12:00 PM - 12:15 PM
[ACG34-12] Trend in Southern Ocean CO2 sink – a review and outlook
Keywords:Southern Ocean CO2 flux, Inverse modelling, Global ocean CO2 flux models
The Southern Ocean (south of 14oS) has played significant role in removing carbon dioxide (CO2) from the atmosphere in the past decades. There several studies which discussed a slowing down of CO2 uptake rate or even an opposite trend in CO2 uptake since the late 1980s through the early 2000s, in contrast to that is expected from the pCO2 increase in the atmosphere and the ocean biogeochemical models. The JAMSTEC’s inverse model, based on MIROC4-ACTM forward simulations, climatological sea-air CO2 flux and atmospheric CO2 observations, shows more or less systematic increase in sink uptake since the late 1980s to the present. We suspect there are issues with data coverage with time that is posing challenges in the pCO2 observation based upscaling products; and for the global ocean biogeochemical models difficulty in the simulation of seasonal cycle of sea-air pCO2 difference may inadvertently lead to a wrong climate sensitivity.
In this presentation, we will review some of these issues and possibly come up with a revised assessment of the trends in Southern Ocean CO2 flux.
The figure below shows the CO2 flux anomalies calculated by 6 cases of MIROC4-ACTM inversions, by varying the prior flux model (flxUnc vs GPPunc) or network of CO2 observations (S34, S22, S22+ship). Although the magnitude of flux anomalies vary between the inversions, the long-term trends and phase of the variability are consistently derived for the overlapping inversion periods.
In this presentation, we will review some of these issues and possibly come up with a revised assessment of the trends in Southern Ocean CO2 flux.
The figure below shows the CO2 flux anomalies calculated by 6 cases of MIROC4-ACTM inversions, by varying the prior flux model (flxUnc vs GPPunc) or network of CO2 observations (S34, S22, S22+ship). Although the magnitude of flux anomalies vary between the inversions, the long-term trends and phase of the variability are consistently derived for the overlapping inversion periods.