PS 65-87
Effects of repeated fire and environmental variation on the efflux of CO2 from soils of a subtropical woodland savanna

Friday, August 15, 2014
Exhibit Hall, Sacramento Convention Center
Susanne Wiesner, Biological Sciences, University of Alabama, Tuscaloosa, AL
Andres Baron, Forest Ecology Laboratory, Joseph Jones Ecological Research Center, Newton, GA
Christina L. Staudhammer, Biological Sciences, University of Alabama, Tuscaloosa, AL
Gregory Starr, Biological Sciences, University of Alabama, Tuscaloosa, AL
Background/Question/Methods

Subtropical longleaf pine (Pinus palustris) savannas are important ecosystems, in terms of their biodiversity (> 40 species per m2) as well as ecosystem services, such as carbon sequestration. The structure and function of these ecosystems are maintained with frequent low intensity fires. However, the role that cyclic fire plays in soil respiration is relatively unknown. Soil respiration is a major component in the global carbon cycle and greatly dependent on air temperature and soil moisture. Approximately half of the soil respiration is attributed to metabolic activity in plants, supporting root growth.

We collected soil respiration measurements in longleaf pine savannas in southwestern Georgia, USA, over a four-year period, to increase our understanding of carbon dynamics in these ecosystems. The study included monthly observations of soil respiration (Fsoil), air temperature (Tair) and soil moisture (VWC), over three fire cycles, at three sites with differing water holding capacities from 2009 to 2013. We used mixed model approaches to answer the following research questions: 1) How does Fsoil contribute to the net ecosystem exchange (NEE) of longleaf pine savannas? 2) How does VWC affect Fsoil in longleaf pine ecosystems? 3) Does seasonal fire alter Fsoil rates at the three sites?

Results/Conclusions

Our results show that Fsoil accounts for approximately 60% of NEE at all sites. All sites vary significantly in terms of VWC, ranging on average from 7% at the xeric site to 11% at the mesic site, with higher variability at the mesic and lower variation at the xeric site. Our mixed model indicated that Fsoil was significantly different by site and year, and varied with VWC and Tair (α = 0.05). We found that lower VWC resulted in higher Fsoil, and that Fsoil was higher following drought years (2012). However, lower VWC at the xeric site did not cause higher Fsoil rates, in contrast to the mesic site. Whereas Tair was positively correlated with Fsoil, VWC showed negative correlation with Fsoil. We did not find significant effects of seasonal burn on Fsoil. We conclude that Fsoil largely depends on the prevailing weather conditions and soil texture. Hence, drier soil conditions as one consequence of global climate change may lead to increased Fsoil in these ecosystems and therefore alter NEE.