How the K-d approach undermines ground water cleanup

Environmental scientists have long appreciated that the distribution coeffi cient (the "K-d" or "constant K-d") approach predicts the partitioning of h eavy metals between sediment and ground water inaccurately; nonetheless, tr ansport models applied to problems of environmental protection and ground m ater remediation almost invariably employ this technique, To examine the co nsequences of this practice, we consider transport in one dimension of Pb a nd other heavy metals through an aquifer containing hydrous ferric oxide, o nto which many heavy metals sorb strongly, We compare the predictions of mo dels calculated using the Ii, approach to those given by surface complexati on theory, which is more realistic physically and chemically. The two model ing techniques give qualitatively differing results that lead to divergent cleanup strategies. The results for surface complexation theory show that w ater flushing is ineffective at displacing Pb from the sorbing surface. The effluent from such treatment contains a persistent "tail" of small but sig nificant levels of contamination, Subsurface zones of Pb contamination, fur thermore, do not migrate rapidly or far in flowing ground water These resul ts stand in sharp contrast to the predictions of models constructed using t he K-d approach, yet are consistent with experience in the laboratory and f ield.