TRAIT-BASED ECOSYSTEM MODELS: FROM INDIVIDUALS TO BIODIVERSITY IN AQUATIC COMMUNITIES

There is a pressing need to develop predictive aquatic ecosystem models to better understand and mitigate effects of global warming, eutrophication and fisheries. Contemporary models a limited number of bulk-biomass functional groups; typically two phytop lankton (flagellates and diatoms) and a few zooplankton groups (micro and mesozooplankton), or alternatively, structured models of single key species. However, the structure, states and function of natural systems are more dynamic, with a broad continuum of functional groups competing for limited resources through strongly size-structured interactions. The dominant forms and biodiversity of marine ecosystems are variables, not rigid parameters as in present models. Therefore, the next generation of aquati c community and ecosystem models must allow functional groups to be predictions from and not input to the models. Here we suggest ways to develop models where the structure and diversity of the microbial and the mesozooplankton communities emerge from bas ic traits at the level of individuals. Because these models have a mechanistic rather than empirical foundation, they provide robust predictions that stand up to variations in environmental forcing. We need no longer be so reliant on black-box parameters such as empirically derived growth and mortality rates. In modelling parlance, the theories of adaptive behaviour provide a closure where individual scale processes and their consequences can be modelled directly according to fundamental rules. The tradi tional functional group approach will be replaced by a trait-based method where morphology, life-history and behaviour of organisms emerge from basic rules. The project takes a novel approach to biodiversity and biogeochemistry, and will adress the import ant question of hos size-structure in microbial and mesozooplankton communites are formed.