Strong direct and inconsistent indirect effects of fishing found using stereo-video: Testing indicators from fisheries closures

Candidate indicators of the direct and indirect effects of fishing can be developed by investigating fisheries closures. We tested a suite of such indicators in areas open to fishing but with suspected differences in effort, using baited remote underwater stereo-video methods. In particular, we predicted that greater fishing would result in decreased biomass of high risk target species and indirectly increase the biomass of small-bodied non-target species. As predicted, the biomass of target species was found to be greater in areas of lower fishing effort and in deeper waters. However, no indirect effects of fishing were detected and any community-level effects were driven by differences in the biomass of target species. In particular, assemblage length class analysis of covariance (ANCOVA), size spectra analysis and the abundance-biomass comparison (ABC) method did not provide any evidence of indirect effects of fishing. The magnitude of the differences in fishing effort between the two areas sampled, may be sufficient to significantly affect target fisheries species, but insufficient to lead to indirect effects on non-target populations. It is also possible that the predicted indirect effects do not occur in this assemblage, due to weak trophic linkages between species. Differences observed using the ABC method were attributed to variation in the abundance of large herbivorous fishes, which are not fished. We also found assemblage length class ANCOVA and size spectra to be insensitive to the direct effects of fishing where large numbers of non-target individuals are sampled along with fished species. We suggest diet studies and comparisons across stronger gradients in fishing pressure to further investigate the indirect effects of fishing in this assemblage.