Environmental Monitoring at King's Lynn - Rob Smith and Malcolm Lillie

Chapter 5. Environmental Monitoring by Malcolm Lillie and Robert J. Smith Introduction The impacts of piling activities on archaeological deposits from urban centres have been highlighted in recent years (Dalwood et al. 1994; Davis 2003); and the extent of these impacts has also been demonstrated in previous research (e.g. Nixon 1996; Williams et al. 2006). However, the majority of this work has been associated with the impact of individual piles on the archaeological resource, with their cumulative impact rarely being considered (Butcher and Williams 2008). A number of previous studies have indicated the importance of monitoring of the burial environment prior to, during and after pile insertion, as part of development planning (Davis 1998; Environment Agency 2006a and 2006b; Nixon 1996; Williams and Corfield 2002). It has been noted that there are a number of physical, hydrological and geochemical changes which can affect the status of the burial environment during development; and these processes can impact upon any potential archaeological remains contained within sediments (Williams and Corfield 2002). As a consequence of the potential damage that piling may cause to the archaeological resource, the installation of monitoring points in order to collect hydrological and geochemical data (redox potential and pH values) before, during and after development, can help in assessing whether any changes to the environmental equilibrium of the site can be identified. The significance of any change can then be determined in relation to the potential impact upon the archaeological resource (Davis 1998). Before the piling programme is initiated, there are a number of pre-construction impacts that may have the potential to damage the archaeological deposits. These impacts will primarily result from physical disturbance and hydrological and geochemical changes to the burial environment (Williams and Corfield 2002). For example, physical disturbance of the soil matrix (topsoil stripping, ground improvement and stabilisation measures) can promote oxygen ingress into previously undisturbed layers, potentially increasing decay and corrosion; and de-watering of the site can lead to the deposits drying out, further increasing aerobic conditions and consequently promoting microbiological degradation (Williams and Corfield 2002). Although the impact of construction activities on the buried archaeological resource can, in certain instances, promote sustainability and be used as part of the in situ preservation process (Williams and Corfield 2002; Williams and Butcher 2006), there are several problems associated with piling. Piles are divided into two different types, displacement and non-displacement piles. Displacement piles are particularly damaging to archaeological deposits due to their disturbance of the ground that surrounds the pile (Williams 2006). Of further concern is the potential for the pile to degrade over time, and as a consequence, allow contaminants to penetrate into the burial environment. Non-displacement piles can also cause changes in the permeability around the pile/soil interface, which results The following report presents a synthesis and evaluation of data recorded from four monitoring locations associated with development work undertaken at the Vancouver Centre, King’s Lynn. As part of the mitigation process, environmental monitoring of the burial environment has been carried out at this site over a two year period. This includes the recording of data relating to water levels, redox potentials, pH, temperature and dissolved oxygen content (DOC) across a period that encompasses the pre-development stage of site works (6 months of monitoring), and the development and post-development stages. In this instance the development that is being assessed is the insertion of 923 driven piles into the sediments beneath the Vancouver Centre. The data generated across the monitoring programme, between 6 August 2004 and 6 September 2006, has been tabulated and is presented graphically below. Comparisons are made between the parameters, both in relation to the piling programme, and in relation to each other. Trends in the data are assessed, and conclusions are drawn based on the evidence for any reactions in the parameters studied in relation to the development process. Final conclusions have been made in light of the current study, and recommendations are offered for future monitoring in the King’s Lynn region. Previous studies Previous research has shown that the effects of de-watering on waterlogged archaeological remains are detrimental to their long-term survival (Lillie 2004; Lillie and Smith 2007; Welch and Thomas 1996). This observation is of particular significance when considered against the fact that the waterlogged component of the archaeological record has been shown to be of considerable importance to our understanding of past human societies. In effect, organic remains provide a more holistic picture of the range of materials used in the past; an aspect that is rarer on dryland sites (e.g. Caple 1996; Coles and Coles 1996; Van de Noort and Davies 1993). As numerous archaeological sites exist in widely varying environments, the identification of the best ways in which these sites can be preserved for the future is of fundamental importance to heritage managers. It has generally been acknowledged that water tables and the maintenance of waterlogged/saturated conditions are essential to the long-term survival of the archaeological and palaeoenvironmental resource (e.g. Chapman and Cheetham 2002; Corfield 1996). As a consequence, this report will review the monitoring approach which has been used in order to determine the effect that piling may have upon the water table and the potential impact that this may have upon the buried archaeological resource. 83 underlying strata. The pile size ranged from 240–300mm. A total of 923 piles were inserted over the site. Two boreholes (3 and 4) are located to the east and west of New Conduit Street, respectively, outside the northern boundary of Zone D (see Figure 5.1). The remaining boreholes (1 and 2) are situated within the Zone D boundary, north of Union Lane. The Majestic Cinema lies to the east of BH 2. The four boreholes were sunk in August 2004. Groundwater gauging and monitoring commenced on 6 August 2004 after borehole installation. Monitoring was undertaken every two weeks for a period of six months, then monthly for the next six months, and quarterly thereafter. Characterising any burial environment is extremely complex and requires the measurement of a number of biological, chemical and physical parameters (Caple 1996; 1994). It is essential to describe the relevant aspects of the burial environment in order to understand the conditions necessary for the preservation of archaeoorganic material (Caple 1996). There are a number of biological, chemical and physical measurements which can be used in order to characterise the burial environment; these include soil hydrology (which relates to water table dynamics), soil chemistry (which relates to the measurement of redox potentials and pH within the soil profile) and microbiological assessment of the soil profile (Caple 1994; Hobson 1988). The environmental parameters which were used in order to determine the nature of the burial environment in Zone D of the Vancouver Centre included temperature, pH, oxygen reduction (redox) potential and dissolved oxygen content (DOC). These parameters are standard techniques which have been used by previous researchers to assess the physico-chemical nature of sediments, and as a consequence, enable a greater appreciation of the conditions necessary for the preservation of archaeological material (Brunning et al. 2000; Caple 1996; 1993; Caple and Dungworth 1997; Lillie and Smith in press; Smith 2005; Smith and Lillie in press). The measurements obtained from Zone D during the two-year monitoring programme, using the environmental parameters highlighted above, were taken from groundwater samples located within the four boreholes (Figure 5.1). Although comparable data may be obtained between in situ soil and groundwater measurements in order to characterise the burial environment at the site, recent research has shown that groundwater monitoring can produce unreliable results (Matthiesen et al. 2004). Furthermore, if no water is present within the boreholes it is impossible to monitor the other parameters outlined above (i.e. temperature, pH, redox potential and DOC). Consequently, this method of sampling, as employed in this monitoring programme, may inhibit the potential for