Issuu

ENGINEERING GEOLOGY SPECIALIST GROUP PAPERS, SERIES NO. 3 AUGUST 1984

INCORPORATES PAPERS PREPARED FOR THE 6th & 7th AUSTRALIAN GEOLOGICAL CONVENTION HELD IN CANBERRA ( 1983) AND SYDNEY ( 1984 ) RESPECTIVELY.

EDITED BY J. C. BRUMLEY

PUBLISHED BY THE ENGINEERING GEOLOGY SPECIALIST GROUP OF THE GEOLOGICAL SOCIETY OF AUSTRALIA INCORPORATED.

iii

CONTENTS

PREFACE

ENGINEERING GEOLOGY SPECIALIST GROUP

PART 1

PAPERS FROM THE 6TH AUSTRALIAN GEOLOGICAL CONVENTION, CANBERRA 1983

SYDNEY AREA TERRAIN ANALYSIS A. A. Finlays on

CHANNEL CHANGES IN AUSTRALIAN RIVERS SINCE EUROPEAN SETTLEMENT R. F. Warner

PART 2

PAPERS FROM THE 7TH AUSTRALIAN GEOLOGICAL CONVENTION, SYDNEY 1984

THE FORMATION OF TALC FROM MICA EFFECTED BY THE BREAKDOWN OF PYRRHOTITE by W. F. Cole and C. J. Lancuc ki

RAPID DECAY OF SLATE ROOFING TILES IN SERVICE A. Shayan and C. J. Lancuc ki

WATER RESOURCES MANAGEMENT IN MINING EXPERIENCES OF THE BHP COMPANY R. J. Frost

ISSN 0728-72 24

iv PREFACE The Engineering Geology Group was established in 1979 as one of the various Specialist Groups within the Geological Society of Australia; its foundation meeting took place on 8th March 1979. The aims of the Group are to promote engineering geology in the widest sense, including environmental and urban geology, hydrogeology, geology of quarried materials, geology of coastal processes, and geology as applied to civil engineering or mining engineering and related disciplines. To achieve these aims, the Group organises meetings for the presentation of papers, field excursions, and technical sessions at the Australian Geological Conventions. This publication follows the Series No. 1 and No. 2 Papers published in January 1980 and May 1982 respectively, It incorporates papers prepared for the Sixth and Seventh Australian Geological Conventions held in Canberra (1983) and Sydney (1984) respectively.

ENGINEERING GEOLOGY SPECIALIST GROUP COMMITTEE 1983-1985

Secretary

c. Brumley Mr, A. s. Power

Treasurer

Mr. w. Peck

Conunittee Members:

Dr.

Chairman

Mr. J.

c. M. Barton

Mr. J. L. Neilson State Representatives : - Australian Capital Territory

Mr. E. J. Best

- New South Wales

Dr. M. J. Knight Mr. R. B. Smith

- Northern Territory

Mr. H. A. Qureshi

- Queensland

Dr. G. w. Hofmann Mr. J. E. Martin

- South Australia

Mr. L. Wylde

- Tasmania

Mr. B. D. Weldon

- Victoria

Dr. w. F. Cole Mr. A. A. Finlayson

PART 1 PAPERS FROM THE 6th AUSTRALIAN GEOLOGICAL CONVENTION, CANBERRA 1983.

ENGINEERING GEOLOGY SPECIALIST GROUP PAPERS SERIES NO. 3, AUGUST 1984

SYDNEY AREA TERRAIN ANALYSIS Alan A. Finlayson CSIRO, Institute of Energy and Earth Resources , Division of Geomecha nics P.O. Box 54, Mount Waverley, Victoria Australia , 3149

ABSTRACT The P.U.C.E. (fattern-Q_ nit-_Q_om ponent-~v aluation) system for terrain analysis, based upon classific ation and subsequen t assessmen t has been applied to the Sydney Area, N.S.W. The terrain has been analysed, classifie d and assessed for a number of planning and engineeri ng geology purposes. INTRODUCTION The beneficia l use of earth science informati on "packaged " in a suitable manner is becoming increasin gly recognize d by engineers and planners. One aspect of this growing trend includes the implemen tation of terrain analysis technique s to provide a framework against which may be based rational engineeri ng decisions at feasibili ty, planning and construction stages. Most workable systems of terrain analysis are conducted on the understan ding that areas of similar geology (within the same broad climatic environme nt) will tend to show similar soil and rock parameter s and, as a result, similar planning and engineeri ng propertie s and problems. The developme nt and scope of the main systems _in use for terrain analysis, especiall y in America, Britain, South Africa and Australia have been described in reviews such as Stewart (1968), Woloshin (1968) and, more recently, Mitchell (1973), Ollier (1977) and Davidson (1982). The complete process of terrain analysis involves the ability firstly, to classify terrain on the basis of similarit y or homogene ity of certain attribute s and, secondly, to assess (qualitat ive) or evaluate (quantita tive) like areas for the propertie s of the terrain that are significa nt for the desired purpose. The scheme adopted for analysing the terrain of the Sydney area is known as the P.U.C.E. (.!'._attern- Q_nit-_Q_o mponent-!v aluation) System, which has been developed at the Division of Geomecha nics, CSIRO, Melbourne . The System was originall y proposed and discussed by Aitchison and Grant (1967), further elaborate d by Grant (1968) and, more recently, described in detail with consequen t amendmen ts, by Grant (1975a and 1975b), Grant and Finlayson (1978a) and Finlayson (1981 and 1984). The classific ation phase of the P.U.C.E. System is formalize d and well documente d, and the procedure s used to implement it remain the same irrespective of location or purpose. In fact, once the terrain in any one area

3 has been classified to a particular level within the system, the classification can be used validly as the basis for any number of assessments or evaluations that can be matched to the appropriate level within the hierarchy of terrain classes. In contrast, each assessment or evaluation remains individual to its purpose. That is, every assessment or evaluation is essentially different from all others. The individual nature of assessments or evaluations prevents standardiza tion of procedures and makes it necessary to specify exactly what is required from each, so that the field-work and testing may be designed accordingly . There is a popular misconcepti on that the process of terrain analysis will reveal all that there is to know about the terrain and its properties . It will be appreciated , however, that an almost endless variety of assessment or evaluation schemes may be devised depending upon the desired purpose. It is clearly not possible to assess or evaluate the terrain for every conceivable requirement at any one time. The particular requirement s relevant to any one purpose must be fully realised in the initial stages, and the set of desired information must be specified and listed as early as possible within the process. When the specificatio n has been finalized it is possible to incorporate the collection of relevant data and any in situ assessments or evaluations efficiently into the field-work phase. The P.U.C.E. System for terrain classificat ion has been used as the basis for assessments and/or evaluations for a number of purposes, including: (a) location of materials and routes for railway and road construction (Grant & Ferguson, 1979); (b) water resources (Grant & Ferguson, 1979); (c) engineering geology evaluation (Grant, 1976; Finlayson, 1982); (d) urban and rural planning (Grant et al., 1981); (e) regional development (Grant et al . , 1982); and (f) military and off-road mobility (Grant et al., 1979). In addition, examples of its use as a basis for:

(a) highway engineer-

..!!!& have been discussed by Grant (1971); (b) urban and regional planning have been discussed by Grant & Aitchison (1970), Arnot & Grant (1974) and Grant & Finlayson (1977a and 1978b); (c) definition of flood-prone land have been discussed by Arnot & Grant (1977); (d) road and highway route location have been discussed by Grant & Finlayson (1977b); (e) land capability assessment and aesthetic landscape appreciatio n have been suggested by Arnot & Grant (1981); and (f) planning within developing countries have been discussed by Grant (1982).

APPLICATION TO THE SYDNEY AREA The terrain classificat ion with correspondi ng assessments of the Sydney area are outlined, therefore, as an example of the transmissio n of engineering geological and pedological information (pertinent to engineering and planning objectives at the feasibility study level) through the medium of terrain analysis. The Sydney area (Figure 1), as defined in the study, corresponds to the area covered by the Sydney (SI56-5, 1:250 000 AAS) topographic map sheet. It occupies some 14 500 square kilometres and is situated on the Central Coast of New South Wales. It is bounded on the east by the coastline extending from Botany Bay in the south to Lake Macquarie in the north. In the west, the area extends from Kanangra Walls in the south, to the north-east of Glen Davis. It includes the Cumberland Plain in the south-east, the Hornsby Plateau in the north and north-east and the Blue Mountains Plateau in the west. Drainage

Cessnock•

Sydney

Area Lake Macquarie

I I

LITHGOW

I I I

•

I I I I I I I I I I

I I I

"'7•-•.,.._~ B"oTany-Say- - - - - - _j

•Moss Vale 100

Kilometres

•Marulan

Figure 1. Location Map (showing extent of the Sydney 1 :250 000 map sheet area)

is dominated by the Hawkesbury/Wa rragamba/Wollon dilly River system and its tributaries, such as the MacDonald, Capertee/Colo, Grose, Nepean, Cox's and Kowmung Rivers. The in i tial opportunity to apply the P.U.C.E. System to the Sydney area came in 1974 et seq from collaboration between the Division of Geomechanics CSIRO and the Civil Engineering School of the University of New South Wales, in establishing a post-graduate course in terrain engineering (Grant & Finlayson, 1977c). This collaboration took the form of:

5 (a) (b)

(c)

lectures and practical classes given at the University; analysis and classificat ion of the terrain of the: (i) Sydney-Picto n-Wollongon g area in 1974; {ii) Sydney-Kato omba-Lithgo w-Kurrajong area in 1975; (iii) Sydney-Gosf ord-Wiseman 's Ferry area in 1976; followed by excursions in each of the above areas, to demonstrate the techniques and scope of terrain analysis, assessment and evaluation to the students of terrain engineering at the University.

The analysis and classificat ion of the terrain of these areas involved the preparation of provisional terrain pattern maps and reports using aerial photograph interpretat ion, but with no field-work at that stage. As a substantial part of the Sydney map sheet area had be~n investigated by this process, a natural extension of these isolated exercises was to: (a) complete the photo-inter pretation of the remainder of the aerial photographs covering the whole of the Sydney map sheet area by delineating province and terrain pattern boundaries and recognizing landform units; (b) analyse and classify the terrain; (c) conduct the necessary field-work for validation of the classificat ion, including the investigatio n of the topography, soils, vegetation and land-use in each landform unit, in order to define terrain units, as required by the P.U.C.E. System; (d) produce a terrain pattern map and terrain pattern/ter rain unit description s for the whole of the Sydney map sheet area; and (e) as a demonstrati on, use this classificat ion as a framework against which to assess or evaluate, in the field, the terrain for a number of factors considered to be relevant to engineering in the Sydney area. Such an extension was commenced in 1977 by the author as a postgraduate thesis study topic in engineering geology (Finlayson, 1978). In addition to further illustrating the scope of such terrain analysis techniques and their uses - for example, by engineers and planners - the study attempted to demonstrate that, if required, such techniques could be conducted quite rapidly at a feasibility- type level using minimum resources and personnel. In this case, one engineering geologist with aerial photointerpretat ion and terrain analysis experience, aided by one assistant in both the laboratory and the field, completed the work in a period of six months, using as basic equipment a stereoscope , a hand soil auger, a clinometer, a geological hammer, a four-wheel drive vehicle and a few reference books. In the circumstanc es it was decided that an assessment rather than an evaluation would be the end result. The specificatio n for, and contents of the chosen assessment are given in the following section. The terrain was analysed and classified, to terrain unit level, in accordance with the latest definitions and modificatio ns of the P.U.C.E. System (Grant & Finlayson, 1978a). When the relevant information pertinent to topography, soil profile and vegetation formation had been collected, it was possible to classify fully and enumerate each terrain unit accordingly . On completion of the field-work (conducted during August-Sept ember and November-De cember, 1977) the aerial photographs were re-examined . Minor amendments were made where found necessary and the final province and terrain pattern boundaries were transferred to the 1:250 000 topographic base map to produce the terrain pattern map. To accompany the terrain pattern map (and essentially acting as an extended legend to it) a report was compiled describing the mapped terrain hierarchica lly in terms of provinces, terrain patterns and terrain units.

