Report No: 10/1994A

Peter Jolly

Darryl Chin

Geoff Prowse

Michael Jamieson

HYDROGEOLOGY OF THE

ROE CREEK BOREFIELD

WATER RESOURCES BRANCH, ALICE SPRINGS

October 1994

•

PREFACE

In 1990 Peter Jolly, Geoff Prowse and Daryl Chin, all of Water Resources Division, PAWA, Darwin, presented a paper to the International Conference on Groundwater in Large Sedimentary Basins that was being held in Perth. The paper was entitled "The Amadeus Basin Mereenie Sandstone Aquifer - Regional Modelling Based on the Diagenetic History of the Mereenie Sandstone". In April 1991 Jolly and Chin presented a second paper "The Roe Creek Barefield Alice Springs" at the Arid Zone Water: A Finite Resource conference in Alice Springs. This incorporated much of the first paper, but took a different perspective and incorporated information not used in the first paper. These papers were later summarised and appeared in the AWWA Journal, Water, in June 1991, as "Alice Springs - The Mereenie Sandstone Aquifer".

The following report was compiled in 1993/94, based on these papers and a collection of notes, diagrams and figures, used in the preparation of these papers. Referral was also made to previous WRD reports concerning the Roe Creek Borefield ..

The purpose of this report is to collate the work done in developing the hydrogeological models associated with the Roe Creek Borefield and make this information more accessible to WRD and PAWA .

1

ABSTRACT

Alice Springs water supply requirements are obtained from the Roe Creek borefield, located 15 km south southwest of the town. This borefield was commissioned in 1964 when the existing supply from the alluvial sediments beneath the town became inadequate. The borefield extracts water from four discrete aquifers located within the following formations- Mereenie Sandstone, Pacoota Sandstone (two units), Shannon Formation.

These formations form part of the Amadeus Basin, an intracratonic depression of Late Proterozoic to Middle Paleozoic age. For the lifetime of the borefield these aquifers can be considered as discrete and not interconnected. Due to economics, these aquifers have been investigated and developed close to their area of outcrop.

The borefield supplies a mean daily water demand of thirty megalitres (or ten thousand megalitres annually) with more than 80% coming from the Mereenie Sandstone. Peak daily extraction is about fifty-five megalitres.

Regional investigations have shown that the Mereenie Sandstone will continue to be the major source of Alice Springs water supply. This will be through the existing Roe Creek borefield and the proposed Rocky Hill borefield.

To determine the economical sustainable yield of the Roe Creek borefield, data obtained from oil wells has been compared with that obtained from water bores and diamond drill holes. This has enabled groundwater modelling to incorporate porosity and permeability effects resulting from diagenesis. This technique was applied to the most important of the aquifers - the aquifer developed in the Siluro­Devonian aged Mereenie Sandstone. Local and regional computerised models were developed, showing that a mining situation exists in the Roe Creek bore field (i.e. water is being drawn from local storage, limiting the life of the borefield) . This has resulted in a linear relationship being developed between drawdown and cumulative extraction, enabling planners to determine the economical sustainable yield of the Roe Creek borefield.

2

•

LIST OF TABLES

LIST OF FIGURES

LIST OF ABBREVIATIONS

CONTENTS

1. INTRODUCTION

2. HISTORY OF THE ROE CREEK BOREFIELD

3. REGIONAL GEOLOGICAL HISTORY OF THE AMADEUS BASIN

4. THE MEREENIE SANDSTONE

4.1 DEPOSITIONAL HISTORY OF THE MEREENIE SANDSTONE

4 .1.1

4 .1.2

4 .1.3

UNIT A - QUARTZ SANDSTONE WITH SILTSTONE INTERBEDS

UNIT B - PURE QUARTZ SANDSTONE

UNIT C - LITHIC QUARTZ SANDSTONE

4.2 POST DEPOSITIONAL DIAGENESIS

4.3 HYDROGEOLOGY OF THE MEREENIE AQUIFER SYSTEM

4.3.1

4.3.2

4.3.3

4.3.4

ZONE 1 CHARACTERISTICS

ZONE 2 CHARACTERISTICS

ZONE 3 CHARACTERISTICS

REGIONAL FLOW MODEL AND ISOTOPE CHEMISTRY

4.4 SUMMARY OF MEREENIE DEPOSITION AND HYDROGEOLOGY

4.5 MODELLING OF THE MEREENIE

4.5.1

4.5.2

4.5.3

LOCAL ROE CREEK MODEL

REGIONAL MODEL

MODEL OF DRAWDOWN BASED ON EXTRACTION

4.6 HYDROGEOLOGY OF THE PROPOSED ROCKY HILL BOREFIELD

5. THE PACOOTA SANDSTONE

6 . THE SHANNON FORMATION

7. DISCUSSION AND CONCLUSIONS

8. REFERENCES

9. APPENDIX A. WATER LEVELS IN BORES FROM THE MEREENIE SANDSTONE

3

•

LIST OF F:IGQRES

1. REGIONAL GEOLOGY OF STUDY AREA

2. ROE CREEK PRODUCTION BORE LOCATIONS

3; DEPOSITIONAL ENVIRONMENT OF UNIT A - HIGH SWL

4. DEPOSITIONAL ENVIRONMENT OF UNIT B - LOW SWL

5. DEPOSITIONAL ENVIRONMENT OF UNIT C - HIGH SWL

6. EROSION PROCESS PRIOR BREWER CONGLOMERATE DEPOSITION AT IWUPATAKA EROSION FEATURE

7. CLOSE OF BREWER CONGLOMERATE DEPOSITION

8. PLANATED PRESENT AGE LANDSCAPE - ROE CREEK BOREFIELD AREA

9. HYDROGEOLOGICAL ZONES IN STUDY AREA

10. REGIONAL POTENTIOMETRIC CONTOURS FOR THE MEREENIE SANDSTONE

11. REGIONAL SALINITY AND RADIOCARBON DATING

12. POLLEN RECORD (130,000 years) FOR LYNCH'S CRATER, QLD

13. ROE CREEK BOREFIELD LOCAL MODEL

14 . COMPARISON OF MODEL AND MEASURED DRAWDOWNS

15. PLOT OF BOREFIELD DRAWDOWN VS. CUMULATIVE EXTRACTION.

16. GRAPHICAL RELATIONSHIP BETWEEN ANTICIPATED BOREFIELD EXTRACTION AND RESULTANT BOREFIELD DRAWDOWN

17. NORTH - SOUTH CROSS SECTIONS THROUGH MEREENIE SANDSTONE

Al. WATER LEVELS -RN 4693

A2. WATER LEVELS -RN 4502

A3. WATER LEVELS -RN 4687

A4. WATER LEVELS -RN 3609 ,'

AS. WATER LEVELS - RN 3600

A6. WATER LEVELS - RN 5457/11772

A7. WATER LEVELS -RN 5730/5731

5

BMR

c

em

em/sec

Kh

km

K,

L

m

m2/day

ML

ML/d

mm

mm/yr

mg/L

No

RN

sq.km

yr

2D

LIST OF ABBREVIATIONS

Bureau of Mineral Resources

Celsius

centimetre

centimetres per second

horizontal hydraulic conductivity

kilometre

vertical hydraulic conductivity

litre

metre

{metres per day) for metres of depth

megalitres

megalitres per day

millimetres

millimetres per year

milligrams per litre

number

registered number of bore

square kilometres

year

two dimensional

6

•

l.J:NTRODUCTJ:ON

Alice Springs is located close to the geographic centre of Australia (refer Figure 1) . It lies on a small plain on either side of the Todd River. With a mean annual rainfall of 285mm, Alice Springs is only moderately arid. Most of the rainfall occurs during the summer period. Temperatures in the region are moderate to high throughout the year, above 30 degrees C for six months of the year during the day. Potential evapotranspiration is high and has been estimated at between 1250 and 1500mm per year (Mabbutt).

The Amadeus Basin is the remnant of a succession of Late Proterozoic to Middle Paleozoic sediments with a maximum preserved thickness of about 14000 metres. It has an east west length of 800 kilometres and an area of about 170000 square kilometres.

The Roe Creek borefield supplies Alice Springs with a mean daily water supply of 30 ML/d, catering for a population of 24000. It is located 15km south southwest of the town and is in the northeastern section of the Amadeus Basin, at the eastern end of an east-west trending broad syncline. This syncline covers an area of approximately 50000 square kilometres (Refer Figure 1) . This syncline may be considered as a continuous series of discrete or poorly interconnected aquifers. Regional groundwater flow is from west to east.

The Roe Creek borefield currently has producing bores in four separate aquifer systems in three formations - the Mereenie Sandstone, Pacoota Sandstone (two aquifers), and Shannon Formation, with the Mereenie Sandstone producing more than 80% of the Alice Springs demand.

Details of the development and hydrogeology of the borefield follow .

