Underestimation of oceanic carbon uptake in the Arctic Ocean: ice melt as predictor of the sea ice carbon pump

<p>The Arctic Ocean is generally undersaturated in CO<span class="inline-formula">₂</span> and acts as a net sink of atmospheric CO<span class="inline-formula">₂</span>. This oceanic uptake is strongly modulated by sea ice, which can prevent air–sea gas exchange and has major impacts on stratification and primary production. Moreover, carbon is stored in sea ice with a ratio of alkalinity to dissolved inorganic carbon that is larger than in seawater. It has been suggested that this storage amplifies the seasonal cycle of seawater <span class="inline-formula"><i>p</i></span>CO<span class="inline-formula">₂</span> and leads to an increase in oceanic carbon uptake in seasonally ice-covered regions compared to those that are ice-free. Given the rapidly changing ice scape in the Arctic Ocean, a better understanding of the link between the seasonal cycle of sea ice and oceanic uptake of CO<span class="inline-formula">₂</span> is needed. Here, we investigate how the storage of carbon in sea ice affects the air–sea CO<span class="inline-formula">₂</span> flux and quantify its dependence on the ratio of alkalinity to inorganic carbon in ice. To this end, we present two independent approaches: a theoretical framework that provides an analytical expression of the amplification of carbon uptake in seasonally ice-covered oceans and a simple parameterization of carbon storage in sea ice implemented in a 1D physical–biogeochemical ocean model. Sensitivity simulations show a linear relation between ice melt and the amplification of seasonal carbon uptake. A 30 % increase in carbon uptake in the Arctic Ocean is estimated compared to ice melt without amplification. Applying this relationship to different future scenarios from an earth system model that does not account for the effect of carbon storage in sea ice suggests that Arctic Ocean carbon uptake is underestimated by 5 % to 15 % in these simulations.</p>