REPRESENTATIVE PATHWAY LENGTH FOR CONTAMINANT TRANSPORT IN PERFORMANCE ASSESSMENT MODELS

摘要

Performance assessment models, in the context of high-level waste management, commonly incorporate a "representative" element concept as a simplification approach. For example, on the basis of a representative waste package, radionuclide release away from a single waste package is computed and scaled up to derive an estimate of the total release from an ensemble of waste packages. This computation is an approximation, as each waste package may release radionuclides at different rates controlled by local conditions in or around the waste package. This paper considers several approaches for defining an averaged property, length of transport pathway in this example, so that scaled-up releases are consistent with releases that would be computed by individually simulating each waste package. Results (total release rates) from a detailed formulation are compared to those considering common averages for the derivation of the representative property, and these can differ significantly. This implies that representative properties should be carefully selected in general. The example in this paper illustrates that, even in a simple case, detailed determination of an appropriate representative property can be a complex endeavor. This is particularly true for cases where the representative element (waste package in this example) is intended to summarize multiple properties that exhibit spatial variability. Therefore, depending on the intended use of the performance assessment model, elements may need to be included to explicitly model spatial variability.