Break-through in predictive modelling opens new possibilities for aquatic ecology and management – a review

Abstract

Due to the complex nature of ecosystems, it has long been argued that process-based dynamic models will never predict well, and numerous studies and critical model tests have also shown this and that simple regression models often predict better for less work. A new generation of dynamic models have, however, been presented that invalidate previous statements about the predictive power of more comprehensive process-oriented dynamic models. These new dynamic models predict important ecosystem variables very well from few and readily accessible driving variables. This paper gives a review of these new models (mass-balance modelling for lakes, rivers and coastal areas and foodweb modelling based on functional groups) and highlights some important reasons for this break-through in modelling in terms of predictive power, wide applicability and practical use. This open new possibilities in aquatic ecology and ecosystem management, e.g., (1) to predict ecosystem effects of pollutants, (2) to estimate changes in the structure of aquatic foodwebs related to future climate changes, (3) to predict consequences of fish kill catastrophes and biomanipulations and (4) to develop new approaches to set fish quota to complement the methods used today where fish quotas are set from fish catch statistics, and not from the amount of food available for fish and for the prey of the fish, i.e., from the presuppositions given by the aquatic foodweb.