We measured carbon emissions and water chemistry in 50 headwater rivers draining 780,000 km2 of the Tibetan Plateau.

The rivers flow in landscapes underlain by continuous permafrost and landscapes in which the permafrost has retreated since the last glacial maximum. They collect organic carbon from soils and dissolved inorganic carbon that is fixed by rock-weathering.

Where the permafrost cover is continuous, rivers emit CO2 from degrading (permafrost) soil carbon. Weathering reactions in these catchments are relatively slow.

In landscapes with (almost) no permafrost, carbon fluxes from weathering are faster than CO2 emissions from rivers.

Thus, as permafrost landscapes transition to landscapes without permafrost cover, chemical weathering reactions may play an ever more important role in riverine carbon cycling.

Interestingly, weathering can affect the carbon cycle in different ways. Where sulfide minerals are present, weathering reactions can emit CO2. Where silicate minerals dominate, weathering draws down CO2 from the atmosphere.

About: Zhang, L.^§, Bufe, A. ^§, Dean, J.F., Rocher-Ros, G., Sponseller, R.A., Stanley, E.H., Karlsson, J., Butman, D.E., Liu, R., Hou, L., Ding, J., Piao, S., Xia, X., Battin, T.J. (2026). Rock weathering can counteract river CO2 emissions induced by permafrost thaw. Nature. Journal Link
^§ L.Z. and A.B. contributed equally to this work.

The paper summarizes a four-year effort of 4 PhD students, 4 MSc students and many helping hands within the SNF/DFG-funded Hillscape project.

We measured 39 different physical, biological, and chemical soil-properties on glacial moraines that are 30 – 13,000 years.

Many of the measured characteristics show very similar changes through time. This observation led us to define a “Hillslope development Index” (HDI) that summarizes the overall state of physical and biological development of the hillslopes, and we find that the HDI saturates around 5000 – 10,000 years in both glacier forefields.

About: Greinwald, K., Bufe, A., Blume, T., Demand, D., Egli, M., Gebauer, T., Hartmann, A., Maier, F., van Meerveld, I., Musso, A., Scherer-Lorenzen, M., Weiler, M. (2026). Co-development of physical and biological variables on glacial hillslopes over millennia. Earth and Planetary Science Letters, 688, 120100. Journal Link

Sen Xu analyzed the changes in weathering fluxes from the headwaters of the Yangtze River over a period of seven years. During this time, climate warmed substantially, and Sen’s data show a substantial increases in solute fluxes from evaporite weathering (especially in the permafrost-rich headwaters of the JSR river). These changes could enhance secondary carbonate precipitation and lead to CO2 release. In addition, sulfide weathering becomes relatively more important than silicate and carbonate weathering reactions. This finding implies that warming and hydrological changes lead to a shift toward less CO2 drawdown and more CO2 release from rock weathering.

About: Xu, S., Bufe, A., Kemeny, P.C., Klaus, M., Zhong, J., Ma, T., Li, D., Li S.L., (2025). Climate warming and strengthened hydrologic cycle accelerate CO₂ release from rock weathering. Geochimica et Cosmochimica Acta, 407, 174-192. Journal Link

A new paper on the fluxes and sources of CO2 outgassing from different floodplain deposits is out now. Sophia measured CO2 that is emitted from river channels, riverbank sediments and older floodplain deposits along the Rio Bermejo in Argentina. She found the highest fluxes occur away from the modern channel and riverbanks. Here, emissions are highly seasonal (highest fluxes during the wet season) and are influenced by the respiration of recent and old organic matter.

Based on her measurements, Sophie estimated that sediments and soils in the Rio Bermejo floodplains emit almost 450 tons of carbon per square kilometer and per year.

About: Dosch, S., Hovius, N., Bufe, A., Repasch, M., Scheingross, J., Vieth-Hillebrand, A., Sachse, D. (2025). CO₂ Fluxes Driven by Floodplain Morphology and Seasonality at the Rio Bermejo, Argentina. Journal of Geophysical Research: Biogeosciences, 130(8), e2024JG008517. Journal Link

About: Erlanger et al., (2024), Nature Geosciences (link)

In our new paper, we use stream water chemistry in two river catchments of the central Apennines to infer the CO₂ fluxes from surficial weathering reactions as well as the CO₂ degassing from depths.

In the east, where the crust is thick and cold, carbon fluxes from silicate weathering dominate the carbon budget. In contrast, the western catchment is underlain by thin and hot crust. Here, carbon fluxes are dominated by CO₂ degassing from the crust and mantle, and these fluxes are up to 50-times higher than the carbon drawdown from silicate weathering.

You can check out a more detailed press-release by the GFZ here.

About: Xu et al., (2024), Environmental Science & Technology (link)

In a new paper spearheaded by Sen Xu from Tianjin University, we investiagate the role of warming for the export of dissolved inorganic carbon (DIC) in two major rivers that drain the eastern Qinghai−Tibetan Plateau.

