Wednesday, August 8, 2007

PS 43-59: Fine-scale variation in CO2 exchange of coastal wet tundra in Alaska in response to hydrological gradients

Paulo Olivas, Steven F. Oberbauer, and Andrea Kuchy. Florida International University

As result of climate change, temperatures in the Arctic are expected to increase substantially.  These changes in temperature will increase ecosystem evapotranspiration and soil thaw above the permafrost, both lowering water tables.  Tundra plant communities are strongly tied to fine-scale variation in topography, and therefore, position relative to the water table.  Changes in water table have the potential to be the most important climate change factor for carbon exchange in coastal wet tundra in the near term.  Increased temperature and lower water tables will increase ecosystem respiration, reducing the capacity of tundra to sequester carbon dioxide.  To evaluate carbon balance of coastal tundra to wetting and drying, a NSF Biocomplexity of the Environment project is carrying out a large-scale hydrological manipulation in a drained thaw lake basin in Barrow, Alaska.  The basin is dominated by wet meadows, but there are also low-center polygons and pond habitats.  Here we report on fine-scale variation in chamber-level net CO2 and component fluxes in randomly stratified plots situated across the basin during the pretreatment growing seasons of 2005 and 2006.  With each flux measurement, we also assessed water table depth, thaw depth, and plot NDVI (Normalized Difference Vegetation Index).  Although no differences were found among plots in the lake basin for seasonal mean net CO2 fluxes, plots located at the top of polygon rims had higher gross ecosystem production.  However, these plots also had a significantly higher respiration compared to wetter sites.  Gross photosynthesis was significantly correlated with NDVI.  Differences in NDVI were more apparent than differences in components of CO2 exchange among the different basin plant communities.  Areas with higher moss cover and drier conditions tended to have significantly higher NDVI values than sites with higher graminoid cover and wetter conditions.