B23C-0216:
What determines the magnitude of CO2 – land carbon feedback in ESMs?

Tuesday, 16 December 2014
Tomohiro Hajima1, Kaoru Tachiiri1, Akihiko Ito2 and Michio Kawamiya1, (1)JAMSTEC Japan Agency for Marine-Earth Science and Technology, Kanagawa, Japan, (2)CGER-NIES, Tsukuba, Japan
Abstract:
CO2 increase in the atmosphere stimulates plant growth, and hence promote carbon uptake by land ecosystems (so called “CO2 fertilization effect”). This process forms a negative feedback loop between atmospheric CO2 concentration and terrestrial carbon amount (CO2 – land carbon feedback). This feedback can have a major impact on climate projections with an uncertain magnitude. This research focuses on the CO2 – land carbon feedback, and analyze the mechanisms how the feedback magnitude is determined in earth system models (ESMs).

By examining the simulation results from ESMs, we confirmed that multiple ESMs driven by a common scenario show a large spread of the feedback strength among models. By analyzing the behavior of the carbon fluxes and pools of the models in detail, we found that the sensitivity of plant productivity to elevated CO2 is the dominant factor to determine the strength of the CO2–carbon feedback, although increasing CO2 stimulates other carbon cycle processes. This result suggest that intensive research for plant production response to increasing CO2 is very important for making precise projection for long-term climate change.

Simulations with a single ESM driven by different CO2 pathways demonstrated that carbon accumulation increases in scenarios with slower CO2 increase rates. Using both numerical and analytical approaches, we showed that the difference among CO2 scenarios is a time-lag of terrestrial carbon pools in response to atmospheric CO2 increase—a higher rate of CO2 increase results in smaller carbon accumulations than that in an equilibrium state of a given CO2 concentration. These results demonstrate that we cannot ignore the dependence of the CO2 – land carbon feedback on the carbon storage state, and suggest that the concentration feedback can be larger for future scenarios where the CO2 growth rate is reduced.