2002 Denver Annual Meeting (October 27-30, 2002)

Paper No. 5
Presentation Time: 2:30 PM


HEVESI, Joseph A.1, FLINT, Alan L.2, FLINT, Lorraine E.1, BEDFORD, David R.3 and MILLER, David M.4, (1)U.S. Geological Survey, Placer Hall, 6000 J Street, Sacramento, CA 95819, (2)U.S. Geol Survey, Placer Hall, 6000 J Street, Sacramento, CA 95819, (3)U.S. Geological Survey, 345 Middlefield Road, MS 975, Menlo Park, CA 94025, (4)U.S. Geol Survey, 345 Middlefield Road, MS 975, Menlo Park, CA 94025, jhevesi@usgs.gov

A numerical representation of the hydrogeologic framework of the upper unsaturated zone (UUZ) is required for a deterministic watershed model used in the Mojave Desert region to analyze basin-scale water budgets, estimate spatially and temporally distributed soil moisture, predict the timing and volume of runoff, and help define the recharge boundary conditions for groundwater flow models. This watershed model represents both local- and regional-scale hydrologic processes and is useful for estimating the hydrologic response of arid and semi-arid basins to monthly and daily climate conditions. Model design depends on a conceptualized representation of the UUZ (from ground surface to depths of 6 to 10 meters), and the resolution of the data available to define model parameters. The purpose of this study is to evaluate model sensitivity and uncertainty to various representations of the UUZ, based on differences in model discretization and spatially distributed soil and bedrock hydrologic properties. Spatial discretization is dependant on available digital elevation models. Spatially distributed basin characteristics include topography, bedrock properties, soil properties, and vegetation properties. Vertical discretization, used to represent heterogeneity in both root zone characteristics and the hydrogeologic framework of the UUZ, is developed using estimates of soil thickness, soil layers, soil properties, and root density. Daily climate records are included as model input to simulate time series and spatial distributions of soil moisture, evapotranspiration, runoff, streamflow, and net infiltration. Results indicate that estimates of the spatial and temporal distribution of soil moisture, wetting fronts, and net infiltration on alluvial fans and basins, including stream channels and mountain front areas, may require a more detailed representation of the UUZ than is currently available using data such as the State Soil Geographic database. For example, net infiltration along alluvial fans can be sensitive to the spatial delineation of impermeable soil layers and variations in soil properties across channels. Detailed representation of the UUZ for upland areas, which tends to be the source of most of the recharge and runoff generation, is not as important due to thin soils and shallower root zones.