The tale is embedded in the footprints. Along the edges of a vanished ice age lake are the fossilized tracks of people who lived among the mammoths, giant ground sloths and other Pleistocene mammals of ancient New Mexico. There were so many prehistoric pedestrians here that their feet pressed the seeds of a local plant called spiral ditch grass into their tracks, and these plant remnants are what has given archaeologists a possible time for when people lived here. Radiocarbon dating puts the age of the seeds at between 23,000 and 21,000 years ago—far older than expected.

Tracking the peopling of the Americas has been a difficult task, and archaeologists disagree over how and when humans arrived. Most agree people were present on these continents by 13,000 years ago, represented by what researchers refer to as the Clovis culture. But evidence from potentially older archaeological sites is often controversial and can be difficult to verify. The radiocarbon dates for the seeds from the New Mexico tracks would thus be the clearest evidence yet that people were making ancient North America their home more than 20,000 years ago.

The tracks, reported on Wednesday in Science by geologist Matthew Bennett of Bournemouth University in England and his colleagues, dot the flat lake deposits in New Mexico’s White Sands National Park. At least seven footprint sites have been found, including one with 37 prints. Most are from smaller-statured people with foot anatomy just like that of modern humans. This indicates, Bennett and his co-authors hypothesize, that most of the tracks were left by adolescents and children.

Following an initial assessment in 2019, U.S. Geological Survey researchers Kathleen Springer and Jeff Pigati visited White Sands in 2020 to dig trenches through the lake sediments. The goal was to better understand the area’s stratigraphy, or rock layers, and to find a way to accurately date ancient lake sediments. Prospects looked good for establishing a date for the tracks. “Multiple footprint horizons were in place, in section, with seed layers smashed by footprints in some cases,” Springer says. These spiral ditch grass seeds were radiocarbon-dated, delivering the time range of 23,000 to 21,000 years ago at the center of the new study.

“I think the evidence is very convincing and extremely exciting,” says University of Oxford radiocarbon expert Tom Higham, who was not involved in the new study. Often, he notes, getting radiocarbon dates from plant material can be challenging because of something called the reservoir effect: carbon absorbed by aquatic plants can carry an older signature than the carbon in the atmosphere, causing plant material to register incorrectly old dates. But Higham says the new paper considered and corrected for this possibly confounding effect.

The older-than-expected age of the White Sands tracks raises a litany of questions about how the people who left them might have arrived in the Americas. The tracks’ time range falls within what archaeologists and paleontologists know as the last glacial maximum (LGM), the period when the world’s ice sheets were at their greatest extent. The Bering land bridge, which at times allowed passage from Eurasia to North America, was entirely under ice during the LGM. This means that people living in White Sands at the time had to have crossed the bridge before the ice expanded, traveled along the ice sheets’ coastal edges or arrived by some other route. “In truth, this site is a bombshell,” says University of Alberta anthropologist Ruth Gruhn, who was not involved in the new study.

The ancient footprints may also inform discussions of what happened to North America’s large animals at the end of the Pleistocene. For decades, paleontologists, archaeologists and ecologists have been debating about the fate of animals such as the American mastodon and saber-toothed cats. Climate change, humans hungry for game or a combination of both are all considered possible causes. Finding old-enough footprints might associate people with some ice age extinctions that were thought to have happened before humans arrived in the Americas, Bennett and his co-authors propose—but then again, a longer cohabitation between people and megafauna would take some bite out of the idea that newly arrived humans rapidly decimated North America’s large mammals.

No single site can answer all the questions. But the age of the White Sands tracks means there are probably other very ancient archaeological clues waiting to be discovered. “Evidence from several other sites such as Meadowcroft Rockshelter [in Pennsylvania], Cactus Hill [in Virginia], Bluefish Caves [in the Yukon] and Gault [in Texas] shows a similar age range to White Sands,” Higham says, even though the archaeological interpretations of these places remain controversial. The new research could spur archaeologists to take another look and reconsider how and when people arrived in the Americas.

The White Sands tracks will join a growing number of clues that may substantially revise what archaeologists thought about how people came to live in these continents. “The number of pre–last glacial maximum sites is steadily increasing and creeping northward,” Gruhn says. Even within the park, there are more clues to find. “This basin is large and there are tracks everywhere,” Springer says. Archaeologists are only just beginning to follow in the footsteps of these early inhabitants of North America.