Meeting Abstract
Reef fish assemblage is important in understanding fishery management. Fish survey data collected by the Reef Environmental Education Foundation was used with the following restrictions: (1) analysis limited to 26 reefs along the Florida Keys with a large number of “expert surveys”, and (2) only “moderately common” fish species considered. Using several cluster analysis techniques, dendrograms were constructed; Ward’s method, which is distinct from other methods because it uses an analysis of variance, gave the most distinct separation. Clusters were tested using discriminant analysis to “back-predict” a reef’s cluster, using latitude, distance of a reef to the nearest neighbor reef, and shortest distance of a reef to one of the island chain Keys (an approximate index of depth). One significant factor was found; using that factor, 73% of the reefs would have been correctly classified if unknown. Additional physical characteristics might be used in the future to get even clearer partitioning.
