Issuu

Aus tra lian Coal Geo log y

CONTENTS Volume 5 )>

-I

::z,

G. R. Holdgate: Stratigraphy, Structure and Brown Coals of the Moe area

r)>

z n

0 )> rC)

P. G. Flood:

m 0

Facies Study of the Callide Seam, Central Queensland: Implications for Mine Planning and Design

JOURNAL OF THE COAL GEOLOGY GROUP

A. C. Hutton: Organic Petrography of the Rundle-Stuart Oil Shale Deposit, Queensland

GEOLOGICAL SOCIETY OF AUSTRALIA

:s: m

G. R. Holdgate: Latrobe Valley Brown Coals - their Geometry and Facies equivalents as a guide to Depositional Environment

(11

G. R. Hatherly, P. F. Irwin and G. E. Holt: ~

Application of In-seam Seismic methods to Coal Mine Planning

(,0

0) (11

VOLUME 5

1985

ISSN 0725-4938

COAL GEOLOGY GROUP OF THE GEOLOGICAL SOCIETY OF AUSTRALIA

Chairman: Dr. C.W. Mallett Vice-Chairman: Mr. J.C. Anderson Hon. Secretary: Mr. L.R. Grimstone Hon. Treasurer: Mr. P.E. Balfe Committee: Dr. L.H. Hamilton Mr. E.A. Pryor Hon. Editor: Mr. J.W. Beeston

Address for editorial matters: Hon. Editor Australian Coal Geology GPO Box 1820 Brisbane. 4001 Address for other matters: Hon. Secretary Coal Geology Group 16 Dome Street Eight Mile Plains. 4123.

Price to non-members, $10.00 per issue

EDITORIAL It is now six years that this journal, then called simply 'Coal Geology', first appeared. Since then, not only the name has changed, but there have been a number of cover designs, formats etc., each reflecting the flavour of the Divisional Executive that produced it. Some of these changes can be confusing. The name change (along with corresponding ISBN number) means that some libraries store different editions on different shelves. As well, the style changes tend to make the series appear somewhat eratic on the bookshelf. But what of the contents. The journal has, with this edition, made available ninety-five papers relevant to Australian coal geologists, and together with additional abstracts, has amassed contributions from over one hundred and seventy authors. Perhaps the cosmopolitan appearance of the journal is only fitting, for within a venue supposedly belonging to a 'specialist' group, we find topics as diverse as palaeobotany, geophysics, oil shale petrology, mine planning, coal measure sedimentation and electric power generation. The authors, too, include lecturers, explorationists, government geologists, mine geologists, consultants and conservationists. And that is as it should be. The journal was begun to disseminate ideas. It has succeeded so far, and will continue in the future for as long as optimism continues to be an integral component of the industry.

CONTENTS Volume 5

G. R. Holdgate: Stratigraphy, Structure and Brown Coals of the Moe area

., •

P. G. Flood: Facies Study of the Callide Seam, Central Queensland: Implications for Mine Planning and Design

A. C. Hutton: Organic Petrography of the Rundle-Stuart Oil Shale Deposit, Queensland

G. R. Holdgate: Latrobe Valley Brown Coals - their Geometry and Facies equivalents as a guide to Depositional Environment

G. R. Hatherly, P. F. Irwin and G. E. Holt: Application of In-seam Seismic methods to Coal Mine Planning

•I

STRATIGRAPHY, STRUCTURE AND BROWN COALS OF THE MOE AREA By G. R. Holdgate

ABSTRACT A program of stratigraphic and regional groundwater drilling in the eastern half of the Moe Swamp Basin has outlined an area of new coal reserves with economic potenial. * These reserves occur within a thick sequence of sediments which overlie the Thorpdale Volcanics towards the centre of the Moe Basin. The most economic coal sequence occurs along the Moe Monocline, a newly defined subsurface structure in the eastern part of the basin. Some 100 million tonnes of coal are considered to be economically recoverable. The coal seams are sediments interseam and stratigraphically equivalent to the Yallourn Formation of the Latrobe Valley.

INTRODUCTION AND METHODS In the latter half of 1980 the Exploration and Geological Division of the State Electricity of Victoria (SECV) began an investigation of the Moe Swamp Basin (Figure 1) in terms of the geology, and its relationships with and possible sources of groundwater recharge to the Latrobe Valley Depression. Geophysical and strata logs of all government and private water bores were examined. A private bore drilled for groundwater 5km west of Moe was relogged using a down hole gamma-ray logger. This confirmed the shallow occurrence of a 36m thick coal seam within 10m of the surface. An SECV drilling rig was placed near this bore to obtain samples of the coal, and drilled to deeper the investigate to 121~ stratigraphy. The bore did not penetrate any further major seams. The moisture content of the main seam was found to be similar to the Yallourn Seam at Yallourn but the ash content to be higher, at around 5.0% (dry basis). In 1981 a drilling programme was initiated to investigate the stratigraphy and structure of the eastern half of the Moe Swamp Basin and, at the same time to identify the extent and quality of the se~m.

At about this time the Department of Minerals and Energy zoned the western half of the Moe Swamp Basin, including Trafalgar and Yarragon, outside a newly pro_claimed Government Coal L~ase Area, which then became open to private companies for investigation. The successful lessee, Esso Australia Ltd. has subsequently completed an investigation of the coal deposits in this area. Reports on this work have been submitted to the Department of Minerals and Energy (DMEV) - Fraser (1982/1983). The present study has consequently been confined mostly to the eastern half of the Moe Swamp Basin, within the Government Lease Area. To date 37 bores totalling about 3000m have been completed in this program. Some 2000m of contract drilling was done by percussion methods which provided reliable samples of the coal measures, and gave accurate definition of the strata below the coal where aquifer pressured sands occurred. The remaining metreage was undertaken by SECV percussion and rotary drilling rigs. Three of the deeper bores were established as observation bores to complement the groundwater program of pump bores established at Trafalgar and Willow Grove (Brumley & Holdgate, 1983). All coal samples were submitted to SECV Coal Science Laboratory for analysis on the usual basis of bulking 3m intervals for moisture and ash, 6m intervals for minerals and inorganics, and 12m intervals for proximate and ultimate analysis.

* As defined by the Brown Coal Resources Inter-Departmental Committee (BCRIDC) July 1977.

EASTERN

HIGH LAND S

SOUTH

GI PPSL ANO

HI GHL AND S

VICTO RI A

S( AL[ OF ~ ILO M[T IU S

Figure 1: Moe Swamp Basin locality map.

GEOLOGY Regional Setting The Moe Swamp Basin covers an area of approximately 400km2 and is divided off from the western end of the Latrobe Valley Depression by the uplifted Haunted Hills west of Yallourn. It is located approximately 100km to the east of Melbourne (Figure 1).

The Basin is ovoid in shape (Figure 2), and is ringed by Palaeozoic rocks of the Central Highlands to the north, the elevated basalt capped plateaus of the Heath Hill and Warragul Blocks to the west and the Mesozoic rocks of the Strzelecki Ranges to the south. Low lying swamps along the southern margins of the basin lie

at an elevation of 60m above sea level which is some 10m above the base level of the Latrobe Valley Depression. The swamps now have been largely drained eastwards into the Latrobe River by the Moe Main Drain.

•I I Stratigraphy

The Tertiary sedimentary and volcanic material within the basin exceeds 300m in places althought over most of the northern half is probably less than 200m. The se- • quence as shown by the generalised cross section (Figure 3) comprises four main units. From the top these include -

•l

I I

thick of Pliocene clays, sand clays, sands and gravels usually of a mottled appearance, which correlate with the Haunted Hill Gravel Formation in the Haunted Hills. This unit has been incised into and partly removed from the southern side of the basin where the main low lying swamplands occur. The edges of this downcut area are marked by distinct terracing. The Haunted Hill Gravel Formation unconformably overlies the Latrobe Valley Group with varying degrees of angular discordance depending on structural location. The unconformity spans most of the late Miocene.

termed the Yarragon Formation* , comprise the major coal bearing sequences. This formation up to 200m thick includes interbedded sands, clays and coal seams with sands predominating towards the base, and clays and coals predominating towards the top. Lateral facies changes reflect the fluvial dominated nature of this interval and have a marked effect on the lateral limits of the major coal seams. Outcrops of this formation may also occur along the Yarragon Monocline on the southern mar~·of the basin (P. Bolger pers comm). Over m eh of the northern half of the basin, and a Moe, this unit has been largely removed by the late Miocene erosion, so that in many places the Haunted Hill Gravel Formation rests directly upon the Thorpdale Volcanics.

2. The post-basaltic Latrobe Valley Group sediments of Middle Miocene age herein

* Refer to Appendix II.

1. A flat lying surface veneer up to 90m

LEGEND

D D ~

RECENT ALLUVtUH

HAUNT£0 Hill GRAVELS

FORMATION Tt-lCRPO ALE VOLCANICS ..UIO

~ CHILDERS FM. IUNOIFFERENTIAHOI

vvvvv

~ STRZELECKI GROUP f77'/1 PALAEOZOIC ROCKS tLLLI IUNOiFFE RENTIAHOJ

VYYIIVV

vvvvvvv vvvvvvv vvvvvvv vvvvvvv vvvvvvv vvvvvvv vvvvvvv

-----:'

v-.1,. r..OV RN O P [ ,.i

cu,

--- ---- ·' • Figure 2: Geological map of the Moe Swamp Basin based on the 1;250 000 Warragul Sheet (Victorian Geological Survey, 1971).

MOE

SWAMP

.,,

BASIN

E"_:::;

::, u

"'~

C: 0 C: 0

O:E

-E ::,

::,

a. 0 C: L

::, 0

V V

V V V

Mo rwel l F m. 1 km '--'

■ B ro w n Coal Seam s

100

200 ve rtical exaggeration x 10

Figure 3: Simplified E-W geological section of the Moe Swamp Basin. ♦

3. Late Oligocene to Early Miocene basalts of the Thorpdale Volcanics are included within the Latrobe Valley Group, and underlie the above units. Limited palynological evidence (Partridge in Fraser, 1982) suggests there may be a break between the Miocene sediments of the Yarragon Formation and the Thorpdale Volcanics. The volcanics consist of an almost continuous basaltic sequence up to 180m thick in the western half of the basin which splits eastwards into an interbedded basalt and sediment sequence near Moe. Thin coals within the interflow sediments here were correlated by Blake (1972) with the Morwell lB and Morwell 2 seams. Three or more basalt layers at Moe are separated by up to 40m thick intervals of clays, sands and minor coals. The upper surface of the Thorpdale Volcanics is invariably deeply leached and altered to clay regardless of depth. Similar weathered to fresh volcanics outcrop extensively on the hills surrounding the Moe Swamp Basin, and have been dated by Wellman (1974) as between 26.8 and 19.3Ma. Similar ages were obtained on core samples of basalts from some of the Esso bores (Fraser, 1982). 4. Underlying the Thorpdale Volcanics is a thin interbedded gravel, sand, clay and coaly sequence about 20m thick which

overlies basement rocks of either Mesozoic and Palaeozoic age. This unit can be correlated by stratigraphic position with the Oligocene aged Childers Formation, and is the lowest stratigraphic unit of the Latrobe Valley Group in this area. Coal seams in this unit may be up to 8m thick. Outcrops of this unit also occur under basalts in the hills peripheral to the basin. 5. Basement rocks underlie the Childers Formation or Thorpdale Volcanics unconformably. In five basement bores north of the Gippsland highway four encountered Palaeozoic rocks including Lower Devonian dark grey shales of the Wilsons Creek Shale (Bolger, 1982). The fifth bore at Moe encountered Mesozoic sediments known as the Strzelecki Group. South of the highway, three basement bores have encountered Strzelecki Group sediments, and one at Yarragon bottomed in Palaeozoic rock. As there are no records of Strzelecki Group outcrops on the northern margins of the Moe Swamp Basin it is presumed that the Mesozoic sequence is thin to absent over all but the very southernmost margins of the basin where it rapidly increases in thickness south of the Yarragon Monocline. Bores at Coalville in the Strzelecki Hills have recorded over 900m of Strzelecki Group sediments.

• II

• I

•

Structure The Moe Swamp Basin is structurally controlled by monoclinal warping and faulting on its southern, western and eastern margins. The major flexures such as the Darn um and Yarragon Monoclines form topographic edges to the highlands of the Balook Block to the south and the Warragul Block to the west and in places have folded the Tertiary sediments outcropping along the edges. The Latrobe Valley Group sediments withing the basin also show similar evidence of folding. The northern edges of the basin appear to be less structurally controlled. Here the Tertiary cover appears to be gradually eroded off from a southward dipping basement surface. Drilling has located a number of subsurface structures in the eastern half of the Moe Swamp Basin. These include the NNE trending Moe Monocline (new name) which downthrows the Latrobe Valley Group 5km west of Moe, by up to 200m into a NNE trending structure known as the Trafalgar Syncline which reaches its maximum thickness beneath Trafalgar (Figure 2). The directions of these subsurface structures accord to the dominant trend of structures in the Latrobe Valley, although the sense of downthrow is westerly rather than the more usual easterly direction. Other subsurface structural features occur in the western half of the basin (Fraser, 1983). The Moe Swamp Basin appears structurally related to the Latrobe Valley Depression, and the definition of these subsurface structures remains the prime aim of any drilling programme. The structural location significantly effects the extent of economic coal, and the assessments to be made of groundwater movements. Thick and extensive coal seams provide marker horizons which are invaluable in defining these structures, whereas thin bedded sand and clay sequences are more difficult to correlate from bore to bore.

• Detailed Stratigraphy of the Coal Seams and Interseams All the economic coal resources occur within the upper stratigraphic levels of the Yarragon Formation. In the eastern half of the Moe Swamp Basin the majority of this

resource is concentrated along the upthrown edge of the NNE trending Moe Monocline (Figure 2 and 3). East of the monocline the seams initially flatten, and then thin rapidly due to later erosion over about a one kilometre width in the direction of Moe. At the monocline they plunge steeply westwards to depths up to 150m quickly passing beyond economical ove/ burden criteria, and simultaneously split and become lenticular into the sands and clays in the Trafalgar Syncline (Figures 4 and 5). As a consequence of this facies change any coal seams which may subcrop within the study area on the northwestern side of the Trafalgar Syncline are probably too thin to be considered economic. However, the coal seams thicken again on the southwestern flank of the Trafalgar Syncline where a broad domal structure containing economic coal has been located by Esso in the Yarragon township area. Tlie economic coal zone near Moe occurs over a strike length of more than eight kilometres, trending NNE from the intersection of Kenny Road and the Princes Highway to the Westbury - Willow Grove Road, where the seams become lenticular and are eroded (Figure 6). Drilling along Kenny Road south of the highway indicates that the seams may subcrop in a narrow zone on the uplifted edge of the Yarragon Monocline, but this edge remains to be better defined. The coal resource occurs as two main seams, comprising an upper A seam 5-15m thick, and a lower B seam 25-36m thick separated by a clay and sand intervai averaging 6m in thickness. The seams tend to thicken to the SSW where the interseam is at a minimum. West of the monocline the interseam interval increases to over 30m gradually replacing both coal seams as shown by cross sections (Figures 4 and 5) . Within the economic coal zone (Figure 6) the A seam is partially or completely eroded, so that most of the economic resource occurs within the B seam. In this seam the upper 2-3m and the lower 3-5m is usually inferior coal with an ash content greater than 10%. As a consequence the average economic coal thickness used in the 11 bores for reserve calculations is take at 21.3m.

Thinner inferior coals up to 5m thick occur both above and below these two main seams in the Yarragon Formation, but have not been considered in the economic calculations. Coals interbedding with the Thorpdale Volcanics beneath Moe, and those presumed to occur within the Childers Formation in the study area similarly have not been considered as part of the resource either because they are too thin, or have basalts present in their overburden section or both. In the economic coal zone between the Moe Main Drain and the Princes Highway the overburden thickness varies between 7 and 17m. North of the Main Drain' an abrupt 20m rise in the topography marks the southern limit of an extensive terrace of Haunted Hill Gravel Formation which covers the northern half of the Moe Swamp Basin. Consequently, the economic coal

zone north of the terrace edge is more deeply buried, with overburden thicknesses of about 25m. West of the Moe Monocline overburden thicknesses increase over a very short distance to exceed the current 90m overburden limit criteria.

Coal Quality Seven bores within the economic coal zone have been analysed for their moisture and ash content. The weighted average moisture content in the B seam for coal below 10% in ash is 64.4%, with a weighted average ash content of 5% (dry basis). The moisture content in the B seam tends to decrease to less than 60% west of the Moe Monocline, probably due to folding of the seam coupled to an increase in overburden. There is also a tendency for the moisture content to reduce by a few percent towards the southern end of the field .

. 1

• I

...

en;

! T!o!ORl'tl,t.U

:.=:~C1S i-= 5

i:1

~NE

• -

ca,.i.u ... , _ . , , . • ..,.,

l'.ifD IW[ltlOIII CO,,, l I- • ""-' ,0'11, I t.., c::::J s.MID • - ~Ml>! ~ ~ C UY

K(Mll ll0NTAL5C.\LE CID 100 lOQ 400 500 KIO 100 100 100 1000

a. ..,

MOE CROSS SE CTIONS (PARISH Of YAAR AGO NJ

,For locol 1Mol uc.t loM ue FIGUIIEI

Figure 4. Moe cross sections. 6

The A seam has a slightly lower average moisture content of 63 .7% but this may be attributable to a higher average ash content of 6.8%. A typical full analysis of the coal is given in the Yarragon 14 bore - hole, the results of which are presented in Table 1. This analysis indicates the coal is broadly similar to Yallourn Seam coal in the Latrobe Valley but has higher ash constituents of calcium,

M,11 ... Ro,u

S•"''"

silica, aluminium and iron. The remaining constituents are similar to typical Yallourn Seam coal. The higher silica, aluminium and iron may indicate more elastic sediment (clays) intermixing in the coal than for the Yallourn Seam at Yallourn. The higher ash content results in a gross dry specific energy value of 24.7 MJ/kg for the coal compared to that of 25.8 MJ/kg for the Yallourn Open Cut coal.

MCN~2f1NE

~"="--~"~"-- - - - - -- - -- - - - • ©

------"-"---- ---- ~i t

I•

------ ·---------ltO'lllONTAL S C,1,Lf

V✓..l

IMf U UOltCD,\.LIMor • lhon10 '11.& lu • lh o nlO'll. a PI I

C ) S ANO ~

MEt RE S

SAMOT CLAY

i!l:m CL AY

MOE CROSS SECTIONS !PARISH Of YA RRA.GONI tHOfl,-Q,l,Lt_

--"

Figure 5. Moe cross sections.

