Vegetation Cover Thresholds for Soil Loss and Their Spatial–Temporal Patterns in the Loess Plateau, China

ABSTRACT Identifying vegetation cover thresholds for soil loss and accurately determining spatiotemporal threshold variations are crucial to controlling soil loss across the Loess Plateau, China. However, few studies have quantified thresholds associated with soil loss, particularly those focusing on the spatiotemporal distribution of the identified thresholds. In this study, the normalized difference vegetation index (NDVI) and the revised universal soil loss equation (RUSLE) are utilized to represent vegetation cover and calculate soil loss, respectively. We then analyzed the nonlinear relationship between soil erosion and NDVI, calculated NDVI thresholds, and determined the spatiotemporal patterns of NDVI thresholds on the Loess Plateau between 1990 and 2020. We showed the following: (1) There are NDVI thresholds for soil loss in the Loess Plateau. The relationship between soil loss and NDVI is not a simple linear relationship. Small changes in NDVI near the thresholds will lead to abrupt changes in soil erosion. (2) Spatially, the NDVI thresholds for soil loss vary across different regions of the Loess Plateau, with the northwest exhibiting lower thresholds compared to the southeast. There is a noticeable increasing trend in NDVI thresholds for soil loss from the northwest to southeast. (3) Temporally, the NDVI thresholds for soil loss across various locations in the Loess Plateau have generally increased from 1990 to 2020, and the average NDVI threshold for soil loss across the plateau has also risen. As the average soil loss in the Loess Plateau first decreased and then stabilized over this time period, the increase in the NDVI thresholds for soil loss implies that the risk of soil loss in the region is increasing. Our findings provide an important reference for vegetation restoration practices.