The return of fire in a transitioning forest: Ecosystem processes and soil biodiversity in fire-excluded, southern Appalachian forests

Melanie K. Taylor, Mac A. Callaham Jr. and Joseph O'Brien, Southern Research Station, USDA Forest Service, Athens, GA, Melanie K. Taylor and Nina Wurzburger, Odum School of Ecology, University of Georgia, Athens, GA, Dana O. Carpenter, Wetlands, Baker University, Lawrence, KS, Irenee Payne, Environmental Sciences, Oregon State University, Corvallis, OR, Irenee Payne, Hall County Soil & Water Conservation District, Gainesville, GA

Background/Question/Methods
Over the past century, eastern North American forests have experienced fire exclusion, atmospheric N deposition, and the loss of tree species to invasive pests and pathogens. Consequently, eastern forest composition is transitioning from fire-adapted and ectomycorrhizal (ECM) tree species to fire-intolerant and arbuscular mycorrhizal (AM) tree species. It is critical to understand how this transition affects soil biodiversity and associated ecosystem processes. Of particular interest is how these new forests will respond to the reintroduction of fire, either as wildfire or as a management action. To investigate these questions, we established plots across a gradient of ECM tree species inside and outside the perimeter of the 2016 Rock Mountain wildfire in the southern Appalachian Mountains. We measured decomposition rates of AM and ECM litter using litterbags and quantified soil and litter macroinvertebrate communities. We hypothesized a reduction in litter decay rates and macroinvertebrate abundance and diversity with increasing dominance of ECM trees in unburned plots. In burned plots, we predicted an overall reduction in litter decay rates and macroinvertebrate diversity and abundance, but we expected no relationship of these with ECM dominance. Overall, we hypothesized that ECM litter would decay more slowly than AM litter.
Results/Conclusions
We found that litter decay rates decreased with increasing abundance of ECM trees in unburned plots, as expected. However, we observed two unexpected patterns in litter decay. First, litter decay rates increased with increasing ECM dominance in burned plots. Second, AM litter decomposed more rapidly in burned than in unburned plots, while ECM litter decay did not differ with fire status. These findings suggest that fire alters the biogeochemical context for litter decomposition, particularly for fast-decomposing AM litter. In the litter decomposer community, earthworm abundance was not affected by increasing ECM percentage or burning, suggesting this group is resilient to impacts of fire. Additional identification and analyses of other soil and litter macroinvertebrates are ongoing and will be reported. Our results suggest that both fire and changes in forest composition may have profound impacts on litter decay rates, which may influence SOM formation and ecosystem C storage. Understanding these effects are increasingly important as eastern forests continue to change and fire is reintroduced to the landscape.