SWAT-Based Estimation of Curve Numbers for Runoff-Potential Zoning

Accurate estimation of runoff potential is essential for watershed planning, flood assessment, and sustainable land and water management. Although the Curve Number (CN) method is widely used for surface runoff estimation, most existing studies still depend on static LULC–soil matrices or empirical CN values that overlook spatial heterogeneity and hydrological variability at the watershed scale. This introduces a methodological gap in deriving dynamically representative CN estimates that capture actual catchment responses. In this study, the Soil and Water Assessment Tool (SWAT) was employed to generate spatially explicit CN values across a diverse set of Hydrologic Response Units (HRUs) by integrating land use/land cover, soil hydrological groups, and slope classes. The analysis was conducted for the Ong watershed, an important tributary of the Mahanadi River basin in eastern central India, covering an area of 4,650 km². The model-simulated CN values were subsequently utilized to delineate runoff-potential zones within the watershed. Calibration and validation of SWAT-simulated runoff against observed streamflow strengthened the reliability of surface runoff parameterization. The spatial assessment revealed distinct patterns of low, moderate, and high runoff potential, predominantly governed by variations in LULC and soil texture. Built-up and agricultural areas exhibited higher CN values, while forested and permeable zones consistently showed lower runoff potential. Overall, the results demonstrate that SWAT-based CN derivation overcomes the limitations of conventional CN assignment by producing hydrologically consistent and spatially distributed runoff-potential maps. This systematic and scalable framework can support improved conservation planning, watershed prioritization, and climate-stress resilience assessments.

Keywords: Hydrological modeling; Curve Number; Hydrological Responses Unit; Runoff potential; Watershed management