The potential vegetation can reflect climatic influence on vegetation type change, and provide a scientific reference and guide for restoration and reconstruction of vegetation ecosystems. Meteorological observation data from 1961 to 2017 and predicative data under the Representative Concentration Pathway (RCP2.6, RCP4.5, RCP6.0 and RCP8.5) during 2030s, 2050s and 2080s, in conjunction with comprehensive sequential classification system (CSCS) model and Geographic Information System (GIS) technology, were used to analyse the spatio-temporal distribution and variation of Sensitive Regions of Potential Vegetation (SRPV) in China. Results suggested that SRPV presented a pattern of total dispersion and partial agglomeration under all scenarios. The agglomerate regions spread from north-eastern China (e.g. Inner Mongolia Plateau, Greater Xing’an Mountains, and North-east Plain) to south-western China (e.g. Loess Plateau, Qinling–Huaihe belt, Nanling Mountains, Jiangnan hills, Qinghai–Tibet Plateau, and Tarim Basin). From the 2030s to the 2080s, SRPV exhibited characteristics of expansion and migration to the north under RCP scenarios. The distributed area of SRPV increased with increased radiation emission intensity. These results both further expand the research on potential vegetation using the CSCS, and also provide reference for governments to ensure vegetation ecological protection.