Modifications in the strength of the biological pump as a consequence of ocean acidification, whether positive or negative, have the potential to impact atmospheric CO2 and therefore climate. So far, most plankton community perturbation studies have been performed in nutrient-rich areas although there are some indications that CO2-dependent growth could differ in nutrient-replete vs. -limited regions and with different community compositions. Two in situ mesocosm experiments were performed in the NW Mediterranean Sea during two seasons with contrasted environmental conditions: summer oligotrophic stratified waters in the Bay of Calvi vs. winter mesotrophic well-mixed waters in the Bay of Villefranche. Nine mesocosms were deployed for 20 and 12 d, respectively, and subjected to seven CO2 levels (3 controls, 6 elevated levels). Both phytoplankton assemblages were dominated by pico- and nano-phytoplankton cells. Although haptophyceae and dinoflagellates benefited from short-term CO2 enrichment in summer, their response remained small with no consequences on organic matter export due to strong environmental constraints (nutrient availability). In winter, most of the plankton growth and associated nutrient consumption occurred during the 4-day acidification period (before the experimental phase). During the remaining experimental period, characterized by low nutrient availability, plankton growth was minimal and no clear CO2-dependency was found for any of the tested parameters. While there is a strong confidence on the absence of significant effect of short-term CO2 addition under oligotrophic conditions, more investigations are needed to assess the response of plankton communities in winter when vertical mixing and weather conditions are major factors controlling plankton dynamics.