966: Theoretical considerations of sealing and nonsealing faults. Am. Assoc. Pet. Geol. Bull. 50, 363-74. SMYTH, M. & SAXBY, J. D., 1981: Organic petrology and geochemistry of source rocks in the Pedirka-Simpson Basins, central Australia. APEA J. 21(1), 187-99. TISSOT, B. P. & WELTE, D. H., 1978: Petroleum

formation

and

occurrence. Springer-Verlag, Berlin. TORGERSEN, T. & CLARKE, W. B., 1985: Helium a c c u m u l a t i o n in

groundwater, I: an evaluation of sources and the continental flux of crustal 4 He in the Great Artesian Basin, Australia. Geochim. Cosmochim. Acta, 49, 1211-8.

376

M. A. HABERMEHL

TOTH, J., 1970: Relations between electric analogue patterns of groundwater flow and accumulation of hydrocarbons. Can. J. Earth. Sci. 7, 988-1007. TOTH, J., 1980: Cross-formational gravity-flow of groundwater: a mechanism of the transport and accumulation of petroleum (The generalized hydraulic theory of petroleum migration); in Roberts, W. H. & Cordell, R. J. (eds) Problems of petroleum migration. Am. Assoc. Pet. Geo I. Stud. Geo I. 10, 121-67. VINE, R. R., 1976: Eromanga Basin; in Knight C. L. (ed.) Economic geology of Australia and Papua New Guinea, 3. Petroleum. Australas. Inst. Min. Metall. Mongr. 7, 306-9.

The potential for groundwater recharge through red, yellow, and grey earth profiles in central north Queensland. Aust. Water Resour. Cone. Conf.

WILLIAMS, J. & COVENTRY, R . J., 1981:

Ser. 3, 169-81. WOPFNER, H . , FREYTAG, I. B. & HEATH, G. R., 1970: Basal Jurassic-

Cretaceous rocks of western Great Artesian Basin, South Australia: stratigraphy and environment. Am. Assoc. Pet. Geol. Bull. 54, 3 8 3 - 4 1 6 .

Geological Society of Australia Special Publication No. 12, 377-384

Hydrocarbon flushing in the Eromanga Basin—fact or fallacy? Tony Williams & Kevin Moriarty 1

Santos Ltd, 39 Grenfell Street, Adelaide, S.A. 5000; present address Western Mining Corporation Ltd, P.O. Box A8, Perth, W.A. 6001; present address P.O. Box 458, Norwood, S.A. 5067. 1

ABSTRACT

The hydrodynamic regime in the Eromanga Basin theoretically imposes a tilt of less than 0.1° to any oil water contact. Mapped closures on oil fields such as Jackson, Merrimelia, Dullingari and Strzelecki are at least 1 so the current regime cannot flush oil from these or any other structures with similar closures. Review of the basin's history indicates that palaeogroundwater gradients would have been substantially lower than those observed today. During and prior to the accepted oil migration phase (mid Cretaceous to early Tertiary), structures with dips of no more than 0.3° have been effective traps. As eastern intake areas were uplifted, thus increasing gradients, concomitant tectonism steepened structural closure, maintaining ability to trap oil. Thus, providing sufficient structural closure was present prior to migration, flushing of traps by past and present flow regimes was and is impossible. This, together with the fact that large numbers of stratigraphic traps exist within the reservoirs, indicates that long distance migration of petroleum within the Eromanga Basin is highly improbable. The growing number of discoveries supports earlier estimates of high Eromanga Basin source potential. Most of the hydrocarbons generated are still trapped within the basin and the problem is now to find traps which existed during the migration phase or earlier.

INTRODUCTION

The Eromanga Basin is a sedimentary basin occupying some 1 x 10 km of central and western Queensland and adjoining areas of the Northern Territory, South Australia and New South Wales (Fig. 1). It comprises a thick sequence of up to 3,000 m of Mesozoic sandstone, siltstone and shale which blankets a variety of older sediments ranging in age from Precambrian to Triassic. The most important of these from the petroleum point of view are the Permo-Triassic sediments of the Cooper Basin, within which significant reserves of gas and liquid hydrocarbons were discovered during the 1960s and early 1970s. The fact that encouraging hydrocarbon shows were not found in overlying Eromanga Basin sediments during the initial Permian exploration phase led to the hypothesis that the basin had been flushed by artesian waters. The reason that only insignificant shows were located initially was that wells were often off-structure at Jurassic reservoir levels, and drill stem tests naturally resulted in flows of water e.g. Merrimelia 1, Strzelecki 1 (Fig. 2). Thus, for a time, many exploration wells were not fully monitored through the Cretaceous-Jurassic section. However, in 1976, gas was located in Namur 1 within Late Jurassic sediments, and attention was immediately re-focused on this part of the stratigraphic section. Subsequently hydrocarbons have been discovered in various Jurassic reservoirs over a wide area (e.g. Dullingari, Strzelecki, Marabooka, Merrimelia, Moorari, Jackson; Fig. 2), indicating the previously popular flushing concept to be somewhat suspect. The purpose of this paper is to illustrate that 6

consideration of basic hydrodynamic principles precludes flushing of hydrocarbons within the Eromanga Basin reservoirs, under both past and present groundwater flow regimes.

