Mapping Arctic cetaceans from space: A case study for beluga and narwhal

Abstract

Emergence of new technologies in remote sensing give scientists a new way to detect and monitor wildlife populations. In this study we assess the ability to detect and classify two emblematic Arctic cetaceans, the narwhal (Monodon monoceros) and beluga whale (Delphinapterus leucas), using very high-resolution (VHR) satellite imagery. We analyzed 12 VHR images acquired in August 2017 and 2019, collected by the WorldView-3 satellite, which has a maximum resolution of 0.31 m per pixel. The images covered Clearwater Fiord (138.8 km2), an area on eastern Baffin Island, Canada where belugas spend a large part of the summer, and Tremblay Sound (127.0 km2), a narrow water body located on the north shore of Baffin Island that is used by narwhals during the open water season. A total of 292 beluga whales and 109 narwhals were detected in the images. This study contributes to our understanding of Arctic cetacean distribution and highlights the capabilities of using satellite imagery to detect marine mammals.

Whales from space dataset, an annotated satellite image dataset of whales for training machine learning models

“…Detecting whales in the imagery is either conducted manually 1 , 4 , 5 , 7 , or automatically 2 , 3 . A downside of the manual approach is that it is time-demanding, with manual counter often having to view hundred and sometimes thousands of square kilometres of open ocean.…”

Remote sensing techniques for automated marine mammals detection: a review of methods and current challenges

“…Although UAS provide less spatial coverage than satellite images, they can allow more control over survey timing and surveying in overcast conditions, as well as higher spatial resolution, which can be of use for smaller marine mammals like sea otters ( Enhydra lutris ) or dolphins of the Cephalorhynchus genus. Nevertheless, the manual analysis of satellite and UAS imagery is time- and effort-intensive, for which automation can help ( Linchant et al, 2015 ; Hollings et al, 2018 ; Charry et al, 2021 ; Höschle et al, 2021 ). While Thums et al (2018) provided a comparative measure of time saved by their workflow, we recommend future automated marine mammal studies do the same, i.e.…”

“…For example, the analysis of 425 km 2 (still a relatively small area in terms of marine mammal distribution) took 24 h to be analyzed with three replicates (eight hours per person) for the work of Thums et al (2018) , corresponding to 1.13 min per km 2 . Furthermore, Cubaynes et al (2019) reported three hours and 20 min to scan 100 km 2 at a 1:1,500 m. scale (2 min per km 2 ), and Charry et al (2021) reported an even slower rate of approximately 2.5 min per km 2 (at a 1:536 scale). While satellite images have allowed continental-scale abundance and distribution studies in remote locations ( e .…”

“…Automation of image analysis reduces the level of effort and time needed, especially for large-scale projects ( Hollings et al, 2018 ; Charry et al, 2021 ; Höschle et al, 2021 ). Even semi-automation in Thums et al (2018) shortened 24 person-hours to 34 min to run the algorithm and an additional 20 min to revise sub-sections with whale detections and discard those that were false (a reduction of more than 96% of the analysis time).…”

Annotating very high-resolution satellite imagery: A whale case study

“…Research using VHR satellite images to monitor cetaceans has increased since Abileah (2002) [13] and Fretwell et al. (2014) [14] pioneering studies, highlighting how VHR satellite imagery may help gather missing information about whales, and complement boat and aircraft surveys [3 , [15] , [16] , [17] , [18] , [19] , [20] , [21] , [22] , [23] . There have been developments in using this technology in remote regions to estimate whale density [17] , detect strandings [21 , 24 , 25] , and count cetaceans [18] .…”