Introduction
Living cover of reef-building corals has declined on Caribbean reefs by 50 to 80 percent since systematic monitoring began in the late 1970s [1,2] with many reefs transformed from coral- to algal-dominated habitats [2,3]. Coral declines have been attributed to multiple anthropogenic stressors including fishing, land-based pollution, and global warming as well as epizootics afflicting corals and urchins [2-4]. The mass mortality of the long-spined sea urchinDiadema antillarum in the early 1980s removed the last abundant herbivore from reefs that were largely devoid of herbivorous fish after decades to centuries of overfishing [5-8], precipitating an explosion of macroalgae on reefs across the Caribbean [2]. Outbreaks of “White Band Disease” appeared on many reefs around this same time, eventually killing over 80% of elkhorn coral Acropora palmatawhich previously dominated reef crest zones and staghorn coralAcropora cervicornis which previously dominated midslope zones [9-11]. These events were followed by regional coral bleaching in the 1990s, leading to further increases in coral diseases and in some instances a further replacement of corals by macroalgae [12]. These factors acted synergistically to rapidly transform coral communities. A recent Caribbean-wide analysis of Acropora presence and dominance from the pre-human period to the present revealed that the loss of these corals initially began in the 1950s and 1960s, decades before the first recorded observations of coral disease and bleaching [13].
While the long-term decline of Caribbean acroporid corals has been documented on a regional scale [13], the effects of Acroporaloss on Caribbean reef coral community composition and ecosystem functioning are unknown. Although isolated surveys of altered reefs have found an increase in the relative abundance of low-relief “weedy” species such as Porites and Agaricia over the past few decades [14-17], the magnitude and geographic extent of community change are unresolved. Assessing long-term change in the full reef-building coral community will shed light on the ecological causes and consequences of recent coral declines and assist in guiding appropriate management interventions. We compiled an extensive dataset on the prevalence of common hermatypic coral species and genera at thousands of individual reef sites across the Caribbean spanning the Late Pleistocene epoch (~131,000 ybp)—when humans were absent from the Americas—up to 2011. By pinpointing the initial timing of major community transformations, we inferred general anthropogenic causes of change (i.e., local human stressors and/or climate change). To provide additional insight into the ecological context of change, we also tracked change in (a) major coral ecological guilds based on various life history traits and (b) inter-site community heterogeneity.