After decades of negotiation, the United Nations’ High Seas Treaty has finally become a reality. First adopted in 2023, the agreement reached a historic milestone when the required 60 countries ratified it. Its entry into force will provide long-awaited protections for international waters — vast, largely unregulated ecosystems threatened by overfishing, deep-sea mining, pollution and climate change.
Covering nearly two-thirds of the global ocean, yet with only 1 percent currently protected, the high seas contain more than 90 percent of ocean habitat and harbour extraordinary biodiversity.
The High Seas Treaty reinforces the principle that safeguarding critical ocean areas is essential for human survival. Healthy high seas regulate currents and stabilize ocean temperatures — benefits that can be preserved while still supporting human livelihoods.
Once it enters into force in 120 days, the treaty will allow countries to establish marine protected areas beyond national jurisdictions, require environmental impact assessments for industrial activities, and ensure that scientific discoveries in the high seas are shared equitably. Its passage is crucial for achieving global goals such as protecting 30 percent of the world’s oceans by 2030.
Canada: Protecting North Atlantic right whale
Another example of marine conservation activities is the protection of the North Atlantic right whale (NARW), one of the world’s most endangered marine mammals.
Once abundant along the eastern coast of North America, the species was driven to the brink of extinction by centuries of commercial whaling. Although whaling has long been banned, the population has struggled to recover due to a combination of human-driven threats and the species’ own slow reproductive rate.
Transport Canada is actively working to protect the endangered North Atlantic right whale in the Gulf of St. Lawrence by reducing the risks of vessel collisions in one of their key habitats.
Since 2017, targeted measures have been implemented in a high-traffic area where the species is frequently observed. These include the designation of Vessel Traffic Management Zones (VTMZs), mandatory and voluntary speed and access regulations, monitoring programs, and enforcement of compliance.
Mandatory protection measures apply to vessels 13 metres or longer within the VTMZs. Voluntary measures ask vessels to reduce speed to 10 knots or avoid sensitive areas such as the Cabot Strait, Roseway Basin and certain Gulf zones during periods of whale presence.
Technology to support marine conservation
International agreements and regulations play a significant role in achieving broad societal goals, but operationalizing them requires more detailed, concrete work.
Marine conservation efforts need to be supported by increased monitoring of vessel operations and their impacts in port waters, optimizing vessel performance and port resources, and adjusting routes to mitigate the effects on protected marine species and ecosystems.
GSTS’ experience with high-resolution satellite imagery for vessel detection combined with our vessel tracking and analytics capabilities led to an idea –
“Can we adapt our vessel detection AI models to instead identify whales at surface, and bring this data into an operational vessel tracking system?”
Unlike many studies, which classify satellite data for conservation tasks and ultimately result in published papers with limited real-world utility, GSTS has developed an operational model to detect, validate, and integrate space-based whale detections for marine conservation.
SmartWhales: AI working for marine conservation
Working as a team as part of the Canadian Space Agency’s SmartWhales programme, GSTS has created a whale detection and monitoring system.
Data from various satellite providers is ingested into GSTS’ OCIANA® platform and processed into a form suitable for ‘object detection’ models (i.e., the same technology that can automatically identify people in your mobile phone’s photos).
SmartWhales Space-Based Detection System architecture diagram. Source: Semi-Automated Detection of Right Whales (Eubalaenaspp.) in Very High-Resolution Satellite Imagery
The task of detecting and labelling whales in an image is significantly more difficult due to the scarcity of training samples: whales spend most of their time below the surface, and for NARW, even spotting one is rare. This issue was resolved by increasing the sample size using two different methods.
Through partnerships within the whale monitoring and research community, both within and outside the SmartWhales programme, a wide array of samples was obtained to provide a representative suite of whales as seen from space, under various conditions.
Concurrently, image augmentation methods were employed to increase the sample size available for model training artificially. Through the combined effect of these strategies, a model was trained with high accuracy to detect NARW in satellite imagery.
Left: Locations of imagery collected in Cape Cod Bay that were collected and used for machine learning model training (or-ange) and testing (green). Right: Location of imagery collected in the Gulf of St. Lawrence that was used for model testing (green). Source: Semi-Automated Detection of Right Whales (Eubalaenaspp.) in Very High-Resolution Satellite Imagery
More features for conservation
In addition to building the AI models, OCIANA® offers a dynamic whale risk model that uses environmental data on the predicted location and abundance of its primary food source (copepods of the genus Calanus), as well as historical vessel traffic. For detections that have occurred, we use vessel path prediction tools to identify potential mammal collision risks.
Artificial intelligence provides a brand new set of tools to generate insights into some of the world’s most pressing problems. From reducing greenhouse gas emissions and marine conservation to mitigating maritime risks and optimizing operations, GSTS offers a comprehensive range of capabilities. Contact us to request a demo.
English