This article is more than 1 year old
Save the whales – with, uh, artificial intelligence?
When a Klingon Bird-of-Prey just won't cut it
Bright yellow buoys running AI software have been deployed in an attempt to deter cargo ships from running over nearby whales.
Cargo ship collisions are the top cause of whale death, Douglas McCauley, professor of ocean science at the University of California, Santa Barbara, told The Register this week. California is home to some of the busiest US ports, and whale collisions (in which the mammal invariably comes off worse) are increasing as shipping traffic levels grow. McCauley reckons up to 80 endangered whales are killed along the US West Coast every year.
McCauley directs the Benioff Ocean Initiative, a philanthropic effort funded by billionaire Salesforce founder and co-CEO Marc Benioff and his wife Lynne. The group is working together with the non-profit Marine Mammal Center to expand Whale Safe, an AI-powered system designed to alert cargo ships to incoming whales. Slowing vessels down reduces the risk of deadly collisions with the animals.
Each buoy carries a computer and listens out for whale sounds using an underwater microphone. The audio is fed into an AI algorithm, running onboard, that can detect the clicks and squeals of specific whale species. When the software identifies these sounds, they're relayed back to base for scientists to double check and record in a log. The data used to train the system was collected by Ana Sirovic, a professor at the Norwegian University of Science and Technology, who has amassed decades worth of whale audio recordings.
"Different whale populations have different dialects," Professor McCauley told us. "To make this AI work in California, it required training the AI specifically using calls from these California whales. The buoy detects blue, fin, and humpback whales – because these are the three endangered whales in our region. These signals are then transmitted via satellite every two hours, a scientist will review the software's results once a day to check it is identifying whale sounds correctly."
Separate machine-learning models analyzing water conditions and records of whale sightings are used along with the audio detection logs to approximate the locations of these animals. An alert is sent out to cargo ships urging them to slow down if they're cruising in an area predicted, from all this information, to be a whale hotspot. The AI acoustic model is capable of detecting blue and fin whales up to 25 kilometers away; humpback whales are quieter, however, and can be found within five kilometers away from the buoy.
- US Senators want more AI, while Microsoftie Paul Allen wants to use it to save wildlife, etc
- Russian-trained spy whale spooks Norwegian fishermen
- Salesforce chief Marc Benioff gets clear view of pandemic, 'new balance' from guru pal
- Japanese sat tech sinks Sea Shepherd anti-whaling activists' hopes
Each Whale Safe buoy costs around $250,000 to build and $200,000 to maintain. The first one was deployed in a pilot study, to assess the technology in the Santa Barbara Channel near the Los Angeles and Long Beach Ports two years ago. Now, the Benioff Ocean Initiative and Marine Mammal Center have installed a second one near San Francisco.
"Whales have been on our planet, being beautiful, majestic, and powerful parts of the ocean, for 50 million years," said Professor McCauley. "We want to make sure that as the oceans become busier with more human commerce, they keep being all those things for another 50 million years. They deserve a place on our planet."
The large sea mammals are also crucial for maintaining a healthy ocean, and have a big impact in creating oxygen, recycling nutrients, and supporting food chains.
"Whale Safe is on a mission to help save the incredible mammals who have ruled the oceans for tens of millions of years," Benioff said in a statement.
"This is a triple win for the planet – we save the whales, fight climate change, and promote community health by cutting air pollution. We need more solutions like this coming out of alliances between science and business."
The team hopes to deploy more AI-powered whale-saving buoys in other coastal areas around North America, such as Seattle, Vancouver, and San Diego, and believes the technology could one day be used around the world in other wildlife hotspots, such as Sri Lanka, too. ®