7

2. HISTORY OF THE ROE CREEK BOREFIELD

Alice Springs was first established in 1871. Until 1964 the town obtained its water supply from alluvial sediments associated with the Todd River. In 1964 the Roe Creek borefield was commissioned, drawing water solely from the Mereenie Sandstone until 1984 when two production bores were commissioned in the Pacoota Sandstone. Annual extraction from the Roe Creek borefield is in the order of ten million cubic metres obtained from production bores pumping from up to 170 metres below ground level. Peak daily consumption approaches 55 ML. Eighty percent of production comes from bores in the Mereenie Sandstone. The set out of the borefield is shown on Figure 2. Individual bores may yield over 8ML/day with less than 5 metres well loss. Production bores Pl to P21 and P24 to P25 have been constructed to extract water from the Mereenie Sandstone. Bores Pl to P7 have since been decommissioned. Production bores P22 and P23 extract water from the upper units of the Pacoota Sandstone, P26 from the lower units. Production bore P27 extracts water from the Upper Shannon Formation. All production bores, except P27, extract water from aquifers developed in porous sandstone. P27 extracts water from solution enlarged fractures within a dolomite aquifer. The chemical quality of water from all aquifers is good with total dissolved solids in the range 350 to 650 mg/L and hardness in the range 130 to 230 mg/L.

The aquifer formations tapped in the Roe Creek borefield present significant problems in the acquisition of hydrogeological data for two reasons:

(1) They dip steeply to the south. Depths exceed 2000 metres within 10 kilometres of the Roe Creek borefield;

( 2) The overlying Pertnjara Group transgresses across the Mereenie Sandstone, resulting in it being absent to a depth of up to 1000 metres (or more) within the area between eight and 62 kilometres west along strike of the borefield.

The only hydrogeologic data previously available on the Mereenie Sandstone aquifer to the south and west was its confirmed continuity at depth from seismic data. This lack of data, combined with the aquifer boundary to the north, had resulted in previous workers being unable to arrive at a model that accurately predicted aquifer performance.

The approach taken during the current study was to investigate the diagenetic history of the Mereenie Sandstone and relate this history to a probable range of porosity and hydraulic conductivity parameters for the aquifer system. Further refinement of this range of parameters was then undertaken by developing a computerised groundwater model, first for the vicinity of the Roe Creek borefield, and then for the broad syncline under natural conditions. The knowledge thus gained then enabled a basic predictive model to be developed for the Roe Creek bore field.

8

•

•

3. REGIONAL GEOLOGICAL HISTORY OF THE AMADEUS BASIN

The Amadeus Basin, an intracratonic sedimentary structure, was infilled in periodic episodes from the Late Proterozoic to Mid Paleozoic times. It took the form of a connecting marine trough, the Larapinta Seaway, joining the Great Artesian Basin of eastern Australia with the Canning Basin of Western Australia.

Basin sediments are almost entirely shallow marine and terrestrial deposits. It is a proven petroleum province with two producing hydrocarbon fields.

Tectonic deformation throughout the northern Amadeus Basin took place during the Late Devonian aged Alice Springs Orogeny (350 million years ago). Within the study area the essentially flat bedded formations were progressively folded along the east west trending fold axes. Dips of up to 60° on the southern flanks of the Macdonnell Ranges resulted. The area has since remained relatively stable. Mesozoic and Tertiary erosion and weathering have generated the present landscape.

This "mountain building" event is known as the Alice Springs Orogeny. It is believed to have provided the environment for fracturing and deep chemical weathering that has resulted in the formation of the high yielding aquifers in the Mereenie, Pacoota, and Shannon Formations. Aquifer development in the vicinity of the Roe Creek borefield has been controlled by the extent of fracturing and weathering associated with:-

1. The deep valley erosion feature south of Iwupataka (refer Figure 1) to the west of the borefield, for the Mereenie and Pacoota Sandstones. Intense fracturing of the sandstone to the west of Roe Creek is most likely associated with chemical weathering and "mass" wasting along the flanks of this valley.

2. The Tertiary trough to the south east of Alice Springs for the Shannon Formation.

The major producing aquifer is the Mereenie Sandstone and this is described in detail in the next section.

9

4. THE MEREENIE SANDSTONE

The most significant aquifer system in the Amadeus Basin has been developed in the Mereenie Sandstone of Siluro - Devonian age. This aquifer system provides 80% of Alice Springs water needs through bores in the Roe Creek borefield and will continue to be important for the town supply through the proposed Rocky Hill borefield development.

In the borefield area the Mereenie Sandstone has a true thickness of approximately 365 metres (drilled thickness 425m) . Varying thicknesses have been noted in other locations where oil wells have intersected the formation. This is because different lithological controls have been used to mark the transition between the Mereenie and sediments of the Pertnjara Group. Often basal sandstones of the Pertnjara Group have been derived from erosion of the Mereenie Sandstone making determination of the contact almost impossible in oil and water wells.

4.1 Depositional History of the Mereenie Sandstone

Deposition of the Mereenie Sandstone commenced with shallow marine sediments in the west of the Amadeus Basin contemporaneously with the broad epirogenic uplift of the eastern sector (the Rodingan Movement). Subsequent erosion and deposition of fluvial marker beds across the Alice Springs area occurred. Arid desert conditions became widely established across the whole basin and up to 600 metres of mature quartz sand was deposited in this aeolian environment.

The Mereenie Sandstone is generally a white, pale brown to red brown, exceptionally pure quartz sandstone, generally fine grained, rarely medium, well rounded and well sorted. Minor accessory tourmaline, zircon and feldspar occur within the top and bottom few metres of the formation. Shale and siltstone lenses do occur. Sedimentary structures include thin massive bedding, characteristically cross-bedded with trough cross sets up to 10m thick. This suggests aeolian deposition. Ripple marks are frequently present and sediment transportation was dominantly towards the west and south. Palaeontological evidence for the age is lacking. The measured age of the Mereenie results in large depositional gaps between it and adjacent formations, which is unlikely, so several major depositional breaks probably occur within the sequence.

In the borefield area, data from cored holes RN 15020 and RN 15021 and observations of outcrop, particularly near Pine Gap, confirm the general conclusion that a terrestrial and mostly aeolian, at times fluvial to lacustrine depositional environment existed. Deposition commenced with coarse fluvial sediments, followed by a mixture of lacustrine to aeolian sediments.

Although Lau has divided the formation into five units based on composite petrographic work, data suggests that in a hydrogeological context the formation in the Alice Springs area can be divided into three units. These are, going upwards from the unconformity base, units

10

•

A (comprising Lau's basal 3 units}, Band C. Contacts between the units appear to be gradational. The divisions are based on grain size and compositional differentials. using airlifted disaggregated samples Ferguson previously had also divided the Mereenie Sandstone into 3 units I, II and III. To the west of Roe Creek these units approximately correlate to units A, B and C. To the east of Roe Creek Unit I roughly correlates to Unit B and Units II and III roughly correlate to Unit C. Table 1 in Section 4.4 summarises the petrographic data and the following sections explain the depositional interpretation of this data for the hydrogeological units .

4.1.1 Unit A Quartz Sandstone with Siltstone Interbeds

This is the lowermost unit and has a true thickness of approximately 150m. It represents the transition from a marine to a dominantly arid terrestrial environment for the Amadeus Basin. The base is recognised {C Powell personal communication} by a fluvial cobble bed above the erosional contact with the underlying Pacoota Sandstone (marine}. Interbedded siltstones to gradational mixtures reflect the transitional climatic environment. The depositional environment of this unit is shown in Figure 3.

The unit is a variable to well-sorted quartz sandstone with average grain size 0.1 to 0.3 mm, comprising some bimodal sections. It contains up to 10% lithics and clay and up to 2% tourmaline and zircon. Thin laminar bedding is the norm. Diagenesis has resulted in some quartz overgrowths and grain boundary contact pressure solution effects, resulting in mechanically competent rock properties.

Compaction and calcareous {calcite} and dolomitic cement reduced the porosity to less than 10% {Palm valley No 1 and West Waterhouse No 1}. It is probable that a high water table within a low dune relief landscape prevailed to allow the development of these cements. Alkaline conditions were likely because of sediment input from erosion of the older formations following the Rodingan Movement uplift in the north east.

4.1.2 Unit B Pure Quartz Sandstone

The middle unit has a true thickness of lOOm and is a clean, poorly cemented white quartz sandstone with high porosity and permeability characteristics. It is generally well sorted and has an average grain size of 0.2mm, typical of aeolian dune features. There is less than 3% lithics and clays and only accessory kaolinite, often occurring as vuggy-like accumulations. With deeper burial, minor calcite has been noted {Alice No 1 oil well}. The primary grain bonding mechanism is provided by grain boundary contact and kaolinite. Generally the unit is poorly competent. The greater median grain size typifies high wind velocity and, thus, high relief dunal systems.

These aspects, in addition to the lack of interbedded clay and siltstone beds, and the disaggregated nature of the rock when drilled

11

in water bores and deeper oil wells, suggest an extremely arid depositional environment.

The relatively rapid accumulation of sediments in the basin would have led to a deeper water table, minimising the development of a cementing matrix (see Figure 4).

4.1.3 Unit C Lithic Quartz Sandstone

The true thickness of this unit is approximately 115m. It is a distinctively cemented orange brown coloured sandstone and is finer grained than Unit B. It is typically well sorted with a median grain size of 0.15rnm, with some intervals being bimodal about 0.1 and 0.2rnm grain sizes (indicative of a fluvial phase). Thin laminae bedforms of a few rnillimetres are characteristic with restricted shaley partings, and multicoloured siltstone beds. Within the borefield it is well indurated with particularly competent properties imparted by a pervasive iron cement. Petrographic studies reveal a brown translucent to sometimes opaque ferruginous limonitic to possibly geothitic mineral rimming grain boundaries and interstitial grain voids and covering most grain surfaces. The rock is a lithic quartz sandstone with up to 10% lithics/clay and containing some aggregates of white kaolinite.

