Individual-based movement behaviour in a simple marine reserve—fishery system: why predictive models should be handled with care

The problem of overexploitation and unsustainability is a major issue in global fisheries. Marine reserves or protected no-take zones have been suggested as a possible solution that would maintain yield and protect stocks indefinitely. A key factor in the effectiveness of a marine reserve-fishery system is the rate of exchange of biomass between reserve and fishery: if the rate of exchange is too low then the fishery is not viable, but if the rate of exchange is too high then stocks may be exploited unsustainably and the reserve is rendered ineffective. The rate of exchange is determined by both the physical design and shape of the reserve, and the movement and dispersal behaviour of both the adult and larval-stage fish. Previous models looking at optimal reserve design usually only consider a diffusive population scale movement and dispersal, even though most animal movement is more realistically modelled as being correlated at the individual level. In this article, a deliberately simple simulation of a theoretical marine reserve-fishery system is used to demonstrate the danger of making predictions using only a population-level simplistic diffusive movement model. Further predictions based on the population average of a more realistic correlated movement model are also shown to be inaccurate. This result is due to both the high levels of individual variability in movement behaviour, and the heterogeneity of the environment. This suggests that in future studies, individual-based (rather than population-level) simulations and models are likely to give more useful insights into the dynamics of the marine fishery environment. © 2008 Springer Science+Business Media B.V.