6 The terrain analysis and classification to terrain unit level (together with the terrain pattern map) is reported fully in Finlayson (1982), however one possible way of concisely summarising that analysis with exemplary, thematic assessments is illustrated in Table 1. SCHEME FOR TERRAIN ASSESSMENT The Specification In this investigation of the Sydney area, no single organization (as is usually the case) was responsible for the specification of the information to be collected. A scheme consisting of a broad range of categories, thought to be relevant to the Sydney area, was initially suggested, and evolved into its final form after discussions with several Sydney-based organizations had prompted modifications, additions and re-arrangement of the initial scheme. This scheme is presented here as one example of terrain assessment. The following factors were decided upon and each terrain unit was systematically assessed in relation to them. (a) Source of borrow for: (i) water-retaining embankments (dams); (ii) non-water retaining embankments (road, rail, etc.); (iii) fill. (b) Resources: (i) clean sand or gravel. (c) Suitability for: (i) road subgrade; (ii) shallow building foundations; (iii) trenching and/or tunnelling; (iv) septic tanks; (v) untreated roads. (d) Urban development: (i) potential; (ii) limiting factors. (e) Route (road and rail) location. (f) Liability to flooding in the natural state. (g) Relative susceptibility to tunnelling or gullying type erosion. (h) Slope stability. The Basis for the Terrain Assessment Following an encouraging similarity between the field determination and the corresponding laboratory soil identification (based on test results) of the same samples obtained in the Moreton Region, Part A, Queensland study (reported in Grant et al., 1975), it was decided (as with the assessment of the Moreton Region, Part B, conducted in 1977) to base the assessment largely on the Unified Soil Classification (U.S.C.) Groups (U.S. Department of the (In the Moreton Region, Part A study about 90% of the Interior, 1974). U.S.C. groups as identified in the field were found to be the same as the laboratory assigned groups; the 10% discrepancy was explained mostly by borderline cases between the u.s.c. groups CL and CH, or between CL, silty and MH.) The u.s.c. groups were identified in the field by following the set of procedures outlined in U.S. Department of the Interior (1974), modified slightly to comply with Australian Standard 1289 (Standards Association of Australia, 1977). The Unified Soil Classification System has been used as a basis for indicating the engineering properties of the soils falling into various u.s.c. groups. Relative ratings schemes (sometimes numerical) have been devised as a guide for the comparison for various purposes of soils in different u.s.c. groups (e.g., U.S. Department of the Interior (1974), British Code of Practice CP 2001 (1957, revised 1981), Wagner (1957), Clifford (1977) and Charman (1978)). Such a scheme was devised (being mainly a compromise of existing examples) for use in the Moreton Region Study (by Pappin & Richards, in Grant et al., 1975 and Grant et al., 1982).

Table 1: Example of One Possible Terrain Analysis Summary and Thematic Assessment D E Te=in Pattern 41.004

Location The Sldllion and Long Reef

Topography

Rock Type

Smooth, sloping tetrain Sediments (sandstone, bounded on one side shale, tu!f, claystone) (seaward) by cliff:s

C R

D O M I

N A

Soil Profile

Vegetation

Engineering Problem, (includillg associated te=in wills)

Duplex or gradational clayey silt or silty cl.ay OYer clay. Rock outcrop, boulder strewn

Grassland, scattered trees, bare

Rock outaop (2. 1.00, 3. 2.00), poor to average for road subgrade and building foundations, swelling ooil could be a problem, unsuitable for septic tanl<s, poor to unsuitable fer untreated roads (1.6. -- ); slopes> 10° (1.6.--, 3.2.00); undercutting by wave action could cause instability ( 3. 2. 00); flood liability - tidal inundation ( 2. 1. 00)

41.004

N. E. of the area, r-: . of Tuggerah Lake

Gently to moderately undulating te=in, minor dissected slopes

Sediments (sandstone, shale, tuff, claystone)

Duplex silty sand and gravel over sandy clay or clayey gravel

Grassland, scattered trees, wex>dland, open forest

Unsuitable for septic tanl<s (1.2.--, 1.3.--, 3.4. --, 3.5 . --) ; swelling soil could be a problem, unsuitable for untreated roads (3. 5. --); weathered rock could be erosion problem (3. 4. --); flood liability (9. 2.--)

41.004

N. of Newnes Junction, S.E. and N. of Katoomba

Undulating, sometimes Sediment., (sandstone, sloping temiin, minor shale, tuff, claystone) dissected slopes

Rock outcrop, shallow sand. Gradadonal sandy clay over clay or silty sand over clayey sand. Uniform silty sand/gravel

Grassland, scattered trees, shrubland, woodland, open/ closed forest

Rock outcrop (1.4.00, 3.5.00); compaction and sharing could be required, poor for untreated roads (1.3. 7-, 1.4.0-, 1.4. 7-, 1. 7. --) ; unsuitable to poor for septic tanl<s (1 . 4. 2-, 1. 4. --, 1. 7. --, 3. 5. --); swelling could be a problem (1.4.2-, 3.5.--J; high susceptibility to gully erosion (1. 4. --, 1. 7. --); flood liability (9. 2. 0-)

41.004

N.E. andN.W. N3D'Ow, steep of Uthgow, below dissected slopes, Groo,e Sandstone leading to drainage escarpment

Rock outcrop, pockets of silty/clayey sand or gravel. Minor pockets of duplex clayey silt over gravelly cla y

Bare, ,cattered trees, woodland, open/ closed forest

Rock outcrop (3 . 7. 00); slopes > 10° (3. 7. --); flood liability (9. 2. 0-)

41.004

Sandstone cliffs of the Blue Motmtains

Sediments (sandstone) Strongly dissected, strongly eroded terrain, with rough rocky and steep dissected slopes, incised gullies and cliffs

Bare, scattered trees, Mostly heavily Jointed rock woodland, open/ clooed outcrop, shallow pockets of silty sand and gravel. Occasion- forest al pockets of shallow duplex

Meetly rock outcrop, slopes > 10°, undercutting may cause instability of Joint controlled blocks ( 2. 9. --, 3. 2. 00, 3. 7. 0-); flood liability (9.2 .0-, 9.4.0-)

Sediment., (sandstone, shale, tuff, claystone)

silty ,andovw elayeysand

41.004

Widespread, especially W. and N. E. of the area

Steep dissected terrain

Sediments (sandstone, shale, tuff, cla ystone )

Rock outcrop, pockets of silty sand and gravel, shallow uniform gravelly clay, shallow duplex silty gravel over clayey gravel. Duplex or gradat!onal clayey/silty sand over c lay

Shru bland, woodland, open/ closed forest, minor rain forest, bare

Rock outcrop (2. 6. 00, 3. 2. 00, 3. 5.00, 3. 7. 00); unsuitable for septic tanks (2.6. --, 3. 5. --, 3. 7. --); slopes visually unstable (3. 5. --, 3. 7.--); slopes> 10° (2.6.--, 3.2.--, 3.7.--); flood liability (9. 2. --)

51.019

S. and E. of Richmond

Gently undulating temiin

Sediment., (older alluvium - sand, silt, clay, gravel)

Duplex silty sand over clay or clayey sand. Uniform or gradational sand, sometimes over clayey sand. Uniform clayey sand, gravel, cobbles

Grassland, scattered trees, woodland, open forest

Swelling soil could be a problem, unsuitable for septic tanl<s and untreated roads, high NSCept!billty to gully erosion (1. 2. 3-); shoring could be required, unsuitable or poor for septic tanks and untreated roads (1. 2. --); shoring required (1. 3. 6-, 1. 3. 8-, 1. 4. 6-, 1. 4. 8-); unsuitable for septic t=ks, high susceptibility to gully erosion (1. 3. 71, 3. 1. 71 ); shoring required, unsuitable for septic tanks (3 .1. 8-); slopes > 10° (3.1. --); flood liability ( depressions, lowe:r slopes, 9. 2. --)

52.002

Fringing the coastline

Flat tetrain

Sand

Deep uniform sand

Bare, shrubland, woodland, salt water swamp forest

Compaction could be required, shoring required, poor for untreated roads (1.1. 7-); subject to possible aeolian erosion if bare or de-vegetated; likely to contain areas of artificial fill with varying properties

52 . 002

Adjacent to the coastline

Gently sloping beaches, Sand often backed by low sand dunes

Deep uniform sand

Bare, low shru bland, scattered ttees

Shoring required, unsuitable for untreated roads (1 . 5. 70, s. 9. 7-) ; subject to aeolian and marine erosion; flood liability - tidal inundation (1 . S. 70); dunes subject to aeolian action especially when disturbed and de-vegetated

52. 002

Fringing the coastline

Strong Iy undulating temiin

Deep uniform sand

Bare, shrubland, scattered trees, woodland, open forest

Com pact!on could be required, shoring required, unsuitable for untreated roads, slopes> 10° (1. 8. 7-); subject to aeolian action especially when disturbed and de-vegetated

Sand

8 In a somewhat simplified version, this scheme was used as the basis for the assessment of sources of borrow, resources and suitability for the various engineering factors conducted in the Sydney area. The engineering suitability ratings (based upon the U.S.C. soil groups) that were used are shown in Table 2. The number of ratings was cut to five (i.e., two ratings better, and two worse, than AVERAGE) in all but two cases: (a)

in assessing material for water-retai ning embankments , it was considered that the soil was either suitable or unsuitable for that particular purpose and rated accordingly ; and

(b)

in assessing material for clean sand and gravel resources, a four point rating system was used depending upon the presence or not of sand or gravel, and if present, the amount and characteris tic of the fines (silt and clay) associated with it.

The remainder of the assessment including urban development , route location and flood liability, was based upon terrain features, and the relative susceptibil ity of the soil to tunnelling/ gullying erosion was based upon a field dispersion/ slaking test (adapted from Emerson, 1967). A small number of standard alternative ratings and answers, for each engineering factor, were evolved and these are set out in Table 3. It should be noted that the ratings (empirical only, based largely upon considerati ons of the shear strength characteris tics of the soil material alone) for slope stability given in Table 2 were not used in the final assessment, because the ratings did not take into account such in situ conditions as, e.g., individual slope characteris tics, underlying lithology and structure or possible disturbance and de-vegetati on of the slope, etc., that would also contribute to slope instability . A rating based on observation was adopted therefore and an indication was given on whether existing slopes were visually stable, potentially unstable or unstable. The complete set of assessment sheets is included in Finlayson (1978 and 1982). SIGNIFICANT RESULTS FROM TIIE CLASSIFICATION AND ASSESSMENT The terrain of the Sydney area has been analysed, classified, described and assessed in terms of 17 provinces, 46 terrain patterns and 170 basic topographic al landform units, which were further sub-divided on the basis of soil and vegetation differences into 1010 terrain units. One possible terrain classificat ion summary, (i.e., province, terrain pattern, location and dominant topography, rock type, soil profile and vegetation formation) together with selected possible engineering problems (within the limits of the particular assessment) , referred to specific terrain units, are given in Table 1. In this particular summary, a thematic approach has been adopted in which a terrain unit was considered to have a likely engineering problem only if: the suitability rating for road subgrade, building foundations , septic tanks and untreated roads was below AVERAGE (i.e., POOR or UNSUITABLE); (b) the slopes were greater than 10° or visually unstable; (c) shoring was required for trenching; (d) it was liable to flooding in the natural stage; (e) the soil had a high susceptibil ity to tunnelling/ gullying erosion; or rock outcrop existed. (f) Of course, many other criteria could have been chosen if so desired. (a)

Table 2: Engineer ing Suitabilit y Ratings based upon Unified Soil Classific ation SOL RCE OF BORRO \\ L '-IFIED OIL CLA IFICATIO:-. T~ r1 ..:al '.'tame s

L .S.C. Group S)mbols

SLITABILITY FOR

RESOL RCES Emban~ments

F,11

Clean sand gra, el

Groups

Road subgrade

Building foundations

(Slop, stabilit~)

Tren.:h1ng tunnelling

Scpllc tanks

Cntreated

GOOD

AVERAGE

roads

re1aining

S on-v.ater retaining

L'.'ISL ITABLE

EXCELLEST

XCELLE:-.T

GOOD

EXCELLENl EXCELLE'.'IT (EXCELLENT) SHORING

L'.'ISLITABLE

.• SER.-\GE

EXCEL LEST

GOOD

EXCELLENT EXCELLE'.'IT (AVERAGE)

SHORl'.'IG

EXCELLENT t:NSL ITABLE

G~(

LSSLITABLE

A\ ER.-\GE

GOOD

A\.ERAGE

EXCELLENT EXCELLENT (AVERAGE!

SHORING

AVERAGE

Cla)ey gra,els. poorly graded gra\elsa nd-day mixtures

SUITABLE

AVERA GE

GOOD

POOR

GOOD

EXCELLENT (AVERAGE)

Well graded sands, gra,elly sands: li1tle or no fines

UNSU ITABLE

EXCELLENT

GOOD

EXCELLENT (EXCELLE T) SHORING

Poorly graded sands, gra-,11) sands; liule or no fines

UNSU ITABLE

AVERAGE

GOOD

AVERAGE

GOOD (density important)

(AVERAGE)

Silty sands. poorly graded sand -silt mi,tures

s~,

SUITABLE (with compaction)

AVERAGE

A\ ERAGE

AVERAGE

GOOD (density important)

(AVERAGE)

Clayey sands , poorly graded sandclay mixtures

SUITABLE

AVERAGE

AVERA E

POOR

AVERAGE

GOOD

(AVERAGE)

Inorganic silts and very fine sands, rock nour, silty or cla yey fine sand s "ith slight plasticity

UNSU ITABLE

POOR

AVERAGE

NOT APPLICABLE

AVERAGE

problem

Inorga nic clays of lo" to medium plasticil) . gra, ell~ clays, sandy clays, silty clays. lean clays

SUITABLE (erosion protection?)

GOOD

A\ERAGE

NOT APPLICABLE

AVERAGE

POOR

NOT APPLICABLE NOT APPLICABLE

Water

Well graded gra-=ls. gra, cl -sand m1\lures: little or no fines Poorl~ graded gra,els. gra,el-sand

mh:tures: little or no fines Silt~ gra,el1-. poorl~ graded gra,el-

!.and-s ilt mI,1ures

Organic silts and organic silt•clays of lo" plasticit y

GOOD

UNSUITABLE EXCELLENT

GOOD

AVERAGE

GOOD

UNSUITABLE

POOR

GOOD

UNSUITABLE

GOOD

(POOR)

SHORING

AVERAGE

UNSU ITABLE

AVERAGE (swelling?)

(GOOD)

GOOD

UNSUITABLE

POOR

POOR (swelling?)

