The role of N₂-fixation to simulate the pCO₂ observations from the Baltic Sea

Abstract. Measurements in the central Baltic Sea have shown that dissolved inorganic carbon (DIC) concentrations in the upper water column continue to decrease even after complete depletion of dissolved inorganic nitrogen (DIN). To explain this observation, a new external supply of nitrogen is required without a concomitant supply of inorganic carbon. The primary proposed candidate process is N₂-fixation. In order to address this question in the eastern Gotland Sea, a biogeochemical model containing nine state variables including diatoms, flagellates, and nitrogen fixing cyanobacteria, was coupled to a 1D physical model. The results from the winter period until the onset of the spring bloom of 1997 gave reasonable surface water values for partial pressure of CO₂ (pCO₂) compared with measurements of the pCO₂. However, the model failed to simulate the observed pCO₂ drawdown for the period from the end of the spring bloom until late summer. Even after introducing a seasonal dissolved organic carbon (DOC) excess production and varying different process parameterization the simulated pCO₂ values did not improve. Only the shift from a sharp to a moderate temperature dependency in addition to an increase in the C:P ratio of the nitrogen fixing cyanobacteria made it possible for the model to match the pCO₂ observations. The resulting total nitrogen fixation (167 mmol m^-2 a^-1) exceeds previous measurement-based estimates but is in good agreement with recent rate measurement based estimates.