AI Powers New Era Of Conservation

AI Unlocks Secrets of Elusive Wildlife The Nighthawk Study

Studying nocturnal birds, especially those skilled in camouflage, presents significant challenges for researchers. Elly Knight, a researcher at the University of Alberta, experienced this firsthand while studying common nighthawks. These medium-sized birds are highly mobile, making them incredibly difficult to track and understand, especially in vast areas like the boreal forest where, beyond traditional ecological knowledge, little is known about them.

To overcome these hurdles, Knight tapped into an extensive collection of wildlife sound recordings. Modern conservation efforts often involve autonomous recording units, and agencies had deployed many such devices across northeastern Alberta. While Knight had access to this rich audio data, a major obstacle remained: the sheer volume of recordings. Manually sifting through these sounds to isolate nighthawk calls from the myriad of other noises was impractical. Knight estimated that expert analysis could realistically cover only about 1 percent of the recordings, leaving a wealth of potential information untapped.

This is where artificial intelligence came into play. By applying AI to the massive acoustic dataset, Knight managed to shed considerable light on the nighthawks' world. The AI-driven analysis revealed crucial information about where the birds lived, their seasonal presence, and differences in their foraging and nesting habitats.

Further refinement of A.I. will expand its applications, providing a far more detailed portrait of species being studied.

Knight believes this technology significantly broadens the scope of what can be studied. Currently, AI can determine species presence or absence, offering valuable ecological insights. However, she anticipates that further advancements will enhance its capabilities, including identifying individual birds, which would provide a much more detailed ecological picture. Knight is already working on extending this AI-based approach to a wider range of boreal bird species.

The AI Revolution in Conservation A Paradigm Shift with Caveats

The use of artificial intelligence is rapidly expanding throughout conservation science, ushering in swift, dramatic changes and the promise of more to come. A recent paper, titled “The Potential for AI to Revolutionize Conservation,” underscores this trend. Elly Knight concurs, calling it a “paradigm shift.”

However, some scientists caution that AI's increasing role in conservation carries notable drawbacks. Beyond the technology's well-documented high consumption of water and energy, there are concerns about its potential to perpetuate errors and biases. Since AI synthesizes existing information largely from web-based, Western academic sources, it may overlook or exclude vital traditional and Indigenous knowledge.

A gray wolf howls in Yellowstone National Park, where a new project will use A.I. to analyze sound recordings of wolves. Credit: Dennis W. Donohue

AI has also faced criticism for potentially acting as a technological barrier, distancing people from direct engagement with wildlife and their natural environments. Hamish van der Ven, a professor at the University of British Columbia and a prominent critic of AI's proliferation in conservation and other fields, stated, “If I could wave a wand and un-invent A.I., I would.”

Despite these concerns, the momentum behind AI in conservation is undeniably strong.

AI in Action Global Conservation Success Stories

Thousands of researchers globally are employing AI to advance biological research and conservation projects. For instance, BioCarbon Engineering in the U.K. uses AI-equipped drones to map forests and strategically plant seeds in optimal habitats. Around the world, AI is also instrumental in tracking diseases affecting wildlife.

In Yellowstone National Park, a new initiative by Colossal Biosciences and Yellowstone Forever aims to integrate audio and visual data to create acoustic fingerprints of wolves. This involves identifying individual howls, chorus howls, growls, and barks to noninvasively monitor packs, their movements, and behaviors. The system can also detect the sound of gunshots, enabling rapid response to potential illegal wolf killings.

Harnessing Big Data How AI Empowers Citizen Science and Management

“The bottleneck has shifted from hard-to-collect data to making sense of the enormous amount data at our fingertips.”

Millions of images collected by individuals using nature apps like iNaturalist and eBird are contributing to the mountains of raw data that AI algorithms can process. Ali Swanson, director of nature, tech, and innovation for Conservation International, observes, “We’re drowning in data, and one of the big challenges is making sense of the information. The advancements we’ve seen in the last three to five years have really blown the top off what is possible with A.I.”

iNaturalist, often called iNat, is a smartphone application that empowers anyone to gather photos of global biodiversity—from plants and insects to birds and mammals. These photos are instantly analyzed by AI, which identifies the species for the user. iNat has become a significant force in biodiversity research. Its users have rediscovered species unseen for decades and have been discovering approximately one new species a month. Recently, an app user in Australia found a new praying mantis species, which was subsequently named Inimia nat in honor of the app.

• Inimia nat, a species of praying mantis discovered using the iNaturalist A.I., which analyzes images of wildlife. Credit: Brendan James*

The iNat library now boasts half a billion images. This data, freely available to scientists, has supported over 6,000 scientific studies. The key to unlocking this vast kingdom of data is AI's ability to rapidly find and process information from these images.