GEOLOGY

The geology of the Eromanga Basin has been adequately described elsewhere and will not be discussed here. Figure 3 shows broad stratigraphic correlations throughout the basin and the more significant hydrocarbon occurrences in context. Further aspects are dealt with in following sections.

HYDROGEOLOGY

The Eromanga Basin comprises the major part of a hydrogeologic entity known as the Great Artesian Basin (Fig. 1). This latter basin covers an area of about 1.7 x 10 km or about one fifth of Australia, and incorporates other sedimentary basins of similar age such as the Surat and Carpentaria Basins. Interflow with deeper, older basins such as the Cooper and Pedirka Basins is also inferred. Since the central and western parts of the Great Artesian Basin are essentially coincident with the Eromanga Basin it is relevant that we study what is known about the hydrogeology of the former. The Great Artesian Basin has been well-described by Habermehl (1980), who summarized the results of a seven year study which included development of a large-scale computer model of the basin's hydrodynamic flow regime. In brief, the basin is a multi-layered confined aquifer (reservoir) system, 6

T. WILLIAMS & K. MORIARTY

Fig. 1. Location of Eromanga and associated sedimentary basins. Inset shows hydrogeological Great Artesian Basin.

"A.

/ • POOLOWANNA 1

EROMANGA

( \

I COOPER BASIN

WACKETT

MOORARI 3

BASIN

^lERRIMELIA

• JACKSON 1

I NAMUR1

MOOMBA«V

i * D U L L I N G A R I NTH. l /

^-BK5XLAKE26*^STRZ:ELECK.3 ^ MARABOOK A 1 | l

COWARD.oSPRINGS 136*

138°

l\ V

140*

142*

Fig. 2. Significant hydrocarbon discoveries in the Eromanga Basin (refer Fig. 1 for location), together with location of Coward Springs.

HYDROCARBON FLUSHING

SOUTHWEST (S.A.)

WEST, NORTH (Q,S.A,NT.)

CENTRAL, EAST, SOUTH

DELHI SANTOS (S.A,Q.)

(Q,N.S.W.)

MINOR GAS SHOWS

EARLY CRETACEOUS CADNA-OWIE FM.

ALGEBUCKINA

SST.

CADNA-OWIE FM.

HOORAY SST. WESTBOURNE FM. ADORI SST.

CADNA-OWIE FM.

TRANSITION BEDS MURTA MBR.

HOORAY

BIRKHEAD FM. HUTTON SST.

HUTTON SST.

SST. WESTBOURNE FM. ADORI SST. BIRKHEAD FM.

[KhJ

BASAL ?,wJURASSIC

MINOR SHOWS DULLINGARI, JACKSON

NAMUR MBR. WESTBOURNE FM. ADORI SST. BIRKHEAD FM.

HYDROCARBON OCCURRENCES

HUTTON SST.

NAMUR, MARABOOKA STRZELECKI, JACKSON, MERRIMELIA, DULLINGARI JACKSON MINOR OIL SHOWS MOORARI, BIG LAKE, WACKETT MERRIMELIA, STRZELECKI STRZELECKI, MERRIMELIA, JACKSON

BOX VALE SST. EVERGREEN FM. PRECIPICE SST.

POOLOWANNA

VARIOUS TRIASSIC UNITS E.G. UNNAMED UPPER TRIASSIC, MOOLAYEMBER FORMATION, NAPPAMERRI FORMATION ETC. Fig. 3. Summary of Eromanga Basin stratigraphic nomenclature (in part after Habermehl, 1980). Stratigraphic context of significant discoveries is also indicated.

