Frontiers of Biogeography

We propose and define the “Habitat Persistence Hypothesis” (HPH) to explain the biogeographical distributions of organisms (especially fishes, invertebrates and algae) inhabiting tropical coral reefs. Both published and unpublished sources indicate that species occurringon deep coral reefs show higher rates of endemism and a less apparent biodiversity gradient across the Pacific Ocean than their counterparts inhabiting shallow coral reefs. The HPH accounts for these biogeographical differences by stipulating that deep reefs are relatively unaffected by sea level changes associated with glacial-interglacial cycles. Shallow-reef habitats may persist across sea level changes in regions with sloped bathymetry (e.g., continental regions and large islands), but are largely extirpated in regions with steep bathymetry (e.g., coral atolls). The HPH suggests that regions with habitat persistence are characterized by higher rates of endemism, and that patterns of attenuating diversity with increasing distance from centers of species richness are shaped by relatively recent recolonization of less persistent habitats from regions with greater habitat persistence. Whereas most existing hypotheses that attempt to explain biogeographical patterns observed on coral reefs (especially in the Indo-Pacific region) relyon observations limited to shallow (<30 m) coral-reef habitat and invoke processes operating on speciation time-scales (107 – 108 yr), the HPH incorporates patterns observed within the remaining 80% of coral-reef habitat (30 – 150 m) and invokes processes operating on time scales associated with sea-level changes (105 – 106 yr). The HPH posits seven specific predictions about coral-reef biogeography that can be directly tested to distinguish it from previous hypotheses. Our intention is to describe the rationale and qualitative support for the HPH with the hope of providing a framework for accumulating sufficient quantitative data to test the predictions, which we anticipate will require decades of robust field surveys.