With an estimated 20,000 whales dying from ship strikes each year, the Woods Hole Oceanographic Institution in Falmouth, Mass. is working to use artificial intelligence and thermal-imaging cameras to diminish these unnecessary deaths.
The independent organization, dedicated to ocean research, technology and education, said that the technology can spot whales more than 4 miles away and alert captains before fatal strikes occur.
“This technology uses thermal imaging to detect a whale’s body or spout and provides real-time information on their distance and bearing from a vessel,” the Woods Hole Oceanographic Institution announced in an October press release. “The goal is to provide ship captains with enough time for vessels to be able to change course or slow down.”
Woods Hole Oceanographic Institution is working with the Matson Navigation Company in Hawaii to better protect marine mammals from ocean liners, commercial fishing vessels and cargo ships. The Hawaii-based shipping company Matson Navigation has pledged $1 million to support WHOI’s research and development of the whale-detection cameras.
The company is also taking part in the research by attaching the detection systems to one of its ships traveling the Hawaii trade lane, as well as another in Alaska.
In three months of trial tests, the technology has identified 1,169 whales and dolphins. WHOI Associate Scientist Daniel Zitterbart, who specializes in ocean physics and engineering, said that the cameras’ algorithms filter out thermal signatures of boats, birds, and waves and only send alerts for possible whale detections.
“A highly stabilized thermal imaging camera is mounted on a ship and monitors the surface of the water for whale surfacings and exhalations, or blows,” Zitterbart said. “If a whale blows, its thermal signature is recognized by the integrated AI, and also alerts vessel crews within seconds to the presence of whales up to several kilometers away.”
When a potential whale sighting occurs, the AI-assisted cameras send an alert to a human within 15 seconds, the organizations said. The person then verifies or discredits the sighting to ensure that captains don’t receive false alerts.
Cascadia Research Collective’s Senior Research Biologist John Calambokidis, who has followed the progress of this technology for the past decade, told KTLA sister station KOIN that this is the first study he’s seen that combines a detection system with a practical verification method.
“You don’t want to have a lot of false negatives, meaning things identified as whales that aren’t whales,” Calambokidis said. “Because then the captains and ships are going to start avoiding things. Or once they learn it’s not working, they’re going to start ignoring it. So that was kind of a neat new addition.”
The waters off the coast of Hawaii and along the West Coast of the U.S. are home to as many as 16 species of whales. The whales most at risk of ship strikes off the West Coast include blue, fin and humpback whales – which are all listed as endangered species – and gray whales, which are protected under the Marine Mammal Protection Act.
“Blue, fin and humpback [whales] have been the primary species [of concern regarding ship strike],” Calambokidis said. “They’re all kind of large, coastally distributed, sometimes occurring in high density and areas of high shipping traffic, and seem to be particularly vulnerable to ship strikes. In other oceans like in the North Atlantic, the primary concern is the North Atlantic right whale.”
Vessel strikes and entanglement are some of the leading causes of injury and death to marine animals like whales. Although the new technology shows promise, Calambokidis said that this is just one method of reducing ship strikes.
“I’m certainly hopeful about technologies like this,” he said. “And you know, seeing this actually tested and implemented and actually being adopted in part by a shipping company, those were encouraging directions along this line. I think the reduction in ship strikes is going to take multiple strategies.”
