Document Type : Research Paper
Author
Assistant Professor, Department of Range and Watershed Management, Arsanjan Unit, Islamic Azad University, Iran
Abstract
Rainfall erosivity is one of the effective factors on rainfall erosion which is very important in water erosion. Preparing rainfall erosivity mapping is an important issue in soil conservation, erosion control and land management. The objective of this research is evaluation of geostatistical methods for rainfall erosivity mapping in Fars province. First, 92 meteorological stations having at least 20 years data and relevant distribution over the region were selected and the modified Fournier index was calculated for each station. Then several interpolation methods including Ordinary Kriging (OK), Simple Kriging (SP) and deterministic methods such as Inverse Distance Weights (IDW), Radial Basis Function (RBF), Local Polynomial Interpolation (LPI) and Global Polynomial Interpolation (GPI) were used in order to extend point data to spatial information in the process of preparing rainfall erosivity map. The evaluation of the results based on higher R value and lower MAE and RMSE values showed that among the studied methods, the OK method has higher R value and lower RMSE and MAE values (R=0.87, MAE=13.09, RMSE=18.52), then it was selected as the most appropriate method for rainfall erosivity mapping in Fars province. Also the results revealed that 0.24% of the province has negligible erosivity, 27.06% has low erosivity, 43.62% has moderate erosivity, 14.48 have high erosivity and 14.6% has very high erosivity.
Keywords
)