Scott Loarie, iNaturalist’s executive director, explains, “One researcher looked at 10,000 pictures of flowering Joshua trees and ran that through an A.I. model to understand how climate change was impacting the phenology or morphology or changing its distribution.” He adds, “A.I. is really good at looking for patterns in big messy data sets that are unstructured,” like iNat's dataset, which is messy yet large due to its volunteer-driven nature. “We are helping biodiversity enter the big data world,” Loarie states. “Biodiversity is still in this world where you go to a museum and open a drawer and pull out a couple of specimens. We have hundreds of millions of records representing one of four named species on the planet.”

AI is also incredibly effective for making timely management decisions. Sara Beery, an MIT assistant professor specializing in AI and conservation, cites an example: “Idaho Fish and Game collected 2 million camera trap images a year.” Previously, analyzing this data to determine population levels took so long that “they were making hunting quotas that were five years out of date.” With AI, four people can now process 18 million images collected over a year in just a couple of weeks. “Now they are making their policies and decisions the year the data is collected,” Beery says, “which is incredibly important given how quickly everything is changing.”

Enhancing Protection and Prediction AIs Role in Safeguarding Species

In Madhya Pradesh’s Kanha-Pench Corridor in India, TrailGuard AI camera traps are deployed to protect vulnerable wildlife, including over 300 tigers—Central India's largest population. This protected area is also home to some 600,000 people. When tigers prey on livestock, retaliatory killings can occur. Now, when a TrailGuard camera captures an image of wildlife, it instantly identifies the species and transmits this data to forest rangers. If a tiger or another predator is detected, rangers can quickly inform local livestock owners, enabling them to move their animals to safety.

Predictive modeling for conservation also receives a significant boost from AI. By analyzing a multitude of variables, AI can generate far more complex and accurate models of potential ecological outcomes. These models can then guide decisions on land preservation or investments in resource protection.

A.I. can analyze images to identify not just distinct species, but individual animals, tracking their movements and posture. Credit: Tuia et al.

Conservation International’s Swanson illustrates this: “We could integrate diverse data — like coastal elevation maps, historical storm patterns, soil hydrology, and human population density — to simulate how restored mangrove ecosystems might mitigate flood risk under different climate trajectories, and compare them to traditional engineered approaches.”

Critical Perspectives Navigating AIs Limitations and Ethical Dilemmas

Despite its strengths, AI-generated content and approaches have limitations, experts warn. In a recent study, van der Ven and his students asked AI chatbots to outline causes, consequences, and solutions for nine environmental challenges. He found that “because it’s been trained on past data, if you ask a chatbot what should we do about biodiversity loss or climate change, you get bits of actions that have previously been attempted,” such as public education and awareness campaigns.

“Those kinds of solutions are drastically outside the scope of urgency of many different [environmental] challenges,” van der Ven noted. “There’s only so far you can get looking to the past to describe future actions, which is what large language models [the most commonly used type of A.I.] do.” Human ingenuity and critical decision-making remain indispensable.

Because A.I. relies heavily on existing data from wealthy countries, answers it produces are skewed toward that perspective.

Another significant concern is data bias. Since AI models heavily rely on existing data, predominantly from affluent countries with a Western academic focus, their outputs can be skewed, potentially discounting alternative approaches like traditional ecological knowledge.

Some scientists argue that much is lost when nature is viewed solely as data points in a computer model. Denver Holt, an experienced owl biologist who has studied long-eared owls for 37 years and snowy owls for 33 years, acknowledges technology's utility but stresses the importance of fieldwork. “You can get a much better understanding of the animal and the ecosystem if you go out in the field,” he says. “I know people who are modeling owls and owl habitat who have never seen an owl.”

“I know people who are modeling owls and owl habitat who have never seen an owl,” says a biologist who studies owls in the field.

Being in the field, Holt explains, “we generate new ideas and new questions.” A recent paper, titled “Extinction of Experience Among Ecologists,” warns that “a decline in fieldwork could hinder scientific progress in some areas of ecology, especially those that rely heavily on direct wildlife observation, such as behavioural ecology, species inventories, and biodiversity monitoring.”

Van der Ven further criticizes the tendency to laud AI's conservation applications while overlooking its broader societal impacts. “There’s been a deluge of academic research on applications of A.I. to conservation,” he stated. “But critical reflections on what are the costs to conservation in terms of what A.I. is being more commonly used for — things like driving conspicuous consumption, getting people to follow through to recommended links on Amazon and buying more stuff, and targeted advertising — is lacking. A lot of the environmental challenges we face today are the consequence of growth-oriented capitalism.”