TABLE 1 MOE COAL QUALITY - YARRAGON NO 14 BORE

B SEAM (15 .5-64.8m). Results below given for coal < 10% ash (21.9-46.5 m) weighted average. Moisture % a.r. Ash(d.b.) % Volatiles (d.b.) % Fixed Carbon (d.b.) % Carbon (d.b.) % Hydrogen (d.b.) % Iron (total) (d.b.) % Titanium Oxide (d.b.) % Calcium (d.b.) % Potassium Oxide (d.b.) % Chlorine (d.b.) % Sulphur (d.b.) % Silica (d.b.) % Aluminium Oxide (d.b.) % Magnesium (d.b.) % Sodium (d.b.) % Oxides (d.b.) % Calorific Value (MJ/kg) Gross Dry Gross Wet Net Wet

Coal Quantities Figure 6 shows the area (4 900 000m 2 ) where coal meets the economic criteria ( <90.0 overburden, inclusion of < 15.0m seams as overburden, and <2.1 overburden to coal ratio). This area was multiplied by the averaged borehole thickness of economic coal (21.3m) to provide a reserve figure of 104 370 000 tonnes. A total figure of 572 783 000 tonnes inferred all coal was calculated for seams > 3.0m thickness within the study area without overburden, depth limit or ratio criteria applied. Some small additional reserve may be found along the less drilled southern edge of the basin between Kenny Road and Trafalgar.

65 .0 5.3 51.6 43.1 63.5 4.5 0.89 0.01 0.72 0.003 0.08 0.25 0.75 0.42 0.35 0.06 3.55 24.7 8.6 7.1

Correlation with the Latrobe Valley Coal Measures An interprete d stratigrap hic correlation between the Tertiary sediments of the Moe Swamp Basin and those in the Latrobe Valley Depression is shown on the east-west cross section on Figure 3. This illustrates that the Morwell (Latrobe) seams in the Latrobe Valley are warped and faulted over the Yallourn Monocline and then split and pinch out beneath the Haunted Hills into Thorpdale Volcanics at Moe and Newborough townships.

• •

~/~

~ o :;?

~£

~Ro

I f , 7 ~1 MOE COAL FIELD TOTAL COAL & INFERIOR COAL ISOPACHS, WITH AREA OF ECONOMIC COAL SHOWN IN STIPPLE

/-~

ISO PACH IN TERVAL IN METRES

4lil-- --J

•

-.L--

~<11---

I I

--==---===

1000

TRAFALGAR

4 1 - - - - 4 1 Sec t ion Lin e For cross sect io ns see Figure s 4 & 5

~00

1000

2000 m.tres

Figure 6: Moe coal field. Total coal and inferior coal isopachs, with area of economic coal shown in stipple.

•

The Yallourn Seam is faulted and eroded at the Yallourn Monocline and is not present beneath the uplifted Haunted Hills Block where presumably it has been eroded. From cross-sections it would appear that a stratigraphically similar seam re-emerges west of Moe to form the Moe Coalfield. Evidence to support the contention that this seam is a continuation of the Yallourn Seam includes the similiarity in moisture content, and the presence of the zonal pollen fossil T. bellus within sediments interbedding and underlying the Moe coal seam, indicating a Middle Miocene age similar to that of the Yallourn Seam (see Appendix 1). A continuation of these coals of age similar to the Yallourn Seam occurs on the western side of the Moe Swamp Basin near Yarragon (Fraser, 1983). Whether the Yallourn Seam and the Yarragon Formation coal seams in the Moe Swamp Basin were ever physically con-

nected prior to uplift of the Haunted Hills is difficult to prove due to subsequent erosion. Because no physical connection is evident, a different formation name (the Yarragon Formation) is herein proposed for a stratigraphic unit which is probably wholly • equivalent to the Yallourn Formation in the Latrobe Valley Depression.

CONCLUSIONS A drilling programme to investigate the stratigraphy and groundwater of the Moe Swamp Basin, and to follow up recorded occurrences of brown coal in shallow private water bores has shown taht the upper stratigraphic levels of the Latrobe Valley Group sediments (herein defined as the Yarragon Formation) comprise interbedded coals, clays and sands with the thickest coal seam up to 36m thick. The Yarragon Formation is unconformably overlain by the

Haunted Hill Gravel Formation which erode the formation on the northern and eastern margins of the basin often entirely removing this sedimentary sequence. In the eastern half of the Moe Swamp Basin the Yarragon Formation is folded along the 8 km long Moe Monocline which has resulted in the coal seams being at economic depths. Within this economic zone some 100 million tonnes of coal are available for open cut mining, with an average moisture and ash content of 64.4% and 5.0% respectively. The ash is comprised of inorganics, probably introduced as clays (silica, aluminium, iron) and intermixed with the peat environment at the time of deposition.

Correlations with the Latrobe Valley to the east indicate that the Moe coal seams may be westerly extensions of the Yallourn Seam, althought any physical connec_tion across the intervening Haunted Hills has been removed by subsequent erosion.

ACKNOWLEDGEMENTS The author wishes to acknowledge the Principal Geologist, Dr. Colin Barton and others of Exploration and Geological Division, Fuel Department, SECV for assistance in the completion of this programme. This paper is published with the permission of the State Electricity Commission of Victoria.

•

REFERENCES BCRIDC, 1977: The Designation and Protection of Economically Winnable Deposits of Brown Coal in Central Gippsland. Blake, W.J .R. , 1972: A feasibility study of a groundwater supply for the Moe township - including aspects of the stratigraphy, structure and hydrogeology of the Moe Basin, Unpub. Rep. Geol. Surv. Viet., 1972/49 (unpub.) Bolger , P., 1982: Examination of basement rock Trafalgar area. SECV Geology Laboratory Report GL 82/83. Brumley, J ., & Holdgate G.R. , 1983: An assessment of the confined groundwater resources of the Moe Swamp Basin, Victoria, Australia. Papers of the International Conference on Groundwater and Man, Vol 1: the Investigation and Assessment of ,Groundwater R esources, Australian Water R esources Council Conference Series No. 8, 41-50. Fraser, N., 1982: Half yearly report to Department of Minerals and Energy, Victoria, for period ending 24th March 1982. Esso Australia Ltd. Coal and Synthetic Fuels Department, Unpub. Rept.

Fraser, N., 1983: Exploration Licence 1800 Trafalgar Final Report to Dept. of Minerals and Energy, Victoria, Esso Aust. Ltd., Coal and Synthetic Fuels Dept. Unpub. Rept., Oct. 1983. Gloe, C.S. , 1976: Brown Coal, in Douglas, J .G. and Ferguson, J .A. (Eds) Geology of Victoria, Geological Society of Australian Inc. , 378-389. Stirling, J ., 1899: Report on the brown coals and lignites of Victoria. Progress Report No. 10 1899, 73-74. Wellman, P ., 1974: Potassium-argon ages on the Cainozoic Volcanic rocks of eastern Victoria, Australia. J. Geol. Soc. Aust. , 21 , 359-368.

G. R. HOLDGATE Exploration and Geological Division Fuel Department State Electricity Commission of Victoria 15 William Street, Melbourne, 3000.

. '

APPENDIX I

Palynological age datings and comments referring to SECV bores Yarragon 18 and 19 by A D Partridge of Esso Australia Ltd; ref 61 :ADP/me/SUP 24 September 1982. "Two additional samples have been examined for palynology from the Moe Swamp Sunkland as part of the Trafalgar Coal Project. The samples come from the coal bores Yarragon-18 at 64.8 metres and Yarragon-19 at 77 metres. These two bores are situated approximately six kilometres east of Esso Australia Ltd's area of investigation and 13 kilometres west of the Yallourn Open Cut. Diverse, well preserved spore-pollen assemblages were obtained from both samples, which are assigned to the Triporopollenites bellus Zone of Middle Miocene age. This age agrees with that obtained for the coal measures sequence in the Trafalgar project ar ea".

APPENDIX II Definition of a new subsurface stratigraphic unit in the Moe Swamp Basin. YARRAGON FORMATION. Derivation of name: Yarragon township (AMG grid reference 418000 E, 5770500 N, Moe Sheet 8121). Previous usage: new name. Distribution: occurs subsurface throughout southern half of the Moe Swamp Basin. Type Section: SECV Yarragon 49 bore (grid reference 429495.6 E, 5773875.7 N Moe Sheet 8121). Occurs between the depths of 20.4 and 220.0 metres. Lithology: lnterbedded medium to coarse quartz sands, ligneous silty clays, grey clays and brown coals. There is an increase in the number and thickness of brown coals and ligneous silty clays towards the top of the unit and an increase in the number of sand units and a coarsening of grain size towards the base of the unit. The brown coal beds thicken up onto structural highs such as the Moe Monocline. Thickness: range: 0 - 199.6 metres. average: 100 metres. type section: 199.6 metres. Relationships and Boundary Criteria : Within the Moe Swamp Basin, the Yarragon Formation overlies the Thorpdale Volcanics disconformably and is in turn overlain unconformably and eroded by the Pliocene Haunted Hill Gravel Formation. Palynological dating (Appendix I) indicates the Formation is probably wholly equivalent to the Yallourn Formation int the adjacent Latrobe Valley Depression.

•

.....

FACIES STUDY OF THE CALLIDE SEAM, CENTRAL QUEENSLAND: IMPLICATIONS FOR MINE PLANNING AND DESIGN By P. G. Flood

ABSTRACT

I •

Detailed facies analyses of the Callide Seam of the late Triassic Callide Coal Measures illustrate the changing pattern of peat formation through time. Six intervals of peat accumulation may be recognised. The area of the peat swamp varied considerably as a low sinuosity river system encroached upon the peat swamp progressively from the west, from the east and then from the north. Finally, plant growth within the peat swamp was not able to compete with the influx of sediment and elastic sedimentation prevailed. The river system left as a record of its existence sedimentary deposits which occur as two major partings within the seam. Crevasse splay deposits also occur between several of the coal plies.

Before future mine cuts are designed it is considered imperative to know the precise location of these partings which in the past have produced mining difficulties. Facies analysis is an inexpensive technique that can predict the 3-D geometry and lithology of these partings down dip from existing open cuts.

INTRODUCTION The Callide Basin Coalfield is located approximately 100km southwest of the Port of Gladstone (Figure 1). This Coalfield covers an area of about 18 000 hectares and is located within a northwest-southeast trending graben structure which is 22km long by 8km wide. The Coal Measures dip at about 4 ° towards the basin's axis (Grimstone and Roach, 1978).

OLD GLADSTONE•

•

-7

i I

CALLIDE BAS IN .

LLIDE BASIN

BRISBANE

,---------- -------/ i

100

Kilometres Figure 1: Location of the late Triassic Callide Basin, Central Queensland.

\.· . ~

.''

•

To Gladstone ___:_:_ ·oAKY C _AREA · • .

6 . : •.

··_. ·_ ..•

. ..

•__ •• •. • . _·· ·.

. ·_

· ·

• •

·1 0

-••

_· . ·_ 12

•

\A .

- . . .• . .. .. -

. . : .

,z .·... .·.

~14 . .

·.·..·..

Biloela

• A

.• :\ .2

·,

•.

• 14 • • •

0 I

TRAP GULLY' \ AREA '

.- . :

Kms

PRESENT MINE ARE i 'Aq4/f'' PETERSON GULLY- DUNN CK

Coal Isopleths in metres ; Simplified from Grimstone & Roach,1978 Figur e 2: Aerial extent of the Callide Basin, the location of the present mine area, and the isopachs of the Callide Seam.

Present open-cut mining operations are restricted to the Peterson Gully-Dunn Creek area and the Trap Gully area at the southern edge of the Basin and at Boundary Hill in the northwest (Figure 2). Mining conditions within the Callide Basin are typical of established Australian open-cut steaming coal mines. Firstly, the topography is rugged with steep sided hills and deeply eroded gullies. Secondly, a considerable thickness (up to 70m) of overburden needs to be removed. This overburden consists mostly of massive sand-

stone or interbedded sandstone and siltstone. Thirdly, two major coal splits occur and these are separated by partings of variable thickness consisting of sandstone, siltstone and carbonaceous mudstones up to 10m in thickness. The net result is that it is essential to be able to precisely predict the nature and characteristics of the intra seam and overburden sediments so that the mine design can be varied to suit the dragline capabilities (see Leveritt, 1978). In excess of 1 000 drill holes, totalling more than 65 000 metres and including

about 10 000 metres of core have been completed within the Coalfield. The pattern of drill holes and the location of the highwalls within the Peterson Gully-Dunn Creek area are shown in Figure 3. This is the area

which is discussed in the present paper. Results of the author's compilation and interpretation clearly illustrate the importance of sedimentary facies studies in mine planning and design .

. . .·.... •• ··•, 12 ·::.:.,

• 9- ,._ '. s.- A.REA • :.- • ; ...

•. ' ...

·-:

'>-.

••

·.-. .. .,·.. . .

Orillhole of Th1ess Bros P/ L

NS Holes • see Cameron(l 972 )

•

1979 Holes. 896 - 921 ; Electric Logs

Figure 3: Distribution of drill holes and highwa lls within the present mine area.

GENERAL GEOLOGY The regional geological setting of Callide Basin has been discussed by Dear and others (1971), Cameron (1972), Svenson and Hayes (1975), Grimstone and Roach (1978), Gould and Shibaoka (1980), Smyth (1980), and Stevens (1981). STRATIGRAPHY The Callide Coal Measures are about 120m thick consisting of a series of cycles of fluvial channel, levee bank, and peat swamp deposits. The Coal Measures have not been formally sub-divided into subgroups, formations, coal members, members etc. Within the present study area they contain several coal seams previously designated from youngest to oldest as: Marker A, Marker B, Callide, Sawmill and Bottom seams. Only the Callide Seam is at present of economic importance and in the Peterson Gully-Dunn Creek mine area it averages 16m in thickness. An idealised stratigraphic section is shown in Figure 4 and geophysical logs of the Callide Seam are illustrated in Figure 5.

These logs may be readily compared with existing seam lithotype profiles (e.g. see Fig. 2 of Svenson and Hayes, 1975). In places, the Callide Seam is divided into at least two major splits and the spatial arrangement of the partings is indicated in Figures 6 and 7. The compilation of bore hole information together with results of highwall mapping has enabled the various plies and partings to be recognised across the entire 5km east-west section of the Basin and work in progress is successfully extending this correlation to the north, down dip, away from the existing mine area. NOMENCLATURE The geotechnical group of Callide Mine have devised an informal nomenclature for the various recognisable subdivisions of the Callide Seam which is designated C. The upper split is referred to as the Cl split, the lower as the C2 split. The alpha-numeric code has been extended to the piles of each split as follows: C21 is the uppermost ply of the second (or lower) split of the Callide Seam, etc. and the individual plies may be

further subdivided (e.g. C141). Field mapping in recent excavations has shown that the Cll ply does not belong to the Cl split and is therefore referred to as the ''high ash tops". In the A-area this ply can be observed joining the base of the Marker B Seam. The parting which occur within the Seam are referred to the plies above and below the parting (e.g. the C21/C22 parting) or in one instance when a parting occurs between two splits, it is referred to as the Cl/C2 parting. In order that the reader is not confused by the nomenclature, the various stages in

the development of the Callide Seam are referred to intervals of time of peat accumulation. These are numbered from oldest to youngest as Intervals I to VI. The justification for considering time intervals is that elastic sedimentation was occurring within the fluvial channels which are now represented by the partings, whilst peat was accumulating in swamps. The Intervals may be thought of as representing chronostratigra phic units. (Clearly the lithostratigrap hic units may be time transgressive.)

PRECIPICE SANDSTONE

- - - - - - - - - - --

< w

a: <

MARKER B

I I

INTERVALS

OF PEAT

HIGH ASH TOPS White Claystone _ _Band I ___ _

::E

C12

i=-

w w

I I

::,

I-

0 N

CALLIDE

a: Cl)

< w

SEAM

::E ...J

...J

ci:i <

:i: E

::,

- - - - - - - -I - - - -

C2 1

◊WEST C22

C23 PLIES

I-

•

SAWMILL BOTTOM

ACCUMULATION

I-

a..

I I I

Cl)

I I I

I C13 I I I I C14 I I ◊EAST.6 I

0 0

_ __ _ Q

DISCONFORMITY

MARKER A

Cl)

V IV

III

I I

I I I I I I I I

II I

I •l

I ~

DISCONFORMITY Marker Bed

.6 Fining upwards cycle ◊

BASAL UNITS(TRIASSIC)

Major parting

Q Braided stream deposit Figure 4: Idealised stratigraphic column of the Callide Seam compiled from data available within the present mine area. (Coal. black; sediment, stippled).

•I

COAL QUALITY & SEAM THICKNESS LOGS CALLIDE SEAM

PLIES

600

J 200

271 ----

CALIPER

GAMMA

BED RESOLUTION DENSITY

HOLE 913C

----

White Claystone Band en

~ 4)

C12

E C

-----

-- -

(I)

I•

C13

I• ---

c, 41 C14

Fusain Horizon

C142

WMe Claysto] Ho,r;; -

C21

40 Nodular Ironstone Horizon

C22

• C23

NCPS

270

200

API

Figure 5: Representative electric logs of the Callide Seam ; position of diamond drill hole 913C shown on Fig. 3.

FACIES ANALYSIS The .term facies refers to the geometry ~nd mternal characteristics (such as lithology and structures) displayed by the sedimentary succession. It may be used · both in a descriptive and interpretative manner. Analysis of the sequence of facies usu~y indicate changes in the depositional enVll'onment through time (see Reading 1978; Walker, 1979). ' This fa,cies analysis of the Callide Seam is based upon a detailed study of available highwall maps, bore cores, and other subsurface information. Recognition criteria are now well established which enable the interpretation of the composition and geometry of sedimentary bodies in terms of

the depositional processes operative within a coal basin (see Allen, 1965, 1970; Beaumont, 1979; Etheridge and Schumm 1978· Horne and others, 1979; Howell ani Ferm: 1980; Jackson, 1978; Miall, 1978; Rust, 1978; Smith and Eriksson, 1979· Smith and Smith, ~9~0). Using information gained from a limited number of data points it is possible to predict the spatial arrangement of the sedimentary facies. Therefore, in the subs~face, one may extrapolate the likely physical and chemical properties of the c?al, the intra ~am sediments (i.e. partmgs), and the mter seam sediments (i.e. overburden). The result of facies analysis in the mine area are illustrated in Figures 7 and 8.

, I

• I

Me1res

250 -'

< u ;::

200

ffi>

,;-if1ooom

p~ 800m

150

600m 100

400m 50 200m

VER TICAL EXAGGERAT ION x4

200m

Fi~e. 6: Isome~ic diagram showing the partings within the Callide Seam and the plies.

