Irina Forkutsa1, Molly Gribb2, Jirka Simunek3, Jim McNamara1, and David Chandler4. (1) Geosciences, Boise State University, 1910 University Drive, Boise, ID 83725-1535, (2) 1910 University Dr., Boise State University, Boise State University, Center of Environmental Sensing, Boise, ID 83725-2075, (3) Environmental Sciences, University of California-Riverside, Bourns Hall A135, Riverside, CA 92521, (4) Civil Engineering, Kansas State University, 2109 Fiedler Hall, Manhattan, KS 66506-5000
Groundwater recharge in mountain blocks is controlled by infiltration of water through the shallow soils overlying the bedrock, which is difficult to measure. The hydraulic properties of these shallow soils (<90 cm depth) are critical controls on the timing and magnitude of groundwater recharge in the mountain block.
A field and modeling study to investigate the spatial and temporal variability of soil hydraulic properties was conducted in a small catchment with an ephemeral stream draining 0.02 km2 in a range of hills near Boise (Idaho, USA), known as the Boise Front. Soils are formed from weathering of the granitic bedrock and are classified as coarse-loamy, mixed mesic Ultic Haploxerolls. The soil hydraulic properties were obtained from four estimates: (1) Field water content and tensiometer data fitted with RETC; (2) Multistep outflow tests fitted with RETC; (3) Rosetta (Schaap, 1999); (4) Pedotransfer function that Schaap (2006) created for the Hanford reservation soils (Ye et al., 2006), and subsequently, calibrated with HYDRUS-1D model (Simunek et al., 2005) for two rain events in 2003 at two depths of a single soil profile. Use the fitted field data, Rosetta and Hanford did lead to the relatively good water content prediction, especially for the deep layer since it was highly driven by the selected “variable pressure head” bottom boundary conditions. According to simulation results, there is no difference which of the four estimates to use in the inverse simulations, because all parameter sets, when used as initial estimates, result in a similar prediction of water content and close RMSEs. Soil water fluxes for both layers were calculated. To test the usefulness of the inversely estimated effective soil hydraulic properties, four more soil profiles at research site were selected for model simulations. Also, simulations were completed for 2003 year.
