Geologists may have finally solved a longstanding mystery surrounding the Colorado River’s largest tributary, which appears to have defied gravity and flowed uphill when it first formed.
The Green River originates in Wyoming and links up with the Colorado River in Canyonlands National Park in Utah. Around 8 million years ago, the Green River carved its way through the 13,000-foot-tall (4,000 meters) Uinta Mountains in northeastern Utah and northwestern Colorado instead of flowing around the formation. But in a new study, researchers argue this isn’t possible without a mechanism to lower the mountains.
The Green River flows through the Canyon of Lodore, where it has eroded a ravine with 2,300-foot (700 m) walls. Two competing theories have previously tried to explain why the river ran this course, but neither is particularly convincing, Smith said.
One hypothesis is that the Yampa River to the south of the Uinta Mountains cut northward through the formation and created a channel for the Green River. This would have required a tremendous amount of force, which the Yampa River is unlikely to have produced, because it isn’t particularly big. “If this were credible, then you would expect giant canyons running through all mountain ranges, but that’s not the case,” Smith said.
The other theory is that sediments accumulated and temporarily elevated the Green River so that it overtopped the Uintas and carved its path through them, but the available evidence doesn’t support this either. “The sediments that you find here aren’t as high as the Canyon of Lodore,” Smith said.
Instead, the researchers behind the new study suggest the Uinta Mountains subsided to the point where the Green River could flow over them. The researchers propose that a phenomenon called a “lithospheric drip” tugged the mountains down before a rebound effect caused the landscape to rise upwards once more, resulting in the topography we see today.
The findings were published Monday (Feb. 2) in the Journal of Geophysical Research: Earth Surface.
Lithospheric drips are high-density regions that can form directly beneath mountains, where Earth’s crust meets the top of the mantle — the layer of the planet between the crust and the outer core. The weight of the mountains increases the pressure at the base of the crust, forming minerals like garnet that are heavier than mantle rocks. Eventually, these minerals form a blob that drips from the base of the crust, dragging the mountains down and reducing their elevation at Earth’s surface.
Lithospheric drips trigger a rebound effect when they finally detach and sink into the mantle. The concept of these drips is relatively recent, but evidence of them has been found in several places, including the Andes. “They can happen wherever you have had a mountain range form, and they can happen at any time,” Smith said.
A telltale sign of lithospheric dripping is a bullseye-like pattern of uplift on Earth’s surface. Smith and his colleagues modeled geological processes in the Uinta Mountains based on the unusual profiles of rivers there and found such a pattern.
The researchers also analyzed seismic tomography images — 3D maps of Earth’s interior that are created using seismic waves — from a previous study. They found a blob 120 miles (200 kilometers) deep in the mantle beneath the Uintas that looked very much like an old lithospheric drip, providing strong evidence for this mechanism, Smith said.
Next, the researchers used the observed drip’s depth and size to calculate when it detached from the bottom of the Uinta Mountains. They found that it likely broke free between 2 million and 5 million years ago, which fitted with the model’s predictions of when the mountains rebounded and matches estimates of when the Green River first cut through the mountains.
The drip lowered the mountains so much that they became “the path of least resistance,” Smith said. Once the Green River started flowing over the Uintas, it kept incising the mountains, creating structures like the Canyon of Lodore, he added.
Other experts who weren’t involved in the research suggested this explanation could ultimately solve the longstanding mystery.
Mitchell McMillan, a research geologist at the Georgia Institute of Technology, said that lithospheric dripping is a plausible explanation for why the Green River flows the way that it does.
“The most exciting aspect of this study is that it uses clues on Earth’s surface to understand mantle processes and how they might affect mountain belts,” McMillan told Live Science in an email. “Whether or not the drip hypothesis ultimately ends up being correct here, this study is a valuable demonstration of such an approach.”
Smith, A., Fox, M., Miller, S., Morriss, M. & Anderson, L. (2026). A lithospheric drip triggered Green and Colorado River integration. Journal of Geophysical Research: Earth Surface. https://doi.org/10.1029/2025JF008733