In the Jinsha River that has 51% of its catchment underlain by continuous permafrost, DIC fluxes increase substantially over the past 40 years. Changes in river discharge play a negligible role for that increase in flux. Instead, the increase in DIC fluxes correlates most strongly with the temperature increase.

The Yalong river that is situated at lower elevation and has only 14% permafrost cover does not show a substantial increase in DIC fluxes. This observation suggest that the presence or absence of permafrost may strongly modulate the sensitivity of inorganic carbon fluxes to global warming.

About: Turowski et al., (2024), Earth Surface Dynamics (link)

What sets the width of river valleys? In a paper published today, we propose a new model for the width of river valleys. It considers valley width as a competition of lateral channel motion and the uplift and erosion of valley walls. Here is the equation:

W is valley width
q_L is the lateral sediment transport capacity
q_H is the lateral input of sediment from hillslopes
U is the uplift rate
W₀ is the channel belt width
W_C is the channel width

A dimensionless “mobility-uplift number, M_U” expresses that competition where:

The model implies that valley width varies between two extremes: At a minimum, valleys are as wide as the channel. Such narrow valleys occur where rivers drain rapidly uplifting landscapes. At a maximum, valleys encompass wide channel belts. These two extremes are connected by a logarithmic function of the mobility-uplift number.

Despite its conceptual simplicity, the model compares surprisingly well to several datasets including experiments and a large compilation of valley widths in the Himalaya.

This model explains valley width in a landscape that has reached steady state. How do valleys evolve over time and what sets the maximum width of valleys? We are working on these questions in an upcoming publication.

About: Bufe et al., (2024), Science (link)

In this new paper, we analysed weathering data from different mountain ranges. We found that silicates, carbonates and sulfides had different non-linear erosion sensitivities. The behaviors are very similar in all study areas. As a result, all datasets show that CO₂ drawdown from rock-weathering is at a maximum at moderate erosion rates of ~0.07 mm/yr.

You can read press-releases here with more info.

About: Xu et al., (2024), Geochimica et Cosmochimica Acta (link)

A multitude of different minerals are exposed at the surface of the Earth. Under the influence of acid waters, these minerals slowly dissolve and transform. These ‘chemical weathering’ reactions release nutrients, and they change move carbon between rocks, water, and the atmosphere. Rivers collect elements dissolved in soils. Therefore, we can use river chemistry to study the weathering reactions that occur within landscapes.

In large drainage basins that host many different rock-types, it can be a challenge to interpret the chemistry of rivers. In particular evaporite (“salt”) minerals can strongly dominate the weathering budget, and their contribution is difficult to distinguish from that of silicate, carbonate, or sulfide minerals.

In our work, we used a series of isotopes and major element chemistry to obtain a weathering budget in the headwaters of three of the largest rivers in the world – the Yangtze, Mekong, and Salween Rivers. We then analyzed how this weathering budget depends on erosion rates, rainfall and permafrost extent. We found that mountain building and attendant erosion play a major role in weathering of the studied rivers. Erosion boosts weathering reactions that may move CO₂ from the rock-record to the atmosphere.

About: Roda-Boluda et al., (2023), JGR-Earth Surface (link)

Where rocks are uplifted, they get eroded by wind, water, ice, and gravity. Erosion creates large volumes of sediment that are transported from the mountains to sedimentary basins. The rates of erosion are fundamentally driven by mountain uplift. However, the climate can also impact the breakdown and movement of rock. For example, heavy and sustained precipitation can trigger landslides, glaciers grind their bases to a fine powder, and cycles of freezing and thawing can efficiently break down solid bedrock. Geologists currently debate how climate affects erosion on the scale of an entire mountain range.

The Southern Alps of New Zealand are a fantastic place to dig deeper into the link between climate and erosion. Along a narrow range, metamorphosed sandstones are lifted up at multiple millimeters per year – making the Southern Alps one of the fastest deforming mountain ranges on the planet. A relief of over 3000 m captures the westerly winds and leads to heavy rains on the Western Southern Alps with yearly precipitation of 2 – 10 meters. Moreover, the Southern Alps are subject to so-called “paraglacial” (conditioned by recently retreated glaciers) and “periglacial” (in a zone where temperatures fluctuate around 0ºC) erosion processes: Where glaciers recently retreated, hillslopes have become unstable and temperatures around freezing cause efficient freeze-thaw cycles.

During one month in the field, we sampled sand from a number of rivers that drain the Western Southern Alps. Measuring the concentration of cosmogenic beryllium-10, we estimated the average erosion rate upstream of each sample point. Then, we studied how erosion rates vary with different topographic and climatic parameters.

We found that erosion rates were highest in those rivers that had a substantial portion of their catchment at an elevation of 1500 – 2000m. At these elevations para- and periglacial processes are particularly strong in the Southern Alps. In contrast, rainfall and erosion rates did not correlate well.

Overall, the pattern of erosion is set by the uplift of the rocks. However, our data suggest that these erosion rates can be modulated substantially by processes related to freeze-thaw and glacier retreat.