7.

SOILS, GROUNDWATER AND SALINITY

7.1 SUMMARY This section presents the results of soils, salinity, groundwater and preliminary geotechnical investigations carried out in the North and South Dunheved Precincts, St Marys NSW. It addresses the soils, groundwater and salinity control requirements of the St Marys Environmental Strategy 2000 (EPS 2000) and the Sydney Regional Environmental Plan No 30 St Marys (SREP 30). The report is based on walkover observations; a review of previous groundwater and geotechnical reports; the logs of 20 test pits and 5 shallow boreholes; laboratory and field testing; and an electromagnetic (EM) geophysical survey to map the ground salinity. The Dunheved site covers about 30ha of high-level alluvial terrace, composed of very stiff to hard silty clay with depths of more than 3m and up to 8m, but averaging 5-6m. Patches of ironstone gravel may be present close to the base of the clay. The underlying bedrock, which is nowhere exposed within the two precincts, is weathered shale. Groundwater inflows were not encountered in any of the test pits, which were mostly about 3m deep, and only in one borehole (at depth 6.5m). The five boreholes were drilled to depths of 6-11m and two were extended into shale bedrock by rotary coring. The salinity and standing level of the groundwater could not be measured at the time of drilling, but two boreholes were completed as observation wells (piezometers). Subsequent measurements recorded water at a depth of 4.6m in BH1 and no water in BH2. The salinity of the groundwater in BH1 was found to be in the order of 3500mg/L total dissolved salts (TDS), which is in the lower range for the St Marys property. Field and laboratory measurements of soil conductivity indicate that the soil profile is generally non-saline down to a depth of at least 1m. An EM survey carried out by Douglas Partners suggests greater subsurface variations, but generally non-saline to low salinity ground conditions down to a depth of 5-6m. The reason for this variability is that the EM equipment measures apparent (bulk) conductivity and is influenced by factors such as clay content, soil depth, degree of saturation and ironstone content, as well as soil and groundwater salinity. Hence salinisation is unlikely to affect the surface of the completed Precincts (which would be 4-5m above the low salinity water table when the depth of filling is included). Although soil salinity is unlikely to pose environmental problems at Dunheved, we suggest a number of groundwater management measures to meet the requirements of SREP 30 and the St Marys EPS 2000. The prime purpose of these measures is to reduce runoff infiltration and so keep the water table as low as possible.

SINCLAIR KNIGHT MERZ I:\INFR\Projects\IN08708\Deliverables\Reports\Final 28 April\Final Report v06.doc PAGE 71

The results of soil testing on four bulk samples from test pits indicate that the alluvial terrace clay is generally of medium plasticity, low dispersivity and only moderate shrink-swell potential. The latter was confirmed by field observations of widely spaced and narrow surface cracking, despite the present drought. Foundation conditions at Dunheved are greatly influenced by the thick layer of very stiff silty clay overlying low strength weathered shale bedrock. The clay depth is generally greater than 3m, averages perhaps 5-6m, and was found to be up to 8m in one borehole. The presumed bearing capacity of the clay is at least 200kpa. The underlying shale is presumed to be equivalent to foundation Class IV, of allowable bearing capacity 1Mpa. For conceptual design purposes, bored piles founded at depth 12m are suggested. However this must be confirmed by site-specific investigations prior to building foundation design. We understand that most of the Dunheved site is to be filled to a depth of 1-1.5m. The in situ alluvial clay is expected to provide a good foundation for this embankment. Provision should be made for groundwater drainage beneath the fill; this could most readily be achieved by placing free-draining rockfill in the lowest of the embankment layers. 7.2 INTRODUCTION This report presents the results of preliminary soils, salinity and groundwater investigations carried out in the Dunheved Precincts within the St Marys Property (SMP). The work was carried out by Sinclair Knight Merz (SKM) P/L for the Maryland Development Company as part of a multidisciplinary planning study for a proposed employment subdivision. A feature of this subdivision will be land filling to raise the present ground level by 1-1.5m. At this stage the source of the fill material has not been decided. The objective of this work was to compile factual information and provide interpretation on ground conditions (soils, bedrock and groundwater) likely to influence planning, design and construction within the Dunheved Precincts area. The work was also intended to satisfy the requirements of SREP 30 and the St Marys EPS 2000 with respect to soils, groundwater and land salinity issues, which include: There should be no significant rise in the water table or in groundwater salinity as a result of this development; Soil constraints (such as erodibility and shrink-swell potential) should be identified and the proposed development should have minimal impact on soils and salinity; Boundaries and development capacities of the soil types to be identified; An electromagnetic induction (EM) survey of the site should be carried out; and A Groundwater Management Strategy is to be prepared for the site.
SINCLAIR KNIGHT MERZ I:\INFR\Projects\IN08708\Deliverables\Reports\Final 28 April\Final Report v06.doc PAGE 72

