Quantification of Streambed Sediment Phosphorus Storage in the Maumee River Using Machine Learning

This repository contains a Random Forest machine learning model for predicting fine sediment storage and sediment-bound phosphorus concentrations across 60-m stream segments in the Maumee River network. The model uses geospatial predictor variables to estimate fine sediment (less than 63 µm) storage concentrations and calculates associated sediment-bound phosphorus (SedP) storage. Results are aggregated at the HUC12 watershed scale. The archive includes the model source code as a Jupyter Notebook.
This repository supports the following data release:
Mendoza, W., Kreiling, R,M., Williamson, T.N., Fitzpatrick F.A., and Broerman, H.M., 2026, Quantification of Streambed-Sediment Phosphorus Storage in the Maumee River network Using Machine Learning: U.S. Geological Survey data release: U.S. Geological Survey data release, https://doi.org/10.5066/P14M2VJM.