Thawing Permafrost Reveals a Hidden River Carbon Sink
A Nature study finds that chemical weathering of rock on the Tibetan Plateau absorbs about 35 percent of river CO2 emissions, a process that permafrost degradation appears to accelerate.
- Chemical weathering of rock absorbs about 35 percent of river CO2 emissions on average across the studied part of the Tibetan Plateau.
- The research covered 50 rivers on the Qinghai-Tibet Plateau, the largest high-altitude cryosphere outside the poles.
- In some catchments with discontinuous permafrost, geological carbon uptake exceeded 100 percent of river CO2 emissions.
- The study appeared in Nature, with the announcement released on June 17-19, 2026.
A team of researchers from Umea University in Sweden and East China Normal University, led by Jan Karlsson and Liwei Zhang, published a study in Nature in mid-June 2026 that reshapes how we think of rivers as sources of carbon dioxide. The research shows that rivers on the Tibetan Plateau do not only release CO2 into the atmosphere; geochemical processes within the same catchments also absorb part of that carbon. Measured across 50 rivers, chemical weathering of rock bound on average about 35 percent of river CO2 emissions.
In climate budgets, rivers are usually treated as conduits that move land-derived carbon toward the atmosphere and the ocean. Water flowing through soil and sediment picks up carbon produced by the breakdown of organic matter and later releases it at the surface as CO2. That flux is well documented. Less studied is what happens when the same water, rich in carbonic acid, reacts with minerals in rock. In a process known as chemical weathering, certain rocks consume CO2 as they dissolve, neutralizing part of the emissions before they reach the air.
Why is the Tibetan Plateau key to this measurement?
The Qinghai-Tibet Plateau is the largest high-altitude cryosphere outside the polar regions, a vast area covered by ice, snow, and permafrost. That makes it well suited for observing how warming acts on the carbon cycle. According to the study, when permafrost thaws and degrades it exposes fresh, reactive minerals that were not previously in contact with water. Those minerals accelerate the weathering that consumes CO2. In other words, a process we usually associate with carbon release from frozen ground can, at the same time, set in motion a mechanism that binds part of that carbon.
The results are not uniform. In some catchments with discontinuous permafrost, where frozen and unfrozen ground alternate, geological carbon uptake exceeded 100 percent of river CO2 emissions. That means those rivers, viewed through the chemistry of rock weathering, were on a net basis binding more carbon than they released. The roughly 35 percent average for the whole studied region shows how large local differences are and how strongly they depend on rock type and the state of the permafrost.
What this does not prove
It is worth separating what the measurements show from what one might wish them to show. The study documents that in one specific region part of river CO2 emissions was offset by rock weathering, and that the process is linked to the state of the permafrost. This is not proof that thawing permafrost reduces emissions globally, nor that rivers are becoming net carbon sinks at the planetary scale. As it thaws, permafrost also releases large amounts of methane and CO2 from accumulated organic matter, and that effect is not cancelled here.
This concerns one drainage basin, one plateau, and a process whose sign varies locally. The authors themselves stress that geochemical carbon uptake in rivers has so far been overlooked in models, so the results should be read as a call to include this factor in calculations, not as reassuring news. Replication in other cold regions, from Siberia to the Andes, will show how widespread the mechanism is. Do we understand rivers well enough to know when they are a source and when they are a sink of carbon?
