Paper No. 7
Presentation Time: 10:05 AM


NIMMO, John R., U.S. Geological Survey, 345 Middlefield Road, Menlo Park, CA 94025, MIRUS, Benjamin B., U.S. Geological Survey, Geologic Hazards Sciences Center, Golden, CO 80401 and PERKINS, Kim S., U.S. Geological Survey, 345 Middlefield Rd, MS-420, Menlo Park, CA 94025,

The Idaho National Laboratory (INL) is an important example of the many sites worldwide where the dynamic behavior of subsurface water is a major control on the spread of environmental contamination. Preferential flow complicates this problem and poses a challenge because of our incomplete understanding of the process and its potential to enhance the speed and severity of contaminant spread. The unsaturated zone at this semiarid location, typically deeper than 100 m, comprises fractured basaltic rock interbedded with layers of sediment. Perched water occurs commonly with material contrasts, further complicating the possible contaminant transport modes and increasing the difficulty of characterization.

A newly developed preferential flow model, called source-responsive because water throughout the unsaturated zone can respond sensitively to changing water-input conditions, is useful for these flow complexities. It requires two characterizations: internal macropore facial area as a function of depth, representing a capacity for preferential flow; and an active-area fraction, indicating how much of that capacity is active at given depth and time.

We have investigated several preferential flow problems at the INL, including: perched water-level dynamics in response to ephemeral streamflow; irregular water content increases generated by ponded infiltration, measured through 50 m of fractured basalt; and various subsurface responses discerned from piezometer, tensiometer, and temperature data. Model results suggest that source-responsive flow through a limited number of connected fractures contributes substantially to the perched-zone dynamics. Values of the source-responsive properties estimated through inverse calculations suggest when and where the consequences of preferential flow are likely to be most pronounced. The model and its results are useful for predicting rapid water table responses, nonsequential increases of water content, and travel times and fluxes relating to contaminant-transport vulnerabilities.