H13L-05:
Use of Weighted Regressions on Time, Discharge, and Season to Assess Effectiveness of Agricultural and Environmental Best Management Practices in California and Nevada, USA

Monday, 15 December 2014: 2:40 PM
Joseph L Domagalski1, Brandon Schlegel2 and Juliet Hutchins2, (1)USGS, Sacramento, CA, United States, (2)California State University, Sacramento, Department of Geology, Sacramento, CA, United States
Abstract:
Long-term data sets on stream-water quality and discharge can be used to assess whether best management practices (BMPs) are restoring beneficial uses of impaired water as required under the Clean Water Act. In this study, we evaluated a greater than 20-year record of water quality from selected streams in the Central Valley (CV) of California and Lake Tahoe (California and Nevada, USA). The CV contains a mix of agricultural and urbanized land, while the Lake Tahoe area is mostly forested, with seasonal residents and tourism. Because nutrients and fine sediments cause a reduction in water clarity that impair Lake Tahoe, BMPs were implemented in the early 1990’s, to reduce nitrogen and phosphorus loads. The CV does not have a current nutrient management plan, but numerous BMPs exist to reduce pesticide loads, and it was hypothesized that these programs could also reduce nutrient levels. In the CV and Lake Tahoe areas, nutrient concentrations, loads, and trends were estimated by using the recently developed Weighted Regressions on Time, Discharge, and Season (WRTDS) model. Sufficient data were available to compare trends during a voluntary and enforcement period for seven CV sites within the lower Sacramento and San Joaquin Basins. For six of the seven sites, flow-normalized mean annual concentrations of total phosphorus and nitrate decreased at a faster rate during the enforcement period than during the earlier voluntary period. Concentration changes during similar years and ranges of flow conditions suggest that BMPs designed for pesticides also reduced nutrient loads in the CV. A trend analysis using WRTDS was completed for six streams that enter Lake Tahoe during the late 1980’s through 2008. The results of the model confirm that nutrient loading is influenced strongly by season, such as by spring runoff from snowmelt. The highest nutrient concentrations in the late 1980’s and early 1990’s correlate with high flows, followed by statistically significant decreases in loading from most streams under different flow conditions. The results of the WRTDS model indicate a clear reduction in nutrient loading of nitrogen and phosphorus in all six streams. However, some streams show an increase in nutrient concentrations after 2000, suggesting the possible need for changes to the nutrient reduction management practices.