(LlN SU ITABLE)

SHOR ING

POOR

UNSUITABLE

POOR

UNSU ITABL E (swelling?)

(POOR)

SHOR I NG

AVERAGE

UNSUITABLE

POOR (swelling?)

(AVERAGE)

GOOD

U'.'ISUITABLE UNSU ITABLE

LlNSU ITAB LE LlNSU ITABLE

SHOR! G

Mostly GOOD (but shoring may ~ required)

Liquefaction

UNSU ITABLE UNSU ITABLE

UNSU ITABLE

POOR

Inorga nic days ot high pla s11city, fat cla~s

SL I TABLE (erosion protection?)

AVERAGE

UNSU ITABLE

NOT APPLICABLE UNSUITABLE

Organic day;; of medium 10 high plastici1~

L:-ISLITABLE

UNSU ITAB LE

UNSU ITABLE

(U:-1 UNSL' ITABLE NOT SU ITABLE) APPLICABLE UNSUITA BLE (s"elling?)

SHORING

Peat and other highly orga nic soils

UNSUITABLE

UNSLl l TABLE

UNSLllTABLE

(UN UNSU ITABL E NOT SU ITABL E) APPLICABLE UNSU ITABLE (swelling?)

SHOR ING

Inorganic silts, mi aceou s or diato• maceous fin~ sandy or si lly soil s. elastic silts

AVERAGE

UNSUITABLE

Table 3: Scheme for Engineering Aaaeaament ahowlng Alternative Ratlnga, Anawera and Limiting Factor•

PRO VINCE No. TERRAIN PATTERN No.

This ~crrain pa lltrn 1s tremo1e) tor hm1ttd ntent) !has d1Hicull acctssl

RESO U RCES

SOURCE OF BORROW TERRAIN

EmbankmenlS

UNIT

Water

retaining

Non.water retaining

Fill

No .

SOIL: SUITABLE

SOIL: EXCELLENT

SOIL: EXCELLEST

UNSU ITABLE

GOOD

(d<p!hl

AVERAGE

Erosion

POOR

pro1cc1ion cou ld be: required, (Stt dispersion test result)

NOT APPLICABLE (if soil dc:pth < 0 .2m . or rock outcrop) (Limned extent) ( Remote:) (Oifficull access)

UNSU ITABLE (dep!h)

NOT APPLICABLE (if soil dc:pth <0.2m, or rock outcrop) (Li mited c:xtc:nt ) ( Remote:) (Difficult accc:u)

Clean sand / gravel

Road subgrade

founda1ions

Slope srabili1y

Trenching / 1unnelling

SUITA BILITY for Sep1ic tanks

SOIL : GOOD

SOIL: EXCELLENT

Eus1mg slopes

\ISuall> stable ,

SO IL: GOOD

SOIL : EXCELL EST

SL'ITA BILITY for

unstablc

sand gra\·tl -

(with liulc or no fines)

GOOD

AVERAGE

AVERAG E POOR sand1gra\·cl (vd1h appreciable L~St;ITABLE amounts of UNSU ITABLE silt) ftnc:s (depth) ...,ashing (dc:pth ) required) Rock ou1c rop NOT POOR APPLICABLE Slope> 10' sand l gra vc:I (if soil dc:pth (wHh appreciable < 0 . 2m, NOT amounts of or rock APPLI CABLE clayey fines outcrop) (1f terram umt washm& auoc,atc:d ,,.,,th required) (Limited drainage extent) S)SIC:ms) (depth) ( Remot e:) (O1ff1cuh :-,iOT access ) APPLI CABLE f1f soll dc:pth 0 .2m, or 1f no s11mf1can 1 amounts of sand or ara\c:I, 1.e silt. clay o r or aan1c soi l) POOR

IL1m1tc:d exten t) ( Rc:mo tc:) 1Diff1cuh accesS)

f Based on Table 21

(Buc:d on Table: 2)

Building

IBasc:d o n Table: 2)

1Buc:d on Table 21

SHORl:-,/G t..:ould be a problem 1f undcrcu11

t;NSUITABLE

(AH~rage slope angle:)

(d epth)

:-,/QT APPLI CABLE

~tn1mum foundaiion depth

(if terrain unit associated "'Ith drainage or na1 surfaces)

Untreated

Po1en1ial

roads

SOIL : EXCELLENT

location

No maJor

Soil t)pc S011 depth

CO MMERCIAL

Roc k out.:rop

ln :-tucnccd b) bedroc k

Steepness of slo pe

Cut and fill required

restra1n1s

GOOD AVERAGE

(including

POOR

offices. h11h me

POOR

fac1or

Roure

HOUSING

GOOD

(depth to

Limiting

UNLIMITED

AVERAGE

required bedrock)

POOR

URBAN DEVELOP:'>IE~T

Liabili1y 10 nood

in natural st ate

L ~ St; ITABL E

UNSU ITABLE

tdepth to bedrock)

(depth fO bedrock)

Rod o ut crop

Rock ou:crop

Slope > 10'

Slope> 10 '

~OT APPLICABLE

Note: 1f terrain unn auoc1atc:d ,,., Ith drainage systems

(dep!h)

s .... c:lhng 1 l1quc:fact1on could be: a problem

(1f terrain unit associat ed ,,.,1th dramagc: S)Stc:ms)

Dens1 I)' 1mponan1 compa ct ion co uld be: rc:qu1rc:d

Flatnes s

INDUSTRIAL SERVICES OPEN SPACE (non -bu1h -up, large: areas of na1 terrain, rccrc:a11on)

Water hazard (SUbJeC1 to flooding) Slope ln )t ab1lit) Roc k 1ypc: L1m11c:d (''(len t Rc:mo tt D1ffo: ult <ICCeSS '.'lone:

Roc k outcrop

gullying

YES

HIGH

:-.o

\IODERATE

SLIG HT :-.O~E (d1spcrnon slak1na assc:umc:ntl

bu1ldin1s , e1c . ) (nppmg blaSllng required)

Susccp11bili1y 10 1unnelling/

Embankment rc:quarc:d Embankment ...,,th cuhen bndgc: required Cuhc:rt bridge: maJo r bridge: required A\Old due: 10: topoaraphy , soil roc k propcr11c:s. slope ms1ab1llt) . tc:ndc:nq to

nood

Slope> 10'

NOT APPLICABLE t1f 1c:rram unit aswc1a1c:d """h dramagc: \)\I C:ffl\ )

( Based on Table: 2)

I Based 1Jn Table 21

!Based on obsc:r\lll0R\I

I Ba sed on Table: 2)

l Basc:d o n Table: 21

(Based on Table: 21

tBasc:d on terrain fc:a1urc:S)

I Based o n 1c:r ra1n fea1urei1

1Ba sc:J o n terrain fc:a1urc:sl

1Buc:J " " :c:rrllin

1Ba,eJ on fic:ld J1\~r ,1on ,ta.,,n, 1c:,1

IU IU IC\I

,.... 0

CONCLUSION The example given has attempted to show the applicability of terrain analysis as a tool for engineers, planners, and designers. With appropriate techniques for assessment and evaluation, the most rational and most economic engineering and planning decisions (with respect to the terrain at least) can be made objectively with limited personnel and resources. ACKNOWLE r:x;EMENTS The help and advice of Keith Grant, Brian Richards, Jack Pappin, Leo Connelly, Gordon Aitchison (currently or formerly CSIRO), Owen Ingles (University of N.S.W.) and the many Sydney-based organizations who gave freely of information are gratefully acknowledged. REFERENCES Aitchison, G.D. and Grant, K., 1967. The PUCE Programme of Terrain Description, Evaluation and Interpretation for Engineering Purposes. Proceedings of the 4th Regional Conference for Africa on Soil Mechanics and Foundation Engineering, Vol. 1, pp.1-8. Arnot, R.H. and Grant, K., 1974. Land Classification for Urban Growth. Application of a System of Terrain Classification to Urban and Regional Planning. Archetype, Vol. 4, No. 2, May, pp. 28-32. Arnot, R.H. and Grant, K. 1977. Land Classification for Land Use Planning. Terrain Analysis by the PUCE System in relation to Definition of Flood Liable Lands. Royal Australian Planning Institute Journal, Vol. 15, No. 3, August, pp. 106-108. Arnot, R.H. and Grant, K., 1977. Environmental Planning - The Application of a Method for Terrain Analysis to Economic Land Capability Assessment and Aesthetic Landscape Appreciation. Landscape Planning, Vol. 8, pp. 269-300. British Standard Code of Practice, CP 2001, 1957. Site Investigations. (The Council for Codes of Practice, British Standards Institution, (Recently revised as British Standards Institution, London), 123 pp. Code of Practice for Site Investigations BS 5930:1981). Charman, P.E.V. (Ed.), 1978. Soils of New South Wales. Their Characterization, Classification and Conservation. Soil Conservation Service Technical Handbook No. 1, (Soil Conservation Service of New South Wales, Sydney), 236 pp. Clifford, J.M., 1977. Compaction of Road Materials Part 3 : Soil Classification Systems and the Development of a Compaction Plant Recommendation Chart. National Institute for Transport and Road Research Technical Report RC/7/77, September, (National Institute for Transport and Road Research, South Africa) 27 pp. Davidson, D.A., 1982. The Assessment of Land Use Capability. Proceedings of the First International Symposium on Soil, Geology and Landforms Impact on Land Use Planning in Developing Countries (LANDPLAN 1), pp. G2.l-G2.57. Emerson, W.W., 1967. A Classification of Soil Aggregates based upon their Coherence in Water. Australian Journal of Soil Research, Vol. 5, pp. 47-57. Finlayson, A.A., 1978. Terrain Analysis, Classification and an Engineering Geological Assessment of the Sydney Area, New South Wales (Unpublished). Thesis, Dept. of Earth Sciences, Monash University. Finlayson, A.A., 1981. The PUCE System for Terrain Analysis - A Summary. Australian Geomechanics News No. 2, pp. 37-44. Finlayson, A.A., 1982. Terrain Analysis, Classification and an Engineering Geological Assessment of the Sydney Area, New South Wales. Volume 1 Terrain Analysis and Classification, 258 pp. Volume 2 : Terrain Assessment, 199 pp. CSIRO Aust., Div. Appl. Geomech. Tech. Pap. No. 32.

12 Finlayson, A.A., 1984. Land Surface Evaluation for Engineering Practice : Applications of the Australian PUCE System for Terrain Analysis. Quarterly Journal of Engineering Geology, London, Vol. 17 (in press). Grant, K., 1968. A Terrain Evaluation System for Engineering. CSIRO Aust., Div. Soil Mech., Tech. Pap. No. 2, 27 pp. Grant, K., 1971. Terrain Evaluation for Engineering Purposes. Proceedings of the Symposium on Terrain Evaluation for Highway Engineering, Special Report 6. Australian Road Research Board, Melbourne, pp. 81-107. Grant, K., 1975a. The PUCE Programme for Terrain Evaluation for Engineering Purposes, Part I. Principles. CSIRO Aust., Div. Appl. Geomech., Tech. Pap. No. 15, 2nd Edition, 32 pp. Grant, K. 1975b. The PUCE Programme for Terrain Evaluation for Engineering Purposes. Part II. Procedures for Terrain Classification. CSIRO Aust., Div. Appl. Geomech., Tech. Pap. No. 19, 2nd Edition, 68 pp. Grant, K., 1976. Terrain Classification and Evaluation for Engineering CSIRO Aust., Div. Purposes for the Canberra Area, A.C.T. and N.S.W. pp. 266 22, No. Pap. Tech. Appl. Geomech., Grant, K., 1982. Land Capability Assessment and Evaluation for Land Use Planning. Proceedings of the First International Symposium on Soil Geology and Landforms - Impact on Land Use Planning in Developing Countries (LANDPLAN 1), pp. G3.l-G3.45. Grant, K. and Aitchison, G.D., 1970. Terrain Studies for Urban Development. UNIBEAM - University of Singapore, Building and Estate Management Annual Magazine, 1970/71, pp. 59-64. Grant, K. and Ferguson, T.G., 1979. Terrain Analysis, Classification and Assessment for Road and Railway Location, Construction Material and Water Resources of the Carpentaria Area, Queensland. CSIRO Aust., Div. Appl. Geomech., Tech. Rep. No. 100, 214 pp. Grant, K. and Finlayson, A.A., 1977a. The Siting of Australian Cities and Towns. Transportation Conference, 1977. The Way Ahead - New Cities or Bigger Cities. Orange, 24-26 October. The Institution of Engineers (The Institution of Australia. Preprints of Papers, pp. 75-77. Engineers, Australia, National Publication No. 77/10). Grant, K. and Finlayson, A.A., 1977b. The Roads around Sydney - Past, Studies from The School of Civil Engineering Present and Future. UNICIV Report No. R-170, June. (University of New South Wales, Sydney) 39 pp. Grant, K. and Finlayson, A.A., 1977c. The PUCE System for Terrain Analysis, Assessment and Evaluation as used for Urban and Regional Development. Proceedings of 5th Australian Conference on Urban and Regional Planning Information Systems (URPIS FIVE), Canberra, pp. 6-70 to 6-71. Grant, K. and Finlayson, A.A., 1978a. The Assessment and Evaluation of Geotechnical Resources in Urban or Regional Environments. Engineering Geology, Vol. 12, No. 3, pp. 219-293. Grant, K. and Finlayson, A.A., 1978b. The Application of Terrain Analysis to Urban and Regional Planning. Third International Congress of the International Association of Engineering Geology. Madrid, Spain, September. Section 1, Vol. 1, pp. 79-91. Grant, K., Ferguson, T.G., Finlayson, A.A. and Richards, B.G., 1981. Terrain Analysis, Classification, Assessment and Evaluation for Regional Developmental Purposes of the Albury-Wodonga Area, New South Wales and Victoria. Volume 1: Terrain Analysis and Classification by K. Grant, T.G. Ferguson and A.A. Finlayson, 123 pp. Volume 2: Terrain Assessment and Evaluation by K. Grant, T.G. Ferguson, A.A. Finlayson and B.G. Richards, 103 pp. CSIRO Aust., Div. Appl. Geomech., Tech. Pap. No. 30.

