THE ROLE OF SPATIAL AGGREGATION IN CORAL REEF BENTHIC DYNAMICS

Modeling potential reef futures necessitates an understanding of coral reef benthic dynamics and the spatial processes that drive them. Simulations using cellular automaton models explicitly capture the intense spatial interactions governing emergent benthic reef structure. Nevertheless, most reef benthic models neglect to include vital spatial aggregation patterns embedded within the ‘ecological memory’ of current reef states. Ecological memory consists of the biological characteristics of an area that determine the trajectory of the ecosystem into the future, especially over short decadal time scales. Instead, many models initialize with a default random arrangement of relevant functional benthic groups (e.g. coral, macroalgae, turf algae) on a grid lattice. To investigate the importance of accounting for spatial aggregation patterns, we compared dynamical behavior and differences in the amount of time to equilibrium for initial states with various aggregation levels (quantified using variance: mean ratios). One hundred simulations over a range of aggregation levels (uniform to random to aggregated) revealed that it takes a significantly greater amount of time for the system to reach the attractor as the level of coral aggregation increases. This might occur due to the model having to decompose pre-existing patterns before arriving at equilibrium. We also found a consistent ‘overshooting’ beyond equilibrium for the lower aggregation levels (i.e. uniform configurations), a trend that could be erroneously misinterpreted by stakeholders. These results emphasize the importance of accounting for aggregation patterns in models simulating potential reef futures over short time scales as well as the role of space in the reef benthos.