The Colorado River’s largest tributary flows ‘uphill’ for over 100 miles — and geologists may finally have an explanation for it
Geologists may have finally solved a long-standing mystery surrounding the Colorado River’s largest tributary, which appears to have defied gravity and flowed upwards when he first formed.
THE Green River originates in Wyoming and joins the Colorado River in Canyonlands National Park, Utah. About 8 million years ago, the Green River cut its way through the 13,000-foot-high (4,000-meter) Uinta Mountains in northeastern Utah and northwest Colorado, instead of flowing around the formation. But in a new study, researchers say this isn’t possible without a mechanism to lower the mountains.
The Green River crosses the Lodore Canyon, where it has eroded a ravine with walls 700 m high. Two competing theories have already attempted to explain why the river took this course, but neither is particularly convincing, Smith said.
One hypothesis is that the Yampa River, south of the Uinta Mountains, flowed north through the formation and created a channel for the Green River. This would have required enormous force, which the Yampa River probably would not have produced, because it is not particularly large. “If this were credible, then you would expect giant canyons running through all the mountain ranges, but that’s not the case,” Smith said.
The other theory is that sediments accumulated and temporarily raised the Green River, so that it overtook the Uintas and cut its way through them, but the available evidence does not support this hypothesis either. “The sediments you find here are not as high as those in Lodore Canyon,” Smith said.
Instead, the researchers behind the new study suggest that the Uinta Mountains have subsided to the point where the Green River could flow through them. Researchers suggest that a phenomenon called “lithospheric flow” dragged the mountains downward before a rebound effect caused the landscape to rise again, resulting in the topography we see today.
The results were published Monday February 2 in the Geophysical Research Journal: Earth Surface.
Lithospheric blobs are high-density regions that can form directly beneath mountains, where the Earth’s crust meets the top of the mantle – the layer of the planet between the crust and the outer core. The weight of mountains increases 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 on the Earth’s surface.
Lithospheric drops trigger a rebound effect when they eventually break away and sink into the mantle. The concept of these drops is relatively recent, but traces have been found in several places, including the Andes. “They can happen anywhere a mountain range forms, and they can happen at any time,” Smith said.
A telltale sign of lithospheric flow is a target-like uplift pattern on the Earth’s surface. Smith and his colleagues modeled geologic processes in the Uinta Mountains based on unusual river profiles and discovered such a pattern.
The researchers also analyzed seismic tomography images – 3D maps of the Earth’s interior created using seismic waves – from a previous study. They found a blob 200 kilometers deep in the mantle beneath the Uintas that looked a lot like an old lithospheric blob, providing strong evidence for this mechanism, Smith said.
Next, the researchers used the depth and size of the observed drop to calculate when it broke away from the base of the Uinta Mountains. They found that it likely broke free between 2 and 5 million years ago, matching the model’s predictions of when the mountains rebounded and matching estimates of when the Green River first flowed through the mountains.
Drip irrigation lowered the mountains so much that they became “the path of least resistance,” Smith said. Once the Green River began flowing over the Uintas, it continued to incise the mountains, creating structures like Lodore Canyon, he added.
Other experts who were not involved in the research suggested that this explanation could finally solve the long-standing mystery.
Mitchell McMillanresearch geologist at the Georgia Institute of Technology, said lithospheric flow is a plausible explanation for why the Green River flows the way it does.
“The most interesting aspect of this study is that it uses clues about the Earth’s surface to understand mantle processes and how they might affect mountain belts,” McMillan told Live Science in an email. “Whether the drip-drip hypothesis ends up being correct or not, this study is a valuable demonstration of such an approach.”
Smith, A., Fox, M., Miller, S., Morriss, M., & Anderson, L. (2026). A lithospheric blob triggered the integration of the Green River and the Colorado River. Geophysical Research Journal: Earth Surface. https://doi.org/10.1029/2025JF008733
