Prioritization of Effective Factors on Gully Erosion and Determination of Sensitive Areas in Kermanshah Province Using MaxEnt Model

  The aim of this study was to determine the areas sensitive to gully erosion and prioritize the factors affecting it in Kermanshah province, using the maximum entropy method (MaxEnt model). 9 factors affecting gully erosion potential in the area Including Elevation, Slope, Slope Direction, Lithology, Land Use, Topographic Moisture Index, Surface Curvature, Drainage Density and RSP was reviewed. Of the total gullies studied in the study, 70 percent were randomly classified as test data and 30 percent as validation data. To determine the most important parameters, the Jackknife diagram and the relative performance detection curve ROC were used to predict the model. The results showed that there is no alignment between the factors used. According to the Jackknife diagram, the lithological layers, slope, height and  drainage density were the most important factors influencing the susceptibility of gully erosion, respectively. ROC curve showed 91percent accuracy of the maximum entropy method in the training phase and 72 percent in the validation phase. According to the moat susceptibility zoning map, about 23.39 percent of the area has low sensitivity, 48.14 percent very low, 9.31percent high and 4.71percent very high and 14.46 percent has moderate sensitivity.