DC resistivity and induced polarisation investigations at a waste disposal site and its environments

We report the results from some geoelectrical surveys carried out to monitor the spread of contamination in underlying aquifers due to a landfill site. The geophysically determined electrical properties of the aquifers have subsequently been used to estimate hydraulic conductivities which are required for modelling contaminant transport. The type of waste deposited and the influence of the geological environment were the crucial factors investigated, employing the DC resistivity and time domain induced polarisation (IP) methods. The landfill was mainly a liquid disposal site with existing borehole information showing that the waste contained high concentrations of not only inorganic material (chlorides, sulphates) but also organic matter (indicated by high values of chemical oxygen demand (COD) and total organic carbon (TOC)). The measured fluid resistivities rho(w) from the surrounding boreholes showed values as low as 0.25 Ohm m, leading us to expect low bulk resistivities rho(o). The aquifer system in the study area, with an eastwards regional groundwater flow, consists of three sand aquifers with intervening semi-pervious clay aquitards. Clay particles are also present in the sand formations and expected to influence the overall bulk resistivity and chargeability. The presence of organic waste is another factor suggesting that the IP method could be employed as a diagnostic tool. Our resistivity measurements along the survey lines perpendicular to the groundwater flow show systematic reductions of resistivities relative to the control line, the effect decreasing progressively eastwards from the landfill. The resistivities of these contaminated sections were higher than expected and one possible explanation for this could be the presence of the organic waste. However, an alternative explanation could be the low porosities in the sand formations. Low porosity may imply reduced fluid content, and therefore increased bulk resistivities, as well as lower hydraulic conductivity values. Field and core hydraulic conductivity measurements in the area from previous investigations had indicated the relatively low hydraulic conductivity values of 9.3 x 10(-7) and 6.2 x 10(-6) m/s for the top and middle aquifers respectively, thus, favouring the low porosity hypothesis. The IP measurements showed high apparent chargeability values (80-120 ms) on top of the landfill, possibly due to the presence of disseminated solid metallic waste or the high organic load of the liquid waste disposed. The IP line parallel to the groundwater flow direction, and close to the landfill, produced chargeability anomalies which may be associated with a plume of organic waste. No chargeability anomalies are observed on the second IP line, further away from the landfill and in the SE direction. The bulk resistivities rho(o) obtained from the resistivity inversions and the fluid resistivities rho(w), from adjacent boreholes, allowed hydraulic conductivities to be estimated. The intrinsic formation factor was first determined from rho(o) and rho(w) and was then used in conjunction with a range of Archie' s parameters appropriate for sands to evaluate porosity and its likely bounds of error. The hydraulic conductivities obtained through the Kozeny-Carmen-Bear equation, for the geophysically determined range of porosities, give plausible values agreeing within an order of magnitude with each other and with reported values for the formation. (C) 2000 Elsevier Science B.V. All rights reserved.