Most models of marine ecosystems can be seen in terms of a set of equations of standard form:
{∂Y/∂t=-∇ΦY+βY}
and their numerical solutions. In the equation, Y is a state variable belonging to a set that could include, for examples, temperature, velocity, energy, and concentrations of nutrients and organisms. The first right-hand term is the divergence of physical transport flux vectors at a given spatial location and time; this term conserves the quantities totalled over a simulation domain except for boundary fluxes. The second right-hand term summarises the sources and sinks of each variable at a given location and time. It is not required to be conservative for any single variable, although conservation laws (e.g. for total energy, total nitrogen) may be introduced to constrain conversions between variables. Instances of this equation range from the simple first dynamical plankton model of Riley (1946) to modern, complicated, models such as ERSEM and COHERENS-PROWQM. This paper will consider two fundamental questions:
1) Is marine modelling largely an engineering problem - i.e. taking established scientific truths and using them to build models that can be used for system diagnosis and prognosis - or part of the scientific process - using models as tools for testing hypotheses? We know less than we think about marine ecosystem dynamics, and so need to continue making and testing hypotheses. Nevertheless, there is a pressing practical need for models that can be diagnostically and prognistically.
2) What are the entities which are quantified by the Y variables? The paper will explore how the idea of microplankton (comprising all pelagic bacteria and protists, both autotrophic and heterotrophic) can be used to develope efficient parameterisations of marine pelagic ecosystems at several levels of model complexity. These complicatied simplifications can be further simplified to make a model that looks much like the original model of Riley, and can in turn be developed into to a steady state solution embodying the precautionary principle that is practically useful in diagnosing or predicting the trophic status of water bodies. Examples will draw on modelling work carried out in several EC supported projects: CANIGO, COHERENS, OAERRE and PROVESS. |
|
|