Hydrologic Design of Soil and Water Conservation Structures Using Probability Analysis and Machine Learning Techniques in Saurashtra Region of Gujarat, India

The soil and water conservation structures are constructed to overcome scarcity as a result of interannual rainfall variability paucity the perennial source water. present study was aimed estimate design runoff for efficient hydrologic various soils using machine techniques enabling harvesting available with economical which can support in developing climate resilience Saurashtra region Gujarat, India. at return periods predicted by Annual One Day Maximum Rainfall (ADMR) three technics i.e. Probability Distribution Fitting, Artificial Neural Network (ANN) Gaussian Process Regression (GPR) 11 stations. Various goodness fit tests revealed that ADMR efficiently log-logistic (3P) distribution six stations, generalized extreme value two stations lognormal (3P), gamma one station each. Among ANN GPR, performance indicators GPR has shown higher capability predict compared correlation coefficient ranging from 0.97 0.99, mean absolute error 15 mm 411 root squared 40 494 estimation demonstrated based on suitable like field bunding, terrace outlets vegetative outlets, diversion, permanent masonry gully control structures, earthen dam, etc. SCS-Curve Number method curve number 70 85. is useful researchers, planners engineers implement economical, safe structures.