The importance of seagrass beds and mangroves as a juvenile habitat as opposed to other shallow water habitat types is investigated using a single sampling method on four islands in the western Indian Ocean for Cheilinus undulatus, and on one island in the southern Caribbean Sea (Aruba) for Scarus guacamaia. Both species occur on the Red list of threatened species. Juveniles of Cheilinus undulatus were predominantly found on seagrass beds while adults were limited to the coral reef. The presence of seagrass beds resulted in significantly higher densities of the species on coral reefs in front of these habitats, indicating the importance of seagrass beds as a juvenile habitat. For Scarus guacamaia, juveniles were exclusively observed in mangroves while adults only occurred on the coral reef. Adult S. guacamaia occurred on all coral reefs along the sheltered coast of the island containing mangroves, but no relationship with distance to mangroves was observed. This could indicate the importance of mangroves for the occurrence of adults of this species on the scale of an entire island.
Connectivity is essential for ecosystem functioning, and in particular for the popula- tion dynamics of species that use different habitats during consecutive life stages. Mangrove and seagrass habitats serve to replenish populations of a range of species that live on coral reefs, but we know little about the fate of these early stages and the spatial scale at which adult populations benefit from this enhancement effect. We examined densities of 12 ecologically important Carib- bean fish species across 3 nursery-dependency categories (high, low, none). We tested the hypo- theses that for nursery species, (1) densities and (2) biomass in the adult habitat decrease with dis- tance from nurseries as the enhancement effect is progressively diluted, and (3) densities in the adult habitat are positively correlated with total juvenile abundance in nurseries. Reef density and biomass of the high- and low-dependence species declined rapidly within ~4 km from nurseries, while at a distance of ~14 km densities of most species were close to zero. These patterns were not confounded by local habitat complexity. Density and biomass of the no-dependence species re- mained unchanged with distance. Total abundance of juvenile fishes in nurseries was a good pre- dictor of total adult abundance on adjacent reefs for the high-dependence species. Our results demonstrate that for several species, enhancement of adult reef populations by mangrove and seagrass nurseries is highly localized (less than ~4 km) in terms of abundance and biomass, and the magnitude of this enhancement is highly correlated with juvenile population abundances within the nursery habitats.