How AI Can Help Protect Endangered Animals - Updated

New software alerts boats about nearby whales

  • A new AI-powered system in the San Francisco Bay Area helps keep ships from hitting endangered whales. 
  • A growing number of programs are using AI to monitor and protect animals. 
  • AI-powered audio recorders are used in the rainforest and can detect gunshots and chainsaw sounds to prevent illegal deforestation.
Whale watching boat in Iceland, with a stunning Humpback Whale tail out of water

Francesco Riccardo Iacomino / Getty Images

Whales may be massive, but they sometimes need help from software to keep boats from accidentally hitting them.

The Whale Safe project, newly installed in the San Francisco Bay Area, uses artificial intelligence (AI) and real-time data on how many whales are in an area to alert shipping companies to slow their boats in the presence of marine mammals. It's part of a growing worldwide movement to use AI to help protect endangered species.

"Artificial Intelligence is especially useful to protect animals because it can make patterns in nature visible to humans that would normally remain undetected," Adam Porter, a University of Maryland computer science professor, told Lifewire via email. "AI can handle huge amounts of data and find correlations—e.g., between pollutants and animal behavior—[and] allow for interpretation, enable prediction, and provide valuable insights to researchers.

"These immediate insights can help people and organizations strategize on how best to protect environments for animals, using sound data to drive their decision-making."

Keeping Wales Secure

The creators of Whale Safe technology claim it will allow anyone to go online and, in near real-time, monitor ship speeds and whale presence in Northern California’s coastal waters. More than 50 percent of container ship traffic coming to and from the United States passes through West Coast ports, and blue, fin, humpback, and gray whales are vulnerable to ship strikes as they migrate and feed in areas that overlap shipping lanes and routes. Scientists estimate that over 80 endangered whales are killed by ship strikes off the West Coast each year.

Whale Safe uses an AI-enabled acoustic monitoring system, big data models, and direct whale sightings recorded by trained observers and citizen scientists. The data is combined into a "Whale Presence Rating," ranging from low to very high whale activity. 

"Whale-vessel collisions are a global concern, so when addressing the problem and building the Whale Safe system, we wanted it to be a blueprint to allow for replication and expansion into other regions. We are excited to expand the technology and expertise to the San Francisco Bay region where ship collisions are of high concern for endangered whales," Callie Steffen, Whale Safe project lead at the Benioff Ocean Science Laboratory, said in the news release.

For the Birds

AI isn't only useful to marine life, however, as it can also help monitor vulnerable bird populations. For example, BirdNet, an AI-powered system created by the Cornell Lab of Ornithology and Chemnitz University of Technology, analyzes recordings of bird songs. 

AI can help protect animals by allowing researchers to scale up monitoring to levels that are difficult to achieve with traditional in-person survey methods, Jerry Cole, a staff biologist at The Institute for Bird Populations, told Lifewire in an email interview.

"Often the easier task in wildlife monitoring is collecting monitoring data, either via photographs, recordings, or other means," he added. "But when a large amount of data is collected, it can take thousands of hours and a skilled observer to process that information—whereas, with a well-trained AI, a researcher can quickly comb through mountains of data for a species of interest."

Male Resplendent Quetzal leaving nest

©Juan Carlos Vindas / Getty Images

The non-profit Rainforest Connection has a system of small, solar-powered audio recorders that are deployed in rainforest canopies and can detect gunshots and chainsaw sounds in real-time and send a message to notify rangers of illegal activities such as logging or poaching. In addition, using recordings from these units, an AI can identify which species are present in different regions of the rainforest.

AI is also helping protect vultures in Namibia. The prototype of a new animal tag system at a Berlin Zoo recently completed its maiden flight. The tags are equipped with sensor-based AI, a camera, energy-efficient electronics, and satellite-based communication technology. The tags detect and transmit animal behavior and can act as an early warning system for ecological changes.

"The processing of the data by an AI directly on the tag is a key feature, as it can be used to decide in real-time which of the collected data are relevant enough to send to the satellite," wildlife researcher Nina Holzer said in a news release. "Sending all the collected data in real-time is virtually impossible, a satellite connection doesn't allow for that and would consume too much power."

With many species headed for extinction, endangered animals need all the technological help they can get, Cole said. "When data can be processed more quickly, we can determine where a species of interest is located and take actions to protect those individuals, for instance, slowing boat traffic when whales are present," he added.

Update: Some researchers are also using drones combined with AI to identify animals from the air, often to count them. The nonprofit WildTrack flies drones to collect data about endangered species without disturbing them by imaging the cryptic ground evidence they leave behind - for example, their footprints.

The group is currently tracking endangered African rhinos to better understand where their numbers are dwindling. WildTrack recently published a study of their methods.

The drones fly overhead and collect video of rhinos, the group wrote on its blog. “An AI algorithm is trained to identify rhinos in the video frames and then reports back via a text message to wildlife managers on the ground who are able, with knowledge of the rhino locations, to protect them more effectively with anti-poaching unit deployment. Until now, it was difficult to do this in a low-connectivity setting, but we worked around this with SMS text messages and more efficient AI models.”

Was this page helpful?