I •I

I I

During Interval I (Figure 8) the peat swamp extended over the entire mine area. Towards the close of Interval I peat accumulation on the eastern sector was interrupted by sediments deposited from a river system. During Interval II (Figure 8) the peat swamp migrated to the east over the upper surface of this sediment. A large easterly migrating river system deposited a considerable thickness of sediment (i.e. C21/C22 parting) over the C23 ply whilst eroding part of the C22. On one occasion the fluvial channel flooded and carried in suspension an enormous quantity of ironrich clay which was deposited across the surface of the peat swamp. Subsequent diagenetic alteration of this clay produced the nodular ironstone horizon. Peat swamp conditions were reestablished during Interval ill and while peat continued to accumulate a major river system migrated westerly over the lower part of the C21 ply (Figure 8). This river deposited a range of sediment types including channel lag conglomerates, epsilon cross stratified point bar sandstones, flaserbedded levee sandstones and siltstones, all of which are now represented as the Cl/C2 parting. On one occasion the river flooded and its waters deposited a 10cm thick layer of mud across the upper surface of the peat swamp. During Interval IV (Figure 7) the peat swamp re-established itself over the top of the Cl/C2 parting. The river system was outside the mine area during the time taken for the deposition of the C14 ply. Occasionally the river levees were breached and in the western area several crevasse splays deposited sediment on the peat swamp. On other occasions the water table in the swamp dropped to a level sufficient to allow the upper surface of the peat to be exposed. These events are indicated by a 30cm fusain horizon which is the boundary between the C141 and C142 subplies. Peat accumulation was interrupted by another major flood which deposited a 3cm thick layer of mudstone. The position of this marker

horizon corresponds to the position of the crevasse splay deposit which occurs to the east. During Interval V peat accumulation continued and plant growth was re-established on the crevasse splay deposits. During this period major channel system encroached onto the peat swamp and severely reduced the area of peat accumulation (Figure 7). For a considerable period of time the position of the channel remained static as can be observed in the highwall of the D-cut where peat accumulation kept pace with the rate of sedimentation in the channel. On at least one occasion the river flooded depositing fine-grained sediment onto the peat swamp thereby producing the crevasse splay in the upper C13 ply. On at least three other occasions it deposited fine grained sediment across the surface of the peat deposit thus terminating peat accumulation for a brief period. Conditions during Interval VI were very similar to those of the previous Interval as the position of peat accumulation remained the same. At the close of this Interval the peat swamp was no longer able to compete with the influx of sediment. A 30cm thick white claystone band marks the top of the Callide Seam and represents the onset of predominantly elastic sedimentation. During the Intervals I to VI the peat swamp managed to maintain almost continuous growth in the immediate vicinity of the north western part of the D-area and the northern part of the B-area (Figure 9). Elsewhere the influx of sediment from, and erosion by, the river systems has produced a complex arrangement of partings in the coal seam. A diagrammatic representation of the conditions of sedimentation towards the close of formation of the Callide Seam is illustrated in Figure 10. Subsequent to Interval VI peat accumulated in two restricted areas, namely the B-area and the S-area, but there was a considerable influx of terrigenous sediment onto these peat swamp areas, hence the term "high ash tops". Finally sediments covered the site of the former peat swamp.

VERTICAL EXAGGERATION APPROX IMATELY SO TIMES

PEAT SWAMP C73

F:tl

CHANNEL SEDIMENTS

PARTING C12IC13

' II PEAT SWAMP Cl• CREVASSE SPLAY SEDIMENTS

Figure 7: Diagrammatic 5 km west-east section of the Callide Seam, Peterson Gully-Dunn Creek area. Facies arrangement during Interval V and IV showing the positions of the C12/C13 and C13/C14 partings respectively. 20

PARTING C13/C14

PARTING C14IC21

PEAT SWAMP C~r

CHANNEL SEDIMENTS

I •

PEAT SWAMP c2;

!'.

CHANNEL SEDIMENTS

PARTING C21IC22

f•

PEAT SWAMP C23

lli±l

CHANNEL SEDIMENTS

!: I, II and Intervals during arrangement Figure 8: Facies III showing the positions of the C22/C23 , C21/C22 and C14/C21 partings respectively.

PARTING C22/C23

...

0 Kms

PEAT SWAMP

FLUVIAL CHANNEL SEDIMENTS & TRACE OF EARLIER RIVER CREVASSE SPLAY DEPOSITS TRACE OF MAIN CHANN EL

Figure 9: Composite facies map showing the temporal and spatial arrangement of peat swamp and channel sediments. ~~~~~ ;;.,en •,c1VC1l ____ -____. . f CZZIC13

.~_t§\{ -'-:. : :. - .:::-~~~,.. E ARE~ ----.-

'-....,

- ~ ,r -->-- -

Ji.. '...

SAR~<'._

--{

czvcn ~- 500

Figure 10: Diagrammatic representation of the relationships between the fluvial channel, meander belt sediments, and the peat swamp during the last stage in the formation of the Callide Seam.

,ooo

CIJ / C 1'

CONCLUSIONS The analysis of the spatial and temporal arrangement of peat swamp and channel facies within the Callide Seam in the has Creek Gully-Dunn Petersen demonstrated the interrelationship between the peat swamp which is now represented as a vertical accumulation of 16m of sub-bituminous coal and the intra seam sediments of the river system. Plant accumulation in the swamp struggled to survive an influx of sediment and erosion from a migrating river system which left a record of its presence as two major partings within the seam. Finally the rate of sedimentation exceeded the rate of peat accumulation. It is obvious that when future cuts are made the mining difficulties produced by the increased thickness of the partings will be even more pronounced than at present. For example the area of the washout in the D-ara will expand enormously and the C21/C22 parting will influence future cuts in the C and B areas. When this occurs, the overburden to coal ratio and not topography alone may be the deciding factor in mine design. The geological details available to mining engineers to date have included isopach maps of the overburden and coal from which the overburden/coal ratio is determined. This ratio has been used as the primary factor influencing mine design: subsequent changes to the mine plan have been made when difficulties were encountered with partings within the coal seam. A more suitable method of decision making would be to use computer-based mapping techniques to generate maps such as:

1. overburden/coal ratio; 2. split/lower coal ratio; and 3. overburden/upper coal ratio, etc. Moreover, the mapping criteria may change (e.g. price of coal increases), requiring the generation of new derivative maps. This integrative mapping technique should provide a rapid tool for mine planning and design. The next phase of this research is to extend the facies analysis technique to the other area of "thick" coal with the aim of producing a synthesis of sedimentation within the entire Callide Basin.

ACKNOWLEDGEMENTS The management of Thiess Bros. Pty. Ltd. (Mining Division), particularly Mr. Ian McCauley, Mine Manager, and Mr. Lance Grimstone, former Mine Geologist, and Mr. Bruce Patrick, current Mine Geologist, are thanked for their co-operation. Fieldwork was undertaken with Dr. Mick Wilson. His unfortunate death in a gliding accident prevented the application of this study to work he was undertaking on the nature of the clay particles in the coal seam and associated sediments in relation to ash precipitation behaviour at the Gladstone Power House and at the Queensland Alumina Refinery. Financial support for the investigation was partly provided by the Julius Kruttschnitt Mineral Research Laboratory, University of Queensland, as part of their Rock Fragmentation in Coal Overburden Project which is funded by AMIRA from NERDDC fund and partly by a NERDDC grant (79/9127) to the Department of Geology, University of New England, Armidale.

REFERENCES

I• Allen, J .R.L., 1965: A review of the origin and characteristics of recent alluvial sediments. Sedimentology, 5, 89-191. Allen, J .R. L. , 1970: Studies in fl uviatile sedimentation: a comparison of fining upwards cyclothems, with special reference to coarsemember composition and interpretation. J.Sedim.Petrol., 40, 298-323. Beaumont, E.A., 1979: Depositional environments of Fort Union Sediments (Tertiary, Northwest Colorado) and their relation to coal. (2), 63 B ull.Amer.A ssoc.Petrol.Geol. , 194-217.

Cameron, J .B., 1972: Callide Coalfield. Rept Geol.Surv.Qd, 68. Dear, J .F., McKellar, R. G., and Tucker, R. M. 1971: Geology of the Monto 1:250 000 sheet area. R ept Geol.Surv.Qd, 46. Etheridge, F.G. and Schumm, S.A., 1978: Reconstructing paleochannel morphologic and flow characteristics:methodology, limitations, and assessment. Mem.Canad.Soc.Petrol. Geol., 5, 703-722 . Gould, R. and Shibaoka, M., 1980: Some aspects of the formation and petrographic features of coal members in Australia, with special reference

to the Tasman Orogenic Zone. Coal Geology, 2(1), 1-30.

Grimstone, L.R. and Roach, C., 1978: Evaluation of the Callide Coalfield. Aus. l .M.M. Conference, North Queensland, 83, 94 . Horne, J.C., Ferm, J .C., Caruccio, F.T. and Baganz, B.P., 1978: Depositional models in coal exploration and mine planning in Appalachian Region. Bull.Amer.Assoc.Petrol.Geol., 62(2), 2379-2411. Howell, D.J. and Ferm, J .C., 1980: Exploration Model for Pennsylvanian Upper Delta Plain Coals, Southwest West Virginia. Bull.Amer.Assoc. Petrol. Geol., 64(6), 938-941. J ackson, R.G., 1978: Preliminary evaluation of lithofacies models for meandering alluvial streams. Mem .Canad.Soc.Petrol.Geol. , 5, 543-576. Leveritt, N., 1978: Dragline operations at the Callide Open Cut Coal Mine. Aus. I.M.M. Conference, North Queensland, 183-188. Miall, A.D. (ed), 1978: Lithofacies types and vertical profile models in braided river deposits; a summary. Mem.Canad.Soc.Petrol.Geol. , 5, 597-604. Reading, H.G. (ed), 1978: Sedimentary Environment and Facies. Blackwell Scientific Publications. Rust, B.R. , 1978: Depositional models for braided alluvium. Mem.Canad.Soc.Petrol.Geol., 5, 605-626.

'.'. 4

Smith, G.L. and Eriksson, K.A., 1979: A flu vioglacial and glaciolacustrine deltaic depositional model for Permo-Carboniferous coals of the northeastern Karoo Basin, South Africa. Palaeog. , Palaeoclim., Palaeoecol., 27, 67-84. Smith, D.G. and Smith, N.D. , 1980: Sedmentation in anastomosed river systems: examples from alluvial valleys near Banff, Alberta. J .Sedim.Petrol. 50(1), 157-164. Smyth, M., 1980: Thick coal members: products of an inflationary environment? Coal Geology, 2(2), 53-76. Stevens, J ., 1981: Palynology of the Callide Basin, east-central Queensland. Pap.Dep.Geol. Univ. Qd. 9(4), 1-35. Svenson, D. and Hayes, S.W. , 1975: Callide Coal Measures, Queensland, in Traves, D.M. and King, D. (eds), Economic Geology of Australia and N ew Guinea, v. 2, Coal. Aus. I.M.M. , Monograph Series No. 6, 283-287 . Walker, R.G. (ed), 1979: Facies Models. Geoscience Canada Reprint Series 1.

P.G. FLOOD, Department of Geology, University of New England, Armidale, N.S.W. 2351.

ORGANIC PETROGRAPHY OF THE RUNDLE-STUART OIL SHALE DEPOSIT, QUEENSLAND By A. C. Hutton

ABSTRACT Rundle and Stuart oil shale deposits are contiguous deposits located in The Narrows Graben, near Gladstone, Queensland. The Tertiary succession has been divided into three formations which comprise in excess of 830 metres of sedimentary rocks. The Worthington Formation, the oldest, contains claystone, the Rundle Formation contains interbedded oil shale (termed lamosite), brown coal, claystone and limestone; the Curlew Formation is a brown coal-dominated sequence. This paper describes the reflected light and fluorescence mode petrography of the lithologies in each formation, with special reference to the liptinite macerals, and discusses the origin of the main oil producing constituents.

INTRODUCTION Oil shale was discovered in The Narrows, 20km northwest of Gladstone in the late nineteenth century (Ball, 1914). During 1918 the first oil prospecting areas were marked (Ball, 1914) and shale was officially sampled from a small shaft in the same year. The Queensland State Government commissioned a drilling programme in 1941 and reserves of 630 x 10 tonnes of shale (at 67 + L/tonne) were calculated. Further exploratory drilling and research was carried out in 1967 by Associated Australian Resources and in 1969 by Carpentaria Exploration Company. A joint venture by Southern Pacific Petroleum NL and Central Pacific Minerals NL commenced detailed drilling in 1974. Two deposits, Rundle and Stuart (Figure 1), were delineated and to date, resources of 2.65 billion barrels and 2.0+ billion barrels, respectively, of shale oil have been calculated (Henstridge and Missen, 1982). Summaries of the regional geology of the area appeared in Kirkegaard et al (1970) and Marshall (1977) and brief summaries of the geology of the area and discussions of

the properties of the oil shale were given by Ball (1914, 1915, 1921, 1946). Geology of the Rundle deposit has been given by Swarbrick (197 4), Lindner and Dixon (1976), Lindner (1980), Henstridge and Missen (1981; 1982), and Henstridge and Coshell (1984). Transmitted light petrography was carried out on Rundle oil shale by Beasley (1945a) but he was unable to determine the nature of the organic matter in the oil shale. Preliminary data (Hutton, 1978) from reflected light and blue-ultraviolet (blue-UV) fluorescence mode microscopy showed that liptinite derived from algal precursors was the dominant form of organic matter in the oil shales. It was evident from that study that a more detailed study was needed to elucidate any downhole and lateral variations in either the type and quantity of organic matter. This paper summarizes a detailed investigation of the Rundle-Stuart deposit (Hutton, 1982) that was carried out in response to the above findings. Data were obtained from both spot core samples and representative crushed samples (taken from 2m half core splits) from twenty five drill holes in the Rundle deposit and fourteen drill holes in the Stuart Deposit.

•

The Narrows Beds (Figure 2) have been subdivided into three formations (Henstridge and Missen, 1981; 1982; Henstridge and Coshell, 1984). Each formation contains a characteristic assemblage of organic matter which is different to the assemblage in either of the other two units . The oil shale-bearing Rundle Formation is characterized by lamosite and carbonaceous lamosite interbedded with brown coal and claystone whereas the underlying Worthington Formation is dominantly claystone that contains only minor vitrodetrinite. The Curlew Formation (the youngest unit) comprises interbedded liptinite-bearing brown coal, carbonaceous shale and claystone.

2fi

and Cypris (Palmieri and Foster, 1980). Beasley (1945b) used ostracode valves to suggest a probable Miocene age but Palmieri (1975) gave no definite age other than Early Tertiary to Late Tertiary. Higher plant microflora is dominantly an angiosperm assemblage with minor gymnosperm and pteridophyte components. Palynological studies have not produced consistent probable ages for the sequence. For example, Paten (1967) suggested an Eocene to Oligocene age where Hekel (1972) suggested a Palaeocene to Middle Oligocene age. Foster (1979) used the occurrence of Anacolosidites sectus to indicate a Late Eocene to Late Oligocene age for younger stratigraphic units. Azolla capricornia sp. nov. was identified from the Curlew Formation (Foster and Harris, 1981) and its occurrence was taken to indicate a Middle to Late Eocene age.

Curtis

Island

SDD-25•

Mt Larcom

OIL SHALE TERMINOLOGY

■

MACERALS

0 ;Mt l ■ a

•

Figure 1: Location of the Rundle and Stuart Oil Shale Deposits.

PALAEONTOLOGY AND AGE Planorbis include Macrofossils gastropods, Chelodina-like tortoise shell, fish bones and scales (Lutjanus, Percalates, and and Epiceratodus) Scleropages crocodilean teeth (Lindner and Dixon, 1976). Ostracode valves are abundant and commonly accumulated to form thin ostracodite beds. Preservation is usually poor but genera recognised include the freshwater forms Cypridopsis, Candona and Virgatocypris as well as Stenocypris

Oil shales contain organic matter derived from algae and terrestrial plants. Maceral terminology as defined by the International Committee for Coal Petrology (I.C.C.P., 1963; 1971; 1975) is adequate when referring to the organic matter derived from terrestrial plants. Consequently, macerals other than algini te are used as defined by the I.C.C.P. (1963; 1971; 1975) and Stach et al. (1975; 1982). However, the I.C.C.P. lists only one maceral, alginite, for organic matter derived from algae. Previous petrographic studies of oil shale (Hutton et al., 1980; Cook et al., 1981; Hutton, 1982a) have shown that organic matter derived from algae can be divided to two groups. and telalginite submacerals, Two lamalginite, were introduced (Hutton, 1982). Telalginite (formerly alginite A of Hutton et al., 1980 and Cook et al., 1981) is spheroidal, ovate or disc-like alginite (derived from large colonial or thick-walled unicellular algae) with generally strong fluorescence at low rank and distinctive external structure, and in many cases internal structure, when viewed in sections perpendicular to bedding. It is derived from large colonial or thick-walled, unicellular algae braunii (in such as Botr_yococcus

I •

CURLEW FORMATION Kerosene Creek Member Telegraph Creek Member Munduran Creek Member Cl)

w m

:r:

CJ>>

a: 0~ a: t-

.• •. RUNDLE FORMATION

•••••

•••• •• •• •• •• •• •

ctW zt-

Brick Kiln Member

•

•••• • •• •

a: a:

Humpy Creek Member

Ramsay Crossing Member Teningie Creek Member Monte Christo Member

WORTHINGTON FORMATION

CURTIS ISLAND GROUP (PALEOZOIC)

(not to scale) Figure 2: Stratigraphic Section, Rundle-Stuart Deposit, The Narrows Graben, Queensland. (after Henstridge and Coshell, 1984)

torbanites), Gloeocapsomorpha prisca (in kuckersite), Tasmanites punctatus (in tasmanite oil shales) and Foerstia (in marinite from Kentucky). Lamalginite (formerly alginite B of Hutton et al. and Cook et al.) is lamellar alginite (derived from small unicellular algae, thinwalled, colonial algae, benthonic algae or planktonic algae) generally with weak to moderate (rarely strong) fluorescence at low rank and a distinctive lamellar form , with little recognisable structure, in sections perpendicular to bedding. Structure can be recognised in some lamalginite if viewed in sections parallel to bedding or in sections, perpendicular to bedding, of some rocks (such as carbonates) where the lamalginite is infilled with fine-grained authigenic mineral matter. LITHOLOGIES In studies of Australian oil shales of Tertiary age, six lithologies (lamosite, carbonaceous lamosite, brown coal, claystone, limestone and carbonaceous shale) are recognised. This terminology is also suitable for oil shales from other countries. The terms can be applied to either handspecimens or to layers within samples. In many samples, which contain two or more lithologies, the boundaries between layers are gradational. The type and abundance of organic matter are the criteria for distinguishing each lithology. Definitions of the six lithologies are: i) lamosite - laminated or massive oil shale composed of at least 5% lamalginite with < 5% vitrinite (including inertinite and liptinite derived from terrestrial plants if either is present); ii) carbonaceous lamosite - · a massive or laminated oil shale with > 5% lamalginite and > 5% vitrinite (including inertinite and liptinite derived from terrestrial plants if either is present); iii) claystone - laminated or massive rock containing < 5% organic matter (but not including limestone); iv) limestone - laminated or massive carbonate rocks containing < 5% organic matter;

vi)

brown coal - massive to laminated rocks containing > 50% organic abundant compnsmg matter vitrinite and/or inertinite with lesser amounts of terrestrial liptinite such as resinite, cutinite, liptodetrinite and very rare sporinite but no alginite; and, massive carbonaceous shale rocks comprising < 5% alginite and > 5% but < 50% vitrinite and/or liptinite derived from terrestrial plants.

