Results from the Drilling of Two Wells in the Upper Quaternary Aquifer at the Urrbrae Wetland Site. Paul Pavelic, Peter Dillon and Karen Barry

Transcription

1 Results from the Drilling of Two Wells in the Upper Quaternary Aquifer at the Urrbrae Wetland Site Paul Pavelic, Peter Dillon and Karen Barry CSIRO Land and Water Client Report December 2005

2 Copyright and Disclaimer 2005 CSIRO To the extent permitted by law, all rights are reserved and no part of this publication covered by copyright may be reproduced or copied in any form or by any means except with the written permission of CSIRO Land and Water. Important Disclaimer CSIRO advises that the information contained in this publication comprises general statements based on scientific research. The reader is advised and needs to be aware that such information may be incomplete or unable to be used in any specific situation. No reliance or actions must therefore be made on that information without seeking prior expert professional, scientific and technical advice. To the extent permitted by law, CSIRO (including its employees and consultants) excludes all liability to any person for any consequences, including but not limited to all losses, damages, costs, expenses and any other compensation, arising directly or indirectly from using this publication (in part or in whole) and any information or material contained in it. Water for a Healthy Country The Water for a Healthy Country Flagship is a research partnership between CSIRO, state and Australian governments, private and public industry and other research providers. For more information about the Flagship go to: Cover Photograph Rotary mud drilling of a production well at the Urrbrae wetland site. The image shows the drilling operator (Mr. George Michopoulos) on the platform controlling the rate of penetration of the drill-bit whilst his co-worker (Mr. Adam Michopoulos) collects the cutting samples by screening the drilling fluid flowing out of the hole. Cover Photograph: From CSIRO Land and Water Photographer: Paul Pavelic

3 Results from the Drilling of Two Wells in the Upper Quaternary Aquifer at the Urrbrae Wetland Site Paul Pavelic, Peter Dillon and Karen Barry CSIRO Land and Water, Adelaide CSIRO Land and Water Client Report December 2005 Page i

4 Acknowledgements This work was financially supported by the Water for a Healthy Country Flagship, Australian Centre for International Agricultural Research (ACIAR), City of Mitcham, Patawalonga Catchment Water Management Board and the River Murray Catchment Water Management Board. We would like to thank: Dr. Declan Page (CLW) for his assistance during the coring phase of the drilling; Stephen Howles (DWLBC) for generously assisting in the analysis of the pump test data with the PARADOX software program; Hajrudin Zulfic (DWLBC) for providing data and Dr. Allin Hodson (Urrbrae Agricultural High School and Urrbrae Wetland Education Centre) for arranging site access and logistical support. Page ii

5 Executive Summary This report documents the results of drilling, completion and testing of two adjacent shallow production wells to depths of around 25m as part of a study to demonstrate the water quality requirements for sustainable stormwater ASR operations in alluvial aquifers. Due to the lower than anticipated transmissivity and extreme heterogeneity of the upper Quaternary aquifer at the Urrbrae site, the likely scale of ASR is considered too small to justify proceeding with an ASR trial in this aquifer via these wells. Drilling of further production wells with a higher anticipated yield in the deeper Tertiary sands will be required in future at this site. Page iii

6 Table of Contents 1. Introduction Hydrogeological setting and aquifer selection Drilling, core sampling, geophysical logging and well completion Aquifer pump testing Groundwater quality Conclusions and future work References...6 Appendix 1: Geological descriptions... 7 Appendix 2: Drillers logs... 9 Appendix 3: Particle size distribution data for three samples Appendix 4: Geophysical logs Appendix 5: Step-drawdown pumping test data Appendix 6: Groundwater quality data Appendix 7: Selection of photographs from different stages of well... construction and testing List of Figures Figure 1. Location of Quaternary Wells A, B and C and pre-existing deep ASR well (Source of Basemap: Dr. Allin Hodson, Urrbrae Wetland Education Centre). The map shown in the lower right hand corner indicates that the study site lies on the margin of an area where the upper Quaternary aquifer is suitable for small-scale ASR. (Pavelic et al, 1992)...2 Figure 2. Design of Wells A and B... 4 Page iv