In addition, a preliminary-stage foundation investigation based on widely spaced drillholes and limited geotechnical testing was carried out. Comments on foundation conditions and earthworks are included with the report. 7.2.1 Work Program The soils and groundwater investigation program included: A review of results from previous geotechnical and groundwater investigations relevant to the Dunheved Precincts, including those for the previous St Marys Western and Eastern Precincts; Walkover surveys, airphoto interpretation and terrain evaluation to identify groundwater and soil features relevant to foundation conditions, infiltration and discharge areas; Excavation, logging, sampling and in situ testing of 20 backhoe-dug test pits; Drilling five boreholes to depths of 6-11m, to investigate soils and shale bedrock below backhoe reach; Completion of 2 boreholes as observation wells (piezometers) for measuring groundwater levels, as well as for permeability testing and water sampling if required; Submitting bulk soil samples to two NATA-registered laboratories for determination of engineering index properties and erodibility; Recording groundwater intersections and soil moisture state in test pits and boreholes, and performing on-site conductivity (salinity) testing on soil samples; Collating DLWC borehole records, water quality data and previous on-site piezometer and water quality measurements; and Performing an Electromagnetic (EM) conductivity survey over the precincts, and correlating the results with test pit, borehole and laboratory test data (this is the subject of a separate report). 7.3 SITE DESCRIPTION

7.3.1 Topography The Dunheved Precincts occupy about 30ha of alluvial terrace lying between South Creek and Ropes Creek, north of Golf Links Road and the existing Dunheved (St Marys) industrial area. This land surface is planar, rising generally southwards between RL 19m AHD and RL 22m. Two minor gullies, which are tributaries of South Creek, drain towards the northwest and have cut down 2-4m below the prevailing terrace level. At the time of the investigation, which followed several months of generally dry weather, the more westerly gully consisted of a train of shallow pools and swampy areas. The other channel was running full with treated effluent from the St Marys STP, which adjoins the eastern corner of the Dunheved Precincts. The surface of the alluvial terrace is nearly level to undulating, with a number of very shallow wet depressions (relief 0.2-0.4m), resembling gilgais. They differ from gilgais in that the soil is not
SINCLAIR KNIGHT MERZ I:\INFR\Projects\IN08708\Deliverables\Reports\Final 28 April\Final Report v06.doc PAGE 73

noticeably expansive despite the drought, shrinkage cracks are relatively uncommon and generally less than 10mm wide, with no significant ground heaving. It was evident that many of these gilgai-like wet patches were much diminished in area as a result of the drought and some have been reduced to bare earth. Test pit D20, dug on one such bare scald, was dry to its full depth at 2.5m. 7.3.2 Bedrock The bedrock in the vicinity of the Dunheved Precincts is indicated as Bringelly Shale on the Penrith 1:100,000 geological sheet (Jones and Clark, 1991), although within the project area it is shown as being overlain by Pleistocene and Tertiary alluvial sediments. No surface outcrops of the shale bedrock were noted during the walkover surveys and the depth of clay cover in test pits and boreholes is everywhere greater than 3m. The Penrith geological map also shows a major geological structure, known as the Narellan Lineament, running in a north-south direction beneath the property. This lineament could be a zone of either closely-spaced jointing or faulting, which defines the straight course of South Creek upstream from the Dunheved industrial area. Within the project area it may be responsible for the deep shale weathering noted in several boreholes and test pits. The Bringelly Shale formation has a maximum thickness of about 300m, although at Dunheved this is believed to be only about 90m, when combined with the underlying Ashfield Shale. Both of these shales in turn overlie the Hawkesbury Sandstone. The Bringelly Shale is composed of shale, mudstone, claystone and some sandstone, although in boreholes drilled for the present investigation the predominant rock type was weathered shale. 7.3.3 Soils Two soil units are shown on the Penrith 1:100,000 soil landscapes sheet (Bannerman and Hazelton, 1990) within the Dunheved project area: Berkshire Park and South Creek soil landscapes (SLs). The first dominates the southern third of the site, while the South Creek SL covers the remainder. Both are clay soil units of alluvial origin, derived ultimately from weathering, erosion and fluvial transport of the Bringelly Shale bedrock. They differ in that the Berkshire Park SL is developed on older (Tertiary age) higher level terraces, while the South Creek SL comprises those clay soils on the near-recent (Pleistocene) and present-day, active flood plain of watercourses such as South Creek and Ropes Creek. Although these soils have many similarities, they differ in that the South Creek SL tends to have a shallower depth to the water table and hence to be more prone to waterlogging, more erodible and subject to more frequent flooding. The Berkshire Park soils are characterised by some degree of mottling, the presence of varying proportion of ironstone gravel (pisoliths and nodules), and are generally drier and stiffer than the soils of the lower terraces. Field observations in test pits suggest
SINCLAIR KNIGHT MERZ I:\INFR\Projects\IN08708\Deliverables\Reports\Final 28 April\Final Report v06.doc PAGE 74

