Verifying Curve Numbers in Arid Environments by Combining Detailed Geomorphic Mapping and Pedotransfer Functions

Rainfall-runoff models, such as HEC-1 and HEC-HMS developed by US Army Corps of Engineers' Hydrologic Engineering Center (HEC), are commonly used in the southwest US to estimate flood discharges because most watersheds in this region are ungaged and do not have stream discharge data. Most rainfall-runoff models do not directly account for the initial abstraction (Ia), which is defined as the amount of precipitation that initially infiltrates into the soil prior to the occurrence of runoff. Rather these models rely on precipitation loss components that are considered to be subbasin area averages, such as the Soil Conservation Service (SCS- now Natural Resource Conservation Service) curve number (CN) approach. The purposes of this study were to characterize soils of a remote basin in southern Nevada, specifically runoff potential and soil hydraulic conductivity (Ks), and to examine the use of a site-specific pedotransfer function (PTF) to relate soil texture and bulk density to CNs. Geomorphic mapping, soil sampling and analysis, rainfall simulation, and tension infiltrometer (TI) tests were ail used as field characterization techniques on six distinct geomorphic surfaces determined within the 22 km² study watershed. Rainfall simulation tests allowed for field-measured CNs to be determined for these surfaces. High CN values occurred on the well-developed desert pavement surfaces of the alluvial fan and lower CN values occurred on the younger and dissected alluvial fan deposits. Field measurements from the TI tests showed higher Ks values on the younger and dissected surfaces-, and lower values on the older and well-paved surfaces. Thus, CN inversely corresponded to Ks. The use of detailed geomorphic mapping significantly reduced the variance in Ks (and hence CN, as well) across the watershed, resulting in statistically distinct hydrologic groups that could be scaled to the watershed. When the average Ks was regressed onto field-measured CNs, a linear relationship was found at R² = 0.928, demonstrating that Ks measurements may be used to estimate CNs in this watershed. In addition, a site-specific PTF method showed that soil particle size distributions and bulk density were good predictors of Ks (R² = 0.890). Therefore, at this field site, less arborous soil characterization data and the site-specific relationships obtained by more rigorous field work, CNs and other parameters for this watershed could be easily estimated. Using field-measured site-specific data to accurately assign CNs is important for verifying hydrologic models used for design of flood hazard mitigation structures. These studies suggest that field verification, which in the past was deemed cost-prohibitive in large remote watersheds, could be cost-effective and efficient using a similar approach.