Spatiotemporal changes in desertified land in rare earth mining areas under different disturbance conditions

摘要: Special mining methods and red soil lead to large-scale land degradation desertification in ion-type rare earth (RE) areas. Therefore, it is crucial for ecological management restoration of areas accurately understand the evolution process desertification. In this study, remote sensing Landsat images from 1986 2019 were used extract desertified information Lingbei areas, Dingnan County, Ganzhou, China. To improve reliability experiment, samples selected Google verification compare accuracy difference index (DDI) model random forest (RF) algorithm extracting information. The results showed that compared with DDI model, overall kappa coefficient RF based on multiple features improved by 7% 9.37%, respectively, indicating its higher applicability. Spatiotemporal change analysis area total increased most rapidly during 1986–1994 reached 60.75 km2. continuously 1994 2004 a maximum 143.08 km2 2004. decreased 50.27 km2, but severe (SDL) 1.69 2004–2011. gradually declined stabilized 2011 2019. Analysis under different disturbance conditions disturbed RE was severely damaged. There still an 16.77 restoration, which 2.24 belongs SDL level. Moderate (MDL) light (LDL) have not been completely contained require attention relevant departments ensure their timely reclamation.