Abstract: Forest fire risk depends on the hazard factors, affected body, and hazard prevention and reduction ability. The integrated risk assessment is the foundation for developing scientific fire mana-gement policies and carrying out the forest fire prevention measures. A forest fire risk assessment model and index system were established based on the classic natural disaster risk model and available data, and the model was used to assess the forest fire risks in past and future. The future climate scenario data included outputs from five global climate models (GFDL-ESM2M, HadGEM2-ES, IPSL-CM5A-LR, MIROC-ESM-CHEM and NorESM1-M) for RCP 2.6, RCP 4.5, RCP 6.0 and RCP 8.5, respectively. Each component index of Fire Weather Index (FWI) system was calculated daily for each grid in 1987-2050 for the historical observations and future climate scenarios according to the maximum temperature, minimum relative humidity, wind speed and daily precipitation. The results showed that areas with high and very high fire danger ratings in 1987-2010 accounted for 21.2% and 6.2%, respectively, which were distributed in Greater Xing’an Mountains and the Changbai Mountain area, most parts of Yunnan, and many fragment areas in southern China. The areas with high and very high burn possibilities were mainly distributed in the northeast and southwest region, accounting for 13.1% and 4.0%, respectively. Compared with the observation period, the areas with high and very high fire danger ratings in 2021-2050 would increase by 0.6%, 5.5%, 2.3%, and 3.5% under RCP 2.6, RCP 4.5, RCP 6.0, and RCP 8.5 respectively, and North China would show significant increase. The regions with high-risk forest fires would also increase due to climate change, with the most significant increase under RCP 8.5 scenario (+1.6%).