Considerable effort has been devoted worldwide to characterizing aquifer heterogeneity and its transport and flow effects, although description of this heterogeneity in terms of sedimentary facies has been less common. Detailed observations of the scale and directional dependence of heterogeneity in sedimentary aquifers have been made for fluvial settings [Davis et al., 1993], and alluvial fan settings [Neton et al., 1994]. Both studies use the concept of architectural elements [Miall, 1985], which are facies-based subdivisions of sedimentary deposits. Other studies (e.g. [Poeter and Gaylord, 1990]) have noted correlations between lithostratigraphy and transport behavior. More recently, models and observations of facies (architectural elements) have noted strong influence of connectivity of elements on contaminant migration. For example, strongly advective transport in interconnected channel facies combined with diffusive storage of contaminant in surrounding overbank facies has been suggested to be the reason for failure of many remediation efforts in alluvial fans [Fogg and Carle, 1996].
Much of this work is based on earlier developments in petroleum geology. Miall was the principal developer of the architectural element concept [Miall, 1985,Miall, 1991]. The need for such an approach had become clear from earlier observation of heterogeneity effects on wellfield productivity [Hartman and Paynter, 1979,Weber, 1982]. Sophisticated modeling techniques have been developed based on statistical characterization of architectural elements and determination of ensemble-average reservoir properties based on stochastic models of element distribution [Lake and Jr., 1986,Lake et al., 1991]. These are now standard tools in petroleum reservoir analysis, particularly in mature oilfields.