FLOW DURATION CURVE ESTIMATION AT UNGAUGED BASIN USING REGIONALIZATION APPROACHES

Abstract

Abstract Long-term hydrological information is essential in many parts of the world because of low density and inadequate spatial distribution of hydrometric networks specifically in Pakistan. The goal of this study to delineate the homogeneous region and to assess the streamflow in ungauged basin using regionalization approaches. Current study introduced a Simple Tyler Skill Score (STSS) technique to ensemble the output of regionalization approaches i.e., Artificial Neural Network, (ANN), Inverse Distance Weightage (IDW) and Stepwise Regression (SWR) for better predication of hydrological information. STSS method is mainly based on the weight derived from coefficient of determination (R2) and hydrological variables. The overall performance evaluation was performed by using coefficient of determination (R2), correlation coefficient (r), root mean square error (RMSE) and percent bias (PBIAS), which revealed that STSS provided more robust estimation of flow duration curve as compared to other methods. Moreover, ANN performance was comparatively better than the SWR and IDW method. The result emphasizes that in upper Indus basin, the characteristics of the watershed matter more than the physical distance between gauged and ungauged areas. This study can provide the direction for the hydrological estimation independent of hydrological modeling in data scarce regions where hydrological conditions are less addressed.