Catchments And Variables Used For Random Forest Classification And Regression Groundwater And Surface Water Models For Nitrate Violation Classes, Violation Concentrations, Or Percent Of Systems In Violation For Public Drinking Water Supplies

• This research used public drinking water nitrate violations data from across the conterminous United States as response variables, and various predictor variables from EPA's StreamCat dataset and other various land use, climate, and N input datasources.
• All predictor and response variables were summarized for each catchment.
• Response variables were either a binary variable (in violation or not or a 1 or 0), mean violation concentration (mg/L) per catchment, or percent of public water systems in violation per catchment.

This dataset is associated with the following publication:
Pennino, M., S. Leibowitz, J. Compton, R. Hill, and R. Sabo. Patterns and predictions of drinking water nitrate violations across the conterminous United States. SCIENCE OF THE TOTAL ENVIRONMENT. Elsevier BV, AMSTERDAM, NETHERLANDS, 722: 137661, (2020).
Organization: U.S. Environmental Protection Agency
Organization URL: https://catalog.data.gov/organization/epa
Last updated: 2019-05-10
Tags: contiguous United States, drinking water, groundwater, machine learning, nitrate, random forest, surface water