Modelling and mapping soil erosion potential in China

Abstract

Soil erosion is an important environmental threat in China. However, quantitative estimates of soil erosion in China have rarely been reported in the literature. In this study, soil loss potential in China was estimated by integrating satellite images, field samples, and ground observations based on the Revised Universal Soil Loss Equation (RUSLE). The rainfall erosivity factor was estimated from merged rainfall data using Collocated CoKriging (ColCOK) and downscaled by geographically weighted regression (GWR). The Random Forest (RF) regression approach was used as a tool for understanding and predicting the relationship between the soil erodibility factor and a set of environment factors. Our results show that the average erosion rate in China is 1.44 t ha−1 yr−1. More than 60% of the territory in China is influenced by soil erosion limitedly, with an average potential erosion rate less than 0.1 t ha−1 yr−1. Other unused land and other forested woodlands showed the highest erosion risk. Our estimates are comparable to those of runoff plot studies. Our results provide a useful tool for soil loss assessments and ecological environment protections.

Keywords

• soil erosion potential
• RUSLE
• mapping
• modelling