with aquifers occurring in terrigenous sandstones of predominantly Jurassic age and with intervening confining beds of siltstone and shale. A thick claystone and siltstone sequence of Cretaceous age f o r m s the m a j o r confining unit. A second, less important aquifer occurs within the upper part of this unit but is not generally recognized in the western portions of the basin. T h e basin f o r m s a large asymmetric synclinal structure uplifted along its eastern margin and tilted southwest. M o d e r n recharge occurs predominantly in the eastern marginal zone where rainfall is at a maximum (Fig. 4). Large-scale groundwater movement in the area of interest is generally towards the southwestern and southern margins (Fig. 4). Natural discharge occurs from m o u n d springs which are associated with faulting in these areas. T h e springs are so n a m e d because of their characteristic m o u n d shape— generally hemispherical, the m o u n d comprising a mixture of physically transported sediment (Cretaceous claystone and siltstone) and chemical precipitate (mainly limestone)—both originating via the spring discharge. Local aeolian derived sand and dust is also incorporated within the m o u n d . As well as discharge via springs, an important c o m p o n e n t to outflow is upward leakage t h r o u g h low permeability confining beds especially in the shallow areas on the southwestern margin of the Great Artesian ( E r o m a n g a ) Basin. Table 1 briefly summarizes the hydrogeology. T h e lower multilayered aquifer system is the more i m p o r t a n t though it should be emphasized that m i n o r discontinuous aquifers d o exist within the upper confining sequence (e.g. the C o o r i k i a n a S a n d s t o n e and its equivalents). T h e main c o n t i n u o u s aquifers within the multilayered system are the H o o r a y S a n d s t o n e (or N a m u r Sandstone M e m b e r of the M o o g a F o r m a t i o n ) a n d H u t t o n Sandstone. More restricted

aquifers occur within the remaining Jurassic and Lower Cretaceous sequence, for example Cadna-owie F o r m a t i o n , M u r t a Member of the M o o g a F o r m a t i o n , Westbourne Formation, Adori Sandstone, Birkhead F o r m a t i o n and preH u t t o n Sandstone units (see Fig. 3, Table 1). Gradients along the dominant southwesterly flow direction vary f r o m 1:2000 to 1:6000; rather low compared with most other areas of the world, for example, the United States Great Plains Region with 1:250 to 1:500, and i n t e r m o n t a n e basins 1:50 (Hubbert, 1953). In addition, Chiarelli (1978) gives Algerian examples with gradients generally less t h a n 1:1000. Obviously local perturbations in both flow and gradient occur wherever permeability barriers such as faults and sand pinchouts occur. However, it is expected that these would only have minor effects on regional gradients. Flow velocities in the Great Artesian Basin are of the order of 1 to 5 m per year (Habermehl, 1980), thus turnover time along the most direct flow paths is between 200 000 and 1 x 106 years (i.e. for 1 000 km path). Movement through the deeper sections of the basin (e.g. immediately above the Pedirka and C o o p e r Basins) is slower and along more circuitous paths resulting in perhaps double the turnover time or even longer. G r o u n d w a t e r chemistry over most of the Great Artesian Basin is characterized by the dominance of three ions, namely sodium, chloride and bicarbonate ( H a b e r m e h l , 1980; South Australian D e p a r t m e n t of Mines and Energy water bore records; S A N T O S unpubl. data). Water containing a predominance of these ions is known as a bicarbonate-sodium type under the Sulin classification (Collins, 1975). It is typical of meteoric groundwaters and occurs in many oilfields. In contrast, in the extreme western portions of the basin,

T. W I L L I A M S & K .

380

MORIARTY

DIRECTION OF GROUNDWATER FLOW RECHARGE

Eiiii

AREA

•

SPRING CONCENTRATION SPRINGS

BRISBANE

LAKE GAIRDNER\

Fig. 4. Recharge and natural discharge (spring) areas and directions of groundwater flow in the Great Artesian Basin (after Habermehl, 1980, fig. 12). sulphate, w h i c h rarely attains a m o u n t s o f 1 milli-equivalent

a n a b u n d a n c e o f sulphate ( S A N T O S , u n p u b l . d a t a ) . I n the

per litre elsewhere, becomes a m o r e d o m i n a n t i o n a n d m a y

b i c a r b o n a t e - s o d i u m type, it is f o u n d t h a t b o t h c h l o r i d e a n d

in fact exceed the a m o u n t o f b i c a r b o n a t e ( H a b e r m e h l , 1980,

s o d i u m increase w i t h d e p t h whilst b i c a r b o n a t e stays at a

fig. 17). This appears to reflect the a b u n d a n c e o f g y p s u m in

similar level or m a y decline slightly (in a n a b s o l u t e sense).

the soils a n d younger sediments w h i c h cover the reservoir

Similarities in c o m p o s i t i o n exist t h r o u g h o u t the Jurassic,

rocks. It s h o u l d be noted that for a similar reason shallow

Triassic

groundwaters towards the basin centre are characterised by

suggesting the existence o f o n e regional h y d r a u l i c system.

and

Permian

(Youngs,

1975;

Williams,

1981)

TABLE 1. Summary of gross lithology and hydrogeology of Eromanga Basin sequence. Formation (Delhi, Santos useage)

Lithology (major)

UNDIFFERENTIATED Claystone/SiItstone (Winton, etc./ Coorikiana Sst.) with minor Sandstone.

Cadna-owie Formation

Hydrogeology

reservoir- potential

Confining bed, minor aquifers recognised in Queensland but not South Australia. Coorikiana Sandstone has reservoir potential in western and southwestern portions of the basin. Confining bed containing reservoirs.

few

thin

low

permeability

Mooga Formation (Murta M.)

