Robustness of perceptual features used for passive acoustic classification of cetaceans to the ocean environment

Abstract

Passive acoustic monitoring (PAM) is used to study cetaceans in their habitats, which cover diverse underwater environments. It is well known that properties of the ocean environment can be markedly different between regions, which can result in distinct propagation characteristics. These can in turn lead to differences in the time-frequency characteristics of a recorded signal and may impact the accuracy of PAM systems. To develop an automatic PAM system capable of operating under numerous environmental conditions, one must account for the impact of propagation conditions. A prototype aural classifier developed at Defence R&D Canada has successfully been used for inter-species discrimination of cetaceans. The aural classifier achieves accurate results by using perceptual signal features that model the features employed by the human auditory system. The current work uses a combination of at-sea experiments and pulse propagation modeling to examine the robustness of the perceptual features with respect to propagation effects. Preliminary results will be presented from bowhead and humpback vocalizations that were transmitted over 1–20 km ranges during a two-day sea trial in the Gulf of Mexico. Insight gained from experimental results will be augmented with model results. [Work supported by the U.S. Office of Naval Research.]