This data release contains model inputs, outputs, and source code (written in R) for the calculation of daily total phosphorus (TP) concentrations at a sampling site operated by the Klamath Tribes (WR6000), which is located 5 miles downstream of USGS streamgage 11502500, Williamson River below Sprague River near Chiloquin, Oregon. Daily streamflow data from 11502500 were used to calibrate the model along with data provided for the Klamath Tribes for site WR6000. Daily TP concentrations were calculated from 1991-2019 using the USGS Weighted Regression on Time, Discharge, and Season (WRTDS) model. Calibration data are bi-weekly (every two weeks) TP concentrations collected at site WR6000 and analyzed by the Klamath Tribes at the Sprague River Water Quality Laboratory (SRWQL). All concentrations are reported in milligrams per liter (mg/L).

In this data release, we provide the data used to compute total annual phosphorus loads from tributaries and direct drainages to the Great Lakes during 2012-18 using the model load ratio approach and the unit area load approach and the resulting annual loads for 2012–18. This data release consists of: (1) measured loads at 24 sites, computed using the surrogate regression approach (Robertson et al., 2018; Koltun, 2020) that were monitored as part of the Great Lakes Restoration Initiative project, (2) estimated annual loads at point source facilities throughout the Great Lakes Basin obtained from the U.S. Environmental Protection Agency and state agencies, (3) loads subdivided into nonpoint and point source contributions, (4) extrapolation factors for each basin (nonpoint model load ratios, nonpoint yields, point source delivery factors), and (5) estimated annual loads for each tributary and direct drainage area to the Great Lakes during 2012–18. This data release consists of three tab delimited files: (1) ReferenceSite.txt, which contains all of the information (site data and phosphorus loads from SPARROW (Robertson and Saad, 2019) and measured during 2012–18 for the 24 reference (extrapolation sites) used in this study, (2) PointSource.txt, which contains all of the measured or extrapolated point source information for wastewater treatment plants in the Great Lake Basin for 2012–18; and (3) ExtrapolatedLoads.txt, which contains all of the nonpoint model load ratios, point source delivery ratios, measured wastewater treatment plant loads, and resulting annual loads for each tributary and direct drainage area in the Great Lakes Basin for the long-term average and 2012–18.

In this data release, we provide the data used to compute total annual phosphorus loads from tributaries and direct drainages to the Great Lakes during 2012-18 using the model load ratio approach and the unit area load approach and the resulting annual loads for 2012–18. This data release consists of: (1) measured loads at 24 sites, computed using the surrogate regression approach (Robertson et al., 2018; Koltun, 2020) that were monitored as part of the Great Lakes Restoration Initiative project, (2) estimated annual loads at point source facilities throughout the Great Lakes Basin obtained from the U.S. Environmental Protection Agency and state agencies, (3) loads subdivided into nonpoint and point source contributions, (4) extrapolation factors for each basin (nonpoint model load ratios, nonpoint yields, point source delivery factors), and (5) estimated annual loads for each tributary and direct drainage area to the Great Lakes during 2012–18. This data release consists of three tab delimited files: (1) ReferenceSite.txt, which contains all of the information (site data and phosphorus loads from SPARROW (Robertson and Saad, 2019) and measured during 2012–18 for the 24 reference (extrapolation sites) used in this study, (2) PointSource.txt, which contains all of the measured or extrapolated point source information for wastewater treatment plants in the Great Lake Basin for 2012–18; and (3) ExtrapolatedLoads.txt, which contains all of the nonpoint model load ratios, point source delivery ratios, measured wastewater treatment plant loads, and resulting annual loads for each tributary and direct drainage area in the Great Lakes Basin for the long-term average and 2012–18.

This data release contains the R source code, inputs, and selected outputs associated with the turbidity-based regression models used to estimate total phosphorus concentrations and loads for the Missouri River at St. Joseph and Hermann, Missouri, for water years 2008–22. Additionally, the R source code, inputs, and selected outputs for LOADEST models used to estimate daily total phosphorus loads for the study period are included, as these loads were used for days when turbidity data was unavailable.

This data release contains the R source code, inputs, and selected outputs associated with the turbidity-based regression models used to estimate total phosphorus concentrations and loads for the Missouri River at St. Joseph and Hermann, Missouri, for water years 2008–22. Additionally, the R source code, inputs, and selected outputs for LOADEST models used to estimate daily total phosphorus loads for the study period are included, as these loads were used for days when turbidity data was unavailable.

The Archive.zip data set contains R source code and inputs and selected outputs associated with regression analyses used to estimate total phosphorus loads for selected sites on Great Lakes tributaries for water years 2014–2018

The Archive.zip data set contains R source code and inputs and selected outputs associated with regression analyses used to estimate total phosphorus loads for selected sites on Great Lakes tributaries for water years 2014–2018

Three water-quality models were generated to estimate total phosphorus (TP) concentrations and loads from 2002 – 2021 in the Boise River near Parma, Idaho. A Weighted Regression on Time Discharge and Season (WRTDS) model and a Weighted Regression on Time Discharge and Season with Kalman filtering (WRTDS-K) model were generated using all observations within the study period to quantify rate of change in TP concentrations and loads. Further, these two models in addition to the Load ESTimator (LOADEST) model were generated using a subset of the observation data and evaluated on the observations withheld from model generation. The results of the models generated with all data are provided in one file ("full_mod_results.csv"), and the results of the models generated with a subset of the data are provided in a second file ("monthly_model_results.csv"). Mean daily discharge and observed total phosphorus are included in each file for comparison against model output.

Three water-quality models were generated to estimate total phosphorus (TP) concentrations and loads from 2002 – 2021 in the Boise River near Parma, Idaho. A Weighted Regression on Time Discharge and Season (WRTDS) model and a Weighted Regression on Time Discharge and Season with Kalman filtering (WRTDS-K) model were generated using all observations within the study period to quantify rate of change in TP concentrations and loads. Further, these two models in addition to the Load ESTimator (LOADEST) model were generated using a subset of the observation data and evaluated on the observations withheld from model generation. The results of the models generated with all data are provided in one file ("full_mod_results.csv"), and the results of the models generated with a subset of the data are provided in a second file ("monthly_model_results.csv"). Mean daily discharge and observed total phosphorus are included in each file for comparison against model output.

The dataset contains estimates of annual flow and annual flux in kilograms for six sites in the Red River of the North Basin (identified as basin 1, 2, 3, 4, 5, and 8). The shape files that are part of this data release document the basins and the streamgage locations are also provided in an associated shape file.