| There are numerous interacting elements within the contaminant transport regime that have critical effects on the behaviour of constituents of concern. Complete and accurate characterization of these components is generally impractical, particularly in heterogeneous regimes. Uncertainty is therefore a reality that must be accepted in the design and operation of pump-and-treat remediation systems.;Empirical knowledge, with respect to full scale application of pump-and-treat remediation systems, is best summarized by a review of case histories. The published case histories available for review do not reveal a discernible trend of system response versus geology, hydrogeology, or contaminant properties. Heterogeneity of the site and inadequate source removal are cited as two common causes of difficulty in achieving remediation success.;Linking scientific knowledge with an empirical database follows from this exercise. Inference, prediction and direct determination are supplied as three methods of characterization of various aspects of a site. A modelling case study illustrates the application of inference methods in this respect.;Uncertainty is ubiquitous in subsurface remediation problems. The question of how to manage this reality follows from acceptance of the low probability of complete characterization in the design phase. A methodology is suggested that accepts uncertainty and responds to it in a structured manner; anticipating possible deviation from previous site conceptualizations and accommodating their demands into the project design. |
