Vibrant yellow buoy-working artificial intelligence software programs have been deployed to stop cargo ships from working near whales.
Douglas Macaulay, a professor of marine science at UC Santa Barbara, said collisions with cargo ships were the main killer of whales registry during this week. California is home to some of the busiest ports in the U.S., and whale collisions (which are always worse for mammals during this period) are on the rise as tourists at the delivery location increase. Macaulay estimates that as many as 80 endangered whales are killed each year on the U.S. West Coast.
McCauley leads the Benioff Ocean Initiative, a philanthropic initiative funded by billionaire Salesforce founder and co-CEO Marc Benioff and his spouse Lynne. The group is working with the nonprofit Marine Mammal Center to develop Whale Protected, an artificial intelligence-driven system designed to warn cargo ships of approaching whales. Slowing down a boat can reduce the risk of a fatal collision with an animal.
Each buoy carries a computer and uses underwater microphones to listen to the whales. The audio is fed into an artificial intelligence algorithm working on board, which can detect stress and screams from certain whale species. When software programs identify these sounds, they are transmitted to the bottom for scientists to outline and report in a log. The information used to guide the system was gathered by Ana Sirovic, a professor at the Norwegian Academy of Sciences and Professional Studies, who has collected years of whale recordings.
“Completely different whale populations have completely different dialects,” Professor Macaulay told us. “For this AI to work in California, the AI needs to be particularly good at the calls of those California whales. The buoy recognizes blue whales, fin whales and humpback whales — because they are the three endangered whales in our region. Then these alerts Streamed every two hours via satellite TV, scientists will test the results of the software program as quickly as possible throughout the day to ensure it accurately picks up whale sounds.
Separate machine learning models analyzing water environment and whale sighting data were used along with audio detection logs to estimate the areas of these animals. If cargo ships are navigating what is predicted to be a hot spot for whales based primarily on all this information, a warning could be issued to cargo ships urging them to slow down. The AI acoustic mannequin is ready to detect blue and fin whales at distances of up to 25 kilometers; however, humpback whales are calmer and can be spotted within 5 kilometers of a buoy.
Each whale protection buoy costs about $250,000 to build and $200,000 to maintain. The primary was used in a pilot inspection two years ago to assess expertise in the Santa Barbara Channel near the ports of Los Angeles and the Long Coast. Now, the Benioff Ocean Initiative and the Marine Mammal Center have launched their second event near San Francisco.
“Whales have been on our planet for 50 million years and they are amazing, majestic and efficient marine creatures,” Professor Macaulay said. “We want to make sure that when the oceans get busier with additional human trade, they stick around with all these things for 50 million years. They deserve a place on our planet.”
Large marine mammals are also critical to maintaining healthy oceans and have serious implications for producing oxygen, recycling vitamins and supporting dietary chains.
“Whale conservation groups are dedicated to saving the unimaginable mammals that have ruled the oceans for tens of millions of years,” Benioff said in an opinion.
“It’s a triple win for the planet – we’re saving whales, responding to local weather changes, and promoting community well-being by reducing air pollution. We want to gain more options from the alliance between science and business.”
The organization hopes to deploy additional AI-powered whale rescue buoys in different coastal regions of North America, such as Seattle, Vancouver, and San Diego, and believes the expertise may be used in the future in different wildlife hotspots around the world, the equivalent of Sri Lanka. ®
Is it possible to avoid wasting whales with artificial intelligence? • Registry