Ecological studies have rarely been performed at the community level across a large biogeographic region. Sponges are now the primary habitat-forming organisms on Caribbean coral reefs. Recent species-level investigations have demonstrated that predatory fishes (angelfishes and some parrotfishes) differentially graze sponges that lack chemical defenses, while co-occurring, palatable species heal, grow, reproduce, or recruit at faster rates than defended species. Our prediction, based on resource allocation theory, was that predator removal would result in a greater proportion of palatable species in the sponge community on overfished reefs. We tested this prediction by performing surveys of sponge and fish community composition on reefs having different levels of fishing intensity across the Caribbean. A total of 109 sponge species was recorded from 69 sites, with the 10 most common species comprising 51.0% of sponge cover (3.6-7.7% per species). Nonmetric multidimensional scaling indicated that the species composition of sponge communities depended more on the abundance of sponge-eating fishes than geographic location. Across all sites, multiple-regression analyses revealed that spongivore abundance explained 32.8% of the variation in the proportion of palatable sponges, but when data were limited to geographically adjacent locations with strongly contrasting levels of fishing pressure (Cayman Islands and Jamaica; Curaçao, Bonaire, and Martinique), the adjusted R 2 values were much higher (76.5% and 94.6%, respectively). Overfishing of Caribbean coral reefs, particularly by fish trapping, removes sponge predators and is likely to result in greater competition for space between fastergrowing palatable sponges and endangered reef-building corals.chemical ecology | indirect effects | community structure | marine protected areas | trophic dynamics

Caribbean coral reefs have been transformed in the past few decades with the demise of reef-building corals, and sponges are now the dominant habitat-forming organisms on most reefs. Competing hypotheses propose that sponge communities are controlled primarily by predatory fishes (top-down) or by the availability of picoplankton to suspension-feeding sponges (bottom-up). We tested these hypotheses on Conch Reef, off Key Largo, Florida, by placing sponges inside and outside predator-excluding cages at sites with less and more planktonic food availability (15 m vs. 30 m depth). There was no evidence of a bottom-up effect on the growth of any of 5 sponge species, and 2 of 5 species grew more when caged at the shallow site with lower food abundance. There was, however, a strong effect of predation by fishes on sponge species that lacked chemical defenses. Sponges with chemical defenses grew slower than undefended species, demonstrating a resource trade-off between growth and the production of secondary metabolites. Surveys of the benthic community on Conch Reef similarly did not support a bottom-up effect, with higher sponge cover at the shallower depth. We conclude that the structure of sponge communities on Caribbean coral reefs is primarily top-down, and predict that removal of sponge predators by overfishing will shift communities toward faster-growing, undefended species that better compete for space with threatened reef-building corals.

Interest in the ecology of sponges on coral reefs has grown in recent years with mounting evidence that sponges are becoming dominant members of reef communities, particularly in the Caribbean. New estimates of water column processing by sponge pumping activities combined with discoveries related to carbon and nutrient cycling have led to novel hypotheses about the role of sponges in reef ecosystem function. Among these developments, a debate has emerged about the relative effects of bottom-up (food availability) and top-down (predation) control on the community of sponges on Caribbean fore-reefs. In this review, we evaluate the impact of the latest findings on the debate, as well as provide new insights based on older citations. Recent studies that employed different research methods have demonstrated that dissolved organic carbon (DOC) and detritus are the principal sources of food for a growing list of sponge species, challenging the idea that the relative availability of living picoplankton is the sole proxy for sponge growth or abundance. New reports have confirmed earlier findings that reef macroalgae release labile DOC available for sponge nutrition. Evidence for top-down control of sponge community structure by fish predation is further supported by gut content studies and historical population estimates of hawksbill turtles, which likely had a much greater impact on relative sponge abundances on Caribbean reefs of the past. Implicit to investigations designed to address the bottom-up vs. top-down debate are appropriate studies of Caribbean fore-reef environments, where benthic communities are relatively homogeneous and terrestrial influences and abiotic effects are minimized. One recent study designed to test both aspects of the debate did so using experiments conducted entirely in shallow lagoonal habitats dominated by mangroves and seagrass beds. The top-down results from this study are reinterpreted as supporting past research demonstrating predator preferences for sponge species that are abundant in these lagoonal habitats, but grazed away in fore-reef habitats. We conclude that sponge communities on Caribbean fore-reefs of the past and present are largely structured by predation, and offer new directions for research, such as determining the environmental conditions under which sponges may be food-limited (e.g., deep sea, lagoonal habitats) and monitoring changes in sponge community structure as populations of hawksbill turtles rebound.

Fundamental theories of resource allocation for terrestrial plants predict that species investing in chemical defenses should be slow growing or less fecund, while undefended species should mitigate consumer effects by diverting energy to growth and reproduction, thereby also enhancing their ability to colonize open space. As for some plant communities, sponges on Caribbean coral reefs include chemically defended species and undefended species that tolerate consumer damage. We surveyed the sponge community on the 155 m long wreck of the USS 'Spiegel Grove' (~1583 m 2 deck surface), which was intentionally sunk 4 yr previously in 2002 as an artificial reef off Key Largo, Florida, USA, to determine the relative abundance of undefended and defended species on previously uncolonized habitat compared to an adjacent reef at a similar depth and with similar topography. As predicted by theory, chemically undefended sponge species were significantly more abundant (96.0%) and larger on the shipwreck than on adjacent coral reef (15.2%; G-test, p < 0.0001). On a subsequent survey 18 mo later, the first recruits of 6 chemically defended sponge species were discovered, suggesting that the sponge community on the wreck is in transition toward that occurring on adjacent reefs. Although more definitive replicated experiments remain to be performed, these results corroborate the resource trade-off hypothesis as applied to the evolution of chemical defenses among sponge species on Caribbean reefs.

Consumer-mediated indirect effects at the community level are difficult to demonstrate empirically. Here, we show an explicit indirect effect of overfishing on competition between sponges and reef-building corals from surveys of 69 sites across the Caribbean. Leveraging the large-scale, long-term removal of sponge predators, we selected overfished sites where intensive methods, primarily fish-trapping, have been employed for decades or more, and compared them to sites in remote or marine protected areas (MPAs) with variable levels of enforcement. Sponge-eating fishes (angelfishes and parrotfishes) were counted at each site, and the benthos surveyed, with coral colonies scored for interaction with sponges. Overfished sites had >3 fold more overgrowth of corals by sponges, and mean coral contact with sponges was 25.6%, compared with 12.0% at less-fished sites. Greater contact with corals by sponges at overfished sites was mostly by sponge species palatable to sponge predators. Palatable species have faster rates of growth or reproduction than defended sponge species, which instead make metabolically expensive chemical defenses. These results validate the top-down conceptual model of sponge community ecology for Caribbean reefs, as well as provide an unambiguous justification for MPAs to protect threatened reef-building corals.An unanticipated outcome of the benthic survey component of this study was that overfished sites had lower mean macroalgal cover (23.1% vs. 38.1% for less-fished sites), a result that is contrary to prevailing assumptions about seaweed control by herbivorous fishes. Because we did not quantify herbivores for this study, we interpret this result with caution, but suggest that additional large-scale studies comparing intensively overfished and MPA sites are warranted to examine the relative impacts of herbivorous fishes and urchins on Caribbean reefs.