Sediment yield modelling using SDR and MUSLE with high resolution satellite precipitation dataset in an ungauged basin

Abstract

Erosion has become one of the extreme ecological dangers in up to date. Decrease of minerals in the upper layer of soil leads to failure in sustainable crop practices. Many researchers have developed prediction models of Sediment Yield (SY) in gauged basins. But modelling for an ungauged basin is very difficult due to the difficulty in validating the predicted model. The primary goal of the study was to identify the sedimentation in the study area using multiple(SY) methodologies, including Sediment Delivery Ratio (SDR) and Modified Universal Soil Loss Equation (MUSLE), in an ungauged basin with a high-resolution satellite precipitation dataset.. Different attributes such as LULC (land use / land cover), soil texture, precipitation, topography, etc. was incorporated to estimate the SY in Ponnaniyar river basin,. The generated SY map from SDR and MUSLE was evaluated by the receiver operating characteristic curve (ROC). The SDR model was found to be an efficient method for determining the SY for Ponnaniyar river basin, and also satisfied the criteria of AUC value of 0.752. The severely affected sub-watershed of Ponnaniyar river basin was identified with the help of erosion and yield spatial map. The obtained results will help prioritize the sub-watershed for locating water harvesting structures in further studies. This study suggests placing the gauging station in Ponnaniyar river basin to monitor the daily observation of discharge and SY estimation to prevent soil loss during flash flooding.