that the Berkshire Park soils in fact cover almost all of the Dunheved Precincts except for the immediate vicinity of South and Ropes Creeks, below about RL 18m, rather than simply the southern third as shown on the Penrith SL sheet. 7.3.4 Groundwater Two regional scale groundwater systems, neither of them very permeable, operate within the former ADI land at St Marys and both appear to be present below the Dunheved Precincts. The upper one comprises groundwater contained within the alluvial clay and in the shale weathering profile itself, to a typical depth of 3-10m. The lower and much more extensive groundwater system occurs in shale bedrock below the base of weathering. Neither would normally be regarded as a true aquifer, meaning a soil or rock layer able to transmit groundwater in sufficient quantity and of adequate quality to support producing wells. However they are aquifers in the sense that they can discharge small volumes of saline water, generally along stream lines or low in the landscape, and thereby create a nuisance which may only become apparent at 5-10 year intervals. The general characteristics of the shale bedrock aquifer include: Low, but variable permeability, limited storage and very low well yield typically less than 1 L/s or 0.1 ML/day. The water-bearing fractures are non-persistent, widely spaced and poorly interconnected. This aquifer can be visualised as a complex of stacked perched water tables rather than a single saturated zone, and a continuous water table may not exist. Boreholes and piezometers may appear to be dry when first drilled, yet gradually fill with water over several weeks. Piezometer recovery time following bailing is very slow and standing water levels (SWLs) may fluctuate by a number of metres between wet and drought years. SWLs in piezometers 100-200m apart may differ by 1-3m, and seasonal variations of 3-4m in a single piezometer are common. Nearby wells may differ greatly in salinity, say from 3000 to 10,000mg/L total dissolved salts (TDS), and a single well may show the same changes from wet to dry season.

The shallow or soil aquifer system is made up of residual soil, floodplain alluvium and weathered saprolite all clays derived from the shale bedrock. It includes, in places, lateritic ironstone gravel developed within the alluvial clay or on the shale parent rock. The soil aquifer ranges in thickness here from about 3m to 6m and is composed of very stiff to hard silty clays, with some thin and discontinuous clayey sand layers and shaly clay towards the base. Soil and shale permeability testing was not included within the present investigation program, but indicative results from previous work in the Western Precinct are:
SINCLAIR KNIGHT MERZ I:\INFR\Projects\IN08708\Deliverables\Reports\Final 28 April\Final Report v06.doc PAGE 75

Upper (soil) aquifer Unweathered shale

10-6 to 10-10 m/s 10-7 to 10-9 m/s

The low end value for the soil aquifer reflects the intrinsic impermeability of the clay substance, but results of 10-6 to 10-7 m/s may indicate the presence of fissures or ferricrete bands. The latter are commonly encountered along the soil/rock interface, indicating a permeability contrast here and former discharge from cracks in the rock. However some permeability test results from the soil aquifer can be misleadingly high because the drillhole, and hence the piezometer response zone, extend a short distance into more pervious shale bedrock. 7.4 INVESTIGATION METHODS AND RESULTS

7.4.1 Test Pits The initial subsurface component of the fieldwork was carried out on 11th and 15th June 2004. This comprised the excavation, logging and backfilling of 20 test pits to depths between 0.9m and 3.5m, using a JCB tractor-mounted wheeled backhoe equipped with a 0.3m wide digging bucket. These pits were located so as to cover the whole of the study area, and all of its typical landforms except for the gullies, and to intersect as many soil materials as possible. Most pits were taken down to depths of 2.5m to 3.5m and were terminated when digging progress became very slow or when the backhoe was approaching its limit of reach, rather than at bucket refusal. Only two pits (D4 and D20) encountered shale bedrock and even at these sites there was some uncertainty due to the effects of weathering. Groundwater was not struck in any pit and the general consistency of the clay penetrated was very stiff to hard, both consequences of the present drought. All pits were backfilled and compacted by the tractor wheels immediately after logging and sampling. It was noted that the backfill did not mound above the surrounding ground surface on completion, an indication that the in situ soil mass is not especially compact. Pits were located by taping from objects visible on the 1:2,000 base map and in some cases by GPS position-finding. Surface elevations were interpolated from 0.2m contours on the site survey plan. Pit locations are shown on Figure 7-1. 7.4.1.1 Test Pit Results Test pit logs indicate that the predominant soil present down to a depth of about 3m is a dirty yellow to brown clayey and fine sandy silt, which grades to a silty clay in places and, rarely, to a clayey sand. A dry, grey brown silt topsoil is present in most pits and is even more conspicuous in gully walls and erosion scars, but otherwise there is little layering visible. At the time of the survey the clay and silt subsoil was only slightly moist and of very stiff to hard consistency. In places, most notably in test pit D5, pisolithic ferricrete (ironstone pea gravel) occurs as discontinuous
SINCLAIR KNIGHT MERZ I:\INFR\Projects\IN08708\Deliverables\Reports\Final 28 April\Final Report v06.doc PAGE 76

horizons within the alluvial clay, especially close to the top of the underlying weathered shale bedrock. Three test pits (D8, 11,17) encountered angular cobbles of grey high strength crushed volcanic breccia, similar to railbed ballast. This is assumed to be imported rubble, which was laid along access tracks (some of which are still visible on old airphotos) to improve wet season trafficability when the site was operated as an ammunition plant. Other portions of the ground surface, in the vicinity of D11 and D12, appear to have been deeply ripped by bulldozer tynes. This furrow and

