Abstract

There is a great need to develop indicators of the status of exploited reef fish communities in the context of Ecosystem-Based Fisheries Management. Preferably, these indicators should be simple and intuitive, so that they can be effectively used by any stakeholder group. In this study, we use Atlantic and Gulf Rapid Reef Assessment (AGRRA) data to explore relationships among four simple community metrics conventionally used to detect fishing impacts (total fish biomass, total fish density, fish mean weight and fish species richness) from 415 underwater reef-surveys across the Caribbean, performed between 1998 and 2004. Our ultimate goal is to determine which of these metrics, if any, can stand alone and is most specific to fishing impacts. In order to do this, we examined how these metrics co-varied across the region and identified their environmental correlates, while control-ling for potentially confounding spatial effects. We found that total fish density and fish mean weight, the two components of fish total biomass, were uncorrelated and thus conveyed different information. In contrast, total biomass was correlated with the other three metrics and was least informative once the latter were considered. Fish density and fish mean weight displayed the strongest associations with environmental factors, but they differed in the nature of their associations. Fish mean weight was negatively related to human population density, whereas fish density was positively related to a reef rugosity index. Overall, our findings support the idea that, at the community level, mean fish weight is the most specific indicator of fishing impacts.