The data are comprised of input and output data from Machine Learning models that were developed to predict watershed health (WH) values in HUC-10 sub-watersheds within three major Midwest river basins. The input data included timeseries of hydro-meteorological and reconstructed WQ parameters (sediment, nitrogen, and phosphorus) as well as GIS shape files of watershed attributes (soil, landcover/land use, geomorphology, drainage classes, fertilizer sale data, etc. ). The output data is ensemble-model estimated annual WH values in HUC-10 sub-watersheds within the three river basins. The ensemble-model predicted WH values are derived from WH values obtained from three trained and validated machine learning models. This dataset is associated with the following publication: Mallya, G., M.M. Hantush, and R.S. Govindaraju. A Machine Learning Approach to Predict Watershed Health Indices for Sediments and Nutrients at Ungauged Basins. WATER. MDPI, Basel, SWITZERLAND, 15(3): 586, (2023).

The data are: 1) compilation of field observed nutrient and hydrometeorological data for the Upper Fish River Watershed (UFRW); 2) wetland and GIS data downloaded from national repositories for UFRW; 3) wetland nutrient data generated by the models for the UFRW; 4) output data (nutrient loads and removal rates) produced by the SWAT-WetQual (watershed-wetland) model framework for the UFRW; 5) global wetland nutrient function data obtained from literature; and 6) model data used in developing statistical regression relationships for nutrient removal rates and efficiencies. Nutrients: Nitrate and Orthophosphate.

The files and data included in this archive allow readers to inspect and reproduce the model results reported in Neff et al. (2020). Please refer to the included ReadMe file for a further explanation of individual files and step-by-step instructions for running the models.

The dataset supported findings in the study: "Evaluation of SWAT reservoir, ponds, and wetlands tools in water and sediment simulation in the Rock River watershed". Results of this study demonstrate the impact of impoundments in SWAT modeling.The dataset includes sources of the SWAT input data. This dataset is associated with the following publication: Jalowska, A., and Y. Yuan. Evaluation of SWAT Impoundment Modeling Methods in Water and Sediment Simulations. JOURNAL OF THE AMERICAN WATER RESOURCES ASSOCIATION. American Water Resources Association, Middleburg, VA, USA, 55(1): 209-227, (2019).

This dataset includes input and output data from streamflow and water-quality regression models used to characterize streamflow and water-quality conditions in the Upper White River Basin, Colorado, from 2000 to 2020. All input data, including discrete and continuous streamflow records and discrete concentrations of inorganic nitrogen, total nitrogen, and total phosphorus, were compiled from the U.S. Geological Survey (USGS) National Water Information System (NWIS) database. Input data were used in multiple models including Maintenance of Variance Extension Type 2 (MOVE.2) and Weighted Regressions on Time, Discharge, and Season (WRTDS) to estimate continuous streamflow records, daily concentrations and loads, and streamflow-normalized annual mean concentrations and loads of selected water-quality constituents.

This dataset includes input and output data from streamflow and water-quality regression models used to characterize streamflow and water-quality conditions in the Upper White River Basin, Colorado, from 2000 to 2020. All input data, including discrete and continuous streamflow records and discrete concentrations of inorganic nitrogen, total nitrogen, and total phosphorus, were compiled from the U.S. Geological Survey (USGS) National Water Information System (NWIS) database. Input data were used in multiple models including Maintenance of Variance Extension Type 2 (MOVE.2) and Weighted Regressions on Time, Discharge, and Season (WRTDS) to estimate continuous streamflow records, daily concentrations and loads, and streamflow-normalized annual mean concentrations and loads of selected water-quality constituents.

This U.S. Geological Survey (USGS) data release contains model scenario input and output files and nine hydrology and water-quality models developed using the Soil and Water Assessment Tool (SWAT). The models represent nine watersheds in eastern New York that drain to Lake Erie or the Niagara River and were created, calibrated, and validated for hydrology, sediment, and nutrients as baseline scenarios. Twenty-six additional scenarios were created to explore the effects of agricultural and urban best management practices, point source discharges, and green infrastructure on the water quality of tributaries to Lake Erie/Niagara River. Model documentation and scenario development are described in Merriman and others (2024).

This section of the data release supports an archive of the models used in the associated publication. The U.S. Geological Survey and the University of Wisconsin – Green Bay collected hydrologic and water-quality data to assess the effectiveness of agricultural conservation management practice (CMP) implementation at Mainstem Plum Creek and West Plum Creek in northeastern Wisconsin. Monitoring data from 2010–2020 at Mainstem Plum and 2013–2020 at West Plum were used to detect changes in hydrologic and water-quality responses during runoff events. Runoff events were defined by hydrographers and used to compute event loads and event flow-weighted mean concentrations of total phosphorus and total suspended solids – all of which are included in the associated data release. Additionally, changes in these parameters were assessed between two time periods (“initial” and “post-CMP implementation”) using the models included in this model archive. Because event discharges, loads, and concentrations are influenced by factors such as weather and the conditions preceding events, random-forest and regression models were developed to control for these factors and to elucidate water-quality changes more directly associated with CMP implementation. Residuals from random-forest models were used to detect changes between the two time periods via Wilcoxon signed-rank tests, and multiple linear regression models were used to determine percent change in responses via time-period dummy variable coefficients. Results indicate statistically insignificant changes in most responses during runoff events.

This section of the data release supports an archive of the models used in the associated publication. The U.S. Geological Survey and the University of Wisconsin – Green Bay collected hydrologic and water-quality data to assess the effectiveness of agricultural conservation management practice (CMP) implementation at Mainstem Plum Creek and West Plum Creek in northeastern Wisconsin. Monitoring data from 2010–2020 at Mainstem Plum and 2013–2020 at West Plum were used to detect changes in hydrologic and water-quality responses during runoff events. Runoff events were defined by hydrographers and used to compute event loads and event flow-weighted mean concentrations of total phosphorus and total suspended solids – all of which are included in the associated data release. Additionally, changes in these parameters were assessed between two time periods (“initial” and “post-CMP implementation”) using the models included in this model archive. Because event discharges, loads, and concentrations are influenced by factors such as weather and the conditions preceding events, random-forest and regression models were developed to control for these factors and to elucidate water-quality changes more directly associated with CMP implementation. Residuals from random-forest models were used to detect changes between the two time periods via Wilcoxon signed-rank tests, and multiple linear regression models were used to determine percent change in responses via time-period dummy variable coefficients. Results indicate statistically insignificant changes in most responses during runoff events.

This data release contains inputs for and outputs from hydrologic simulations of the southeastern U.S. using the Monthly Water Balance Model, the Precipitation Runoff Modeling System (PRMS), and statistically-based methods. These simulations were developed to provide estimates of water availability and statistics of streamflow for historical and potential future conditions for an area of approximately 1.16 million square miles. These model input and output data are intended to accompany a U.S. Geological Survey Scientific Investigations Report (LaFontaine and others, 2019); they include four types of data: 1) model input parameters, 2) model output statistics, 3) GIS files of the model hydrologic response units and stream segments, and 4) statistically-based streamflow estimates for headwater watersheds. LaFontaine, J.H., Hart, R.M., Hay, L.E., Farmer, W.H., Bock, A.R., Viger, R.J., Markstrom, S.L., Regan, R.S., and Driscoll, J.M., 2019, Simulation of Water Availability in the Southeastern United States for Historical and Potential Future Climate and Land-Cover Conditions: U.S. Geological Survey Scientific Investigations Report, 2019-5039, 83 p., https://doi.org/10.3133/sir20195039.