7 1. Introduction International experience has shown that higher levels of pre-treatment of source water are required for injection into alluvial aquifers than for limestone aquifers to avoid excessive well clogging problems (Pavelic and Dillon, 1997). Most stormwater ASR operations in Australia target limestones as they are more tolerant of poorer source water quality due to the offsetting effect of matrix dissolution on clogging. Opportunities to enhance groundwater resources are being lost in regions where unconsolidated alluvial aquifers represent the predominant target unit and stormwater runoff is in seasonal abundance. Such areas include Adelaide, Melbourne, and the Philippines. The only known attempt to inject passively-treated stormwater into an unconsolidated aquifer at the Urrbrae wetland site in metropolitan Adelaide in the late 1990s, resulted in a significant decline in recharge rates and the cessation of operations within the first season. Exhaustive efforts using chemical and mechanical methods to remediate the clogged well failed. Consequently a series of research projects have commenced to address the design of ASR wells (Segalen et al, 2005) and on methods of pre-treatment to improve the quality of the injectant, including the use of roughing filtration (Lin et al, in prep.) and biofiltration (Page et al, in prep.). These methods were selected for their simplicity, low cost and potentially low maintenance requirements, making them suitable for use in developing countries and in Australia for urban stormwater harvesting. The current research project at the Urrbrae wetland site aims to demonstrate the requirements for sustainable ASR operations with urban stormwater in alluvial aquifers by incorporating these two additional passive pre-treatment steps aimed at removing colloidal matter and key bio-available nutrients from the recharge water. This report documents the results of drilling and testing of two upper Quaternary wells at the Urrbrae wetland site. Whilst these wells were constructed with the intention that they serve as ASR wells, it will later be shown that poorer than anticipated yields will necessitate further drilling of production wells in a deeper formation at this site in future to progress the broader study. 2. Hydrogeological setting and aquifer selection It was decided to target the upper Quaternary aquifer rather than the deeper Tertiary Carrisbrooke Sand and Port Willunga Formation of the previous ASR trial on the basis of the lower cost (30% of the cost of the deeper wells), the acceptable rates of injection achieved in the upper Quaternary only 400m away at the Kingswood rainwater ASR site (Barry and Dillon, 2005), and the geological log of a pre-existing shallow well at the study site that had indicated 4.5m of gravel within the saturated part of the profile. The aquifer targeted for ASR is situated within the first 25 m of the Hindmarsh Clay Formation, which is known to extend to 60m in the local area (John Botting and Associates Pty Ltd and Lisdon Associates, 1998). Within the study area the aquifer consists of thin bands of interbedded sand and gravel amongst clay deposits of fluvial origin (Gerges, 1999). Groundwater usage is limited due to brackish water quality and relatively low yield (<2 Ls -1 ) (Pavelic et al, 1992). The depth to water table in the area ranges from approximately 11 to 13m below ground surface (Barry and Dillon, 2005). The new wells were drilled by Olympic Boring Contractors (drilling supervisor/operator: Mr. George Michopoulos) as replicate wells only 4.8 m apart at the southern end of the rubber-lined detention basin and as close as possible to the available power board and concrete pad (Figure 1). They have been assigned Department of Water Land and Biodiversity Conservation (DWLBC) Unit Numbers (UN) and but for convenience will be referred to as Wells A and B respectively. The pre-existing observation well (Well C; UN ), drilled in 1996 and completed to a depth of 17m, is situated around 30m down-gradient (north-west) of Well B and adjacent to the wetland (Figure 1). 1

8 Figure 1. Location of Quaternary Wells A, B and C and pre-existing deep ASR well (Source of Basemap: Dr. Allin Hodson, Urrbrae Wetland Education Centre). The map shown in the lower right hand corner indicates that the study site lies on the margin of an area where the upper Quaternary aquifer is suitable for smallscale ASR. (Pavelic et al, 1992). 2