the generation of valuable environmental data. in a pathway for the transport of oxygen and contaminants (Westcott et al. 2001; Williams and Butcher 2006). Archaeological deposits can also be affected by contaminants being driven down during pile insertion. On development sites that possess a high potential for archaeological remains, or are waterlogged in nature, continuous flight augered piles are preferred. When the correct pile depth has been reached during auger insertion into the ground, it is withdrawn, taking up the soil during the movement, while at the same time concrete is pumped into the hole which is left behind. Background As part of the mitigation strategy for the redevelopment of the Vancouver Centre it was necessary to undertake a programme of environmental monitoring in order to assess the impact of piling upon the character of the underlying sediments. Particular emphasis was placed on the potential of pile insertion into the sediment to decrease the preserving qualities of the soil strata, subsequently causing the degradation of any potential organic component contained within the soil matrix. It is anticipated that the data generated during the monitoring programme will help to inform future planning applications within the area. Ground investigations CL Associates were instructed by Oxford Archaeology to undertake the drilling of four boreholes in order to permit the long-term monitoring of a number of environmental parameters over a two-year period. The boreholes are situated in Zone D, and their location is shown on Figure 5.1. No data exists on the precise nature of the sedimentological sequences identified during the emplacement of the piezometers used in the monitoring programme. King’s Lynn itself is superimposed upon deeply stratified sequences of reclaimed estuarine and marine sediments. Approximately 10m of deposits underlie the area; these comprise material relating to former tidal/intertidal environments such as silts, sands and clays, peats and channel sequences, alongside several freshwater channels — the Millfleet, Purfleet, Nar and Gay which debouch into the Ouse. The surface of the marine sequences occurs at between 1–2m below the modern ground surface. In general the sequences underlying the town are considered to represent natural accumulations of tidal/intertidal sediments, along with areas of saltern activity. The sequences identified during excavation appear to be natural in derivation. One of the key problems with the lack of sediment logs is the fact that the water table reactions identified may be influenced by the specific sediment characteristics in the immediate vicinity of the piezometer clusters. The boreholes were sunk to 4.5m depth at locations which would remain accessible after development for monitoring purposes. Dual piezometers were installed in all four boreholes, one with a response zone of 1.5–2.5m and the other with a response zone of 3.5–4.5m. The piles were driven into the sediment to a depth of between 14m and 24m, depending upon the nature of the Results Introduction This section addresses the results which have been obtained by CL Associates as part of the ground investigation of Zone D at the Vancouver Centre, King’s Lynn (CL Associates 2006). Each environmental monitoring parameter will be discussed separately in 84 Figure 5.1 Zones A–D borehole and pile locations 85 Figure 5.2 Water level depths obtained from the shallow boreholes over the duration of monitoring Figure 5.3 Water table depths obtained from the deep boreholes over the duration of monitoring visits identified dry conditions within this borehole, indicating that the water table was located below the lower reaches of the response zone. The deeper boreholes (Figure 5.3) exhibit water table heights of c. -1m to -4m below the modern ground level prior to pile insertion, and there is clear evidence for comparability between the upper and lower boreholes at each location. This indicates that the earlier monitoring data is demonstrating that the water table is being mirrored at both depths, and that synergy in response to hydraulic head is occurring. As will be outlined below, the data generated after pile insertion demonstrates that the hydrological environment in the vicinity of the piling works is severely disrupted after pile insertion takes place. It should be noted that the lack of sedimentological data for the boreholes, accompanied by a lack of detailed stratigraphical information across the area being studied, limits the discussion of water table activity to a generic overview of water table responses to pile insertion. Greater resolution of the data discussed below would be forthcoming had transect coring and sediment logging been undertaken in advance of development (Smit et al. 2006, 41–2), as sediment type and stratigraphic relationships will influence water movement through the soil profile. The observations below are made against the pre-pile insertion data for all aspects of the burial environment (water table activity, redox potential, dissolved oxygen content, pH and temperature). In general order to assess the impact of piling on the character of the underlying sediments (i.e. water levels, redox potential, dissolved oxygen content, pH values and temperature). Water levels Water level monitoring of the shallow and deep boreholes began on 06/08/04. Six monitoring visits were undertaken prior to the insertion of piles into Zone D. This monitoring continued fortnightly for a period of six months, then monthly for the next six months and quarterly thereafter for the following year (Figs 5.2 and 5.3). Water level monitoring ceased on 06/09/06. The water level depths obtained from the shallow boreholes (BH 1–3) prior to pile insertion (Figure 5.2) fluctuate between -1.2m and -2.1m below the ground surface. The exception to this trend is BH 4 which displays values that are consistently lower than the other three boreholes (-2.9m) and below the designated response zone for boreholes located at this depth (i.e. between -1.5m and -2.5m) (CL Associates 2006). Insufficient information exists regarding the viability of obtaining measurements from BH 4 below -2.5m, particularly in light of the fact that the deepest part of the response zone is -2.5m (which should be the bottom of the piezometer tube). In addition, the first three monitoring 86 Figure 5.4 Annual rainfall data for England and Wales, January 2003 to May 2007, obtained from the Met Office website recorded. Borehole 1 is clearly being influenced by shifts in the hydrology across the period February to October 2005. The shallow and deep piezometers are showing differing patterns of response, suggesting that the shallow piezometer is failing to reflect the dominant groundwater activity at this location, and is instead, recording the near surface disruptions caused by piling. The data for BH 2 (shallow and deep piezometers) is consistent through until May 2005, after which no data is recorded in the shallow piezometer until May 2006, and the deep piezometer only has two records for this entire period. It is likely that BH 2 is showing seasonal shifts in water table dynamics, with the removal of water beyond the response zones of both piezometers after May 2005. As these response zones range from 1.5–4.5m overall, the data would suggest that the cumulative effect of pile insertion is the reduction of the water levels to >4.5m below the site at this location. On a cautionary note, it should be observed that the data points after May 2005 move from monthly to quarterly monitoring, which is both smoothing and masking trends in the data after this time. The fact that BH 2–4 all fail to show any (or only minimal) water table activity across the period April 2005 to May 2006 (Figs 5.2 and 5.3) indicates that severe disruptions in water table responses have occurred as a result of the piling activities at this site. This is reinforced by a comparison of the data recorded prior to pile insertion, wherein the water table activity commensurate with good levels of saturation is in evidence. For the equivalent period in 2005 (August-November), no water table data are recorded for BH 2–4. The rainfall data supplied by the Met Office (Figure 5.4) shows that in general 2005 is broadly equivalent to 2004 in terms of measured rainfall, although August 2004 was considerably wetter than 2005. The rainfall data demonstrates that precipitation is not the cause of the marked lowering of water tables at this site in 2005. It should be noted that the differences in water levels recorded by the deeper piezometers are probably a reflection of the fact that the groundwater table is not flat, and that these piezometers