Modelling of ecosystems
Understanding trophic interactions between different parts of the ocean's planktonic food web is the main objective of marine ecosystem modelling. This is also required in order to better compute the biogeochemical cycling of elements. Modelling trophic interactions means to simulate those biological processes which are responsible for the transfer of mass (e.g. nitrogen or carbon) within the food web. Hence, the mass exchange rates in the model are parameterizations of the responses of planktonic organisms to their environment. The chosen parameterizations need to be in accordance with physiological studies and the results of laboratory or mesocosm experiments. The coupling between different elemental cycles depends on the physiological state of the algae (acclimation) as well as on the species composition. Phytoplankton species can be classified into functional groups (e.g. silicifying and calcifying algae as well as nitrogen fixers), influencing elemental cycles in different ways. In our group we work on an ecological model which includes the simulation of calcifying photoautotrophic organisms such as coccolithophores. One of the main challenges in model development is to avoid an "overparameterization". The model needs to remain testable. An evaluation of the model performance can only be achieved by validating model against measured data. To date, useful datasets for the evaluation of ecosystem models come from mesocosm and laboratory experiments and open-ocean time-series studies.
Large-scale fluxes of carbon between the ocean and atmosphere are tightly coupled to ocean circulation and to primary production in the ocean and thereby to the cycling of nutrients (nitrate, phosphate, but also micronutrients, such as zinc, iron). Simulating the marine carbon cycle therefore requires coupling of an ocean general circulation model to tracer transport equations and a representation of the biological productivity, e.g. by an ecological model. We are interested in the cycling of iron in the upper ocean. The main focus is the role of dust deposition as an iron source for phytoplankton. Iron is a highly particle-reactive element and shows an interesting photochemistry. We are currently coupling a model of these processes to an existing ecosystem model that has been optimized for the location of the US-Bermuda Atlantic Time series Study (BATS). Later on we intend to use the model to derive a quantitative assessment of the global input of bioavailable iron from dust.