Long-term surface pCO(2) trends from observations and models

• Published in: Tellus B: Chemical and Physical Meteorology
• Main Authors: Tjiputra, Jerry F., Olsen, Are, Bopp, Laurent, Lenton, Andrew, Pfeil, Benjamin, Roy, Tilla, Segschneider, Joachim, Totterdell, Ian, Heinze, Christoph
• Format: Article in Journal/Newspaper
• Language: English
• Published: Co-action Publishing 2014
• Subjects: surface pCO(2); ocean CO2 sinks; Earth system models; CMIP5 projections; ocean biogeochemistry; Southern Ocean; Pacific; North Atlantic
• Online Access: https://archimer.ifremer.fr/doc/00291/40228/38691.pdf; https://archimer.ifremer.fr/doc/00291/40228/38692.pdf; https://doi.org/10.3402/tellusb.v66.23083; https://archimer.ifremer.fr/doc/00291/40228/

We estimate regional long-term surface ocean pCO(2) growth rates using all available underway and bottled biogeochemistry data collected over the past four decades. These observed regional trends are compared with those simulated by five state-of-the-art Earth system models over the historical period. Oceanic pCO(2) growth rates faster than the atmospheric growth rates indicate decreasing atmospheric CO2 uptake, while ocean pCO(2) growth rates slower than the atmospheric growth rates indicate increasing atmospheric CO2 uptake. Aside from the western subpolar North Pacific and the subtropical North Atlantic, our analysis indicates that the current observation-based basin-scale trends may be underestimated, indicating that more observations are needed to determine the trends in these regions. Encouragingly, good agreement between the simulated and observed pCO(2) trends is found when the simulated fields are subsampled with the observational coverage. In agreement with observations, we see that the simulated pCO(2) trends are primarily associated with the increase in surface dissolved inorganic carbon (DIC) associated with atmospheric carbon uptake, and in part by warming of the sea surface. Under the RCP8.5 future scenario, DIC continues to be the dominant driver of pCO(2) trends, with little change in the relative contribution of SST. However, the changes in the hydrological cycle play an increasingly important role. For the contemporary (1970-2011) period, the simulated regional pCO(2) trends are lower than the atmospheric growth rate over 90% of the ocean. However, by year 2100 more than 40% of the surface ocean area has a higher oceanic pCO(2) trend than the atmosphere, implying a reduction in the atmospheric CO2 uptake rate. The fastest pCO(2) growth rates are projected for the subpolar North Atlantic, while the high-latitude Southern Ocean and eastern equatorial Pacific have the weakest growth rates, remaining below the atmospheric pCO(2) growth rate. Our work also highlights the importance and need for a sustained long-term observing strategy to continue monitoring the change in the ocean anthropogenic CO2 sink and to better understand the potential carbon cycle feedbacks to climate that could arise from it.