14C.3 – Physiological Responses of Tropical Forests to Rising Atmospheric CO2 as a Driver of Continental Asymmetries in Precipitation Change and Runoff Extremes


James T. Randerson
University of California Irvine
Gabriel J. Kooperman
University of California Irvine
Baird Langenbrunner
University of California Irvine
Michael S. Pritchard
University of California Irvine
Abigail L. S. Swann
University of Washington
Forrest M. Hoffman (forrest at climatemodeling dot org)
Oak Ridge National Laboratory
Keith Lindsay
National Center for Atmospheric Research
Charles D. Koven
Lawrence Berkeley National Laboratory


Session 14C Piers Sellers’ Legacy – Part I
Thursday, January 11, 2018 11:00 a.m.–11:15 a.m.
Salon J (Hilton) Austin, Texas


Changes in tropical precipitation during the latter half of the 21st century are likely to have profound effects on ecosystem function, vegetation dynamics, food security, and flood risk. All of the earth system models participating in the Coupled Model Intercomparison Project Phase 5 (CMIP5) show a strengthening precipitation asymmetry across different tropical continents by 2100 for the Representation Concentration Pathway 8.5. In the models, precipitation increases across equatorial Asia are offset by neutral or declining trends in lowland forests across South America. Analysis of a subset of these models used for carbon cycle feedback evaluation reveals that the direct effects of rising atmospheric CO2 on plant physiology is a dominant contributor to the continental asymmetry in the precipitation response. Specifically, the coupling of photosynthesis and stomatal conductance yields a response to rising CO2 that considerably modifies surface energy fluxes—reducing latent heating and increasing sensible heating. This coupling and its impact on surface air temperatures were first described in work led by Piers Sellers using the Simple Biosphere Model coupled with the Colorado State University climate model. While the influence of this coupling on precipitation has been explored in past work, an important outstanding research question is to understand why plant physiological responses to rising CO2 induce contrasting impacts on precipitation across different continents. Using a series of fully-coupled Community Earth System Model simulations, we investigated this question by confining the the effects of rising CO2 to forests on different continents. We find that distinctly different regional circulation and moisture flux anomalies arise on each continent in response to mostly similar declines in stomatal conductance and transpiration. Moisture convergence remains about the same in the Amazon in response to rising CO2 and is not enough to compensate for decreases in evapotranspiration and precipitation recycling. In contrast, moisture convergence increases considerably across the Maritime Continent, yielding higher precipitation rates over land. These experiments indicate that dynamical responses to rising CO2 may contribute to both positive and negative feedbacks with the carbon cycle, and that the sum of local responses mostly explains the pan-tropical asymmetry in precipitation change. Exploring downstream changes in the hydrological cycle, our analysis indicates that plant physiological processes are a dominant contributor to intensifying runoff extremes, whereas the radiative effects of CO2 are more important for precipitation extremes. More broadly, our analysis suggests that forests in South America may be more vulnerable to rising CO2 than forests in Asia or Africa.

Forrest M. Hoffman (forrest at climatemodeling dot org)