Oil well data shows that calcareous and dolomitic cements reduced porosity to less than 10% (Palm Valley No 1 and Orange No 1). The iron cement is thought to develop in early diagenesis because of infiltration of wind blown dust following desert storms. Oxidised Fe3

•

could be liberated from ferrous silicates within the lithics. Data from cored hole RN 15021 shows well-preserved fluvial bed forms near the top of the Mereenie (which preceded lacustrine sediments of the younger Hermannsburg Sandstone) . Therefore it is thought that a wetter environment existed, especially towards the close of the period, which raised the continental water table to the near surface (see Figure 5). This unit represents the transition from an arid to a very wet, humid environment.

4.2 Post Depositional Diagenesis

The most significant regional post depositional diagenetic episode occurred contemporaneously with the "mountain building" event known as the Alice springs Orogeny. Steep relief accompanied by high rainfall resulted in deep chemical weathering in areas of, and next to, outcrop. This deep weathering established much of today's primary porosity in the Mereenie. Before this deep chemical weathering the -Mereenie had lower porosity, due to chemical cementation during its deposition.

Contacts with the Mereenie from formations both above (eg Hermannsburg Sandstone of Pertnjara Group) and below (eg Pacoota Sandstone) are regionally gently unconformable. Localised marked unconformities do, however, occur because of erosion along axes and whole areas of tectonic movement. The most significant of these unconformities occurred south of Iwupataka. In this sector (central along longitude

12

•

•

•

133°30') erosion has removed all units of the Mereenie and also older and younger formations (refer to figure 1 with cross-section C-D which shows the location of this Iwupataka Erosion Feature) .

In a localised context, this erosion feature south of Iwupataka (see Figures 6 and 7 for an explanation) has significantly affected porosity and permeability. It provided a localised mechanism for very deep chemical weathering (up to 1000 metres below present land surface) and re-exposed the Mereenie Sandstone on the flanks of the erosion feature to sub-aerial weathering. The resulting chemical breakdown of the lithic grains (iron silicates) and redistribution of iron resulted in a more competent sandstone. Subsequent deformation in the latter stages of the Alice Springs Orogeny has resulted in a higher level of fracturing in this more brittle rock and hence greatly enhanced permeability. In Roe Creek borefield Mereenie Sandstone thus affected occurs immediately to the west of Roe Creek.

During the Tertiary (Figure 7), iron-enriched weathering profiles developed over much of the present day outcrop of Mereenie Sandstone. Hydrogeologically, the only effect of this weathering was probably a slight further enhancement of porosity in the Mereenie Sandstone close to outcrop.

Figure 8 shows the present landscape at the Roe Creek borefield with regard to the underlying geology.

4.3 Hydrogeology of the Mereenie Aquifer System

The depositional and diagenetic history of the Mereenie Sandstone (detailed in sections 4.1 and 4.2) controls the hydrogeology of the aquifer system.

On this basis, the study area has been subdivided into the three zones shown on Figure 9. The hydrogeological characteristics of these regional zones follow and hydrogeological parameters for each zone are summarised in Table 2, Section 4.4.

zones 1 and 2 have been extensively drilled and tested within the Roe Creek, Rocky Hill, Deep Well and Waterhouse Range areas. Information on Zone 3 is available primarily from oil wells drilled in the region.

In a regional context Zone 2 and Zone 3 are extensive, while Zone 1, which covers only a small area, is important primarily because most of the producing bores in the Roe Creek Borefield are sited within it .

4.3.1 zone 1 Characteristics

This is a very high hydraulic conductivity and porosity zone. Sandstone has had carbonate and other non-siliceous cement leached out and iron redistributed during physical and chemical weathering associated with erosion and subsequent aerial exposure during the Late Devonian period. Further deformation during the Alice Springs Orogeny

13

has resulted in significant fracturing of this more brittle sandstone.

Ten plug samples from bore hole RN 15020 were analysed by AMDEL (Lau 1989) for permeability to air and porosity data. These gave average vertical (k,) and horizontal (kh) hydraulic conductivities of 1. 5 x lo-• and 2.9 x lo-• em/sec respectively. Porosity values averaged 23.2%.

Three cores from RN 15020 tested for k, and kh (Lau) resulted in average values of 3.6 x lo-s em/sec and 6.2 x lo-• em/sec. Porosity averaged over 21%. Test pumping results indicated hydraulic conductivities for bores producing from the fractured sandstone between 2 x 10 -2 and 1 x 10 -1 em/sec. Values for storage coefficient averaged 3 x 10-3 (Roberts 1978) .

Results therefore indicate that the aquifer in this zone is controlled by fracture permeability with hydraulic conductivities in the range 2 x 10-2 to 10-1 em/sec. The ratio of Kh to K, is between 2 and 17 for unfractured core. Where confined, the aquifer has a storage coefficient of 3 x 10-3 • Bulk porosity must exceed 23% since the porosity measured in the core takes no account of storage within large vugs and fractures in the bulk rock. Specific retention values of between 2.3% and 7% (averaging 5%) have been measured for this sandstone, in seven plug samples from holes RN 15020 and RN 15021. Therefore, where unconfined, the aquifer would have a specific yield between 17 and 21%.

4.3.2 Zone 2 Characteristics

Zone 2 is a high hydraulic conductivity and porosity zone. Sandstone has had carbonate and other non-siliceous cement leached out during chemical weathering associated with erosion during the Late Devonian period. Minor fracturing is associated with the Alice Springs Orogeny particularly in the north east.

Sixteen plug samples from core hole RN 15021 were analysed by AMDEL for permeability to air and porosity data. These gave average values for K, and Kh of 2.2 x lo-• and 4.1 x lo-• em/sec respectively. Porosity values averaged 23.3%. Twenty-two plug samples from cores from various bores were analysed by BMR (Woolley 1966) for permeability to air and porosity data. These gave average values for K,, Kh in direction of dip, and Kh in direction of strike, of 3.6 x lo-•, 3.7 x lo-• and 4.4 x lo-• em/sec respectively. Porosity values averaged 22.6%.

Three cores from RN 15020 tested by Dames and Moore for K, and Kh resulted in average values of 2.5 x lo-s and 1 x lo-• em/sec. Porosity values averaged over 20%.

Five cores from Alice Oil No 1 gave values forK, and Kh averaging 3.4 x lo-s and 1.5 x lo-• em/sec. Porosity values averaged 20.3%. Testing of the Mereenie Sandstone in Wallaby No 1 yielded sonic porosities in the range 6 to 24%, averaging 22%. Test pumping results indicated hydraulic conductivities between 7 x lo-• and 7 x 10-3 em/sec. Storage

14

•

coefficients were between 3 x 10-3 and 1 x lo-• (Eggington 1966, Verhoeven et al. 1979, Macqueen and Knott 1982}.

Results therefore show that the aquifer in this zone is influenced by fracture permeability with hydraulic conductivities in the range 7 x lo-• to 7 x 10-3 em/sec. The ratio of Kh to :Kvis between 1.2 and 4.4. Where confined the aquifer has a storage coefficient in the range 3 x 10-3 to 1 x lo-•. Porosity values are over 22%. For similar reasons as for Zone 1, where unconfined the specific yield of the aquifer would be in the range 17 to 21% .

4.3.3 zone 3 Characteristics

A low hydraulic conductivity and moderate porosity zone. Deposition continuous through the Devonian period with sediments of the Pertnjara Group. Moderate permeability and porosity restricted to Unit B of the sandstone which has porosities over 20% due to a lower degree of carbonate and non-siliceous cementation, resulting from near surface diagenesis, during deposition in a more arid climate than existed for Units A and C.

Data for this zone has primarily been acquired from inspection of well completion reports for Palm Valley No 1, Palm Valley No 2, Palm Valley No 3, west Waterhouse No 1 and Orange No 1 oil wells. Most of these wells were drilled through the Mereenie Sandstone using air drilling techniques and therefore yield significant information on the relative permeabilities of the three units. It is difficult to correlate the different reported drilled thicknesses of Mereenie Sandstone. Differing criteria were used in each instance to determine the contact with the overlying Pertnjara Group sediments.

However, based on airlifted water yields and the necessity to change to the use of "aerated water", significant permeability was noted in an average thickness of only 120 metres in each well. This high permeability section closely correlates with Unit B of the Mereenie Sandstone.

Complete porosity data is available only for Palm Valley No 2 oil well. Porosity values of 15 to 25% occur only over an interval of 113 metres. Values of 5 to 15% occur over 238 metres. Porosity values of less than 5% occur over the bottom 160 metres and the upper 180 metres. The first substantial inflow of water noted in drilling was coincident with the top of an 85-metre thickness of sandstone with porosities consistently exceeding 15%. In these oil wells significant amounts of calcareous and dolomitic cement were recorded. This suggests that lower porosities are associated with the presence of original cementing material which results in a more competent sandstone. Drilling and geological data also indicate the presence of only minor fractures in the Mereenie Sandstone in these wells. In Orange No 1 oil well a flow rate estimated at "forty barrels of fresh water per hour" (2 litres per second} was noted after 220 metres of Mereenie Sandstone had been intersected. Flow rates increased to "200 barrels of fresh water per

15

hour" (10 litres per second) at this depth and continued to increase over the next 122 metres. This data is not inconsistent with the other oil wells and suggests that the high fracture permeability intersected in Zones 1 and 2 is absent.