Siltstone, shale, thin Sandstone

Confining bed with few thin, low to high permeability reservoirs

Mooga Formation (Namur M.)

Sandstone, thin Siltstone

Excellent reservoir with high porosity and good permeability. Upper section may be very fine tight.

Westbourne Formation

Siltstone, shale and Sandstone

Mainly confining bed, some poor to moderate reservoir quality in part.

and

Adori Sandstone

Sandstone, Siltstone

Moderate reservoir though few hydrocarbon shows.

Birkhead Formation

Siltstone, shale and Sandstone

As for Westbourne Formation

Hutton Sandstone

Sandstone with minor Siltstone and shales

Probably the best reservoir in the basin - having both high porosity and permeability. Again, top may be marked by poorer quality fine grained tight sandstone which may affect lateral continuity of reservoir.

Basal Jurassic

Sandstone, Siltstone and shale

Some sand exhibit good reservoir potential. a good seal.

Many lack

HYDROCARBON FLUSHING

fix

POTEN1 lOMt TRJC SURFACE

[dh

PIEZOMETERS v \

"xo^sr oil_

% \ *

* /

Zo tan e

? w dh Xo pw-eo dx ' Zo ? w dh Xo " ? w-fo dx

Fig. 5. Hydrodynamic effects on petroleum occurrences, illustrating Hubbert's (1953) equation.

BASIN HISTORY Before discussing palaeohydrogeology it is relevant to review basin history, since tectonics play such an important role in the development of palaeo-hydrodynamic regimes which are critical to the migration and entrapment of hydrocarbons within the basin. The tectonic framework and history of the Eromanga Basin has been outlined in a broader study of Australia's Phanerozoic Basins by Veevers et al. (1982). Basinal history in Queensland is discussed by Senior et al. (1978) and in South Australia by Wopfner et al. (1970), Wopfner et al. (1974) and Wopfner (1974). Numerous other authors have contributed to the subject (e.g. Sprigg, 1961; Forbes, 1966). Lastly, reference is drawn to papers on timing of the uplifts in the Eastern Highlands area by Brown et al. (1958), Wellman & McDougall (1974), Wellman (1979), Oilier (1982) and Moore & Pitt (1984). The following outline is drawn from relevant portions of the above contributions to the history of the Eromanga Basin. Jurassic fluvial sediments were laid down on a pre-existing peneplaned surface comprising a variety of lithologies ranging in age from Precambrian to Late Triassic. Extensive PermoTriassic sediments accumulated in basins (e.g. Cooper, Pedirka, Galilee) between ridges of upfaulted older rocks (economic basement). Early sedimentation was confined to areas above these Permo-Triassic depocentres and differential compaction over local pre-existing highs may have created some drape relief; minor tectonism along such highs could also have led to similar relief. Whatever the origin, such relief can be measured on seismic sections by reference to the strong event known as the 'C' horizon (near base Cretaceous—top Cadna-owie Formation) and deeper reflectors. By the end of the Jurassic, both stratigraphic and structural traps (albeit of small vertical extent) existed, favourable for the accumulation of hydrocarbons and disruption of long distance migration to basin edges. Fluvial depositional cycles ended by the beginning of the Cretaceous when a shallow sea began to transgress the area. This eventually resulted in deposition of up to 1800 m of Aptian-Albian marine claystone and siltstone. A late AlbianCenomanian regression resulted in the re-establishment of a fluvial, lacustrine and paludal environment similar to that of the Jurassic, though provenance differed with finer grained, labile constituents dominating lithologies. Again deposition

381

was concentrated above older basins such as the Cooper and Pedirka Basins. Major sedimentation essentially ceased early in the Late Cretaceous (Cenomanian), and was followed by a period of gentle uplift and erosion characterized by deep chemical weathering and limited deposition. In Late Cretaceous to Early Tertiary times a further phase of uplift rejuvenated structures within the basin and around the margins, resulting in accentuated folding of older structures and influx of quartz-dominated sediment. This change in provenance was probably brought about by uplift and erosion of Jurassic sediments around the basin margin. Estimates of uplift for the southeastern highlands are given as 300 m by Wellman & McDougall (1974) and, although timing and mechanism were probably different to the north (Wellman, pers. comm., 1982), one would expect a similar order of uplift in the eastern and northeastern highlands. Obviously, the amount of uplift decreased basinward. A period of quiescence followed until about mid-Miocene time when further uplift of the order of 200 m took place along the eastern highlands (Wellman, 1974). Additional movement around the eastern margin occurred during Pleistocene time, probably related to the Kosciuskan Uplift. Sedimentation during the latter part of the Cainozoic again took place in a fluvio-lacustrine environment which has persisted to the present, although aeolian deposition has lately predominated in the western portion of the Eromanga Basin.

