46-Year (1973–2019) Permafrost Landscape Changes in the Hola Basin, Northeast China Using Machine Learning and Object-Oriented Classification

Land use and cover changes (LUCC) in permafrost regions have significant consequences on ecology, engineered systems, the environment. Obtaining more details about LUCC is crucial for sustainable development, land conservation, environment management. The Hola Basin (957 km2) northernmost part of Northeast China, a boreal forest landscape underlain by discontinuous, sporadic, isolated permafrost, was selected case study. analyzed using Landsat archive satellite images from 1973 to 2019. A thematic change detection analysis performed combining object-based image (OBIA) Support Vector Machine (SVM) algorithm. Four types (forest, grass, water, anthropic) were extracted with an overall accuracy 80% >90% 1986, 2000, Forest, dominant class (750 km2 1973), declined 88 (11.8%) 1986 but had recovery 78 (12.5%) 2000 Grass, second-largest (187 increased 86 (46.5%) between decreased 90 (40%) anthropic continuously 10 (1973) 37 (2019). Major features are attributed rapid population growth, resource exploitation, agriculture intensification, economic frequent fires. Under pronounced climate warming, these drivers been accelerating degradation subsequently triggering natural hazards deteriorating ecological This study represents benchmark management Basin, China.