Background Soil erosion has seriously threatened the social and economic development in many regions of the world. The Pearl River Delta is one of the most dynamic economic zones in China and even the world. The Pearl River Basin is located in the south subtropical region, with the disadvantages of heavy precipitation and thin soil. Coupled with the interference of human activities, soil erosion has become more severe. However, it is still insufficient in the current research on the spatial differentiation characteristics and main controlling factors of soil erosion in the Pearl River Basin, which is not suitable for the prosperous economy. Therefore, it is urgent to make use of modern scientific and technological means to systematically study soil erosion in this area.

Methods In this study, the Pearl River Basin was selected as the study area, and China Water Conservancy Survey (WRI) was used to set up sampling survey units. The Chinese Soil Loss Equation (CSLE) was used, based on the sample survey unit data interpreted by high-resolution remote sensing images and thematic data on soil erosion factors. The map algebra and spatial interpolation were used to calculate the soil erosion rate map in the study area, and to analyze the spatial pattern of soil erosion in the Pearl River Basin. Furthermore, the geo-detector method was used to quantitatively identify the main influencing factors.

Results 1) Among the factors affecting soil erosion rate, R and LS are generally higher, which were the main inducing factors of soil erosion. The B factor value was low, while the E and T factor values were high. The study area had good vegetation coverage and was the main inhibitor of soil erosion. 2) The soil erosion in the Pearl River Basin mainly concentrated in Guizhou and Yunnan provinces, central Guangxi and coastal areas of Guangdong province. The areas of intense and extremely intense erosion were relatively small and distributed on relatively scattered sloping farmland. The mean values of soil erosion rate mapped by map algebra and spatial interpolation were 791.78 t/(km²·a) and 615.37 t/(km²·a) respectively. The mean values of the two soil erosion rates were relatively close. The average value of soil erosion rate calculated by map algebra method was higher, which might be related to the difficulty of integrating water conservation measures into the calculation. 3) Except for hilly and mountainous area of Zhejiang-Fujian, land use pattern was the main controlling factor of soil erosion, and the influencing factor Q value was above 47%. Secondly, B factor had a great influence on soil erosion rate and its spatial distribution. T factor ranked the third, and K factor and R factor had little influence. That was related to the high vegetation coverage in the study area, which was dominated by forest and grass vegetation.

Conculsions Two methods of map algebra and spatial interpolation are used to map soil erosion in the Pearl River Basin. Soil erosion is mainly controlled by land use and biological measures. Adjusting land use structure and optimizing the soil conservation function of vegetation are the main direction of future management. This study may provide a reliable scientific basis for ecological environment restoration and soil conservation in this region, and promote the development of soil erosion research in subtropical region.