PALAEOHYDROGEOLOGY The earliest significant groundwater flow regime in the Eromanga Basin was established in the Early Jurassic when fluvial sedimentation began to extend southwest from the area overlying the Galilee Basin to that overlying the Cooper Basin, (Senior et al., 1978). Since surface water flow was in a general southwesterly direction, so was the groundwater flow. By the end of the Jurassic, when deposition had extended to at least today's basin margin, there was a regional flow regime with a major southwesterly component of flow and minor contributions from the north and west. This is similar to today's flow regime except that the asymmetric axis of flow would probably have been displaced eastward and the degree of groundwater confinement would have been much less. Some local unconfined groundwater flow to streams and lakes would have existed especially in areas with limited confining beds in the section (e.g. the southwest). Relief around the basin margins is considered to have been of a similar order of magnitude to that observed today, such that similar gradients were in operation (note that Oilier (1982) considers relief may have been much less). Outlet to groundwater flow was probably via the area now occupied by the Murray Basin (inset, Fig. 1). Inundation by the shallow Early Cretaceous sea and deposition of the thick marine sequence of shale and siltstone obviously affected the above flow regime. Whether the reservoir system was partially or completely inundated by sea water is open to speculation, and depends on the extent and continuity of the transgression. At this stage there does not appear to be any evidence of sea-water-type brines within the basin, though some moderate salinities have been reported (Senior & Habermehl, 1980). Whatever the case, by the time of deposition of Winton Formation sediments, the hydrogeological regime was again similar to that described for the Jurassic, the finer grained sediments indicating that gradients were probably lower. By this time the whole Jurassic reservoir system was confined.

382

T. WILLIAMS & K. MORIARTY

Uplift along the eastern margin of the Eromanga Basin of HYDRODYNAMIC EFFECTS ON PETROLEUM up to 300 m and possibly more during the Late Cretaceous OCCURRENCES IN THE EROMANGA BASIN and Early Tertiary rejuvenated groundwater gradients and maximum height of a hydrocarbon column which may tilted the basin's axis to the southwest. Flow through the be The trapped within a reservoir system is controlled mainly by reservoir system would have been much slower because of capillary pressure with the prevailing hydrodynamic gradient longer (deeper) flow paths, reduced permeability through acting to increase decrease the height. This can be expressed diagenetic changes, and faulting associated with the eastern by the following orequation adapted from Stone & Hoeger uplift. Nevertheless it is likely that complete flushing of (1973) and Schowalter (1979): residual Jurassic-Early Cretaceous groundwaters took place d _ d during this period except in zones of stagnation (e.g. fault Zmax + traps, sand pinchouts and diagenetic traps (Senior & 0.433(pw - pp) 0.433(pw - o) Habermehl, 1980). This flow regime changed little until Plio-Pleistocene times, where Zmax = maximum height of hydrocarbon column apart from slight modification to gradients by erosion, and P<% = displacement pressure of barrier rock variation in rainfall over the recharge area (note that increased P^R = displacement pressure of reservoir rock rainfall may add only metres to a few tens of metres of inlet 0.433 = units conversion factor head and increase the gradient by approximately 10%). pw = water density at subsurface conditions (see Table 2) Further uplift, which was associated with the Kosciuskan po Uplift, took place along the eastern margin of the basin and 2) = hydrocarbon density at subsurface conditions (see Table elevated the recharge areas to levels similar to, or slightly hydrodynamic pressure drop across the oil higher than those observed today, about 500-700 m above A^HD sea level. Groundwater gradients would have been at a accumulation. The hydrodynamic effect may be written as in Berg (1975): maximum at this time. Gradients within the basin would have declined by a few per cent since Pleistocene time because erosion of the intake area, coupled with decline in rainfall, = pw dh , X o has probably exceeded downwarping in the main outlet area pw - o dx (i.e. Lake Eyre region). where Zo = change in height of hydrocarbon column due In summary, it appears that the groundwater regime of the to hydrodynamic effect Great Artesian Basin in its central, western and southwestern = hydrodynamic gradient portions has been essentially the same since Early Cretaceous times. There appears to be no evidence that regional gradients Xo = width of oil accumulation. have been greater in the past, especially during the initial period of primary hydrocarbon migration, (i.e. mid Cretaceous This equation is a rearrangement of that derived by Hubbert to Early Tertiary (Poll, 1981; Moore & Pitt, 1984)). Higher (1953), i.e., gradients may have existed during the Jurassic but it is pointed pw dh out that to achieve a gradient some 5-10 times greater than that tan e = • — observed today would require a mountain range along the w-o dx present eastern coast of Australia somewhat similar to the Himalayas. Evidence for such a high altitude range does not where © is angle of tilt of oil water contact imposed by exist (Oilier, 1982). hydrodynamic gradient. P

T A B L E 2. Fluid data at reservoir and surface conditions for selected Eromanga Basin oilfields (refer Fig. 2 for locations). A.

