PS 16-154
How is N fixation regulated in fire-adapted ecosystems?

Monday, August 10, 2015
Exhibit Hall, Baltimore Convention Center
Michael Ament, Odum School of Ecology, University of Georgia, Athens, GA
Nina Wurzburger, Odum School of Ecology, University of Georgia, Athens, GA
Background/Question/Methods

The capacity for nitrogen (N) fixing organisms to up-regulate fixation in N-poor environments allows ecosystems to overcome N limitation. However, we poorly understand the ecological controls on N fixation in ecosystems, particularly those characterized by frequent fire. Longleaf pine savannas are fire-dependent ecosystems that offer an ideal context to investigate how disturbance and chronic N loss affect N-fixing organisms. Here, we hypothesized that increased fire frequency stimulates leguminous fixation and suppresses free-living fixation. To test these hypotheses, we quantified the activity of free-living fixers (O horizon mass and acetylene reduction assays) and legume abundance and diversity in 23 1 ha plots of sandhill longleaf pine forests distributed along a fire frequency gradient (fire return intervals of 2, 3, and 15+ years) at Fort Benning (GA) and Eglin Air Force Base (FL).

Results/Conclusions

Increased fire frequency resulted in substantial increases in legume abundance and concomitant declines in free-living fixation. Mean legume abundance in savannas with 2-year fire return intervals was 46% and 362% greater than those with 3 and 15+ year fire return intervals, respectively. These results suggest that increasing fire frequency triggers N limitation and promotes legume abundance. We also observed strong species-specific trends in the legume community with fire frequency. Low-capacity N-fixers, such as Rhychosia cytisoides, dominated legume communities at high fire return intervals (62%), but were scarce in savannas that burned frequently (10%). Conversely, high-capacity N-fixers, such as Tephrosia virginia and Galactia regularis, reached peak abundances in frequently burned savannas (19% and 37%) but were notably absent in savannas that burned infrequently. Free-living fixation increased linearly with years since fire (r2=.71), but fixation rates were low. These results suggest that free-living fixers facilitate small inputs of N in savannas that experience low N losses. In contrast, the legume community is highly responsive to fire frequency, suggesting that leguminous N fixation is the primary mechanism replenishing N losses in longleaf pine savannas.