Detecting Whales from Space: the development of a powerful monitoring tool!
In the past decade, the ability for satellites to collect images anywhere on Earth at a sub-meter resolution has opened new perspectives for scientists to study wildlife on a large scale. To date, commercial satellites such as WorldView-3 can now collect images at a resolution of 30cm per pixel, which can be brought down to 15cm using post processing technology! To name only a few, very high-resolution satellite imagery has been used to detect and/or monitor elephants, polar bears, Weddell seals, south Atlantic right whales, finback and blue whales. The pandemic further increased our interest in the use of this new platform, giving us the opportunity to collect data while we had to remain at home in the past year.
Beluga individuals detected from WorldView-3 satellite images at a 1:265 scale. Image Source: Mapping Arctic cetaceans from space: A case study for beluga and narwhal
The detection and monitoring of wildlife populations is a fundamental step to track up and downtrends of a population and distribution shifts within species. Scientists have been using an array of tools and methods to study distributions and fluctuations of populations, such as camera traps on land, aerial surveys, acoustic surveys, boat surveys, and satellite tagging. Each of these methods provide valuable information at different scales and introduce their own challenges. For instance, satellite tagging provides very precise data on the movement and behaviour of an animal — but this method is invasive, and in order to make inferences at a population level we need to be able to tag lots of individuals. Aerial surveys can help extract distribution, abundance, and demographic data when used in conjunction with digital cameras, but they are limited in scale and can be subject to double-counting individuals, due to their inability to capture a true snapshot of a large area. Satellite imagery, by contrast, captures a true snapshot and can now provide some information at a much larger scale, filling the gaps in places scientists are not able to access or monitor often.
Will satellite imagery replace traditional surveying methods such as aerial surveys? The quick answer is no! Like any monitoring method, satellite imagery comes with its own caveats. In order to detect whales from space, we need pristine environmental conditions: this means little or no cloud cover (although radar sensors might provide a more weatherproof solution as technology advances), in small swell seas, preferably with no white caps. Nevertheless, more and more projects are pushing the use of satellite imagery and AI to detect and monitor whales: for example, smartWhales in Canada, which uses satellite imagery to detect individuals of the endangered North Atlantic Right Whale; and the SPACEWHALE service in Europe, to detect great whales.
Satellite imagery has a bright future ahead as a surveying method. However, it is still in its infancy, and recent studies, while promising, highlight the need to pursue our effort to understand not only the capabilities of this new tool but also how to implement it in order to get the most value out of it.
Here, at Whale Seeker we recently published an article in a peer-reviewed journal demonstrating the first-ever detection of belugas and narwhals from space. This study is an important addition to the scientific community, demonstrating the potential use of this tool in the future not only for large whales but also medium-sized whales.
Scientists need to use all the tools they can, in order to piece together the most accurate picture of our wildlife populations. At present, satellite surveys need to be thought of as a valuable addition to our toolbox, not a replacement.
