AI Is Paying Attention to the Howls of Yellowstone Wolves — and It Is Helping in a Big Way
Yellowstone National Park could be swarming with wolves again, all thanks to artificial intelligence. It seems that the researchers have hit the bullseye with their new approach to conservation or revival. The national park once inhabited a large population of wild gray wolves. Although Yellowstone – the first national park of the U.S – was established in 1872, the existence of the wolf species in that region dates back thousands of years. However, increased hunting activities in the area led to a rapid decline and eventual extinction.
In a TED Talk titled "Can AI Help Us Speak with Wolves?", biolinguistics expert Jeffrey T. Reed explained the science behind wolf communication. When a wolf howls, they are essentially saying: “Here I am; where are you?” He also revealed that wolves can identify one another from their howls. "With the help of AI, we are learning more about how wolves howl," he added. "Software algorithms help us find signal in that noise," Reed explained. Yellowstone's earlier managers were either indifferent or unaware of the importance of wolves in an ecosystem.
They only saw threat and danger in that species in the early 20th century; therefore, conservative efforts didn't seem necessary. Moreover, the hunting practices surged, and by the 1940s, the wolf population in the Yellowstone area and the lower 48 states of the country was nearly extinct, as per Good Good Good. In the late 1990s, conservation efforts and reintroduction programs revived the species, but it has been challenging ever since. Although responsible hunting practices have been imposed, the threat to the wolf population hasn't stopped.
Reed believes that the conservation approach is outdated and not enough to revive a large population and keep them protected. The biolinguist has founded The Cry Wolf Project, a project that combines AI and bioacoustics while helping government, biologists, and conservationists learn more about wolf populations. The project has collected wild wolf vocalizations from around the world, the largest ever repository of wolf sounds. The sounds of wild wolves are captured through camouflaged battery-operated autonomous recording units. After the base data is collected, artificial intelligence takes over.
Spectrogram – a visual representation of audio – is created with the collected data, and AI looks for patterns that can help scientists map the location of these wolves. The recording helps detect wolves in the nearby areas, how many are present at one location, and what they might be communicating among one another. “Every six kilometers, there are recorders running 24/7. We know that wolves from one pack can count or estimate the number of wolves howling in a rival pack without seeing them," Reed told the outlet.
"So, we’re trying to see if we can count wolves just from audio recordings and improve census estimates of wildlife," he added. Reed's initiative is not funded by the government but by the national park's non-profit organization, Yellowstone Forever. In some states, wolves are protected, while in others, they might face removal from the Endangered Species Act. That's why coming up with an effective conservation method is especially important in these regions. Hidden cameras could be efficient to a certain extent, but the Autonomous Recording Unit (ARU) can collect wolf howls up to six miles away, making it a better approach.
More on Green Matters
Researchers Put GPS Tags on Animals in Yellowstone — and Found Surprising Responses to Rising Heat
Voyageurs National Park Wolf Population Just Hit an 11-Year Low, And There's One Key Reason For It
How Does AI Use Water and Energy? Unpacking the Negative Impact of Chatbots