The use of geoelectrics and test wells for the assessment of groundwater quality of a coastal industrial site

Using geoelectrical depth soundings, potential groundwater pollution was assessed on a former shipyard of the Newport Naval Base, Rhode Island, USA. Besides an increase of the total dissolved solids (TDS), toxic organic compounds and metals were expected in the saturated zone. Two nested test wells were used to support the geoelectrical interpretation and a third one was located in a poorly accessible place. Non-invasive and cost-effective geoelectrics is representative for a sizeable volume, but the field data are non-unique with respect to the interpretation. Test wells, on the other hand, provide precise data, but they are laterally confined, so that a combination of both methods optimizes the results. This survey concentrates on the freshwater-saturated zone of glacial sediments that range in thickness between 1 and 15 m. The sediment is a poorly sorted till that is described as silty, gravelly sand with a low hydraulic conductivity of considerable heterogeneity. The major effort concentrated on a 12-m wide strip located between a sheet pile bulkhead along the shore and a building in which chemicals were stored and handled. This strip was the only place where depth soundings with a depth penetration extending below the freshwater-saturated zone were logistically possible. Other places were unsuitable because of conductive constructions at the surface, concrete pavements, and adjacent buildings. In spite of poor data quality, four of the five geoelectrical depth soundings were identified to be of QH-type (ρ1 \u3eρ2 \u3eρ3 \u3cρ4), with the following hydrogeological correspondence: unsaturated-, freshwater-saturated, seawater-saturated zone, and non-conductive tight glacial till and shale bedrock. The target of this investigation is the freshwater-saturated zone, which is qualitatively recognized on four of the five sounding curves. To avoid an underestimation of the interpreted layer resistivities, the lateral effect from the seawater and iron sheet pile bulkhead was corrected. It was approximated with a semi-infinite vertical boundary between the aquifer and a conducting medium using the image method. Low aquifer resistivities are usually associated with potential pollution in terms of high concentrations of TDS. A semi-quantitative attempt is described to correlate bulk layer with pore water resistivity and finally with equivalent TDS. Because of the presence of clay-sized particles in the sediment, the bulk electrical conductivity is caused by water conduction through the pore channels and the surface conductivity of the particles. Lab sample measurements of resistivity on samples recovered from the test wells provided a calibration curve for the bulk conductivity-water conductivity relationship. In spite of complications due to surface constructions and the heterogeneity of the freshwater-saturated till, it was possible to identify high conductivity groundwater with a conductivity on the order of 3000 μS/cm that is equivalent to approximately 1950 ppm TDS. It is not possible to determine whether or how much of the TDS originates from seawater overwash and how much is attributable to shipyard activities. Test wells, however, showed objectionable amounts of dissolved metals and adsorbed toxic organic compounds. © 2002 Elsevier Science B.V. All rights reserved