Geographic Information Systems (GIS) analysis of high-resolution multibeam bathymetry and remotely operated vehicle data to model rockfish habitat preference

Abstract

Management of declining rockfish stocks requires effective tools capable of providing accurate stock assessments of near-shore, high-relief habitat. Multibeam bathymetry, when analyzed with GIS landscape analysis tools, can create models which can identify preferred habitat based on species-specific parameters. For this study, high-resolution multibeam data of the Del Monte shale beds in Monterey Bay, California were analyzed in GIS for slope, rugosity, and relative topographic position to assess rockfish (Sebastes spp.) habitat preference. Video transects collected by a remotely operated vehicle (ROV) provided habitat ground-truth and fish distribution data. A series of habitat suitability models was created in GIS by combining different suitability factors from multibeam-derived grids: slope, rugosity, topographic position index (TPI), and depth. Distance to preferred categories for each parameter were determined for eight rockfish species. Of these, distance to peak features identified by TPI50 proved the most effective means of modeling fish distribution, successfully predicting an average of 80% of the eight rockfish species. Using fish distribution information, stock estimates were calculated for the study area. By combining GIS landscape analysis tools with multibeam bathymetry and ROV video data, we have created a predictive tool that can locate areas of most suitable habitat given rockfish-specific parameters.