Abstract: The spatial distribution and combustibility of forest fuels are decisive for the occurrence and spread of forest fires. In this study, stand structure characteristics and fuel spatial distribution in 16 forest stands from Guangzhou City were obtained by terrestrial laser scanning. The surface fuel load was measured by ground measurement and standard sampling methods. Leaves of dominant tree species in different stands were collected to measure physicochemical properties. One-way ANOVA was used to compare the spatial distribution of fuels and the differences in the physicochemical properties of leaves of dominant tree species among different stands, to analyze the potential fire behaviors and fire dangers. The comprehensive scores of combustibility of different stands were calculated by principal component analysis (PCA), while the fire risk grading was classified by cluster analysis. The results showed that there were significant differences in spatial distribution of fuels and the physicochemical properties of leaves of dominant tree species among 16 forest stands. The 16 forest stands could be classified into four fire danger grades based on combustibility: (1) Pinus massoniana forest and Cunninghamia lanceolata forest had the very high fire risk; (2) pine and broadleaved mixed forest, Cunninghamia lanceolata and broadleaved mixed forest, Castanopsis chinensis forest, Eucalyptus forest, and Acacia confusa forest had high fire risk; (3) Schima superba forest had the moderate fire risk; (4) broadleaved mixed forest, Castanopsis fissa forest, Acacia auriculiformis forest, Castanea henryi forest, Elaeocarpus apiculatus forest, Machilus chinensis forest, Schefflera octophylla forest, and Liquidambar formosana forest had low fire risk. According to the combustibility and fire risk grade of different stands, the corresponding fuel management measures and forest fire prevention countermeasures were proposed, which would provide theoretical support and scientific guidance for the deployment of forest fire prevention resources and the construction of fire prevention forest belt in Guangzhou.

Key words: fuel load, combustibility, forest fire behavior, forest fire risk, terrestrial laser scanning