What Really Happened to the Woolly Mammoth?

Woolly mammoths roamed Eurasia and North America for roughly 300,000 years before disappearing in stages, with the last known individuals dying on a remote Arctic island about 4,000 years ago. Their extinction wasn’t a single event but a drawn-out collapse driven by climate change, human hunting, habitat loss, and ultimately, genetic deterioration in their final small populations.

A Gradual Disappearance, Not a Sudden One

Mammoths survived multiple ice ages and warming periods over hundreds of thousands of years. The trouble began around 12,000 to 10,000 years ago, when most mainland populations vanished. But pockets held on much longer. On the Taymyr Peninsula in Arctic Siberia, mammoths persisted until at least 6,000 years ago in small areas that still had suitable habitat. On Wrangel Island, a small landmass off northeastern Siberia, a population became isolated around 10,000 years ago when rising seas cut the island off from the mainland. Those mammoths survived for over 200 generations before finally dying out roughly 4,000 years ago, making them contemporaries of the ancient Egyptians building pyramids.

A separate island population lived on St. Paul Island in Alaska. That group disappeared about 5,600 years ago, and their extinction story is unusually well understood: rising sea levels shrank the island’s area and dried up its freshwater sources. As lakes became smaller and shallower, the mammoths themselves likely made things worse by crowding around the remaining water, degrading its quality. Vegetation on the island stayed stable, and no humans arrived until 1787 CE, so the cause was purely environmental: the mammoths ran out of water.

The Landscape That Vanished

To understand why mammoths disappeared, you need to understand the world they were built for. During the Pleistocene, much of the Arctic and subarctic was covered by what ecologists call the “mammoth steppe,” a vast, dry grassland ecosystem unlike anything that exists today. Despite being colder than the modern Arctic (summers were 2 to 5°C cooler, winters 10 to 20°C colder), this landscape supported enormous herds of grazers: mammoths, horses, bison, and woolly rhinoceroses.

Between 12,000 and 10,000 years ago, this grassland underwent one of the most dramatic vegetation shifts in recent geological history. The dry grass steppe converted into the wet, mossy tundra we see today. The grasses that had sustained mammoths were replaced by mosses, which hold moisture in the soil and offer almost no nutrition for large grazers. This transition was partly driven by a warming climate, but research suggests it was also self-reinforcing. Large grazers like mammoths had been maintaining the grassland by trampling moss, compacting snow, and promoting grass growth through their feeding. When their numbers dropped, the loss of grazing pressure accelerated the shift to tundra, which in turn made the habitat even worse for the remaining animals. It was a feedback loop: fewer mammoths meant less grass, which meant even fewer mammoths.

Humans Played a Clear Role

The archaeological record of humans hunting elephants and their relatives stretches back an astonishing 1.8 million years to East Africa, where early human ancestors butchered ancient relatives of elephants. As human populations expanded northward over hundreds of thousands of years, a consistent pattern emerged: evidence of hunting proboscideans (the group that includes elephants, mammoths, and mastodons) appears repeatedly at the edges of the expanding human range.

This pattern fits what researchers call “overkill” rather than climate-driven extinction. If climate alone were responsible, you’d expect extinction events clustered by latitude and temperature zone. Instead, the evidence shows a more linear spread that tracks human migration. As behaviorally modern humans pushed into the high Arctic and eventually crossed into the Americas, mammoth exploitation followed. Archaeological sites across Venezuela, the western United States, and Arctic Siberia all show mammoth kills dating to roughly 11,000 to 12,500 years ago, coinciding with human arrival in those regions.

This doesn’t mean humans hunted every last mammoth. The more likely scenario is that human hunting reduced already-stressed populations below the point where they could recover, especially as their habitat was simultaneously shrinking.

Built for the Cold, Then Undone by It

Mammoths were exquisitely adapted to frigid environments. They carried specialized versions of hemoglobin, the protein in blood that delivers oxygen to tissues. Normal elephant hemoglobin holds onto oxygen more tightly in cold temperatures, which would have been a serious problem for an animal wading through Arctic snow. Mammoth hemoglobin had specific amino acid changes that allowed it to release oxygen efficiently even in extreme cold, reducing the need to generate extra body heat. This adaptation, identified by resurrecting mammoth hemoglobin genes in the lab, helped explain how an animal descended from tropical African ancestors could thrive near the Arctic Circle.

These adaptations made mammoths specialists, perfectly tuned to a world of cold grasslands. When that world changed, their specialization became a liability.

Genetic Collapse on Wrangel Island

The final chapter of the mammoth story played out on Wrangel Island, and it was not a graceful ending. Genomic analysis of one of the last surviving mammoths, dated to about 4,333 years ago, reveals a population in serious genetic trouble. Compared to mammoths from the mainland tens of thousands of years earlier, the Wrangel mammoths carried an excess of harmful mutations: broken genes, premature stop codons that cut proteins short, and deletions that disrupted important DNA sequences.

Large numbers of olfactory receptors, the genes responsible for the sense of smell, had lost function. Two independent mutations knocked out a gene called FOXQ1, which likely gave these late mammoths an unusual “satin” coat texture. Normal mammoth fur had air-filled cores in each hair shaft that provided insulation, similar to polar bear fur. Losing that structure would have compromised their ability to stay warm.

This accumulation of genetic damage is what biologists call a “mutational meltdown.” In small, isolated populations, harmful mutations that would normally be weeded out by natural selection can spread simply by chance. With no new individuals arriving from elsewhere, there’s no way to flush out the bad DNA. The Wrangel mammoths were trapped in a genetic dead end, their genome slowly degrading over thousands of years. Interestingly, researchers found no sign of a sudden accelerating decline at the very end. The population appears to have been slowly eroding rather than abruptly crashing.

What the Mammoth’s Loss Left Behind

The disappearance of mammoths didn’t just remove one species from the Arctic. It fundamentally changed the ecosystem. Without large herds of grazers trampling snow, breaking up moss, and recycling nutrients, the Arctic shifted toward the mossy, waterlogged tundra that now covers vast stretches of Siberia, Alaska, and northern Canada. Beneath that tundra lies permafrost packed with carbon, and as the Arctic warms today, that permafrost is thawing and releasing methane, a greenhouse gas far more potent than carbon dioxide.

Some researchers have argued that restoring large grazers to the Arctic could help slow permafrost thaw by compacting snow in winter (which lets cold air penetrate the ground) and maintaining grassland over tundra. This idea underpins projects like Pleistocene Park in Siberia, where bison, horses, and other herbivores have been introduced to test the concept. It also motivates Colossal Biosciences, a company working to use gene editing to create a cold-adapted elephant with mammoth traits. Their plan involves editing Asian elephant cells to incorporate mammoth genes for cold tolerance, then using nuclear transfer and surrogate elephants to produce calves after a 22-month gestation.

Whether such a project could meaningfully affect permafrost at a landscape scale remains deeply uncertain. While the Arctic is releasing methane as it warms, scenarios involving sudden, catastrophic releases of vast quantities of the gas appear unlikely based on current evidence. The mammoth steppe covered millions of square miles. Restoring even a fraction of that ecosystem with lab-created animals would take decades, if it’s possible at all.