A new way to count wildebeest Until now, scientists have used airplanes to count the herds from above. They flew along set lines, photographed the animals below, and used math models to estimate how many were in the areas they didn\u2019t see directly. It was the best method available, but not perfect. <\/p>\n Wildebeest are constantly on the move. If a group shifted from one spot to another, they could be counted twice \u2013 or missed altogether. That\u2019s where satellites come in.<\/p>\n Researchers turned to high-resolution satellite imagery that can snap photos covering hundreds of thousands of square miles in one shot. These images don\u2019t scare the animals and don\u2019t require dangerous low-flying planes. <\/p>\n But there\u2019s a catch: there are way too many images for humans to sift through and count animals by hand. So, the team turned to artificial intelligence<\/a>. Two powerful deep-learning models were trained to recognize the tiny shapes of wildebeest from space. <\/p>\n The shapes appear as blobs just 6 \u2013 12 pixels wide, which is barely visible even when zoomed in. The models, called U-Net and YOLOv8, were trained on over 70,000 examples of labeled wildebeest and reached high accuracy scores of up to 0.83, which is considered excellent for this kind of work.<\/p>\n What the AI found<\/p>\n Once trained, the AI models got to work analyzing satellite photos taken in August 2022 and 2023 over the Masai Mara region of Kenya and Tanzania. The results came as a surprise. <\/p>\n In 2022, the models counted between 324,202 and 337,926 wildebeest. In 2023, they found between 502,917 and 533,137. That\u2019s a huge gap compared to the long-standing estimate of 1.3 million wildebeest \u2013 a number that\u2019s been around since the 1970s.<\/p>\n The models were created and run by a team that included researchers from Oxford University\u2019s Wildlife Conservation Research Unit. They believe the new counts are more reliable because satellite images don\u2019t require any guesswork about where the animals might be.<\/p>\n Where did the wildebeest go?<\/p>\n There\u2019s no clear answer yet about what happened to the \u201cmissing\u201d wildebeest. The study was led by Dr. Isla Duporge from the University of Oxford<\/a>.<\/p>\n \u201cThe sheer difference between traditional estimates and our new results raises questions about where the \u2018missing\u2019 wildebeest might be,\u201d said Dr. Duporge.<\/p>\n \u201cBased on data from GPS tracking surveys, we are confident that most of the herd were contained within the surveyed area. And whilst some individuals may have been obscured by tree cover, it seems unlikely that such a large number \u2013 on the order of half a million \u2013 would have been concealed in this way,\u201d <\/p>\n Possible changes in migration<\/p>\n It\u2019s possible that wildebeest aren\u2019t vanishing \u2013 they may just be moving differently than before. Changes in rainfall due to climate change could be shifting where the best grazing areas are. <\/p>\n Human activities like farming, roads, and fences are also blocking traditional migration<\/a> paths. If the herds are splitting or rerouting, they may be harder to spot in one place.<\/p>\n Also, current satellite images can\u2019t easily tell a wildebeest apart from other similar animals like zebra<\/a> or eland. So the count might actually be slightly higher than it should be, not lower.<\/p>\n Importance of accurate data<\/p>\n The Great Migration<\/a> doesn\u2019t just involve wildebeest. Lions, crocodiles, hyenas, and other predators rely on this moving buffet for survival. The migration also draws visitors from all over the world and brings in millions of dollars to local communities in Kenya and Tanzania through tourism.<\/p>\n If the herds really are shrinking, it could ripple through the entire ecosystem. That\u2019s why having accurate data is so important.<\/p>\n \u201cThe field of wildlife conservation relies on having accurate data on wildlife population numbers,\u201d said Dr. Duporge.<\/p>\n \u201cCombining earth observation satellite data with deep learning, this study has revolutionized our understanding of migratory wildebeest numbers, and could open the floodgates for surveying other species using this method.\u201d<\/p>\n What else can AI spot from space?<\/p>\n This isn\u2019t the first time AI and satellites have teamed up for conservation. The same research team had earlier trained a model to spot elephants from space. But this new wildebeest study is the first to count individual animals in a large, spread-out population.<\/p>\n According to the team, the same technology could soon be used to track other herd animals like reindeer<\/a>, camels, and zebra. They\u2019re even working on a new method to detect African rhinos<\/a> \u2013 one of the most endangered animals on the planet.<\/p>\n Professor David Macdonald, a co-author of the study, noted that the most basic fact to know as a foundation for conserving any species is how many of them there are. <\/p>\n \u201cThe technological breakthrough of our study \u2013 satellite-based wildlife monitoring, powered by AI \u2013 potentially revolutionises the answer for wildebeest, besides opening up incredible possibilities for monitoring other large species,\u201d said Professor MacDonald.<\/p>\n This new way of counting animals could change how we protect wildlife. But for now, it leaves us with a big, unsettling question: what happened to hundreds of thousands of wildebeest that we thought were there? We may need to rethink what we know about one of the planet\u2019s most iconic migrations.<\/p>\n The full study was published in the journal PNAS Nexus<\/a>.<\/p>\n \u2014\u2013<\/p>\n Like what you read? Subscribe to our newsletter<\/a> for engaging articles, exclusive content, and the latest updates.\u00a0<\/p>\n Check us out on EarthSnap<\/a>, a free app brought to you by Eric Ralls<\/a> and Earth.com.<\/p>\n \u2014\u2013<\/p>\n","protected":false},"excerpt":{"rendered":"Every year, across the vast grasslands of East Africa, over a million wildebeest are believed to thunder across…\n","protected":false},"author":2,"featured_media":23810,"comment_status":"","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[26],"tags":[111,139,69,147,406],"class_list":["post-23809","post","type-post","status-publish","format-standard","has-post-thumbnail","category-wildlife","tag-new-zealand","tag-newzealand","tag-nz","tag-science","tag-wildlife"],"_links":{"self":[{"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/posts\/23809","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/comments?post=23809"}],"version-history":[{"count":0,"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/posts\/23809\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/media\/23810"}],"wp:attachment":[{"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/media?parent=23809"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/categories?post=23809"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.newsbeep.com\/nz\/wp-json\/wp\/v2\/tags?post=23809"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}
\n
\n ![\"EarthSnap\"\/](\"https:\/\/www.newsbeep.com\/nz\/wp-content\/uploads\/2025\/09\/1757887565_481_earthsnap-banner-news.webp.webp\")
\n<\/a><\/p>\n