Aerial imagery surveys are commonly used in marine mammal research to determine population size, distribution and habitat use. Analysis of aerial photos involves hours of manually identifying individuals present in each image and converting raw counts into useable biological statistics. Our research proposes the use of deep learning algorithms to increase the efficiency of the marine mammal research workflow. To test the feasibility of this proposal, the existing YOLOv4 convolutional neural network model was trained to detect belugas, kayaks and motorized boats in oblique drone imagery, collected from a stationary tethered system. Automated computer-based object detection achieved the following precision and recall, respectively, for each class: beluga = 74%/72%; boat = 97%/99%; and kayak = 96%/96%. We then tested the performance of computer vision tracking of belugas and manned watercraft in drone videos using the DeepSORT tracking algorithm, which achieved a multiple-object tracking accuracy (MOTA) ranging from 37% – 88% and multiple object tracking precision (MOTP) between 63% – 86%. Results from this research indicate that deep learning technology can detect and track features more consistently than human annotators, allowing for larger datasets to be processed within a fraction of the time while avoiding discrepancies introduced by labeling fatigue or multiple human annotators.

As interest in tourism and conservation grows worldwide, whale-watching has become a popular means of educating the public about wildlife conservation. The short-term impact of ecotourism industries on observed species has been widely studied with findings that indicate responses are most often behavior alterations or avoidance. Close vessel interactions with beluga whales (Delphinapterus leucas) are a major draw for whale-watching ecotourism in Churchill, Manitoba, Canada. As the Churchill River estuary and surrounding waters are assessed for a Marine Protected Area, information on the response of belugas to vessels are needed to inform management. To assess this, an oblique time-lapse camera system with a 5-minute photo interval was set up overlooking a section of the Churchill River estuary that is shared by belugas and tourist vessels. Measurements calculated from photos were used to compare the distance between belugas and kayaks, paddleboards, motorboats, and Zodiac whale-watching vessels. These distances were compared to an expected distribution generated from locations of belugas in photos without the presence of vessels. We found evidence that belugas are attracted to kayaks, avoid paddleboards, and are neutral regarding motorboats and Zodiacs. This is the first study to quantify the behavioral response of cetaceans to tourist vessels using a camera system and a distance-based analysis. Results could inform the development of a site-specific management system that accounts for beluga-vessel relationships.

River estuaries along western Hudson Bay, Canada, are important summer habitats for beluga whales (Delphinapterus leucus), and subject to increasing industrial development activities including vessel traffic. The feasibility of establishing a National Marine Conservation Area (NMCA) in western Hudson Bay is under consideration, requiring baseline studies and habitat monitoring. In this study, beluga whale locations were identified using aerial photographs collected during summer 2018 of the Seal, Knife, Churchill, and Nelson River estuaries. Sentinel 2 wavelength bands were used to outline river plume boundaries for the Seal, Knife and Churchill Rivers. Multiple discriminant analysis was used to differentiate between beluga habitat areas according to their environmental characteristics including concentration of total suspended sediments (TSS), and colored dissolved organic matter (CDOM). The Seal River estuary, Knife River estuary, Churchill River outer estuary, Churchill River estuary and Nelson River estuary were identified as distinct habitat areas. Resource selection functions and model selection were used to determine that habitat variables related to prey availability were important for beluga habitat selection, including TSS, CDOM, and the distance to river mouth or river plume. Identification of preferred habitat and habitat areas in this study are imperative for future management decisions including establishment of a NMCA.