Characterizing the occurrence of wildland-urban interface fires and their important factors in China

Urbanization and climate change have led to an increasing number of wildfires in China in recent years, posing a direct threat to life, property, and social development. These areas facing severe wildfire risk are mainly concentrated in the Wildland-Urban Interface (WUI). Although much research has been conducted on wildfire management in China, one of the critical constraints is that there is not yet a WUI distribution map that can meet the needs of forest fire management in China. The present study proposes an approach to identify WUI areas in mainland China between 2008 and 2020 by analyzing the impact of wildfire events on communities. Using historical fire occurrence data, we explored the spatial and temporal distribution patterns of WUI fires in various geographic regions. Using the random forest model, we predicted the spatial distribution of WUI fires and identified the key factors affecting their occurrence. The results show that the WUI area in China has been rapidly expanding since 2008, with a growth rate of 15.2 × 103 km2/a, especially in the densely populated and economically developed eastern regions. The seasons of WUI fires varied across regions. Both spring and autumn are major seasons for WUI fire occurrence in northern China. In contrast, more than 80 % of WUI fires in southern China occurred in winter and spring. According to the findings generated by the random forest analysis, weather patterns marked by low levels of rainfall and elevated temperatures during both spring and fall seasons are the primary atmospheric contributors to the incidence of WUI fires. Our study indicated that repeated human activities within southern China also increased the probability of fire occurrence. We suggest that better prediction of fire-prone weather and improved regional forest fire management can help decrease the incidence of WUI fires. In addition, future efforts should focus on enhancing the long-term monitoring and predictive capabilities of fuel moisture content.