Localization of North Atlantic Right Whales Using a Deformable Sonobuoy Array

Abstract

North Atlantic right whales are critically endangered, with an estimated 356 (+7/−10) individuals remaining as of 2023. Current mitigation measures include fishing area closures and vessel speed slowdowns when a right whale is detected acoustically or visually. Although the primary means of detection is from aerial surveys, passive acoustic monitoring has the advantage of being long-term and can be persistent without requiring continual human observation. However, acoustic detections currently used in Canada do not include localization estimates, which could enable better management decisions. This thesis examines three localization algorithms: bearing localization, time-difference-of-arrival, and matched-field processing. They were evaluated according to the precision they provided with the location estimates and the assumptions needed to use them. Based on the results, the bearing and time-difference-of-arrival algorithms showed promise in improving mitigation procedures with their localization estimates. However, given its uncertainty and assumptions, matched-field processing was deemed untrustworthy.