TYPES OF OIL SHALE MEMBERS Brown coals from Tertiary oil shale deposits such as the Rundle-Stuart deposit are canneloid coals that contains abundant liptinite such as resinite and cutinite. These coals are oil shales in their own right: many samples have been assayed by Fischer pyrolysis at more than 80 litres per tonne. Because the brown coals are oil shales, it is convenient to define two types of oil shale members in Tertiary deposits: brown oil shale members - these i. are dominated by lamosite and the shale oil is derived mostly from alginite, and carbonaceous oil shale units 11. these are dominated by canneloid brown coal; the shale oil is derived from terrestrial liptinite and vitrinite. ORGANIC PETROGRAPHY OF THE RUNDLE-STUART DEPOSIT The Narrows Beds comprise a Tertiary sequence of lamosite, claystone and coal of brown coal rank with thin interbeds of limestone, siltstone and sandstone. Three formations , the Worthington, Rundle and Curlew Fromations have been described and each formation has a distinctive organic assemblage.

WORTH! GTON FORMATION The Worthington Formation is the oldest of the Tertiary units and is composed of white or mottled (iron-stained) claystone and siltstone. These rocks are barren of liptinite but contain sparse to rare vitrodetrinite.

I 2

RUNDLE FORMATION The Rundle Formation comprises eight members (Figure 2) all of which contain significant resources of lamosite. Kerosene Creek Member has the highest average shale oil yield; the Brick Kiln Member is the thickest. Telegraph Creek Member, the least prospective member, is dominated by claystone but has several oil shale seams that have shale oil yields comparable with those of oil shale seams in other members.

•

'I •

Lithologies Lamosite. In the Rundle-Stuart deposit lamosite is generally yellowish-brown o; olive grey to dark greyish-brown, moderately hard, sectile and laminated or massive. Lamalginite is the dominant maceral (Plates 1 to 4) and it possesses a range of optical properties. It is divided into three types: lamalginite I, lamalginite II and lamalginite III. Lamalginite I (Plates 1 to 6) is the dominant alginite and generally comprises from 10 to 60 percent by volume(% by vol) of the lamosite; however, in one sample (from SDD-34 , 23.4m depth) lamalginite con-

stituted 86% by vol of the rock. Lamalginite I exhibits a variety of morphological forms and fluorescence colours (Table 1). Orange fluorescing lamalginite is dominant over yellow flourescing lamalginite throughout most of the Rundle Formation. Yellow fluorescing lamalginite mostly occurs in samples from the lower part of the sequence (especially the Ramsay Crossing and Teningie Creek Members) and in a few thin layers of the Telegraph Creek Member. P!olonged irradiation of samples, for periods of more than fifteen minutes, with blue-UV light results in an apparent increase in fluorescence intensity and a shift in wavelength towards the blue end of the spectrum (orange fluorescence normally changes to yellow; yellow fluorescence does not change). However, preliminary spectral analysis of the emitted radiation from lamalginite in one sample (Figure 3) showed a ~ec:ease in intensity and no significant sh~ft 1~ wavelength of the emitted energy. It 1s evident that further studies are needed to fully detail changes in lamalginite fluorescence after prolonged irradiation.

Figu re 8: Spectral Anal:,sis of Em itted Rad ia ti on from Lamalgin ite in One Rundle Sample. (Data from Dr H. Pillar) •

60 ~

,i C

c• .• ··

t - time In mlnut'°'

0 ---:-!'::------.---- -- - - - . - - - - - 5?0 600 800 70 0 Wa v el e ngt h (nm)

Tabld . Prooerties of Lamalginite. c., 0

Lamalginite I

Lamalginite II

Lamalginite I II

Precursors

Pediastrum and other planktonic algae

Cleistosphaeridium

Planktonic and benthonic algae of ununknow~ affinity

Algal Form

stelldte disc, 2 to many cells, processes with bifid termination (0.005 to 0.007mm long); grantexture on cell walls; in one species raised margin between cells

thin hollow sphere, numerous fine processes (0.002 to 0.003mm long), commonly hirsute

thin hollorl spheres or thin sheets

orange to yellowishorange or yellow mostly moderate but occasionally weak positive - increase in intensity and wavelength change towards the red end of the spectrum

green to yellowishgreen moderate to strong

yellow to · yellowishgreen strong to very strong

none

0.03 to 0.05mm

0.05 to 0.3mm 0.001 to 0.02mm ( thicker forms composed of several lamellae 0.001 to 0.002mm th i ck)

<10, mostly ~1 (minor component of lamosite)

<5, mostly <.2 (minor component of lamosite)

Fluorescence Colour Intensity Alteration

Size Length Thickness

Abundance ( as ~6 of total rock)

0.03 to 1.2mm <0.0ll5mm

<5 to 86

(dominant crnnponent of lamo~ it e)

< 0.01mm

. Plate 1: Moderately high grade lamosite, from Rundle, containing lamalginite I (1), vitrodetrinite (V) and fragments of ostracode shell (0). Field width = 0.28mm, fluorescence mode, section perpendicular to bedding.

I•

Plate 2: Lamalginite I (1), vitrodetrinite (V) and mineral grains (M) in a moderately high grade lamosite from Stuart. Field width = 0.28mm, fluorescence mode, section perpendicular to bedding.

Plate 3: Lo"· g-rade lamosite with small lamalg-inite I ( ll. lamalg-inite III (3) and small B otryococcu telalg-inite (T). Field "·idth = 0.-1 -lmm. fluorescence mode. section perpendicular to bedding-.

Plate -1: Lamalg-inite I lamella. Field width = 0.07mm. fluorescence mode. section perpendicular to bedding-.

Plate 7: Cleistosphaeridium; precursor of Jamalginite II in Rundle-Stuart lamosite. Field width - 0.06mm, nuorescence mode, strew mount.

Plate 8: Lamalginite m of unknown affinity; Field width = 0.28mm. nuorescence mode. strew mount.

Plate 9: Botryococcus; precursor of telalginite. Field width = 0.28mm, fluorescence mode, strew mount.

Plate 10: Pollen grain; precursor of sporinite. Field width = 0.06mm; flourescence mode, strew mount.

After prolonged irradiation, samples with a brown to dark brown groundmass exhibit fluorescence emanating from the groundmass. Part of this fluorescence may originate from alginite under the surface but the major part probably emanates from fined-grained algal or other organic matter in the mineral groundmass. Lamalginite II (Plate 7) generally occurs in only a small number of samples, and in quantities of much less than 1% by vol. It has much stronger fluorescence intensity than lamalginite I (Table 1). Lamalginite III (Plates 3 and 8) mostly constitutes less than 1 % by vol of samples but in several samples, thin layers of less than 1mm thickness contain up to 5% by vol lamalginite III. Minor organic constituents of lamosite include telalginite (Plate 3) and liptinite derived from terrestrial plants. Telalginite occurs as spheroidal to ovoid colonies < 0.02 to 0.3mm diameter and constitutes < 1 vol % of the bulk rock. It is ubiquitous in most samples but may comprise up to 5% by vol of some thin layers. Botryococcus (Plate 9) is the precursor of this alginite. Fluorescence intensity of telalginite is typically strong and it normally shows negative alteration if subjected to prolonged blue-UV irradiation. The state of preservation varies from excellent (colonies have excellent internal structure) to poor (colonies have little internal structure) in which case identification of the telalginite is based on features such as shape, fluorescence colour, fluorescence intensity and colour in reflected light) rather than structure. Many colonies that are poorly preserved are infilled with, and/or partly replaced by, fine-grained anhedral pyrite and nonfluorescent quartz and/or authigenic clay minerals. The degree of mineralization varies from low (with little change in the optical properties) to severe (where almost total replacement occurs and the alginite has a much reduced fluorescence intensity and a granular texture). The degree of mineral replacement/infilling in any sample may vary from zero to total replacement of the algae. No apparent relationship between the type of lamosite, volume of lamosite, or fluorescence colour of lamosite, and the degree of mineralization has been. noted. Thus neither the occurrence of telalginite, nor the degree of 36

alteration of colonies, is useful as an indicator of lamosite type, probable shale oil yield, or prevailing conditions during deposition. Components of lamosite that are derived from terrestrial plants include: Sporinite - yellow fluorescing liptinite derived from pollen and spores (including microspores and megaspores) of plants other than algae (Plate 10); . Resinite - yellow, yellowish-green or orange fluorescing liptinite _ derived from waxes, fats, oils and lipids that have been exuded by plants; resinite usually occurs in carbonaceous lamosite in association with ulminite or larger vitrodetrinite; Liptodetrinite fragments of other liptinite macerals; Bitumen - weak to moderate orange of yellow fluorescing organic matter derived from the alteration of other organic matter; it occurs as pods and blebs dispersed throughout samples or as infillings in the interstices of mineral aggregates; bitumen is common in samples clo~e to intrusions. Vitrodetrinite - two forms are recognised in lamosite phytoclasts (subequant vitrodetrinite) that are grey or black and opaque and stringers (thin filamentous to lensoidal vitrodetrinite) that are brown and translucent; the phytoclasts are probably derived from fragments of terrestrial plants whereas the brown stringers are possibly derived from fungae or other non-vascular plants. The Rundle-Stuart deposit contains abundant lamosite that is best termed breccia. Henstridge and Missen (1981 : 1982) recognised sedimentary and injection breccias in drill core; the latter were identified by injection structures cutting across bedding planes. Petrographically, three types of breccias are recognised - erosional breccias, peletal breccias and compaction breccias. Table 2 summarizes features of each type of breccia. Carbonaceous Lamosite. This rock is a laminated to massive lamosite containing > 5% by vol vitrinite. Two types are recognised: that composed of interlayered

I e

lamosite and brown coal, and that which is predominantly a lamosite but one that contains a much higher percentage of dispersed vitrinite than is generally found in lamosite. In the former the coal layers contain the same macerals as brown coal, that is, ulminite, corpohuminite, attrinite, densinite and liptinite. In the latter type, vitrodetrinite and rare ulminite are dispersed throughout the sample. Brown Coal. Brown coal is a minor lithology in all members (except the Humpy Creek Member) and is interbedded with lamosite. Boundaries between the two are gradational with the intermediate zone typically a carbonaceous lamosite. In handspecimen the coal is dull black with minor bright laminae up to 5mm thick. The coal is composed of layers of vitrite (which are either ulminite-dominated or have co-dominate ulminite and corpohuminite), clarite, and clarite-related carbominerite. Leaf resinite is the most abundant liptinite in clarite and clarite-related carbominerite. It comprises up to 5% by vol of many layers. Fluorescence colours range from green, greenish-yellow or yellow to orange. Discrete resinite bodies (up to 0.2mm diameter), cutinite, miospores, megaspores and liptodetrinite are minor constituents. Weak brown fluorescing, structureless liptinite occurs as a ground mass in attrinite-rich clarite layers. This non-structured liptinite, which is intimately associated with the attrinite and thus could either be poorly coalified, cellulosic organic matter or bituminite. Weakly fluorescing suberinite with only poorly preserved cellular structure is associated with corpohuminite in several samples. Porigelinite, sometimes with weak yellow to orange fluorescence, is a minor constituent of vitrite and clarite layers. It mostly occurs with corpohuminite. In this association much of the corpohuminite has a porigelinite core and a homogeneous rim of corpohuminite. Most samples have 1 to 2% by vol porigelinite. Limestone and Calcareous Claystone. Limestone and calcareous claystone, commonly grading from one to the other, occur sporadically throughout the sequenc. Limestone contains rare to sparse lamalginite (much less than 1% by vol of

the bulk rock) with lesser amounts of bitumen and liptodetrinite. Botryococcus telalginite is sparse to rare in many limestones; lamalginite III and vitrinite are very rare. Cell structure is well-preserved in telalginite but lamalginite I is usually diffuse with weak fluorescence. Other features in limestone include stylolites (perpendicular, oblique and parallel to bedding) and brecciated textures that are probably formed by differential solution and reprecipitation of carbonate. In "Brecciated" limestone "clasts" are generally angular. Claystone. Organic matter in claystone comprises sparse to rare lamalginite, Botryococcus telalginite, liptodetrinite and bitumen. Lamalginite is most abundant in yellowish-grey claystone. Claystone commonly grades into lamosite. Characteristics of Members The eight members of the Rundle Formation can be divided into three types brown oil shale members, carbonaceous oil shale members and mineral dominated members. Brown oil shale members include the Kerosene Creek Member, Munduran Creek Member, Brick Kiln Member, Ramsay Crossing Member, Teningie Creek Member and the Monte Cristo Member. These members contain abundant lamosite with minor carbonaceous lamosite, brown coal, claystone and limestone. Most of the shale oil formed during pyrolysis is derived from lamalginite. The Humpy Creek Member is a carbonaceous oil shale member. It is composed of abundant brown coal and carbonaceous lamosite with lamosite and minor claystone and limestone. Much of the shale oil formed during pyrolysis of carbonaceous lamosite is derived from terrestrial plant macerals such as resinite, cutinite and vitrinite; a significant proportion of the shale oil is also derived from lamalginite. Thus the composition of the shale oil from any retorted sample would depend on the relative proportions of terrestrial and algal matter in the shale. The Telegraph Creek Member is a mineral dominated member. Claystone is the dominant lithology although several lamosite layers are interbedded with the claystone layers.

c;.:, 00

Table 2 . Characte ristics of Brecc iat ed Lam os ite, Rundle-Stua rt Deposit.

Erosional Breccia Clasts

Groundmass

Other Features

Peletal Breccia

Compaction Breccia

Type

lamosite; carbonaceous lamosite; claystone; carbonate

carbonate; rarely lamosi te and clays tone

claystone; low grade lamosite

Form

angular to rounded; spherical to flattened discs

rounded; spherical to ovoid

sub-angular to rounded; spherical to flattened discs

Size (mm}

< 1 to 50

.:: 1 to 20

,< 1 to 80

Alginite content (% by vol)

0 to 80

0 to 10, rarely 15 to 20

0 to 50

Alginite fluorescence

yellow to orange, moderate to intense

orange, moderate intensity

yellow to yellowish-orange, weak to moderate intensity

Mineral matter

quartz; clay minerals; carbonate

carbonate

quartz; clay minerals; carbonate

Composition

clay minerals; quartz; huminite content Oto 25 vol %; gastropod fragments; ostracods, mostly broken

carbonate; minor clay minerals; huminite absent

no groundmass (see below)

Alginite Content (% by vol)

0 to 40

Alginite

yellow to orange, weak to moderate intensity

orange, moderate intensity

boundaries of layers sharp, flat or undulating, bedding of clasts parallel or discordant to seam bedding, clasts transported some distance before deposition

boundaries of layers sharp or gradational, clasts formed by precipitation of carbonate around a nucleus which may or may not contain lamalginite, little movement of clasts after t~ey are formed

0 to 5, rarely to 10

boundaries of layers sharp or gradational, breccia forms as sediments are dewatered during compaction and a brecciated texture forms; boundaries of "clasts" defined by orientation of alginite

CURLEW FORMATION The Curlew Formation, the youngest formation of The Narrows Beds, comprises interbedded brown coal, carbonaceous shale and claystone. Minor amounts of calcareous sandstone and limestone (up to 0.7mm thick) also occur. Brown coal is liptiniterich, (with resinite dominant in most samples), dull to semibright and with lenses, andJess commonly, thin laminae of bright coal. Dull to semibright layers are composed of clarite or clarite-related carbominerite whereas the bright laminae and lenses are composed of vitrite. Vitrite layers are characterized by three maceral assemblages: dominantly ulminite with a central (a) eu-ulminite core surrounded by a texto-ulminite zone that, in rare layers, has a cutinite rim; corpohuminite (including phloba(b) phinite and pseudophlobaphinite), porigelinite and with minor leaf resinite (always < 5% by vol); and, subequal ulminite and cor(c) pohuminite with < 1 % by vol resinite and liptodetrinite. The three maceral assemblages represent sections of leaves, branches and stems. Ulminite is derived from gelified plant tissue. Cell walls are tightly packed together and, due to homogenizing reactions during coalification, original cell structures are indistinct. Corpohuminite is derived from phlobaphenitic cell excretions and secondary humitic cell infillings that have not undergone gelification (Stach et al., 1975; 1982). Primary corpohuminite, termed phlobaphinite, is derived from wood and root tissue of conifers and the Lauraceae family in which cell excretions are either oxidation or condensation products of phlobaphenes (I.C.C.P., 1971). Phlobaphenes are essentially tannins. Pseudophlobaphinite or secondary corpohuminite occurs as cell infillings derived from humic gels. Brigh laminae generally have microfractures that are perpendicular to bedding. These fractures are probably formed during desiccation after the samples have been polished; most samples develop a polygonal fracture pattern if the surface is left unprotected for two or more days.