are reflecting topographic variation, which while unproven, is probably associated with subsurface features. The insertion of 923 piles across the area of the Vancouver Centre has resulted in considerable disruption to the water table beneath this site, this is especially marked across 2005, and is demonstrated by the lack of data from all of the monitoring points across this period. This is unsurprising given the nature of the piles used (driven piles), and the density of piles in a relatively small area. The variability in response demonstrated between BH 1 and the remaining boreholes appears to be an artefact of location, but this would require confirmation by stratigraphic survey. The lag time associated with the lowering of the water table from early in 2005 may be due to the re-establishment of equilibrium after vertical displacement at depth, both in relation to the sediment and the water body, during pile insertion. Water level readings obtained from 20/10/05 to the end of the monitoring programme appear to indicate that the groundwater is returning to some measure of equilibrium at BH 1 and 2, and that the levels are commensurate with those obtained at the start of the monitoring programme. The lack of data from BH 3 would suggest data generation in advance of development has been advocated (Lillie 2007), and in particular data of sufficient resolution to facilitate comparisons between annual/seasonal variability is necessary. As such, baseline data of a minimum duration of twelve months would ensure that realistic comparisons are made when assessing the impacts of development at a specific location. The baseline data generated at the Vancouver Centre does facilitate a degree of comparability, especially as the subsequent monitoring is of a relatively long duration. In the shallow piezometers, the water levels that were obtained on 16/11/04, after pile insertion on 06/11/04 adjacent to BH 1–3 (as shown in Pile Sheet No. 044393), display shifts in relation to the general pattern of water table activity prior to this date. These data indicate that the insertion of the piles has changed the groundwater dynamics. The shifts that are immediately apparent in the water levels are that BH 1 decreases by 0.25m, while the water level in BH 2 increases in depth by 0.27m, subsequently rising through to February 2005. BH 3 exhibits a decrease in depth of 0.15m, and continues this trend of decreasing water level through until December 2004. The water level data that was obtained from BH 4 on 30/11/04 (shallow piezometer) indicates that the insertion of piles adjacent to the monitoring point on 26/11/04 (as shown in Pile Sheet No. 044386), caused immediate shifts in relation to the pattern identified previously. The water level decreases by c. 0.5m. However, by the time of the next monitoring visit (12/12/04), the water levels are generally equivalent to those measured prior to pile insertion. When the data from the shallow piezometers is compared to that from the deeper piezometers at this site (Figs 5.2 and 5.3) it is immediately apparent that while BH 1 and 2 are exhibiting a lowering of water levels commensurate with the impact of piling (Figure 5.2) at the pile insertion date (06/11/04, as shown in Pile Sheet No. 044393), the data from BH 3 and 4 fails to register any impacts. This would suggest that the degree of saturation of the sediment profile at BH 3 and 4 is already lower than at BH 1 and 2. In effect, the water table indicated by the data from the deeper piezometers at BH 3 and 4 is a more accurate reflection of the water levels at these locations. While the shallow piezometer at BH 3 exhibits a water level height that is commensurate with that in evidence at BH 1 and 2, it is likely that this is a consequence of the head potential at BH 3. Taken together, the data from each of the monitoring points, recorded after pile insertion, reinforces the observation that BH 1 and 2 are showing shifts in their hydrological characteristics that are reflecting the impacts of pile insertion on the sediment sequences, and the disruption of water movement in and around the site (Figures 5.2 and 5.3). In general piezometers 1 and 2 show an increase in water levels, after pile insertion. This occurs through until May 2005 in the case of the shallow piezometer at BH 1 and until February in the case of the shallow piezometer at BH 2. The lack of data for BH 2 between February and May is accounted for by the matching drop in water levels recorded by the deep piezometer (Figure 5.3) at this location, indicating that the head potential is reduced at this point in the monitoring. The data for BH 1 and 2, from February 2005 onwards (through until the end of the monitoring), shows two contrasting scenarios in the data 87 30/11/2004 13/12/2004 12/01/2005 08/02/2005 30/03/2005 82.5 -16.8 94.6 118.3 104.4 93.7 67.5 53.3 5.5 15.2 42.8 92.4 -36.6 51 BH 3 7.7 -42.9 26.8 108.3 80.1 140.5 192.4 191.6 176.4 75.3 BH 4 83.7 -54.7 -38 -114.5 -28.7 -64.2 -34 129.5 -105.6 -21.5 -30.6 -116 7.8 05/09/2006 16/11/2004 55.2 22.6 04/05/2006 01/11/2004 137.4 21.2 30/01/2006 15/10/2004 110.3 -1.2 20/10/2005 29/09/2004 -15.3 BH 2 07/07/2005 15/09/2004 BH 1 12/05/2005 24/08/2004 06/08/2004 Shallow boreholes Original redox potential values -40 130.2 71 94.7 69.9 64.2 -14.3 31 74.7 -54.3 33.3 BH 3 49 48.8 9.1 96.5 56.5 122.8 186.5 173.9 152.8 57.6 71.9 05/09/2006 -34.5 -12.2 04/05/2006 71.9 29.7 30/01/2006 25.7 4.9 43.9 -67.4 -61.6 -138.1 -99.5 -70.1 -39.9 94.1 -123.3 -67.8 1.9 20/10/2005 107.5 -26 07/07/2005 80.8 51.9 12/05/2005 90.9 BH 2 30/03/2005 08/02/2005 12/01/2005 13/12/2004 30/11/2004 16/11/2004 29/09/2004 01/11/2004 15/09/2004 BH 1 BH 4 15/10/2004 24/08/2004 06/08/2004 Adjusted redox potential values 42.4 -110.1 -51.8 106.6 16/11/2004 13/12/2004 12/01/2005 08/02/2005 12/05/2005 07/07/2005 20/10/2005 -54.3 -70.7 65.6 -40.7 -47.8 -47.3 -14.5 -16.2 -32.6 -84.7 -131.8 -84 -107 11.5 57.4 35.8 48.9 90.5 -62.7 115.4 138.1 29.9 -12.2 -42.7 BH 3 -44.9 -71.8 29.2 20.1 40.1 49.8 40.6 47.6 34.4 150.6 BH 4 -15.1 -29.2 4.9 104 30.6 -6.2 35.9 -36.4 -71.9 -36.4 -50.8 -56.3 -23.3 -108.9 -11.2 05/09/2006 01/11/2004 -3.9 BH 2 04/05/2006 15/10/2004 BH 1 30/01/2006 29/09/2004 30/03/2005 15/09/2004 30/11/2004 24/08/2004 06/08/2004 Deep boreholes Original redox potential values -37 -90.4 -119 -99.4 -48 -42 21.3 117.1 04/05/2006 05/09/2006 42 64.3 -34.5 71 26.3 69 4.9 37.1 -12.2 -14.3 109.5 125.9 -24.3 33.3 -62.1 -102.4 -155.4 -107.8 -138.1 -147.2 -120.7 -42.9 BH 3 37.7 -77.7 11.5 2.4 10.6 20.3 17 24 4.9 138.8 BH 4 20.3 52.8 -24.6 86.3 -4.8 29.8 18.2 -65.9 -101.4 -60 -60.4 -80.3 -74 -8.2 30/01/2006 -94.3 57.4 20/10/2005 66.1 129.5 07/07/2005 39.3 BH 2 12/05/2005 BH 1 30/03/2005 08/02/2005 12/01/2005 13/12/2004 30/11/2004 16/11/2004 01/11/2004 29/09/2004 15/10/2004 15/09/2004 24/08/2004 06/08/2004 Adjusted redox potential values -146.9 -96.3 -113.1 -42 87.6 -123 -65.7 Table 5.1 Original and adjusted redox values for the shallow and deep boreholes recorded from August 2004 to September 2006 One final observation that needs to be highlighted at this point is that the bi-weekly data obtained during the monitoring programme exhibits a considerable degree of variation when compared to post-pile insertion data recovered at monthly/quarterly intervals. This smoothing of the monthly/quarterly data is restrictive from an interpretation point of view and when readings are obtained at the later dates during the programme, subtle changes occurring as a result of seasonality or location are being lost. Bi-weekly monitoring of data relating to the physico-chemical characteristics of the burial environment are advocated here, and have been proposed by recent studies (e.g. Smit et al. 2006, 65). that at monitoring points BH 3 and 4, the reductions in water table indicated by the 2005 data are sustained, and that the water table at both of these locations is now placed at c. -3 to -4m below the modern ground level. These reductions may be anticipated as having an impact at BH 3, but are equivalent to those in evidence at BH 4 prior to the insertion of piles at this location. The water level results obtained from the deep boreholes at monitoring points 1 and 2 over the duration of the monitoring programme have shown broadly similar patterns to the water levels taken from the associated shallow boreholes, which occur between 06/08/04 and 06/02/05, and after 04/05/06 through to the end of the monitoring period. The loss of the water table across the period May to October 2005 is considered to be a direct result of the insertion of piles into the surrounding sediment, as this event is mirrored at all of the monitoring points and in the response zones of both