The impacts of climate change on biodiversity and its adaptation will directly affect the efficiency of biodiversity conservation. Predicting spatial variation of suitable habitats of threatened species under future climate change has important theoretical and practical significance for biodiversity conservation. In this study,we predicted the suitable distribution of Elaeagnus mollis,an endemic endangered plant in China,under climate change at regional scales. Then,we simulated the spatial variation and migration tend of suitable distribution under different climate change scenarios by spatial analysis. The results from Maxent model showed that the two suitable distribution areas of E. mollis presented different migration trends under the future climate change scenarios: the suitable areas of Lyuliang Mountain would fluctuate slightly in latitudinal direction,while that in Zhongtiao Mountain would migrate to high elevation. Analysis of the spatial pattern change of the suitable areas indicated that the areas with obvious change occurred at the boundary of the suitable areas of E. mollis,including new suitable area and lost suitable area. The new suitable areas were scattered in the marginal of the original,with the increase rate of 9.1% to 20.9%,and the lost suitable areas were concentrated in the northern Lyuliang Mountain suitable areas and the southeast Zhongtiao Mountain suitable areas,with the loss rate of 16.4% to 31.1%. These regions were more sensitive to climate change. Using the classification statistical tool of Zonal,we found that the central points of the Lyuliang Mountain suitable areas showed southward migration trend under the future climate change,with the maximum migration distance of 7.451 km,while the center point of the Zhongtiao Mountain suitable areas showed migration trend to the northwest,with the maximum migration distance of 8.284 km. Our results indicated that the response of E. mollis distribution in Shanxi to climate change was intense.