Data for this zone shows that significant hydraulic conductivity exists only· for a thickness of 120 metres correlatable with Unit B. This hydraulic conductivity is equal to that of the sandstone, due to the absence of fracturing. Data obtained from cores in Zone 2 (particularly that from Alice Oil No 1) gives the range for possible values of Kv as 2.5 x lo-s to 3.6 x lo-• em/sec and for Kh as 1 x lo-• to 4.4 x 1o-• em/sec.

4.3.4 Regional Flow Model and Isotope Chemistry

Regional potentiometric level contours have been constructed for the Mereenie Sandstone aquifer (refer Figure 10). These contours are based on the limited potentiometric water level data that is available (see Appendix A), the hydraulic parameters given in the previous section, and the results of regional computerised groundwater flow modelling (briefly outlined in the following section) . Available data suggests that recharge is primarily from small rivers and creeks that run along strike o.f outcropping or subcropping Mereenie Sandstone. Discharge is controlled by the base level of present day geomorphology with major discharges being evapotranspiration from where the Finke River, Deering Creek, Ellery Creek and Todd River cut across the Mereenie Sandstone and from the subcropping Mereenie Sandstone in the Deep Well area.

Water chemistry data suggests that the aquifer system developed in the Mereenie Sandstone acts discretely with insignificant interconnection between it and the aquifer systems developed in underlying or overlying strata. All systems, however, have similar undisturbed potentiometric levels because the base level of present day geomorphology controls the recharge-discharge relationships of all these aquifer systems.

Carbon isotope chemistry has been used to help derive a regional water balance for the Mereenie Sandstone. Using parameters derived in the previous section, the total volume of water in storage in the system is estimated to be in the order of 3 x 1011 cubic metres (300 million ML). Carbon-dating data contained in a paper by Jacobson et al. and in another by Calf (1978) (Refer Figure 11) suggests a mean age for groundwater in the system of 20000 to 50000 years. This age equates to an average annual recharge in the order of 1 to 2. Scm per year through the outcropping Mereenie Sandstone. Water level monitoring in the Roe Creek floodout area over a thirty year period indicates that a recharge event resulting in a rise in water level in the range 0.2 to 0. 5 metres occurs once in every five years. This equates to an average annual recharge in the order of 0.8 to 2cm per year.

Historical biological climatic data from northern Australia (Kershaw, 197 6, 1980) indicates the probable variation in the mean annual rainfall over the preceding one hundred and twenty three thousand years

16

(refer Figure 12). This variation can be used as a guide to the likely magnitude and duration of lower and higher rainfall periods in Central Australia. Preliminary modelling of such a rainfall variation suggests that the most probable effect would be a variation in discharge by means of evapotranspiration at the base level of the geomorphology relevant for that period. Available data suggests that the stability of the landscape has resulted in very little change to this base level, at least over the preceding seventy thousand years.

Ranges of values for hydraulic conductivity at depth in Zone 3 can be obtained both by consideration of movement of water away from recharge areas in the west of the system and by the movement of water between the McDonnell Ranges and West waterhouse No 1 oil well. Water from this oil well has been dated at 30900 years (Jacobson et al. 1989). Values thus obtained are in the range 4. 6 x lo-s to 4. 5 x lo-• em/sec. The ratio of along-strike to across-strike hydraulic conductivity obtained is approximately five. These figures are the Zone 3 (regional) figures shown in Table 2.

4.4 Summary of Mereenie Deposition and Hydrogeology

The Mereenie Sandstone has been subdivided into three stratigraphic units. The depositional history can be summarised as:

Unit A:

Unit B:

Unit C:

Deposited in an environment that increased in aridity with time. Underlying sediments were deposited in a marine environment. In the Roe Creek - Ooraminna Anticline area, the base is marked by well-rounded pebbles and cobbles deposited in a fluviatile environment. It is characterised at depth by being well cemented by carbonate and silica cement, probably emplaced at time of deposition.

Deposited in a very arid environment. Uniform grain size with very little cementation. Carbonate noted in some samples that may have resulted from remobilisation of sedimentary carbonate.

Deposited in an environment of decreasing aridity. Sediments of fluviatile origin become more common with the sandstone having thin laminar bedding and occasional shale partings. It is characterised at depth by being well cemented by carbonate and silica cement, probably emplaced at time of deposition.

The petrography of these units is summarised in Table 1.

The Mereenie Sandstone aquifer system has also been subdivided areally into three zones, based on the different depositional and diagenetic history of different parts of the basin. Hydrogeological parameters for each of these zones are shown in Table 2.

17

TABLE 1 PETROGRAPHIC CORRELATION DATA

PETROGRAPHY UNIT A UNIT B UNIT C

AVERAGE GRAIN 0.1-0.3mm, 0.2mm Mostly 0.15mm SIZE minor cobbles (0.1 - 0 .4mm)

MODALITY AND Poorly to Generally well Poorly to well SORTING well sorted. sorted and sorted.

Unimodal, unimodal Unimodal, some minor bimodal bimodal sections

LITHICS/CLAY Up to 10% Trace to 3% 2 - 25%

IRON RIM CEMENT Partial near Nil Moderate and base partial to high

degree

MECHANICAL Competent Disaggregated in Competent COMPETENCY parts. Generally

poorly competent

TABLE 2 HXDRQGEOLOGICAL PABAMETERS FOR THE MERE&NIE SANDSTONE

ZONE HYDRAULIC '&/K. POROSITY STORAGE SPECJ:FJ:C CONDUCTJ:VJ:TY (%) COEFFJ:CJ:ENT YIELD

(em/sec) (%)

1 2X10-2 to 2 to 17 >23 3x10-3 17 to 21 lXl0-1

2 7xlo-• to 1.2 to 4.4 >22 lxlo-• to 17 to 21 7xl0-3 3x10-3

3 2. Sxlo-s to 4.4 10 3x10-3 3 to 7 MECHANICAL 4. 4xlo-•

3 4. 6x10-s to 5 REGIONAL 4 .5xlo-•

Variations in yield and competence of the Mereenie Sandstone have been attributed to a combination of how the sandstone was originally deposited and the weathering processes it has since been subjected to. Unit A ( the first deposited) represents the transition from a marine to an arid terrestrial environment and is relatively low yielding. Unit B was deposited under extremely arid conditions and is the least competent of the units. Formation stability problems encountered in this unit are associated with deep chemical weathering that took place during the Alice Springs Orogeny and later during the early Tertiary period (ie fifty million years ago). Unit C represents the transition from an arid to a very wet, humid environment. It is also relatively low yielding except in a small area to the west of Roe Creek. It is,

18

however, in this area where most of the production bores are located. High yields in Unit C in this area are associated with large open fractures believed to have resulted because of the "brittle" nature of the sandstone in this area. It is probable that this "brittleness" resulted from the redistribution of iron during physical and chemical weathering of the Mereenie Sandstone during the Alice Springs Orogeny. At this time the Mereenie formed the flanks of the valley that is now evident as the erosion feature south of Iwupataka.

Detailed studies of permeability and porosity characteristics have only been undertaken for the Mereenie Sands tone. The borefield ··studies have shown that the aquifer in the Mereenie Sandstone is controlled by fracture permeability with hydraulic conductivities in the range 0.02 to 0.1 em/sec. Porosity values average in excess of 22%. Examination of data for the Mereenie Sandstone where it has been intersected at depth in oil wells shows that significant permeability and porosity only occurs in Unit B over most of the basin. Hydraulic conductivities of the order of 10 em/sec and average porosities of 10% occur in this situation.

4.5 Modelling of the M9reenie

The mathematical models available to model groundwater flow solve the mathematical equations representing the physical process using either finite difference or finite element numerical methods. The finite difference numerical method was chosen in this instance for several reasons. They included the user's familiarity with the mathematics of the method, the availability of a finite difference computer software package and the ease of application of the package to the conceptual model.

The particular package used was a version of the us Geological Survey Modular three-dimensional Modflow programme. The programme uses the three dimensional modular finite difference groundwater flow model, Modflow (McDonald and Hasbaugh) . It uses pre- and post-processors for the setting up of the input data files and the manipulation of the output nodal heads and/or drawdowns to obtain contour plots.

Three models have been established as tools in further refining the hydraulic parameters of the Mereenie Sandstone aquifer and to predict future performance of the Roe Creek Borefield. Two of the models use the Modflow computer package. The first is a localised model that investigates the behaviour of water levels within 20km of the Roe Creek borefield. The second is a regional model used to help understand hydraulic gradients and potentiometric levels on the regional scale.

Results from the first two models suggested that a "tank model" was sui table for the Mereenie Sandstone. The third model is this "tank model". It results in the development of a relationship between cumulative extraction and drawdown in the borefield. The graphical representation of this linear relationship that occurs between

19

cumulative extraction and drawdown when an aquifer is mined, can be used as a simple but effective means of predicting the future water levels within the Roe Creek borefield.

4.5.1 Local Roe Creek Model

This model was established with the aim of reproducing the measured drawdowns within the area of pumping influence of the Roe Creek borefield. It is a simplified 2D finite difference flow model that covers a plan area of 351.75 sq. km, being 33.5 km from east to west and 10.5 km from north to south (refer Figure 13). The eastern extent of the model lies in the area of Rocky Hill (obs. bore RN 11340) and the western extent lies approximately 5 km west of Pine Gap. The northern boundary of the model is coincident with the northern edge of outcrop of Mereenie Sandstone within the Roe Creek Barefield area and the model then extends 10.5 km south. The size of the model has been chosen such that the effects of the no flow model boundaries, which incorrectly represent an actual physical continuum of the aquifer, are not significant within the time-frame of modelling.

