Mapping gully affected areas based on Sentinel 2 imagery and digital elevation model

Gully poses a great threat to agricultural production and ecological environment. Mapping accurate gully affected areas plays an important role in regional environmental monitoring and management. In this paper, an object-based image analysis method based on Google Earth Engine (OBIA-GEE) for mapping gully affected areas was proposed by using Sentinel 2 imagery and AW3D 30 DEM data. The method utilized the simple non-iterative clustering (SNIC) segmentation and the random forest (RF) algorithm to segment imagery and map gully affected areas. And terrain skeleton lines were further used to optimize the mapping results. The proposed method was applied to five study areas with different landform types on the Chinese Loess Plateau, and the results showed that the method achieved good performance with the overall accuracy of 86.44%, user’s accuracy of 84.97%, and producer’s accuracy of 83.90%. The OBIA-GEE method provides the possibility of large-scale gullies mapping, which is beneficial to monitor gullies and manage soil erosion.