Clarite layers have abund.a nt attrinite dens~te (both composed of finely divided h1:11111c matter). ulminite and corpohuminite with rare porigelinite and gelinite. Liptinite constitutes greater than 5% by vol of the_ t?tal organic matter, and comprises leaf resnnte, rounded to ovate resinite bodies suberinite, cutinite, bituminite lip: todetrinite, and very rare Botryo~occus telalginite. The latter is only found towards the base of the formation. Three maceral associations occur in clarite layers: layers containing an inner ulminite (a) core and an outer corpohuminite zone with interstitial leaf resinite and a cutinite rim (liptinite always totals < 5% by vol); ulminite, corpohuminite with abun(b) dant leaf resinite, suberinite, resinite bodies in a groundmass of attrinite , densinite and liptodetrinite; and weakly fluorescing layers compos(c) ed of attrinite, minor ulminite and resini te bodies. Type (a) maceral association probably represents a single stem or leaf whereas types (b) and (c) represent comminuted terrestrial detritus. Crassicutinite and tenuicutinite is found in assemblages (b) ' and (c). Porigelinite occurs in many vitrite and clarite layers and is mostly associated with corpo~uminite. Corpohuminite commonly con1:ams a rounded to irregularly-shaped core of porigelinite with an outer homogeneous rim of corpohuminite. Elsewhere irregular blebs of porigelinite with distinctive red or green internal reflec: tions, a granular texture and weak to moderate, yellow to orange fluorescence ' occurs in ulminite. Clarite layers contain rare sclerotinite including unicellular to hexacellular sclerotia and very rare, thick-walled fungal tissue. . Percentages of macerals vary considerably throughout the formation but a change in either vitrinite or resinite content with depth is not apparent. Vitrinite content in clarite layers ranges from 30% to . 80% by vol. Ulminite is generally dommant over corpohuminite. S_~ch et al._<;975; 1982) divided primary res1mte, (resm1te secreted by plant tissue

during the plant's lifetime) into terpene resinite and lipid resinite. Terpene resinites form as cell secretions and generally assume the shape of cells that, when the cell tissue is destroyed, remain as individual entities. Resinite bodies in Rundle coals are similar to the terpene resinite illustrated by Robert (1979). Approximately 5% by vol of the resinite bodies in RundleStuart coal has numerous gas bubbles that impart a distinctive yellow to pale brown colour in reflected light. Terpene resinite is particularly abundant in Tertiary coals of Europe where conifers are significant precursors. The occurrence of terpene resinite and phlobaphinite may indicate that some of the vitrinite and resinite is derived from conifers or members of the Lauraceae family. Lipid resinite is derived from fats and waxes (Star.h et al., 1975). Both constituents accumulate in plant tissue especially stems. Lipid resinite does not display the cell shape. The leaf resinite in the Rundle-Stuart brown coal is probably lipid resinite that is derived from resin and other lipid-rich substances that are secreted by mesophyll cells of leaves and stems. In Rundle-Stuart coal, leaf resinite is usually interstitial to corpohuminite and less commonly interstitial to ulminite. Suberinite is derived from the suberin layer of cell walls of cork tissue of bark and to a lesser degree roots, stems and fruit. Its presence in thise coals is indicative of significant quantities of terrestrial vascular plants. Many of the clarite and clarite-related carbominerite layers have weak orange to brown fluorescing liptinite (bituminite) which occurs as stringers or as a grountf mass between the huminite macerals. In some ulminite-rich layers it is commonly interlaminated with thin stringer-like ulminite.

REFLECTANCE DATA Mean maximum reflectance (R0 max) values for vitrinite (measured using monochromatic light of 546 nm wavelength in oil immersion at 23 ± 1 °C using glass standards of 0.53, 1.06 and 1.82% reflectance for calibration) in the Curlew Formation and Rundle Formation are summarized in Tables 3 and 4 and Figures 4 to 7.

The average R0 max values for ulminite and corpohuminite in coal from the Curlew Formation are 0.33% and 0.41 % (figure 4). Reflectance values for semifusinite range from 0.40 to 0.66%, those for inertodetrinite range from 0 .36 to 1.18% and those for sclerotinite 0.32 to 0.66% . In most samples from the Rundle Formation vitrodetrinite has a bimodal reflectance distribution related in part to maceral size. For example in one sample from RDD-86 (71.5m), R 0 max for small vitrodetrinite in one lamosite layer was 0.18% (range of 0.14 to 0.23%) and for larger vitrodetrinite in an adjacent carbonaceous lamosite layer, R0 max was 0.31 % (range of 0.27 to 0.39%). Both layers have similar alginite contents. Within the carbonaceous lamosite layer, reflectance for layers and lenses of ulminite ranged from 0.27 to 0.39% (Ramax = 0.33%) and the reflectance for smaller vitrodetrinite ranged from 0.15 to 0.34% (Ramax = 0.20%). Reflectance is also related to type of organic matter. Angular to rounded, equant grey vitrodetrinite has a higher reflectance than thin, lenses of brown, weakly brown fluorescing vitrodetrinite. The former is probably derived from the terrestrial plants whereas the origin of the latter is not known. Likely precursors for the latter include the floating fern Azolla, fungal tissue or nonvascular plants such as moss. Figure 5 shows reflectance values for inertodetrinite in lamosite. The highest values are much higher than those for inertodetrinite in coal of the Curlew Formation. It is possible that in lamosite some of the highly reflecting inertodetrinite, may be secondary inertodetrini te. Reflectance data for ulminite and corpohuminite in samples from GSQ Rockhampton-2 indicate an increase of 0.12% per 100m (Figure 6). Increasing reflectance with depth is also illustrated in Figure 7 where the reflectances of vitrodetrinite in samples from SDD-1 are given.

DISCUSSION LAMALGINITE In most samples the fluorescence of lamalginite I is dominantly orange to yellowish-orange in most samples; a small number of samples have mostly yellow

•

CURLEW FM

■

KEROSENE CREEK MEMBER

...... Ill

Cl)

•

200

TELEGRAPH CREEK MEMBER

Cl)

•

~ C

.r:. ....

Cl)

.r:.

.. .... ......

0. ~

c,, ~

...

Z MUNDURAN CREEK MEMBER HUMPY CREEK MEM. <

■

...

I-

■

a: u. w BRICK KILN MEMBER 0

■

..J

z::::,

"' 400 a:

...

RAMSAY CROSSING MEMBER TENINGIE CREEK MEMBER

■

•

MONTE CHRISTO MEMBER

•

eoo-------- ---------------------0.4 0.2 0.3

0.1

% Reflectance Figure 4: Average Romax Values for Vitrinite in Rundle-Stuart Deposit. (circle - vitrodetrinite, triangle - ulminite, square - corpohuminite; bar gives the range of values for each average)

Table 3 . Summary of Ref lecta nc e Data for th e Rundl e Formation, Rundle-Stuart Deposit.

Kerosene Creek Seam

Average Reflectance Data (~)I Lamosite 0.24 Mean Rvmax 0,22 to 0.31 Range

Telegraph Creek Seam

Munduran Creek Seam

Humpy Creek Seam

Brick Kiln Seam

Ramsay Crossing Seam

Teningie Creek Seam

0,24 0.21 to 0.26

0,26 0.23 to 0.30

0.21 0, 18 to 0.31

0.25 0,21 to 0.31

0,24 0.21 to 0,31

0,24 0,22 to 0.28

Brown Coal Rumax

Mean

0.36*

0.34 0.31 to 0,37

0,34*

0,42*

0.36*

Rcmax

Mean Range

0,40*

0,40 0,39 to 0.42

0,43*

0,39* 0,36 to 0,41

0.46*

Rvmax

Mean

Carbonaceous Lamosite Mean Rumax Range Rcmax Rvmax

Mean Range Mean Range

0,32* 0.33 0,30 to 0.35

0.30 0.22 to 0,36

0.24*

0. 31 0,28 to 0.35 0,38 0.35 to 0,45

0,37 0,28 to 0.43 0.30 0.23 to 0,38

0.35 0.3 ~ to 0,37 0,27 0,21 to 0,32

0.28 0.22 to 0,33 0,33

0,35 0,29 to 0,41

Excludes data from RDD-45 1 or 2 values only

u = ulminite = corpohuminite V = vitrodetri nite

0,33* 0,26*

Tabl e 4 . Summ ary of Refl ec t ance Dat a of the Curl ew Form ation .

Mean of R0 max Values ( ~o)

------- Range of------R0 max Standard Values (%) Deviations

Number of Samples

Hole

Maceral

RDD-48

ulminite corpohuminite

0.32 0.42

0.29-0.34 0.38-0.44

0.02-0.03 0.02-0.05

14 15

RDD-66

ulminite* corpohuminite* ulminite corpohuminite

0.26 0.31 0.33 0.38

0.20-0.35 0.24-0.39 0.32-0.36 0.35-0.41

0.03-0.07 0.02-0.07 0.03-0.05 0.02-0.05

29 27 8 8

RDD-78

ulminite corpohuminite

0.36 0.43

RDD-94

ulminite corpohuminite

0.31 0.39

Data from Hutton ( 1978 )

1 1 0.20-0.35 0.30-0.48

0.03-0.04 0.02-0.08

8 9

fluorescing lamalgini te. Where fluorescence colour is predominantly orange , moderately strong orange fluorescence is associated with large, discrete lamellae and weaker, pale orange or yellowish-orange fluorescence is emitted from shorter and thinner lamellae. Thus both intensity and colour are related, in part, to lamellar size and thickness. Some samples containing large lamellae have strong, bright orange fluorescence with weaker pale orange or yellowish-orange extremities. Several low grade samples have small, weakly pale orange fluorescing lamalginite with smooth, thin walls. This smaller lamalginite (normally 0.02mm long) has similar properties to the smaller lamalginite in samples from Condor and Byfield deposits (Hutton, 1982). Significant morphological differences between the smaller lamalginite I and the dominant lamalginite I indicate different precursors (Table 1). Yellow fluorescing lamalginite is much less abundant than orange to yellowishorange fluorescing lamalginite. It has similar form to orange fluorescing lamalginite but is usually more diffuse and lamellae are commonly longer (up to 1mm long). Fluorescence intensity of yellow fluorescing lamalginite is generally weak to very weak.

The difference in fluorescence colour may be related to differences in: (a) precursor type; (b) coalification path; (c) alteration during diagenesis; (d) mineralogy of the respective samples; (e) differences in preservation; or (f) differences in the degree of biological or mechanical degradation. Of these, (a), (b) and (c) are considered to be the most likely causes of the different fluorescence characteristics. Lamalginite III and Botryococcus telalginite occur in most samples that have orange to yellowish-orange fluorescing lamalginite. Abundance of both is variable within each seam. Lamalginite ill is most abundant in the Munduran Creek, Ramsay Crossing and Teningie Creek members but rarely exceeds 2% by vol in any sample. However, in a few samples, Botryococcus and lamalginite ill are locally abundant in some laminae; each may constitute up to 5% by vol of these laminae. Despite these restricted local concentrations neither alginite contribute significantly to shale oil production.

.•"' C

:;;

Reflectance

Figure 5: Reflectance Values for Inertodetrinite in Lamosite. Rundle-Stuart deposit. (272 data points)

Telegraph Creek Member

- - Munduran Creek Member

- - - - -

Humpy Creek Mem.

• •

0 ~

• Corpohumlnlte

Brick KIin Member

•

200

II.

•

- - - - -

Ramsay CroHlng Member

- - - - Tenlngie Creek Member

-...

..I

- -

::, a;

Monte Christo Member

I I

•• • • I

Q. t)

Ulmlnlte

400

• ·1 I

•

I ■

•

•I • z

0 z ... 0

i ~

600

i: ,c ... :I a;

• 0

II.

0·6

0 ·2

0·8

% Reflectance Figure 6: Reflectance Data for Drillhole GSQ Rockhampton-2 .

KEROSENE CREEK MEMBER

• •• • •

• TELEGRAPH CREEK MEMBER

•

100

.•-• • •

MUNDURAN CREEK MEMBER

•

.-·. • _._• • • • • •· ' •• -

HUMPY CREEK MEMBER

200

•• •• • •

Q. 4)

•

• BRICK KILN MEMBER

•••

300

•

•••

•

MONTE CHRISTO MEMBER

•

• • •

•

TENINGIE CREEK MEMBER

-?-

•

·-

• ••

•

• •

•

• 4000

0.1

0 .2

0 .3

% Reflectance Figure 7: Reflectance Data for Vitrodetrin ite in La mosite, Drillhole SDD-1, Stuar t . (75 data points)

0 .4

0 .5

VERTICAL VARIATION IN ORGANIC MATTER in the type and abunvariation Vertical dance of organic matter in the RundleStuart succession occurs on three scales: large-scale facies changes which (a) are characterised by changes in seam type; for example, changes f~om lacustrine to peat swamp env1roi:iments of deposition; medium-scale changes in alginite (b) content, within the oil shale members; these changes are easily observed in hand-specimen and are marked by the alternation of rock types (for example, the alternation of claystone-siltstone-limeston e and lamosite, including carbonaceous lamosite, layers; layers range in thickness from one centimetre to several metres; and small-scale changes in alginite con(c) tent (marked by laminae up to several millimetres thick) that can only be detected microscopically. Three major facies changes can be recognised in the Rundle-Stuart succession. The first is at the base of Monte Christo Member where the stratigraphically lowest occurrence of alginite occurs. Henstridge and Missen (1982) suggested the Worthington Formation was deposited on an outwash alluvial fan. In such an environment the rate of elastic sedimentation would be too rapid to permit the growth of algal blooms and any organic matter settling to the bottom would be rapidly consumed by benthonic organisms or oxidized. Another facies change occurs at the base of the Humpy Creek Member where the volume of terrestrial plant matter increases sign~ficantly. This results in a higher proportion of carbonaceous lamosite and brown coal than in other members. Lamosite of the Humpy Creek Member also contains higher percentages of ulminite vitrodetrinite., leaf resinite and sporinite: although the mcrease in the latter two components does not show up in the point count data (Hutton, 1982). Much of the vitrinite ~s allochthono1;1s and was probably brought mto the basm from marginal, heavily vegetated stream banks and lake margins. The third major change is at the base of the Curlew Formation where there is a

marked and rapid change from lamosite to brown coal. Whereas in the Humpy Creek seam, abundant lamosite layers and lenses are interbedded with vitrite,clarite and clariti~ carbominerite layers, only rare lamos1te layers occur in the Curlew Formation, and then, only in the lower six metres. This facies change probably resulted from decreased water levels within the lake (possibly climatically and/or tectonically controlled) and consequently the development of peat swamps. S_m:ill-scale and medium-scale changes in algm1te content reflect changes in the volume of organic matter that accum1:lated, and was preserved, with the elastic and authigenic mineral matter. Alginite productivity is probably the dominant factor controlling these changes. Medium-scale changes in the volume of alginite are manifested in the alternation of rock types. Thus within any high grade seam claystone, limestone, coal, carbonaceous lamosite and lamosite may occur. Layers of each lithology may be as thin as 10mm. Figures 8 and 9 show the variation in alginite content for spot core samples from drillholes SDD-25 (Stuart) and RDD-66 (Rundle). Mean seam grade is a function of both the relative proportion of each lithology, and the relative proportion of alginite in each lithology: Small-scale variations in alginite content are manifested in the interlamination of alginite-rich and alginite-poor laminae within samples that show little variation in either hand-specimen colour or texture. Laminae may be as thin as 0.25mm and boundaries between laminae are either gradational or sharp.

RDD66 'I,

ORGANIC

MATTEH

LTOM

'I,

OR GAN IC M ATTE R

r --.-- -ro -~~ 200 0 , - - - ~ - - ~ - - ~ - - ~------. '00

-------- --- ---------· ----------------·

..................... --------· -- ......

iinii-.;,~~=-

3J"''

0 0

....z "' "' ~ a:

Cl)

cl' Oy

385 . 1 m

Figure 8: Distribution of Organic Matter in Drillhole RDD-66, Rundle. (Samples on the right hand side are spot core samples, samples on the left hand side were taken at 10cm spacing over 2m intervals. LTOM Fischer assay in litres per tonne at 0% moisture; dotted line - % terrestrial liptinite in coal; dashed line - % vitrinite in coal; continuous line - % lamalginite in lamosite)

SDD-25 LTOM

2·0

1· 3

% ORGANIC

MATTER

o__________~so

200

1/J " 1/J

:I: Q. c(

(!) 1/J ...J

1/J

f--

216-01===~=-=

...... -

- -

. E=

234·01=---~-~-= - . - .- . _... _. -

244·01- -

~~;~i_[I

2700KX~~'.--l i

272· 01 ~=-~ ~~~~

276-01 ;- ; --{ •

...

-... -

Figure 9: Distribution of Organic Matter in Drillhole SDD-25, Rundle. Samples were taken at 10cm spacing over 2m intervals. (SG - specific gravity; other symbols as in Fig. 8.) 49

Minor variations occur in the type of lamalginite in each seam. These variations in lamalginite type are of two kinds variations in the fluorescence colour of the dominant lamalginite I (that is, yellow or yellowish-orange to orange fluorescence) and variations in the type (and quantity) of minor constituents. A relationship betweend depth and either fluorescence intensity or fluorescence colour of lamalginite is not evident. Minor organic components show the following variations with depth: (a) greater abundance of Cleistosphaeridium in the Teningie Creek, Ramsay Crossing and Brick Kiln members and its rare occurrence in the Kerosene Creek Member; (b) increased occurrence of bitumen in the coaly layers, especially in the seams below the Humpy Creek Member; (c) lower abundance of Botryococcus in low grade lamosite and claystone of the Telegraph Creek Member; (d) decreased abundance of vitrodetrinite in the Telegraph Creek Member. The combined volume of minor components such as Botryococcus telalginite, lamalgini te II derived from Cleistosphaeridium and lamalginite III is usually less than 1% by vol in most samples. Larger volumes (up to 10% for Cleistosphaeridium and 5% by vol for others) or these components occur in thin layers throughout the sequence. Such variations are not systematic and are probably related to local variations in conditions at the time of deposition.

SUMMARY The Rundle-Stuart oil shale deposit contains in excess of 830 metres of interbedded oil shale (termed lamosite), brown coal, claystone and limestone. Three formations, the Worthington Formation, the Rundle Formation and the Curlew Formation, have been recognised by Henstridge and Missen (1981; 1982). Each formation has an organic assemblage that is different to the organic assemblage in the other units.

Using the type and abundance of organic matter, three types of members are recognised within the Rundle Formation. Brown oil shale members, such as the Kerosene Creek Member, contain dominantly lamosite with minor brown coal, claystone and limestone. The Humpy Creek Member is a carbonaceous oil shale member and contains co-dominant brown coal, carbonaceous lamosite and lamosite. The Telegraph Creek Member is a mineral dominated member and is composed mostly of claystone with minor lamosite. Rundle-Stuart lamosite contains three types of lamalginite. The dominant lamalginite is termed lamalginite I; it has yellowish-orange to orange, or less commonly, yellow fluorescence. Lamalginite with orange to yellowish-orange fluorescence is derived mostly from the colonial green algaPediastrum. Lamalginite II has green fluorescence and is derived from the acritarch Cleistosphaeridium . Lamalginite III has more intense green to yellow fluorescence than the previous types; it varies from single lamellar, thickwalled forms to composite lamellar forms with up to ten lamellae. Its precursors are of unknown origin. Minor constituents of lamosite include telalginite derived from Botryococcus, vitrodetrinite and liptinite derived from terrestrial plants. Vertical variation in the type and abundance of organic matter occurs on three scales; i. large-scale variations such as abundant brown coal in the Curlew Formation and Humpy Creek Member, abundant lamosite in the Rundle Formation, the restricted development of lamosite in the Telegraph Creek Formation and the lack of liptinite in the Worthington Formation; large-scale variations represent changes in facies type and are manifested as changes in the type and abundance of liptinite and vitrinite. ii. medium-scale (hand-specimen) variations; and iii. small-scale (microscopic) variations. The medium-scale and small-scale changes are manifested as changes in rock type within the formations. These variations represent changes in the rates of growth of algae and changes in the rates of accummulation and preservation of organic matter.

