URBANA, Ill., Mar 19: A team of international researchers has developed a novel method to enhance the accuracy of hydrological models by addressing uncertainties in precipitation data, a major factor in water resource planning and management.
Hydrological models, which simulate water movement in natural systems, are highly sensitive to the quality of rainfall data.
“Precipitation is very variable in space and time,” said Jorge Guzman, research assistant professor in the Department of Agricultural and Biological Engineering at the University of Illinois Urbana-Champaign. “If you enter rainfall as a single value, it can distort the model’s representation and affect predictions for water yield, sediment production, and flood management.”
The study, conducted in collaboration with researchers from the Universidad Industrial de Santander, Colombia, and the Federal University of Lavras, Brazil, focused on improving model calibration for watersheds with limited rainfall measurement infrastructure. The team developed a stepwise back-correction algorithm that adjusts precipitation inputs to better reflect actual rainfall patterns.
The algorithm was tested in the Sangamon River watershed in Illinois and in the Grande River and Jequitinhonha River watersheds in Brazil. Researchers applied it to three widely used hydrological models: the Soil and Water Assessment Tool (SWAT), MIKE-SHE, and Distributed Hydrological Model (MHD). Results showed improved model performance across all platforms, with SWAT achieving up to 18% higher accuracy than traditional approaches.
“This research addresses a common limitation in hydrological modeling by integrating parameter calibration with dynamic precipitation correction, producing significant improvements in model performance,” said Dany Hernandez, co-author from Colombia.
The back-correction tool is now freely available for other researchers, along with software and application instructions, promoting broader adoption in hydrology and environmental management studies.
The findings are published in Environmental Modelling & Software in the paper, “A stepwise back-correction function for precipitation representation in hydrologic models” [DOI: 10.1016/j.envsoft.2026.106908].
Funding for the research was provided by the USDA National Institute of Food and Agriculture Hatch Program, the College of ACES Office of International Programs, and the Grupo de Investigación en Recursos Hídricos y Saneamiento Ambiental at Universidad Industrial de Santander, Colombia.
