Submitted to: Water Resources Bulletin
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 10/1/1996
Publication Date: N/A
Interpretive Summary: Subsurface water is a major source of water in the United States. In semi-arid and arid areas, it may be the only water source for domestic use and irrigation. In many mountainous watersheds, snowmelt recharge via shallow subsurface flow is the primary source of streamflow, but ground water response to snowmelt is often unappreciated. Snowmelt from deep mountainous snowpacks is seldom rapid enough to produce direct runoff. To adequately understand the hydrologic processes occurring in snowfed mountainous watersheds, the interaction between surface snowmelt and sub-surface ground water flow needs to be addressed. The purpose of the present study was to drive a detailed ground water with simulated snowmelt input to test the viability and applicability of such a coupling for simulating the hydrologic response within a mountainous areas. Simulation results reasonably predicted measured ground water level response to snowmelt in the Upper Sheep Creek Watershed located withing the Reynolds Creek Experimental Watershed. However, difficulty with the numerical solution to the ground water flow equations limited the applicability of the ground water model. Incorporation of a more robust solution procedure will be necessary prior to coupling a snowmelt model with the ground water model or to applying the model to larger areas.
Technical Abstract: Snowmelt from deep mountainous snowpacks is seldom rapid enough to exceed infiltration rates; thus, the source of streamflow in many mountainous watersheds is snowmelt recharge through shallow ground water systems. The hydrologic response and interaction between surface and sub-surface flow processes in these watersheds, which is controlled by basin structure, the spatial distribution of snowmelt and the hydrogeology of the subsurface, are not well understood. The purpose of this study was to test a three-dimensional ground water model using simulated snowmelt input to simulate ground water response to spatially distributed snowmelt on the Upper Sheep Creek Watershed located within the Reynolds Creek Experimental Watershed in Southwestern Idaho. The model was used to characterize the mountainous aquifer and to delineate the subsurface flow mechanisms. Difficulty in finding a reasonable combination of grid spacing and time stepping within the model was encountered due to convergence problems with the Picard solution to the non-linear variably-saturated ground water flow equations. Simulation results indicated that flow may be either unconfined or confined depending on inflow rate and hydrogeologic conditions in the watershed. The flow mechanism had a much faster response time when confined flow occurred. Response to snowmelt from a snow drift approximately 90 m away took only a few hours when flow was confined. Simulated results showed good agreement with piezometer measurements both in magnitude and timing, however convergence problems with the Picard solution limited applicability of the model.