the shallow and deeper piezometers, with the exception of the deep piezometer at BH 1 (as discussed above). Redox potential To present the results from the acquisition of the redox data it is necessary to adjust each redox value to pH 7 in order to remove pH variability between different types of sediments (Smith 2005; Smith and Lillie in press). This is achieved by adding a correction factor of -59 per unit of pH for values above pH 7, or subtracting the same correction 88 more reducing conditions within the environment. However, the redox value obtained on 30/11/04 indicates that the burial environment has re-established its previous equilibrium. The redox values obtained from BH 2 (16/11/04) immediately after pile insertion (06/11/04) display similar values to those readings taken since the start of the monitoring programme. However, during the next two monitoring periods (30/11/04 and 13/12/04) redox values show a shift towards more oxidising conditions (-14mV to 74mV) although the category of redox potential still remains the same (i.e. reducing). This trend suggests that the insertion of the piles adjacent to BH 2 has slightly increased the oxidising nature of the burial environment during this period, and that the elevated water levels recorded across this period are oxygenated when compared to the period prior to pile insertion. The redox values obtained from BH 3 (16/11/04) show an immediate increase in oxidising conditions from 57mV to 123mV after the insertion of piles on 06/11/04. Redox values continue to rise during the next monitoring period (30/11/04) from 123mV to 187mV. Subsequent redox values indicate a gradual decrease until the reducing nature of the burial environment is attained on 08/02/05. The initial increase in redox potential values is caused by the introduction of oxygen into the sediment by pile insertion. However, the subsequent uptake of oxygen by the microbial communities present within the sediment promotes the re-instatement of more reducing conditions in February 2005. There are insufficient redox readings for BH 4 to allow for the interpretation of any trends within the data. The lack of water within the borehole makes it impossible to monitor the redox potential of the groundwater. However, although six water table measurements were obtained between October 2004 and January 2005, only one redox measurement (30/03/05) was collected. It is not possible to identify why this lacuna in data exists. From 08/02/05 to the end of the monitoring programme, the redox values obtained from BH 1 and 2 show values that are reflecting the general condition within the burial environment, but the data for BH 1 needs to be treated with caution as the impacts of piling on the water table dynamics at this location have been brought into question (see ‘Water levels’ above). There are insufficient data to interpret the redox values that are associated with BH 3 from 08/02/05 until the end of the monitoring programme due to the lack of a water table at this point. The redox potential values obtained from the deep boreholes over the duration of the monitoring programme are displayed in Figure 5.6 below. Although the redox values for BH 1, 3 and 4 vary from one monitoring visit to another, they still remain in the same category of redox potential. As a consequence of this, the results indicate that prior to, during and after the insertion of piles adjacent to BH 1, 3 and 4 until 30/03/05, no significant impact can be identified. The redox values generally fall between -100mV and +100mV, which are indicative of reducing conditions (with the exception of BH 3 on 08/02/05). This situation probably reflects that fact that the deeper boreholes are in equilibrium with the water table. factor below pH 7 (Bohn 1971; British Standards Institute 1990). Intermediate pH unit corrections are made proportionately, i.e. for pH 6.2 a factor of 0.8 x 59 would be adjusted from the redox value. Table 5.1 highlights the original and adjusted redox values for each of the monitoring periods, for the shallow and the deep boreholes. The original values are taken from the results of ground investigations by CL Associates (2006, tables 1–18). The standard classification used to define the redox status of soils originated from the research by Patrick and Mahapatra (1968) for well-drained and waterlogged soils during studies of rice production. Table 5.2 (below) presents the categories of redox potential. Redox potential (mV) Category >+400 Oxidized +100 to +400 Moderately reduced -100 to +100 Reduced -300 to -100 Highly reduced Table 5.2 Categories of redox potential Numerous subsequent studies have used these standard categories in order to measure the redox potential of burial environments (Brunning et al. 2000; Caple 1996; 1993; Caple and Dungworth 1997; Hogan et al. 2001; Lillie and Smith in press; Smith 2005; Smith and Lillie in press). It is on this basis that the same scheme will be employed to describe the redox results obtained from the groundwater investigation at Zone D of the Vancouver Centre. The redox potential values obtained from the shallow boreholes over the duration of the monitoring programme are displayed in Figure 5.5 below. The redox values taken prior to the insertion of piles adjacent to BH 1–3 display measurements between c. +100mV and -30mV; these values are indicative of a reducing environment. Although many redox values vary between monitoring periods, they are still within the same category outlined above. One must be aware that when interpreting redox data, the use of redox values can only serve as a generic indicator regarding the type of conditions occurring within a variety of different soils, owing to the use of Eh creating unavoidable inaccuracies associated with the measurement of mixed potentials within the burial environment (Bohn 1971). The redox values obtained on 16/11/04, after pile insertion on 06/11/04 in close proximity to BH 1 (as shown in Pile Sheet No. 044393), display shifts in relation to the pattern prior to this date. These figures indicate that the insertion of piles has changed the redox potential of the burial environment. The redox value obtained from BH 1 on 02/11/04 was 72mV, while that obtained on 16/11/04 was 35mV. These data suggest that the insertion of the piles adjacent to BH 1 has promoted Figure 5.5 Redox potential values obtained from the shallow boreholes over the duration of monitoring. Values adjusted to pH 7 89 Figure 5.6 Redox potential values obtained from the deep boreholes over the duration of monitoring. Values adjusted to pH 7 Figure 5.7 Dissolved oxygen values obtained from the shallow boreholes over the duration of monitoring Figure 5.8 Dissolved oxygen values obtained from the deep boreholes over the duration of monitoring 90 Figure 5.9 pH values obtained from the shallow boreholes over the duration of monitoring Figure 5.10 pH values obtained from the deep boreholes over the duration of monitoring identified in BH 1; suggesting that a measure of seasonality is occurring in the dataset over the study period. The redox values obtained from the remaining borehole (BH 2) display similar trends to those highlighted above, until the insertion of the piles adjacent to this borehole on 06/11/04. Although pile insertion does not change the redox values associated with the subsequent monitoring period (16/11/04), they do however affect the redox values obtained on 30/11/04 and 13/12/04. Measurements taken from these two monitoring periods indicate that the reducing conditions have been replaced by moderately reducing conditions (i.e. 110mV on 30/11/04 and 126mV on 13/12/04). Subsequent redox values obtained from BH 2 indicate a return to the measurements obtained prior to pile insertion. The remaining redox values recovered after pile insertion, to the end of the monitoring period, can only be interpreted from BH 1 as insufficient values exist from BH 2, 3 and 4 due to the lack of a water table. However, the redox values associated with BH 1 could be compromised as a result of pile insertion. Despite this, the fact that the data are recovered from the deeper borehole, with a response zone between 3.5m and 4.5m, could be used to argue that some validity in this data set might be anticipated. If this is indeed the case, then saturation is showing that there appear to be lower redox values associated with highly reducing conditions during the winter months and higher values associated with reducing conditions in the summer months. The redox values obtained from BH 2 (between 30/01/06 and 05/09/06) and BH 4 (between 04/05/06 and 05/09/06) highlight similar trends to those Dissolved oxygen content The dissolved oxygen content (DOC) values obtained from the shallow boreholes are shown in Figure 5.7. The general trend which can be identified from the data prior to, during and up to one month after adjacent pile insertions indicates DOC levels of between 1 mg/l and 5 mg/l. These values are indicative of reducing conditions and are below the average measurements obtained in water (between 6 mg/l and 14 mg/l). In October and November 2004 the measurements obtained (1 mg/l to 2 mg/l) indicated the presence of hypoxic conditions (ANZECC/ARMCANZ 2000). The exception to the general trend is the value obtained from BH 3 (9 mg/l) at the start of the monitoring (06/08/04). It is suggested that this outlying value may be associated with contamination of the groundwater during building development and/or water ingress from above (promoting an increase in the dissolved oxygen content of the water). There are insufficient data available after 08/02/05 until the end of the monitoring programme from which an interpretation of the DOC values can be generated, especially from BH 3 and 4. In contrast, the values from the remaining two boreholes (1 and 2) are reflecting DOC levels that mirror the water level data across the period February 2005 to 91 Figure 5.11 Temperature values obtained from the shallow boreholes over the duration of monitoring Insufficient information exists on the pH values associated with BH 3 and 4 after 30/03/05 for interpretation purposes. February 2006. The reliability of these data has been brought into question in light of the discussion relating to the differences between the shallow and deep water table data generated at monitoring point 1 (BH 1). The available data are of insufficient resolution to allow for a realistic assessment of the DOC across this period. After February of 2006 it could be argued that the low DOC content values evident in the winter months (between 0 and -3 mg/l) contrast to those values obtained during the summer months (between 0 and 5 mg/l). This is usually considered to be due to a decrease in the temperature of the water in winter, which increases the potential for oxygen dissolved within it. The low dissolved oxygen content values in the winter months indicate hypoxic to anoxic conditions (ANZECC/ARMCANZ 2000). Figure 5.8 shows the DOC values for the deep boreholes over the duration of the monitoring period. The values display a similar trend to those identified in the shallow boreholes (as shown above), that is, some suggestion of seasonal variation exists, with values of between 0 and -3 mg/l occurring in the winter months (anoxic conditions) and values of between 0 and 5 mg/l occurring in the summer months (hypoxic/anoxic conditions). There are insufficient data for interpretation of the values obtained from BH 3 and 4 from 08/02/05 and 12/05/05, respectively. In general, the results from the shallow and deep boreholes highlight the similarity that exists between the dissolved oxygen concentrations of the groundwater samples obtained from different depths on the site. Temperature The temperature values obtained from the shallow boreholes over the duration of monitoring are shown in Figure 5.11. As can be seen, the temperature of the groundwater from BH 2 and 3 between October 2004 and February 2005, and from BH 1 between October 2004 and April 2005, exhibit consistency in the seasonal trends highlighted. The temperature values decrease over the winter period from 17ºC to 9ºC across 2004–5, and exhibit a similar pattern across the winter of 2005–6 in BH 1 and 2. During the summer months (August to September 2004 and May to September 2005–6) temperature values increase from 11ºC to 20ºC. Insufficient data exists after 08/02/05 until the end of the monitoring period from BH 3 and 4, for any discussion of significance to be undertaken. The insertion of piles adjacent to these boreholes is not reflected in changes in the temperature of the groundwater. Figure 5.12 presents the temperature values obtained from the deep boreholes over the duration of the monitoring period. These values mirror the results obtained from the shallow boreholes which again appear to indicate seasonal changes in temperature between October 2004 and April 2005, and October 2005 to April 2006 (the winter months); and August to September 2004 and May to September 2005–6 (the summer months). Once again, as shown above, insufficient data exists for the accurate interpretation of the temperature values obtained from BH 3 after 08/02/05 and BH 4 after 30/03/05. pH values The pH values obtained from the shallow boreholes over the duration of the monitoring programme are shown in Figure 5.9. The pH values obtained from all four boreholes remain between pH 6.9 and pH 7.5 over the two-year period. There are however two exceptions to this trend; neither of which are associated with the insertion of piles during November and December 2004. The first exception concerns the pH readings obtained from BH 1, 2 and 3 on 24/08/04. This may be due to contamination of the groundwater from site construction, the products of which may percolate down the borehole and/or adjacent to it. The second exception is the increase in pH values on 30/01/06 for BH 1 and 2. It is suggested that in this period, groundwater levels were particularly high (as shown in Figure 5.2). This would either promote the influence of a different level of sediment upon the water body, or be influenced by an increase in the percolation of rainwater through the profile of the borehole (possibly as a result of a significant rainfall event); hence causing a shift in pH values during this period. Insufficient information exists on the pH values associated with BH 3 and 4 after 08/02/05 for accurate interpretation. Figure 5.10 displays the pH values obtained from the deep boreholes over the duration of the monitoring programme. The trends identified mirror those pH values obtained from the shallow boreholes, i.e. the pH readings obtained from BH 2–4 on 24/08/04 are lower than the general trend, and the pH values obtained from BH 1 and 2 indicate an increase in alkalinity (pH 9.1 and pH 9.3 respectively) on 30/01/07. Due to the similarity of timing between these events in both the deep and shallow boreholes it is likely that the same causal factors are responsible. Discussion Comparison between water levels and adjusted redox potential values Figure 5.13 displays the combined results obtained from the water level depths and adjusted redox potential measurements for the shallow boreholes over the duration of the monitoring period. As has previously been stated in the water levels section, BH 1–3 display shifts in groundwater depths immediately after the insertion of adjacent piles on 06/11/04. BH 4 shows a change in groundwater level associated with pile insertion, but this is only of a very limited duration. The insertion of the piles into the sediment in close proximity to the monitoring boreholes results in fluctuations in the level of the groundwater, and a shift in the overall pattern of groundwater activity when contrasted against the data generated prior to pile insertion. The results of the borehole monitoring on subsequent site visits indicate that 92 Figure 5.12 Temperature values obtained from the deep boreholes over the duration of monitoring piling, the data is unlikely to be an accurate reflection of the general conditions in the burial environment across the site between February and August 2005. The water level depths obtained between 26/11/04 and 08/02/05 for BH 3 display similar patterns of displacement to those previously identified in BH 1. In contrast, the water levels obtained from BH 2 display a pattern which is commensurate with seasonality in the data as both shallow and deep piezometers are reacting in the same way through until May 2005. It is therefore likely that from 08/02/05 to the end of the monitoring programme, the limited redox values obtained from BH 2 are values associated with seasonal trends. The available redox values from this period indicate that the water level measurements obtained from both boreholes at this monitoring point (BH 2) contained more oxygen, which is considered to be occurring as a direct result of the insertion of piles into the surrounding sediment. The limited information available regarding the presence of groundwater in BH 2, 3 and 4 from 08/02/05 to the end of the monitoring programme (and in particular from April 2005 to May 2006) indicates that the insertion of piles has resulted in severe disruptions in water table response at this site. There is insufficient data to interpret the redox values which are associated with BH 3 from 08/02/05 until the end of the monitoring programme, due to the