9 3. Drilling, core sampling, geophysical logging and well completion The choice of drilling technique and method of completion were recognized to have a significant effect on the hydraulic performance of ASR wells, particularly in low-transmissivity environments. Recommendations made by Segalen et al, (2005) that ASR wells should be rotary drilled and subsequently reamed and completed with the largest aperture stainless steel screen suitable for the formation were followed in an attempt to maximize well yields. Drilling operations commenced on 12 September 2005 and concluded on 21 September For both wells the top 12m was drilled with hollow-stem rotary mud drillrig with 229mm (9-inch) drill pipe. Samples of cutting material were collected from the recirculating mud every 0.5m. Only a small amount (~0.5kg) of the biodegradable polymer mud ( Bio-Vis, Australian Mud Company Ltd) was used due to the high clay content within the profile. Below 12m an alternative rig with 203mm (8- inch) split-core sampler and wire-line recovery system was used to collect continuous core samples. Coring terminated at 24m for Well A and at 22m for Well B (2m less than Well A due to timeconstraints). Recoveries from the 0.5m length core barrel ranged from 60% to 100% for Well A and 20% to 100% of the cored interval. Both holes were then carefully reamed to a diameter of 229mm and depth of 24m. Following completion the residual mud was removed by airlifting until clear water was recovered. Detailed geological logs derived from the cutting and core samples are given in Appendix 1. The Drillers Well Construction Report for Wells A, B and C are given in Appendix 2. Particle size distribution (PSD) data for 3 samples are given in Appendix 3. The Well A profile consists largely of mottled sandy clay grading to clay throughout most of the saturated part of the profile. A thin band of clayey gravel from m was the only interval to encounter any substantial fluid loss during drilling suggesting it to have the highest permeability. Well B consists largely of sandy clay grading to clay interspersed with discreet lenses of sand between 17 and 19m depth. The lower core recovery for this well is indicative of more permeable sediments. The contrast in the geological profiles between the two wells over a lateral distance of only 5m suggests significant spatial variability over even small separation distances. This is characteristic of fluvial environments where the transport of sediment by water in the landscape can produce significant vertical and lateral heterogeneities, often associated with specific depositional facies. For example, the permeable zones identified at m in Well A and between 17 to 19m in Well B appear not to be laterally connected. The Geophysical Technical Services group of DWLBC ran the following suite of logs down both holes prior to well completion: Gamma, Neutron, Spontaneous Potential, Point Resistance, Resistivity, Density and Caliper. The Gamma and Neuton logs in particular, were used in conjunction with core samples to identify the most permeable parts of the formation for the selection of the interval to install the screen. Plots of the Gamma and Neuton logs are given in Appendix 4. A standard screen length of 6m and aperture of 0.8mm was used in both cases. This aperture size is greater than that suggested by the PSD data (~0.3mm based on 50% retention of aquifer particles; Appendix 3) so as to maximize well efficiency. On the basis of the core samples and geophysical logs the screened interval was set at 11-17m below ground level for Well A and from 13-19m for well B. Completion of the wells was carried out with 152mm (6-inch) diameter Class 12 PVC casing attached to a 6m continuous length of 203mm (8-inch) diameter, 0.8mm aperture wire-wrapped stainless steel screen with a 1 m sump at the base. Artificial gravel packs with grainsize diameters of 5-7mm were used to stabilize the aquifer and minimize the pumping of sand. Both wells included a 25mm conduit pipe attached to the outer side of the casing and screen and slotted over the screened interval to allow monitoring of piezometric pressures within the gravel pack and in the event that jetting of the screens and gravel pack became necessary. A 300mm bentonite seal was 3

10 placed directly above the gravel pack and the remaining annular space around the casing cemented to the surface. The total depth of Well A was 19 m and Well B was 21m (including the sump). Thus the design of both wells were identical in all respects apart from the 2 m shallower screened interval and total depth for Well A (Figure 2). The wells were developed by airlifting before and after installation of the casing. The final air lifting was for half a day each and both wells largely produced clear water. The final airlift yields obtained by the driller were 0.13 Ls -1 (100 gallons/hour) for Well A and 0.19 Ls -1 (150 gallons/hour) for Well B. Neither well experienced problems with pumping sand implying the selection of the larger aperture size appropriate. Well A Well B 0 cement Depth bgs (m) bentonite access tube screen gravel pack sump 20 Figure 2. Design of Wells A and B 4

