OOS 31-8
Burn severity influences soil microbial responses to wildfire in Alaskan boreal forests

Friday, August 9, 2013: 10:30 AM
101A, Minneapolis Convention Center
Sandra R. Holden, Ecology and Evolutionary Biology, University of California, Irvine, Irvine, CA
Brendan M. Rogers, Earth System Science, University of California, Irvine, Irvine, CA
James T. Randerson, Department of Earth System Science, University of California, Irvine, Irvine, CA
Kathleen K. Treseder, Ecology and Evolutionary Biology, University of California, Irvine, Irvine, CA
Background/Question/Methods

Wildfires are a chronic disturbance in boreal forests, and climate warming is likely to increase wildfire frequency and severity. Forest fires create a patchy distribution of burn severities across boreal landscapes. Although previous studies have documented differences in soil microbial communities between burned and unburned boreal forests, few previous studies have accounted for within-site variation in burn severity. In this study, we asked how soil microbial responses to fire differ as function of burn severity. To address this question, we sampled fibric and humic soils from 19 locations within a burned boreal forest in interior Alaska. We also collected soils from an adjacent unburned boreal forest. The Landsat-derived difference Normalized Burn Ratio (dNBR) was used to estimate burn severity at each sampling point. We measured microbial biomass and basal respiration in each soil sample. In addition, we used pyrosequencing of 18s fungal DNA to characterize fungal community composition at each sampling location.

Results/Conclusions

Burned fibric soil samples had significantly lower microbial biomass per gram of soil than unburned soil samples (P < 0.0001). Differences in burn severity (dNBR) among sampling locations explained a significant amount of the variation in microbial biomass across fibric soil samples, with more severely burned soils having lower microbial biomass (r2 = 0.834, P < 0.0001). In contrast, microbial biomass in humic soils was not significantly different between burned and unburned locations (P = 0.256) and showed no relationship with burn severity (r2 = 0.135, P = 0.108). Changes in basal respiration among soil samples mirrored differences in microbial biomass. High severity fires may elicit greater reductions in microbial biomass in fibric soils due to higher burn temperatures or increased post-fire carbon and water limitation. Forthcoming analyses will determine how fungal community composition varies as a function of burn severity. Taken together, these results suggest that burn severity is an important factor modulating soil biogeochemical responses to boreal forest fires, and they highlight the importance of accounting for landscape-level variation in disturbance severity in future studies.