The three-dimensional basin geometry of the Mereenie Sandstone aquifer has been simplified for modelling purposes into two dimensions. This assumption, made to enable a two-dimensional model to be used, is valid because the groundwater flow is primarily within the Mereenie Sandstone and not across formations. A single layer, 70 x 40 node rectangular grid was overlaid on the model area and the aquifer boundaries, parameters and the location of pumping bores were then translated to the nodes within the grid. The grid size in the east-west direction varies from 250m within the borefield area to 1250m immediately east and west and up to 5000m at the extreme western edge. A smaller grid was necessary in the borefield area to obtain the spatial definition required to locate production and observation bores and the aquifer boundaries. In the north-south direction a uniform grid spacing of 250m is adequate for both spatial definition and coverage.

The model boundaries are all no-flow boundaries. Within the model the northern edge of the Mereenie Sandstone outcrop has also been modelled as a no flow boundary (ie no interconnection between the Pacoota and Mereenie Sandstones). Provision has been made within the model to simulate throughflow. Initial water levels used in the model reflect a natural throughflow prior to pumping of 250 cubic metres per day.

Those areas of the aquifer where the water table intersects Mereenie Sandstone are modelled as unconfined with the remainder modelled as confined aquifer conditions. (Refer Figure 13).

The hydraulic conductivity as given in Table 2, varies with direction. The ratio of horizontal to vertical hydraulic conductivity can be equated to the ratio of east-west transmissivity to north-south transmissivity. Transmissivity also varies areally, as discussed in previous sections due to fracturing and weathering. Using data given in Table 2, a west to east transmissivity in the borefield and related

20

highly fractured areas of 3000m2/d was chosen for the first run. Within the outcrop or weathered areas a west to east transmissivity of 1000 m2 /d was used. The remaining confined areas were given a west to east transmissivity of 300m2 /d. A ratio of 10:1 was first utilised as the ratio of transmissivity west-east to that north-south. In the initial run a specific yield of 0.20 was assigned to the unconfined area. For the remaining confined area of the aquifer an initial storage coefficient of 0.001 was assigned.

Before the Roe Creek borefield was commissioned in 1964 the Mereenie Sandstone aquifer was in an unstressed condition. During this time there was only limited water level data collected (see Appendix A). Therefore, the approach adopted for modelling was to use the first eleven years of extraction and water level data to calibrate the model and the second ten years to verify the model. The monitoring bores used in the calibration and verification processes were all located in the outcrop of the Mereenie Aquifer (refer Figure 13).

The process followed to calibrate the model was to first superimpose any available pre-pumping standing water levels coarsely on the model as starting or initial heads. The model was then run for ten years to stabilise and then stressed for eleven years with the borefield extraction rates from 1964 to 1975. Various combinations of parameters within the ranges shown in Table 2 were trialled to establish the combination that best reproduced the observed water levels.

The model that most closely reproduced the actual drawdowns aquifer evolved through thirteen calibration runs. The combination of parameters was as follows:

Specific yield of 0.20 for the unconfined area, Storage coefficient of 0.003 for the confined area,

in the final

West-east transmissivity of 10000m2/d in the borefield area, west-east transmissivity of 3000m2 /d in the remaining unconfined area west-east transmissivity of 300m2 /d in the confined area. The ratio of west-east to north-south transrnissivities was 5:1.

Verification of this model then involved the input of extraction rates from the borefield from 1976 to 1985. Selected outputs of this final model are graphically presented in Figure 14 as log distance versus drawdown and log time versus drawdown plots.

A significant finding was that the contribution from up-dip flow and across strike flow from Rocky Hill was small (about 0.5 million cubic metres/ annum) in comparison to the borefield annual extraction of ten million cubic metres.

4.5.2 Regional Model

A regional model for the syncline was set up with the aim of confirming the reduction in permeability at depth by reproducing the regional potentiometric levels which result from natural recharge and discharge

21

from the system. This simplified 2D finite difference model covers a plan area of 9600 sq. km, being 240km from east to west and 40 km from north to south. The model has been set up as a single layer, 60 x 40 node rectangular grid. East to west the grid spacing is a uniform 4 km and north to south it is 1 km. The model encompasses the Mereenie Sandstone aquifer with the model no-flow boundaries coinciding with the edges of Mereenie Sandstone outcrop. Again, where the water table intersects the Mereenie Sandstone (ie next to outcrop) the aquifer was treated as unconfined with the remainder being confined. Natural recharge (stream concentration and infiltration) and discharge occurs only over unconfined areas of the aquifer. Recharge (stream concentration and infiltration) occurs over the MacDonnell, James and Waterhouse Ranges. Discharge occurs by transpiration over the subcropping Mereenie Sandstone to the east, and as river base flow where the Finke River system dissects outcropping Mereenie Sandstone. The recharge module of the Modflow package was used to simulate the required recharge while discharge was simulated using discharge wells. Initial parameters used were; specific yield of 0.20, storage coefficient of 0.003, west to east transmissivities in unconfined areas of 2500m2/d, and in confined areas 100m2 /d. The ratio of west-east to north-south transmissivities was 5:1. Recharge into the sys.tem was set at lOrrnn/yr.

This model was calibrated by reproducing the known regional water levels as shown on Figure 10. The process followed was to use various recharge rates and hydraulic parameters for a period of 1000 years, and withdraw from the system the equivalent volUme of water to maintain a net storage change within the system of zero. Various combinations of parameters within the ranges shown on Table 2 were trialled to establish that combination that reproduced the best match of a potentiometric surface with the set of known regional levels.

The best fit was obtained with the following parameters;

a recharge rate of lOmm/yr, storage coefficient of 0.003, specific yield of 0.20, confined transmissivity west to east of 100m2 /d, unconfined transmissivity west to east of 3000m2 /d, east -west to north -south transmissivity ratio of 10:1

The regional model suggested regional throughflow into the Rocky Hill area was in the order of 5000 cubic metres per day.

4.5.3 Gr8Dhical Borefield Model

The flow regimes produced by the two models made it clear that extraction from the borefield was not intercepting regional throughflow, which would imply a sustainable resource, but that extraction was coming from "local" storage. Modelling indicated that with the current annual extraction rate of ten million cubic metres, inputs from recharge, up dip flow and across strike flow from the Rocky

22

Hill area were only in the order of 0.5 million cubic metres. It was therefore concluded that the hydraulic performance of the Roe Creek borefield could be conceptualised by·a tank model. This results in a predictive error for water levels of less than 5%, as seen when borefield drawdown was plotted against borefield cumulative extraction (see Figure 15). Drawdowns from three representative monitoring bores for the period 1964 to 1989 were plotted versus the cumulative extraction from the borefield. Lines of best fit were then drawn through the points and a statistical analysis carried out. Figure 15 shows drawdowns in monitoring bores with RN's 3609, 5731/5798 (data sets combined) and RN 3600, versus cumulative extraction.

Extrapolation of the linear plot and the inclusion of future estimated extraction results in a simple graphical relationship that can be used to predict the medium term (30 year) performance of bores in the Mereenie Sandstone at the Roe Creek borefield (see Figure 16).

4.6 Hydrogeology of the Proposed Bocky Hill Barefield

The proposed borefield development in the Mereenie Sandstone aquifer in the Rocky Hill area is significantly different to the Roe Creek Barefield. The dip of the sandstone is much shallower (refer Figure 17) and consequently greater storage is available for exploitation from a shallower depth. Production bores will need to be constructed to extract water from the poorly cemented Unit B.

It is believed that previous workers have mistakenly identified Unit C as the Hermannsburg Sandstone. This was due to the lower than expected yields intersected in that unit in the Rocky Hill area (similar to the east of Roe Creek in the Roe Creek borefield) .

23

5. THE PACOOTA SANDSTONE

The Pacoota Sandstone consists of interbedded shallow marine sandstones, siltstones, shale, minor conglomerates, carbonates and mudstones in characteristic " shoaling" upward sequences on geophysical downhole gamma logs (Lau, 1989). It is the main oil and gas reservoir for the Palm Valley and Mereenie hydrocarbon fields. It is of Late Cambrian - Early Ordovician age. Two poorly interconnected aquifers have been identified in predominantly sandstone sequences separated by a sequence of mudstone. The sandstones are characterised by secondary porosity after solution of carbonate shell fragments (Lau, 1989).

Diagnostic features of the Pacoota Sandstone are its paleontology, its • 'shoaling upwards' cyclic gamma log, its shelly sandstones and brecciated and intraclastic sandstones. The presence of glauconite in the formation suggests a slow sedimentation process.

Sandstones range from friable to compact. Compaction and reduction of original porosity have occurred by pressure solution effects on the quartz grains rather than by squeezing of lithics into intergranular spaces or by deposition of authigenic kaolinite. Sandstones are also characterised by secondary porosity after solution of carbonate shell fragments. It is possible that friable sandstones have also had carbonate cements removed by solution.

A 40m thick khaki mudstone separates the 2 Pacoota Sandstone aquifers. This mudstone forms a strong aquiclude. Other mudstones occur but are generally red, silty, thinner and more permeable than in this interval.