11 4. Aquifer pump testing The aquifer transmissivity was determined through a three-stage step-drawdown test. This is a test in which the well is pumped at a constant flow rate until the rate of drawdown begins to stabilize and then the pumping rate is increased, and so forth until pumping stops and the piezometric head is then allowed to recover. The length of each step was 30 minutes and the pumping rates were Ls -1 in the first step, 0.15 Ls -1 in the second and 0.21 Ls -1 in the third. Both wells were tested on 20 October 2005 after a preliminary test was conducted earlier in that week to set pumping rates and gauge the likely response. A 2.7 hour rest period between tests allowed waterlevels to fully recover. The initial depth to watertable on the day was 10.70m below ground surface (bgs) for Well A and 10.76m bgs for Well B. Drawdown data are presented in Appendix 5. For Well A, the drawdown reached 5.5m, the depth of the pump, after 42 minutes of pumping and the test was prematurely stopped. For Well B, the total drawdown after 90 minutes of pumping at rates of up to 0.21 Ls -1 was 4.6m. The estimated transmissivity from the graphical Hazel method is <1 m 2 day -1 for Well A and ~1 m 2 day -1 for Well B. Some uncertainty in the transmissivity value for Well A is suggested by the non-linear relationship between the inverse specific capacity and flow rate (used to derive one of the well equation coefficients). This may be due to boundary effects, the multi-layer nature of flow through the aquifer, the varying length of screen contributing flow due to dewatering or a combination of these factors. Both values are extremely low irrespective of this uncertainty. The maximum indicative pumping rate, based an available drawdown of 5m for Well A is <0.1 Ls -1 and 7m for well B is ~0.15 Ls -1. These values are probably upper limits for injection since they neglect the clogging which could result in increased headloss and flow rate declines of up to 50%. However during injection the full screen length will be submerged so it is possible that this effect could offset clogging. The likely volume of water that could be injected into Well B assuming four months of continuous injection at 0.1 Ls -1 is only around 1ML/yr. The low yields overwhelmingly reflect the absence of substantial permeable deposits within the top 25 m of the profile intersected by these two wells and the resultant low transmissivities. 5. Groundwater quality A water sample was collected during each stage of the pumping test for the analysis of EC and turbidity (Appendix 6). In both cases the salinity of the groundwater was brackish with EC values of around 3900 μs/cm. Marginally lower values during the first pumping step indicates a small volume of residual fresh wetland water introduced during drilling was rapidly flushed. Water extracted from Well B was initially relatively low in turbidity (3 NTU), but increased with each successive step up to 17 NTU in the last sample due to greater mobilization of fines with increased pumping rate. Extracted water from Well A was far more turbid than Well B ( NTU), with values also increasing in the second step. The higher turbidity for Well A is attributable to the higher clay content in the profile (Appendix 1). Downhole EC profiles from Well C range from 3900 to 4100 μs/cm with depth below the top 2m which appears to have been due to the absence of a cap having allowed rainwater to enter the well. The temperature is stable with depth at 19 o C. At the Kingswood ASR site the corresponding values for ambient groundwater were 4000 to 4200 μs/cm and ~20 o C (Barry and Dillon, 2005). This implies reasonable spatial uniformity in the hydrochemistry in the local area. 5

12 6. Conclusions and future work At the Urrbrae wetland site the upper Quaternary aquifer is of lower than anticipated transmissivity and extremely heterogeneous, even over spatial scales as small as 5m. The likely scale of ASR is considered too small (1 ML/yr or less) to justify proceeding with a trial in this aquifer via these wells. Drilling of further production wells to the Tertiary sands at depths of 80-90m, with an anticipated yield of around 3 Ls -1, will be required at this site to progress the broader objectives of the project. This activity will be delayed for at least 12 months while current efforts focus on refining and testing pre-treatment methods and in establishing a related ASR trial at the Rossdale site in Melbourne (CSIRO and SKM, in prep.). 7. References Barry, K and Dillon, PJ (2005) Domestic-scale ASR with rainwater at Kingswood, South Australia. Proceedings of the 5 th International Symposium on Management of Aquifer Recharge (ISMAR5), Berlin, June, CSIRO and SKM, (in prep.) Developing aquifer storage and recovery (ASR) opportunities in Melbourne: Preliminary hydrogeological assessment of ASR at Rossdale Golf Club. Gerges, NZ (1999) The geology and hydrogeology of the Adelaide metropolitan area. PhD Thesis, Flinders University of South Australia. John Botting and Associates Pty Ltd and Lisdon Associates, (1998) Urrbrae Environs Water Management Plan. Stage 2A Aquifer Injection Investigations. Unpublished Report. Lin, E, Hutson, J, Le Gal La Salle, C, Dillon, P, Page, D and Pavelic, P (in prep.) Assessment of roughing filtration for pre-treatment of urban wetland waters, Proceedings of the 4 th International Slow Sand and Alternative Biological Filtration Conference, 3-5 th May 2006, Müllheim and der Ruhr, Germany. Page, D, van Leeuwen, J, and Dillon, P (in prep.) Review of slow sand filtration, CSIRO Land and Water Technical Report. Pavelic P, Gerges NZ, Dillon PJ and Armstrong D (1992) The potential for storage and re-use of Adelaide s stormwater runoff using the upper Quaternary groundwater system. Centre for Groundwater Studies Report No. 40. Pavelic, P and Dillon, PJ (1997) Review of international experience in injecting natural and reclaimed waters into aquifers for storage and reuse. Centre for Groundwater Studies Report No. 74. Segalen, A-S, Pavelic, P and Dillon, PJ (2005) Review of drilling, completion and remediation methods for ASR wells in unconsolidated aquifers. CSIRO Land and Water Technical Report 04/05. 6

