Can terrestrial biosphere models capture the response of atmospheric CO2 growth rate to ENSO?
Yuanyuan Fang1,2, Anna M Michalak2,3, Christopher Schwalm4, Deborah N Huntzinger5, Yaxing Wei6, Robert B Cook6, Kevin M Schaefer7, Andrew R Jacobson8, Philippe Ciais9, Joshua Fisher10, Daniel J Hayes11, Maoyi Huang12, Dr. Akihiko Ito13, Atul K Jain14, Huimin Lei15, Chaoqun Lu16, Fabienne Maignan17, Jiafu Mao18, Nick Parazoo19, Shushi Peng17, Benjamin Poulter20, Daniel M Ricciuto18, Xiaoying Shi11, Hanqin Tian21, Ning Zeng22, Fang Zhao22, Weile Wang23 and MsTMIP modeling team, (1)Stanford University, Stanford, CA, United States, (2)Carnegie Institution for Science, Department of Global Ecology, Washington, DC, United States, (3)Stanford University, Stanford, United States, (4)Northern Arizona University, Flagstaff, AZ, United States, (5)Northern Arizona University, School of Earth Sciences and Environmental Sustainability, Flagstaff, AZ, United States, (6)Oak Ridge National Laboratory, Oak Ridge, TN, United States, (7)National Snow and Ice Data Center, Cooperative Institute for Research in the Environmental Sciences, University of Colorado at Boulder, Boulder, Colorado, U.S.A, Boulder, United States, (8)University of Colorado at Boulder, CIRES, Boulder, United States, (9)LSCE Laboratoire des Sciences du Climat et de l'Environnement, Gif-Sur-Yvette, France, (10)Jet Propulsion Laboratory, California Institute of Technology, Pasadena, United States, (11)Oak Ridge National Laboratory, Oak Ridge, United States, (12)National Weather Service, National Oceanic and Atmospheric Administration, Office of Science and Technology Integration, Silver Spring, MD, United States, (13)The University of Tokyo, Graduate School of Agricultural and Life Sciences, Tokyo, Japan, (14)University of Illinois at Urbana Champaign, Urbana, United States, (15)Pacific NW Nat'l Lab-Atmos Sci, Richland, WA, United States, (16)Auburn University at Montgomery, Montgomery, AL, United States, (17)CEA Saclay DSM / LSCE, Gif sur Yvette, France, (18)Oak Ridge National Laboratory, Environmental Sciences Division, Oak Ridge, TN, United States, (19)NASA Jet Propulsion Laboratory, California Institute of Technology, Pasadena, CA, United States, (20)Spark Climate Solutions, Greenbelt, United States, (21)Auburn University, Auburn, United States, (22)University of Maryland, College Park, MD, United States, (23)CSUMB & NASA/AMES, Seaside, CA, United States
Abstract:
Previous studies have highlighted ENSO as a key driver of the interannual variability of atmospheric CO
2 growth rate (AGR) through its influence on the biospheric carbon cycle. The biophysical mechanisms leading to this influence remain unclear, however. Understanding and correctly representing those mechanisms would provide crucial diagnostic tools to improve predictions of future changes to the global carbon cycle.
Here we analyze the correlation between annual AGR and the Nino 3.4 index during 1959-2010 to elucidate the response of the biospheric carbon cycle to ENSO. We further compare these results with the responses implied by 11 process-based models participating the Multi-scale Synthesis and Terrestrial Model Intercomparison project (MsTMIP). We find that the annual AGR is strongly correlated with the ENSO index during the preceding September to February, with stronger land CO2 sources following stronger El Nino signals. This response results from teleconnections between tropical temperatures and ENSO, as well as from the influence of tropical temperatures on the biospheric carbon cycle. MsTMIP models capture this correlation, but overestimate it. This is due to an unrealistically high sensitivity of simulated NEE to tropical precipitation. In particular, the response of AGR to ENSO becomes asymmetric under positive and negative phases of ENSO, with their correlation with ENSO index peaking at different times for post-El Nino and post-La Nina years. This asymmetric response is not captured by models, and the simulated responses for post-El Nino years are highly inconsistent across models as well as between models and AGR. Models therefore appear to have problems in simulating the biophysical mechanisms after El Nino years, mechanisms that are likely associated with anomalously dry conditions. As stronger and more frequent El Nino events are projected under climate change, these results suggest that model response to ENSO variability needs to be improved in order to robustly predict future climate-carbon interaction.
