THere’s a line of thought, from sci-fi movies to Stephen Hawking, that suggests artificial intelligence (AI) could be the downfall for humans. However, conservationists are increasingly turning to AI as an innovative technical solution to address the biodiversity crisis and mitigate climate change.
A recent report by Wildlabs.net found that AI is one of the top three emerging technologies in conservation. From camera traps and satellite imagery to audio recordings, the report states: “AI can learn how to recognize which photos contain rare species from thousands; or locate an animal call from hours of field surveys – vastly reducing the manual labor required to collect vital conservation data.”
AI helps protect species as diverse as humpback whales, koalas and snow leopards, and supports the work of scientists, researchers and rangers on vital tasks from anti-poaching patrols to species monitoring. With machine learning (ML) computer systems that use algorithms and models to learn, understand, and adapt, AI is often able to do the work of hundreds of people and produce faster, cheaper, and more effective results.
Here are five AI projects contributing to our understanding of biodiversity and species:
1. Stop poachers
Kafue National Park in Zambia is home to more than 6,600 African savannah elephants and stretches over 22,400 km². Therefore, stopping poaching is a major logistical challenge. Illegal fishing in Lake Itezhi-Tezhi on the park’s border is also a problem, and poachers pose as fishermen to enter and exit the park undetected, often under cover of darkness.
The Connected Conservation Initiative, by Game Rangers International (GRI), Zambia’s Department of National Parks and Wildlife and other partners are using AI to improve conventional anti-poaching efforts and erecting a 19km virtual fence over Lake Itezhi-Tezhi. Forward-looking infrared (FLIR) thermal imaging cameras record every boat entering and exiting the park, day and night.
The cameras installed in 2019 were manually monitored by rangers, who could then respond to signs of illegal activity. FLIR AI has now been trained to automatically detect boats entering the park, increasing effectiveness and reducing the need for constant manual monitoring. Waves and flying birds can also set off alarms, so the AI is taught to eliminate these false readings.
“For a long time, there have been insufficient resources to secure protected areas, and having people under 24/7 surveillance from multiple cameras is not enough,” says Ian Hoad, special technical adviser at GRI. “AI can be a game changer as it can monitor illegal boat crossings and alert ranger teams immediately. The technology has enabled a handful of rangers to conduct 24/7 surveillance of a massive illegal access point over Lake Itezhi-Tezhi.”
2. Water loss tracking
Brazil lost more than 15% of its surface water a crisis in the past 30 years that only came to light with the help of AI. The country’s rivers, lakes and wetlands are under increasing pressure from a growing population, economic development, deforestation and the worsening impact of the climate crisis. But nobody knew the extent of the problem until last August, when the MapBiomas water project, using ML, published its findings after processing more than 150,000 images collected from 1985 to 2020 by NASA’s Landsat 5, 7 and 8 satellites on a Area of 8.5 sq km were generated km of Brazilian territory. Without AI, researchers could not have analyzed water changes across the country at the scale and level of detail required. AI can also differentiate between natural and man-made bodies of water.
The Negro River, a major tributary of the Amazon and one of the 10 largest rivers in the world by volume, has lost 22% of its surface water. The Brazilian portion of the The Pantanal, the world’s largest tropical wetland, has lost 74% of its surface water. Such losses are devastating for wildlife (4,000 plant and animal species live in the Pantanalincluding jaguars, tapirs and anacondas), humans and nature.
“The AI technology gave us a startlingly clear picture,” says Cássio Bernardino, director of WWF-Brazil’s MapBiomas water project. “Without AI and ML technology, we never would have known how serious the situation was, let alone had the data to convince people. Now we can take steps to address the challenges this loss of surface water poses to Brazil’s incredible biodiversity and communities.”
3. Find whales
Knowing where whales are is the first step in taking measures like marine sanctuaries to protect them. It is difficult to visually locate humpback whales across vast oceans, but their distinctive song can travel hundreds of miles underwater. at National Oceanic and Atmospheric Association (Noaa) Pacific Island Fisheries, acoustic recorders are used to monitor populations of marine mammals on remote and hard-to-reach islands, says Ann Allen, Noaa’s research oceanographer. “In 14 years we have collected around 190,000 hours of acoustic recordings. It would take an exorbitant amount of time to manually identify whale vocalizations.”
In 2018, Noaa entered into a partnership with Google AI for social purposes bioacoustics team Create an ML model that might recognize humpback whale song. “We have been very successful in identifying humpback song throughout our dataset and identifying patterns of their presence in the Hawaiian Islands and the Mariana Islands,” says Allen. “We also found a new occurrence of humpback whale song at Kingman Reef, a site that had never previously documented humpback whale presence. This comprehensive analysis of our data would not have been possible without AI.”
4. Protect koalas
Australia’s koala populations have plummeted due to habitat destruction, domestic dog attacks, road accidents and bushfires. Without knowing their numbers and whereabouts, rescuing them is a challenge. Grant Hamilton, Associate Professor of Ecology at Queensland University of Technology (QUT), has created one AI Center for Conservation with federal u Country maintenance Australia Funding to count koalas and other endangered animals. Using drones and infrared imagery, an AI algorithm quickly analyzes infrared footage and determines whether a heat signature is a koala or another animal. Hamilton used the system after Australia’s devastating bushfires in 2019 and 2020 to identify surviving koala populations, particularly on Kangaroo Island.
“This is a landmark project for koala conservation,” says Hamilton. “Powerful AI algorithms are able to analyze countless hours of video footage and distinguish koalas from many other animals in the dense bushland. This system will enable landcare groups, conservation groups and organizations working to protect and monitor species to survey large areas across Australia and send the data back to us at QUT for processing.
“We will increasingly see AI being used in conservation,” he adds. “In this current project, we just couldn’t do it as quickly or as accurately without AI.”
5. Count species
Saving endangered species in the Congo Basin, the second largest rainforest in the world, is a daunting task. In 2020, data science company Appsilon partnered with the University of Stirling in Scotland and the Gabon National Parks Authority (ANPN) to develop the Mbaza AI image classification algorithm for a large-scale biodiversity monitoring in the national parks Lopé and Waka in Gabon.
Conservationists had used automated cameras to capture species including African forest elephants, gorillas, chimpanzees and pangolins, which then had to be identified manually. Classifying millions of images can take months or years, and in a country that loses about 150 elephants to poachers every month, time is of the essence.
The Mbaza AI algorithm was used in 2020 to analyze more than 50,000 images collected by 200 camera traps spread over 7,000 square kilometers of forest. Mbaza AI classifies up to 3,000 frames per hour and is 96% accurate. Conservationists can monitor and track animals and quickly spot anomalies or warning signs, allowing them to act quickly if necessary. The algorithm also works offline on an ordinary laptop, which is useful in places with no or poor internet connection.
“Many Central African forest mammals are threatened by unsustainable trade, land use change and the global climate crisis,” says Dr. Robin Whytock, Postdoctoral Fellow at the University of Stirling. “Appsilon’s work on the Mbaza AI app enables conservationists to quickly identify and respond to threats to biodiversity. The project started with 200 camera traps in Gabon’s Lopé and Waka National Parks, but hundreds more have since been deployed by various organizations in West and Central Africa. In Gabon, the government and national park authority are aiming to deploy cameras across the country. Mbaza AI can help all these projects to accelerate data analysis.”