Tuesday, August 5, 2008 - 9:50 AM

COS 24-6: A multiple-scale analysis of air temperature and monsoonal precipitation as controls over CO2 fluxes across a woody encroachment gradient: Implications for an interaction Between climatic and vegetative cover change

Greg A. Barron-Gafford1, Russell Scott2, G. Darrel Jenerette3, Anna P. Tyler1, and Travis E. Huxman1. (1) University of Arizona, (2) United States Department of Agriculture, Agricultural Research Service, (3) University of California Riverside

Background/Question/Methods

The physiognomic shift in ecosystem structure from a grassland to a woodland may alter the sensitivity of CO2 exchange of both dominant plants and the entire ecosystem to variations in growing-season air temperature and precipitation inputs. Previous work has shown that deeper-rooting mesquite (Prosopis velutina) trees and shrubs have greater access to groundwater than sacaton bunchgrass (Sporobolus spp.), and this access results in CO2 exchange becoming relatively decoupled from variation in monsoonal precipitation. We used a combination of leaf-level measures of gas exchange and ecosystem-level eddy covariance techniques to quantify the responsiveness of a grassland, grassland-shrubland mosaic, and a woodland to changes in air temperature and the onset of the dominant period of precipitation input. Measurements were conducted prior to the onset of the monsoon, a few weeks into the monsoon, midway through this rainy period, and after the monsoon rains had ended for the year. Temperature optima were determined by fitting curves to the net CO2 uptake data as a function of air temperature.

Results/Conclusions

Throughout the entire growing season, light-saturated net CO2 exchange exhibited an optimum within the range of 26 -31oC for the mesquites sampled at all sites. The mesquites within the woodland site had the narrowest range of temperature optima, and the shape of the response curves varied little in response to the onset of monsoonal precipitation. Rates of net CO2 assimilation dropped significantly as temperatures derivated from the calculated optima for the more recently encroached mesquites found within the grassland and mosaic sites. Temperature optima were generally greater in the sacaton grasses than the mesquites found within each site, and sacaton temperature optima were much more plastic in response to precipitation input. Responses of net CO2 exchange to inputs of monsoonal rainfall therefore depend on vegetation composition and the connection of dominant plants to the water table. Our analysis revealed that the woodland site, which has undergone the greatest level of woody encroachment, are optimally suited for net CO2 assimilation within a relatively small range of “optima”, but continue to assimilate high levels of CO2 throughout the growing season. Only when growing season temperatures exceeded 44oC did rates of net CO2 assimilation drop to 80% of their maximal rates. This high capacity for CO2 assimilation across a range of temperatures may have implications under projected patterns of climate change for the arid southwest which include increased as increased atmospheric temperatures and altered precipitation patterns.