OIL Field

Formation

Surface Conditions Density (gm/cc)

Temp. (°C)

Reservoir Conditions

Pressure (kPa)

Density (gm/cc) 0.676

Dullingari

Murta Member Mooga Formation

0.784

101 (atmosph)

Moorari

Birkhead Form.

0.762

101 (atmosph)

Pressure (kPa)

Surface Density x 100 Reservoir Density

106

14020

116

Merrimelia

Hutton Sandstone

0.760

101 (atmosph)

Strzelecki

Hutton Sandstone

0.813

101 (atmosph)

0.745

1. Avge.

0.780

(density changes between 16 and 21°C are insignificant.)

116

16700

Using 1. and 2. by assuming surface oil density is approximately conditions, an average reservoir density is about 0.69 gm/cc. B.

Density (surface) x 100

Temp. (°C)

109 2. Avge. 113

13%

higher

than

that

reservoir

WATER

Water salinities within the Eromanga Basin are generally less than 7000 mgl total dissolved solids so subsurface densities are very close to 1 (see Fig. 2, Schowalter, 1979).

HYDROCARBON FLUSHING Obviously, when dh/dx = 0 i.e. under hydrostatic conditions, the angle of tilt is zero and there is no hydrodynamic effect. When groundwater gradients do exist within the reservoir, they not only affect the total thickness of hydrocarbon column which may be trapped, but tilt the oil water contact in the direction of flow such that if reservoir dips are less than this angle of tilt, some hydrocarbons will be lost from the trap (see Fig. 5). The significance of this equation in relation to the Eromanga Basin was first pointed out by Holmes (pers. comm., 1980) and later mentioned by Bowering (1982). However, their papers did not evaluate the effects of palaeoflow regimes on hydrocarbon accumulations. Although both authors calculated tilt angles and hence trap angles for present day flow regimes it is worthwhile to repeat the exercise. Present day gradients in the central and southwestern Great Artesian Basin (i.e. Eromanga Basin) vary from 1:2000 to about 1:6000. Insertion of these values, and of densities shown on Table 2 in the equation, yields the following results: GRADIENT TRAP ANGLE REQUIRED 1:2000 (steepest regional) 0 °5 ' (0.09 °) 1:6000 (shallowest regional) 0 °2' (0.03 °) 1:400 (for comparison) 0 °27' (0.46 °) Figure 5 illustrates the tilt angle in relation to the dip of the structure. Considering that all Eromanga Basin structural traps located so far have closures greater than 0.5 it can be seen that it would be theoretically impossible to flush hydrocarbons under the present regime—even in local steeper gradient zones. An attempt has been made to measure the tilt of the oil water contact on some of the Jurassic fields using petrophysical log data but the errors involved in measurements of depth are too great to detect such small tilts. It has already been concluded that as far back as the mid to Late Cretaceous (i.e. the accepted time of commencement of primary migration) palaeo-hydrodynamic gradients were less than half those of today, because maximum uplift only occurred in the mid-Late Tertiary. Bowering (1982) suggests that the high-level abandoned mound springs around the southwest basin margin (Fig. 4), are indicative of higher hydrodynamic gradients in the past, however, these were deposited on a higher level Pleistocene landscape and indicate that the gradient may in fact have increased since that time because the discharge level is now lower. Bowering (1982) also suggests that palaeogradients were higher because of increased rainfall. This was probably the case but the increased inlet elevation would not exceed 30-40 m, i.e. the distance between the current water table and the ground surface in the present recharge areas. Once this zone was fully saturated, no matter how much the rainfall may have increased, the majority would end up as runoff. Such an unlikely increased inlet head would increase the gradient by about 10 per cent. It could be argued that structural closures during mid Cretaceous times were less than those observed today. This is almost certainly the case because significant folding is recorded during the mid-Tertiary phase of tectonism in the Eromanga Basin (Senior et al., 1978; Wopfner & Twidale, 1967). However, pre-Early Cretaceous (Cadna-owie Formation) dips of more than 1 ° are recorded for structures such as Jackson implying presence of suitable structural traps at that time. Even if most existing structures throughout the Eromanga Basin had lower dips prior to the Early Cretaceous, it is considered that palaeo-gradients would have been unable to flush a structure with more than 0.03 ° of dip in the direction of flow. Further, as gradients increased in response to uplift around the eastern basin margin, flushing of most low angle

383

traps would have been prevented by concurrent structural growth within the basin proper.

