Detecting Marine Mammals at Sea in Atlantic Canada: Comparing Detections Made Using a Rotating Thermal Imaging Camera with Traditional Visual and Acoustic Methods

Underwater sound is frequently introduced into the marine environment during industrial activities such as oil and gas exploration and port and windfarm developments. To successfully mitigate the effects of these introduced... [ view full abstract ]

Underwater sound is frequently introduced into the marine environment during industrial activities such as oil and gas exploration and port and windfarm developments. To successfully mitigate the effects of these introduced sounds on marine mammals, it is often necessary to first detect marine mammals at sea.

Our study compares marine mammal detections made using two traditional methods, marine mammal observers (MMOs) making visual observations and passive acoustic monitoring (PAM), with detections made using a thermal imaging based automatic whale detection system. During the summers of 2015 and 2016, we made visual observations of marine mammals concurrent with the collection of thermal imaging data at a shore-based observation site at Cape Race, Newfoundland. These shore-based data were used to optimize and then validate the detector/classifier used by the automatic system for use in the thermal regime of Atlantic Canada. During summer 2017, we conducted ship-based fieldwork offshore of Nova Scotia and Newfoundland during which we made concurrent observations of marine mammals using the three detection methods. We encountered a variety of marine mammal species, and collected data in a range of environmental conditions. The results of our combined field seasons will be presented and the performance of the various detection methods under different conditions will be discussed. This research is intended to assist in the development of best practice guidelines for marine mammal mitigation and monitoring during seismic surveys and other activities in Canada.