Evaluating the Efficiency of the USLE, RUSLE, USLE-M and AUSLE Models in Estimation of Soil Loss at Plot Scale in a Semi-Arid Region

This study was conducted to investigate the capability of the Universal Soil Loss Equation (USLE) and its versions (RUSLE, USLE-M and AUSLE) models in predicting soil loss at the plot scale in a semi- arid region in Zanjan. Toward this, twenty two plots from 1m to 22.1m with the same widths (1.83m) were installed on a uniform slope 10 percent. The plots were maintained under the USLE standard plot conditions for a fourteen-month period (from March 2013 to Jun 2014). Based on the results, an increasing trend was observed in soil loss with increasing plot length from 1 m (0.01 ton/ha.year) to 6 m (0.07 ton/ha.year), which was associated with increasing runoff and its transport capacity. Mean estimated soil loss using the USLE, RUSLE, USLE-M and AUSLE for twenty two plots was 80.2, 65.4, 4 and 113.5 times bigger than the measured values, respectively. The USLE-M appeared the lowest error in soil loss estimation as compared to other two models. In this model, erosivity factor is determined using runoff along with rainfall characteristics. This study revealed that the calibration of the empirical soil erosion models using long period data is essential for predicting soil loss in the plot scale.