Global ocean-mapping project bolstered by Canadian-made AI tool

A new Canadian-made software tool could speed up a global effort to create a definitive map of the ocean floor.

Teledyne CARIS, a developer of marine mapping software, has created a new data-processing tool in partnership with the United Kingdom Hydrographic Office (UKHO), a world-leading surveyor of the oceans. Together, they have entered into a technical cooperation agreement with Seabed 2030 — a collaborative effort to map the world’s oceans by 2030 led by the Nippon Foundation and GEBCO (General Bathymetric Chart of Oceans).

The companies say that more precise maps are needed for climate change modeling, fisheries management and conservation, and the sustainable development of the Blue Economy.

Seabed 2030 has been endorsed as part of the United Nations Decade of Ocean Science for Sustainable Development (2021–2030).

Vessels around the world are using sonar to measure the depth and shape of the seafloor and are contributing those data to Seabed 2030, where it is being compiled into the freely available GEBCO Ocean Map.

In 2017, when the project was launched, just six percent of the seafloor had been mapped to modern standards. Today, bathymetric data have been collected on more than 20 percent.

“We're bringing AI into this project in an effort to solve a data processing challenge and build the map faster. And we're doing that in partnership with the UKHO,” said Andy Hoggarth, vice president of sales and marketing of Fredericton-based Teledyne CARIS.

“The British Admiralty has hundreds of years of experience in ocean mapping, and therefore they are a very good benchmark to see whether AI can satisfy a need of work that started with Captain Cook in the late 1700s.”

Starting this August, Seabed 2030's network of researchers around the world will have access to Teledyne CARIS's new software to process bathymetric data. It uses an artificial intelligence algorithm designed to separate meaningful depth soundings from the noise, which until now has been a highly manual process.

Over the next 12 months, the researchers will test the algorithm during their ocean mapping cruises and provide Teledyne CARIS and the UKHO with feedback. If it performs as expected, the software could speed up data handling to help Seabed 2030 achieve its goal.

Teledyne aims to commercialize the new software tool to support surveying and cabling for high-speed Internet and offshore wind farms, oil and gas pipelines, as well as marine safety and navigation.

It could also help deliver real-time data to help ensure the safe, secure and sustainable operation of autonomous cargo ships, which are already undergoing trials in some parts of the world.

“A quicker turnaround of data could underpin a move in the next five to 10 years toward autonomous shipping,” said Hoggarth. “If we can turn around data quickly by using AI to take raw measurements and process accurate measurement, then that's going to be beneficial to that work."

The Seabed 2030 project builds on research supported in part by the Ocean Supercluster through its Accelerated Ocean Solutions Program that makes its marine mapping software available in the cloud.

R$