COS 20-3
Aridity and plant functional type control soil carbon pools across an Australian aridity gradient

Tuesday, August 12, 2014: 8:40 AM
302/303, Sacramento Convention Center
Heather L. Throop, Biology Department, New Mexico State University, Las Cruces, NM
David J. Eldridge, Evolution and Ecology Research Centre, School of Biological, Earth, and Environmental Sciences, University of New South Wales, Sydney, NSW 2052, Australia
Michaela Buenemann, Geography Department, New Mexico State University, Las Cruces, NM
Keska Kemper, Cleveland High School, Portland, OR
Background/Question/Methods

Arid and semi-arid (‘dryland’) ecosystems cover about 40% of the earth’s terrestrial surface.  Although soil organic carbon concentration in these systems is typically low, they strongly influence the global carbon cycle as they account for nearly a third of global soil organic carbon and terrestrial net primary productivity. Large land cover changes in drylands have occurred globally over the past 150 years, with the potential for substantial impacts on the global carbon cycle. An enhanced understanding of the influence of these on-going land cover changes on carbon cycling is therefore critical.  The impacts of dryland land cover change on belowground carbon pools has long been difficult to characterize.  However, a recent synthesis suggested that the relative differences in soil organic carbon pools between woody vegetation and intercanopy spaces increase in a predictable manner with increasing aridity (defined here as a ratio of mean annual precipitation:mean annual temperature).  We tested this hypothesis in open woodlands across an aridity gradient in New South Wales, Australia.  We sampled soils for organic carbon and labile carbon content, comparing values from co-located trees, shrubs, and intercanopy herbaceous areas at 100 locations across the gradient. Remote sensing analyses were used to assess changes in aboveground vegetation. 

Results/Conclusions

The expected pattern of an increase in the relative influence of woody plant canopies on soil organic with increasing aridity (mean annual precipitation:mean annual temperature) was upheld.  However, this relationship was not as strong across this 1500 km aridity gradient as had been suggested by a recent global synthesis.  For soil organic carbon, the relationship was much greater for trees (r2 = 0.18) than it was for shrubs (r2 = 0.10).  Similar patterns were found for labile carbon (r2 = 0.21 for trees and r2 = 0.12 for shrubs).  Analyses of remote sensing data found strong increases in woody cover with decreasing aridity, although soil and landform differences also played a role. These data suggest that aridity plays a significant role in affecting the biogeochemical consequences of dryland land cover change.