Forest fire is one of the main disturbances in the Great Xing'an Mountains. Revealing temporal and spatial distribution characteristics of fire occurrence in this region was important for local fire management and fire prediction under climate change scenarios. Generally,distribution patterns of fire occurrence may vary at temporal and spatial scales. In this study,we used historical fire ignitions occurred between 1967 and 2006 to derive fire clustering characteristics at various spatial and temporal scales in the Great Xing'an Mountains. We employed the coefficient of variation and Allen Factor to derive the temporal clustering scope of fire occurrence,and used the Ripley's K function and Kernel density methods to estimate the spatial clustering characteristics of fire occurrence. The results showed that at a fine temporal scale,fire distribution was in a tendency of stochastic process. The time-clustering phenomenon was clearly visible as the observation time increased. Spatially,the fires showed aggregated distribution characteristics within certain distances. Exceeding these distances,fires showed stochastic Poisson distribution characteristics. The forest fires were located in certain areas,where there was a significant high occurrence density across the landscape. These results indicated temporal and spatial distribution patterns of forest fires in the Great Xing'an Mountains,yet lacked an in-depth research about the impact mechanism of the fire patterns,which remains to be solved.