Water Resources Research Act Program

Details for Project ID ND_2023_Chu_1

Enhancing Machine Learning Methods for Rainfall-Runoff Forecasting

Institute: North Dakota
Year Established: 2023 Start Date: 2023-09-01 End Date: 2024-08-31
Total Federal Funds: $4,653 Total Non-Federal Funds: $4,694

Principal Investigators: Xuefeng Chu

Project Summary: Rainfall and runoff are essential hydrologic processes. Researchers strive to create precise models to capture their dynamic relationships and control factors. Generally, the quality and complexity of such models are influenced by a vast number of parameters. Nowadays, many efforts have been made to utilize data-driven models for hydrologic predictions. However, data-driven models often suffer from certain critical issues and limitations. Consequently, the development of an accurate rainfall-runoff model using the emerging techniques can benefit hydrologic forecasting and water resources management. The goal of this endeavor is to develop a new data-driven model that integrates machine-learning and soft-computing techniques to improve the accuracy and efficacy of rainfall-runoff prediction. The Turtle River watershed in North Dakota will be selected to demonstrate the applicability of the new modeling methods. This study will provide a novel approach for improved hydrologic modeling, which can be further utilized to address flood and drought issues.