p CO2 effects on species composition and growth of an estuarine phytoplankton community.

The effects of ongoing changes in ocean carbonate chemistry on plankton ecology have important implications for food webs and biogeochemical cycling. However, conflicting results have emerged regarding species-specific responses to CO2 enrichment and thus community responses have been difficult to predict. To assess community level effects (e.g., production) of altered carbonate chemistry, studies are needed that capitalize on the benefits of controlled experiments but also retain features of intact ecosystems that may exacerbate or ameliorate the effects observed in single-species or single cohort experiments. We performed incubations of natural plankton communities from Narragansett Bay, RI, USA in winter at ambient bay temperatures (5-13 °C), light and nutrient concentrations under three levels of controlled and constant CO concentrations, simulating past, present and future conditions at mean CO levels of 224, 361, and 724 μatm respectively. Samples for carbonate analysis, chlorophyll , plankton size-abundance, and plankton species composition were collected daily and phytoplankton growth rates in three different size fractions (<5, 5-20, and >20 μm) were measured at the end of the 7-day incubation period. Community composition changed during the incubation period with major increases in relative diatom abundance, which were similar across CO treatments. At the end of the experiment, 24-hr growth responses to CO levels varied as a function of cell size. The smallest size fraction (<5 μm) grew faster at the elevated CO level. In contrast, the 5-20 μm size fraction grew fastest in the Present treatment and there were no significant differences in growth rate among treatments in the > 20 μm size fraction. Cell size distribution shifted toward smaller cells in both the Past and Future treatments but remained unchanged in the Present treatment. Similarity in Past and Future treatments for cell size distribution and growth rate (5-20 μm size fraction) illustrate non-monotonic effects of increasing CO on ecological indicators and may be related to opposing physiological effects of high CO and low pH both within and among species. Interaction of these effects with other factors (e.g., nutrients, light, temperature, grazing, initial species composition) may explain variability among published studies. The absence of clear treatment-specific effects at the community level suggest that extrapolation of species-specific responses or experiments with only present day and future CO treatments levels would produce misleading predictions of ocean acidification impacts on plankton production.