Increasing risk of global forest loss from extreme wildfires under climate change

Climate change has intensified the frequency and severity of forest fires, driving a global increase in fire-induced forest losses. Although extreme forest fires have been extensively studied at regional scales, the global characteristics of extreme fire-induced forest loss (EFFL) remain poorly understood. Here, we used high-resolution remote-sensing data to assess the spatiotemporal dynamics, ecological impacts, and climatic drivers of EFFL events worldwide. Our results show that from 2001 to 2023, EFFLs were widespread across global forest ecosystems, with boreal forests having the highest frequency and extent, contributing to 57.53% and 58.37% of the global EFFLs, respectively. During this period, the frequency of EFFLs increased 2.02-fold. Moreover, 8.79% of endangered forest vertebrates (662 species) experienced EFFLs in >10% of their geographic range, and 1.76 × 109 tons of aboveground biomass (AGB) was affected. High-risk regions, including the Amazon rainforest, Southeast Asia, and southeastern Australia, face severe threats to species conservation and carbon stocks. The occurrence of EFFLs was consistently associated with dry and hot anomalies. Our findings underscore the urgent need for enhanced forest monitoring and conservation, and provide a scientific basis for developing targeted fire prevention and management strategies.