SINCLAIR KNIGHT MERZ I:\INFR\Projects\IN08708\Deliverables\Reports\Final 28 April\Final Report v06.doc PAGE 77

ridge soil surface topography, now hardened by desiccation, is presumed to have resulted from contaminant clean-up operations some years ago. 7.4.2 Boreholes Five boreholes (BHs 1-5) were drilled to depths between 6.0m and 11.2m on 12-13 July 2004 by means of a truck-mounted Edson 3000 rig. These holes were located so as to cover as much of the site as possible within the limited time available, and were supervised and logged by a qualified geotechnical engineer. Most boreholes were sited beside or close to test pits, which had been previously excavated and backfilled, in order to provide the maximum amount of information on the shallow soil materials. Drillholes were advanced through soil materials using 125mm diameter continuous flight augers equipped with V-bits or tungsten carbide (TC) bits. The auger string was withdrawn at intervals for soil logging and to carry out standard penetration tests (SPTs). Auger drilling was terminated when the rate of advance became very slow in weathered shale, at depths of 6-10m. In some cases this slow drilling approached refusal, but definite V-bit or TC bit refusal on strong rock did not occur. Two holes (BHs 1 and 3) were deepened, to 11.2m and 10.5m respectively, by rotary coring to prove bedrock and obtain continuous samples of the shale. Boreholes BH1 and BH2 were completed with PVC casing and screened intervals as groundwater observation wells (piezometers). Four standard penetration tests (SPTs) were performed in BHs 3 and 4, to provide an indication of strength within the alluvial clays; other boreholes were not tested, in order to make best use of the remaining rig time. These tests were carried out using a standard open-drive sampling shoe, which also provided soil samples for logging. Test N-values were in the range 10-41, indicating stiff to hard clay, which is consistent with observations of very stiff to hard silty clay and clayey silt made during test pit logging. All boreholes were backfilled immediately after drilling and logging, except for the piezometers holes BH 1 and 3. Drilling sites were located relative to features visible on the 1:2,000 site plan and approximate collar levels were interpolated from 0.2m contours on these maps. Borehole locations are shown on Figure 7-1 and their logs are appended. 7.4.2.1 Borehole Results The augered sections of the boreholes indicated that alluvial silty clays and clayey silts, of stiff to hard strength and low to medium plasticity, extend to depths of 5-8m, though 5-6m is more common. This is consistent with results from the test pits, which revealed that the depth of the alluvial clay is generally deeper than about 3m, the limit of backhoe bucket reach.
SINCLAIR KNIGHT MERZ I:\INFR\Projects\IN08708\Deliverables\Reports\Final 28 April\Final Report v06.doc PAGE 79

The alluvial clay appears to be underlain by 1-2m of extremely weathered shale, described as shaly clay on the auger logs because it is thoroughly ground up by the auger bit. In the cored sections of the boreholes most of the core losses are likely to have been in layers of extremely weathered (XW) shale. This XW shale is presumed to be similar in engineering properties to a very stiff to hard fissured clay, though it might equally be described as a very low strength rock. The cored bedrock sections in BHs 1 and 3 are predominantly dark grey slightly weathered Bringelly Shale, but with core losses where more weathered shale was intersected. The strength of this shale has been estimated as generally very low to low (UCS 0.6-6Mpa), but with some harder hands. 7.4.3 Groundwater and Salinity Results As noted previously, no groundwater inflows were struck in any of the 20 test pits, even in those deliberately excavated close to wet surface depressions or in the floor of clay pans (D4 and D20 respectively). The soils encountered in these pits, down to a maximum depth of 3.5m in D4, were clays and silts of low permeability with no significant pervious sand or gravel layers. In almost every pit the soil moisture content appeared to be below the plastic limit as a result of the drought. The soil surface likewise appeared to be drier than in a normal season, with wheel ruts hardened and bare soil scalds where shallow reed ponds would normally be present. Groundwater was cut in only one borehole, at depth 6.5m in BH4. This occurred along the interface between the alluvial clay and the shale bedrock, as is common in western Sydney shale areas a consequence of either gravel at the base of the alluvial clays, or of fractured shale below a clay blanket. This water did not rise during the time that the borehole remained open (less than one hour) and the hole was subsequently backfilled. No water was detected in piezometer borehole BH2 (shallow alluvial clay aquifer) up to the time of its completion. Piezometer borehole BH1 was drilled using water as a fluid and hence the water level on completion was close to the top of the hole, and not representative of the true depth to the water table. It is normal for borehole standing water levels to take days or even weeks to stabilise in shale and clay aquifers of western Sydney because of their low permeability (water has to drain from the shale to fill the borehole to its SWL). Soil salinity tests were carried out in the field during test pit sampling on 1:5 soil in water suspensions, using a Horiba conductivity meter. Samples were also taken for laboratory tests, carried out in the DIPNR soils laboratory at Scone NSW. Results from both sets of testing are summarised on Table 10.

