B13F-0695
Vegetation Responses to Future Climates: Global Variability in Water Use Efficiency and Primary Productivity in a CMIP5 Multimodel Ensemble

Monday, 14 December 2015
Poster Hall (Moscone South)
Sergio Bernardes, Organization Not Listed, Washington, DC, United States
Abstract:
Climate projections for the 21st century predict substantial changes in temperature and in the quantity and spatiotemporal distribution of precipitation for large regions on the planet. Reductions in water availability resulting from decreased precipitation and increased water demand by the atmosphere can negatively affect plant metabolism, reduce carbon uptake, and limit services of entire ecosystems. Future increases in temperature and in atmospheric carbon dioxide concentrations may help offset some of these impacts on vegetation. In particular, plants may adjust their water use efficiency (WUE, plant production per water loss by evapotranspiration) in response to changing environmental conditions. We assessed an ensemble of thirteen models from the Coupled Model Intercomparison Project 5 (CMIP5) and analyzed future changes in climate variables, carbon mass in vegetation, vegetation primary productivity and WUE. The analysis considered two representative concentration pathways (RCP4.5 and RCP8.5) for the period 2006-2099 and compared projections to historical values (1850–2005). Results include differences between historical and projected conditions for global, regional and latitudinal distributions of model outputs, for both RCPs. We observed significant intermodel variability when representing changes in WUE over the century, including high model sensitivity to different greenhouse concentration scenarios. Model agreement varied with RCP (higher agreement for RCP4.5), as well as regionally (higher agreement in Southeast Asia, lower agreement in arid areas, including the Sahara and Western Australia). The majority of models point to an increase in GPP and WUE for most of the planet under both concentration pathways, with changes ranging from zero to 100% of their historical values. These increases were observed to be consistently higher for RCP8.5. In addition, higher increases in GPP and WUE are predicted to occur over higher latitudes in the northern hemisphere (boreal areas), with WUE usually following GPP in changes. Modeling results show also decreases in productivity and WUE (from zero to -50%) mostly in the tropics, affecting tropical forests in Central America and in the Amazon.