Grant, K., Finlayson, A.A., Richards, B.G. and Pappin, J.W. l982. Terrain Analysis, Classification, Assessment and Evaluation for Regional Development Purposes of the Moreton Region, Queensland. Volume 1: Terrain Analysis and Classification by A.A. Finlayson and K. Grant, 336 pp. Volume 2: Terrain Assessment and Evaluation by J.W. Pappin, B.G. Richards, A.A. Finlayson and K. Grant, 203 pp. CSIRO Aust., Div. Appl. Geomech., Tech. Pap. No. 23. Grant, K., Finlayson, A.A., Spate, A.P. and Ferguson, T.G., 1979. Terrain Analysis and Classification for Engineering and Conservation Purposes of the Port Clinton Area, Queensland, including the Shoalwater Bay CSIRO Aust., Div. Appl. Geomech., Tech. Pap. Military Training Area. No. 29, 185 pp. Grant, K., Richards, B.G., Finlayson, A.A. and Pappin, J.W., 1975. Terrain Classification and Evaluation for Engineering Purposes of the r.t:>reton Region, Queensland (Part A). Volume 1 : Terrain Classification by K. Grant and A.A. Finlayson. Volume 2 : Terrain Evaluation by J.W. Pappin and B.G. Richards. CSIRO Aust. , Div. Appl. Geomech., Project Rep. No. 1 to the Cities Commission, Canberra and the Moreton Region Strategy Investigations, Brisbane (unpublished). Mitchell, c. W. , 1973. "Terrain Evaluation - An Introductory Handbook to the History, Principles and Methods of Practical Terrain Assessment". Longman, London, 221 pp. Ollier, C.D., 1977. Terrain Classification : Methods, Applications and In J. R. Hails (Ed.) "Appl i ed Geomorphology" , Elsevier Principles. Scientific Publishing Company, Amsterdam, pp. 277-316. Standards Association of Australia, 1977. Australian Standard 1289. Methods of Testing Soils for Engineering Purposes. Standards Association of Australia, Sydney. Stewart, G.A. (Ed.), 1968. "Land Evaluation", Macmillan, Melbourne, 392 pp. U. S . Department of the Interior, Bureau of Reclamn tion, 1974. Earth Manual •n d as Construction - A Guide to the Use of Soils as Foundation o n, U.S. Government i t Edi Materials for Hydraulic Structures. 2nd Printing Office, Washington, 810 pp. Wagner, A.A., 1957. The Use of the Unified Soil Classification System by the Bureau of Reclamation. Proceedings of the 4th International Conference on Soil Mechanics and Foundation Engineering, Vol. 1, pp. 125-134. Woloshin, A.J., 1969. Evaluation and Comparison of Terrain Classification Methods - Final Report, Geonautics, Inc., Falls Church, Virginia.

ENGINEERING GEOLOGY SPECIALIST GROUP PAPERS, SERIES NO. 3, AUGUST, 1984 CHANNEL CHANGES IN AUSTRALIAN RIVERS SINCE EUROPEAN SETTLEMENT Robin F. Warner Department of Geography University of Sydney, N.S.W., 2006 Australia

ABSTRACT Channel changes occur naturally, either as lateral meander migration or as responses to natural regime variations. The latter modify water and sediment discharges which influence channel size and shape. Catchment changes induced by human activities have similar effects in that they too modify channel-forming discharges. These include the effects of deforestation, grazing, cropping, urbanization and conservation, and are known as indirect modifications. Direct changes involve channel alterations such as dams, dredging, channelization and improvements. Such measures affect channel size, slope, alignment and so on, and these influence the conThis paper seeks to make distinctions tinuity of throughput discharges. between natural and human-induced changes, although in practice they are often indistinguishab le. General historical sequences are outlined to indicate the nature of likely changes since European settlement. Finally, two N.S.W. examples of channel changes are given: these are for the Georges and Nepean Rivers. In the former, a tidal-channel bed has been dredged for commercial sand, together with flood-plain areas which have now become tidal lakes. Increased channel capacities and a much-modified tidal prism have greatly influenced tidal flushing in this urban environment. In the latter, although sand and gravel dredging has been much more extensive below Penrith, the impacts of dams, weirs, subsequent sediment starvation and high, recent natural discharges have been responsible for significant changes in and below sandstone gorges. INTRODUCTION River channels are seldom stable for long. Even under relatively stable hydrologic conditions, they are in a form of dynamic equilibrium. In unstable regimes which are very common in Australia or those involving climatic change, channels adjust through changes in morphological dimensions to new runoff and sediment-yield conditions from their catchments. Human activity can also lead to catchment instability through alteration of the catchment's surface. Also by changing channel form, processes operating within the channel can be modified. This paper is concerned with discussion of river channel changes since European settlement in Australia. These involve not only those induced by human activity, but also the impacts of dynamic equilibrium and of secular changes in climate. The latter also accelerate geomorphic processes. Human impacts can be indirect and direct, with the former involving catchmentchanged hydrologic responses and the latter actual changes to channels. Some of these have been conscious endeavours, like channel realignment: others unconscious as with the impacts of dams and dredging. The latter usually occur up-and downstream of the change and involve both erosion and deposition.

Detailed historical sequences of channel changes have yet to be worked out. Thls is mainly because initial and subsequent data on channel morphologies were not collected. However, with more knowledge of the processes involved, some assessment of the sequence of changes may be possible, with knowledge of land clearance and use, climate, flooding and qualitative reports on changes. In some cases, local di mensional data were collected and these can be used in more quantitative assessments of changes and their causes. Examples of such works are summarised to show the nature and implications of change. NATURAL CHANNEL CHANGE Channel changes occur naturally in all fluvial systems. The most connnon change is in lateral migration of meanders involving erosion of the concave bank and accretion on the convex (Leopold et al., 1964). Even in the short term, this may mean 10m per year or more, where the channel is able to meander freely on alluvial plains. Generations of cutoffs on modern alluvium attest to these (Pickup et al., 1979). However, such migration is constrained where the channel is entrenched in bedrock and older alluvia. Natural changes also occur when the hydr ologic regime changes. This may involve short- or long-term increases or decreases in precipitation. Temperature changes can also alter the effec tiveness of precipitation. For instance, a colder Pleistocene climate lowers evapotranspiration losses and renders the rainfall more effective as runoff (Galloway, 1965). Such changes influence amounts of water and sediment reaching channels. The former influences the channel size, found by many to equate with the mean annual flood (Leopold et al., 1964; Dury, 1969; Woodyer, 1968), while the latter, in terms of sediment size, affects the shape (Schunnn, 1960). High silt-clay percentages in the perimeter cause deep narrow, sinuous channels and low percentages increase the width-depth ratio in straighter channels. Under stable regimes, there is assumed to be a basic harmony between size and shape and the catchment's hydrology. Thus channel morphology, involving width, depth, meander wavelengths, sinuosity, slope, sil t -clay percentages and so on is related to water and sediment discharges (Schunnn, 1971 and 1977). Such stability in Australian rivers may be less common than was thought, becaus e of the great variations in stream discharges (Pickup and Rieger, 1979; Abrahams and Cul l, 1979). Even precipitation in areas where seasons are marked, as in the monsoonal sunnners of the north and winter rain dominance in the south, is very variable. Thus it seems probable that morphological responses will be higher, with high flows in wet years giving more flooding and larger channels. These may suffer some decrease in drier years, given some sediment movement. "Noise" within the natural system also includes secular changes in climate. Increases from the late 1940s to the late 1970s have greatly altered channels (Pickup, 1976; Bell and Erskine, 1981; Erskine and Bell, 1982; Erskine and Melville, 1983; Warner, 1983 and 1984a). Other increases from the late 1860s to the early 1900s may also have had similar effects. HUMAN-MODIFIED PROCESSES AND FORMS Be f ore the arrival of Europeans, the only human-induced changes involved aboriginal burns. They used fire extensively and it has been argued (Hughes and Sullivan 1981; Sullivan and Hughes, 1983) that the first major changes may have been related to such events. Natural fires produced charcoal which has been used to date Holocene and Pleistocene deposits (Warner, 1970). European land-use practises altered catchments more permanently and they also had to build dams and weirs for water supplies (eg. Liverpool Weir, NSW, 1823). Thus catchment changes, involving deforestation, ringbarking, grazing,

TABLE 1 HYDROLOGIC 1 GEOMORPHIC AND MORPHOLOGIC CHANGES LIKELY TO OCCUR IN RESPONSE TO MAJOR CHANGES CAUSED BY THE LAND-USE CHANGES INDICATED Geomorphic Processes

Hydrologic Processes Pr Deforestation 1 Grazing

Cropping

Urbanization

Pere

GWF

Conservation 3 Mining Afforestation

w d L

Qp +

- +

RO +

Q +

0 4

Morphologic Change

+ 0

Water Harvesting 5

s m + +

F +

+ ±

+ +

0 = no change,+= increase, - = decrease Pr= precipitation, Ic = Interception, In= Infiltration, Pere= Percolation, If= Throughflow, GWF = Ground Water Flow or Base Flow, RO= Runoff, Q = Total Discharge, Qs = Total Sediment Discharge, Qp = Flood Peaks, W = Width, d = Depth, Lm = Meander Wavelength, s = Channel Slope, P = Sinuosity, F = Width-Depth Ratio, M = Weighted Silt-Clay Percentage. 1. 2. 3~ 4. 5.

particularly with overgrazing urban conditions are of high sediment yield in early construction stages and perhaps later. However with stable high runoff "concrete" environment, the Qs may decrease; this will caused, P to increase and Lm (if free) S, F and M to decrease. strip mining and severe disturbance some increase in Q and Qs have been noted where high runoff from roads and drainage ditches occurs Cumberland Plain (Crofts, Geddes and Carter,1963).

1-'-::ri-'S I-'-( en en 1-'-~n ::l'< Ill ro ::iroo c: ::llll11 0.11 cu n 111-'-11 n rto 8 11()Q::, 11 ro i-'H>l1 (t)'t) 1piro11'tl ro ooro 1-'111-'11elll>(Snll> ..,enQl0(t)OlQl::l "C0 ro::i::-:: ::J"OQn::lrortl-'o 1-'-::J"C: o.ro en en o.o. cu 11rtrtC" i-'nP>11> c:oroen::i :;:li-'rt o o. ro n 11 11 o 11> OQ cu ::, 1-'-::lrl>H>0::lrt>ll>QlO. OQC:1-'-C: ()0.1-'i-'• 0. cnroens11 eno cu ro ro O" en rt o 1-'0Q ::lQl (t)Q)O,.l-3(1)0( n o : : , en ::ro.11 cu 1-'-rt::l 1-'(t)t-'•ll>en::l ::TSO. rt'<N ,...._11SOQ Q)0 11 11 ::T ~ (t) (t) rt> 11 (t) t-'•11> ro11rort P>'tlt-'-C:H>::l n ::J"OQ 11 o r t ( o.n 1-'l-'Ql C: rt> 0.0 rt t-'•::T'< 11 roo.enn"Oen::l ronro .. (t)rt t-'11 11 ,-.nen110.l-'-l-'t-'•'tl o ::, n ::r n · ro en 11 H> cu ::r o. C" ::, n ::r rt o I-' 111l> t-'•n0\0::lll>C:'< enro::, nro0'--'::l11n ':Jleno,. .. (t)OQrl>::Trtll>OQ 1-'rl>O.P>0enrt> • en 1111roo. ro"O(len C" o. cu o. en I-' o H> ::r ,. Ill C: 0 t-'•'< (t)OQ 11rl>Ql'< t-'•i-'11::l t-'•::l O.EI n I-' .. «-'::, o. ... 0.11 F<-110 ::,en 0 O.• P>::l(t)(rtll> I-' OQ Ql ro 11 en n I-' t-'• P>0 en11P> en n 811C:OQ ll>l-:300.(t) 1-'-0.1-'-"0(t)ent,,'rtll> 1-'-C"::J"en .. ::ll-'O.rt>n H>rl>CO. ::ll-'Qln 11 OQ ro <: ::J"C"t-'•O.n ronro roP>c:1-' 11 rt rt cu o. ::r en cu ,-. Ill "O 11 ::l t-'• OQ I-' en en::, ro"O cu o.C"::lo '<::l::l oronenrort roen rt11 .. P>l-'-ro ::,;-o::lrtl-'rt> o.enro en ::in::r P>I-'t-'• 11ron 11oenH10 ::rrt 1111c roc:ro cuen11P>o nrto.ro .OH>rtrtrt>::l rt 1-'-n rtC 1-'rt> ro SOQl1 o ::ro 1-'-0 en o.o ro Ill <: en 1-'<: 0 • ::l Ill enP>• ::i ::in cu o. t-ti 01-' 0OQlii-' Ill t-t,rl)(t)rt>O.ZI-'enpi ::ll-'-01-'01-:31-' rt rt en rt en H> ::r I-' n ro n ::J"t-'•rt> I-'rt rt::J"0 ll>::l 11H>::l 0Ql::l ::, 111-'ll>rt>0";;l OQ OQ'<::lro11s ro o.ens .. ro ro en en 11 •

'° °'

......