The brown coal of the Curlew Formation is generally dull to semibright with thin bright laminae. Dull coal layers are composed of clarite and clarite-related carbominerite whereas bright laminae are

composed of vitrite in which ulminite is the major component. Vitrite reflectance data indicate that the deposit has a rank equivalent to that of soft brown coal.

REFERENCES

Ball, L.C., 1914: Tertiary oil shales of The Narrows, Port Curtis district. Queensland Government Mining Journal, 15, 73-76. Ball, L.C., 1915: Oil shales of the Port Curtis District. Geological Survey of Queensland Publication 249. Ball, L.C., 1921: Oil shale, Curtis Island. Queensland Government Mining Journal, 22, 186. Ball, L.C. 1946: Oil shales of The Narrows, Central Queensland. Queensland Government Mining Journal, 47, 176-179. Beasley, A.W., 1945a: The petrology of some Queensland oil shales. Memoirs Queensland Museum, 12, 124-133. Beasley, A.W., 1945b: Ostracods from some Queensland Tertiary basins, their bearing on the correlation of strata. Proceedings Royal Society Queensland, 56, 95-124 . Cook, A.C., Hutton, A.C. and Sherwood, N.R., 1981: Classification of oil shales. Bulletin des Centres de Recherche Exploration-Production ElfAquitaine, 5, 353-381. Foster, C.B., 1979: Tertiary miospores from the Rundle Oil Shale Deposit. Queensland Government Mining Journal, 80, 442-443. Foster, C.B. and Harris, W.K., 1981: Azolla capricornis, sp. nov. First Tertiary record of Azolla Lamarck (Salviniaceae) in Australia. Transactions Royal Society of South Australia, 105, 195-204. Hekel, H., 1972: Pollen and spore assemblages from Queensland Tertiary sediments. Geological Survey of Queensland Publication 353, Palaeontological Paper, 30. Henstridge, D.A. and Missen, D.D., 1981: The geology of The Narrows Graben near Gladston, Queensland, Australia. 51st ANZAAS Congress, Brisbane. Henstridge, D.A. and Missen, D.D., 1982: Geology of oil shale deposits within The Narrows Graben, Queensland, Australia, American Association of Petroleum Geologists Bulletin, 66, 719-731. Henstridge, D.A., and Coshell, L., 1984: Definitions of the eight members of the Rundle Formation, The Narrows Graben, Central Queensland. Queensland Government Mining Journal, 83, 94-99. Hutton, A.C., 1978: Aspects of the petrology of the Rundle oil shale The Narrows, Queensland. University of Wollongong BSc (Hons) thesis, (unpublished). Hutton, A.C., 1982: Organic Petrology of Oil Shales. University of Wollongong, PhD thesis, (unpublished). Hutton, A.C., Kantsler, A.J., Cook, A.C. and McKirdy, D.M., 1980: Organic matter in oil shales. Australian Petroleum Exploration Association Journal, 20, 44-67. International Committee for Coal Petrology, 1963:

International Handbook of Coal Petrology. 2nd edition, Centre National de la Recherche Scientifique, Paris. International Committee for Coal Petrology, 1971 : International Handbook of Coal Petrology. Supplement, 2nd edition, Centre National de la Recherche Scientifique, Paris. International Committee for Coal Petrology, 1975: International Handbook of Coal Petrology. 2nd Supplement, 2nd edition, Centre National de la Recherche Scientifique, Paris. Kirkegaard, A.G., Shaw, R.D. and Murray; C.G., 1970: Geology of the Rockhampton and Port Clinton 1:250 000 sheet areas. Geological Survey of Queensland Report, 38. Lindner, A.W., 1980: Rundle oil shale deposit Synopsis of geology; in Hoffman, G.W., (ed.), Field Conference Mount Morgan Rundle Range Yeppoon area. Geological Society of Australia, Queensland Division Field Conference 1980, 72-78. Lindner, A.W. and Dixon, D.A., 1976: Some aspects of the Geology of the Rundle oil shale deposit, Queensland. Australian Petroleum Exploration Association Journal, 12, 165-172. Marshall, J.F., 1977: Marine geology of the Capricorn Channel areas. Bureau of Mineral Resources Bulletin, 163. Palmieri, V., 1975: Micropalaeontological report on core samples from boreholes drilled at The Narrows. (unpublished report for Southern Pacific Petroleum NL). Palmieri, V., and Foster, C.B., 1980: A palaeoecological appraisal for The Narrows Beds, Rundle oil shale, Eastern Queensland. (unpublished memorandum). Paten, R.J., 1967: Memorandum to Carpentaria Exploration Manager on the age of the oil shale at The Narrows. (unpublished). Robert, P., 1979: Classification des matieres organique et fluorescence application aux roches-meres petrolieres. Bulletin du Centre du Recherches Exploration-Production, 3, 223-263. Stach, E., Mackowski, M.-Th., Teichmuller, M., Taylor, G.F. , Chandra, G. and Teichmuller, R., 1975: Stach's Textbook of Coal Petrology. 2nd edition, Gerbruber Borntraeger, Berlin. Stach, E., Mackowski, M.-Th., Teichmuller, M. , Taylor, G.F., Chandra, G. and Teichmuller, R., 1982: Stach's Textbook of Coal Petrology. 3rd edition. Gebruder Borntraeger, Berlin. Swarbrick, C.F.J. , 1974: Oil shale resources of Queensland. Geological Survey Queensland Publication, 83. 1-84.

A. C. Hutton Department of Geology The University of Wollongong P.O. Box 1144 WOLLONGONG, N.S.W. 2500

LATROBE VALLEY BROWN COALS-THEIR GEOMETRY AND FACIES EQUIVALENTS AS A GUIDE TO DEPOSITIONAL ENVIRONMENT By G. R. Holdgate

•

ABSTRACT Some of the thickest and most continuous brown coal seams in the world occur in the Latrobe Valley, Victoria. A series of coal isopachs has been prepared from borehole data to show the three-dimensional geometry of the seams and their subsurface extent at the time of deposition. The four main seams in stratigraphic order are the Morwell 2, Morwell lB, Morwell lA and Yallourn seams. In the Latrobe Valley the average surface area of each seam is nearly 600km"with about 20% of each ar~a containing coal greater than 80m m thickness. In general the major coal depocentres occur to the south of the Latrobe River and west of a marine interface with the Gippsland Limestone. The depocentres are not generally spatially coincident, but sh~w relative shifts in time, probably due to differential compaction. Facies equivalents to the major coal seams comprise kaolinitic clays with sa~ds predominating closer toward~ the ?Iarme interface. The clays occur m thick sequences, and are interpreted as being deposited in lakes which effectively surrounded the peat swamps and protected them from the more destructive fluvial inputs from higher ground beyond the Latrobe Valley. As a result the peat swamps became stabilised for long peri~ds of time and have produced extremely thick brown coal seams containing under 5% ash. Nutrients were brought into the coal swamps via lateral diffusion from the lakes and by rising groundwater from the underlying aquifer systems of the Traralgon Formation.

INTRODUCTION The location of the Latrobe Valley in Victoria is shown on Figure 1 and the major

structures and cross section locations are shown on Figure 2. The main descriptive works on the stratigraphy and coal geology of the Latrobe Valley Depression can be found in Thomas and Baragwanath (1949, 1950, 1951) and Gloe (1960, 1976, 1980). ~~t detailed borehole knowledge on spec1f1c coalfield areas was available for these reports, knowledge of the deeper stratigraphy in the synclinal areas throughout the Latrobe Valley was sparse. Commencing in 1976, the deeper areas of the Latrobe Valley Depression were investigated by the State Electricity Commission of Victoria (SECV) and Department of Minerals and Energy, Victoria (DMEV) for groundwater and stratigraphic purposes. Some of the DMEV results have been described by Thompson (1980) and also in joint SECV-DMEV publications (Gloe et al 1979 and Brumley et al 1981). Additional stratigraphic information is now available from this series of bores, which were drilled to basement and which now cover much of the deeper Latrobe Valley areas on spacings between 5 and 10km. In addition, more detailed drilling has been undertaken by the SECV in the Traralgon Syncline between the Morwell and Loy Yang coal fields for the purpose_of stratigraphic, groundwater and coal quality investigations. From this new information, together with all previous SECV borehole data, isopachs were produced of all main ~oal seams in the Latrobe Valley Depress10n. This extended the work previously undertaken by the SECV in specific coalfield areas only. From the isopachs and crosssections, the geometry of the main_ coal seams were determined, the mterrelationship to the non-coal sediment equivalents was affected, and the latter related to the depositional setting of the coals.

SOU TH

,~ /

GI PP SL ANO

Q<,;- /

HI GH L A OS

'/ (

• YARRAH

Figure 1: Latrobe Valley Depression locality map.

<I)

SUBCROP OF VALLOURN

{=I+

EDGE OF MAR IN E LIMESTONE

&, MOR WELL SEAM S

77). BASEMENT

Figure 2: Major structures and cross section lines in the Latrobe Valley.

METHODS To produce isopachs for all main coal seams in the Latrobe Valley Depression as shown in Figures 3 to 6, the following methods were used: a ALL SECV standard borehole crosssections at a vertical scale of 1:5000 were examined and the coal seam intersections identified under the four main seam categories of Yallourn, Morwell lA, Morwell lB and Morwell 2 Seams. In cases where the Morwell 1 Seam (a combined MlA and MlB seam) or the Latrobe Seam (a combined MlB and M2 seam) was represented, the constituent parts were allotted to the MlA, MlB or M2 categories. Parts of the Traralgon Seam also occur within the Latrobe Valley but were not studied in this survey as most of the seam lies outside the confines of the Latrobe Valley area. b

• •

Thicknesses were measured deleting splits and seam endings less th~ 3m. Inferior coals (10-30% ash) were mcluded in the isopachs where they occur within a seam but in most cases these would constitute less than 7% of the seam thickness. Where parts of the seams have been eroded by the late Tertiary Haunted Hill Gravel Formation, the isopachs were interpolated across the eroded areas from uneroded bore data in the near vicinity. To accentuate the facies relationships, the borehole logs were projected onto a base line approximating to the base of the Yallourn Seam. By this method the late Tertiary folding of the coal seams was removed and the seams restored to their original flat lying disposition.

GEOMETRY OF LATROBE VALLEY COALS Figures 3 to 6 show the isopachs for Yallourn, Morwell lA, lB and _2 in the Latrobe Valley in their pre-eros10n state, which are discussed in stratigraphic order as follows:

Morwell 2 Seam (Figure 3) The major site of accumulation for the Morwell 2 seams occurs in the vicinity of the Latrobe River extending north-east from the Yallourn Open Cut area where massive thickening up to 140m occurs. Near Glengarry it decreases rapidly. An additional Morwell 2 accumulation of lesser overall coal thickness occurs in the Loy Yang/Flynn area where up to three seam splits (the Morwell 2A, 2B and 2C), when aggregated, form 80 to 100m of coal. . These two major sites of accumulat10n are separated by a coal-poor sediment interval which trends north-easterly along the Traralgon Syncline. The Morwell 2 seam is also largely absent in the Yinnar area and to the north and north-east of Traralgon township. A small additional accumul~tion occurs in the Rosedale to Flynn Railway area and probably in the Gormandale area, although in the latter case the seams have largely been removed by erosion. The total area for uneroded Morwell 2 Seam amounts to approximately 528 sq km with a further 58 sq km of eroded and projected seams. Hence, the total original depositional area for the Morwell 2 seams may have amounted to some 586 sq ~ of which about 16% (95 sq km) contams a single seam or an aggregate of splits ,which total more than 80m in thickness. b

Morwell lB Seam (Figure 4) A major shift of depocentres occurred subsequent to the Morwell 2 depo_sitional period. The Morwell lB seam mamly accumulated to the south of the Latrobe River with the n;iajor portion centred around the Loy Yang/Flynn area. Lesser centres_occur in the Morwell to Yinnar area and m the Flynn Railway area. In the north, the 20m isopach broadly coincides with the Latrobe River and only minor Morwell lB seam splits extend into the northern areas of the Latrobe Valley Depression. No significant coal seams extend east of the township of Rosedale. To the south of the Loy Yang/Flynn and Yinnar areas, the seam rapidly reduced in thickness. The total area for uneroded Morwell lB seam amounts to some 596 sq km with a further 101 sq km of eroded and projected coal. Total original depositional area for the Morwell lB seam could have amounted to

c;, Ol S794000.1N

-- --

---- "---------- --- --

LEGEND

r - - - , 20·1.0

r - - 7 1.0 · 60 L___j

L___j

,.i 7 1

B ] ,o- eo -

100 · 120

12011,0

11.0,

C ) ao - 100 CONTOUR INTERVAL

Figure 3: Isopachs of total Morwell 2 seam (preerosional).

•

I N M[ TRl S

697 sq km of which about 28% (197 sq km) contains a single seam or seam split aggregates totalling more than 80m in thickness. A maximum seam thickness of greater than 120m occurs in the Traralgon Creek area and in the Flynns Creek Syncline. c

Morwell lA Seam (Figure 5) Both the depositional area and the overall seam thickness of the Morwell lA seam is considerably less than for the Morwell lB seam. However, the major centres remained essentially the same, being located around the Morwell, Loy Yang/Flynn and Flynn Railway areas. The major contractions when compared with the Morwell lB seam are obvious in the Yallourn/Maryvale, Churchill/Yinnar, and South Flynns Creek Syncline areas where thick intervals of sediments, mainly clays, occur rather than coal seams. Maximum thicknesses of coal in the main centres generally do not exceed 80m. Total area for uneroded Morwell lA seam is 465 sq km (c.f. Morwell lB at 596 sq km) with an additional eroded and projected area of 85 sq km. This gives a total area of deposition for Morwell lA seam of 550 sq km of which only 7% (42 sq km) exceeds 80m in aggregate seam thickness.

Yallourn Seam (Figure 6)

A recurrence of peat accumulation at the earlier Morwell 2 seam depositional site occurred during deposition of the Yallourn Seam. The major depocentre occurs in the Yallourn/Maryvale area, decreasing northeasterly towards the Glengarry township and southwards towards Churchill and Yinnar. The Yallourn Seam appears to have been less well .developed in the Loy Yang area, although the greater extent of erosion has removed most of the evidence. Nevertheless indications are that, due to splitt ing, the Yallourn Seam was poorly developed over the Loy Yang Dome. West of Yallourn, the Yallourn Seam may well have extended across the Haunted Hills where it is now absent, as a seam of similar age occurs in the Moe Swamp Basin (Holdgate, 1985). Maximum uneroded thicknesses of 100m occur in the Yallourn Seam just to the east of the Morwell Monocline. Total area for uneroded Yallourn Seam is 421 sq km, and 144 sq km for eroded and projected coal, but not including any presumed extensions west of Yallourn into the Moe Swamp Basin. Hence, total area for Yallourn Seam in the Latrobe Valley Depression amounted to some 565 sq km of which some 16% (94 sq km) exceeded 80m in aggregate seam thickness.

TABLE 1: AREAS OF COAL SEAMS IN THE LATROBE VALLEY DEPRESSION IN SQ KM

Seam Name

Coal in situ

Eroded Coal (subcrop)

Projected Coal (beyond subcrop)

Total All Coal

Area Of Depocentre* and % in Brackets

Yallourn Morwell lA Morwell lB Morwell 2

420.8 464.9 596.1 528.3

126.1 32.1 70.2 12.4

18.4 53.4 31.2 45 .7

565.3 550.4 697.4 586.4

94.1 (16.6) 42 .0 (7.6) 197 .3 (28.3) 94.6 (16.1)

* Seam thickness exceeds 80.0 M. Table 1 shows the areas for the four major coal seams, and that when compared, the total all coal for Yallourn, Morwell lA and Morwell 2 areas are very similar in size but the Morwell lB is considerably larger. The shifting of the major coal depocen-

tres with time is best illustrated if the areas with greater than 80 m of coal are overlain (Figure 7). This indicates that within the Latrobe Valley Depression, only rarely do more than two of the five major centres coincide in any one area, with exceptions being the Loy Yang Dome and at Morwell. 57

<:11 CX>

5794000 111 N

Ul rngorry

IZI

CAHK AA[A

rz,J Gormandolt

0 LEGEND L J 20·40

□ l.0·60

,00110

120,

~ 60 • 80 C J eo - 100 COt-.TOUA lfll[RVAL IN METRES

Figure 4: Isopachs of total Morwell lB seam (preerosional).

S79f.OOOmN

--- "

....----.. ,/

"''"'"'' I /I / ezi

I I

/ I

LEGEND

ll:'.J

Juralang Junc tion

C J 20·40

,oo .

□ 4060

EEIIm eo 60 -

C J 110 -100 CONTOUR INTERVAL IN METRES

°'

Figure 5: Isopachs of total Morwell lA seam (preerosional).

S1'1.000111N

-..,.__

HO(

SWAMP

_,,,. ;.

.......... .

--:

,.,,,.. ,

\ \

LOY YANG

OP[N CUT

FLYNN AREA

.._ _______ _;

t \\0 s1oll

THE SE SC ALE 1

8KM

Gor11ondole

rlll

LEGEND r = J o-20

r = J 20-"°

L J ao -10o

......"' C0NTOl.fl lNTEAVAl IN MElAfS

111

}ouTCROP EOGE Of' BASI N

Figure 6: Isopachs of total Yallourn seam (preerosional).

