US Department of Energy

 

Date of this Version

2003

Citation

Journal of Hydrology 275 (2003) 141–161; doi:10.1016/S0022-1694(03)00039-8

Abstract

The processes governing physical nonequilibrium (PNE)—coupled preferential flow and matrix diffusion—are diverse between humid and semi-arid regions, and are directly related to climate and rock/sediment type, and indirectly related to subsequent soil profile development. The fate and transport of contaminants in these variably saturated undisturbed media is largely a function of the influence of PNE processes. Large cores of laminated silts and sands were collected from the US Department of Energy Pacific Northwest National Laboratory (PNNL) in semi-arid south central Washington. Additional cores of weathered, fractured interbedded limestone and shale saprolite were collected from the Oak Ridge National Laboratory (ORNL) in humid eastern Tennessee. PNNL cores were collected parallel (FBP) and perpendicular (FXB) to bedding, and the ORNL core was 30° to bedding. Saturated and unsaturated transport experiments were performed using multiple nonreactive tracers that had different diffusion coefficients (Br, PFBA, and PIPES), in order to identify the influence of PNE on the fate and transport of solutes. In the ORNL structured saprolite, solute transport was governed by coupled preferential flow and matrix diffusion, as evidenced by tracer separation and highly asymmetric breakthrough curves (BTC). BTCs became more symmetric as preferential flowpaths became inactive during drainage. Tracer separation persisted during unsaturated flow suggesting the continued importance of nonequilibrium mass transfer between flowpaths and the immobile water that was held in the soil matrix. No evidence of PNE was observed under near-saturated conditions in the semi-arid region (PNNL) laminated silts and sands. Unsaturated flow in cores with discontinuous layering resulted in preferential flow and the development of perched, immobile water as evidenced by early breakthrough and separation of tracers. Conversely, transport parallel to laterally continuous beds did not result in preferential flow, the development of perched water, or tracer separation regardless of water content. These observations suggested that desaturation had two effects: (1) grain size variations between individual beds resulted in different antecedent water contents, and (2) the exchange of water and solutes between individual sedimentary beds was subsequently inhibited. Under unsaturated conditions, these effects may promote either stable lateral flow, or unstable vertical finger flow coupled with the development of perched, immobile water.