Ocean acidification and warming in the Baltic Sea: effects on diazotrophy and pelagic biogeochemistry

Nitrogen (N) is an essential element for cellular functioning in all living organisms. However, the most abundant formof nitrogen (N2 gas) in not bioavailable, or fixed, and therefore N availability restricts primary production in large parts of the ocean. Dinitrogen (N2)-fixing, or diazotrophic, organisms possess a nitrogenase enzyme which converts N2 into bioavailable forms. Aquatic N2-fixation is a source of new nitrogen, hence where fixed N availability conditions, N2-fixation can relieve the N-supply bottleneck in the non-diazotrophic organisms and support increased production - provided there is enough phosphate, light, and warm temperatures to sustain the energetically demanding N2-fixation. The Baltic Sea is a semi-enclosed water body under considerable anthropogenic pressure due to the highly populated drainage basin and limited water exchange. The spring-bloom draws down inorganic nutrients leading to seasonal N-limitation and provision of a diazotrophic niche with excess phosphate, increasingly stratified water column and warming sea surface temperatures. This is a seasonal niche occupied by filamentous diazotrophic cyanobacteria such as Nodularia spumigena and Aphanizomenon flos-aquae, which commonly form extensive surface blooms during the summer period. N2-fixation is particularly important process in the region as it balances N loss processes and supports an estimated 20 - 45% of primary productivity during the summer season. Increasing atmospheric CO2 concentrations due to anthropogenic activity leads not only to warming of the atmosphere and oceans, but also to measurable shifts in seawater carbonate chemistry, termed ocean acidification. Single-strain culture studies have shown that N2-fixation and diazotroph growth is sensitive to changes in the seawater temperature and CO2 concentrations. Until now only a few short-term experiments have been completed to probe changes in fitness of diazotrophic species in situ. In addition, comparatively little is known about the response of low nutrient plankton communities to ocean acidification as more commonly nutrient induced blooms have been studied. This doctoral dissertation presents the results from two independent mesocosm studies on naturally present summer plankton communities in the Baltic Sea. The aim was to investigate the impact of ocean acidification (increased CO2 concentration and decreased pH) as well as the combination of ocean acidification and ocean warming (increased seawater temperature) on the abundance and activity of diazotrophic organisms and on N-limited plankton communities. In the first study, pelagic mesocosms were deployed off the south-western tip of Finland in the Archipelago Sea. To observe differences in organic matter pools and fluxes under realistic ocean acidification scenarios, CO2 concentrations were adjusted to give a range between 365 and 1231 μatm (average during study period) and the plankton community and biogeochemical elemental pools were sampled over the 47-day long study. Approximately three weeks after the initial CO2-manipulation, CO2-related differences in pelagic particulate and dissolved matter pools became clear. These differences were sustained for a further three weeks until the end of the experiment. Higher particulate matter and dissolved organic carbon and chlorophyll a concentrations, and lower dissolved inorganic phosphate (PO 3–4) concentrations under increased CO2 concentrations were driven by the positive response of picophytoplankton (<2 μm). These CO2-related differences in the water column could not be traced into the sinking particle flux within the study period. There were no significant differences in A. flos-aquae abundances, the dominant filamentous diazotrophic cyanobacterium present, or in diazotroph activity. Hence the positive response of plankton community biomass, could not be attributed to changes in fixed N supply. In the second study using the indoor mesocosm facility in Kiel, the interactive effects of elevated CO2 and temperature on new N inputs through diazotrophy were studied in four week long experiment. Here the dominant diazotrophic filamentous cyanobacteria was N. spumigena. There was a strong negative effect of pCO2 on N. spumigena abundances which was exacerbated in the higher temperature treatment. This supports results from culture experiments with N. spumigena and shows that the negative response to increased CO2 concentrations may not be overridden by biotic interactions such as grazing pressure and resource competition within the plankton community. In both studies, abundances of filamentous diazotrophic cyanobacteriawere too lowto distinguish any potential influence on biogeochemical element pools. However, there were noticeable effects of temperature and CO2 on one of the two common filamentous diazotrophic cyanobacteria species. Growth of N. spumigena may become restricted at the summer bloom peak in future, even though the period where blooms occur during summer may expand. How this interacts with shifts in pCO2 and spring bloom dynamics remains unclear. The more coastal-dwelling species, A. flos-aquae, seemed better adapted to variable CO2 concentrations, indicating that future CO2-related changes in abundance in this species are not expected. Previous culture studies have also shown a diverse response of diazotrophic taxa. The results included in this dissertation indicate that picoplankton may be able to sustain higher biomass under ocean acidification despite very low N availability. Hence, there is potential that this sustained response in picoplankton may shift food web structure with consequences for long-term changes in organic matter fluxes.