,0 - 110

"'° '

FACIES EQUIVALENTS TO LATROBE VALLEY BROWN COALS The isopachs of the Yallourn and Morwell lA, lB and 2 coal seams indicate that for each of these four seams there occurs on average a maximum coal thickness of between 80 and 100m. This consistency in stratal thickness appears to continue beyond those areas of maximum coal deposition. As a result, areas where the coal seams thin and split still retain a representative interval if interbedded coals and sediment, the combined total of which still aggregates 80 to 100m in thickness. This relationship can be illustrated when borehole cross-sections are projected onto a flat baseline placed within the coal seam sequence so as to r_emove the late Tertiary folding (Figure 8). The interface between the Yallourn and Morwell lA seams was chosen as the most convenient base line because it is one of the more regionally consistent breaks between these two main seams, it is usually present in most areas of the Latrobe Valley Depression, and it usually correlates to a significant (up to 5%) in moisture content and density. On Figure 8 the Haunted Hill Gravel Formation (Pliocene unconformity) is shown to truncate the Yallourn and Morwell lA Seams on structural high areas such as the Morwell Open Cut area. Where all seams remain intact and thick non-coal intervals occur above the Yallourn Seam, these relate to present synclinal areas e.g. the Traralgon Syncline. Thick post-Yallourn but prePliocene unconformity sediments relate in age to the Yallourn seam (Bolger, 1984a) but essentially do not contain any coal. In most directions to the north, west and south, the main coal depocentres for the Yallourn and Morwell seams predominantly grade laterally into interbedded clay/coal sequences which then grade to mainly clay sequences. In some cases the Morwell 2 seams grade laterally into interbedded clays and volcanics. Sands tend to be a minor constituent except in localised areas, or in the sediments underlying the Morwell 2 Seam. However, in an easterly direction down the centre of the Latrobe Valley, as shown on Figure 9, sands become increasingly dominant as a lateral facies equivalent to the coals. Further east they become the major sediment component at the interface between the marine Gipp-

sland Limestone and the Yallourn and Morwell Formations. Figure 8 illustrates typical borehole examples of the facies transition between the thick coal areas at Morwell and the coalpoor areas at Churchill and Yallourn North. Here the clays which are the major sedimainly comprise component ment kaolinite~. Towards the coal seams they darken m colour with increased carbonaceous content becoming ligneous clays (up to 70% carbonaceous content). Away from_ the coals the clays lighten in colour grading from greys and browns to light fawn greys and whites. In the more distal areas yellowish disseminations occur within the clays. These are high in iron and contain disseminated siderite. The facies changes were complex and facies shifted laterally with time. In consequence ~oals, ligneous clays and grey to fa~n/white clays occur as vertically alternatmg sequences within any single borehole. . One major expansion of areas receiving mterbedded coal/clay and clay sediments occurred during the period of deposition of the Morwell lA Seam effectively constricting the major coal depositional sites compared to the underlying Morwell lB seam as shown in Figure 8. . The goegraphical distribution (Figure 5) illustrates that during the time of the deposition of Morwell lA coal seam thick accumulations were largely confi~ed to Morwell, Loy Yang/Flynn and Flynn Railway areas. Between the three coal areas and also bordering the northern and southern basin margins were extensive areas of interbedded coal/clay and clay sediments up to 140m in thickness. The main southern area for clay deposition approximates with the Traralgon Syncline. It trends northeasterly from Yinnar through Jeeralang and Churchill and probably extends across the southern flanks of the Loy Yang Dome as evidenced by the thinning of coal on this southern flank and into the Flynns Creek Syncline. Within the thicker parts of the MIA clay facies cemented siderite bands up to lm thick occur at regular stratigraphic levels and have widespread regional extent (see figure 8). The number of bands tends to increase towards the thickest part of the clay depocentres. Petrographic studies indicate most of these siderite bands comprise a 61

cryptocrystalline aggregate of sparry siderite in some cases cementing and replacing fine grained quartz grains (Bolger, 1984b). The major clay depocentre in the north of the basin trends eastwards from the Yallourn open cut and opens out to the

north of Rosedale eventually connecting to the marine interface. Similarly to the southern clay centre, carbonate cemented bands occur in the Tyers and northeast Maryvale area but appear to comprise mainly dolomites and ankerites.

til

. -~ 0

s I

TA,t,JULC,ON

S'YMCllN(

/'OllafU

IURYVAL£

(H\..lt(Htl

JH~LA" G

All_itllL_

0,1111,

,-,an,

PIHS{lilTIIIA'IJltlLSURfA(f

lli1U

.,, §:::J TICAl

oor.r,...,

r=:::J(LAY

THOAPQ,t,lf YCN..C.llU(S

, n - PII( • f(Atlllol!T SASOfOO

Similar clay sequences occur as facies equivalents to the Yallourn, Morwell lB and Morwell seams and broadly occur in the same areas as for the Morwell lA facies equivalents. For the Yallourn seam (Figure 6), the clay facies equivalents can be seen to develop on the Yallourn Open Cut Western batters (Webster, 1983), in the Tyers area, east of Yinnar, and to the north and west of the Loy Yang open cut. In many areas, they have been removed by erosion. The facies equivalents of the Morwell lB seam (see Figure 4) are less well represented due to erosion on the basin margins. Neverthless, areas where the Morwell lB seam splits and thins into clay occur east of Yinnar, west of the Yallourn Open Cut and in the vicinity of the Rosedale Monocline. The thick Morwell 2 Seam depocentre under Maryvale and Tyers (see Figure 3) grades rapidly to the northeast into thick clay sequences under Glengarry township. In the Traralgon Syncline and Loy Yang/Flynn area, the Morwell 2 seam splits into three subseams (Morwell 2a, 2b and 2c). In the Moe/Newborough area, west of Driffield and along the Rosedale Monocline, the Morwell 2 interval grades into interbed-

Figure 8: North south generalised cross-section through the major coal deposits of the Latrobe Valley Depression prior to Late Ter tiary Folding. Note: base line drawn through Yallourn/Morwell l A seam interface. Section line shown on Figure 2.

ded sediments and basalts of the Thorpdale Volcanics. DISCUSSION Sedimentation Rates The broad coincidence in the size of areas over which the four main seams occur and their similar (80-lO0m) thicknesses indicate a constant rate of subsidence during deposition over a large part of the central and southern Latrobe Valley Depression. This continued for long periods during the Middle Tertiary and kept pace with a constant and equal rate of peat formation and clay sedimentation. The similarity of seam thicknesses and their lateral facies equivalents imply the interseam breaks occurred at the end of regular time intervals. Major changes to the peat swamps at these times included a shifting of depocentres accompanied by changes in coal quality (ash constituents) which occur between the different seams, and indicate the interseam breaks may represent significant diastems which could span greater time intervals than their thicknesses would suggest. Their regularity of occurrence suggests they are the upper hierarchy of cyclic-sedimentation to the 63

lithotype cycles which occur in the coals (Mackay et al, 1984). The cyclic nature may be related to eustatic cycles of sea level change (Vail et al, 1977). The broad spaced spore-pollen zonation used in the Latrobe Valley is inadequate at this stage to either define sedimentation rates further or make any estimates on the time interval represented by the postulated diastems. Migration of Coal Depocentres Differential compaction appears to best explain why the major coal depocentres rarely coincided. By this mechanism, as shown diagrammatically in Figure 10a, an initial thick pile of peat (proto-Morwell 2 seam) consolidates faster than for its surrounding sediment equivalents. Ultimately a relative ground surface low over the thicker peat areas is created, becomes flooded, creating a site for lacustrine sedimentation and a sediment sink for inflowing fluvial material. Slower compacting areas over thick sediment equivalents to the Morwell 2 are left at higher levels and therefore become peat accumulation sites for the subsequent proto-Morwell lB and lA seams. (Figure 10b and 10c) Gradual expansion of the lacustrine systems during Morwell lA finally became sufficiently widespread to terminate this coal depositional phase. Differential compaction created a lowering of ground surface accompanied by lacustrine flooding over the thick Morwell lA and lB sites. Deposition of the Yallourn seam tended to revert back to the relatively higher ground where the Morwell 2 deposition centre had been (Figure 10d). Structural Control on Coal Depocentres In the Latrobe Valley Depression, most of the coal-bearing areas contain thick coals interspersed with thick non-coaly intervals. However, the vertical stacking of more than two coal depocentres in a continuous succession does occur on some of the present day structural highs. These include the Morwell Anticline (where Yallourn, Morwell lA and lB each exceeded 80m thickne~s), the Loy Yang Dome (where Morwell lA, lB and 2 each exceed 80m thickness) and to a lesser extent the Flynn Railway Dome (where Morwell lA and lB can be up to 80m each and where Yallourn is between 60 and 80m in aggregate thickness).

This suggests that more continuous coal rather than alternating clay-coal deposition tended to occur around and on some of the present day structurally positive features. These areas must therefore have been relatively positive at the time of coal deposition to maintain a base level consistently high enough to prevent flooding by lacustrine sedimentation. Probably these were areas of slower sinking within the Latrobe Valley Depression. However, other positive structures such as the Rosedale Monocline, the Hazelwood Dome and the Yinnar Dome appear to have been less continuously active, and thick coal/clay sequences intersperse between thick coals. Eustacy and Migration of Coal Depocentres Eustatic sea level changes effecting the coal depositional sites requires maximum westwards marine transgressions during times of Morwell 2 and Yallourn seam deposition, as these are the more westerly located seams. Conversely, sea level regressions are required during deposition of Morwell lA and lB to allow for their eastward expansion (Figure 10). However, the Cem1zoic sea level curves of Vail et al (1977) show that shortly before Morwell 2 deposition commences, sea levels were falling and then rose in stages throughout most of the depositional time for remaining Morwell and Yallourn seams. They began to fall towards the end of deposition of Yallourn seam. These changes in sea-level contradict the borehole results on Figure 9. Here the maximum westerly transgression of the marine limestones appears to occur at a time approximating to the end of deposition of Morwell 2 seam and was followed by a gradual eastwards regression of limestones coeval with deposition of the Morwell lB, lA and Yallourn seams. Therefore, there appears to be little correlation between the overall eustatic sea level curves and the changing of coal depocentres. The eustatic sea level changes appear to have been barred from significantly effecting the Latrobe Valley by the blocking effect of a long standing and stable barrier system developed at the eastern end of the Latrobe Valley. Nevertheless, it is conceivable that if a more detailed spore-pollen zonation system were to be defined correlations between eustatic sea levels and coal seam development may exist.

··· @

-2

-JO -400 -SOO -600 -100

-800 ~

>O DS

(t U S OALS

• •;::,• ~ :':· .

· 900

l· "" ESTO" l

.-.u.

· 1000

ltOC!lONTAL SOU

Mf~lOI{

tw:ni--t:

.......

I Ko

__J

Figure 9: Latrobe Valley west to east cross-section along central line (modified from Holdgate, 1979). Section line shown on Figure 2.

LEGEND COHTlf!IUOUS COAl FOIIMIN' SMAJoU"S

•

=--=--=-.I

=-=-=--

(MTS

00.,u,uu,,ru LAOJSTIIIIHC

HlB

NOIIW(Ll II SE AN

l'IINIWELL 2 SE AM

CHUltCMILL

rNVlltONM(HTS

-. 1xco LACI.ISTIIIN{ "L UV IAL

ilPtNO KA.IIIN{ IA.RIIIHI [H"llto..M[NTS

-_'.j

vvv

To\ LLOUflN KAM NORWELL I A SEAM

SCALE IN KM.

,\L T[lbl.At IN!, COAL SNA#S

I\ICI LACUSTIIIIH£ [..VIit

y HlA

"'111W(LL

IIOS£0 AL E

HIAAi\L!;ON

wot.CANICS

Figure 10: Palaeoenvironment maps and sections showing stages of development for Latrobe Valley coal seams. 65

Fluvial Influences on Coa Depocentres Other than during interseam times, fluvial influences appear to be relatively minor, with most of the lateral facies and interseam sediments comprised of clays which probably represent lacustrine environments of deposition. Thick sand intervals, typically found in fluvial dominate envir nments rarely occur, and many that do ar-e not in proximity to tlj.e . major coal seam areas (see Figure 8). Rather most appear to trend towards the thick clay areas possibly representing lake edge deltas. o washouts or major erosion of seams occur at seam boundaries which tend to show graded contacts into clays. Isopachs of interseam sediments such as occurring during Morwell lA time show closed basin-like areas dominantly of clay more typical of lacustrine lakes than the more sinuous and elongated shapes of a typical fluvial channel system. The abundance of carbonates and presence of algae within the thicker clay areas, whilst not exclusive to lacustrine environments are nevertheless more common than for fluvial systems (Pick and High, 1972). Facies equivalents and thick seam deposition The non-coaly facies equivalents of all Morwell and Yallourn seams are thick, prec¥,minantly clay sequences which were deposited between the i:,at swamps of the Latrobe Valley Dep*ssion and the highlands beyond the basin margins to the north and south, in peripheral lacustrine lakes. These lakes constituted an effective barrier between the basin and the swamps, filtering out the river derived coarse sediment in lake edge deltas. Suspended clay matter that travelled further across the lakes would have tended to flocculate and settle near to the swamp margins where acidic waters prevailed. As a result of the above, the swamps received negligible amounts of fluvially derived inorganic sediment constituents directly, which explains their overall very low inorganic ash content of less than 5% (dry basis), and their very great thickness. However, at regular intervals over which enough peat accumulated to eventually form 80-lO0m of brown coal, the lacustrine and some localised fluvial environments

become sufficiently widespread to terminate these coal forming periods. At the eastern eel@._ of the major coal areas along the central Latrobe Syncline a well preserved section shows facies changes out to the edges of the marine boundary (Figure 9). The coals tend to grade laterally into sands and clays associated with the marine barrier system described by Thompson (1980) as the Balook Formation. Quality data from coals along the Rosedale Monocline and in the Latrobe yncline have shown higher ash, including sulphur content which is considered to be related to the marine influence (HolJle et al 1978). To sustain thick and continuous swamp development, abundant quantities of mineral salts are considered to have been required for plant nutrition. These may have been provided in two main ways. The interbedding of lake-edge clays with the peats could have enabled the lateral penetration by diffusion of nutrient-rich lake waters through to the swamp. As the swamps would have been perched on relatively high ground relative to the lakes most of this lateral diffusion would have occurred as shallow groundwater percolation through the peat masses underlying the growing vegetation. Otherwise vertical leakage of nutrient rich groundwater from confined aquifers underlying the coal swamps (such as the Traralgon Formation aquifers) could have occurred at slower rates through the confining beds of overlying coals, peats and clays. Such slow discharge is considered to have played a significant role in maintaining the recent Moe Swamp and hence may also have provided a source of sediment free water during peat swamp growth in thei Latrqbe Valley (Brumley and Holdgate, 1983).

CONCLUSIONS As an initial step to understanding the depositional environments under which Latrobe Valley coal seams formed, a series of coal isopachs were prepared from borehole data, to show their threedimensional geometry and subsurface extent. The four main seams in stratigraphic order include the Morwell 2, Morwell lB, Morwell lA and Yallourn seams. Each seam was found to cover a surface area averaging 600 km 2 • About 20% of these areas con-

tains seams of thicknesses greater than 80m. The four major coal depocentres were found to occur predominantly to the south of the Latrobe River where a series of Late Tertiary folding events largely of post-coal deposition age caused uplift and erosion of seams over what is now currently defined as economic coal fields.

In the Latrobe Valley the coal depocentres (coal > 80.0m) for each seam are not spatially coincident, but show evidence of migration from the western end of the valley (the Morwell 2 seam) to a central valley position for the Morwell lB and lA seams, and then a return to the westlery location for the Yallourn seam. This shifting of depocentres appears to be a consequence of more rapid compaction of the coal relative to its lateral fades equivalents. Depressions forming over thick, more rapidly compacting coal became in turn sediment sinks, and less rapidly compacting areas overlying thick sediments became in turn sites for new coal deposition. Tertiary eustatic seal level changes appear to have had little effect on these processes, barred as they were during the main periods of coal deposition by a longstanding and stabilised barrier system at the eastern end of the Latrobe Valley.

)

Kaolinitic clays constitute the dominant sediment lithology associated with the coal seams, except for sandier sediments localised near the marine barrier system. Thick ( > 80m) clay sequences which are a fades equivalent to the coals, occur between the main coal areas and the basin margins. They resulted from sedi,rnent inputs from the higher surrounds of the Latrobe Valley. The coal sequences interface the clay sequences in a complex intertonguing of coal and clay. Carbonaceous content decreases into the clays, which become lighter in colour away from the coal. Within the main clay depocentres thin but extensive finegrained sparry carbonates (siderite, ankerite and dolomite) appear as discrete

bands at regular stratigraphic intervals, and as localised disseminations. Algae have

been identified within these clays. The thick clay sequences are considered to indicate prolonged periods of lacustrine sedimentation which transgressed and regressed with time. Major expansions in the lacustrine regime occurred at the end of each major coal forming period, and during the deposition of the Morwell lA seam. These expansions temporarily interrupted the major coal forming periods. In some areas such as Loy Yang, Flynn and Morwell, base level for coal deposition appears to •have been fractionally higher (possibly due to local tectonics), so that these expansions had less effect, and coal deposition continued uninterrupted.

The effective surrounding by lacustrine lakes of the main coal swamps over large areas of the Latrobe Valley has probably been the major factor in the stabilisation of the coal swamp environments. This prevented major influxes of sediment from the more destructive fluvial sources off the surrounding hills. Most sand inputs from these sources were deposited on the outer margins of the lake systems, as localised deltas. The suspended clay matter would have tended to flocculate and settle under the more acidic waters off the lake-coal swamp margins. Nourishment for the vegetation growing in the swamps was probably maintained laterally from the nutrient-rich lake waters, and by upwards movement of groundwaters from the extensive aquifer systems of the Traralgon Formation that underlie the Morwell 2 seams.

ACKNOWLEDGEMENTS The author wishes to acknowledge the Principal Geologist, Dr. C. M. Barton and Mr. R. L. Webster and Mr. P . G. Bolger for their useful comments and discussion, and others of Exploration and Geological Division, SECV, for assistance in the completion of this report. This paper is published with the permission of the State Electricity Commission of Victoria.

REFERENCES Bolger, P.G., 1984(a): Lithofacies Assemblages and Boundary Relationships of the Tertiary Latrobe Valley Group and Overlying Strata in the Latrobe Valley, Victoria. MSC Thesis. Department of Earth Science, Monash University, Victoria. 1984, unpub. Bolger, P.G ., 1984(b): Petrography of siderites from the Yinnar/Hazelwood area. E&G Geological Laboratory Report 1984/8, unpub. Brumley, J.C. , and Holdgate G.R. , 1983: An assessment of the confined groundwater resources of the Moe Swamp Basin, Victoria, Australia. Inter. Conf. on Groundwater and Man. Sydney 1983, 41-50. Brumley, J .C., Barton, C.M., Holdgate, G.R. , Reid, M.A. 1981: Regional Groundwater Investigation of the Latrobe Valley 1976-1981. Joint report SECV and DMEV 1981, unpub. Gloe, C.S., 1960: The Geology of the Latrobe Valley Coalfield. Proc. Australas. Inst. Min. Metall., 194, 57-125. Gloe, C.S., 1976: Brown Coal. In Douglas J .G. and Ferguson J.A. Eds., Geology of Victoria. Geol. Soc. Aust. Spee. Pub/. No. 5, 378-386. Gloe, C.S., 1980: The economically winnable brown coal reserves in the Latrobe Valley. SECV Fuel Dept, E&G Div. Geological Report No. 26, unpub. Gloe, C.S., Barton, C.M., Lawrence, C.R., Nahm, G.Y. , Reid, M.A. , 1979: Regional Groundwater Investigation of the Latrobe Valley - Initial Data. of Compilation and Review SECV/DMEV Report 1979/81, unpub.

