COS 165-6 - Simulating future climate and management effects on fire and vegetation to inform conservation

Thursday, August 9, 2012: 3:20 PM
Portland Blrm 257, Oregon Convention Center
Jennifer K. Costanza1, Adam Terando2, Todd S. Earnhardt1 and Alexa McKerrow3, (1)Department of Biology, North Carolina State University, Raleigh, NC, (2)Biology Department, NCSU, Biodiversity and Spatial Information Center, Raleigh, NC, (3)Core Science Systems, United States Geological Survey, Raleigh, NC
Background/Question/Methods

Projections of future landscape dynamics are essential for guiding conservation strategies. To inform conservation in fire-dependent ecosystems, it is important to know how future wildfires may affect the landscape, and how much fire management may be necessary to maintain important habitat. We examined the future of wildfires and prescribed burning in longleaf pine ecosystems of the southeastern US. These ecosystems, when frequently burned, provide vital habitat for plants and animals, but recent fire suppression has severely decreased the amount of fire-maintained longleaf pine habitat. Future conservation of the ecosystem will depend in part on how much fire the ecosystem experiences. Therefore, it is important to understand potential wildfire dynamics, especially under climate change, as well as the potential for management actions like prescribed burning to restore and maintain fire-dependent longleaf habitat. For ecoregions in the coastal plain of Georgia, Alabama and Florida, we asked: (1) How will the effect of climate change on the fire regime impact longleaf vegetation dynamics in the absence of prescribed burning? (2) Given climate change, how much fire-dependent habitat will be restored under plausible future scenarios of prescribed burning? To simulate the future wildfire regime, we developed an empirical relationship between recent wildfires and climate, and projected that relationship to the year 2100 under climate change. We accounted for uncertainty in climate projections by incorporating Bayesian Model Averaging of climate model output. We used a spatially-explicit vegetation dynamics model to simulate the landscape effects of the future fire regime. We then simulated the effects of three potential prescribed burning scenarios: one that emulated recent levels of burning, as well as an increase and a decrease in burning. 

Results/Conclusions

On average, model results show that under climate change without prescribed burning, there will be only a slight increase in fire, and fire-maintained longleaf habitat will increase by less than 1% by 2100. Under a doubling of prescribed burning over recent levels, with climate change, the extent of fire-maintained habitat will increase, but will still be substantially smaller than historic amounts. These results suggest that the effects of climate change on the fire regime will be minor compared to anthropogenic fire suppression, and that if prescribed burning increases, it will likely have a greater effect on the ecosystem than climate change. These results point to a critical role of prescribed fire and thus provide important information for developing conservation strategies for the longleaf pine ecosystem.