The ocean plays a major role in the global carbon cycle, exchanging CO₂ with the overlying atmosphere. The uptake of atmospheric CO₂ by the oceans is driven by physicochemical processes as well as biological fixation of inorganic carbon species. The biogenic production of organic material and carbonate minerals in the surface ocean and their subsequent transport to depth are termed the "biological carbon pumps". Two different biological carbon pumps can be distinguished. Photosynthetic carbon fixation and the flux of organic matter to depth, termed organic carbon pump, generates a CO₂ sink in the ocean. In contrast, calcium carbonate production and its transport to depth, referred to as the carbonate pump, releases CO₂ in the surface layer. Therefore, the relative strengths of these two processes largely determine the biologically-mediated ocean-atmosphere CO₂ exchange.

At present an increase in atmospheric CO₂ concentrations is observed, which inevitably changes the seawater carbonate chemistry when more CO₂ is taken up by the surface ocean. If CO₂ concentrations keep rising at the present rate, it is expected that surface ocean CO₂ concentrations will have increased to 3-fold relative to preindustrial values by the end of this century. This would cause carbonate concentrations and pH to drop by ca. 50 % and 0.35 units, respectively.  Changes in the seawater carbonate chemistry of this magnitude are expected to affect the production of marine microalgae. One of the questions addressed by our group is whether rising atmospheric CO₂ concentrations affect primary production and calcification by marine phytoplankton. In laboratory experiments we observed that CO₂ -related changes in seawater carbonate chemistry strongly affect calcification and primary production of marine coccolithophorids.

In monospecific cultures of two dominant coccolithophorid species, Emiliania huxleyi and Gephyrocapsa oceanica, the ratio of calcification to photosynthetic carbon fixation decreased significantly with pCO₂  increasing from pre-industrial levels (270 ppmv) to values expected by the year 2100 (700 ppmv). This was both due to an increase in photosynthetic carbon fixation and a decrease in calcium carbonate production. In accord with the culture data, natural phytoplankton assemblages from the Subarctic North Pacific also showed severely reduced ratios of calcification to organic matter production at experimentally elevated CO₂ levels. These results suggest that changes in seawater carbonate chemistry expected to occur over the next century may slow down the production of calcium carbonate in the surface ocean and its subsequent transport to the deep sea. CO₂ -related changes in coccolith formation have profound implications for the oceanic carbon cycle, both in the future as well as in the geologic past, and may influence the ecology of calcareous phytoplankton. Model calculations indicate that a corresponding decrease in calcite flux would significantly increase the oceans capacity to store atmospheric CO₂, resulting in a negative feedback to the present rise in atmospheric CO₂. We also tested CO₂ effects on a natural bloom of Emiliania huxleyi in a mesocosm experiment. The experiment was carried out at the Large Scale Facilities in Bergen, Norway, in summer 2001 over a four week period. In 9 mesocosms of approximately 11 m³, each covered by gas-tight tents, atmospheric and seawater pCO₂ were manipulated. Three different CO₂ levels were achieved, i.e. 180 ppmv, 365 ppmv and 650 ppmV, resepectively corresponding to glacial, present and year 2100 atmospheric CO₂ concentrations.