SINCLAIR KNIGHT MERZ I:\INFR\Projects\IN08708\Deliverables\Reports\Final 28 April\Final Report v06.doc PAGE 80

Table 10: Summary of soil salinity testing results

Test pit No

Depth (m)

PH

Conductivity (mS/cm)

Temp. (C0) 19.0 16.9 -15.7 15.0 15.8 --19.0 16.9 ---20.0 18.2 ----

TDS (mg/L) (%)

Field/Lab

D2 D2 D2 D13 D13 D13 D14 D14 D15 D15 D15 D16 D16 D18 D18 D18 D20 D20

020 0.50 0.75 0.25 0.35 0.55 0.15 0.70 0.20 0.50 0.75 0.15 0.45 0.30 0.60 0.90 0.15 0.45

10.06 8.70 9.53 10.68 11.06 11.50 5.8 5.9 10.06 8.70 9.53 5.8 5.4 11.6 9.41 10.63 6.0 5.4

0.19 0.11 0.15 5.61 0.13 0.16 0.17 0.37 0.19 0.11 0.15 0.28 0.38 0.14 0.20 0.23 0.07 0.37

122 70 96 3742 88 107 109 237 127 74 100 179 243 93 133 153 45 237

F F F F F F L L F F F L L F F F L L

SINCLAIR KNIGHT MERZ I:\INFR\Projects\IN08708\Deliverables\Reports\Final 28 April\Final Report v06.doc PAGE 81

7.4.4 Geotechnical Soil Testing Four bulk samples of about 20kg were taken from the upper parts of test pits D13, 15, 17 and 19. These specimens were considered to be typical of the range of alluvial clays present in the upper 2m of the existing soil profile at the Dunheved Precincts. However it is understood that they will probably be covered by 1-1.5m of fill from a source yet to be determined. The four samples were submitted for testing in a NATA-certified laboratory for the following procedures: Maximum dry density and optimum moisture content under standard compaction; California bearing ratio (CBR) on compacted and four-day soaked specimens; Atterberg consistency limits (liquid and plastic limits, plasticity index) plus linear shrinkage; Particle size distribution for the >75 micron fraction.

Results from this suite of tests are summarised on Table 11 below. These results indicate that the alluvial clays are generally of medium plasticity and low shrinkage potential, probably due to a relatively high silt content. Sample D19 is anomalous (PI of only 7) because of its high sand content and a small proportion of gravel. Field moisture contents are below the plastic limit, though the difference is smaller than what was estimated during test pit logging.
Table 11: Summary of soil geotechnical testing results

Test pit Number

Depth

Atterberg Limits LL PL 18 16 18 16 PI 23 29 25 7

MDD (t/m3) 1.72 1.68 1.66 1.96

OMC (%) 18.5 20.0 20.5 13.0

FMC (%) 14.2 19.0 16.7 13.5

CBR (%) 4.0 4.0 3.5 11.0

LS (%) 12.5 14.0 13.0 0

D13 D15 D17 D19

0-1.4m 0-1.5m 0.6-2.0m 0-1.8m

41 45 43 24

LL = liquid limit PL = plastic limit

SINCLAIR KNIGHT MERZ I:\INFR\Projects\IN08708\Deliverables\Reports\Final 28 April\Final Report v06.doc PAGE 82

PI = plasticity index MDD = maximum dry density (standard compaction) OMC = optimum moisture content CBR = California bearing ratio 7.5 ELECTROMAGNETIC SOUNDING An electromagnetic (EM) sounding survey was carried out across the site by Douglas Partners in June 2004, with the primary aim of mapping variations in subsurface salinity, since this was assumed to be the main contributor to ground conductivity. The full results of this work are given in their report (Douglas Partners, 2004), which is summarised below. The survey was carried out by means of an EM31 conductivity meter mounted on a 4WD quad bike. The nominal 100m by 200m grid was distorted due to access limitations and obstacles, and the eventual traverse lines totalled 12.5km, with readings at approximately 1m intervals. Location control was provided by a differential GPS system mounted on the quad bike and linked to the EM31. The results indicate low apparent conductivities (ECe <125mS/m or 0.13mS/cm) adjacent to gullies and in areas of shallow depressions on the alluvial terrace surface, and higher conductivities (125-200mS/m) beneath more elevated ground. Overall, the EM results indicate that the subsurface is non-saline to slightly saline. However they also showed greater variability than the soil salinity measurements listed in Table 10, which were uniformly low. The reason for this discrepancy is believed to be that the soil testing was limited to samples taken in test pits at depths less than 1m, whereas the EM31 measures bulk conductivity to a depth of several metres, possibly down to 3-6m in this case. The EM31 results are believed to be a response to a number of factors affecting the overall ground conductivity: Variations in the clay mineral content and the depth of alluvial clay (and hence depth to shale bedrock); Variations in moisture content and degree of saturation within the clay blanket, and in the salinity of this pore water; and The presence or not of conductive lateritic ironstone in the subsurface.

