Model Analysis of Vertical Carbon Export in a Mesocosm Experiment

Marine biogeochemical models can be improved by developing a deeper understanding of the necessary degree of model complexity, which includes for example the investigation of non-unique solutions in parameter optimization problems. The examination of uniqueness and uncertainties of optimal parameter estimates might disclose the relevance of individual processes. With our data based model analysis we will explore the possibility of explaining similar patterns in observations that could possibly be explained equally well by different parameter settings or even different parameterizations. Here, we show the results of an optimality based plankton ecosystem model (Carbon:Nitrogen-Regulated Ecosystem Model with Coccolithophores, CN-REcoM&Co). The model includes carbonate chemistry (with air-sea flux of carbon dioxide) and its setup was designed to simulate plankton dynamics observed during a mesocosm experiment (PeECE III, Bergen, Norway, 2005). A special model feature is the explicit consideration of extracellular gels that form from coagulation of algal exudates. These macrogels interact (aggregate) with detritus and become incorporated into sinking particles. In our analysis we focus on those parameters that affect photosynthesis, exudation of polysaccharides, grazing, particle aggregation and sinking. We will discuss how variations of these parameter values induce variability in chlorophyll a, dissolved inorganic carbon (DIC), and particulate organic carbon (POC) concentrations, as well as carbon export flux. Our primary concern is to disclose the uniqueness of the model solution that explains the observed DIC drawdown as well as the build-up and sinking loss of POC from the upper layers.