Permeability and porosity are at a maximum near outcrop (ie where the formation occurs at the surface) . Permeability and porosity markedly decrease with increasing depth due to the closing up of fractures with the increasing weight of overlying material and the decreasing amount of chemical weathering. The Pacoota Sandstone appears to have both primary storage characteristics in some sandstones and fractured rock transmissivity in others, with a combination in particularly favourable intervals.

24

.

6 • THE SRANNQN FQRMATJ:ON

The Shannon Formation is of Middle to Late Cambrian age and is characterised by the rhythmic alternation of siliclastic mudrocks (shaley half - cycles) and carbonates (carbonate half cycles) of great lateral continuity. The term Shannon Formation is used to describe this rock unit instead of the localised terms, Jay Creek Limestone and Hugh River Shale. In practice, the Upper Shannon Formation is equivalent to the Jay Creek Limestone, the Lower Shannon to the Hugh River Shale. The Upper Shannon Formation consists of interbedded dolomite and shale. Production bore P27 has been constructed to intercept an aquifer in a predominantly dolomitic sequence where fractures have been enlarged by chemical weathering .

The Upper Shannon Formation consists of interbedded carbonate and mudstone with a true thickness of approximately 140m. Petrographic descriptions show the carbonate to be dolomite rather than limestone. The dolomite is hard and pyritic in part, stromatolitic and thrornbolitic in some intervals. Secondary features are stylolites, which are generally subparallel to bedding, and vugs.

Two distinct sets of joints are apparent dolomite. One is subvertical, possibly due to normal to bedding, probably an axial-plane flexural slip folding.

in the more competent unloading. The other is cleavage formed during

Pyrite is common with calcite in vugs and on fractures and probably contributes to high sulphate values (178mg/L in RN 15102).

Yield from the Upper Shannon Formation decreases to the west of Roe Creek. There is also a steep increase in potentiometric gradient here. This is because deep chemical weathering in this location was confined to the early Tertiary period and then only adjacent the Tertiary trough located to the south-southeast of Alice gprings next to the Todd River. The erosion feature to the south of Iwupataka does not transgress this formation.

The permeability and porosity of the formation are at a maximum near outcrop (ie where it occurs at the surface). Permeability and porosity markedly decrease with increasing depth due to the closing up of fractures with the increasing weight of overlying material and the decreasing amount of chemical weathering.

25

7 DISCUSSION AND CONCLUSIONS

The aquifer system developed in the Mereenie Sandstone is primarily controlled by two diagenetic episodes. The first took place contemporaneously with deposition and resulted in the partial loss of primary porosity by compaction and cementation. This reduced porosities to below 10% over most of the formation except for a 120 metres thick unit of uniform grain size (0.2mm) sandstone. This unit was most likely deposited in an environment of extreme aridity and high wind velocities. Porosities as high as 25% were maintained in this unit. The second episode occurred during the Late Devonian "mountain building event" known as the Alice Springs Orogeny. Steep relief combined with high rainfall enabled deep weathering to occur, thus reestablishing high porosities (20 to 30%) in areas next to outcropping Mereenie Sandstone over the entire formation thickness. Fracturing associated with this orogeny has also enhanced hydraulic conductivity.

During the Alice Springs Orogeny a major valley system developed (Iwupataka Erosion Feature) immediately to the west of the present position of Roe Creek. This valley system enabled a unique weathering imprint to be placed on the sandstone that formed the flanks of the valley system. This has resulted in zones with an anomalously high degree of fracturing, and associated high hydraulic conductivities, on either side of the Feature. The high yielding production bores of Roe Creek borefield situated to the west of Roe Creek are sited in such a zone.

Computerised modelling has simulated the natural system and the effects of extraction from the Roe Creek borefield. Modelling has incorporated the depositional and diagenetic history of the Mereenie Sandstone by applying separate hydrogeological parameters to different parts of the aquifer. Results obtained from this exercise have suggested that the performance of the Mereenie Sandstone in the Roe Creek borefield can best be represented by a linear relationship between drawdown and cumulative extraction. This equates the system to a mining situation. Modelling of the other three aquifer systems tapped by the Roe Creek borefield is required. A similar approach to that adopted for the Mereenie Sandstone is considered appropriate.

The regional flow system is controlled by the base level of present day geomorphology. Estimates for evapotranspiration and throughflow rates suggest that all water in the system can be accounted for with average recharge rates in the range 0.8 to 2.5 em per year.

Production bores in the Mereenie Sandstone aquifer within the Roe Creek borefield are currently mining groundwater at the rate of ten million cubic metres per year. Inputs to the system from recharge, up dip flow and across strike flow from the Rocky Hill area are estimated to be in the order of 0.5 million cubic metres per year.

A basic linear relationship between drawdown and cumulative extraction has been developed which will enable planners to determine the economical life time of the Roe creek borefield.

26

8. REFERENCES

CALF, G.E. 1978.

The Isotope Hydrology of the Mereenie Sandstone Aquifer, Alice Springs, NT, Australia. Journal of Hydrology, 38: 343-355.

DOUGLAS, I. AND SPENCER, T.

Environmental Change and Tropical Geomorphology.British Geomorphological Research Group.

EGGINGTON, H.F. 1966.

Pumping Capacity of the Alice Springs, Mereenie Aquifer. Water Resources Branch, NT Administration.

FERGUSON, K. 1988.

Lithology Sandstones Division,

and Stratigraphy of the Mereenie and Pacoota in the Roe Creek Borefield, Water Resources

Department of Mines and Energy, NT.

HUQ, N. 1980.

Alice Springs Flood Study, Water Division, Department of Transport and works, NT.

JACOBSON, G., CALF, G.E., JANOWSKI, J. AND McDONALD, P. 1989.

Groundwater Chemistry and Basin, Central Australia. 266.

Palaeorecharge in the Amadeus Journal of hydrology, 109: 237-

JOLLY, P.B., PROWSE, G.W. & CHIN, D.N. 1990.

The Amadeus Basin Modelling Based on Sandstone. Water Authority, NT.

Mereenie Sandstone Aquifer Regional the Diagenetic History of the Mereenie

Resources Division, Power and Water

KENNARD, J.M., NICHOLL, R.S. AND OWEN, M. 1986.

Late Proterozoic and Early Paleozoic Depositional Facies of the Northern Amadeus Basin, Central Australia. Geology and Geophysics, BMR, Canberra.

KERSHAW, A.P. 1976.

A Late Pleistocene and Holocene Pollen Diagram from Lynch's Crater, North eastern Queensland, Australia. New Phytol. 77, 469-98.

27

KERSHAW, A.P. 1980.

Evidence for Vegetation and Climatic Change in the Quaternary. In The Geology and Geophysics of North Eastern Australia, R.A. Henderson and P.J. Stephenson {Eds), 398-402, Brisbane. Geological Society of Australia.

LAU, J.E. 1989.

Logging of Diamond Drill Core from Roe Creek Barefield, Alice Springs. Report for Water Resources Division, Power and Water Authority, NT.

LEEDER, M.R. 1985.

Sedimentology, Process and Product. George Allen and Unwin, London.

LLOYD, J.W. AND JACOBSON, G. 1987 .

. The Hydrogeology of the Amadeus Basin, Central Australia, Journal of Hydrology, 93: 1-24.

MABBUTT, J.A. 1977 Desert Landforms, ANU Press, Canberra.

MCDONALD, M.G. AND HARBAUGH, A.W. 1984.

A Modular Three Dimensional Finite Difference Groundwater Flow Model. Scientific Publications, Washington.

MACQUEEN, A.D. AND KNOTT, G.G. 1982.

Groundwater in the North Eastern Part of the Amadeus Basin. Water Division, Department of Transport and Works, NT.

MACQUEEN, A.D. 1978.

Roe Creek Barefield Development. Water Division, Department of Transport and Works, NT.

MAGELLAN PETROLEUM NT PTY LTD.

Well Completion Reports for Oil Wells Alice No 1, 1964, Orange No 1, 1967, Palm Valley No 1, 1965, Palm Valley No 2, 1971, Palm Valley No 3, 1973, West Waterhouse No 1, 1970.

PANCONINTENTAL PTY LTD.

Well Completion Reports for Oil Wells Dingo No 1, 1982, Wallaby No 1, 1982.

28

PROWSE, G.W. AND WIEGELE, A. 1989.

Photography of Diamond Drill Strata Core, RN's 15020, 15021, 15022, 15043 and 15091. water Resources Group, Power and Water Authority, NT.

ROBERTS, K.P. 1974.

Analysis and Prediction of the Effects of Pumping from the Mereenie Sandstone Aquifer, Alice Springs. Water Resources Branch, Department of the NT.

ROBERTS, K.P. 1978.

Alice Springs Town Water Supply, Report on Activities at Roe Creek Barefield 1975 and 1976. Water Resources Branch, Department of the NT.

SCHOLLE, P.A. AND SCHLUGER, P.R. 1979.

Aspects of Diagenesis. Society of Economic Paleontologists and Mineralogists, Tulsa.

TUCKER, M.E. 1981.

Sedimentary Petrology, An Introduction. Blackwell Scientific Publications, Oxford.

VERHOEVEN, T.J., MACQUEEN, A.D. AND READ, R.E. 1979.

Alice Springs Water Supply, Future Development. Division, Department of Transport and Works, NT.

VERHOEVEN, T.J., PAIGE, D.B. AND DONOHUE, L.J. 1982.

