Initial investigations for a new technology building at a school in central London came up with some slightly worrying results for foundation designers.The building has a footprint of about 1,200m 2, on part of the school playground.
While four of the six boreholes identified very similar ground conditions, comprising up to 1.1m fill overlying up to 1.3m of clay and 10m of dense sandy gravel (Kempton Park Gravel), two on the north eastern edge of the site showed up to 4.8m of fill.
This indicated a likely area of ancient gravel extraction - a borrow pit - 'which had obvious ramifications for the foundation design of the new block, ' says geophysical engineer Kathryn Watson of Stats Geophysical.
With no surface evidence, the extent of the pit could not be established and data from the investigation was limited. The cost and disturbance of further boreholes to define the pit boundary prompted the local education authority to seek an alternative solution from site investigation firm Stats Geophysical, part of the ground engineering division of geotechnical consultant Stats.
Watson says a range of geophysical methods could have been used, 'as there were no obvious sources of geophysical noise generation such as overhead power cables, reinforced concrete, or obstacles to inhibit surveying, such as cars and people.'
An electromagnetic survey was chosen as the most appropriate method.Work, which took only one day, was carried out using a Geonics EM31 ground conductivity meter. The equipment determines the apparent electrical conductivity of the ground in milli Siemens per metre (mS/m), based on the magnitude of eddy currents that are induced to flow within the sub-surface.
The EM31 was configured to record values to two maximum depths: 'shallow' (3m-4m) and 'deep' (5m6m). Measurements were taken on a 1m by 1.5m grid, giving about 800 individual data points.Data was stored on a data logger and then downloaded to a PC on site to produce a contoured map of average ground conductivity.
'Different rock types and sediments exhibit different values of electrical conductivity depending on their porosity, degree of saturation, conductivity of saturating fluid and the percentage of electrically conductive materials, for example clay and graphite, ' Watson explains. This information allowed different materials to be identified across the site.
With groundwater lying at 6m below ground level, Stats anticipated that the uppermost 3m to 4m of Kempton Park Gravel (the unsaturated zone) would exhibit a lower electrical conductivity than the fill, which was assumed to have a relatively lower porosity and contain more clay.
Background conductivity was relatively low in the 'shallow' data, at about 8mS/m, these values coinciding with the boreholes showing only shallow depths of the gravel. Two anomalous features on the contour map were interpreted as an old gas main, which showed as a large (between 100mS/m and 160mS/m) linear feature running along the western edge of the site, and the borrow pit.
The borrow pit, or at least its western edge, appeared as a relatively weaker anomaly along the eastern edge of the site.
'Although reduced in magnitude, it is still virtually double the background response, exhibiting values of 16mS/m to 18mS/m, 'says Watson.This was the area where the two boreholes had identified almost 5m of fill.
The 'deep' electromagnetic survey did not exhibit this anomaly, however.Aside from the broad anomaly generated by the gas main, the site was characterised by background values of between 14mS/m and 16mS/m.
'This relative increase is attributable to the presence of the groundwater table saturating the gravels and hence increasing their bulk electrical conductivity, ' Watson explains. As bulk conductivity increased, electrical contrast with the adjacent fill fell and the pit become undetectable.
All geophysical surveys require some form of calibration with ground truth data, Watson says. A series of dynamic probes was carried out along the pit boundary as determined by the electro- magnetic survey to confirm the results.These were also correlated with the six boreholes.
Dynamic probing confirmed a significant change in ground conditions coinciding with elevated ground conductivity, with conductivity falling when the interface between the gravels and the fill was at shallow depth.
Watson says the electromagnetic survey allowed a second targeted phase of investigation. Dynamic probing could then be used to obtain 'a very detailed definition of ground conditions' 'The high degree of correlation between the nonintrusive and intrusive techniques gave credibility to the geophysical anomaly, such that the extent of the borrow pit could be confidently extrapolated across the site, 'she adds.
'Some of the most cost effective geotechnical site investigations have employed geophysical surveying prior to any large scale intrusive work. The advantages of this combined approach appear to be registering with specialist geotechnical engineers and general civil engineers alike.'