B24B-04
The Carbon Balance Pivot Point of Southwestern U.S. Semiarid Ecosystems: Insights From the 21st Century Drought

Tuesday, 15 December 2015: 16:45
2004 (Moscone West)
Russell L Scott1, Joel Biederman1, Greg Barron-Gafford2 and Erik P Hamerlynck3, (1)Agricultural Research Service Tucson, Tucson, AZ, United States, (2)University of Arizona, Tucson, AZ, United States, (3)USDA-ARS, Tucson, AZ, United States
Abstract:
Global-scale studies indicate that semiarid ecosystems strongly regulate the long-term trend and interannual variability of the terrestrial carbon sink, possibly due to changes in vegetation and an inherent sensitivity to changes in water availability. However, we lack understanding of how climate shifts, such as the ongoing decadal-scale drought in the Southwest US, impact carbon sink functioning in semiarid ecosystems with differing structure. Therefore, we investigated the response of net ecosystem production of carbon dioxide (NEP) to changes in annual water availability in four Southwest US ecosystems varying in relative shrub, tree and grass abundance. Using eddy covariance carbon dioxide and water vapor flux measurements collected over the last drought-impacted decade, we identified a precipitation “pivot point” in the annual carbon balance for each ecosystem type where annual NEP switched from negative to positive. At the three sites with larger amounts of grass, rather than woody plant, cover, pivot points were closer to the drought-period mean annual precipitation (MAP) than MAP over the preceding 30 years, suggesting the carbon pools of these grassier ecosystems have more quickly adjusted to the decadal-scale drought. Current-year water availability, as quantified by evapotranspiration (ET) overwhelmingly drove the response of gross ecosystem photosynthesis (GEP) and respiration (Reco) fluxes. Ecosystem water use efficiency (GEP/ET) increased with water availability and leaf area index, resulting in a more efficient photosynthetic use of water in wetter years and at wetter sites. Grasslands supported a higher leaf area than shrublands at a given water availability, and thus had higher GEP/ET. Differences in GEP/ET were also related to the relative proportion of abiotic evaporation, estimated from the ET intercept in a linear regression of ET and GEP, to total ET at a site, highlighting the importance of ET partitioning for understanding how semiarid rainfall drives plant productivity.