How artificial potential field algorithms can help to simulate trade-offs in movement behaviour of reef fishes

Abstract

IntroductionSpace use patterns in fish result from the interactions between individual movement behaviour and characteristics of the environment. Herbivorous parrotfishes, for instance, are constrained by the availability of resources and different predation risks. The resulting spatial distribution of the fish population can strongly influence community composition and ecosystem resilience.MethodsIn a novel approach, we combine individual-based modelling (IBM) with an artificial potential field algorithm to realistically represent fish movements and the decision-making process. Potential field algorithms, which are popular methods in mobile robot path planning, efficiently generate the best paths for an entity to navigate through vector fields of repellent and attracting forces. In our model the repellent and attracting forces are predation risk and food availability, both implemented as separate grid-based vector fields. The coupling of individual fish bioenergetics with a navigation capacity provides a mechanistic basis to analyse how the habitat structure influences population dynamics and space utilization.ResultsModel results indicate that movement patterns and the resulting spatial distributions strongly depend on habitat fragmentation with the bioenergetic capacity to spawn and reproduce being particularly susceptible processes at the individual level. The resulting spatial distributions of the population are more irregularly distributed among coral reef patches the more the coral reef habitat becomes fragmented and reduced.DiscussionThis heterogeneity can have strong implications for the delivered ecosystem functioning, e.g., by concentrating or diluting the grazing effort. Our results also highlight the importance of incorporating individual foraging-path patterns and the spatial exploitation of microhabitats into marine spatial planning by considering the effects of fragmentation. The integration of potential fields into IBMs represents a promising strategy to advance our understanding of complex decision-making in animals by implementing a more realistic and dynamic decision-making process, in which each fish weighs different rewards and risks of the environment. This information may help to identify core areas and essential habitat patches and assist in effective marine spatial management.