13 Appendix 1: Geological descriptions Table A1. Geological log for Urrbrae ASR Well A (UN ; PN ) Depth bgs (m) SI* Description Sample Type Subsample ID / Depth Clay, dark brown Cutting Clay, dark brown Cutting Clay, brown, with intercalated sand/gravel Cutting Clay, brown/grey mottling, Cutting Clay, brown/grey mottling, slight gravel Cutting Clay, brown/grey mottling, slight gravel Cutting Clay, brown/grey mottling Cutting Gravelly clay, red brown Cutting Clay, grey/brown, some gravel/boulders Cutting Clay, slight gravel content Cutting Clay, slight gravel/boulder content Cutting Clay, slight gravel/boulder content Cutting Gravel/boulders, grey/brown, some clay Cutting Clay, brown/grey mottling Core A1 ( m) Clayey sand, slight gravel, red brown Core A2 ( m) Clay, grey/brown mottling Core A3 ( m) Clay, grey/brown mottling (60% core recovery) Core A4 ( m) Sandy clay, intercalated gravel more prevalent towards bottom of core Core A5 ( m) Sandy clay to clayey sand, grey/brown mottling Core A6 ( m) Sandy clay to clayey sand, grey/brown mottling Core A7 ( m) Sandy clay, red brown/orange/grey mottling Core A8 ( m) Sandy clay, intercalated gravel to sandy clay, red Core A9 ( m) brown/orange/grey mottling Sandy clay to clayey sand, red brown/orange/grey mottling Core A10 ( m) Clay, grey/brown mottling to un-mottled grey clay (70% core recovery) Core A11 ( m) Sandy clay, red brown/orange/grey mottling Core A12 ( m) ** Sandy clay, intercalated gravel, friable red/grey mottling Core A13 ( m) Clay, grey/brown mottling Core Sandy clay, grey/brown mottling Core A14 ( m) Sandy clay, grey/brown mottling (70% core recovery) Core A15 ( m) Sandy clay, variably friable, grey/brown mottling Core A16 ( m) Sandy clay, grey/brown mottling, moderately consolidated Core A17 ( m) Sandy clay, grey/brown mottling with some unidentified Core A18 ( m) black precipitate Sandy clay, grey/brown mottling, friable Core A19 ( m) Sandy clay, grey/brown mottling, friable plus consolidated stone with some unidentified yellow precipitate Sandy clay, grey/brown mottling, with some white (carbonaceous) precipitate Sandy clay, grey/brown mottling, friable, with some white (carbonaceous) precipitate Sandy clay, grey/brown mottling, friable, with some white (carbonaceous) precipitate Core Core Core Core A20 ( m) A21 ( m) A22 ( m) Sandy clay, grey/brown mottling Core Sandy clay, grey/brown mottling, friable Core A23 ( m) Sandy clay, grey/brown mottling, friable Core A24 ( m) Sandy clay, grey/brown mottling, friable Core A25 ( m) Sandy clay, brown/grey mottling, friable Core A26 ( m) * shading represents screened interval ** particle size distribution given in Appendix 3 7