However the possibility of higher salinity at depths greater than 3m, probably due to saline groundwater below the water table, cannot be excluded.
SINCLAIR KNIGHT MERZ I:\INFR\Projects\IN08708\Deliverables\Reports\Final 28 April\Final Report v06.doc PAGE 83

7.6

GROUNDWATER AND SALINITY IMPLICATIONS

7.6.1 Existing Groundwater Conditions The hydrogeology of the St Marys property, including the Dunheved Precincts, is summarised in Mackie Martin (1991) and ADI Ltd (1996). The results of boreholes drilled between 1990 and 1996 in or close to the Dunheved Precincts suggest that both the unconfined shallow (soil) aquifer and the confined deep (shale bedrock) aquifer are present. Both aquifers have similar characteristics to those in other parts of the St Marys property in that they are tight, with low to very low permeability and very limited storage capacity. Both probably consist of a series of stacked and sporadically distributed perched water tables in effect, poorly interconnected lenses of saturated ground - rather than a single homogeneous water-bearing layer. The vertical connection between the soil and shale aquifers is poor, to judge by nearly dry soils observed in test pits, and they appear to have different recharge / discharge relations. Recharge to the soil aquifer is by direct infiltration onto the surface of the alluvial terrace (around RL 19-20m), followed by throughflow across the A/B soil profile interface and temporary storage in shallow perched aquifers at depth 0.5-1m. Discharge is by evaporation from puddles in shallow gilgai-like surface depressions, through transpiration by trees and by seepage to shallow pools in the unnamed western gully (at about RL 16m). Limited information in the Mackie Martin (1991) report indicates that the shallow groundwater is of low salinity, less than 500mg/L TDS, although both the surface puddles and the gully pools support halophyte vegetation including salt-tolerant reeds. No saline scalds were observed. At present most infiltration to the shale aquifer is likely to be coming from the unlined effluent discharge channel in the eastern gully, at about RL 15m. This is believed to have raised the water table by perhaps 1-2m and reduced the salinity to about 3500mg/L (BH1), and to be moving slowly through the shale aquifer. It is presumed to ultimately discharge along South Creek at about RL 12m. 7.6.2 Existing Salinity Information on salinity at Dunheved has been drawn from four sources: On-site conductivity testing carried out on 1:5 soil/water suspensions using a Horiba water testing device (results are listed on Table 11 and designated F for field); Similar testing carried out independently by DIPNR under laboratory conditions on soil samples submitted by SKM (results also on Table 10, designated L); Groundwater salinity results from previous boreholes SM2-4 and SM14,15 (Mackie Martin, 1991); and

SINCLAIR KNIGHT MERZ I:\INFR\Projects\IN08708\Deliverables\Reports\Final 28 April\Final Report v06.doc PAGE 84

Electro-magnetic induction (EM) surveys across the precinct area to measure ground conductivity, carried out by Douglas Partners in 2004 and reported separately. Soil Salinity The soil conductivity results on Table 10 are consistently low, all less than 300mg/L TDS with one outlier of 3742mg/L. Values of less than 300mg/L in the top 1m of the soil profile are unusually low for western Sydney, since salt is normally stored within the B-horizon and moved around in throughflow along the A/B horizon interface. Groundwater Salinity The borehole results indicate a pattern typical of western Sydney, with the shallow unconfined aquifer being almost non-saline (<500mg/L) and the deeper shale aquifer water being much saltier. In the latter case results from BH1, SM14 and SM15 range from 3500mg/L to 13,100mg/L. Nevertheless, these are relatively low values for the shale aquifer, whose salinity is typically in the range 10,000-20,000mg/L. This may indicate mixing of the rising shale water with infiltrating rainwater, or it may reflect the influx of low salinity effluent from the nearby St Marys STP outfall channel. EM Salinity The EM survey results present a plot of relative ground conductivity averaged out over a depth of about 5-6m. The EM thus sees to greater depth than the soil tests, which are limited to about 1m below the surface, but is influenced by several factors: Salt stored within the soil B-horizon and in saline groundwater below the water table; Differences in clay content, and in moisture content between saturated and partly-saturated clays; and Differences in depth to the shale bedrock (and hence differences in the thickness of the overlying clay blanket); and The presence or otherwise of lateritic ironstone gravel in the subsurface.

The B-horizon salinity at Dunheved appears to be generally less than 300mg/L, which is lower than elsewhere in the St Marys site. The salinity of the water in the shale aquifer, as noted above, is considerably higher, though still relatively low by the standards of the St Marys property. Impact of Development The amount of salt likely to be added by the proposed development is negligible salinity problems arise when the existing stored salt is brought to the surface by a rising water table, or is washed laterally from the B-horizon by increased infiltration. We consider that though the EM results show variations in the overall ground conductivity, the soil and groundwater test results indicate relatively low salinity overall.
SINCLAIR KNIGHT MERZ I:\INFR\Projects\IN08708\Deliverables\Reports\Final 28 April\Final Report v06.doc PAGE 85

