Following the path of digitalization in Slovenia and Europe: Technology that saves wildlife and mitigates climate change

In a recent report, artificial intelligence (AI) was identified as one of the three most important emerging technologies in nature conservation and protecting endangered species. This technology is increasingly being used to analyse vast amounts of information gathered by conservationists. It can do the work of hundreds of people and thus achieve faster, cheaper, and more efficient results.

In previous articles on artificial intelligence, we explained how this technology works and how it is used to optimise supply chain management. However, in this article, we will present five examples of how artificial intelligence saves wildlife and cares for our planet.

A recent report by the Wildlabs community found that AI was one of the top three emerging technologies in conservation. “AI can learn how to identify which photos out of thousands contain rare species, or pinpoint an animal call out of hours of field recordings – tremendously reducing the manual labour required to collect vital conservation data,” the report says.

Artificial intelligence is helping to protect various species, such as humpback whales, koalas, and snow leopards, supporting the work of scientists, researchers, and rangers in vital tasks, from anti-poaching patrols to monitoring species. With the help of machine learning (ML), which we have already written about, AI can often do the job faster, cheaper, and more efficiently.

Zambia’s Kafue National Park: a virtual ‘fence’ across the lake was created to to stop poaching and illegal fishing. Photo source:
Zambia’s Kafue National Park: a virtual ‘fence’ across the lake was created to to stop poaching and illegal fishing.
Photo source:

Stopping poachers

Zambia’s Kafue National Park is home to more than 6,600 African savannah elephants and covers 22,400 square kilometres, so stopping illegal poaching is a major logistical challenge. Another problem is illegal fishing in the lake on the park’s border, where poachers dress as fishers and thus enter and exit the park, often at night, unnoticed.

The non-profit organisation Game Rangers International (GRI) and Zambia’s Department of National Parks and Wildlife are implementing an initiative to enhance its conventional efforts to prevent poaching. By doing so, they can record every boat crossing in and out of the park, day or night.

Photo source: Game Rangers International.
Photo source: Game Rangers International.

The cameras installed in 2019 were initially monitored manually by park rangers, who could then respond to signs of illegal activity. Artificial intelligence, however, is now capable of automatically detecting boats entering the park, which increases efficiency and reduces the need for constant human surveillance. Waves or birds can also trigger warnings, but AI is being taught to eliminate these false readings.

“There have long been insufficient resources to secure protected areas, and having people watch multiple cameras 24/7 doesn’t scale,” Ian Hoad, special technical adviser at GRI, explained to The Guardian. “AI can be a gamechanger, as it can monitor for illegal boat crossings and alert ranger teams immediately. The technology has enabled a handful of rangers to provide around-the-clock surveillance of a massive illegal entry point across Lake Itezhi-Tezhi.”

Machine learning models were used to recognise humpback whale songs. Photo source: Pixabay.
Machine learning models were used to recognise humpback whale songs.
Photo source: Pixabay.

Finding humpback whales and protecting koalas

Knowing the location of whales is the first step in introducing measures to protect them, such as in marine protected areas. Locating humpbacks visually across vast oceans is difficult, but their distinctive singing can travel hundreds of miles underwater. The U.S. National Oceanic and Atmospheric Association (NOAA) uses acoustic recorders to monitor marine mammal populations in the Pacific Islands, says Ann Allen, a research oceanographer at NOAA. “In 14 years, we’ve accumulated around 190,000 hours of acoustic recordings. It would take an exorbitant amount of time for an individual to identify whale vocalisations manually.”

In 2018, NOAA worked with Google AI for Social Good’s bioacoustics team to create a machine learning model that could recognise humpback whale songs. “We were very successful in identifying humpback song through our entire dataset, establishing patterns of their presence in the Hawaiian Islands and Mariana Islands,” says Allen. She explained that they also found a new occurrence of humpback song in certain areas where humpbacks had not been documented before, adding: “This comprehensive analysis of our data wouldn’t have been possible without AI.”

Another example comes from Australia, where koala populations are in severe decline due to the destruction of their natural habitats, domestic dog attacks, road accidents, and bushfires. Without knowing their number and location, saving them is challenging. Grant Hamilton, associate professor of ecology at Queensland University of Technology (QUT), has set up a conservation AI hub to count koalas and other endangered animals.

The AI algorithm uses drones and infrared imaging to rapidly analyse the infrared footage and determine whether a heat signature is a koala or another animal. Photo source:
The AI algorithm uses drones and infrared imaging to rapidly analyse the infrared footage and determine whether a heat signature is a koala or another animal.
Photo source:

The AI algorithm uses drones and infrared imaging to rapidly analyse the infrared footage and determine whether a heat signature is a koala or another animal. Hamilton used the system after Australia’s devastating bushfires in 2019 and 2020 to determine surviving koala populations.

“This is a gamechanger project to protect koalas. ” Hamilton explained that “powerful AI algorithms can analyse countless hours of video footage and identify koalas from many other animals in the thick bushland”. This system will allow conservation groups and organisations protecting and monitoring species to survey large areas anywhere in Australia and send the data back to QUT to process it. “We will increasingly see AI used in conservation. In this current project, we simply couldn’t do this as rapidly or as accurately without AI.”

Animal rescue and water tracking

Saving species on the brink of extinction in the Congo Basin, the world’s second-largest rainforest, is a considerable task. In 2020, the Polish data science company Appsilon developed the Mbaza AI image classification algorithm for large-scale biodiversity monitoring in collaboration with some other agencies.

Classifying millions of photographs captured by environmentalists with the help of automated cameras and identifying different animal species is time-consuming work. At the same time, time is of the essence, as about 150 elephants die every month in the African region due to wildlife hunters. The Mbaza AI algorithm analysed more than 50,000 images collected from 200 camera traps spread across 7,000 sq. km of forest in 2020. The Mbaza AI algorithm classifies up to 3,000 images an hour with up to 96% accuracy.

Conservationists can thus monitor and track animals and quickly spot irregularities or warning signs, allowing them to act swiftly if necessary. The algorithm also works offline on an ordinary laptop, useful in locations without or with poor internet connectivity. In Gabon, the government and national parks agency plan to install automatic cameras across the country, and Mbaza AI can help speed up data analysis for all these projects.

However, artificial intelligence can also help track water loss. Brazil has lost more than 15 per cent of its surface water in the past 30 years, which has only become apparent with the help of AI. The country’s rivers, lakes, and wetlands are facing increasing pressure due to growing population, economic development, deforestation, and the worsening effects of the climate crisis.

Aerial view of the Pantanal, Brasil.
Photo source: R. Isotti, A. Cambone, Homo Ambiens WWF.

However, no one knew the scale of the problem until last August, when the MapBiomas water project, using machine learning, published its results after processing more than 150,000 images generated by NASA’s Landsat 5, 7 and 8 satellites from 1985 to 2020 across the 8.5m sq. kilometres of Brazilian territory. Artificial intelligence has thus helped identify a large-scale problem that might otherwise have been noticed too late, if at all.

Author: Rok Žontar

Keywords: AI, wildlife, climate change, water tracking, endangered species, monitoring


This article is part of joint project of the Wilfried Martens Centre for European Studies and the Anton Korošec Institute (INAK) Following the path of digitalization in Slovenia and Europe. This project receives funding from the European Parliament. 

The information and views set out in this article are those of the author and do not necessarily reflect the official opinion of the European Union institutions/Wilfried Martens Centre for European Studies/ Anton Korošec Institute. Organizations mentioned above assume no responsibility for facts or opinions expressed in this article or any subsequent use of the information contained therein.