Alternate Title

Factors Associated with Artificial-Reef Fish Assemblages


Visual census (5-min point-count, 100 m2) was used to estimate fish assemblage parameters associated with artificial reef variables from 64 reefs over a 3-yr period in the northern Gulf of Mexico. Dependent variables, recorded by divers [i.e., number of species, number of individuals, length (TL in cm), and species diversity (H')], were analyzed for their associations with potentially underlying environmental attributes using stepwise regression, TWINSPAN, and canonical correspondence analyses (CCA). The fish assemblages (dominated by haemulids, labrids, and serranids) were qualitatively and quantitatively similar to the assemblages described by others from the same general area. Pelagic fishes (carangids and scombrids) associated with the reefs were among the most numerous and were the largest predators in the assemblage. The stepwise regression analysis was able to account for fairly high percentages of the variation in number of species (37%), number of red snapper (43%), and size of red snapper (52%). TWINSPAN allowed the recognition of fish assemblages based on their inshore-offshore biotopes. Vermilion snapper was identified as a key indicator species. The CCA helped identify species groups and factors associated with them. The affinity of pinfish and spottail pinfish with rock jetty was evident, as was the association of triggerfish with vertical steel structure. Although the axis loadings using CCA did not identify a clear species/factor relationship, this analytical method should prove useful in recognizing environmental factors that can be controlled to optimize species-specific artificial reef construction.