An assessment of two visual survey methods for documenting fish community structure on artificial platform reefs in the Gulf of Mexico

Non-extractive visual survey methods are commonly used to assess a variety of marine habitats. The use of Underwater Visual Census (UVC) by SCUBA divers is predominant; however, remotely acquired video data (e.g., cameras systems, remotely operated vehicles (ROVs), submersibles) are becoming more frequently used to acquire community data. Both remote and diver-based surveys are currently used to survey artificial reef habitat in the Northwestern Gulf of Mexico (GOM) and have associated error due to inherent method bias. Because survey methods that most accurately document the occurrence and estimated abundance of several important fisheries species are greatly needed in the GOM, we compared data collected on the same days and sites from both Roving Diver Surveys (RDS) and micro-ROV surveys conducted on reefed oil and gas platforms. The combined datasets identified a total of 56 species from 22 families, and there was no significant difference in measured species richness between a comprehensive 30 min ROV survey and RDS. Five species of federally managed fish in the GOM were more frequently detected by ROV, as were the majority of species in the Lutjanid and Carangid families. However, abundance estimates from RDS surveys were up to an order of magnitude greater. Multivariate analyses indicated that method choice affected community composition, with Lutjanids and Carangids driving the differences. These two fish families in particular are subject to method bias, probably due to inflated abundance estimates with RDS, or alternatively, deflated estimates from ROV. Although our ROV surveys more frequently detected important fisheries species and produced conservative abundance estimates, a further examination of species distributions on these high-relief platform reefs is needed to fully determine the most accurate survey method. In addition, the attraction and/or gear avoidance of certain species to underwater vehicles deserves further investigation. Overall, our data indicate these methods are viable but the choice of survey method can have implications for the management of certain species, and that careful consideration of methodology is necessary to most accurately document species of interest.