COS 131-10
Consequences of changing fire frequency for carbon pools and fluxes in a Longleaf Pine savanna system

Thursday, August 13, 2015: 4:40 PM
347, Baltimore Convention Center
Justin Wright, Biology, Duke University, Durham, NC
Gregory M. Ames, Biology Department, Duke University, Durham, NC
Cari D. Ficken, University Program in Ecology, Duke University

Prescribed fire is an important management tool for maintaining diversity and ecosystem processes in historically fire-dominated ecosystems.  However, determining the optimal burn frequency requires balancing multiple management objectives and complex interactions.  Here we take a system approach to understanding the controls of carbon cycling in a long leaf pine savanna ecosystem and how experimental changes in prescribed burn frequency can alter pool sizes and dynamics. Previous work has shown that the low-intensity ground fires used in prescribed burns in this system can reduce tree growth in this system, but the effects of such fire on other aspects of the carbon cycle are unknown. We measured standing above ground biomass, litter biomass, below ground biomass and soil carbon as well as above and below ground productivity and litter fall across an upland to riparian zone ecotone at each of 12 sites randomly assigned to control frequency (3 year burn rotation), burned annually for 4 straight years, or fire suppression for 5 years.


There were high amounts variation in various components of the carbon cycle along the upland to wetland ecotone within the sites.  Despite this high variation, we found evidence that changing the frequency of fire was beginning to alter the way in which carbon is moving through the system.  In general, the sites from which burning had been removed showed larger carbon storage, with trends towards higher soil carbon, higher understory biomass, and highly significant increases in liter pools compared to sites with annual burns, with 3 year burn sites typically showing intermediate values.  Although our previous results suggested that fire leads to slower tree growth, we saw no differences in basal area between the treatments, although this result is not particularly surprising given the large inertia present in this slow-turnover carbon pool.  These results suggest that management decisions about using fire to maintain diversity might have the potential for decreased carbon storage.  Furtheremore, if natural fires increase in frequency under the hotter, drier conditions predicted under climate change scenarios, net losses of stored carbon could lead to positive feedbacks.