Science Wiki

Silicate weathering is the set of chemical reactions in which carbonic acid and other acids dissolve silicate minerals in rocks and release cations, silica, and bicarbonate to water. Because the acid originates from atmospheric carbon dioxide, the process consumes CO₂. When the dissolved products are carried to the ocean and precipitated as carbonate minerals, carbon is transferred from the atmosphere and ocean to the rock reservoir. Silicate weathering is the main long-term sink of CO₂ in the carbon cycle, and, together with volcanic and metamorphic degassing, it is thought to regulate atmospheric CO₂ and climate on timescales of hundreds of thousands to millions of years. This is the basis of the proposed weathering feedback, sometimes called the carbonate-silicate cycle or the geological thermostat, first described by Walker, Hays, and Kasting in 1981.

The feedback is negative. Warmer and wetter conditions are expected to speed up weathering and CO₂ consumption, which cools the climate, and cooler and drier conditions slow it, which allows CO₂ from volcanoes to accumulate. The strength of this feedback, and the role of other factors such as tectonic uplift, vegetation, and lithology, are still debated. The thermostat is a slow process, and it does not respond on human timescales.

Dissolution by carbonic acid

Carbon dioxide dissolves in water, mostly in the soil, where biological respiration raises the CO₂ concentration far above atmospheric values. It forms carbonic acid (H₂CO₃), which dissociates to bicarbonate and hydrogen ions:

The acid attacks silicate minerals. For the idealized calcium silicate wollastonite, the net reaction is:

Wollastonite is a simplified model. Real rocks contain feldspars, pyroxenes, olivine, micas, and other minerals, which weather to a mix of dissolved cations (Ca²⁺, Mg²⁺, Na⁺, K⁺), dissolved silica, and secondary clay minerals, so the stoichiometry is more complex. For example, the weathering of the plagioclase feldspar anorthite to the clay kaolinite is:

In both cases, two moles of CO₂ are consumed for each mole of Ca²⁺, which is released with two moles of bicarbonate. Magnesium silicates give the same ratio for Mg²⁺. Sodium and potassium silicates consume one mole of CO₂ per mole of cation, but sodium and potassium are not removed in carbonates, so their weathering does not result in a net long-term sink.

Carbonate precipitation

The dissolved products flow through groundwater and rivers to the ocean. There, calcium and bicarbonate are removed from seawater mainly as calcium carbonate:

Combining the two steps gives the net reaction of the carbonate-silicate cycle (the Urey reaction):

The net removal of CO₂ per mole of Ca²⁺ is one, and not two, because the precipitation of carbonate releases one of the two moles of CO₂ that were consumed. Weathering of carbonate rocks (limestone) has a net effect of zero on timescales longer than the ocean's carbonate residence time, because the carbonate precipitated in the ocean returns the carbon that was consumed. This distinction explains why silicate weathering, and not carbonate weathering, controls atmospheric CO₂ over geological time. The precipitation of carbonate is carried out largely by marine organisms such as coccolithophores, foraminifera, and corals, and also by inorganic precipitation. The carbon is buried in sediments on the seafloor, including the deep ocean and the continental shelves, and it can be preserved as limestone.

The sediments are carried by plate motion to subduction zones, where part of the carbon is returned to the atmosphere by metamorphism and volcanic arc degassing, and part is carried into the mantle. Carbon is also released at mid-ocean ridges and by volcanism at hotspots and, over the long term, by the metamorphic breakdown of carbonate in orogenic belts. The present-day degassing flux of carbon from the solid Earth is of the order of 0.1 to 0.3 gigatons of carbon per year, much less than human emissions. Over time, the weathering sink and the degassing source have been roughly in balance, though not exactly, and imbalances over millions of years have contributed to changes in the climate. Oxidation of old organic carbon in rocks (petrogenic carbon) is another source, and the burial of organic carbon from the biosphere is another sink.

The rate of silicate weathering is determined by the supply of fresh minerals, the amount of water, temperature, and biological activity. It can be written in a simple form as a rate that is limited by the reaction kinetics or by the supply of minerals.

Temperature

The rate of a chemical reaction increases with temperature according to the Arrhenius equation:

where E_a is the activation energy, R is the gas constant, and T is the temperature. For silicate dissolution, E_a is of the order of 40 to 80 kJ per mole in laboratory studies. For E_a = 50 kJ per mole, a rise in temperature from 288 K to 293 K increases the rate by exp[(50,000 / 8.314) × (1/288 − 1/293)] ≈ exp(0.35) ≈ 1.4. Field studies of river chemistry give apparent activation energies that are often lower and that differ among catchments, because temperature is only one of the controls. The warming dependence is the basis of the weathering feedback.