HYDROCARBON TRANSPORT BY SOLUTION

It is difficult to estimate the rate of solution transport of oil within the reservoir system because there is little information on the concentration of hydrocarbons in the water. The concentration is very low in groundwaters (up to 3 ppm in areas beneath large oil fields—see Dickey, 1979). Concentration increases with temperature and could be of the order of 20 ppm where water and oil are well mixed (Price, 1976). Natural reservoirs have segregated oil and water i.e. mixing is minimal. However, an attempt is made below to estimate the effect of solution transport of hydrocarbons through the Eromanga Basin. The volume of water moving through a ten metre thick top section of the basin in 5 million years at an average flow rate of 2.5 m per year (Habermehl, 1980) is 1.9 x 10 m per kilometre cross section (at 15% porosity). Over a 100 km section (i.e. similar to the section overlying the Cooper Basin) the volume is 1.9 x 10 m . Assuming an average hydrocarbon concentration of 2 ppm, the volume of oil removed is 3.8 million m (approx. 24 million barrels). Allowing for lower flow rates in the earlier Tertiary (say 0.8 metres per year), some 16 million m (100 million barrels approx.) would have been removed since the main migration phase. This figure is a large overestimate for the following reasons: (i) oil is segregated into traps and not dispersed over the whole top section of the reservoir; (ii) the thickness of reservoir containing hydrocarbons could be more like 1 metre (cf. 10 metres); (iii) with time, the less soluble components will become concentrated at the oil water interface, thus protecting the more soluble components from further dissolution; (iv) the actual concentration of oil in the water is likely to be much less than 2 ppni; (v) where there is a stratigraphic component to the trap there will be little or no water flow or solution (this may have been more important prior to the major folding episodes). In summary, solution transport of at most 100 million barrels of oil is likely to be very insignificant compared with the known source potential of the Eromanga Basin (e.g. Armstrong & Barr, 1982). 10

HYDROCARBON OCCURRENCES IN MOUND SPRINGS

Bowering (1982) describes occurrences of hydrocarbons in core samples from Coward Springs (Fig. 2) and attributes their presence as direct evidence of physical hydrocarbon migration to the basin margin. It is suspected that these occurrences are in fact related to solution transport and not physical migration as can be shown by the following example. Price (1976) estimates that average solubilities of crude oil at 100 °C are around 20 ppm. Even if we assume perhaps an unrealistically low value for the Eromanga Basin (say 0.02 ppm or 0.02 mg/L) over a flow period of say 1000 years (probably a short life for a mound spring), the total amount of so'uble organic matter (TOM) brought to the surface is: TOM = concentration x flow rate x time to surface = 0.02 x 10 x 0.8 x 86 400 x 365 x 1000 = 500 kg (Flow rate for Coward Spring of 0.8 L sec is from Cobb, 1975). 6

384

T. WILLIAMS & K. MORIARTY

Although much of this may be dispersed to the atmosphere and spread over a wide area of the spring, there is still a significant leakage of water flow up through the spring mound. Therefore, significant quantities of heavier hydrocarbons such as the bitumens will be precipitated within the mound sediments along with the carbonates and other salts which often comprise the mound itself. In fact, occurrence of hydrocarbons in mound spring sediments is to be expected, and does not indicate flushing, nor does it directly contribute information which can be used to locate traps in the subsurface.

CONCLUSIONS

Hubbert's equation which predicts the tilt of the oil water interface under hydrodynamic conditions gives a value of less than 0.1° for the Eromanga Basin under today's conditions. A review of the tectonic history and palaeohydrogeology indicates that palaeo-gradients have not been more than 10 per cent higher than present values. The highest gradients probably existed in the late Pleistocene, whereas structural growth had created structures with dips of 1 ° as far back as the Jurassic. The highest regional gradient envisaged for the Eromanga Basin would not remove hydrocarbons from any trap with a dip closure of greater than 0.1°. Local gradients may be higher but there is no evidence of large gradients in the reservoirs tested to date. It is therefore concluded that hydrodynamics have never

played a significant role in Eromanga Basin hydrocarbon migration. Long distance migration of hydrocarbons might be further prevented by the potential abundance of stratigraphic traps found in fluvial sediments. These may conceivably have provided hydrocarbons to later structural traps if sufficient tilting occurred. Even a slight stratigraphic discontinuity will trap oil if the hydrodynamic gradient is favourably aligned. Major faults did not exist in the mid-Late Cretaceous and would not be likely sites for trapped oil. The fact that nearly all of the oil-bearing structures located so far are not filled to spill point (the exception being Naccowlah South) is a compelling argument that much of the oil had migrated prior to the major phase of folding in the Tertiary. Whether the major trapping mechanism was stratigraphic or structural prior to this remains to be argued. Lastly solution transport is insignificant compared to the source potential and known oil reserves of the basin. It will contribute some hydrocarbons to mound spring sediments, but no inference can be drawn from this about major flushing of oil from the subsurface.

ACKNOWLEDGEMENTS

The authors wish to acknowledge Professor J. W. Holmes of Flinders University, South Australia, Professor J. D. Haun of the Colorado School of Mines and exploration staff members of SANTOS for helpful discussions concerning this paper.

