Analysis of Long-Term Trend of Stream Flow and Interaction Effect of Land Use and Land Cover on Water Yield by SWAT Model and Statistical Learning in Part of Urmia Lake Basin, Northwest of Iran

The water yield produced at the outlet of a sub-basin is combination multiple interacting land uses. In majority previous research, while accounting for effect use and cover (LULC) on yield, hydrologic components watershed have been attributed to dominant class within that sub-basin. We adopted an approach investigate interaction LULC (WYLD) using Johnson–Neyman (JN) method. soil assessment tool (SWAT) model was employed in Urmia Lake Basin (ULB) estimate WYLD following successful calibration validation by stream flow. It found each sub-basin, statistically significant only when area rangeland less than 717 ha agricultural lands 633 ha. On other hand, trend flow assessed over 70 years two stations Bayesian Estimator Abrupt change, Seasonal Trend (BEAST). year 1991 turned out be most likely change point both stations. A decrease Lake’s level started 1995, which indicated part this shrinkage caused inflow reduction 4-year time delay. Besides identifying probable seasonal points, method has additional capability analyze uncertainty estimated lacking earlier methods.