B23J-01 – Diagnosing drought in a changing climate (Invited)


Abigail Swann
University of Washington Seattle
Forrest Hoffman (forrest at climatemodeling dot org)
Oak Ridge National Laboratory
Charles Koven
Lawrence Berkeley National Laboratory
James Randerson
University of California Irvine


New Mechanisms, Feedbacks, and Approaches for Improving Predictions of the Global Carbon Cycle in Earth System Models II
Tuesday, December 15, 2015 13:40–13:55
Moscone West 2010


Predictions of future climate impacts such as drought rely heavily on metrics based on changes in rainfall and changes in the demand for water from the atmosphere. However, the underlying driver of climate change is the increasing concentration of atmospheric carbon dioxide (CO2), which simultaneously increases temperature globally and modifies the water needs of plants. Although the influence of CO2 on plant stomatal conductance and transpiration is well established, the relative impact of this physiology on different drought metrics has not been rigorously assessed. We find that predictions of increasing drought stress derived using atmospheric demand metrics in many regions (including potential evapotraspiration and Palmer Drought Stress Index) correspond to places where Earth system models show stable or increasing water availability on land when assessed using the difference between precipitation and evapotranspiration. Approximately 70% of the increase in global water availability is a direct result of the effect of CO2 reducing plant water needs. Current models predict a decoupling of water flux and carbon flux, which require revisions to how aridity is measured and drought is calculated under changing CO2.

Forrest M. Hoffman (forrest at climatemodeling dot org)