Revealing future changes in China’s forest fire under climate change

Forest fires, driven by global change, have become an increasingly severe concern for the future. To mitigate this uncertainty, understanding the projected response patterns of forest fires under climate change can provide valuable insights, especially in regions like China, where forest fires are prone to occur and highly sensitive to climate variations. This study used the Random Forest algorithm to develop predictive models for occurrence probability, intensity levels, and burned area of forest fires in China. These models were applied to future climate simulations under SSP1–2.6, SSP2–4.5, and SSP5–8.5 scenarios, providing predictions for the mid-term (2041–2060) and long-term (2081–2100) periods with a comparative analysis against the baseline period (2001–2020) to assess relative changes. Results show that, by the end of the century, 60 % to 70 % of forested areas in China are projected to face increased fire risk, with the average percentage increase not exceeding 20 %, and most low-risk areas remaining at low risk. Nevertheless, the transition to moderate or high fire risk is particularly significant in the long term under SSP5–8.5, with hotspots such as the Greater Khingan Mountains, central and southeastern hilly regions, and the gorges of the western Yunnan Plateau. While fire intensity in most fire-prone areas is projected to remain at medium-to-low levels, localized intensification is possible in southwestern China. The increase in burned areas is projected to be widespread, covering 63 % to 64 % of fire-prone areas, with the greatest growth occurring in the mid-term and under SSP5–8.5. Compared to the southern regions, northeastern China shows a smaller proportion of areas experiencing increased burned areas but a larger total increase, indicating more concentrated fires in specific areas. These findings reveal the distinct future trajectories of different forest fire characteristics with regional variability in China, offering critical information for their comprehensive consideration and integration into forward-looking, scientifically robust national fire prevention planning.