Water

Alice Springs: Forward Planning of the Water Supply. Part A: Source. Water Division,· Department of Transport and Works, NT.

WELLS, A.T., FORMAN, D.J., RANFORD, L.C. AND COOK, P.J. 1970.

Geology of the Amadeus Basin, Central Australia, Australia, BMR Geol Geophys Bulletin 100.

WOOLLEY, D. 1966.

Geohydrology of the Emily and Brewer Plain Area, Alice Springs NT, Water Resources Branch, NT Administration.

29

•

' GEOLOGICAL CROSS SECTION CD

Figure 1:

-

GEOLOGICAL CROSS SECTION AB

Regional Geology. of Study Area·

A --- 8 GtOl.OG!CAl CROSS St:CTION

CREEl<

ntYER

• OIL WEll - BOREFIELD

•••~•••••••• SYNCLINE

••••••••i••• ANTICliNE

REGIONAL GEOLOGY OF STUDY AREA

LEGEND

DEVONIAN TO CAROONIFEAOUS

SILURIAN TO DEVONIAN

CAMBRIAN TO OA06VtCIAN

LATE PROTEROZOIC

PROTEROZOIC TO PALAEOZOIC

PH!nlara Group Including Har01anuburg Banllolane

lncludlftO la,.plnU Ofld Perlaaar~ta Groupo

Including Periatalaka lotay0t~go, Bluer Sprln11o Far"'allnno and Haavllrao Ouarulta

... mil!ilii!iiiill_.' <£:p,-e•OI ~

Soutea :- lor,..doua Ouln Eul, Wut t:6000DO Se.la 1870 BMR AUca Bpoln~ 1;2~00011 111&3 2nd ad. DMR r.l&edDrv>aD 1:100000 11181 fii.IR/NTGS

w 15102 S' w

15 ' P27 • E 0

0 15091 ,., 0 0 0

0 15100 ;;; "' " "

'

I .~' ' w '!<.<'' /w w

' E ".J'~ ,/ E E 0 ~'./ g g 0 0 ('';' 0 ~

"' .. ' " " "' ,o // " .. " +· / " "

