This image shows mammoth vertebrae in ice in Yukon Territory, Canada. (Kieren Mitchell)

Ever since humans uncovered the fossil records of the mammoths, giant sloths and other extinct megafauna that walked the Earth until about 11,000 years ago, scientists have been trying to solve the mystery of why they vanished. Was it human hunting and activity? Or did climate change bring about their end?

A new study suggests that the key to solving the question lies in a series of abrupt warming events over the course of the Late Pleistocene era, going back at least 50,000 years.

The study, published in Science on Thursday, is the first to link specific climatic events to localized extinctions of megafauna. In this case, the events are called “interstadials,” or short warming periods that occurred throughout the era. The interstadials saw temperature increases anywhere from 4 to 16˚C, the study’s authors explain, and sometimes that warming occurred over just a few decades. Once warming occurred, the Earth would stay warm for hundreds, sometimes thousands, of years.

The finding could reverse a long-held belief that human activity was the primary driver of megafaunal extinction, one of the study’s authors, Alan Cooper of the University of Adelaide, explained in an interview. “You could connect the arrival of humans to extinctions,” Cooper said, “When humans arrived 50,000 years ago in Australia, for instance, the megafauna disappeared soon after.”

The theory is based on the work of American paleontologist Paul Martin, who first proposed the idea in 1985 that advanced human hunting techniques wiped out local populations of megafauna as our species spread around the world.

One of the more famous examples used in support of this theory has to do with the extinction of mammoths in North America, which coincided with when researchers believe humans arrived on the continent. But that extinction happened as part of a whole cluster of extinction events around the world, and not all of them can be explained in the same way. “Several taxa (e.g., mammoth) go extinct on the mainland of Eurasia considerably later than that of the New World, despite a much longer exposure to human hunting,” the study says.

The idea that climate change caused the mammoths to vanish was, for a time, relegated to history, an old-fashioned idea that relied on less reliable methods of dating than those used by Martin and other whose evidence supported the human activity explanation.

Eventually, Cooper said, researchers began uncovering clues that “climate has got to be involved,” somehow. Some localized extinctions, for instance, were shown to occur before the arrival of humans.

Still, researchers assumed that climate had played an auxiliary role at best, and had trouble figuring out where climate could fit into the process. That’s, in part, because those extinctions happened over a long period, meaning that it was difficult to argue that a single climate change event was the source of the extinctions seen in the fossil record.

“Everybody was thinking about the one-off,” Cooper said, referring to a catastrophic climate event that could have caused the extinctions. “Our model is much more of a holistic thing. It’s been going on all the time.”

There was another problem with previous attempts to connect climate change to the extinctions: Those who did believe that the cycles of cooling and warming throughout the era might be causing them were looking at the wrong events. Many researchers originally suspected that the cooling periods, and not the warming periods, were causing species to go extinct. Cooper himself believed this was the case 10 years ago, when he noticed a pattern in the timing of when different species were going extinct.

Some research has previously linked the demise of the Neanderthals to a cold snap, suggesting that the species lost its source of prey due to the cold, eventually leading to their own die-off.

Of the megafauna extinctions, he said: “It seemed at first glance that sudden cold snaps might be responsible.” But soon after Cooper began to look at how a detailed climate record matched up to the localized extinction events, it became clear that the opposite is likely true.

Cooper’s model suggests that climate-driven extinction events happened in a pattern over time, closely tied to warming events that occurred through the era, going back until at least 50,000 years ago. Those patterns, Cooper said, weren’t always discernible from the fossil record. Instead, the team used a combination of DNA and radiocarbon dating to link localized megafauna extinctions to a series of rapid warming events over time.

The research team’s next step is to try and establish which aspect of the warming periods is the driving force behind the extinctions. “We can see the relationship between the warming periods and the extinctions,” Cooper said, “but can’t tell whether its the warming or the pace of change. It’s one of the two.”

Although there’s still more work to do, Cooper hopes his findings might help to illustrate the importance of considering our own climate today. “If we’re right, and these warming events are the key problem,” he said, “it’s quite obvious that the current global warming trends are very worrying, because they in many ways represent the conditions of the start of an interstadial.”