°'

17 Direct changes involve waterway modificatio ns. These change conditions not only in the affected reach, but also up- and downstream. These include: dredging to improve navigation or to remove useful aggregate; dams for water storage, hydroelectr ic power, irrigation, flood mitigation and recreation; weirs to maintain riparian water; and channel ization to improve drainage. These affect the continuity of water and sedi ment movement and lead to channel changes by way of adjustment, as well as causing some environmen tal problems. Interbasin transfers along existing channels also promote changes (Upper Murray Advisory Committee, 1982). Table 2 summarizes some of the expected changes. TABLE 2

Dredging

UPSTREAM, AFFECTED REACH AND DOWNSTREAM MORPHOLOGIC AND PROCESS CHANGES ASSOCIATED WITH DIRECT CHANNEL MODIFICATIONS

CAP

+ +

Dams

+ +

Channelization

D2 Weirs

+ ±

Act like small dam, as local sediment traps

0 = no change; + = increase;

- = decrease

w = width, d = depth, V = mean velocity, A= channel area at bankfull, CAP = channel capacity at bankfull, s = channel slope, Q = total discharge, Qs = total sediment discharge and Qp = peak discharge.

U = upstream, A= affected reach, D = downstream, Dl = scour zone downstream, D = accreting reach. 2

Thus it is now possible to suggest what kind of response may follow changes in the catchment or to the channel. The rate o f adjustment is not always clear. Several decades may be involved and, in view of other changes both natural and human, it is not certain whether adjustments are ever completed. In any area the amount of change, its spatial and temporal occurrences and its impacts are unique. Therefore it is impossible to have a simple historical sequence impact model (Wolman, 1967) especially in a country like Australia with its diverse environmen ts, climates and settlement patterns.

18 HISTORICAL SEQUENCES Australia was settled over a long period of time after the development of Sydney Cove in 1788. The first channel changes were the tanks cut in the local Tank Stream for a local water supply. Pioneer fringes moved inland in different directions and at different rates from initial coastal settlements. The intensity of settlement, and therefore the level of catchment and channel modifications, was dependent on the natural environments, their resource bases, climates and proximity to natural and artificial routeways. There are major differences in occupance densities and therefore in potential human-induced channel changes. While most of the initial settlement took place in the 19th century, it was for different reasons in different places (Jeans, 1972). Coastal valleys were important for timber, grains and then dairy products. Drier tablelands were cleared by ringbarking and tallow, leather and wool were major products, prior to routeway development. On the wetter margins of the arid areas, grains were important, while elsewhere there was grazing. Gold rushes and rail links were often inducements to populate more distant areas. The impacts of all these on rivers have yet to be fully appreciated, and yet the impacts of alluvial gold digging and dredging are still visible. Catchments have also been modified at times by droughts, locusts, rabbits, feral animals, kangaroos and so on. Much of the early assessment on river changes has to be speculative because data on natural vegetation and channels were not recorded. It has not been possible to indulge in reconstruction like Knox (1977) found possible in Wisconsin. One historical source on channel~ known as chains of ponds, has been well documented (Eyles, 1977a, band c). Historical impacts in the Hunter were recorded by Holmes and Loughran (1976). Major impacts resulted from timber clearance and subsequent land use in catchments, which increased water and sediment discharges. These changed chains of ponds to gullies and caused bed accretion and bank erosion in coastal streams. Changes started in mainly drier periods (1800-1860 and 1900-1945) were probably accelerated in wetter regimes (1860-1900 and 1945-1978). The latter period can be documented from early surveys and air-photograph interpretation. The story of change is best unravelled where historical surveys can be found, often by accident. Surveys for water supplies and navigation have been used to assess changes in the Nepean from 1900 and 1911 (Warner, 1983), the Hawkesbury from 1871 (Dury, 1970) and Georges River from 1959 (Warner and Pickup, 1978). A history of flood mitigation and channel relocation has been presented for a Victorian stream by Bird (1980) and currently Erskine is investigating detailed changes in the Hunter (Bell and Erskine, 1981; Erskine and Bell, 1982). Such works, whilst not revealing the full historical sequences, are useful beginnings and tests of changes hypothesized by Schumm (1971) in his concept of river metamorphosis. EXAMPLES FROM THE GEORGES AND NEPEAN RIVERS In recent years channel changes have been investigated in these two New South Wales rivers. These are only two of many studies but they do indicate impacts of changes and problems that they present. The tidal channel of the Georges River extends from Liverpool to Botany Bay. This is the lower part of a 930 km2 catchment, half in sandstone and half in shale. The shale area mainly has been urbanized and is presently occupied by more than 500,000 people. Above Menai the channel floor has been dredged for sand and above Milperra flood plains have also been dredged, such that series of ponds are now connected to the uppertidal basin at Liverpool and Chipping Norton. A re-survey in the mid 1970s revealed increased channel dimensions in the shales upstream of East Hills and in the

19 dredged sandstone gorges above Menai. Below this there had been accretion, reducing the channel capacity and tidal prism. These changes not only affected the channel dimensions, but also tidal dynamics and flushing, the latter being ve3 important to sewage works at Liverpool. The removal of up to 10 mill m has increased the capacity of the channel, lowered the rate of flushing (twice as long under design conditions) in spite of an increase in the upper tidal prism, and reversed the trend of sedimentation associated with large-scale urbanization (Warner and Pickup, 1978; Warner, Maclean and Pickup, 1977). Flushing rates were also found to be very sluggish in a major dye experiment (Warner and Smith, 1979a and b). This channel was being affected by the natural increase in precipitation and runoff after the late 1940s, progressive urbanization with higher, more flashy runoff, sedimentation in tidal sinks (embayments and inlets) and the removal of large volumes of sand. The Nepean River above the Grose Junction is nontidal. It involves more than 11,000 km 2 above Penrith, mainly i n sandstone, with some shales on the Cumberland Plain, and a large area of granites and Lower Palaeozoic rocks in the upper part of the Warragamba catchment. This part of the catchment yields the gravels so important in the extractive industries between Penrith and Richmond. More than 40 mill m3 of sands and gravels have been removed from this system; it has been subject to dam building (4 on the Upper Nepean and the Warragamba cutting off coarse sediments from more than 80% of the catchment), to some catchment modification, and to secular climatic changes. There are also a series of riparian weirs to pond rural water from Penrith to Menangle. Surveys were found of the long profile in 1911, with the inclusion of the weir s, and of 46 cross sections in 1900 for the 5 km above Penrith. Resurvey has revealed changes to the alternating sandstone gorges and shale-alluvial reaches (Warner, 1983 and 1984b). Generally the gorge above Menangle has been scoured by high, post-late 1940s flows which have been devoid of sediments because of the Nepean dams. Similarly the Nepean Gorge at Penrith has been deepened over 16 km, in spite of gravel armouring of part of the bed, by high, postWarragamba flows. The shale reaches above Wallacia have tended to sand up because of the weirs and a progrannne of channel clearance has been necessary to restore weir capacities, to reduce bank damage and to lower the frequency of flooding. In thescourreach above Penrith, the proximity of Warragamba dam and the impacts of the Glenbrook delta (formed in 1943 by a flood) have been sufficient to increase channel capacity in all but the section just above the weir. Some 660,000 m3 have been lost from the 5 km since 1900 (Warner, 1984b). Normally downstream of the scour reach below a dam, the channel is expected to reduce capacity, because of lower water and sediment discharges. This has not happened in the Nepean because of recent higher flows, physical removal of sand and gravel and the presence of weirs. These help to promote bank erosion, as has been experienced along many reaches of the Nepean and Hawkesbury. CONCLUSIONS In this paper natural and the man-induced channel changes have been defined and discussed. Also it has been possible to consider historical sequences of changes that might have followed European settlement of diverse environments, largely in hypothetical terms. This was mainly because the impacts of land clearance, grazing, cropping, urbanization and so on in catchments and channels have not been recorded. Limited, specific studies have been possible in some instances where, for various reasons, detailed surveys were undertaken, have been found and replicated to work out net changes for the intervening periods. Examples of such work for the Georges and Nepean Rivers were given briefly. These revealed not only the nature and the amount of changes but also their impacts on processes, dynamics and management of these systems.

REFERENCES Abrahams, A.D. and Cull, R.F., 1979 - The formation of alluvial landforms along New South Wales coastal streams. Search, 10, 187-188. Bell, F.C. and Erskine, W.D., 1981 - Effects of recent increases in rainfall on floods and runoff in the Upper Hunter Valley. Search, 12, 92. Bird, J.F., 1980 - Geomorph ological implicati ons of flood control measures. Australia n Geograph ical Studies, 18, 169-183. Crofts, F.C., Geddes, H.J. and Carter, D.G., 1963 - Water harvestin g and planned pasture productio n at Badgery's Creek. Universit y of Sydney, School of Agricultu re Report No. 6. Dury, G.H., 1969 - Hydraulic geometry. In Chorley, R.J. (ed.), Water, Earth and Man, Methuen, London, 319-330. Dury, G.H., 1979 - A re-survey of part of the Hawkesbur y River, N.S.W., after 100 years. Australia n Geograph ical Studies, 8, 121-132. Erskine, W.D. and Bell, F.C., 1982 - Rainfall, floods and river channel changes in the Upper Hunter. Australia n Geograph ical Studies, 20, 183-196. Erskine, W.D. and Melville, M.D., 1983 - Impact of the 1978 floods on the channel of the Lower Macdonald River, N.S.W. Australia n Geographe r, 15, 284-292. Eyles, R.J., 1977a - Erosion and landuse in the Burra catchment , Queenbeyan. Journal of Soil Conserva tion, N.S.W., 33, 45-59. Eyles, R.J., 1977b - Birchams Creek: the transitio n from a chain of ponds to a gully. Australia n Geograph ical Studies, 15, 145-156. Eyles, R.J., 1977c - Changes in drainage networks since 1820, Southern Tableland s, N.S.W. Australia n Geographe r, 13, 377-386. Galloway, R.W., 1965 - Late Quaternar y climates in Australia . Journal of Geology, 73, 603-618. Hughes, P.J. and Sullivan, M.E., 1981 - Aborigina l burning and Late Holocene geomorphi c events in eastern N.S.W. Search, 12, 277-278. Holmes, J.H. and Loughran, R.J., 1976 - Man's impact on a river system: the Hunter Valley. In Holmes, J.H. (ed.) Man and the Environme nt: Regional Perspecti ves, Longmans, Melbourne , 96-114. Jeans, D.N., 1972 - An Historica l Geography of New South Wales to 1901. Reed, Sydney. Knox, J.C., 1977 - Human impacts on Wisconsin stream channels. Annals of the Associati on of American Geograph ers, 67, 323-342. Leopold, L.B., Wolman, M.G. and Miller, J.P., 1964 - Fluvial processes in Geomorph ology, Freeman, San Francisco . Pickup, G., 1976 - Geomorphi c effects of changes in river runoff, Cumberlan d Basin, N.S.W. Australia n Geograph er, 13, 188-193. Pickup, G. and Rieger, W.A., 1979 - A conceptua l model of the relations hip between channel characte ristics and discharge . Earth Surface Processes , 4, 37-42. Pickup, G., Higgins, R.J. and Warner, R.F., 1979 - Impacts of waste rock disposal from the proposed Ok Tedi mine on sedimenta tion processes in the Fly River and its tributari es. Departmen t of Minerals and Energy and Office of Environme nt and Conserva tion, Papua New Guinea. Schumm, S.A., 1960 - The shape of alluvial channels in relation to sediment type. United States Geologica l Survey, Professio nal Paper 352-B, 17-30. Schumm, S.A.,' 1971 - Fluvial geomorpho logy: the historica l perspecti ve. In Shen, H.W. (ed.) River Mechanics ,H.W. Shen, Fort Collins 4-1 to 4-30. Schlllilm, S.A., 1977 - The Fluvial System, Wiley, New York. Sullivan, M.E. and Hughes, P.J., 1983 - The geoarchae ology of the Sydney Basin sandstone s. In Young, R.W. and Nanson, G.C., (eds.) Aspects of Australia n Sandstone Landscape s. Australia n and New Zealand Geomorpholog y Group Special Publicati on No. 1, 120-126. Upper Murray Advisory Committee , 1982 - Effects of the Snowy Mountain Scheme on the Upper Murray River. Report by this committee . 1

21 Warner, R.F., 1970 - Radio-carbon dates for some fluvial and colluvial deposits in the Bellingen Valleys, N.S.W. Australian Journal of Science 32, 368-369. Warner, R.F., 1983 - Channel changes in the sandstone and shale reaches of the Nepean River, N.S.W. In Young, R.W. and Nanson, G.C., (eds.), Aspects of Australian Sandstone Landscapes, Australian and New Zealand Geomorphology Group Special Publication No. 1, 106-119. Warner, R.F., 1984a - Man's impacts on Australian drainage systems. Australian Geographer, (in press). Warner, R.F., 1984b - Impacts of dams, weirs, dredging and climatic changes: an example from the Nepean River, N.S.W., Australia. Paper for 25th International Geographical Union Congress, Paris. Warner, R.F. and Pickup, G., 1978 - Channel changes in the Georges River between 1959 and 1973/76 and their implications. Report published by Bankstown Municipal Council, N.S.W. Warner, R.F. and Smith, D.I., 1979a - The use of fluorometric tracers to study effluent movement in the Georges River, N.S.W. Australia. Proceedings 10th New Zealand Geographical Conference and 49th ANZAAS Congress, Auckland, 48-53. Warner, R.F. and Smith, D.I., 1979b - The simulation of effluent movement in the tidal Georges River, N.S.W., Australia, using the dye tracer, Rhoda.mine W.T. Report for N.S.W. State Pollution Control Commission, Published by Bankstown Municipal Council. Warner, R.F., Maclean, E.J. and Pickup, G., 1976 - Changes in an urban water resource: an example from Sydney, Australia. Earth Surface Processes, 2, 29-38. Wolman, M.A., 1967 - A cycle of sedimentation and erosion in urban river channels. Geografiska Annaler, 49A, 385-395. Woodyer, K.D., 1968 - Bankfull frequency in rivers. Journal of Hydrology, 6, 114-142.

