Improving MUSLE performance for sediment yield prediction at micro-watershed level using seasonal classified data

Abstract

Separating erosion data and assessing season-based models are of great importance considering the variation in soil erosion processes in different seasons, especially in semi-arid regions. However, evaluation of an erosion model using seasonal classification of data and at a micro-watershed level have rarely been considered. Therefore, the present study was conducted to evaluate the modified universal soil loss equation (MUSLE): 1) with the seasonal classification of data and 2) with the traditional approach (no classification of data), in the Sanganeh research micro-watershed. This watershed has an area of 1.2 ha and is located in the north east of Iran. The results showed that the original MUSLE overestimated the sediment yield in the study watershed. Also, after calibration of MUSLE, the seasonal classification of data (with a relative estimation error (RE) of 34%) showed its superior performance compared with the traditional calibration approach (with a RE of 62%). In this regard, the obtained REs of 33, 40, and 31% respectively for spring, autumn, and winter are within or close to the acceptable range.