Thursday, October 27, 2011: 7:05 PM
Room A3 (San Jose Convention Center)
Tropical coral reefs are among the most diverse habitats with great ecological, economic, and social importance. Sadly, coral reefs are also one of the most threatened ecosystems often subject to both local and global anthropogenic impacts. Although most monitoring efforts have focused mainly on descriptions of benthic cover, there is an urgent need to concentrate on coral reef benthic dynamics and the processes that drive them. Due to the inherent complexity of the coral reef benthos, ecological modeling has become an increasingly prominent tool to investigate dynamics. Because different model formulations weigh particular dynamics differently, our investigation is twofold: 1) use two prominent coral reef benthic models to explore and compare behavior and predictions of benthic composition and frequency distributions of relevant functional groups (e.g. turf algae, macroalgae, corals) 2) investigate model-specific sensitivity to particular dynamics. The two models differ dynamically not only in their formulation of coral-algal competition, but also in their treatment of spatial interactions (i.e. explicit vs implicit). We parameterize the models by deriving the necessary biological rates (e.g. coral extension rates, mortality rates, recruitment, etc) from a photographic dataset of permanent quadrats annually obtained from the Caribbean island of Curaçao. We gage model predictive accuracy by using cross-validation predictive deviance metrics. This multi-model approach allows us to illuminate the extent to which predictions are model dependent and address the immensely important task of linking demographic and ecological rates to the emergent distribution of the coral reef benthos; thereby permitting us to better inform coral reef management.