Abstract
How tropical forest carbon stocks might alter in response to changes in climate and atmospheric composition is uncertain. However, assessing potential future carbon loss from tropical forests is important for evaluating the efficacy of programmes for reducing emissions from deforestation and degradation. Uncertainties are associated with different carbon stock responses in models with different representations of vegetation processes on the one hand, and differences in projected changes in temperature and precipitation patterns on the other hand. Here we present a systematic exploration of these sources of uncertainty, along with uncertainty arising from different emissions scenarios for all three main tropical forest regions: the Americas (that is, Amazonia and Central America), Africa and Asia. Using simulations with 22 climate models and the MOSES-TRIFFID land surface scheme, we find that only in one of the simulations are tropical forests projected to lose biomass by the end of the twenty-first century - and then only for the Americas. When comparing with alternative models of plant physiological processes, we find that the largest uncertainties are associated with plant physiological responses, and then with future emissions scenarios. Uncertainties from differences in the climate projections are significantly smaller. Despite the considerable uncertainties, we conclude that there is evidence of forest resilience for all three regions.
Original language | English |
---|---|
Pages (from-to) | 268-273 |
Number of pages | 6 |
Journal | Nature Geoscience |
Volume | 6 |
Issue number | 4 |
DOIs | |
State | Published - Apr 2013 |
Externally published | Yes |
Funding
O.K.A., P.M. and J.L. all acknowledge funding from the NERC-UK (NE/F002149/1 and NE/G008531) grants and the ARC-Australia (DP0986823 and DP1093759) grants. C.D.J., B.B.B.B., D.H., G.K., P.G. and R.B. acknowledge joint DECC and Defra Met Office Hadley Centre Climate Programme funding (Ref: DECC/Defra GA01101.) C.N. and J.M. acknowledge support from the Brazilian Research Council CNPq and the Sao Paulo State Research Foundation FAPESP (2008/58107-7). O.L.P. S.L.L. and Y.M. are supported by the European Research Council. S.L.L. acknowledges support from the Royal Society. C.H., P.Z. and L.M.M. thank the CEH Science Budget for support during this analysis. All authors gratefully recognize the many hundreds of people who have developed the climate models contributing to the CMIP3 database. C.H., L.M.M., S.S., S.L.L., E.G., O.L.P. and J.L. acknowledge the UK NERC QUEST, TROBIT and AMAZONICA (NE/F005806/1) initiatives. D.G. acknowledges support from the Moore Foundation.
Funders | Funder number |
---|---|
Brazilian Research Council CNPq | |
NERC-UK | NE/G008531, DP1093759, DP0986823 |
Sao Paulo State Research Foundation FAPESP | 2008/58107-7 |
Seventh Framework Programme | 291585, 282664 |
Henry Moore Foundation | |
Natural Environment Research Council | NE/F002149/1, NE/J010154/1, CEH010023, earth010002, NE/D01168X/1, NE/J011002/1, NE/F005806/1 |
Royal Society | |
European Research Council |