Mackay, G.H., George, A.M., Attwood, D.R. and Gaulton, R.J., 198~: The cyclic occurrence of brown coal lithotypes. Abstracts of 18th Newcastle Symposium, 1984. Picard, M.D. and High, L.R. , 1972: Criteria for Recognising Lacustrine Rocks. Rigby, J .K., and Hamblin, W.K., Eds. Recognition of Ancient Sedimentary Environments. Soc of Economic Paleontologists and Mineralogists, Spee. Pub/. 16, 108-145. Thomas, D.E. and Baragwanath, W., 1949: Geology of the Brown Coals of Victoria. Part 1 - Min. Geol. J. Viet ., 3(6), 28-55. Thomas, D.E. and Baragwanath, W., 1950: Geology of the Brown Coals of Victoria. Part 2 - Min. Geol. J. Viet., 4(1), 36-52. Thomas, D.E. and Baragwanath, W., 1951: Geology of the Brown Coals of Victoria, Part 4 - Min . Goel. J. Viet. , 4(3), 36-50. Thompson, B.R. , 1980: The Gippsland Sedimentary Basin - A study of the onshore area. PhD Thesis Univ. of Melbourne, 1980, unpub. Vail, P.R, Mitchyum, R.M. and Thompson, 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.) to Seismic Stratigraphy, applications hydrocarbon exploration. AAPG, Memoir 26, 83-97. Webster, R.L. , 1984: Yallourn Open Cut, Western Batters edges of Yallourn seam. SECV Fuel Department, E&G Technical Note No. 108, unpub.

Holdgate, G.R., 1979: Latrobe Valley Regional Groundwater bore cross-sections. SECV Fuel Depart, E&G Div. Report July 1979, unpub. Holdgate, G.R., 1985: Stratigraphy, Structure and Brown Coals of the Moe Area. Australian Coal Geology, 5, * * -**. Horne, J.D., Ferm, J.D. , Caruccio, F.T., Baganz, B.P. , 1978: Depositional models in coal exploration and mine planning in Appalachion Region. AAPG Bull. 62, (12), 2379-2411 .

G. R. HOLDGATE Exploration and Geological Division Fuel Department State Electricity Commission of Victoria 15 William Street, MELBOURNE 3000.

)

APPLICATION OF IN-SEAM SEISMIC METHODS TO COAL MINE PLANNING By

G. R. Hatherly, P. F. Irwin and G. E. Holt

ABSTRACT

The in-seam seismic method allows geological conditions within coal seams to be evaluated during exploration programmes or as part of the mine planning process. Typically, the method can be used to identify structures such as faults and dykes, as well as identifying changing seam and roof conditions. Knowledge of the existence of such features is essential for efficient coal mine planning. Surveys are implemented by detonating explosive charges in the coal seam and detecting the the resultant seismic waves travelling through the seam. Analysis is made of waves travelling directly from shot to detector and any waves reflected from anomalous structures. The surveys may be conducted between surface boreholes, between mine and boreholes or between faces within a mine. Results from successful surveys conducted in NSW by the Australian Coal Inare Laboratories Research dustry presented. Faults have been mapped by a borehole-to-borehole survey conducted at Glennies Creek, a face-to-face survey conducted at West Cliff Colliery and a face-toface-borehole survey conducted at Invincible Colliery. Dykes and areas of poor roof condition have been identified by a face-toface survey at Liddell Colliery.

many areas, if the anomalies were identified beforehand, such information could affect the choice of mining method or layout geometry. In marginal operations, the viability of mining itself is affected. All Australian coalfields are disturbed by geological anomalies of some kind. The in-seam seismic method allows the location of such anomalies to be determined prior to mining. In the U.K. and West Germany where the method has been principally developed, it is applied on a routine basis. Interest in in-seam seismic techniques is now spreading world-wide. Face-to-face, face-to-borehole and borehole-to-borehole methodologies are all possible. In Australia, the Australian Coal Industry Research Laboratories (ACIRL), with financial support from the Australian Coal Association, have undertaken a research programme aimed at determining the suitability of in-seam seismic techniques to Australian conditions. This project has obtained encouraging results and it is intended to establish a routine service for the industry. In the two years that the project has been running, trial surveys have been conducted in collieries and from boreholes in both the NSW and Queensland coalfields. Surveys have been conducted using a conventional digital seismograph and the results interpreted using special purpose processing software. General in-seam seismic procedures and the results from four recent surveys are described in this paper.

INTRODUCTION The problem of geological and structural anomalies such as faults and dykes in and adjacent to coal seams is very real, particularly with the present introduction of high technology rapid extraction mining systems such as the longwall method. The unexpected occurrence of an anomaly within a production block of coal can cause serious disruption to extraction operations, and can obviously be very expensive. In

THE IN-SEAM SEISMIC METHOD SEISMIC WAVES In-seam seismic surveys measure the various seismic waves generated by the controlled detonation of explosive shots. The waves propagate through the roof and floor strata and through the seam itself. Each wave has its own characteristic velocity and particle motion. Velocities are also a function of rock type.

__., 0

I I I I I

■

P - Wave V)

u w V) _J

Seam J_} Wave

512 1

Figure 1: A geophone trace gather.

The fastest of the seismic waves recorded in an in-seam seismic survey is a wave travelling through the roof and floor strata. This wave is called a P-wave and has longitudinal particle motion. Energy reaches the geophones by refracting from the roof or floor back into the coal seam. • P-waves are often the strongest of the seismic waves. Waves with transverse particle motion, S-waves, also travel through the roof and floor strata but with lower velocities. S-waves are not easily observed in in-seam seismic surveys. The waves which travel through the coal seam are known as seam or channel waves. These waves may have transverse particle motions (Love-type waves) or coupled longitudinal and vertical particle motions (Rayleigh-type waves). Seam waves have lower seismic velocities than P and S-waves and arrive later on the seismic recordings. Seam waves are caused by interfering multiple reflections between the roof and floor. Their propagation is very complex. One significant property is that their different frequency components propagate with different velocities. This property is known as dispersion. THE SURVEY To set up a survey from a colliery face, it is necessary to drill holes in the face for both the shots and geophones (detectors). Typically 50 shots will be fired and recorded by 12 geophones. Geophones may have up to three recording elements arranged at right angles to each other to allow longitudinal waves to be separated from transverse waves. Roadways for the survey should be at least 200m long. 'L' and 'U' shaped colliery workings with cross cutting targets make good sites for surveys. Borehole surveys require the installation of geophones in the boreholes. The geophones are usually grouted into the seam on completion of the borehole and are useable for several years. Shots for borehole surveys can be either explosive charges or non destructive sources. Non destructive sources necessarily have low energy and the

results of repeated 'shots' are summated to build up useable data. Explosives are still the best of the available sources. BASIC DATA PROCESSING Interpretation of in-seam seismic data requires numerous computer processing stages. The overall intention of the processing is to enhance the desired seismic signal and to decrease the background noise. Noise includes random signals such as those caused by mining operations and any unwanted seismic waves generated by the shots. The most significant of the basic data processing stages are the gathering, trace rotation and prefiltering steps. Gathering involves the merging of seismic data and survey co-ordinates into computer files containing the data recorded by single geophones (geophone gathers), or the data recorded by all geophones from single shots (shot gathers). Trace rotation is a mathematical process which ensures that each seismic trace (the data from a single geophone element) contains either predominantly longitudinal or transverse wave motions. Prefiltering is conducted to remove interference at any predominant frequency from the data. The 50 Hz noise caused by electrical interference from mine power supplies is a common form of noise removed by prefiltering. Figure 1 shows an example of a geophone trace gather. Each trace represents a half second recording from separate shots. The traces are arranged from left to right in increasing shot to geophone distance. Seismic events are characterised by the large amplitude departures from the background level. Both a P-wave and a seam wave are present on this gather. IMAGING Imaging is the processing operation for locating reflecting structures. Faults involving full seam displacement are the best reflectors of seismic energy but reflections

can also be obtained from smaller faults, from dykes, from mine roadways and from structure such as sand channels if they have sufficient vertical extent. To image a structure, there are three processing steps which often need to be applied: 1. The seismic data may need corrections to counteract the effect of dispersion. Dispersion causes channel waves to be recorded as long oscillatory wavetrains. These need to be converted to short pulses for accurate reflection mapping. The corrections may not be necessary if the data can be analysed over a frequency range for which velocity does not vary significantly with frequency. This occurs at the Airy phase and also in situations when the natural absorption of the seismic waves limits the observable frequency range. 2. The velocity of the seismic waves must be determined. Velocities are calculated by knowing the distance between shots and geophones, and travel time of the waves travelling directly between these points. It may be necessary to know both the velocity of individual frequency components (phase velocity) and the velocity of energy peaks (group velocity). 3. The reflections on the seismograms may need to be enhanced. Seismograms are usually dominated by direct arrivals and it may be necessary to reduce their strength to enhance the low amplitude reflections. One way of achieving this is to scale the data on the basis of the signal strength - the weaker the signal the greater the scaling. To illustrate steps 1 and 2, Figures 2, 3 and 4 show examples of group velocity analysis and dispersion corrections on the data shown in Figure 1. In Figures 2 and 3, seismic velocity is plotted against frequency, with contours indicating the strength of seismic signal for particular frequency and velocity. In Figure 2, the variation of velocity with frequency caused by dispersion is clearly evident. After application of dispersion corrections, the results in Figure 3 are obtained. Velocity is now independent of frequency. Figure 4 shows the data from Figure 1 after application of dispersion corrections.

Enhancement of low energy seismic signals is illustrated in Figure 5. The data here were obtained from that shown in Figure 4. The dashed line indicates the arrival of reflected energy.

In the imaging process, it is assumed that the targets are acting as reflectors of seismic energy or, more generally, as diffractors. Generally it is best to assume reflectors and to limit the strike of the structures to a suspected range. If a range is not known, the process is repeated for all possible situations. Results of imaging may be presented in contour map or seismic trace form. Figure 6 shows an example of results in a contour form. To interpret the map, both the amplitude and continuity of the anomalies must be considered. This is because noise can sometimes combine to yield strong amplitude features but rarely will these be continuous. Care must also be taken not to interpret reflection amplitudes as direct indicators of geological conditions. Geometrical focusing of seismic energy, for example, can also lead to variations in reflection map amplitudes. TOMOGRAPHY Tomography is the process of inferring rock properties by analysis of the waves travelling directly between shot and geophone. In a qualitative form, it may be used to infer the locations of faults by the presence or absence of channel waves. In more sophisticated forms, travel time data are analysed by computer to map the variation of seismic velocity across a panel of coal. Tomographic results are presented in contour form and may be for roof and floor strata or for the coal itself. Best results are obtained from surveys across 'U' shaped panels where the seismic waves cross the panel from many different directions. Figure 7 is an example of the results of tomographic analysis. Here, P-wave velocity is contoured in intervals of l00m/s. Low velocities may be indicative of poor roof conditions, higher seismic velocities indicate good conditions.

)

2000.00

dispersed seam wave

g u.J

0...

::::::,

Figure 2. Group velocity analysis - dispersed seam wave.

0.00

FREQUENCY- Hz

0 0::: l.!J

500.00

0.00 2000.00

u 0

......J u.J

Figure 3. Group velocity analysis seam wave.

recompressed

0... ::::::, 0 0::: L:)

0.00

FREQUENCY - Hz

250.00

0.00

I I I I

•

V1 0

z 0

u w

V1 ......J ......J

L I

w L

512 1

Figure 4. Geophone gather, recompression.

Figure 1, after

L.J

Lu V)

.....J .....J

.__

512 1

Figure 5. Enhancement of low energy signals.

-.J Cl1

"' 0 0

dyke reflections

- roadway ref le et ions

\ \

interpolation between roadway outcrops

□□□ DDD □ DDDD~DDDDD • SCALE 1: 6700 Figure 6. Reflection map

C) C)

0--.J

m _.

ER:---4000_

lt---3900----

□

~~2t:::c~~~~~~~J~~~ oc

= =

--:t

If)

1214000 N

co N

1214000 N

0 Q

- Ref lee tor 00 0

'Noise

1213000 N

photo "" □ir lineaments

1213000N

1: 10~00~0'..____ _ _ _ _ _ _ _ __ w

-e00

• N - Reflection Map. 8. Wes t Cliff Colliery Figure

~·

If)

SELECTED SURVEY RESULTS LIDDELL COLLIERY Results from the Liddell Colliery survey have been used in the preceding discussion. The colliery is situated in the Hunter Valley of NSW and is operated by Coal and Allied Industries Ltd. The survey was undertaken around a proposed longwall block to two dykes known to cross the area and to map washouts in the roof associated with a stream channel. Figure 6 shows one of the dyke reflection maps. The location of the seismic reflector ties in well with the known dyke outcrops. Note that the dyke has not been mapped across the whole block. This is due to the lack of reflecting points near U57 panel and does not imply that the dyke is discontinuous. Figure 7 shows the result of the tomographic analysis of the waves travelling through the roof and floor. Velocities across the block vary between 3700m/s and 4000m/s. Notable areas on the map are the areas of low seismic velocity to the left and right. That to the left coincides with an area where a dirt band in the seam thickens appreciably and difficult roof conditions develop. That at the top right is in the vicinity of the stream channel and washouts. Further work is currently underway to map the channel in more detail. GLENNIES CREEK At Glennies Creek, also in the Hunter Valley, an in-seam seismic survey was conducted between boreholes drilled as part of a routine exploration programme. Targets, for this survey, were any faults which might exist between the boreholes. To conduct this survey, geophones were

grouted at seam depth into four holes spaced between 250m and 500m apart. Shots were then fired, also at seam depth, from three holes located to the west, north and east of the line of geophone holes. The resulting seismograms were then examined for the presence of the characteristic seismic seam waves. From two of the shot holes, seam waves propagated to all geophones. The coal seam is inferred to be continuous between all of these holes. From the third shot hole, however, no seam wave propagation was seen at all. Coal seam discontinuity is thus implied between the shot hole and the closest of the geophone holes. A fault with a throw greater than the seam thickness is inferred to be present. WEST CLIFF COLLIERY West Cliff Colliery is located in the southern coalfields of NSW. As shown in Figure 8, northward development of the mine has been temporarily halted by a NW trending fault. Targets for the survey were two further NW trending faults suspected to be some 1 to 1.5km beyond the present fault. Seismic reflection mapping indicated no trace of the closer target. This fault is inferred not to be present. The further target, however, gave rise to weak reflections. These are at a distance of 1.5km, at the very limit of resolution for an in-seam seismic survey. At this distance the fault is not of immediate concern in mine planning. An unexpected fault was also indentified. This is shown in Figure 8 where there is a strong reflector along the projection of a NS fault in the workings. It aso coincides with an air photo lineament. This fault is definitely of concern in mine planning.

w C C

"'r-, l312000t~

Bo e hole Ge phone J

DOD DO □oo

v o

B88 ooo DO

...... 0

..c V)

BB Scale 1: 12 500 w

• • - Reflection Map. Figure 9. Invincible Colliery

INVINCIBLE COLLIERY Invincible Colliery is located some 100km west of Sydney in the western coalfields of NSW. The in-seam seismic survey conducted at this colliery involved shooting from the mine face and recording at the ground surface from four boreholes located up to 1.5km from the face. The target for the survey was a suspected NS fault zone to the east of the workings. Figure 9 shows the mine plan and the seismic result. A definite seismic reflector coinciding with the expected location of the fault zone has been mapped by the survey. Mine plans are being adjusted accordingly. APPLICATIONS FOR IN-SEAM SEISMIC SURVEYS

From the above results, it can be seen that in-seam seismic surveys can be used in situation ranging from exploration problems to detailed planning of longwall blocks. In the explorl.'tion setting, boreholeborehole surveys allow seam disrupting faults to be located. These same borehole geophones might then be subsequently used for a face-to-borehole survey to map structures affecting mine development. Surveys conducted across a longwall block might be aimed at locating cross-cutting dykes or faults , areas of poor roof, or even features as small as washouts. It must be stressed, however, that inseam seismic surveys are not always successful. There are many pitfalls and seemingly obvious targets might not be mapped. An intelligent application of the method is needed and it should be used in conjunction with all the other tools available to the mining engineer and geologist. Longhole drilling is one obvious supplementary procedure. So too is careful geological mapping and understanding of the mine geology. At present, in-seam seismic surveys cannot differentiate between dykes or faults. Nor can they indicate the throw of a fault. All they can reliably indicate is the presence of a disrupting feature. It is for the geophysicist, geologist and engineer, working together, to determine what these features might represent. In the future, it can be expected that the in-seam seismic method will have much greater versatility. It is more than likely

that all of the problems mentioned above will be solved. So too might the present limitations that arise with difficult site access. Certainly increased computing power and knowledge of seam wave propagation will increase capabilities. Two of the developments which will probably soon be production tools are repeatable down-hole sources and the capability to infer geotechnical rock properties from the in-seam seismic data. Low cost repeatable sources will be particularly useful in borehole-borehole work. Conceptually, such sources could be placed in boreholes together with a geophone, thus greatly increasing the range of possibilities for borehole-borehole surveys and face-toborehole surveys. The inference of geotechnical rock properties is possible from a knowledge of P and S-wave velocities. All that is required to make geotechnical predictions, is a sound understanding of seismic wave propagation, good geological control and sufficient experience to enable the predictions to be made with confidence. CONCLUSIONS

ACIRL's investigation into the applicability of in-seam seismic methods to Australian coalfields is showing that there is a definite application for the technology. Our experience is that in-seam seismic surveys can be conducted quickly and efficiently without any disruption to mining or exploration operations. Results showing the locations of dykes, faults and areas of changing geological conditions are all possible. Used in the correct manner, these permit effective mine planning and exploration.

P. J . HATHERLY P. F. IRWIN G.E.HOLT Australian Coal Industry Research Laboratories,, P.O. Box 83, North Ryde, NSW, 2113, Australia.