The occurrence frequency of forest fires in China is high, causing great losses and serious damage to the forest ecosystem，especially for the Qahar Youyi Qianqi, the vegetation coverage rate is low, the fire loss is large, and it is difficult to recover. Therefore, it is of great significance to screen out local fire- resistant tree species for forest protection and forest grassland fire prevention in this region. Field investigation and laboratory test were carried out to measure and analyze the fuel moisture content, calorific value, crude fat content and ash content of litter layer and humus layer in different parts of trees and shrubs in the study area. Principal component analysis and ranking of physical and chemical property indexes were conducted by using SPSS 23.0 and Origin 2021 software. The combustibility of 7 typical tree species in this region from small to large is: Populus davidiana Dode, Ulmus pumila L., Ostryopsis davidiana Decaisne, Caragana korshinskii Kom., Spiraea salicifolia L., Prunus pedunculata Pall., Pinus tabuliformis Carriere. It can be divided into three grades：strong fire resistant tree species（Populus davidiana Dode, Ulmus pumila L., Ostryopsis davidiana Decaisne.），fire resistant tree species: Caragana korshinskii Kom.; Weak fire- resistant tree species（Spiraeasa licifolia L., Prunuspedunculata Pall., Pinus tabuliformis Carriere）. The Populus davidiana Dode forest in this area is the most prone to fire, and as Populus davidiana Dode is the main afforestation species in this area, therefore, afforestation should be appropriate with strong fire resistance of tree species, artificial intervention to prevent the occurrence of fire.

desert steppe, combustibility, principal component analysis, fire resistance