7.6.3 Groundwater Management Management of groundwater, and hence of salinity, to meet the requirements of SREP 30 and the St Marys EPS 2000 implies that the water table will not rise significantly as a result of the proposed development. There should also be no increase in throughflow (lateral movement of water through the soil profile, but above the water table). In practice this means that infiltration to the soil profile and from there to the water table should be reduced by all practical means. The proposed filled landform within the Dunheved Precincts, and the planned employment landuse present opportunities for achieving these goals. The specific measures proposed for groundwater and salinity management at Dunheved are as follows: Shaping the filled landform as a cambered embankment to shed water rapidly and directing this runoff into graded natural watercourses, while avoiding detention in natural and artificial ponds so far as possible. Constructing the base of the embankment of free-draining rock fill and providing subsoil drains (to South Creek or Ropes Creek) where necessary, to prevent water accumulating on the fill / former land surface interface. Making maximum use of paving, especially of car parks and storage areas, to reduce the ground available for rainwater infiltration. It is assumed that much of the Precincts will be built over in any case, for factories and warehouses. Collection of stormwater from paved areas and roofs and directing it through sealed drains to approved discharge points along natural drainage lines. All basins and swales will be lined with an impermeable liner to prevent infiltration into groundwater. Grassing, mulching and tree planting in unpaved areas, with preference given to native species with high water demand (but making allowance for the relatively dry St Marys climate). Preference should also be given to deep-rooted trees and shrubs over shallow rooted grasses. 7.6.4 Groundwater Monitoring In order to judge the success or otherwise of the infiltration reduction strategy outlined above, it will be necessary to monitor fluctuations in groundwater level and changes in water quality over many years. It is recommended that three piezometer pairs (one shallow, one deep) be constructed in an east-west line across the Dunheved Precincts, with one pair in the centre and one each at the eastern gully (recharge area) and one at near South Creek, at the presumed discharge zone. 7.7 SOILS IMPLICATIONS Residual soils derived from weathered shale bedrock in western Sydney are typically of moderate to high reactivity (shrink-swell potential in response to drying and wetting cycles) and moderate
SINCLAIR KNIGHT MERZ I:\INFR\Projects\IN08708\Deliverables\Reports\Final 28 April\Final Report v06.doc PAGE 86

dispersivity (the tendency of sodic soils to erode rapidly when in contact with fresh water). These characteristics are especially well developed where: There is a sharp texture contrast between a silty, low plasticity A-horizon and a high plasticity, sodic and saline B-horizon; Where the soil profile, and especially the B-horizon is relatively thick, say 1-2m; and On low gradient slopes and in low-lying ground, with grass rather than tree cover, where seasonal moisture changes within the soil profile are likely to be greatest.

Test results summarised on Table 11 indicate that the alluvial clays within the Dunheved Precincts area are highly silty and of medium plasticity, with linear shrinkage bar test results in the low to medium range. The salinity results indicate that these clays are of low salinity, at least in the top 1m. The test pit logs demonstrate that the soil profiles, though deep (several metres), are poorly differentiated in terms of horizon development. These results suggest only moderate shrink-swell potential, by the standards of western Sydney clay soils. Surface observations of widely spaced but narrow aperture shrinkage cracks under the present drought conditions confirmed that these clays are of only moderate reactivity, despite the presence of shallow surface depressions resembling gilgais. In other parts of Australia gilgais are associated with the presence of high plasticity, highly reactive clay soils. The relative absence of rill and gully erosion across the site, coupled with the low salinity of the soil B-horizon, suggest that these clays are of low dispersivity and hence comparitively nonerodible. Filling of land within the project area, as proposed, will further reduce the impact of urban development on these soils. As well as protecting the natural soil profile from erosion by running water, the effect of a fill blanket will be to maintain a relatively constant moisture content within the buried clay subgrade, thereby minimising the potential for both swelling and drying shrinkage. 7.8 FOUNDATION IMPLICATIONS

7.8.1 Foundation Conditions It is expected that while some lightweight structures may be founded on well-compacted engineered fill or stiff to very stiff alluvial terrace clay (of presumed bearing capacity 200kpa), most Dunheved buildings will be set upon bored piles augered to shale bedrock at depths below about 6m from the present surface. Based on limited core drilling in BHs 1 and 3, slightly weathered shale of very low to low strength (UCS 0.6-6Mpa) persists to the maximum depth drilled (11.2m and 10.2m respectively).

SINCLAIR KNIGHT MERZ I:\INFR\Projects\IN08708\Deliverables\Reports\Final 28 April\Final Report v06.doc PAGE 87