L. L.

PART2 PAPERS FROM THE 7th AUSTRALIAN GEOLOGICAL CONVENTION, SYDNEY 1984.

ENGINEERING GEOLOGY SPECIALIST GROUP PAPERS SERIES NO. 3, AUGUST 1984

THE FORMATION OF TALC FROM MICA EFFECTED BY THE BREAKDOWN OF PYRRHOTITE* W.F. Cole and C.J. Lancucki CSIRO Division of Building Research Graham Road, Highett, Victoria, 3190 AUSTRALIA

ABSTRACT At Renison Bell, Tasmania, the mine waste from the treatment of the tin-containing metamorphosed sediment contains appreciable amounts of pyrrhotite which readily breaks down in stockpile. The unaltered mine waste is composed of mica, chlorite, quartz, dolomite and pyrrhotite. After alteration, the pyrrhotite is lost, mica and chlorite are reduced in amount, and goethite, gypsum and talc appear. The talc is probably formed at the expense of the mica. The chemical (or mineralogical) changes in the stockpile have been duplicated in the l aboratory by subjecting fresh mine waste to a weekly cycle of wetting and then drying at S0°C for several months, after which time a white layer of talc and gypsum coated the rock particles. The mechanism of the alteratio~ is discussed. Comment is made on the suitability of the mine waste as an ag gregate for use in concrete. Keywords: Mica alteration, Talc f ormation, Pyrrhotite, Renison Bell, Tasmania, Aggregates, Concrete * Originally presented a t the Poster Section of the VI International Clay Conference 1978 held at Oxford, England.

INTRODUCTION At Renison Bell, Tasman ia, (Fie. 1) rocks of the Dundas group (midupper Cambria n) and Crimson Creek Group (lower Cambri an) unconfo rmably overlie a Precam brian sequenc e (Newnham, 1975). The Crimson Creek formati on consis ts of a conform able success ion of shales , quartz ites, Lower argilli tes, carbon ates, hemati tic conglom erates and pyrocl astics. lying nts sedime and ics volcan tic hemati beds compris e a Red Rock Member of the are found beds lowest The beds. ate conform ably between two carbon age, Renison Bell Shale and the Dalcoa th Quartz ite of Younge r Precam brian es, uartzit orthoq d -bedde thickly nes, which consist of shales, siltsto dyke carbon ate rock beds and minor pebble conglom erates. A Devonia n quartz ore with us poraneo contem be to red conside is and e intrude s into the sequenc deposi tion.

TAS MAN IA

Fig. 1

Genera l locatio n map.

Minera lizatio n of the sedime nts was probab ly due to a Devonia n granite source at depth, from solutio ns which rose along fault zones and spread lly latera lly into the carbon ate beds, minera lizing them and metaso matica s sulfide but rite cassite is ed replaci ng them. The only ore minera l extract of t consis They ). (40-70% compri se a high propor tion of the rock te, pyrrho tite, pyrite, arseno pyrite and chalco pyrite with traces of stanni are ls minera Gangue work). ished galena and sphale rite (W. Fander , unpubl e. tourma line, quartz, carbon ate, talc, fluorit e, apatite and chlorit is After the tin is extract ed from the ore body, the waste mine metal as used be could that sizes e discard ed on stockp iles in a range of particl colour changes al materi iled stockp aggreg ate for concre te. Howeve r, fresh al after a few months from grey to brown and, in the proces s, the materi the s, reason these For sulfur. of become s warm to the touch and smells opinion CSIRO Divisio n of Buildin g Researc h was approac hed in 1977 for an Bell Renison from al materi waste the as to whethe r it would be safe to use ent subsequ the of results in concre te. This paper reports the investi gation .

25 EXAMINATION OF UNALTERED STONES A selection was made of all rock types present in a fresh stockpile at Renison Bell, and individual stones were examined by X-ray diffraction methods to establish their mineralogy. At the same time, some were examined in a scanning electron microscope with an energy dispersive analyser to evaluate their chemical composition. The results are summarized in Table 1. A sample representation of the fresh stockpile was then crushed and milled to pass a 200 mesh screen. It gave the X-ray pattern shown in Fig. 2 which indicates a mixture of quartz, mica, chlorite and dravite (tourmaline). This mixture differs from those shown in Table 1 in not containing talc, dolomite, siderite or pyrrhotite, presumably because stones containing these minerals were not present in sufficient amounts for them to be detected by the X-ray method. Table 1: Mineralogical composition of stones from fresh stockpile Carbonate

Mica

Stone type

N µ

•,-l

0 cJ

;:,

Grey-green, hard Dark-green, soft Grey Shiny White White with grey Cream

CIJ

0 .-i

0 .-i 0

cJ .-i

E--1

..c: u

p p

"M C/)

"M µ 0

..c:

H H :>, p...

p p

0 00 0 .-i

..c: p...

µ Q)

p p

p p p p

P = present

a = Quartz M = Mica C = Chlorite D = Dravite

FRESH a

♦

a a

28 CuKa

Fig. 2

X-ray diffraction pattern of aggregate from fresh stockpile.

EXAMINATION OF ALTERED STONES Stones from an old stockpile at Renison Bell which had been severely altered were also taken and the surface coating on the stones removed . This gave the X-ray pattern shown in Fig. 3 which indicates a mixture of talc, quartz, mica, chlorite, lepidocro cite and goethite. Fines ob t ained by sieving old stockpile material through a 200 mesh screen had gypsum in addition. The X-ray pattern of stones freed from their surface coating was essential ly the same as that of fresh material (Fig. 2).

T = Talc

a = Quartz H = Hic a L = Lepidocrocit e G = Goethite ( = Chlorite

SURFACE COAT a ♦

28 [uK 0

Fig. 3

X-ray diffractio n pattern of surface coatings on aggregate from weathered stockpile .

LABORATORY ALTERATION Some fresh stones were cycled twice weekly for a period of four months from a fog room to a room at S0°C and 17% minimum relative humidity in which time a white crust formed on the stones. The c rust had the compositi on at room temperatu re of gypsum, MgS04.6H2 0, talc and a small amount of mica. Other stones containin g large amount of pyrrhotit e, when subjected to cycles of wetting and drying in a weatherom eter, broke down in a few days to a brown rusty mixture of pyrrhotit e and poorly crystalli ne goethite. Equally dramatic changes took place in the powder of the milled fresh stones (used to obtain the X-ray patterns of Fig. 2) after 9 months standing in a constant temperatu re (20°C) and humidity (50%) room . The Xray pattern of the stored material (Fig. 4) showed, compared with the X- ray pattern of the original material (Fig. 2), the generatio n of talc and a correspon ding reduction in mica with no appreciab le change in the quartz, chlorite and dravite contents. The grey colour of the sample had not changed. DISCUSSION The oxidation of sulfide minerals pyrite, chalcopy rite and pyrrhotit e has been studied by Steger and Desjardin s (1978) and pyrrhotit e alone by Steger (1981). Pyrrhotit e is the most unstable of the sulfide minerals and

T = Tal c

a= Quartz H = Hica [ = Chlorite D = Dravite

ALTERED

Q ♦

a ♦

28 [uK 0

Fig. 4

X-ray diffraction pattern of aggregate from fresh stockpile (Fig. 1) after standing in a constant temperature and humidity room for nine months.

quickly oxidizes to goethite and elemental sulfur. There may be a sequence of intermediate reactions that give FeS04 and/or Fe(OH)(S04).xH 2o. In the oxidation of pyrrhotite at Renison Bell although sulfur is smelt it is not recognized in the X-ray pattern of the alteration products so presumably it is oxidized in the presence of water to sulfuric acid. The latter forms gypsum and MgS04.6H20 by reaction with the dolomite of the ore body. The direct oxidation of the pyrrhotite forms coatings of goethite and lepidocrocite on the surface of the stones from the altered stockpile. The appearance of talc at the expense of the mica (probably phlogopite) in the alteration process is unusual but by no means unique. Rimsaite (1975) has reported the natural alteration of phlo gopite-vermicu lite intergrowths to talc. The vermiculitized phlogopite loses its interlayer cations, tetrahedral and octahedral Al, Ti and part of Fe as well as F and Cl ions from the OH.0-sheets. The alteration processes are accelerated in the presence of sulfide minerals. The reactions taking place in the stockpile at Renison Bell can be summarized as follows: 1.

Pyrrhotite (FenSn + 1) + 02 +H2o ➔ Goethite and/or Lepidocrocite (FeO(OH)) + S.

2S + 30 2 + 2H20 ➔ 2H2S04.

Dolomite (CaMg(C03) 2 ) + 2H2S04 + 6H20 ➔ Gypsum (CaS04.2H20) + MgS04.6H20 + 2C02.

Phlogopite Mica (KMg3(AlSi3 010) (OH) 2 ) + H2 S04 ➔ Talc (Mg3(Si 401o)(OH) 2 ) + K,Al sulphates.

The performance of the Renis on Bell aggregate in concrete was not tested in the investigation, although it was strongly recommened that this should be done if it were to be used for this purpose. This recommendation

was based on an investigation, being carried out at that time by the CSIRO Division of Building Research, into the warping of precast concrete flagstones laid on a concrete bed at Parkes Place, Canberra. The warping (D. Beresford, personal communication) was attributed to the adsorption of soluble sulfate salts into the flagstones from below, combining with cement to form an expansive compound, calcium sulfoaluminate (ettringite). The sulfates were contained in the concrete bed and a most likely origin was the coarse aggregate of this member, as the quarry which supplied this aggregate had a recorded history of sulfate effects from the oxidation of pyrite when its products were used in road construction. Another effect that could cause trouble in concrete made with aggregate containing iron sulfide minerals is the heat evolved during the oxidation processes. Roper and Taplin (1965) have shown that a quartzite from Tasmania, containing on the average 1.5 per cent by weigh t of the rock of sulfides, produced concrete that had an adiabatic temperature rise onethird higher than anticipated. This could produce a dangerous source of heat in mass concrete construction. CONCLUSIONS Stockpiles of mine waste at Renison Bell, Tasmania, that contain the iron sulfide mineral pyrrhotite are too unstable to be used for aggregate for concrete. The pyrrhotite, when wet, oxidizes to goethite and/or lepidocrocite and sulfur. The sulfur is further oxidized, in the presence of water, to produce sulfuric acid which reacts with dolomite to produce gypsum and MgSO4.6H 2o, and with mica to form talc. REFERENCES Newnham, L.A. (1975) - Renison Bell Tin Field. An economic geology of Australia and Papua New Guinea. 1. Metals (Ed. C.L. Knight). The Australian Institute of Mining and Metallurgy, Parkville, Victoria, 1975, 581-583. Rimsaite, J. (1975) - Natural alteration of mica and reactions between released ions in mineral deposits. Clays and Clay Minerals, 23, 247255. Roper, H. and Taplin, J.H. (1965) - The potential contribution of sulfide minerals to heat evolution in concrete. Constructional Review, 38, 22-25. Steger, H.F. (1982) - Oxidation of Sulfide Minerals. VII:Effect of temperature and relative humidity on the oxidation of Pyrrhotite. Chemical Geology, 35, 281-295. Steger, H.F. and Desjardins, L.E. (1978) - Oxidation of Sulfide Minerals. 4: Pyrite, Chalcopyrite and Pyrrhotite. Chemical Geology 23, 225-237.

