SwimmingIndividuals: A high-performance agent-based model for marine ecosystems.

SwimmingIndividuals, an open-source mechanistic agent-based model, simulates full life-cycle marine agents at large scales, reproducing diel vertical migration and King Mackerel growth consistent with stock assessments.

• Why it matters: Mechanistic ABMs are needed to project ecological responses to novel conditions, but prior applications have been constrained by computational limits and simplified sub-models.
• What they did: The framework couples visual prey detection, adaptive behaviors like diel vertical migration, a flexible bioenergetics engine with three taxon-specific metabolic architectures, and habitat-seeking pathfinding, using a hybrid CPU/GPU Julia implementation.
• The result: In three Gulf of America simulations, it reproduced DVM, cetacean diving patterns, size-based trophic shifts, and von Bertalanffy growth curves, enabling long-term population dynamics and quantification of fishery harvest control rule impacts across five fleets.

AI summary of the abstract · may contain errors · not medical advice · read the paper.