Effects of experimental coral disturbance on the structure of fish communities on large patch reefs

Communit~es of reef fishes on 10 large isolated patch reefs or ' b o m e s ' (108 to 267 m') were monitored at 1 to 2 mo intervals over 1 yr, and then the hard corals on 5 of the bommies were physically disturbed by breaking the colonies into small pieces with a mallet. Monitoring continued for a further year after the disturbance. Coral cover at the impacted bommies decreased from -66 to -29% by the end of the study, while coral cover had increased by 5 % in the same period at the control bommies. There were significant declines in species richness and total abundance of fishes on the impacted bommies, but multivariate analyses showed that the disturbance had no significant effect on spatiotemporal patterns of adult relative abundance and larval recruitment at the community level. Effects of disturbance on the abundance and species richness of 6 families and 8 distinct ecological categories of fishes were also exammed. There were significant declines in the abundance of 1 family (the Chaetodontidae) and 1 ecological category (water column/substratum feeding Pomacentridae), and significant declines in the species richness of 2 families (the Apogonidae and Scaridae) and 1 ecological category (invertebratc feeders). Although coral disturbance did effert some elements of the fish con~munities, it appears that patterns of spatial heterogeneity In the underlying hard substrata at each bommie were also important determinants of fish community structure. Hence, historical patterns of coral growth and substrate consolidation, together with the contemporary coral community, can affect the structure of fish communities on these large patch reefs. Further studies of fish-habitat associations, together with field experiments which manipulate both the underlying substratum and the abundance of live corals, will be necessary to determine the relative importance of each of these habitat characteristics to reef fishes.