lack of a water table across this period. In general the data obtained from May 2005 to May 2006 do not reflect dominant rainfall patterns for this period (Figure 5.4 above), and the data would suggest that the piling is the primary variable in the reduction of the water table across the area of the survey. Only three monitoring visits to BH 4 from 12/12/04 to the end of the monitoring programme provided evidence of a water table. This information indicates that the insertion of piles adjacent to BH 4 has significantly impacted upon the water table, and it is not possible to identify whether equilibrium will be re-established in the future. The lack of water within the boreholes at this location (BH 4) makes it impossible to monitor the redox potential of the groundwater. Furthermore, even during the water levels in all four boreholes exhibit a degree of re-establishment towards the levels recorded prior to pile insertion, but that BH 2 exhibits a marked rise when compared to BH 1. The redox values also highlight a shift in the general pattern of values when compared to the pre-pile insertion data, and exhibit trends associated with the insertion of piles adjacent to the shallow boreholes. The values obtained from BH 1 and 3 show immediate changes in the redox potential of the groundwater at these points. The redox values obtained from BH 1 show an increase in the reducing nature of the environment (which corresponds to a decrease in water level), while the values from BH 3 indicate a shift to more oxidising conditions (which correspond to a small increase in water levels). Although the redox values obtained from BH 2, immediately after pile insertion, display similar readings to those taken from the start of the monitoring programme, during the subsequent two monitoring periods the redox values show a shift towards more oxidising conditions (as shown in ‘Redox potential’ above). When compared against the water level depths recorded after pile insertion, the redox values do not react in precisely the same way as the data for BH 1 and 3. Despite this, the data from the following two monitoring periods show clear evidence for an increase in water depth which is accompanied by a change in the redox potential of the groundwater towards more oxidising conditions (albeit with a lag time of approximately two weeks). As has been previously outlined, the water level depths obtained for BH 1 indicate that the insertion of piles on 06/11/04 into the adjacent ground changes the level of the groundwater for a period of approximately six months. After this period, the water levels from 12/05/05 onwards display data that is inconsistent with the deeper borehole at this location. The interpretation here is that this separation of water level data reflects near-surface disruption of water table activity as a direct result of pile insertion. Thus, while the redox values obtained after 30/11/04 suggest that the groundwater had re-established equilibrium within several weeks of the cessation in 93 Figure 5.13 Water level depths and adjusted redox potential values obtained from the shallow boreholes over the duration of monitoring. The dashed lines indicate the redox potential values the monitoring period. As has been stated under ‘Water levels’ above, BH 1 and 2 display a lowering of water levels commensurate with the impact of piling, while the groundwater depths associated with BH 3 and 4 do not indicate change immediately after pile insertion. It is suggested that the lack of response in BH 3 and 4 reflects that fact that the degree of saturation at depth is equivalent to the prevailing water table, which is not producing an immediate response to piling. In contrast, in addition to the fact that the redox values obtained from BH 3 and 4 do not display trends associated with pile insertion (see ‘Redox potential’ above), a similar lack of reaction is apparent in BH 1. However, the two sets of redox values obtained from BH 2 immediately after piling and at subsequent two-weekly monitoring visits, indicate that the piling has impacted upon the redox levels for approximately one month after pile insertion. Measurements taken from these two monitoring periods show that the reducing conditions that characterise this location prior to piling have been replaced by moderately reducing conditions. While similar patterns of water table responses occur between 06/08/04 and 08/02/05, and after 04/05/06, the water level data obtained between 08/02/05 and the end of the monitoring programme from the deep piezometer at BH 1 displays a pattern that differs to that obtained from the shallow piezometer at this location (Figure 5.13), and as such, the deep piezometer could be reflecting seasonal trends. This is supported by the observation that this pattern is also reflected in the overall redox values obtained during this period for BH 2. However, contrary to this, the low level of groundwater present in BH 2 (shallow piezometer) suggests that pile insertion on 16/11/04 not only has an initial impact upon the water level, but also a subsequent (cumulative) effect between February and May 2005, when no water is recorded within the piezometer. Insufficient values exist from BH 2 for an accurate interpretation, due to the lack of a water table during this period. The limited water level data available for BH 3 and 4 between 08/02/05 and the end of the monitoring period periods of groundwater measurements (between October 2004 and March 2005), only one redox measurement (30/03/05) was collected. The reason for this anomaly is not known. On the basis of the above observation it is apparent that while the monitoring programme is of some duration, the reliance on data obtained from piezometers has proven to be severely limiting both in terms of data recovery and as a corollary, data interpretation. In addition to the above observation, there are a number of associated limitations between the water table data and the redox potential measurements recorded over the monitoring programme in this dataset. These are listed below: • there was only one redox potential value (BH 2) obtained from the first two monitoring visits, despite the collection of groundwater measurements from BH 1–3 during this period • although redox potential values from BH 1–3 were obtained from a monitoring visit on 30/30/05, there is only one associated water table measurement (BH 1) • during the first nine months of the monitoring programme only three redox measurements were obtained from BH 4. This is in contrast to the seven groundwater readings collected during the same period • redox values were obtained on 12/05/05 (BH 4), 07/07/05 (BH 2), 20/10/05 (BH 3) and 08/02/05 (BH 1) even though there is no record of associated water table measurements. Groundwater measurements were collected on 12/05/05 (BH 2 and 3) and 20/10/05 (BH 2), without associated redox values. It is not possible to ascertain why these datasets are incomplete, as the presence of water within the piezometers should allow for the measurement of associated redox potentials. Figure 5.14 displays the combined results obtained from the water level and adjusted redox potential measurements for the deep boreholes over the duration of 94 Figure 5.14 Water level depths and adjusted redox values: obtained from the deep boreholes over the duration of monitoring. The dashed lines indicate the redox potential values Although there are insufficient data available after 08/02/05 until the end of the monitoring programme, from which an interpretation of the DOC values can be generated, after February 2006 it appears that the values from BH 1 and 2 may be indicating seasonal fluctuations. This trend is mirrored by the associated redox values that were obtained over the same period of monitoring. As with the data generated from the monitoring of redox and water levels, there are again limitations inherent in the data generated for DOC and the redox potentials obtained over the duration of the monitoring programme. These include: indicates that the groundwater has been significantly affected by the insertion of piles in the surrounding area. This lack of data prevents the assessment of redox potentials of the groundwater during this period of monitoring. As has been noted above in relation to the comparison of water level data and associated redox potentials obtained from the shallow piezometer monitoring, a number of limitations are inherent in the dataset relating to the results obtained from the monitoring of the deep piezometers at this site. These are outlined below: • • • there was only one redox potential value (BH 2) obtained from the first two monitoring visits, despite the collection of