Global analysis of influencing forces of fire activity: The threshold relationships between vegetation andb fire

Ruisen Luo,Yingying Dong,Muye Gan,Dejun Li,Shuli Niu,Amy Oliver,Ke Wang,Yiqi Luo
2013-01-01
Abstract:Manylarge scale firestudies considered the relationships between fire and its influencing factors as smooth.However, the responses of fire activity to influencing factors could be abrupt on the global scale, because the hysteretic responses of vegetation to fire and vegetation types are discrete. This study examined the climatic, vegetation, anthropogenic, lightning, and topographic factorsdriving variations in global fire density, and discussedthe thresholds of vegetation on fire activity. Fire density was developed from 7 years of Moderate Resolution Imaging Spectroradiometer (MODIS) active fire data to represent global fire activity, and nine typical influencing variables were selected. The random forest regression tree method was used to identify the relative importance and relationships between fire and the influencing variables. The patterns of global fire density were captured well by the model (78.33% variance was explained), and the related thresholds were identified. Climatic factors played a primary role in determining global fire density. Agricultural land use and topographic roughness were not identified as the most important factors, probably due to the large scale we considered. Three intervals of tree density were identified to have distinct levels of fire density. Intermediate tree density (9%-53%) was related with the highest fire density, but both low and high percent of tree cover were associated with low fire density (7.0 vs. 1.3/0.9 counts per 100 km2 per year). This study could provide further insights into understanding of the threshold effects of influencing factors on fire activity, and contribute to advances in fire modelingand vegetation distribution studies.
What problem does this paper attempt to address?