Automated Classification of Beaked Whales and Other Small Odontocetes in the Tongue of the Ocean, Bahamas

Abstract

Navy sonar has recently been associated with a number of marine mammal stranding events. Beaked whales have been the predominant species involved in a number of these strandings. Monitoring and mitigating the effects of anthropogenic noise on marine mammals are active areas of research. Key to both monitoring and mitigation is the ability to automatically detect and classify the animals, especially beaked whales. This paper presents a novel support vector machine based methodology for automated species level classification of small odontocetes. To date, the algorithm presented has been trained to differentiate the click vocalizations of Blainville's beaked whales (Mesoplodon densirostris) from the clicks produced by delphinids and from man-made sounds. The automated classification capability compliments the detection and tracking tools already developed through ONR funding for the monitoring and localization of whales at the Atlantic Undersea Test and Evaluation Center, Andros Island, Bahamas