IDENTIFICATION OF HYDRAULIC CONDUCTIVITY STRUCTURE IN SAND AND GRAVELAQUIFERS - CAPE-COD DATA SET

This study evaluates commonly used geostatistical methods to assess re production of hydraulic conductivity (K) structure and sensitivity und er limiting amounts of data. Extensive conductivity measurements from the Cape Cod sand and gravel aquifer are used to evaluate two geostati stical estimation methods, conditional mean as an estimate and ordinar y kriging, and two stochastic simulation methods, simulated annealing and sequential Gaussian simulation. Our results indicate that for rela tively homogeneous sand and gravel aquifers such as the Cape Cod aquif er, neither estimation methods nor stochastic simulation methods give highly accurate point predictions of hydraulic conductivity despite th e high density bf collected data. Although the stochastic simulation m ethods yielded higher errors than the estimation methods, the stochast ic simulation methods yielded better reproduction of the measured In ( K) distribution and better reproduction of local contrasts in In (K). The inability of kriging to reproduce high In (K) values, as reaffirme d by this study, provides a strong instigation for choosing stochastic simulation methods to generate conductivity fields when performing fi ne-scale contaminant transport modeling. Results also indicate that es timation error is relatively insensitive to the number of hydraulic co nductivity measurements so long as more than a threshold number of dat a are used to condition the realizations. This threshold occurs for th e Cape Cod site when there are approximately three conductivity measur ements per integral volume. The lack of improvement with additional da ta suggests that although fine-scale hydraulic conductivity structure is evident in the variogram, it is not accurately reproduced by geosta tistical estimation methods. If the Cape Cod aquifer spatial conductiv ity characteristics are indicative of other sand and gravel deposits, then the results on predictive error versus data collection obtained h ere have significant practical consequences for site characterization. Heavily sampled sand and gravel aquifers, such as Cape Cod and Borden , may have large amounts of redundant data, while in more common real world settings, our results suggest that denser data collection will l ikely improve understanding of permeability structure.