Re-evaluating dual-porosity effects at the site of a seminal groundwater modelling study: Tilmanstone, southern England

Abstract

Abstract The first numerical model of solute transport to incorporate Fickian diffusive exchange between mobile fracture water and immobile porewater for an actual case of groundwater contamination at catchment scale was applied to the Chalk aquifer at Tilmanstone in SE England by Bibby ( Water Resources Research , 1981, 17 , 1075–1081). The unconfined aquifer at Tilmanstone had been contaminated by coalfield brine leaking from disposal lagoons operating throughout much of the twentieth century. Recent observations show that the Bibby model underestimates dual-porosity diffusive retardation, and hence underestimates the persistence of contamination, probably by several decades. 2D representation of the aquifer in plan ignored the hydrostratigraphy, and model calibration was limited by the lack of time-variant paired profiles of fracture water and porewater. Vertical profiles through the Chalk determined by packer testing, borehole dilution testing and geophysical logging, together with a new depth profile of chloride concentration in Chalk matrix porewater, are described. The hydrostratigraphy is interpreted in relation to the Chalk lithostratigraphy of SE England, and incorporated into a vertical-section model of chloride transport along the axis of the valley, consistent with the new and historical profiles of fracture water and porewater chloride concentrations. New predictions of the longevity of the chloride contamination at Tilmanstone are presented.