groundwater measurements from BH 1–4 during this period although redox potential values from BH 1–4 were obtained from a monitoring visit on 30/03/05, there are only three associated water table measurements (BH 1, 2 and 4) redox values were obtained on 12/05/05 (BH 4) 07/07/05 (BH 2) and 20/10/05 (BH 3) even though there is no record of associated water table measurements. Groundwater measurements were collected on 12/05/05 (BH 3) 20/10/05 (BH 2 and 4) 30/01/06 (BH 2) and 04/05/06 (BH 3) without associated redox values. • there were two redox potential values obtained from the first two monitoring visits (BH 2 and 4), despite the collection of DOC measurements from BH 1–3 during this period • redox values were obtained on 16/11/04 (BH 4), 12/05/05 (BH 4), 30/03/05 (BH 2), 07/07/05 (BH 2) and 20/10/05 (BH 3), even though there is no record of associated DOC measurements • dissolved oxygen content measurements were collected on 30/03/05 (BH 4) and 12/05/05 (BH 2 and 3) without associated redox values. Figure 5.16 displays the results obtained from the DOC and redox potential values from the deep boreholes over the duration of the monitoring period. The values display similar seasonal trends after February 2006 to those identified in the shallow boreholes (above). In general, the redox potential values obtained from BH 2 indicate the impact that adjacent piling has had upon the groundwater (for a period of one month). The DOC measurements mirror this trend to some degree (as noted above in the shallow piezometer). Throughout the remainder of the monitoring programme (i.e. after 08/02/05) the limited DOC values obtained from BH 1 and 2 appear to reflect seasonal trends in the data. This is mirrored by the redox potential values collected from BH 1 over the same period. Only minimal Comparison between dissolved oxygen content and adjusted redox potential values Figure 5.15 shows the combined results obtained from the DOC and redox potential values for the shallow boreholes over the duration of the monitoring programme. It is apparent that, when compared against the results obtained from the study of the redox potentials associated with BH 1–3, the measurements obtained from the DOC of these boreholes do not demonstrate any significant changes that can be directly related to piling impacts (see ‘Dissolved oxygen content’ above), although BH 2 does exhibit a marked peak shortly after pile insertion. 95 Figure 5.15 Dissolved oxygen content and adjusted redox potential values obtained from the shallow boreholes over the duration of monitoring. The dashed lines indicate the redox potential values data exists from the DOC measurements of BH 3 and 4 until August 2005, limiting the observations that can be drawn from this dataset. There are several limitations in the dataset for the deeper piezometers. These are outlined below: • • there were only three redox potential values obtained from the first two monitoring visits (BH 2–4), despite the collection of dissolved oxygen measurements from BH 1–4 during this period • redox values were obtained on 30/03/05 (BH 3), 12/05/05 (BH 4), 07/07/05 (BH 2) and 20/10/05 (BH 3), even though there is no record of associated DOC measurements pH and temperature values The pH and temperature values obtained over the duration of the monitoring programme have been discussed in the relevant sections above. The results of the pH values recovered from the shallow and deep boreholes indicate that stable, near neutral conditions prevail (between pH 6.9 and pH 7.5) across the monitoring period. The temperature values obtained from the shallow and deep boreholes also display marked similarities throughout the monitoring programme. Seasonal trends in dissolved oxygen content measurements were collected on 12/05/05 (BH 3), 20/10/05 (boreholes 2 and 4), 30/01/06 (boreholes 2 and 4) and 04/05/06 (BH 4) without associated redox values. Figure 5.16 Dissolved oxygen content and adjusted redox potential values obtained from the deep boreholes over the duration of monitoring. The dashed lines indicate the redox potential values 96 DOC values that characterise the burial environment at this location. Although the groundwater levels show an immediate shift in relation to piling activity, there is some suggestion of an interval of approximately two weeks between changes in water levels and associated changes in redox potential and DOC values. While this is not displayed in all of the readings obtained from the redox and DOC values, researchers must be aware of its implication in assessing how the physico-chemical status of the burial environment changes both during and after piling activities. The environmental results indicate that after an initial reinstatement of water levels, which are broadly equivalent to those in evidence prior to development, the cumulative effects of the piling activities become apparent from February to May 2005. After this period, the intrusive nature of the piling displaces the water from the shallow boreholes, and also from deep BH 2–4, until the end of the monitoring period. In general, the evidence indicates that groundwater equilibrium is not re-established before May 2006, and at this point only BH 1 and 2 exhibit some level of equilibrium. Only further water table monitoring will determine whether any reinstatement of water tables will occur at this location. Over the duration of the monitoring programme, the insertion of 923 piles on site has had a significant effect upon the groundwater levels (as measured by the four boreholes) and probably influenced the physico-chemical status of the burial environment in Zone D of the Vancouver Centre. As a consequence, it should be assumed that the in situ burial conditions beneath the Vancouver Centre will be compromised, and as noted, the disruptions will extend down to depths of c. 3 to 4m below the modern ground level at this location. This situation will significantly enhance the potential for the microbiological decay of the organic archaeological deposits within the underlying sediments as oxygen levels increase, promoting a shift from anaerobic to aerobic conditions within the underlying sediment. Furthermore, the data generated up until the end of the monitoring programme indicate that groundwater equilibrium has not been re-established. As a consequence, it is impossible to determine the continued impacts that the piles will have upon the archaeological resource. The data generated are uneven in resolution, and a number of alternative measures for recording in situ parameters in the burial environment are recommended below. However, it should be emphasised that the results discussed above are of sufficient resolution to indicate that the cumulative effect of 923 driven piles in a relatively constrained area has had a significant impact on the water table beneath the Vancouver Centre. The nature of the piles used is considered to be a primary factor in the creation of the negative impacts on the burial environment at this location. both datasets are evident, with lower temperatures in the winter months (October–April), and higher values in the summer months (May–September). Neither the pH nor the temperature values provide direct information which can be used to assess the impact of piling activities on the water table over the two-year monitoring period (and hence the preservation status of the burial environment). However, the pH values have been used indirectly when ‘adjusting’ the redox potential measurements obtained from the groundwater for data comparison. Conclusions The environmental parameters described above, which were used to determine the physico-chemical nature of the burial environment at Zone D of the Vancouver Centre, have been used by many researchers prior to this study (Brunning et al. 2000; Caple 1996; 1993; Caple and Dungworth 1997; Lillie and Smith in press; Smith 2005; Smith and Lillie in press). However, in contrast to many previous investigations all of the measurements assessed in the current study have been obtained from groundwater samples at four monitoring points, comprising paired shallow and deep piezometers (Figure 5.1). Although comparable data can be obtained between in situ soil and groundwater measurements in order to characterise the burial environment, recent research has shown that groundwater monitoring can produce erroneous results which are difficult to replicate (Matthiesen et al. 2004). This situation is further compounded by the lack of water recorded from the boreholes at certain times during the monitoring period. The lack of water makes it impossible to record data for monitoring of the associated parameters such as redox potential, pH, temperature