Modeling the Effects of Agricultural Conservation Practices and Climate Variability on Sediment Yield and Transport in the Colusa Basin, California

Agricultural Best Management Practices (BMPs) have been employed for years as erosion reduction measures on cultivated lands. The ability to model the effects of BMPs at the watershed scale using a standardized process-based method provides watershed managers with an important decision making tool for addressing large-scale water quality concerns. This study uses a process-based BMP simulation method with the Soil and Water Assessment Tool (SWAT) to evaluate five BMPs including grassed waterways, channel stabilization structures, strip crops, cover crops and vegetative filter strips on almond orchards in northern California, USA. The sediment reduction rate of the five BMPs was compared to a base case where no BMPs were applied to the watershed and analyzed for several precipitation scenarios to compare BMP effectiveness. In-channel BMPs which included grassed waterways and channel stabilization structures, reduced sediment load at the watershed outlet by 8% to 14%, respectively, depending on the annual precipitation scenario. Grassed waterways consistently outperformed channel stabilization structures, reducing sediment load by an additional 2% to 5% compared to channel stabilization structures. Upland BMPs including strip crops, cover crops and vegetative filter strips reduced sediment yield by 15 to 100% for the various precipitation scenarios. For years with median and above median precipitation, strip crops were most effective, reducing sediment yield by 63% in both cases, whereas cover crops only reduced sediment yield by 54% and 15%, respectively. For the below median precipitation year, the cover crop reduced sediment yield from fields completely (100%), whereas strip crops and vegetative filter strips only reduced sediment load by 64% and 59%, respectively. For all BMPs, a positive correlation between sediment load/yield and increasing precipitation amount and intensity was observed. The methods presented in this study are easily applicable to watershed scale studies of other basins and for other water quality concerns, such as the fate and transport of agricultural pesticides and nutrients.