TY - JOUR
T1 - Reducing uncertainties in decadal variability of the global carbon budget with multiple datasets
AU - Li, Wei
AU - Ciais, Philippe
AU - Wang, Yilong
AU - Peng, Shushi
AU - Broquet, Grégoire
AU - Ballantyne, Ashley P.
AU - Canadell, Josep G.
AU - Cooper, Leila
AU - Friedlingstein, Pierre
AU - Le Quéré, Corinne
AU - Myneni, Ranga B.
AU - Peters, Glen P.
AU - Piao, Shilong
AU - Pongratz, Julia
N1 - Funding Information:
This paper builds on an analysis started by the late Michael R. Raupach and is a contribution to the work of the Global Carbon Project to understand and better constrain the human perturbation of the carbon budget. We thank Alessandro Anav for providing the data from CMIP5 and Peter Landschützer and Christian Rödenbeck for sharing their ocean flux data. We are grateful to the scientists from NOAA/ESRL and the Scripps O2 Program for making the invaluable data from the long-term measurements of CGR and O2/N2 available to the community. W.L. is supported by the European Commission-funded project LUC4C (603542). P.C. and S. Peng acknowledge support from the European Research Council through Synergy Grant ERC-2013-SyG-610028 "IMBALANCE-P." J.G.C. is supported by the Australian Climate Change Science Program, G.P.P. is supported by the Norwegian Research Council (236296), and J.P. is supported by the German Research Foundation's Emmy Noether Program.
PY - 2016/11/15
Y1 - 2016/11/15
N2 - Conventional calculations of the global carbon budget infer the land sink as a residual between emissions, atmospheric accumulation, and the ocean sink. Thus, the land sink accumulates the errors from the other flux terms and bears the largest uncertainty. Here, we present a Bayesian fusion approach that combines multiple observations in different carbon reservoirs to optimize the land (B) and ocean (O) carbon sinks, land use change emissions (L), and indirectly fossil fuel emissions (F) from 1980 to 2014. Compared with the conventional approach, Bayesian optimization decreases the uncertainties in B by 41% and in O by 46%. The L uncertainty decreases by 47%, whereas F uncertainty is marginally improved through the knowledge of natural fluxes. Both ocean and net land uptake (B + L) rates have positive trends of 29 ± 8 and 37 ± 17 Tg C·y-2 since 1980, respectively. Our Bayesian fusion of multiple observations reduces uncertainties, thereby allowing us to isolate important variability in global carbon cycle processes.
AB - Conventional calculations of the global carbon budget infer the land sink as a residual between emissions, atmospheric accumulation, and the ocean sink. Thus, the land sink accumulates the errors from the other flux terms and bears the largest uncertainty. Here, we present a Bayesian fusion approach that combines multiple observations in different carbon reservoirs to optimize the land (B) and ocean (O) carbon sinks, land use change emissions (L), and indirectly fossil fuel emissions (F) from 1980 to 2014. Compared with the conventional approach, Bayesian optimization decreases the uncertainties in B by 41% and in O by 46%. The L uncertainty decreases by 47%, whereas F uncertainty is marginally improved through the knowledge of natural fluxes. Both ocean and net land uptake (B + L) rates have positive trends of 29 ± 8 and 37 ± 17 Tg C·y-2 since 1980, respectively. Our Bayesian fusion of multiple observations reduces uncertainties, thereby allowing us to isolate important variability in global carbon cycle processes.
KW - Bayesian fusion
KW - Carbon cycle
KW - Decadal variations
KW - Global carbon budget
UR - http://www.scopus.com/inward/record.url?scp=84995653359&partnerID=8YFLogxK
U2 - 10.1073/pnas.1603956113
DO - 10.1073/pnas.1603956113
M3 - Article
AN - SCOPUS:84995653359
SN - 0027-8424
VL - 113
SP - 13104
EP - 13108
JO - Proceedings of the National Academy of Sciences of the United States of America
JF - Proceedings of the National Academy of Sciences of the United States of America
IS - 46
ER -
