National Center for Caribbean Coral Reef Research
University of Miami - Rosenstiel School of Marine and Atmospheric Science
4600 Rickenbacker Causeway, Miami, FL 33149




ABM of Reef Fishes

The advent of powerful and affordable desktop computers allowed fisheries scientists to use computing intensive models to describe the life cycle of the majority of commercial fished species.  This also provided a theoretical foundation to derive parameters of other species (see FishBase for extensive compilation of parameters). These models describe, among other things, instantaneous mortality rates, reproduction, recruitment patterns, and prey-predator relationships (e.g. FiSAT) to help define management parameters.  Attempts had been made to develop top-down models of interacting elements in an ecosystem (e.g. Ecopath with EcoSim). 

Agent-based models will augment the existing tools by its ability to simulate the effects of disturbances in the ecosystem (e.g. introduction of pollutants, demise of coral colonies, changes of food preferences, etc.) and in the process, help define key parameters that are essential in determining the optimal catch rate that will sustain a given stock. Developing bottom-up models using agent-based approaches will answer critical questions that have eluded fisheries scientists for decades. 

The ABM models of reef fishes will initially contain the following classes defining the agent: (i) growth, (ii) reproduction, (iii) mortality, (iv) food preferences, and (v) defenses. More classes will be developed later to better describe the agents. The figure presented in this page is an abridged representation of a simulation of a fish species population in a hypothetical space for one time period. Our vision is to present the 3D-rendered agents on a 3D geographic representation of the study area. 

NOTE: The basic Wa-Tor World model is now available. Click here for more on the model.