We analyzed and compared 17 yr of monthly settlement indices of postlarvae of the Caribbean spiny lobster, Panulirus argus, at two locations on the Mexican Caribbean coast. Along this coast, the shelf is very narrow and the local circulation is driven by waves overtopping subtidal coral reef crests. Settlement occurred year-round, but both locations showed annual rhythmicity, average seasonal maxima during the autumn, and similar interannual variation, reflecting the influence of mesoscale oceanographic processes on settlement. Significant correlations between settlement indices and a number of interrelated physical factors indicate that sea-level variation is the main factor influencing settlement. There was further association between extreme settlement pulses and extreme weather events (tropical storms and hurricanes) during the late summer-autumn. The autumn settlement peak indicates that large hurricane waves enhance the onshore transport of postlarvae that originated during the spring spawning peak. The geographic range of potential sources of larvae that settle along this coast during the periods of March-April and September-October, 5-9 months after hatching (the estimated larval duration for P. argus), was investigated by stochastic simulations integrating passive trajectories backward in time using 6 yr of stable outputs from a well-established numerical model for the entire Atlantic Ocean. Trajectories from the model indicate that the local population of P. argus depends to a large extent on larval supply from many potential source regions across and outside the Caribbean but that some larvae follow pathways that permit selfrecruitment.

Habitat degradation can affect trophic ecology by differentially affecting specialist and generalist species, and the number and type of interspecific relationships. However, the effects of habitat degradation on the trophic ecology of coral reefs have received limited attention. We compared the trophic structure and food chain length between two shallow Caribbean coral reefs similar in size and close to each other: one dominated by live coral and the other by macroalgae (i.e., degraded). We subjected samples of basal carbon sources (particulate organic matter and algae) and the same 48 species of consumers (invertebrates and fishes) from both reefs to stable isotope analyses, and determined the trophic position of consumers and relative importance of various carbon sources for herbivores, omnivores, and carnivores. We found that both reefs had similar food chain length and trophic structure, but different trophic pathways. On the coral-dominated reef, turf algae and epiphytes were the most important carbon source for all consumer categories, whereas on the degraded reef, particulate organic matter was a major carbon source for carnivores. Our results suggest that the trophic structure of the communities associated with these reefs is robust enough to adjust to conditions of degradation.

In Caribbean coral reefs, many crustacean species associate with sea anemones, but only a few are anemone symbionts. We examined several ecological traits of 3 anemone species (Bartholomea annulata, Condylactis gigantea, Lebrunia danae) and their crustacean symbionts (6 species) on a coral reef at Puerto Morelos, Mexico. On average, C. gigantea was the largest and B. annulata the most abundant of the 3 anemone species. Season did not affect the density distribution of any species, whereas reef zone (back reef, fore reef, reef channels) significantly affected density and mean size of B. annulata and C. gigantea, but only density of L. danae. The probability of harboring crusta ceans increased with anemone size in all species, but varied with reef zone and season in B. annulata only. These patterns may be due to different microhabitat requirements, reproductive strategies, or photosynthetic plasticity of dinoflagellate endosymbionts among hosts, and different flow regimes among reef zones. Alpheus armatus and Ancylomenes pedersoni were strongly associated with B. annulata, and Periclimenes rathbunae with L. danae. Thor amboinensis and Mithraculus cinctimanus occurred more often in C. gigantea, while P. yucatanicus was more evenly associated with the 3 hosts. Only Ancylomenes pedersoni and T. amboinensis occurred in conspecific groups more often than expected by chance. Commensal complexes of up to 3 symbiont species occurred in all host species, with symbionts that typically used different parts of the host coexisting more frequently. These results provide a baseline to assess the potential influence of local and global anthropogenic stressors on anemone−crustacean symbioses.

Shortage of natural crevice shelters may produce population bottlenecks in juvenile Caribbean spiny lobsters (Panulirus argus), a socially gregarious species. We conducted a field experiment to test enhancement of a local population of juvenile P. argus with the addition of artificial shelters ("casitas") that mimic large crevices (1.1 m(2) in surface area and 3.8 cm in height). Changes in density and biomass of juvenile lobsters 15-50 mm carapace length (CL) were assessed with a multiple before-after control-impact (MBACI) analysis. Separate analyses were also conducted on small (15-35 mm CL) and large (35.1-50 mm CL) juveniles to assess size-related effects. First, we carried out 13 lobster surveys on nine fixed 1-ha sites over a shallow reef lagoon ("before" period). Then, we deployed ten casitas in each of five sites and left four sites as controls, and conducted 22 further surveys ("after" period). Deployment of casitas resulted in a sixfold increase in juvenile density (76% contributed by small and 24% by large juveniles) and a sevenfold increase in biomass (40 and 60%, respectively). Capture-recapture results revealed that enhancement was achieved not by promoting individual growth but by increasing survival, persistence, and foraging ranges of small and large juveniles. Casitas both mitigated shortage of natural shelter and increased sociality, allowing for cohabitation of smaller, more vulnerable juveniles with larger conspecifics that have greater defensive abilities. Casitas may help enhance local populations of juvenile P. argus in Caribbean seagrass habitats, typically poor in natural crevice shelters. The use of MBACI and the simultaneous assessment of multiple interrelated response variables may be a powerful analytical approach to test shelter limitation in other species and to examine the function of structural habitat in other systems.

No abstract