Phytoplankton biomass is often inferred from chlorophyll, however, since biogeochemistry in the ocean is coupled mostly to carbon, models need to convert chlorophyll to carbon biomass. The chlorophyll to carbon (Chl:C) ratio in phytoplankton is variable, and although there are species specific differences, most of the variability is driven by acclimation to changing nutrient and light-conditions. Many models include variable Chl:C ratios to account for the effects of photoacclimation, however results can be very different depending on the model used. Can we improve the way photoacclimation of phytoplankton to different light conditions is modelled? A diagnostic Chl:C ratio can describe properly the balanced growth of phytoplankton populations, but modelling separately the dynamics of carbon and chlorophyll could help to predict the variability observed matching situations of balanced and not balanced growth. Our current model, REcoM, is a global ecosystem model based on the phytoplankton growth model from Geider et al. (1998) which runs coupled to the MIT global circulation model. We compared the Chl:C ratios from that model with observations, showing that the model was able to represent some general patterns, but also faced some deficiencies. Within the project IPSO (Improving the prediction of photophysiology in the Southern Ocean by accounting for iron limitation, optical properties and spectral satellite data information) the modelling group of the Alfred Wegener Institute aims to explore two ways for refining the model. First, we will compare several phytoplankton growth models that treat differently the chlorophyll synthesis term with in situ data of Chl:C ratios gathered at local scale. In a next step, we will extend this analysis to compare modelling options with satellite chlorophyll data. The improvement of photophysiology description in the model is expected to provide a more accurate estimation of chlorophyll content on a global basis.