East Africa’s “Great Migration” is generally estimated to involve as many as 1.3 million wildebeest. But in reality, fewer than 600,000 of the animals might move across the Serengeti Mara landscape each year, according to an AI analysis of satellite images.

The Great Migration sees wildebeest, zebra and antelopes move between feeding and breeding grounds in Kenya and Tanzania, while also trying to dodge predators including lions, crocodiles and hyenas.

Assessing the number of animals involved is a tough task, traditionally achieved using crewed aerial surveys. Researchers can only survey a small area at a time, however, so they use statistical models to extrapolate densities across unsurveyed regions, which can introduce errors given herds are unevenly distributed and constantly on the move.

Satellite surveys can avoid these problems because one photo can cover hundreds of thousands of square kilometres, reducing the chance of double-counting animals and removing the need for extrapolative projections. It is impractical to manually count wildebeest in such vast images, but the animals can be tallied using AI instead. “Automation through AI does provide greater consistency and accuracy in counting,” says at the University of Oxford.

For the new study, Duporge and her colleagues trained two deep-learning models – U-Net and YOLOv8 – to identify wildebeest using a dataset of images in which 70,417 of the animals had been manually labelled. Both models were then applied to over 4000 square kilometres of high-resolution satellite imagery. The images were captured on 6 August 2022 and 28 August 2023.

The two AI models gave similar results, counting 324,202 and 337,926 wildebeest in 2022, and 502,917 and 533,137 in 2023. The substantial difference between the 2022 and 2023 figures reflects the fact the two surveys were conducted at different times of August: it was already known the number of animals involved in the migration rises and falls through the month. “[But] what’s encouraging is that both deep learning models, which use very different approaches, produced consistent results,” says Duporge. “That consistency strengthens our confidence in the reliability of AI-based counts for tracking wildlife populations.”

Free newsletter

Sign up to The Earth Edition

Sign up to newsletter

The previous estimate of 1.3 million produced from aerial surveys has been in place, largely unchanged, since the 1970s. “Based on our findings, I suspect the true population size is closer to around 800,000 individuals, if we were able to count every individual with zero error,” says Duporge. “I think the aerial count is an overestimate and ours is a slight underestimate. We miss those under the trees and there will be some outside of the study area, but it’s very surprising that we can see no more than 533,137.”

The extraordinary ways species control their own evolutionary fate

Natural selection isn't just something that happens to organisms, their activities also play a role, giving some species – including humans – a supercharged ability to evolve

Lower counts don’t necessarily mean wildebeest populations have collapsed. They may have altered their migration routes. But wildebeest do face significant challenges, including habitat loss and fragmentation due to agriculture. Accurate population estimates are crucial for informing targeted conservation efforts.

The research team previously trained an AI model to recognise elephants from satellite data, but this is the first time the approach has been used to conduct a census of individual mammals in a large, distributed population. The team is currently developing a similar method to detect and count African rhinos.

“I do believe we should be moving toward satellite and AI-based methods for counting wildlife populations, particularly for species that are widely distributed across large landscapes,” says Duporge.

The code for the researcher’s model has been made available at 

A tiny gecko has been rediscovered on RĂĄbida Island in the GalĂĄpagos, where it was believed to have been wiped out by invasive rats.

The leaf-toed gecko (Phyllodactylus maresi), whose adults are just 8 centimetres long, was previously only known to have existed on RĂĄbida from 5000-year-old fossil records. But teams collected live specimens during expeditions in 2019 and 2021, which have now been officially confirmed as this species.

The gecko’s reappearance was down to a successful restoration and rewilding project that began in 2011, led by the US-based non-profit in partnership with Galapagos National Park, Charles Darwin Foundation and The Raptor Center. The project, which also included 10 other islands in the Galápagos, used helicopters to distribute poisoned bait over large areas – a first in South America.

By 2012, invasive rats were confirmed as successfully removed on RĂĄbida, with nature immediately bouncing back, including the geckos. “We conducted extensive monitoring prior to the removal of invasive rats and they weren’t detected at all,” says Paula Castaño of Island Conservation. “We believe a small population held on all those years and then, without invasive predators, it finally had the opportunity to recover and grow its population. You can call it one of the most historic comebacks ever or just a long-overdue reappearance.”

Through DNA analysis, Castaño and her colleagues have determined that the Råbida population is closely related to P. maresi from nearby islands, but classified it as a distinct lineage, technically known as an evolutionarily significant unit, underscoring its importance for conservation.

The case of RĂĄbida demonstrates the benefits of island restoration and removing invasive species, says Castaño. “We found the ‘extinct’ gecko, rediscovered a snail last seen in 1906, as well as two other snail species being recorded on the island for the first time ever that are still being identified, and GalĂĄpagos hawks have returned and are thriving,” she says. “Nature is remarkable – allow it an opportunity to thrive and it comes roaring back.”

Free newsletter

Sign up to The Earth Edition

Sign up to newsletter

Ethiopian wolves feed on the sweet nectar of a local flower, picking up pollen on their snouts as they do so – which may make them the first carnivores discovered to act as pollinators.

The Ethiopian wolf (Canis simensis) is the rarest wild canid species in the world and Africa’s most threatened carnivore. Endemic to the Ethiopian Highlands, fewer than 500 individuals survive.

at the University of Oxford and her colleagues observed wild Ethiopian wolves lapping up the nectar of Ethiopian red hot poker (Kniphofia foliosa) flowers. Local people in the mountains have traditionally used the nectar as a sweetener for coffee and on flat bread.

The wolves are thought to be the first large carnivore species ever to be recorded regularly feeding on nectar.

“For large carnivores, such as wolves, nectar-feeding is very unusual, due to the lack of physical adaptations, such as a long tongue or specialised snout, and because most flowers are too fragile or produce too little nectar to be interesting for large animals,” says Lai.

The sturdy, nectar-rich flower heads of the poker plant make this behaviour possible, she says. “To my knowledge, no other large carnivorous predator exhibits nectar-feeding, though some omnivorous bears may opportunistically forage for nectar, albeit rarely and poorly documented.”

Some of the wolves were seen visiting as many as 30 blooms in a single trip. As they lick the nectar, the wolves’ muzzles get covered in pollen, which they could potentially be transferring from flower to flower as they feed.

“The behaviour is interesting because it shows nectar-feeding and pollination by non-flying mammals might be more widespread than currently recognised, and that the ecological significance of these lesser-known pollinators might be more important than we think,” says Lai. “It’s very exciting.”

Lai and her colleagues at the  now hope to dig deeper into the behaviour and its ramifications. “Trying to confirm actual pollination by the wolves would be ideal, but that would be quite challenging,” she says. “I’m also very interested in the social learning aspect of the behaviour. We’ve seen this year adults bringing their juveniles to the flower fields, which could indicate cultural transmission.”