REFERENCES ARMSTRONG, J. D. & BARR, T. M., 1982: The Eromanga Basin; in

Moore, P. S. & Mount, T. J. (compilers) Eromanga Basin Symposium summary papers. Geol. Soc. Aust. & Pet. Explor. Soc. Aust., Adelaide. BERG, R. R., 1975: Capillary pressures in stratigraphic traps. Am. Assoc. Pet. Geol. Bull. 59, 9 3 9 - 5 6 . BOWERING, O. J. W., 1982: Hydrodynamics and hydrocarbon migration—a model for the Eromanga Basin. A PEA J. 21(1), 227-36. BROWN, D. A . , CAMPBELL, K. S. W. & CROOK, K. A . W., 1958: The

geological evolution of Australia and New Zealand. Pergamon Press, Oxford.

CHIARELLI, A., 1978: Hydrodynamic framework of Eastern Algerian

Sahara—influence on hydrocarbon occurrence. Am. Assoc. Pet.

Geol. Bull. 64, 6 6 7 - 8 5 . COBB, M. A., 1975: Sampling

and measurement of mound springs, Great Artesian Basin, South Australia. Progress report No. 2. Marree, Curdimurka and Billa Kalina Sheets. S. Aust. Dep. Mines Ener. Rep. 7 5 / 9 0 (unpubl.). COLLINS, A. G., 1975: Geochemistry of oil field waters. Elsevier, Amsterdam. DICKEY, P. A. 1979: Petroleum Development Geology. Petroleum Publ. Co. Tulsa. FORBES, B. G „ 1966: The geology of the M A R R E E 1:250 0 0 0 map area. S. Aust. Geol. Surv. Rep. Invest. 28. HABERMEHL, M . A., 1980: The Great Artesian Basin, Australia. BMR J. Aust. Geol. Geophys. 5, 9-38. H U B B E R T , M. K., 1953: Entrapment of petroleum under hydrodynamic conditions. Am. Assoc. Pet. Geol. Bull. 37, 1954-2026. POLL, J. J. K., 1981: The significance of the southwest Eromanga Basin oil and gas discoveries. A PEA J. 21(2), 33-8. PRICE, L. C., 1976: Aqueous solubility of petroleum as applied to its origin and primary migration. Am. Assoc. Pet. Geol. Bull. 60, 213-34.

SCHOWALTER, T. T., 1979: Mechanics of secondary

hydrocarbon migration and entrapment. Am. Assoc. Pet. Geol. Bull. 63^ 723-60. SENIOR, B. R. & HABERMEHL, M. A., 1980: Structure, hydrodynamics and hydrocarbon potential, central Eromanga Basin, Queensland, Australia. BMR. J. Aust. Geol, Geophys. 5, 47-55. SENIOR, B. R., M O N D , A. & HARRISON, P. L., 1978: Geology of the Eromanga Basin. Aust. Bur. Miner. Resour. Geol. Geophys. Bull. 167. SPRIGG, R. C., 1961: On the structural evolution of the Great Artesian Basin. APEA J. 1, 27-56. STONE, D. S. & H O E G E R , R. L., 1973: Importance of hydrodynamic factor in formation of Lower Cretaceous combination traps, Big Muddy—South Glenrock area, Wyoming. Am. Assoc. Pet. Geol. Bull. 57, 1714-33. VEEVERS, J. J., JONES, J. G . & POWELL, C. M C A . , 1982: Tectonic framework of Australia's sedimentary basins. APEA J. 22(1), 283-300. W E L L M A N , P., 1979: On the Cainozoic uplift of the southeastern

Australian highland. Geol. Soc. Aust. J. 26, 1-9.

W E L L M A N , P., & M C D O U G A L L , I., 1974: Potassium argon ages on

the Cainozoic volcanic rocks of New South Wales. Geol. Soc.

Aust. J. 21, 2 4 7 - 7 2 . WILLIAMS, A. F., 1981: Formation

water study, Tirrawarra Field. SANTOS Ltd (unpubl.). 1974: Post Eocene history and stratigraphy of northeastern South Australia. R. Soc. S. Aust. Trans. 98, 1-12. W O P F N E R , H . , FREYTAG, I. B. & HEATH, G. R., 1970: Basal JurassicCretaceous rocks of western Great Artesian Basin, South Australia: stratigraphy and environment. Am. Assoc. Pet. Geol. Bull. 54, 3 8 3 - 4 1 6 . W O P F N E R , H . , CALLEN, R. & H A R R I S , W . K., 1974: The Lower Tertiary Eyre Formation of the southwestern Great Artesian Basin. Geol. Soc. Aust. J. 21, 17-52. YOUNGS, B. C. 1975: The hydrology of the Gidgealpa formation of the western and central Cooper Basin. 5. Aust. Geol. Surv. Rep. Invest. 43.

WOPFNER, H.,