OOS 48-6
Water table depth and nitrogen availability affects productivity and water use in a savanna ecosystem

Friday, August 15, 2014: 9:50 AM
203, Sacramento Convention Center
Robert O. Teskey, Warnell School of Forestry and Natural Resources, University of Georgia, Athens, GA
Chelcy Ford, Coweeta Hydrologic Lab, USDA Forest Service, Otto, NC
Robert Mitchell, Joseph W. Jones Ecological Research Center, Newton, GA
Background/Question/Methods

There is a large variation in aboveground net primary productivity (ANPP) in longleaf pine-wiregrass savanna ecosystems in the southeastern US.  We investigated whether this variation was linked to depth to groundwater and availability of soil nitrogen.  We hypothesized that a) ANPP will increase with increases in  water availability, b) the effect of nitrogen addition on ANPP will depend on water availability and c) longleaf pine ANPP was more coupled with water table depth than ANPP of wiregrass.  The experiment was conducted at the Joseph Jones Ecological Research Center in southwestern Georgia on four 2 ha plots in mesic and xeric sites across the landscape.  At each site a 1 ha area served at the control and the other 1 ha was fertilized with 100 kg N/ha/yr of ammonium nitrate.  Soil moisture and depth to ground water was measured in each plot throughout the study.  ANPP was estimated by biomass increment and leaf litter collection.  Leaf litter was also used to estimate leaf area index. Transpiration was measured using thermal dissipation probes in five trees in each plot. A model was developed to estimate wiregrass transpiration.    

Results/Conclusions

The water table varied from <1m to 10 m depth among the sites.  Xeric sites also had lower soil moisture than mesic sites.  ANPP decreased linearly with increasing minimum annual water table depth.   Mean ANPP on mesic sites was 2.6 Mg/ha/yr compared with 1.5 Mg/ha/yr on the xeric sites.    Nitrogen addition increased ANPP by an average of 54% on both mesic and xeric sites.  Transpiration was three times greater on mesic than xeric sites, resulting in a 72% greater water use efficiency on the xeric sites.  Minimum water table depth and nitrogen availability explained 95% of the variation in ANPP among sites. Wiregrass contributed only a small portion to total ANPP, and its ANPP was not correlated with water table depth across sites. Our results emphasize that savanna ecosystems are co-limited by nutrient and water availability.  While water availability plays a critical role regulating ANPP, the positive response to nitrogen fertilizer across all sites indicates that they are all also nitrogen limited.