Symposium Abstract: Decreasing Habitat Disturbance by Improving Fish Stock Assessments: A New Method of Remote Species Identification and Quantification
D. F. Doolittle, M. R. Patterson, Z.-U. Rahman, and R. Mann
A direct link exists between the quality of fisheries data and the effectiveness of fisheries management. Increasing the quality and quantity of data on which stock assessments and management decisions are based has been cited as a critical national issue (National Research Council, 2000. Improving the Collection, Management, and Use of Marine Fisheries Data. National Academy Press, Washington, D.C.). We approach the challenge of limiting deleterious habitat impacts due to fishing through the creation and demonstration of novel stock assessment and habitat visualization tools. We present here a new method of fish species identification and quantification. The technique uses a Radial Basis Function artificial neural network classifier to discriminate and enumerate selected fish species from high-resolution side scan sonar images. We demonstrate this technology onboard a Fetch! class Autonomous Underwater Vehicle (AUV) and provide examples of how such technologies could augment fisheries stock assessment as well as essential fish habitat determination. Ancillary benefits of this technology include the opportunity to simultaneously characterize surficial bottom types and document habitat utilization by species that are known to the classifier. Such side scan sonar species identification tools would significantly augment current stock assessment methods, provide new insight to habitat usage, and allow more ecologically realistic models to be constructed.