Decadal comparison of a diminishing coral community: a study using demographics to advance inferences of community status.

The most common coral monitoring methods estimate coral abundance as percent cover, either via in situ observations or derived from images. In recent years, growing interest and effort has focused on colony-based (demographic) data to assess the status of coral populations and communities. In this study, we relied on two separate data sets (photo-derived percent cover estimates, 2002-12, and opportunistic in situ demographic sampling, 2004 and 2012) to more fully infer decadal changes in coral communities at a small, uninhabited Caribbean island. Photo-derived percent cover documented drastic declines in coral abundance including disproportionate declines in Orbicella spp. While overall in situ estimates of total coral density were not different between years, densities of several rarer taxa were. Meandrina meandrites and Stephanocoenia intersepta increased while Leptoseris cucullata decreased significantly, changes that were not discernable from the photo-derived cover estimates. Demographic data also showed significant shifts to larger colony sizes (both increased mean colony sizes and increased negative skewness of size frequency distributions, but similar maximum colony sizes) for most taxa likely indicating reduced recruitment. Orbicella spp. differed from this general pattern, significantly shifting to smaller colony sizes due to partial mortality. Both approaches detected significant decadal changes in coral community structure at Navassa, though the demographic sampling provided better resolution of more subtle, taxon-specific changes.