Modelling planktonic ecosystems: parameterizing complexity

This paper explores several simplified representations of complexity or ecological ‘texture’ in models of the marine planktonic ecosystem. It is relatively straightforward to formulate more complex models to include explicitly different functional groups of phytoplankton, zooplankton and bacteria, and to include regulation by multiple nutrients such a nitrate, ammonium, silica, and iron. However, the number of parameters that must be specified from observations increases approximately as the square of the number of compartments and quickly surpasses our ability to constrain them properly from observations. Moreover, ecosystem models often become unstable for small changes in parameter values, and increasing complexity may not lead to increased stability. Here we consider alternative formulations for ecosystem models that try to represent complex interactions, such as the microbial loop, in simpler, less explicit ways. Results are presented demonstrating diagnostic methods to represent (1) multiple size classes of phytoplankton according to total biomass of a single phytoplankton compartment, and (2) partitioning a single compartment for nitrogen into nitrate and ammonium according to the origin of incoming fluxes, nitrification by bacteria, and a relative preference algorithm for uptake by phytoplankton. Ambient concentrations of ammonium are simulated with this model and evaluated against observations from Ocean Station P in the subarctic Northeast Pacific Ocean.