Abstract
Surface soil shear strength is an effective parameter in soil erosion and runoff modeling, and known as soil erodibility index. This study was conducted to investigate and predict soil shear strength in central Zagros region using multiple linear regressions (MLR). The study site was selected in rangeland and degraded-rangeland landuse with an area of 23562 ha. Based on the maps of geology, topography, landuse and soil capability, 18 Land Unit Tracts (LUT) was created. Supervised random sampling was used to collect samples in every land unit. A total of 120 samples were collected in triplicates to determine sample variability in each LUT. Soil samples were collected from the 0-3 cm of soil depth. Routinely measured (available) parameters included particle size distribution, soil organic carbon, CaCO3 in addition to slope, aspect; elevation and normalized difference vegetation index (NDVI) were used. Three MLR models were tested and investigated for soil shear strength in this study. The performances of the different models were evaluated using spearman’s correlation coefficient (r) between the observed and the estimated values, normalized mean square error (NMSE), mean absolute error (MAE), geometric mean of error ratio (GMER) and geometric standard deviation of the error ratio (GSDER). Results showed that land degradation significantly reduced surface soil shear strength. Addition of slope, aspect, elevation and NDVI to soil attributes as input parameters improved the performance criteria for soil shear strength.

Keywords: Surface soil shear strength, Soil erodibility, Central Zgros