Nonlinearity of ocean carbon cycle feedbacks in CMIP5 Earth System Models

Carbon cycle feedbacks are usually categorized into carbon-concentration and carbon-climate feedbacks, which arise owing to increasing atmospheric CO2 concentration and changing physical climate. Both feedbacks are often assumed to operate independently: that is, the total feedback can be expressed...

Full description

Bibliographic Details
Published in:Journal of Climate
Main Authors: Schwinger, J., Tjiputra, J., Heinze, C., Bopp, L., Christian, J., Gehlen, M., Ilyina, T., Jones, C., Salas-Melia, D., Segschneider, J., Seferian, R., Totterdell, I.
Format: Article in Journal/Newspaper
Language:English
Published: 2014
Subjects:
Online Access:http://hdl.handle.net/11858/00-001M-0000-0019-D8D9-0
http://hdl.handle.net/11858/00-001M-0000-0019-D8EC-8
Description
Summary:Carbon cycle feedbacks are usually categorized into carbon-concentration and carbon-climate feedbacks, which arise owing to increasing atmospheric CO2 concentration and changing physical climate. Both feedbacks are often assumed to operate independently: that is, the total feedback can be expressed as the sum of two independent carbon fluxes that are functions of atmospheric CO2 and climate change, respectively. For phase 5 of the Coupled Model Intercomparison Project (CMIP5), radiatively and biogeochemically coupled simulations have been undertaken to better understand carbon cycle feedback processes. Results show that the sum of total ocean carbon uptake in the radiatively and biogeochemically coupled experiments is consistently larger by 19-58 petagrams of carbon (Pg C) than the uptake found in the fully coupled model runs. This nonlinearity is small compared to the total ocean carbon uptake (533-676 Pg C), but it is of the same order as the carbon-climate feedback. The weakening of ocean circulation and mixing with climate change makes the largest contribution to the nonlinear carbon cycle response since carbon transport to depth is suppressed in the fully relative to the biogeochemically coupled simulations, while the radiatively coupled experiment mainly measures the loss of near-surface carbon owing to warming of the ocean. Sea ice retreat and seawater carbon chemistry contribute less to the simulated nonlinearity. The authors' results indicate that estimates of the ocean carbon-climate feedback derived from "warming only" (radiatively coupled) simulations may underestimate the reduction of ocean carbon uptake in a warm climate high CO2 world.