You are here:
Enhancing Hydrological Modeling of Ungauged Watersheds through Machine Learning and Physical Similarity-based Regionalization of Calibration Parameters
Citation:
Bawa, A., K. Mendoza, R. Srinivasan, F. O'Donncha, D. Smith, K. Wolfe, R. Parmar, J. Johnston, AND J. Corona. Enhancing Hydrological Modeling of Ungauged Watersheds through Machine Learning and Physical Similarity-based Regionalization of Calibration Parameters. ENVIRONMENTAL MODELLING & SOFTWARE. Elsevier Science, New York, NY, 186:106335, (2025). https://doi.org/10.1016/j.envsoft.2025.106335
Impact/Purpose:
With increasing climate variability and more frequent extreme weather events, there is a growing demand for accurate hydrological simulations to support decision-making in water resource management. Hydrological modeling is an essential tool for comprehending hydrological processes and managing water resources. These models have been widely applied across various scales, from small local catchments to large river basins, to address research questions and support decision-making processes. This study proposed using physical similarity-based clustering and random forest techniques for parameter regionalization of the physical, process-based SWAT hydrological model.
Description:
This study enhances hydrological modeling in ungauged watersheds by employing physical similarity and machine learning-based clustering for regionalizing the Soil and Water Assessment Tool (SWAT) model parameters at the HUC12 (hydrological unit code) watershed scale within a HUC02 basin. Eleven features, including environmental, topographical, soil, and hydrological properties, were utilized to identify physical similarities for watershed clustering. Machine learning techniques, including random forest and hierarchical clustering, were employed to transfer calibrated parameters from gauged to ungauged watersheds. Validation of parameter transfer over gauged SWAT model projects showed that 88% of the projects achieved calibrated status (KGE ≥ 0.5; PBIAS ≤ 25%). Additional validation using MODIS satellite evapotranspiration measurements confirmed the robustness of the approach. Results indicated that the proposed approach successfully captures physical similarities, and effectively captures flow patterns. Overall, the study highlights the potential of physical similarity-based clustering and machine learning techniques for improving hydrological modeling in ungauged watersheds.
URLs/Downloads:
DOI: Enhancing Hydrological Modeling of Ungauged Watersheds through Machine Learning and Physical Similarity-based Regionalization of Calibration Parameters