ENGINEERING GEOLOGY SPECIALIST GROUP PAPERS SERIES No. 3, AUGUST 1984

RAPID DECAY OF SLATE ROOFING TILES IN SERVICE

A. Shayan and C.J. Lancucki CSIRO Division of Building Research P.O. Box 56, Highett, Victoria, 3190 AUSTRALIA

ABSTRACT It has been found that slate*roofing tiles containing pyrite (8.3%) have undergone a rapid decay after only two years of service in a suburb of Melbourne. The cause of the decay is the oxidation of pyrite and the production of highly hydrated iron sulfates. As well, Fe-rich chlorite contained in the slate is attacked by the very acidic environment resulting from the oxidation of the pyrite, and produces additional highly hydrated sulfates of iron and aluminium. The minerals identified by X-ray diffraction in the decayed tiles, in addition to the original quartz, muscovite mica and iron-rich chlorite, are hydronium jarosite Fe3(S04)(0H)5.2 Hz0, melanterite Feso 4 .7Hz0, and halotrichite FeAlz(S04)4.22H zO. Scanning electron microscopy and energy dispersive Xray analysis were used to confirm the presence of hydronium jarosite rather than potassium jarosite, through the lack of Kin the reaction product. This indicated that the muscovite was probably unattacked by the pyrite oxidation. A milder decay in another slate tile is thought to have been due to the oxidation of small amounts of pyrrhotite, leading to blistering and fretting of the slate. Formation of goethite, gypsum and hydronium jarosite were associated with this form of decay. Great increases in the molar volume of the reaction products relative to that of the original components are responsible for the disintegration of the slate. - This work shows that slate intended for use in buildings must be free of pyrite. Keywords: Slate decay, Pyrite, Hydronium jarosite, Rozenite, Rhomboclase, Melanterite, Holotrichite, Gypsum.

*The geological origin of the slate is unknown.

INTRODUCTION Slate roofing tiles vary in quality and performance depending on t he origin and mineralogic al composition of the rock. Typical slates are usually composed of lenticular crystals of quartz, muscovite and chlorite, which are elongated parallel to the cleavage plane of the rock . Such rocks , if not from a fault zone, are usually sound and resistant to weathering . Behre, Jr. (1931) mentioned that some Amer i can roofing slates have been exposed successfull y for 190 years, and that some old European slates are about 650 years old. However, when slates contain deleterious minerals such as carbonates and pyrite, chemical changes can occur that result in the disintegrat ion of the rock. An examp l e of such a disruption is the interaction of sulfate ions released from pyrite (or from atmospheric S02) wi th col cite to form gypsum. The l ar ger molar volume required for the <'rys tall iz ;1tlon of t he l.1t te r hrin ~s ;1ho 11l

disrupt i vl' pre ssure whi c h

eventua lly results in the breakdown o f the slate. Tests for weather resistance of slates, e.g., ASTM Designation C217-58 (reapproved 1970), are based on the effect of such mechanisms on the softening of the slate. Although chemical changes occurring in slate are regarded as slow processes, a rapid decay of slate roofing t iles has been observed in a suburb of Melbourne, resulting in the disintegrat ion of the tiles after only two years of service. In this paper, the reactions causing the decay are discussed and the reaction products identified. OPTICAL DESCRIPTION OF THE SLATE TILES Two tiles representin g the severely decayed (No. 1) and the blistered and fretted (No. 2) tiles were provided by a consulting agency for examination . The origin of the slate tiles is unknown. Tile No. 1 was moderately dark grey, appeared uniform in composition , and showed extensive softenin g and disintegrat ion on the upper surface. On the lower surface, which had been pro te c ted fr om the r a in, a thi ck globular salt incrustatio n was evident. Tile No. 2 was not uniform and had both a thick (2 mm) and a thin (0.5 mm) hematite-ri ch layer. Although only a few small softened areas containing some salt crystals were present, extensive blistering, fretting and staining were observed on this tile. The elongated blistering occurred along parallel lines in the cleavage plane. While some blisters were still closed, others that had been opened by exfoliation had a distinct ironoxide staining in the opening. Freshly opened cleavages in tile No. 2 showed eye-shaped regions which had a black material among coarse quartz grains (up to 0.5 mm). Where the centre was stained with iron oxide the two ends of the eye contained the black material. This feature was much less obvious and on a smaller scale in tile No. l; instead, particles of pyrite were discernible on the fracture surfaces of its unaltered portions . These differences between the two tiles were also evident in thin sections cut parallel and perpendicul ar to the cleavage planes. In tile No. 1 many very small eye-shaped regions were observed with one or more grains of pyrite included within the quartz aggregation . These regions were also arranged along parallel lines, in a fine matrix of mica and chlorite. Pyrite particles, up to 0.75 mm, were present either singly or in large clusters. In the thin sections of tile No. 2 the eye-shaped regions were larger, less frequent, and with coarser quartz grains. Moreover, fine reddish iron-oxide particles were dispersed throughout this slate and in some regions formed large stained patches. The fine matrix of tile No. 2 consisted of quartz, mica and chlorite.

X-RAY DIFFRACTION (XRD) STUDY The minerals identifie d by XRD in various portions of each tile are given in Table 1. Table 1: Minerals Present in Various Portions of the Slate Tiles

TILE NO. 2

TILE NO. 1 Portion

Minerals present

Portion

Minerals present

Fresh

Q, Ch, M, P, F, 9 .6 A

Fresh

Q , M, Ch , Hern, Go, F, 9.6 A, Ka

Surface stain

Q, M, Ch, Hj' F, 9.6 !

Upper layer

Q, M, Ch, Gy, Hem, Go, F, 9.6 A, Ka

Globular salt crust

Hal

Middle reddish layer

Q, M, Hem, Go

Loose flakes underside of tile

Q, M, Ch, Hj,

Lower layer

Q, M, Ch, Gy, Hy, Go, Hern, 9.6 A, F, Ka

Softened parts on lower surface

Q, Gy, M, Ch, Hj, F, 9.6 A

Transluce nt crust on lower face

Reddish stains in fretted areas

Q, M, Go, Hem, Ch

Earthy black material in eyeshaped regions within the tile

Q, elemental S, M

White spots in the black area

Mel

mica (rnuscovit e); Ch = iron-rich chlorite; P = pyrite; F = feldspar; Hj - hydronium jarosite, Fe3(SO4)2 (OH)5.2H2 O; Hal= halotrich ite, FeAl2(so 4 )4.22H2O; Mel= rnelanter ite, FeSO4.7H2O; Hern hematite; Go= goethite; Gy = gypsum; Ka= kaolinite ; 9.6 A= probably a partially stripped mica. The order of listing is with decreasin g amounts.

Q = quartz; M

Hydronium jarosite was more abundant in the decayed regions of tile No. 1 than in those of tile No. 2 while goethite, hematite and gypsum were found only in tile No. 2. Other specific differenc es are the presence of halotrich ite and melanter ite in tile No. 1 and elemental Sin tile No. 2. As discussed later these differenc es may indicate that either the mechanism of decay or its severity is different from one tile to the other. The 9.6 A phase listed in Table 1 was observed in most samples, but its amount was insuffici ent to produce a reasonabl e diffractio n pattern. It may be a mica partially stripped of the interlaye r cations, but it does not appear to be a product of the decay since it is also present in the fresh portions.

32 SEM AND EDS INVESTIGATION Qualitative energy dispersive X-ray analysis (EDS), using a scanning electron microscope (SEM), helped with the identification of the reaction products in the slate. Although the jarosite lines in the XRD patterns were somewhat closer to those of the hydronium jarosite Fe3(S04)2(0H)5.2H 2o than to those of the potassium jarosite K Fe3(S04)2(0H)6, the presence of the former was confirmed by EDS analysis of the reaction material which showed no Kin the spectra (Fig. 1 A,B).

K Ca

i- - - - _ )

~----: I

K,Kn

[

B A

keV Figure 1. Energy dispersive spectra of various components of slate tile No. 1, and the reaction products formed due to the oxidation of pyrite. A and B for hydronium jarosite, C for chlorite, D for mica, E and F for melanterite and hydronium jarosite coating partially altered chlorite, G for halotrichite. All lines are the Ka line of the elements shown, except for Kand Fe, where Ks line occurs close to Ka line.

The Si, Al and Mg shown in these spectra are impurities due to other minerals such as chlorite which could have been attacked during the decay processes. Chlorite was found to be the only Mg-bearing mineral in the slate (Fig. lC), while muscovite was the K-bearing phase (Fig. lD). The softened parts of the underside of tile No. 1 showed stacks of clay material which appeared to have been chemically attacked and were partially coated with fine materials with the composition shown in Fig. lE, F. This composition could be due to the hydronium jarosite and melanterite with some gypsum, although the latter was not identified by XRD (Table 1). In experiments where fresh slate powder was mixed with additional ground pyrite and alternately wetted and dried, a considerable amount of gypsum was formed around the walls of the reaction dish as well as on the surface of the powder. This observation showed that gypsum could be a reaction product in slate No. 1, but was probably too mobile to remain where it was formed. The proportions of Mg and Al relative to Si are much less in Fig. lF than in Fig. lE probably indicating a more severe attack on chlorite in the former. Slight impurities of Na and Pare from the slate matrix. The globular salt incrustation formed on the underside of tile No. 1 produced a complex XRD pattern which was identified as halotrichite FeA1 2 (SO4)4.22H2O after the EDS gave its composition as shown in Fig. lG. It is likely that Al was derived from chlorite, in which case the Si would be in some amorphous form and the Mg would probably exist as a sulfate. The very small amount of Ca could be present as gypsum. Very little, if any, potassium was present in any of the reaction products, which indicates that mica was not attacked during this process. MECHANISM OF DECAY Although pyrite was identified in tile No. 1, and the decay products could be attributed to the reaction between the slate components and the oxidation products of pyrite, the latter could not be found in any portion of tile No. 2. Instead elemental S was abundant within the eye-shaped regions accompanied by goethite. Steger and Desjardins (1978) and Steger (1982) in a detailed study of the oxidation of pyrite and pyrrhotite found that the former produces, initially, ferrous thiosulfate which in turn oxidizes to ferric sulfate, while the latter undergoes a much more rapid oxidation to goethite and elemental sulfur. It is very likely, therefore, that tile No. 2 contained pyrrhotite in the eye-shaped regions, all of which had oxidized after 2 years of exposure. Hydrolysis and oxidation of the elemental sulfur yields an acidic environment with mobile sulfate ions according to the reaction:

Gypsum was formed when soz- ions reacted with Ca released from the slate components (e.g., feldspar). Gypsum was present in the form of very thin crusts on the underside of the slate and within the neighbourhood of the eye-shaped regions. Both the iron-rich chlorite and the iron oxides could have been the source of Fe for the formation of the hydronium jarosite, which forms readily under acidic conditions. Fretting and exfoliation occur because the much greater molar volumes of goethite, gypsum and jarosite relative to the pyrrhotite, introduce expansion forces in the reaction zones. The matrix already softened due to acid dissolution facilitates the fretting. The pyrite content of the unreacted portions of tile No. 1 was 8.3% as determined by chemical analysis. The melanterite and possibly the hydronium jarosite (Table 1) found in the reacted zones could have formed directly from the oxidation of the pyrite. Mapstone (1954), Penner et al.

34 (1970), and Steger and Desjardins (1978) all describe reactions which yield ferrous and ferric sulfates as a result of the oxidation of pyrite. In fact, large crystals of pyrite stored in jars developed a white powdery coating which was identified by XRD as rozenite, FeS04.4H20. The same pyrite crystals ground to <75 µm and stored at ambient temperature and humidity developed another sulfate phase which was shown by XRD to be rhomboclase FeH(S04)2.4H 2o. This initial phase is expected to be metastable and would finally convert to ferrous or ferric sulfate. Penner et al. (1970) reported that autotrophic bacterial oxidation of pyrite in a calcareous shale resulted in the formation of gypsum and potassium jarosite and caused heaving and disintegration of the shale. Similar processes are responsible for the disintegration of buried concrete sewer pipes, where sulfuric acid generated by the biogenic oxidation of iron-sulfide attacks the calcium silicates and calcium aluminates (Drew and Reilly, 1977; Thornton, 1978). The formation of halotrichite FeAl2(so 4 ) 4 .22tt 2o requires a considerable amount of Al in addition to the oxidation products of pyrite. If the Al had come from the degradation of mica, then the K would have been found among the reaction products and probably K-jarosite would have formed. Direct observation of the mica in the SEM did not show signs of degradation. The iron-rich chlorite on the other hand is very susceptible to acid attack, and SEM and EDS indicated some disintegration of the chlorite. The general structural formula for iron-rich chlorite, e.g., berthierine, is [(Fe 2+, Mg)J-x (Fe3+, Al)x (Si2-xAlx)05(0H)4], the decomposition of which, through H2S04 attack, would provide all the components required for the formation of halotrichite. As shown by Fig. lG, the Mg and Si would result in some forms of impurity. The stoichiometry of such a reaction cannot be determined since the reacting components and the products are not determined quantitatively. The much more severely decayed tile (No. 1) suffered a much greater increase in the molar volumes of the reaction products since the hydration water is 22 molecules for halotrichite and seven molecules for melanterite, and the amount of the hydronium jarosite was much greater in this tile than in tile No. 2. Moreover, the rock matrix would have been attacked and softened to a greater extent because of the larger amounts of H2S04 released by the oxidation of the much greater amounts of pyrite in this tile. It is not clear whether the oxidation of pyrite and/or pyrrhotite was facilitated by any bacterial action. CONCLUSIONS The oxidation of pyrite has resulted in a severe decomposition of slate roofing tiles. The mechanism of decay is the formation of highly hydrated sulfates which were produced as a result of direct oxidation of pyrite and because the H2S04 released by the oxidation attacked the iron-rich chlorite, forming additional highly hydrated sulfates. The weakened matrix after chlorite decompositidn disintegrated due to the pressures introduced by the volume increase of the reaction products relative to the original components. The oxidation of small amounts of iron sulfide, probably pyrrhotite, in another tile resulted in fretting mainly due to the formation of goethite as a result of the oxidation of the sulfide. The fretting was enhanced by the formation of gypsum when H2S04 reacted with Ca-bearing phases. Some disintegration occurred as a result of the formation of gypsum and hydronium jarosite in regions other than the fretted ones.

