Fire effects on soil carbon cycling pools in forest ecosystems: A global meta-analysis

Abstract

Changes in soil carbon (C) pools driven by fire in forest ecosystems remain equivocal, especially at a global scale. In this study we analyzed data from 232 studies consisting of 1702 observations to investigate whether ecosystem type, climate zone, stand age, soil depth, slope, elevation, and the time since fire in influence of forest soil carbon pools to fire regime (fire type, fire season, fire intensity). Additionally, we explored the potential mechanisms of the relationships between multiple response variables to the fire using linear regression and random forest models. On aggregate, fires significantly increased the mean effect sizes of several key soil carbon cycling components-including microbial biomass carbon (MBC), dissolved organic carbon (DOC), total carbon (TC), pyrogenic carbon (PyC), soil organic matter (SOM), soil organic carbon (SOC) by 0.77, 0.89, 0.87, 1.22, 0.97 and 0.93, respectively, compared to unburned forests ecosystems. However, the fire effects on soil C pools vary widely between environmental factors and duration, and are mediated by factors such as tree species, fire type, and soil layer. A correlation analysis displayed the effects of fire on MBC and DOC were significantly and negatively correlated with elevation. Fire effects on the forest floor and mineral soil indicated significantly increased PyC. SOC and TC in coniferous tree species are the most sensitive to fires, thereby altering important feedback relationships with the fire-vegetatale-climate system. Interestingly, latitude has a stronger influence on SOC than mean annual precipitation or elevation, indicating that variations in latitude play a significant role in regulating the amount of SOC in forest ecosystems. Overall, the results illustrated geographic variation in fire effects on soil C cycling underscores the need for region-specific fire management plans, and help us understand the responses of soil C cycling to fire in forest ecosystems, and facilitate decision-making to forest fire management.