14 Table A2. Geological log for Urrbrae ASR Well B (UN ; PN ) Depth bgs (m) SI* Description Sample Type Subsample ID / Depth Clay, dark brown Cutting Clay, plastic, dark brown, slight gravel Cutting Clay, red brown, with intercalated sand Cutting Clay, red brown Cutting Clay, red brown/grey mottling Cutting Clay, red brown/grey mottling Cutting Clay, brown/grey mottling Cutting Clay, brown/grey mottling, slight gravel Cutting Clay, brown/grey mottling, slight gravel Cutting Clay, brown/grey mottling Cutting Clay, grey/brown mottling, slight gravel Cutting Clay, grey/brown mottling, slight gravel Cutting Clay, grey/brown mottling (60% core recovery) Core B1 ( m) Sandy clay to clayey sand, grey/brown mottling with some white (carbonaceous) precipitate Core B2 ( m) Clay, grey/brown mottling Core Clay, grey/brown mottling with some large white (carbonaceous) precipitate Core B3 ( m) Sandy clay to clayey sand, grey/brown mottling with some boulder (non-carbonaceous) at 14.1m Core B4 ( m) Clay, moderate gravel, grey/brown mottling Core Sandy clay, grey/brown (60% recovery) Core B5 ( m) Clay, grey/brown then grading to red mottling below 15.05m Core B6 ( m) Clay, grey/brown mottling Core B7 ( m) Clay, grey/brown mottling Core B8 ( m) Clay, grey/brown mottling grading to sandy clay and brown sand at bottom of core (80% recovery) Core B9 ( m) Clayey sand, grey/brown mottling Core Sand, red brown Core B10 ( m) Clayey sand, grey/brown mottling Core ** Sand Core Gravelly sand and clay Core B11 ( m) Clayey sand, grey/brown mottling Core Clay, grey/brown mottling Core Sand Core B12 ( m) Clay, grey Core Clay, grey/brown mottling (70% recovery) Core ** Sand grading to clayey sand Core B13 ( m) Clay, grey/red brown mottling Core B14 ( m) Sandy clay, grey/brown mottling Core Clay, grey/red brown mottling Core B15 ( m) Clay, grey/red brown mottling Core Sandy clay, consolidated Core Clay, grey/red brown mottling Core B16 ( m) Clay, grey/brown mottling (20% recovery) Core Clay, grey/red brown mottling Core B17 ( m) Clay, grey/red brown mottling Core B18 ( m) Clay, grey, highly friable grading to consolidated carbonate Core B19 ( m) Clay, grey/red brown mottling Cutting Clay, grey/red brown mottling Cutting * shading represents screened interval ** particle size distribution given in Appendix 3 8

15 Appendix 2: Drillers logs Well A 9

16 Well B 10

17 Well C 11

18 Appendix 3: Particle size distribution data for three samples Well A: m bgs Well B: 17.5 m bgs Well B: 18.5 m bgs Size (um) Mass (g) (%) Cum. (%) (%) Cum. (%) Mass (g) (%) Cum. (%) (%) Cum. (%) Mass (g) (%) Cum. (%) (%) Cum. (%) < (1) >< (2) > < (1) > < (2) > < (1) > < (2) > > ) Plotted data excludes >200um fraction: 2) Plotted data includes >200um fraction: Total mass <2000 um (g): Mass of sample (g): Mass recovered after sieving (g): Cum % passing Well A: m bgs including >2000um excluding >2000um Size (um) Cum % passing Well B: 17.5m bgs including >2000um excluding >2000um Size (um) Cum % passing Well B: 18.5m bgs including >2000um excluding >2000um Size (um) 12

19 Appendix 4: Geophysical logs Well A Gamma (API) Well B Gamma (API) Depth (m) Depth (m) Neutron (CPS) Neutron (CPS) Depth (m) Depth (m) Screened interval Gamma Neutron 13

20 Appendix 5: Step-drawdown pumping test data Well A step-drawdown pumptest data 6 5 Pumping 0.1 L/s 0.15 L/s Recovery Drawdown (m) well A (logger) well B (logger) well B (probe) well A (probe) Time (minutes) Well B step-drawdown pumptest data L/s Pumping L/s Recovery 4 Drawdown (m) well B (logger) Well A (probe) well C (logger) well B (probe) Time (minutes) 14

21 Appendix 6: Groundwater quality data Table A6.1. EC and turbidity of extracted water during pumping tests in Wells A and B. Parameter Well A * Well B * EC (μs/cm) Turb (NTU) 3490 (21 mins) 3940 (37 mins) 180 (21 mins) 438 (37 mins) 3740 (28 mins) 3940 (49 mins) 3860 (85 mins) 2.9 (28 mins) 6.9 (49 mins) 17.5 (85 mins) * The times shown in brackets are the duration into the pumping test Table A6.2. EC and temperature profile of Well C on 20 October 2005 Depth below TOC (m) EC * (μs/cm) Temp ( o C) ** * Fresh water in top 2 m may be due to absence of cap on well that allowed some rainwater to enter ** Bottom of the well encountered at m 15

22 Appendix 7: Selection of photographs from different stages of well construction and testing Rotary drilling, 0-12m Rotary drilling, 12-24m Drillhole cuttings, 0-12m Collection of a core sample Intact core sample, Well A ( m) Intact core sample, Well B ( m) Geophysical testing of Well B Pump testing of Well B (Well A in foreground) 16