Improving Flow Estimation Accuracy Through the Integration of Hydrological Methods and Remote Sensing Data: Emphasizing the Role of Soil Texture and Land Use in Unguaged Sites Located Hydrometric Data

Effective water resource management in areas with limited hydrometric data requires the application of innovative and integrated methods to examine hydrological dynamics more accurately. This study investigates and analyzes how flow was estimated in the sub-basins of the Dez in Lorestan Province. Initially, Sentinel-1 and 2 satellite data were used, along with SRCI and BI indices, to extract maps of soil textures, land use, and curve number (CN). Subsequently, Relying on rainfall and discharge data from 1992 to 2023 and statistical analysis, the return period of rainfall and flow for the studied sub-basins was calculated utilizing EasyFit software. The flow for each sub-basin was estimated using the SCS method and multivariate regression. The results indicated that multivariate regression, evaluated using the Durbin-Watson statistic (1.74), the coefficient of determination (0.768), the mean squared error (17.88), and the Nash-Sutcliffe efficiency (0.758) for a 2-year return period, was the most suitable method for estimating flow at ungauged stations within the sub-basins of the Dez River. Overall, this research presents effective approaches for water resource management and the optimization of hydrological in Lorestan Province, To optimize cost and time efficiency, the use of multivariate regression for flow estimation in ungauged hydrometric sub-basins is recommended.