' '\0 ,/ /

Quarantine Yard ,.. ... --'--......_ 'l j' ~~1~ ',~~~ '15043 -, 15022 I

(''------,-~~~~>,- 8P26 ,--'~' ;/ // , ___ , ___ ,---t-

\ PACOOTA SANDSTONE ' ......... :\ • :~oot;-----. / \ -----1_-~~~ /;,·-----:::--"''P22'\I'l/, 7367000mN I ___ , ,--;;-----·--·- T ~

~~~~---------+~--------~----/ ~------~====--r--r----,_--------------+-----~V9 I ,I-., y / , •••

"T1 G) c :0 m 1\)

I I ', \ // c..4' I I \ "-, /" ...., 1 I '\ , / , ,,r.." ~ MEREENtE SAN'DsroNE \ ', P5 ~r ae•e \ ~ r ' ' ' ~ P 12 "'/ ;t,.o -' I I ' ', Y'X .I &..' I / ', '\. J1l,: ~ P11 ,I(...-PUMPING /" ~~-

~1772 - ..@.5457 : 10~02 I ', \O.tSg \~',~r"V STATION

1

/ (j"(. l -{i)- I '..._ • ... • I / <?-\ P'6 I }~. - p

1'::'5' --i- t!L!'2: ~~ P10', ~ + 3600 ;// t7

P20

7 366 OOOmN

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

---------__ , __ ----

8 P19

~18 : P17 [!] [!] P14 P21r ---.:; ......... ~. _________ ..,../

,sP24 P16 1 P2s :, ',,, 1--~-~;:}-,,-=:::---~-----~=;-----" 15020 HERMANNSBURG SANDSTONE \ ~ '-. ..... f>g-- :...!::~'t;;p:-;1------/--/ !i] P2 \ "-.. '' ----- ·r;; P1A --

' ' ' '

LEGEND CREEK 15020 BORE REGISTERED NUMBER

MAJOR ROAD [!)P2 PRODUCTION BORE 200 0 200 .coo 600 000 1000 ROE CREEK BOREFIELD

MINOR ROAD + WATER LEVEL MONITORING BORE metro a BORE LOCATION MAP TRACK • CORED DIAMOND DRILL HOLE

FENCE ® INVESTIGATION HOLE FIGURE 2 INTERPRETED GEOLOGICAL BOUNDARY

"T1 -· (Q .

NORTH EASTERN UPLIFTED AREA

DEPOSITIONAL ENVIRONMENT UNIT A HIGH S.W.L.

w~------------------------------------------------------------------------------~

"T1 (Q .

LARGE HIGH RELIEF DUNAL SYSTEM

REDUCTION/PLANATION OF UPLIFTED ROCK

DEPOSITIONAL ENVIRONMENT UNIT B LOW S.W.L.

,,

~L---------------------------------------------------------------------------~

, -· (Q .

,, ,,

DEPOSITIONAL ENVIRONMENT UNIT C HIGH S.W.L.

m~------------------------------------------------------------------------------------~

., (Q .

FORMATION BOUNDARY CONTACT

,. •

HERMANNSBURG SANDSTONE

MEREENIE SANDSTONE

m~------------------------------------------------------~

., cc .

" .•

' ' ,, '-0,

.. ·.

BREWER CONGLOMERATE

CLOSE OF BREWER CONGLOMERATE DEPOSITION

.•

T HERMANNSBURG SANDSTONE

+ MEREENIE SANDSTONE

~~----------------------------------------------------------------------------------------~

r-------------------------------------------------------------------------------------------~------------------- --"

PACOOTA SANDSTONE

MEREENIE SANDSTONE OUTCROP

MEREENIE SANDSTONE

HERMANNSBURG SANDSTONE

~­---- ./'

--

~ ~-----------------------------------------------~---- Sand Dunes ~~~

- - "')? > --;::tV > ::::;v>

BREWER CONGLOMERATE

--- -----

~ PLANATED PRESENT AGE LANDSCAPE cg

·a

0 0 0 0.

0

ROE CREEK BOREFIELD AREA 00~--------------------------------------------------------------------------------------J

,, •

~

i~ •· •" .. o, •• co• ..,. "'CI

~· -;rc ~: D 1000

, .B . " ~-10110 2

-:rooo

•3000 • •'• •'• ,, ,', " ... " .. .. ..

' j

II. it ..... .,, .. ,.

GEOLOGICAL CROSS SECTION CD GEOLOGICAL CROSS SECTION AB

LEGEND

A---8 O~OlOOIC"'l -cno~s ~fCilfiH lOHr ' CAlUI -AIY(A

ZOtlf f

Oil. Wlll D ION( ) - I!OIU,II!lO

... , ........ IIYNCl!Nl

.. ...... ; ... "'NTICliN.

"TI -· -• HYDROGEOLOGICAL ZONES IN STUDY AREA

~--------------~----------------------------~--~.. ---. "

-n -· = . -=

•• , . .._ __ ===='·=~'• =-=--;iioo? =--'l"'o . ..,

REGIONAL POTENTIOMETRIC CONTOURS FOR THE MEREENIE SANDSTONE AQUIFER

e 011. W(ll

- IIOIIHIElO

•••;•••••••• sV~tcltHl

••••••••t••• .ONTICliNE

LEGEND

PT• S14NOI-.OW•THI UVH ("'AIIDl

-···- CI'IEE~

-111\IEIO

..., Ci" • --

~ •.

~·~:II>'

"

CUll M

HHI'(tl

Oil WlU

- 80FUftHD

·--•·••••••• &Yt.ICliN(

'"""""i•" AllliCIINL

REGIONAL SALINITY AND RADIO CARBON DATING

LEGEND

'" TOTAl DIIIOLVED IOUDS (..,giO

' ' '1. M00£AN ••C

c J "C .O.G£l'r[MIS 8PJ

POLLEN RECORD 1130000 YEARS! FOR LYNCH'S CRATER. QUEENSLAND !AFTER KERSHAW 1976, 19801

Fig. 12

"'" =· • -Co:l

• P>O

• P7 • •

P19 pIS

\

. •. ~p

Pll p 11 • • • . ,. • "''

.,, • .~

P:13 ~

"' • 3800

•••

ALICE SPRINGS

'"

• ,,

D D ------, L_J

ROE CREEK BOREFIELD tnr.Al MODEL

""

II Of CR[[ K non[ rl[l II f>fi(IOIIC 1 ION H(lllf ~

llfOIONAl. I.IONIIOflltHl nonrs

INFERRED GEOLOGICAL BOUNDARY OF Olll­CROPPING OR SUBCROPPINO MEREENIE SANDSTONE

BOUNDARY Of MODEL

AREA OF UNCONFINED AQUIFER WITHIN MODEL

AREA OF CONFINED AQUIFER WITHIN MODEL

INACTIVE NODES WITHIN MODEL

PROPOSED FUTURE BOREFIELD IN ROCKY IfiLL AREA AS RECOMMENDED VERiiOEVEN 1079

Ill co m ~

0

2

4

6

8

10

12 ....... E ....... 14

z 16 3: 0 18 0 3: 20 <(

a: 22 0

24

26

28

30

32

34

36 ., =· . -""""

,,

•

•

TIME (years) 0 ,... m

0 0 0 0

• • • eo .o 0

• • 0

0

•• • • o

0

•

• •

Ill (I)

m

0

• 0

• • • 0

o ACTUAL ORAWOOWN 3537 • MODEL DRAWDOWN 3537

'"

(EAST OF

0 BOREFIELO) 4693

0 11325

0

2

4

6

8

10

12 .......

14 E .......

(WEST OF 16 Z BOREFIELO) 3:

0 18 0

20 3: <(

2 2 a: 0

24

26

28

--ACTUAL ORAWOOWN 30 ---MODEL DRAWOOWN

32

34

~-~L-------~~~----------~~36 1000 10000 100000

DISTANCE FROM PUMPING CENTRE (m)

COMPARISON OF MODEL AND MEASURED DRAWDOWN

"T1

IP ..... (J1

•

0 .... ~ {>, .....

2 •• ~0~

"' ., :: 1 ., E

4

6

8

0

2

4 z s: 01 0 5: 0.:1 a: 0

2

2

6

8

0

2

24

26

28

30 0

•·• .. t.. o''

~~----...... ... ~

...... . ~··· ..... ~·····~.

~ ..... . ...

..... 0 .......... ~ LINE OF REGRESSION

' ... ...........

~ ..... ............ ~ i/ y,0.11X • 0.908

..... 0 . .. • COEFF. OF CORRELATION 0.996 ..... ... ... .. ........

6~ ..... ............. _ .......... ..... .. ............

~ ..... u ..

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

~ ... ,o . ....

........ ............ ~

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

" ......... ~

- LINE OF REGRESSION - y,o 204X - 0.998

LINE OF REGRESSION ___..-- ""' .......... +

.......... COEFF. OF CORRELATION 0.998

y,0.27X • 0.615

""' :...

COEFF. OF CORRELATION 0.997

~ s "-.........

20 40 60 80 100

CUMULATIVE EXTRACTION FROM MEREENIE lx106Jm3

• BORE 3609 !assumed pre-pumping SWL 458.68m AHDI

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

o' ,__ -o --· ..... .....

" ..... ..... .....

120

~

u

140

o BORE 5731 & 5798 !assumed pre-pumping SWL 455.20m AHDI 6 BORE 3600 I assumed pre-pumping SWL 457 65m AHDI PLOT OF BOREFIELD ORA WDOWN v's

BOREFIELD CUMULATIVE EXTRACTION

....... E ....... z ~ 0 0 ~ <( cr: 0

-en

,,

CUMULATIVE EXTRACTION FROM MEREENIE (x 1 06 )m3

0o~--------1~o~o ________ ~2~o~o~------~3~o~o~-------4~o~o~------~5~o~o~------~6~oo 2020

50

100

150

.,./.,/

.,.,­... .....­.,.,..,.,..

A·································· 2oos • • • • • , .

/ : , : ..................... .L .................... .

Y=0.27X+0.615 COEFF. OF CORRELATION 0.997

(/) cr:

1990<( UJ >-

1975

200~--------~----------~----------L---------~----------~------~~1960

---HISTORICAL EXTRACTION DATA

PREDICTED EXTRACTION BASED ON 2% GROWTH AND ALL INCREASE COMING FROM MEREENIE

MONITORING BORE 3600 (assumed pre pumiJ:ing SWL 4 57 .65m AHD)

GRAPHICAL RELATIONSHIP BETWEEN ANTICIPATED BOREFIELD EXTRACTION AND RESULTANT BOREFIELD DRAWDOWN

1000r-------------------------------------------------------------------,

0

-500

ROE CREEK BOREFIELD I I

0 I ..: -1000

E - -1500 _, a:

0 I ..:

E _, a:

0 I ..:

E _, a:

-2000

-2500

-3000

-3500L-----------------------------------------------------------------~

1000

0

-500

1. THROUGH ROE CREEK BOREFIELD

ALICE OIL 1 OORAMINNA RANGES I

.................. . - . . . . . . ......... . . ..... .. . .. . ..

-1000L-----------------------------------------------------------------~

2. THROUGH ALICE OIL 1 WELL

1000 ROCKY HILL BOREFIELD

(PROPOSED) WALLABY 1

500 I I I .. . ..

0

-500

-1000L-----------------------------------------------------------------~

3. THROUGH WALLABY 1 WELL AND PROPOSED ROCKY HILL BOREFIELD

km 0 2 3 4 5 km

~ PROBABLE DEPTH OF ECONOMICAL WATER EXTRACTION

Gill] MEREENIE SANDSTONE

NORTH - SOUTH CROSS SECTIONS THROUGH MEREENIE SANDSTONE

Fig. 17

, (Q

)> .....

•

Old South Road 12/10/1994

460 ro----~--r----·r-------l--~· ~ ~r:·~---~----r--Fl

I I I I I I I

458. ~.o.__ -----~ _ __ _ __ J____ ----4---~-- : I

I I l ,

I I ---~------- ------ !

I I .

6 ! I I I ! I i i !

4541 --~~- .• ~ ~-r-- - . -. ---,-1---- ---------r----~

452 . r~__~__ ____________ __________ ______ __________ I ___

1

_ ,.-~t~-~

450. t~]~~- ~ ~= .~-=t-=t~=-~ ·=-~ =~=-.~.····~---' 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988

, )> 1\)

• • •

Old South Road 12/10/1994

460 ·F-L-H~--~~~=L--~~-----==l-~=-~--4 I I I I i I I I

455 "Ia_ i - --- -"" r=- ~r-----~L----r ! I

450.r-+ . • . I -+----1------~---t--t !

445 . L ___ _______ ___ _ ____ 1~ ----11----H

I I I

440. ) --~- _l_ ' : i I !I I I ,

435. b.o__-f---------------------- --------~-------~-------- ----l-----Hi I I I .

430.t0 :~_~=j--· -----=---~--- ]~~r~ _ _j 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988

, -· cp l> (..)

•

Old South Road 12/10/1994

~----~--------r·------r··--------r-----1-------T-----T----~--------,----_ -r-·T--il

I I i i ! I I I il i I I -----~~~-------L! i I i i i

450 . r--+-----~---:,·------1--------t---_j----t--+---+----~ ~------~! I ! I . I i i I

I I I ; I I i I I I ! I ' i ' I I ' ' I I

I I i l i I ! I I ! I 445. H--j_---r-----1---------~-- --1----1----r i .

I ' I I • I. I 440.r---!---+--+--+--+·-~---t • ~-1-.4697 J :,

l J J -L ... L .1_j ___ 1 __ [ ___ [ L~J-=J .. _U 1966 1968 1970 1912 1974 1976 1978 1980 1982 1984 1986 1988

•

Old South Road 12/10/1994

465 .I --__ - -~----_:-_-_.F_::..-:...-=r.==--·==-1= ===- -·-·-- ==r=-=-:~=r----f--

460. ~ --t---- -+--+--- ----1---1 ··--t~--~-~ 455 .L --Lt-----1-----H

I I I ! I , I

450L __ --+-- ---+- -----l---1- -~-----1---f-; 445 .

1

J]__ ·-··-- ____

1

__________ __[ _ _L __ _I _____ · _j ____ ·+·-·--L·~' I

I I I

i I I I 440. ho_ ______ _ __ t-·- . --t==:-~--l I

435. __ L ____ -~_L ________ -~~ ·--~ ___ ~L __ t..=: =!~ L._ . ., 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988

cc .

, -· (Q

l> c.n·

• .•

Roe Creek Well Field 12/10/1994

' ----r-----·r------r·-------~-.. -------r------~----- ----- --·-·- .. ------,-----·~------r-"1

460. ! .a , ··---t-··--T-.. ---r-- ----~---T--~------.. -l-----~--1---i I L I i! I ! I i

450. ______ . ___

1 __ .. ____

1 _______

1_ .. ____ __ __ _ _ .. _ -·-- · ----1-------

1

.-----_j_ __

I I I I I I I ' i I I i ! I I I I I I I I i I

440 .

1

Q _ _j____ --~-~-~-- ----~--r . . 430 • ~L_ ---j--t--- I +- I _ _j___·---i----+--- ' i

, ! ! I I I I I II ! 1 I

I I I l I I 'l i I I :

420. ~-+-----~--------~---+--- --+- I --+--L--l~~- I I I ' ' ! I I I 1 I I J I r-· ___ _J~ !

' I I i I ! 3~00 ' I ~ 410. Q ___ , ______ _j ________ -----·------ -- -- , __ _L ____ L--:.:::::::. ·-·.=:d. I

··---·· •• 1 ... _ ••••••. ..1. ______ .... 1 ............. ! ····--·····-· j _______ --·------ ____ , ___ j ______ _[ _______ ----···~t _____ ] ____ _ 1962 1964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988

"T1 cc )> (7)

•

NT Water Resources Period 30 Year Plot Start 00:00_01 /01/1965 Interval 15 Day Plot End 00:00_01 /01/1995

463 1 RN005457 115. 00 Line l..ewl 2 RN011772 115.00 Line Level

453

423

413

I NIDI IAHDI

H'II'I.OT we 0111p11 0210211015

1965

RN005457.GW RN011772.GW

"T1 -· (Q . )> ........

Pine Gap

460. F=~- ~=:-- --------=~t- --------------+----------11--

--

~ .

12/10/1994

=r==--~­---t-

I I ·~~-- --r-----r----

1 -·---1 440. I ,-·--·- - - _J r~--T- __ ,. ___ ' 435. ~-~- ___ _ _ T------~--- _

425. _-1:=]=--==--=J--=-L ____ =-~--- ~J=~-~~:_1_ 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988