35 This rapid decay indicates that slates intended for building purposes must be free of pyrite and other sulfide minerals. REFERENCES American Society for Testing and Materials, Designation C217-58 (Reapproved 1970). Standard method of test for weather resistance of natural slate. Behre, Jr., C.H., (1931). The weathering of slate. Symposium on Weathering of Masonry Materials. American Society ,for Testing and Materials 31 (11), 768-774. Drew, W.M. and Reilly, L.A., (1977). Some chemical and biochemical aspects in the design of large diameter effluent pipelines. In symposium: The chemistry of waste water treatment. 48th ANZMS Congress, September 1977, Melbourne, Australia. Mapstone, G.E., (1954). No. 20, 577-578.

The weathering of pyrite.

Chemistry and Industry

Penner, E., Gillott, J.E., and Eden, W.J., (1970). Investigatio n of heave in Billings Shale by mineralogic al and biochemical methods. Canadian Geotechnica l Journal 7, 333-338. Steger, H.F. and Desjardins, L.E., (1978). Oxidation of sulfide minerals. 4: Pyrite, chalcopyrit e and pyrrhotite. Chemical Geology 23, 225-237. Steger, H.F., (1982). Oxidation of sulfide minerals. VII: Effect of temperature and relative humidity on the oxidation of pyrrhotite. Chemical Geology 35, 281-295. Thornton, Jr., H.T., (1978). Acid attack of concrete caused by sulfur bacteria action. American Concrete Institute Journal 75, 577-584.

36 ENGINEERING GEOLOGY SPECIALIST GROUP PAPERS SERIES NO. 3, AUGUST 1984

WATER RESOURCES MANAGEMENT IN MINING - EXPERIENCES OF THE BHP COMPANY R.J. Frost, BE, FIE Aust Civil Engineering Manager, BHP Engineering P.O. Box 1237, North Sy dney NSW 2060 AUSTRALIA

ABSTRACT Mining in Australia is generally characteris ed by the remoteness and the aridity of the mine site. Yet despit e the aridity of much of Australia it is regularly subjected to flooding of considerabl e severity. Adding to these problems of the mining companie s is the o f ten poor quality of the groundwater encountered in mining. The summary of all these circumstanc es is that water must be considered a resource which often must be as carefully husbanded as the ore resource itself. Furthermore , the proper management of water, both on the supply side and on the usage side of the water balance equation is essential. The BHP Company operates or is partner in a diverse range of mines throughout Australia including iron ore, manganese, coal and bauxite. The water management situations and problems at these mines are typical of other Australian mines in similar areas. This paper describes the varied experiences of the BHP Company in this regard. INTRODUCTION Water is man's most essential need; he usually dies of thirst in the desert, rather than from lack of food or shelter. The ancient chemists considered water as one of the four basic elements; additionall y the so-called water signs comprise one quarter of the astrologica l zodiac. The mining industry is as equally concerned with and dependant upon water. This is particularl y so in the arid and semi-arid parts of Australia where significant mining activity is concentrate d. The Broken Hill Proprietary Company Ltd. (BHP) operates alone or is in joint venture with partners in a number of mining activities throughout Australia. These activities have taught the Company the importance of proper water management so as to best utilize this sometimes scarce resource; a resource which is often thought of as water mining when taken from groundwater aquifers. This paper briefly describes the water management experiences of the BHP Company throughout Australia. SCOPE OF MINING ACTIVITIES Figure 1 comprises a map of Australia on which the isohyets of average annual rainfall have been superimpose d. Also shown on Figure 1 are the location of the mines which will be referred to in this paper. Each mine

site is referenced by name; the nature of which is given in Table 1.

BRISBANE

MEDIAN ANNUAL RAINFALL IN MILLIMETRES

\;)HOBART

Figure l - Mine Location Table l - Code of Mines on Figure 1 No.

Mine

Ore

Ownership

Newman

iron

BHP and others

Telfer

gold

BHP and others

Koolan Island

iron

BHP

Worsley

alumina

BHP and others

Groote Eylandt

manganese

BHP

Gregory

coal

BHP and others

Moura

coal

BHP and others

Riverside

coal

BHP and others

Saxonvale

coal

BHP

Whyalla

iron

BHP

BHP is involved in other mines, such as the Ok Tedi gold and copper mine in PNG and underground coal mines in N.S.W. but they are excluded from

38 Table 1 as they are not located in areas as arid as that listing. It is outside the scope and space of this general paper to be specific about detailed experiences on all these mines but some typical examples are illustrative of the issues involved. SUPPLY AND USE OF WATER The basis of water management is the balancing of water supply with water usage and adopting water conservation where necessary or beneficial. Water may be supplied from the following sources which therefore represent the plus side of the water balance: (i)

surface rainfall runoff collected in dams and piped or otherwise supplied to the mine

(ii)

groundwater aquifers

(iii)

recirculation of plant water

(iv)

water contained within the mine ore

Water may be used by or lost from the following activities which therefore represent the minus side of the water balance: (i)

dust suppression on roads, ore piles, waste piles, mining areas

(ii)

domestic/town supply including grassed areas

(iii)

ore processing plant

(iv)

evaporation from open water supply storage

(v)

water contained in the product ore

(vi)

water discharged to waste from the mine site

WATER BALANCE MODELS Many of the items on the supply and usage sides of the water balance equation are subject to variation. This particularly applies to rainfall, evaporation and mine ore output among others. These variations give rise to questions by mine management such as: (i)

how often (or at what return period) will exceptionally wet conditions give rise to environmental problems in the case where the mine and ore product contaminate the inflow water

(ii)

at what return period will exceptionally dry conditions give problems of insufficient water for dust suppression when the external (say piped) water supply is restricted.

The multitudinous calculations (albeit simple) in these assessments are found to be best done by a computerised water balance model. An example of such a model used by BHP on a typical mine is shown in Figure 2.

WASH DOWN DRAl"I

LOSS

WASH DOWN

PAD TOTAL WASH DOWN US[

AH NITI

ROAO WAHR EHERCiENCY

ROAD WATER MAKE UP

TOTAL USE UIO,((OUNU.8LE LOSS WASHfRY WATER □AH

LOSS

WASHERY

TOU.L SPRAY

ABSORB.

·"

-~t~ l.v..--~

~P1 r

PIT

ROUND WATER !UNIT 11

1•o•o

AT[.

i >-

DUHP HOPPER

WATER [X·ltOPPER

RAW COAL WATER

TRANSFER

HOUSE I

. DRAINAGE

Figure 2 - Water Balance Model SOME EXPERIENCES OF WATER SUPPLY (i)

Surface Runoff

Seven of the 10 mines listed in Table 1 are supplied by pipeline from reliable river flows in the vicinity of the mine. The longest individual pipeline installed for this purpose is the 64 km by 434 mm diameter pipeline from the Nogoa River to Gregory mine carrying about 1860 ML/A (megalitres per annum) The pipeline from the Murray River to Whyalla is longer but it is used for town and other supplies along its route. Surface runoff is also collected from the local catchments around the mine and as the mine develops and earthworks are constructed these make an increasingly greater contribution to the total supply needs Moura mine which typically received in its early years about 1500 ML/A by pipeline from the Dawson River some 15 km away is now independent of that source if need be.

40 (ii)

Groundwater Aquifers

Five of the 10 mines listed in Table 1 are fully or almost fully dependant upon groundwater supplies for their operation. Riverside mine is receiving about 2200 ML/A from a borefield located some 100 km away. Newman initially was solely dependant upon the aquifers in the Fortescue River and tributaries. When the safe yield of that source only represented 25% of the 1979 gross requirement of 12,000 ML/A, a dam with a capacity of 32,000 ML/A was constructed so as to store the occasional streamflow before quickly using it for recharge and storage within _t he aquifer. (iii)

Recirculation of Plant Water

All the mines in Table 1 (except Koolan and Whyalla) involve processing of the mine product. The fine material (or tailings) resulting from this processing is stored in a tailings dam and recirculation of the water from this dam makes an important contribution to water supply. Such circulation at Newman provides 10 percent of annual mine requirements. To minimise evaporation losses, tailing dams must be operated with the minimum surface area of stored water. Although contrary to most tailings operational practice, there is merit in discharging the tailings at some high point away from the dam and allowing the water fraction to collect in a small deep pool at the dam wall from whence it is recirculated back to the plant . . (iv)

Mine Ore Water

This source of water can be surprisingly significant. Central Queensland run-of-mine coal from open pits contains about 5% water as percentage of dry weight of coal. This is equivalent to 200 ML/A for a typical 3 MT/A (million tonnes per annum) coal mine. SOME

EXPERIENCES OF WATER USAGE

(i)

Dust Suppression

Dust from haul roads, waste dumps, pit faces, and stockpiles represents a safety hazard, and a mechanical and equipment nuisance and most miners will not work when dust is present. Accordingly dust must be suppressed and this is usually done by spraying with water; the chemical suppressants that are available being generally too expensive for wide use. The arguments for and against fixed reticulation spraying and water truck spraying are outside the scope of this paper but include the usual conflict between capital cost and operating cost. The amount of water required .to lay dust is related to the evaporation rate. Experiments at Newman and experience elsewhere suggest that it is equivalent to about 2 times the net annual pan evaporation rate which is applied to the total road surface area. Water usage at Newman is about 2600 ML/A and at a typical Central Queensland open cut coal mine of 3 MT/A it is about 400 to 600 ML/A, or about one quarter of total usage. (ii)

Domestic/Town Supply

Greenness in a dry environment has soothing and beneficial effects and accordingly outback communities use much water on house gardens, playing fields and the like. Attempts to restrict such usage are mostly unsuccessful. It is better to be more efficient in greening the community, including use of drip irrigation, and watering in the cooler (less evaporative) time of the day. Use of treated waste water from such as sewage plants on playing fields also reduces the gross water needs. Newman uses about 3000 ML/A in

41 the township and about 110 ML/A of treated effluent. (iii)

Ore Processin g Plant

A coal preparati on plant requires about 500 litres of water per tonne of coal so that the Central Queenslan d open cut mine of 3 MT/A requires about 1500 ML/A for coal washing. This almost equals all the other mine uses of water and it is essential that it be recircula ted as much as possible. (iv)

Evaporati on

Evaporati on from open water storages at the mines listed in Table 1 averages about 5 times and varies from 2 to 10 times the annual rainfall; with values such as 10 being applicabl e in the northern half of the continent . At Newman, about 9000 ML net (or about 30%) were lost by evaporati on from the Fortescue storage dam during the 6 months that it took to recharge about 8500 ML into the aquifer. Hence the good practice indicated above for tailing dams should be generally applied, such as by avoiding large shallow areas in water storage. (v)

Ore Product Water

Although all ore products are freedrain ing, sufficien t quantitie s of water are retained within the pores to the extent that they must be taken into account in the water balance. Washed coal for example contains about 6% water which is equivalen t to 180 ML/A at mines such as Gregory. (vi) ►

Waste Water

The layout of the mine and the difficult ies of pumping to a useful central point often mean that rainwater or groundwa ter accumula ting in the open pits must be pumped to waste. Strict control must be placed on this operation to ensure that the discharge water satisfies the local water quality criteria. SOME SPECIAL WATER MANAGEMENT PROBLEMS (i)

Dirty Water Areas

Rainwater falling on coal stockpile s or in the local catchment of coal preparati on plants becomes contamina ted by dissolved salts such as sulphates , According ly, it is often a statutory hydrocarb on wastes and the like. so of rainfall runoff must be collected or mm 75 first the that nt requireme water dams. This water is either dirty so-called in y separatel and stored better quality water or is used with diluted later is , evaporate allowed to special water managemen t require options these All on. suppressi for dust water and clean water dirty several have may Riverside as such care. A mine bunding. and channels drainage separate their with dams catchment (ii)

Groundwa ter

Many areas of Australia , particula rly those sedimenta ry deposits laid down in seawater basins, have high salt content. The groundwa ter entering mine pits sunk into such strata must be specially handled and disposed of in a manner similar to dirty water. FLOODING

•

It is paradoxic al that mines that are concerned with husbandin g their water supplies and demand should also be concerned with flooding of the mine works and the pits from extreme rainfall events. The situation of high flood training banks along a stream that is almost always dry looks strange ·

indeed. The magnitude of the design flood is selected on a benefit: cost basis; where the benefit is the value of lost production avoided, the cost of pumping, mine disruption and the like and where the cost is the construction If flood water and maintenance cost of the flood prevention measures. cannot be easily it t tha extent the to ted entering a pit becomes contamina then a high closed, permanently be may pit discharged to waste so that the flooding. such avoiding of value must be placed on the benefit

•

The outcome of the analysis depends upon the local circumstances at the mine. The return period of the design flood at BHP mines ranges from several times mine life (e.g. Riverside) to Probable Maximum Flood period (e.g. Saxonvale). In those cases where the desirable extent of flood prevention measures is still too costly, it is advantageous to provide increased levels of ore product stockpiles so that at least lost production does not result in lost sales. SUMMARY The general experiences and examples given in this paper are intended to demonstrate and support the importance of proper water management practices on remote mine sites in Australia. Compliance with such practices are often as much or more in the Company's direct (cost effective) interest than satisfying some statutory environmental requirement.

◄

•