Water Resources Research Act Program

Details for Project ID WI_2025_Kalcic

Machine Learning-Guided Assessment of SWAT+ In-Stream Phosphorus Algorithms for Capturing Legacy Phosphorus in the Soils and Streams of Great Lakes Priority Watersheds

Institute: Wisconsin
USGS Grant Number:
Year Established: 2025 Start Date: 2026-08-01 End Date: 2029-07-31
Total Federal Funds: $309,647 Total Non-Federal Funds: $309,648

Principal Investigators: Margaret McCahon Kalcic

Project Summary: This project aims to understand how phosphorus is exported and conveyed from agricultural soils through streams and rivers to western Lake Erie and Green Bay, key Great Lakes embayments where harmful algal blooms are fueled by high concentrations of inflowing phosphorus. We address limitations in the inner workings of the SWAT+ watershed model to more accurately track in-stream phosphorus dynamics. We test these improvements in SWAT+ models derived from the National Agroecosystem Model in data-rich Priority Watersheds. We will use a novel machine learning approach to assess accuracy in model simulation using three in-stream phosphorus routines. Improved prediction of nutrient pollution will allow us to understand the role of legacy phosphorus in soils and streams and aid in the development of more effective nutrient management strategies to protect the Great Lakes and nation for years to come.