Importance of Characterization of Soil Properties and Spatial Distribution in Watershed Modeling.

Simulation of watershed basins is an important field in environmental research, especially for increased understanding of agro-ecosystem land use and practices affecting land surface exchange of mass and energy, nitrogen and carbon cycling, water pollution, and response to climate change. Many of these processes occur within the soil, and simulations need to incorporate both the heterogeneity of the land surface and the complexity of soil properties. We parameterized the spatial distribution of soils in a watershed using pedotransfer functions, and simulated the watershed discharge, infiltration and redistribution with hourly cascade and finite difference methods, which are incorporated into the CropSyst Microbasin Model. The study basin is located in the eastern Palouse region of the Pacific Northwest, a dryland production zone, where saturation-excess runoff is common. Typically, this zone contains silt loam soils in rolling loess-derived hills. The field data were collected from 2011-2014 at twelve locations and at five depths in 30-cm intervals. Sample points were monitored continually (hourly frequency) for the volumetric water content and soil temperature. Soil bulk density (top 1.5 m), clay fraction (top 0.9 m) and restrictive layers were estimated with a VisNIR penetrometer at thirty six locations. Additionally, the water retention curve profile was characterized in a few locations using the evaporation method (HYPROP, UMS GmbH, Munich, Germany), and also the saturated hydraulic conductivity (K_sat) was determined with a KSAT device (UMS GmbH Munich, Germany). We obtained soil maps of the retention curve, K_sat and soil water storage at saturation ( ). The hourly cascade simulation showed good agreement with the measured daily discharge data (Nash-Sutcliffe efficiency = 0.81). We determined that  and the restriction K_sat at particular soil depths most reliably predicted water discharge, suggesting that they are the most important factors to consider in watershed modeling in the eastern Palouse region.