In the patchiest terrain, geology was winning the carbon battle.
High on the Tibetan Plateau, where frozen ground has held ancient carbon for millennia, the thaw is revealing a story more intricate than science had assumed. New research shows that as permafrost melts and releases old carbon into rivers, the same retreat exposes bare rock to air and water, triggering chemical weathering that draws carbon back out of circulation. The planet's carbon ledger, it turns out, has entries on both sides of this geological page — and our models have only been reading one of them.
- Thawing permafrost on the Tibetan Plateau is releasing carbon dioxide more than a thousand years old into rivers, which vent it skyward — a warming feedback already alarming climate scientists.
- A new wrinkle disrupts the simple narrative: the same thaw exposes fresh rock, and the resulting chemical weathering quietly pulls CO2 back out of the water before it escapes to the atmosphere.
- In landscapes where permafrost has broken into patches, rock weathering captures more carbon than rivers emit — flipping the terrain from a carbon source into a net carbon sink.
- The offset is real but fragile: sulfur-rich minerals can reverse the gain, captured carbon may eventually return from the ocean, and the balance shifts depending on how much frozen ground remains.
- Climate models that treat thawing permafrost as a simple one-way emissions source are now shown to be incomplete, and scientists are calling for geological weathering to be built into future projections.
High on the Tibetan Plateau, where frozen ground has locked away carbon for thousands of years, the thaw is doing two things at once. The first is well known: melting permafrost releases ancient carbon into rivers, which carry it downstream as carbon dioxide and vent it to the sky. The second went unmeasured until now — the same retreat exposes fresh rock to water and air for the first time in millennia, triggering chemical reactions that pull carbon back out before it can escape.
Liwei Zhang of East China Normal University led a team of researchers from Sweden and China across the Qinghai-Tibet Plateau, sampling 50 rivers at different stages of thaw. Unable to watch a single river transform over centuries, they used spatial variation as a proxy for time, comparing landscapes at different points in the thawing process. Every river they tested held more than twice the carbon dioxide that air alone would contain, and the gas was ancient — released by microbes consuming thawed organic matter.
But the geological counterforce was striking. As permafrost retreated, exposed minerals underwent weathering that captured dissolved CO2 from the water. The team read this process through river chemistry — measuring calcium and other dissolved minerals as a record of geological activity. Across the region, weathering offset roughly a third of what rivers emitted. In heavily patched permafrost terrain, it captured more than 100 percent, making the landscape a net carbon sink.
The researchers were careful not to overstate the finding. Some minerals release CO2 as they break down, and carbon captured by weathering may eventually return from the deep ocean over thousands of years. Jan Karlsson of Umeå University stressed that biological and geological carbon cycles are inseparable — the math collapses when you treat them in isolation.
The broader implication is clear: current climate models that treat thawing permafrost as a straightforward emissions source are missing half the picture. The net carbon fate of a warming landscape now depends on its rock as much as its soil, and tomorrow's projections will need to account for both.
High on the Tibetan Plateau, where frozen ground has locked away carbon for thousands of years, something unexpected is happening as that ice melts. The thaw releases ancient carbon into rivers, which carry it downstream as carbon dioxide—a well-documented threat to the climate. But the same thaw is doing something else, something that until now went unmeasured: it is exposing fresh rock to water and air for the first time in millennia, triggering chemical reactions that pull carbon back out of the water before it can reach the atmosphere.
Liwei Zhang, a biogeochemist at East China Normal University, led a team of researchers from Sweden and China in studying this paradox across the Qinghai-Tibet Plateau, home to the world's largest expanse of permafrost outside the polar regions. Over multiple field seasons, they sampled 50 rivers at different stages of thaw. Because watching a single river transform over centuries would be impossible, they used a clever proxy: they compared rivers in landscapes at different points in the thawing process, treating the spatial variation as a stand-in for time.
The first part of the story confirmed what climate scientists already knew. Every river they tested held more than twice the carbon dioxide that air alone would contain, and the gas was ancient—over a thousand years old, released from the soil as microbes consumed the thawed organic matter. Rivers were venting this carbon to the sky, a mechanism already documented across the Arctic as permafrost regions warm.
But the second half was new. As permafrost retreated, it exposed bare minerals to water and air. When water seeps across these freshly exposed rocks, it triggers rock weathering—slow chemical reactions that grab carbon dioxide from the water and lock it away. The team detected this process by measuring dissolved minerals like calcium in the river water, reading the geological story from the chemistry rather than watching the reactions unfold in real time. Scientists had suspected that thaw accelerates weathering, but no one had measured whether it could actually offset the carbon that rivers released.
The numbers told a striking story. Across the region, rock weathering soaked up roughly a third of the carbon dioxide that rivers emitted. But the effect varied sharply depending on how much permafrost remained. In areas with continuous frozen ground, weathering offset only about 15 percent of emissions. In landscapes where the ice had broken into patches, weathering captured more than 100 percent of what the rivers released—the rock becoming a net carbon sink rather than a source. As permafrost thinned, the pattern held: in the patchiest terrain, geology was winning the carbon battle.
Yet the researchers were careful not to oversell this finding. Rock weathering is no climate cure. Some minerals, particularly sulfur-rich ones like pyrite, release carbon dioxide as they break down, driving the gas back into the air and potentially canceling out what other rocks store. Even the carbon that weathering captures may not stay locked away forever; far downstream in the ocean, some of it returns to the atmosphere over thousands of years. Jan Karlsson, a professor of ecology and geoscience at Umeå University who helped lead the international effort, emphasized that biological and geological carbon cycles are tightly linked. Pull them apart and the math breaks down.
The implications for climate science are significant. Most current models treat thawing permafrost as a simple one-way carbon source, a producer of greenhouse gases with little else to consider. This study shows that view is incomplete. In much of the Tibetan Plateau region, weathering rock pulls a measurable share of that carbon back. The net effect of a thawing landscape now hinges on the rock as much as the soil. Whether warming permafrost heats the planet or merely shifts carbon between reservoirs cannot be read from the soil alone. Tomorrow's climate models will have to weigh the geology too, redrawing the outlook for a warming world.
Citações Notáveis
River CO2 emissions decline while carbon uptake through rock weathering increases as permafrost cover decreases— Liwei Zhang, East China Normal University
Biological and geological carbon cycles are tightly linked— Jan Karlsson, Umeå University
A Conversa do Hearth Outra perspectiva sobre a história
So the thawing permafrost is still releasing carbon—that part hasn't changed?
No, that's still happening. Every river they tested was venting ancient carbon dioxide. The microbes are doing their work, breaking down the organic matter that's been frozen for thousands of years.
But something else is happening at the same time?
Yes. As the ice melts, it exposes rock that hasn't seen water or air in millennia. When water flows over that bare mineral, it triggers weathering—chemical reactions that actually pull carbon dioxide out of the water and trap it.
How much carbon are we talking about offsetting?
About a third of what the rivers emit, on average across the plateau. But it depends heavily on how much permafrost is left. In the most heavily thawed areas, the rock is actually capturing more carbon than the rivers release.
So in those places, thawing is a net positive for the climate?
Not quite. The researchers are careful about that. Some minerals release carbon as they break down. And even the carbon the rock captures might not stay locked away forever—it could return to the air in the ocean over thousands of years.
Then why does this matter for climate models?
Because current models ignore the rock entirely. They treat thawing permafrost as purely a carbon source. If you're missing a third of the equation, your predictions are wrong. The net effect now depends on both the soil and the geology working together.