Environ. Res. Lett. 7, 024002 (2012)

Key sources of uncertainty in predicting the level of expected global warming stem from difficulties in predicting greenhouse-gas emissions and the sensitivity of the climate to changes in atmospheric greenhouse-gas concentrations. The additional uncertainty in relating carbon emissions to atmospheric carbon dioxide concentrations — resulting from incomplete knowledge of the carbon cycle — is typically estimated to be smaller than those in physical climate feedbacks and emissions.

Ben Booth, from the Met Office Hadley Centre, UK, and co-workers used a fully coupled climate–carbon cycle model to systematically explore the magnitude of the various sources of uncertainty in the land carbon-cycle feedback.

The simulations indicate that the plausible range of climate–carbon-cycle feedbacks might be significantly larger than previously estimated. The range of carbon dioxide concentrations (669–1130 parts per million by the year 2100) that could arise from a single emissions scenario was estimated to be greater than conventionally estimated across the full range of different Intergovernmental Panel on Climate Change emissions scenarios if carbon-cycle feedbacks are ignored. The authors note that this reveals an urgent need for better understanding of plant photosynthetic responses to warming, as these are shown to be key contributors to the magnitude of future carbon dioxide change.