Input data, trained model data, and model outputs for predicting streamflow and base flow for the Mississippi Embayment Regional Study Area using a random forest model

Publication Date

2022-05-05

Westenbroek, S.M., Dietsch, B.J., and Breaker, B.K., 2022, Input data, trained model data, and model outputs for predicting streamflow and base flow for the Mississippi Embayment Regional Study Area using a random forest model: U.S. Geological Survey data release, https://doi.org/10.5066/P9QCK8HY.

Type: Source Code Type: Related Primary Publication

This data release, when combined with code retrieved from an associated software release, is designed to allow for reasonable monthly total and baseflow volumes to be estimated at arbitrary locations within the Mississippi Embayment as part of the Mississippi Alluvial Plain (MAP) Project. The resulting surface water volume estimates were used as inputs to associated groundwater flow models developed as part of the MAP Project. This data release makes the trained random forest model object, input data sets, output data sets, and source code (via software release, hosted on USGS GitLab: https://code.usgs.gov/map/maprandomforest) available for inspection and use. The data (in this archive) and code (in the GitLab software release) will reproduce about 97% of the records originally generated for use by groundwater flow modelers as part of the Mississippi Alluvial Plain Project. The remaining records that cannot be reproduced using the scripts provided here are associated with large rivers; the values for these rivers were obtained directly from the USGS National Water Information System and are not generated by means of the software release.

ScienceBase WMS

• USGS Data Release Products
• USGS Lower Mississippi-Gulf Water Science Center

Data source

Input directly