This weathered bedrock appears to be equivalent to Class IV shale in the classification of Pells et al (1998), with an allowable bearing pressure of 1Mpa, based on a serviceability settlement criterion of <1% of minimum footing dimension. Deeper drilling would have eventually encountered stronger shale, Class III or better, but at an unknown depth. The persistence of relatively weak bedrock with depth was also noted in the other uncored boreholes at Dunheved. No definite refusal was encountered by auger V- and TC- bits, even after 1.5m to 4.5m penetration into the shale, and significant core losses indicate some very weak layers within the upper few metres of the shale. For site planning and conceptual design purposes only, a pile founding depth of 12m in 1Mpa bearing capacity Class IV shale is suggested. The depth of 12m is based on a presumed maximum depth of clay of 8m, though 5-6m thickness is expected to be more common, and a socket depth of 4m in the weathered low strength shale. However these generalised assumptions should be checked by drilling at the locations of specific building structures prior to their design. The actual bedrock encountered at the building sites may be stronger or weaker than that implied by these wide-spaced preliminary boreholes. 7.8.2 Earthworks We understand that the ground surface over most of the Dunheved Precincts is to be raised by 11.5m to place the industrial estate above the 100 year flood level. The following remarks therefore reflect good earthworks practice rather than recommendations based on a particular design. It is presumed that the filled ground landform will also be designed so as to avoid any risk of future salinisation or waterlogging. Prior to commencing embankment construction the existing land surface should be stripped of vegetation and graded, where necessary, to assist drainage from surface depressions. No significant thickness of topsoil requiring removal was noted during test pit excavation. Any soft patches encountered should be proof-rolled and if necessary dug out and backfilled with granular fill. Test pit logs and the limited soil test data indicate that the subgrade material is generally stiff to very stiff alluvial silty clay, with thin (less than 1m) stony fill in places. The clay is of aboveaverage bearing capacity for shallow clay in western Sydney, with CBRs of 3.5-4 (and one CBR 11 result). No signs of organic or other deleterious matter, such as peat or organic clay, were detected in the top 2-3m of this clay in the 20 test pits distributed across the Precincts. The source of fill material for this site has yet to be nominated. Crushed sandstone tunnel spoil is very suitable for this purpose because of its uniformity, high compacted density, stiffness and resistance to infiltration. It is suggested that fines-deficient rockfill, such as ripped sandstone, be placed in the bottom 0.5m of the embankment over low-lying portions of the natural surface, to facilitate drainage and provide a capillary break against rising groundwater.

SINCLAIR KNIGHT MERZ I:\INFR\Projects\IN08708\Deliverables\Reports\Final 28 April\Final Report v06.doc PAGE 88

The bulk of the fill should be placed in layers 200-300mm thick and compacted systematically to at least 95% standard density ratio. The top 0.5m should be compacted to 98% of standard and its upper surface should be cambered to shed runoff and minimise infiltration. 7.9 CONCLUSION Borehole, geophysical and test pit investigations in the North and South Dunheved Precincts indicate that shallow groundwater occurs at depths of 3-6m and is of low salinity. Deeper water in the shale bedrock is moderately saline, in the range 3500-13,100mg/L, which is low by the standards of the St Marys property. It is concluded that the planned development is unlikely to result in surface salinisation and that the remedial measures proposed in the report raising the ground level by filling and limiting infiltration will further reduce this possibility. 7.10 REFERENCES ADI Ltd (1996) Southern Sector West validation report [at ADI site, St Marys NSW]. Unpublished report dated June 1996 to Australian Defence Industries Limited. Anon (2000) Draft 1:100,000 Penrith salinity hazard map [and accompanying unpublished notes, 19p]. NSW Department of Land and Water Conservation (DLWC) Penrith, December 2000. Bannerman and Hazelton, PA (1990) Soil landscapes of the Penrith 1:100,000 sheet. NSW Soil Conservation Service, Sydney, 122p. Department of Natural Resources Qld (1997) Salinity management handbook. DNR, Coorparoo Qld, 214p. Dias, A and Thomas, D (1997) Salinity in the South Creek catchment. DLWC Goulburn, 29p. Douglas Partners (2004) Report on salinity assessment, proposed industrial development Dunheved. Unpublished report No 37217 to Sinclair Knight Merz, dated August 2004. Hayward, S and Mitchell, P (2000) Salinity in the South Creek catchment - the prequel. Proceedings, 1st South Creek Conference (editor PA Hackney), UWS June 2000. Herbert, C (1979) The geology and resource potential of the Wianamatta Group. NSW Geological Survey Bulletin 25, 203p. Jeffery and Katausakas (2000) Soil and groundwater investigation for proposed residential redevelopment at Comland St Marys [Western Precinct]. Unpublished report to Lend Lease Development by EIS division of Jeffery and Katauskas P/L (2 volumes). Jones DC and Clark, NR (editors) (1991) Geology of the Penrith 1:100,000 sheet. NSW Geological Survey, Sydney, 202p.
SINCLAIR KNIGHT MERZ I:\INFR\Projects\IN08708\Deliverables\Reports\Final 28 April\Final Report v06.doc PAGE 89

Mackie Martin and Associates (1991) Groundwater, surface water and sediment sampling, St Marys facility. Unpublished report to ADI Ltd, December 1991 (2 volumes). Old, AN (1942) The Wianamatta Shale waters of the Sydney district. NSW Agricultural Gazette, May 1942, p215-221. Pells PJN, Mostyn G and Walker BF (1998) Foundations on shale and sandstone in the Sydney region. Australian Geomechanics Journal, Dec 1998, p17-29. Smith, V (1979) The Cainozoic geology and construction materials resources of the PenrithWindsor area, NSW. NSW Geological Survey unpublished report GS1979/074, 170p. WSROC (2003) Salinity code of practice. Prepared by R Nicolson for the Western Sydney Regional Organisation of Councils (WSROC) Ltd, Blacktown, March 2003, 58p.

SINCLAIR KNIGHT MERZ I:\INFR\Projects\IN08708\Deliverables\Reports\Final 28 April\Final Report v06.doc PAGE 90