An Assessment of the Empirical Erosion Potential Model in 63 Selected Watersheds in Iran

Accelerated soil erosion and high sediment production are important issues in many parts of Iran. The identification of high erosion risk regions is necessary for the planning of soil and water conservation measures in order to reduce erosion damages. Empirical models are required for estimating soil erosion and sediment yield in the absence of gauged small watersheds. The Erosion Potential Method (EPM), originally developed in Croatia, is a popular erosion estimating model in Iran; however, its efficiency across the country has not been thoroughly assessed. Therefore, a calibration study was carried out by comparison between the model the estimated and measured sediment yields through sediment survey for 63 reservoirs at the outlet of small catchments in the Semnan, Markazi, Isfahan, Chaharmahal va Bakhtiari, West Azarbaijan, Lorestan, Fars, Golestan, and the Khorasan Razavi Provinces. All required maps and information of the study basins for estimating the EPM sediment yield were produced via field survey as well as data analysis. Comparison of the results showed that in most cases, the model estimates (0.19 to 9.71 tons per square kilometer per year) are lower than the measured sediment yields ​​at the reservoir inlets (0.44 to 459.64 tons per square kilometer per year). Thus, we calibrated the EPM by modifying its original tables and evaluated the results using the Nash‐Sutcliffe efficiency coefficient in two scenarios, a, all basins, and b, after dividing into three geographical zones, which significantly improved the estimates. The Nash ‐ Sutcliffe efficiency coefficient for scenario a was 0.66, and 0.68, 0.50 and 0.89 for the Central, Zagros, and Northeast zones, for scenario b respectively. Based on the results, it is strongly recommended to use the modified EPM instead of the original model.