Total phosphorus and suspended-sediment concentrations and loads from two main tributaries to Upper Klamath Lake, Oregon, 2014–20

Abstract

Total phosphorus (TP) and suspended-sediment concentrations (SSC) and loads were computed at two U.S. Geological Survey (USGS) streamgages in the upper Klamath River Basin on the Sprague (USGS site ID 11501000) and Williamson (USGS site ID 11502500) Rivers using high temporal resolution turbidity and streamflow data to develop surrogate regression models. Regression models were updated and validated for TP at the Williamson River site, and additional data improved a prior published TP model, increasing the coefficient of determination (R 2 ) from 0.73 to 0.88. A new TP regression model was developed for the Sprague River site using 2 years of data and showed promising results with an R 2 of 0.93. Suspended-sediment concentration (SSC) surrogate models were also updated at these sites using a longer period of record than the TP models and improved characterization of sediment transport conditions at these monitoring sites. Computations of TP loads were compared to the annual loading capacity dictated by the total maximum daily load (TMDL) for Upper Klamath Lake and showed that the combined TP load of the Williamson and Sprague Rivers approaches the annual loading capacity in water years with high annual streamflow. TP loads were also compared to loads computed by the Klamath Tribes using a long-term dataset and a regression and interpolation algorithm (RIA). The comparison showed that the two methods report similar annual loads, with the surrogate regression method generally reporting lower loads than the RIA, and the RIA annual loads falling within the range of uncertainty of the surrogate regression model results. Determining the effect of habitat and stream restoration on basin-scale TP and suspended-sediment loading is challenging using the surrogate regression method at these sites given the short period of record that TP and suspended-sediment load (SSL) data are available. However, long-term analysis by the Klamath Tribes in their larger monitoring network could provide insight into the impact of restoration at smaller spatial scales compared to the basin-wide assessment produced in this study.