Improving Current Assessments and Future Predictions of Carbon Fluxes in the Southern Ocean as Mediated by the Dynamical Response of Ice-Ocean-Ecosystem Interactions to Climate Change

Sharon Stammerjohn, LDEO, sharons@ldeo.columbia.edu
Chris Fritsen, DRI, cfritsen@dri.edu
Langdon Quetin, UCSB, langdon@icess.ucsb.edu
Robin Ross, UCSB, robin@icess.ucsb.edu
Colm Sweeney, NOAA/UC, colm.sweeney@noaa.gov (Presenting)
Maria Vernet, UCSD, mvernet@ucsd.edu

The Southern Ocean (which is 10% of the Earth’s total ocean area) is estimated to be responsible for up to 20% of the global ocean CO2 uptake of 2.2 Pg C yr-1 (Takahashi et al. 2002; Takahashi et al, in prep). However, a roughly two-fold uncertainty exists in the estimated CO2 uptake, and the uncertainty depends largely on whether polar continental shelves covered by seasonal sea ice act as one-way CO2 pumps: in winter out-gassing of CO2-rich water is prevented by the overlying sea ice cover, while in spring-summer rapid onsets of phytoplankton blooms at the ice-edge provide a CO2 sink. The extent to which phytoplankton blooms will provide a CO2 sink is uniquely dependent on downward carbon export, i.e., the biological pump. To improve current assessments and future predictions of carbon fluxes in Southern Ocean sea ice zones, we will investigate the highly seasonally (and interannually) varying dynamic relationships between sea ice, ocean and the ecosystem using 15 years of data from the western Antarctic Peninsula (wAP) region. We are focused on the wAP region for two reasons: (1) it has been shown to be warming faster in winter than anywhere else on earth and (2) remote and in situ physical, biogeochemical and ecosystem data in this region are more extensive than anywhere else in the Antarctic/Southern Ocean region.



Our approach will involve building a predictive ice-ocean-ecosystem model through careful data synthesis and model testing over contemporary periods (the last 15-28 years). The predictive model seeks to test the hypothesis that the rectification of air-sea gas exchange has and will continue to decrease due to the combination of reductions in sea ice concentration and grazer populations capable of transporting carbon to the deep ocean. The ecosystem component in the model was designed for global predictions and will allow us to easily incorporate our improvements into global-scale modeling efforts. Two ecosystem processes that we will include in our investigation that have not been well-parameterized or studied in previous ecosystem data synthesis and modeling efforts include (1) the role of sea ice biota in enhancing spring phytoplankton concentrations and CO2 draw down during sea ice melt, and (2) the role of macrozooplankton grazing in influencing phytoplankton community composition and export production.