Individual based spatial models with evolved fish behaviour

• Main Authors: Huse, Geir, Giske, Jarl
• Format: Report
• Language: English
• Published: ICES 1998
• Subjects: fish behaviour; fiskeatferd; genetics; genetikk; distribution; utbredelse; VDP::Mathematics and natural science: 400::Zoology and botany: 480::Ethology: 485; VDP::Mathematics and natural science: 400::Geosciences: 450::Oceanography: 452; VDP::Agriculture and fishery disciplines: 900::Fisheries science: 920::Fish health: 923; Barents Sea
• Online Access: http://hdl.handle.net/11250/105851

Due to spatial heterogeneity, the survival and growth of fishes depend upon their habitat. For mobile pelagic fishes, it hence becomes important to analyse the factors responsible for spatial dynamics in order to understand population dynamics better. The current work is an attempt to approach fisheries assessment from a theoretical ecological perspective. We here present individual based single species and multispecies models where spatial behaviour and life history strategies are evolved. Spatial movement for each fish is calculated using an artificial neural network with weights evolved by a genetic algorithm. Behaviour relies on sensory information about food, temperature, light and predators. A concept for modelling life history strategies and migration in fish using this method is presented along with an application of this concept for the Barents Sea capelin. A predator-prey model based on the same principle is also presented. Potential areas of application for these kinds of models are discussed.