and dissolved oxygen content. Groundwater sampling therefore inhibits the generation of environmental data for interpretative purposes during the monitoring programme. The environmental dataset is incomplete, with many values for each of the parameters measured being missing. Although it is not possible to determine why this lacuna occurs, it is somewhat surprising, particularly given the fact that the presence of groundwater enables the measurement of all other associated environmental parameters to take place. As has been mentioned under ‘Previous studies’ above, two types of piles are primarily used in construction; these are displacement/non-displacement piles and flight augured piles. Displacement piles are damaging to archaeological deposits due to displacement of the ground that surrounds the pile (Williams 2006). [Editor’s note — ‘Displacement piles were used at the Vancouver Centre due to the aggressive water table which had contaminants and also was strong flowing, which in seams would wash out the poured concrete, so steel tubes were sleeved and used precast’ — pers. comm. Kevin Malle, Eastern Area Director Alfred McAlpine]. Recommendations As can be seen from the interpretation of the results generated from the two-year monitoring programme of the burial environment at the Vancouver Centre, the insertion of piles into the sediment adjacent to the shallow and deep boreholes has impacted on the groundwater levels, redox potential values, and to a lesser extent, the Due to the restrictive nature of the environmental monitoring of the groundwater present within the shallow and deep boreholes, it is strongly recommended that future monitoring of archaeological deposits in this type of development should include the in situ analysis of redox potentials in the burial environment. This can be achieved 97 points are offered as a minimum requirement for any future monitoring programme where it is anticipated that waterlogged burial environments with the potential to preserve archaeo-environmental remains are encountered: by the insertion of a number of platinum-tipped redox probes of different lengths into the soil matrix at various pre-defined locations, in order to identify the oxidizing-reducing nature of the deposits, without the reliance on the presence of groundwater for assessment purposes. Previous research has shown that in situ redox probes produce more stable, reliable readings than their portable counterparts (Matthiesen et al. 2004). The generation of complete environmental datasets would help the interpretation of the values obtained during the monitoring programme. The lack of data in many of the environmental parameters considered prevents the accurate identification of trends within the data. Although this is partly due to the destruction of BH 3 in January 2006, which prevented further data generation, the majority of the missing values cannot be explained, as data for groundwater values were obtained during associated site visits. In order to determine any seasonal changes in the physico-chemical nature of the burial environment at the Vancouver Centre prior to pile insertion, future investigations should include the collection of baseline data for a minimum of one year before development begins. This information can be subsequently used in order to produce greater accuracy in the identification of piling impacts during the development process. An analysis of borehole transects across the area of development is of fundamental importance for the understanding and interpretation of water table reactions to piling, as the sedimentology of the area will exert an influence on water movements. To enhance the accuracy of the monitoring programme is it also recommended that a greater number of boreholes be installed over the site in question in order to monitor at a higher resolution groundwater fluctuation prior to, during and after pile insertion. The boreholes would be inserted in a grid format over the site to maximise the assessment of pile impacts on the groundwater level. The current locations of the four boreholes which are situated around the perimeter of the site cannot provide accurate information on the hydrology of the burial environment within Zone D of the Vancouver Centre. This is due to the fact that as the water table is displaced by pile insertions adjacent to the monitoring boreholes, incipient shifts within the level of the groundwater will therefore occur in all locations on the site wherever further piling is undertaken. In order to avoid damage to monitoring points, reinforced concrete sectional manholes should be installed, whenever possible, over monitoring locations (as undertaken previously Lillie and Cheetham 2002a and b). Piezometers, redox clusters and coring for the retrieval of pH, temperature and other data could easily be recovered from these locations if necessary. The resolution of the data retrieval would be enhanced by bi-weekly monitoring (as advocated by Smit et al. (2006), as the data smoothing (associated with quarterly monitoring) and lack of data (due the an absence of water caused by disruption of the water table due to piling activities) highlighted in the current study, has proved to be severely limiting in relation to the interpretation of the data generated. As it was anticipated that the data generated during the current monitoring programme would help to inform future planning applications within the area, the following • all parties involved in the planning process need to be on site at the start of the development process in order to agree the programme of monitoring — PPG16 requirements must be adhered to at this stage of the process • a detailed stratigraphic survey of the area to be monitored is required in order to ensure that monitoring points are located strategically (Lillie 2007) • in areas where the activities associated with the development have the potential to impact on monitoring locations, protective measures, such as the use of reinforced concrete sectional manholes, should be a priority • it is essential the seasonal data is generated so that preand post-development impacts can be assessed. As such, a minimum dataset of 12 months duration is needed before any development impacts occur at the site being studied • while dipwells or piezometers are essential for the monitoring of water level activity, and can be used to obtain data on the physico-chemical characteristics of the burial environment, the current study has demonstrated that they are wholly inadequate in situations where disruption to the hydrology of a site might be anticipated. As such, the following measures are recommended: • in addition to dipwells and piezometers, the use of in situ platinum tipped redox probes is essential for the recovery of reliable redox data • the use of boreholes should be considered for the obtaining of pH and temperature data when dipwells/piezometers are dry, in order to ensure continuity in data generation • wherever practicable the resolution of data generation should not be lower than monthly in resolution, and the preferred option would be bi-weekly, in order to reduce the smoothing effects of lower resolution data gathering practices (Smit et al. 2006) • a baseline survey of the development site is required prior to development impacts. This survey should characterise the burial environment through borehole survey and a 12 month programme of environmental monitoring (e.g. water levels, redox, pH, temperature, electrical conductivity, soil moisture content and dissolved oxygen contents — DOC) as a minimum requirement. In addition, in areas where waterlogging of the depositional sequences is identified, an assessment of the physical state of preservation of the archaeo-environmental resource should be undertaken whenever possible. These recommendations represent a minimum requirement for in situ studies, but, while they are not exhaustive, the specific context of the sites investigated will necessitate approaches that either enhance, or perhaps 98 all end users. The current study has shown that the generation of discontinuous, post-development data, severely limits the conclusions that can be drawn from the dataset. In the current study the water table data has provided the most reliable indicator of the impacts of development on the burial environment studied. even reduce, the level of analysis undertaken. Certain parameters can provide overlapping information, while others are time consuming. Finally, it is essential to gain agreement between all parties involved in the development process — this is fundamental to the development of a viable and deliverable monitoring programme that will be of value to 99