Scripts for projecting stream water quality using Weighted Regression on Time Discharge and Season (WRTDS) in the Delaware River Basin
This data release provides R scripts for estimating future stream water quality using Weighted Discharge on Time Discharge and Season for Projection (WRTDS-P). The technique is tested against hold-out data and is applied to four drought scenarios in the Delaware River Basin (DRB). The scripts retrieve WRTDS models from an existing data release for 39 sites in the DRB with observations spanning from 1978 to 2018, run training and testing evaluations of model performance, and run scenarios of drought effects on water quality. Training and testing were conducted for specific conductance, nitrate, calcium, and magnesium (29 sites for specific conductance and nitrate and 22 sites for calcium and magnesium). Datasets were split, with the final 1 to 5 years serving as holdout data for testing, and the remainder serving as training data. Four drought scenarios were used to evaluate the effects of a range of discharge reductions on the specific conductance at 38 sites in the DRB. For each drought scenario, the final 4 years of the discharge record were repeated with discharge being reduced by 25%, 50%, 75%, or the estimated reduction during the 1965 drought. Outputs of the model evaluation and scenarios are summarized in text files and in an R-markdown file for each site, constituent, and evaluation case.
Citation Information
| Publication Year | 2025 |
|---|---|
| Title | Scripts for projecting stream water quality using Weighted Regression on Time Discharge and Season (WRTDS) in the Delaware River Basin |
| DOI | 10.5066/P14MG3EF |
| Authors | Christopher T Green, Robert M Hirsch |
| Product Type | Data Release |
| Record Source | USGS Asset Identifier Service (AIS) |
| USGS Organization | Water Resources Mission Area - Headquarters |
| Rights | This work is marked with CC0 1.0 Universal |