Runoff and hydrology

Water supplies the reactants and carries the products away, which keeps the solution far from equilibrium. Higher runoff generally increases the export of weathering products, but the relation is weaker than linear in many regions, because the concentrations of dissolved solids decrease as runoff increases. A warmer climate has a faster hydrological cycle, which increases global precipitation by about 2 to 3 percent per kelvin, although regional changes differ in sign.

Supply of minerals: tectonics and erosion

Weathering of a rock requires fresh mineral surfaces. In regions where the supply of fresh rock is high, such as steep mountain belts, rapid erosion exposes unweathered minerals, and weathering is limited by the rate at which the minerals dissolve ("kinetically limited" or "supply limited"). In flat, deeply weathered regions, the soil has already lost the soluble elements, and the rate is limited by the supply of fresh minerals ("weathering limited" or "supply limited"). The uplift of the Himalaya and the Tibetan Plateau in the Cenozoic is thought to have increased the global weathering flux, and it has been proposed as a cause of the decline in atmospheric CO₂ and the cooling that led to the Pleistocene ice ages (the "uplift-weathering hypothesis" of Raymo and Ruddiman). This hypothesis is debated, because the sources of the cations in Himalayan rivers include carbonate and silicate rocks, and because the change in the burial of carbon and in the degassing also affect the balance.

Lithology and basalt

Different rocks weather at different rates. Basalts, which are rich in calcium and magnesium minerals that dissolve quickly, weather much faster than granites and gneisses, and a disproportionate part of the CO₂ consumption comes from basaltic provinces, such as the Deccan Traps, Iceland, Hawaii, and the Siberian Traps, and from tropical islands in volcanic arcs. This is the reason why the emplacement of large igneous provinces has been linked to climate changes, as an initial warming from degassing, followed by cooling from the weathering of the fresh basalt. Weathering of the Central Atlantic Magmatic Province and the Deccan Traps has been proposed to have played a role in the recovery from the end-Triassic and end-Cretaceous perturbations, but the evidence is incomplete.

Biology

Plants and soil microbes increase weathering by raising the CO₂ concentration in the soil, producing organic acids, stabilizing soils, and increasing the contact between water and minerals through roots and mycorrhizal fungi. The rise of deeply rooted land plants in the Devonian is thought to have increased the weathering rate and to have contributed to the decline in CO₂ in the Late Paleozoic, but the size of this effect is uncertain.

Stabilization of climate

A temperature-dependent weathering rate stabilizes the climate against changes in the CO₂ flux from volcanoes. If the degassing flux increases, CO₂ and temperature rise, the weathering rate rises, and CO₂ is removed faster until a new balance is reached at a higher CO₂ level. A simple model is a balance between the degassing flux F_v and the weathering flux F_w, which depends on atmospheric CO₂ (_p_CO₂) and temperature:

where β is a parameter of CO₂ dependence, which is often taken to be 0.3 to 0.5 or more, k is the temperature sensitivity, and T₀ and _p_CO_2,0 are reference values. At balance, F_w = F_v. Models of this type were used to argue that the Earth's climate has remained within the range that allows liquid water for billions of years in spite of the increase of the luminosity of the Sun by about 30 percent since the formation of the Earth, the faint young Sun paradox. The mechanism is thought to be an important part of the answer, together with other factors such as the composition of the atmosphere.

Snowball Earth

The weathering feedback is also thought to help to end global glaciations. In the Snowball Earth hypothesis, ice covered most of the planet in the Neoproterozoic (for example the Sturtian and Marinoan glaciations). Weathering would be very slow under ice cover, and volcanic CO₂ would accumulate over millions of years until the greenhouse warming was enough to melt the ice. After the melt, the high CO₂ and temperature would cause fast weathering and the precipitation of "cap carbonates" that are found on top of glacial deposits. The hypothesis has been controversial, and some scientists favor a "slushball" state with open water near the equator.

Limits of the thermostat

The thermostat does not make the climate immune to runaway states. It operates over about 10⁵ to 10⁶ years, and a sufficiently large change in solar luminosity, or in the CO₂ supply, can overwhelm it. About 1 to 1.5 billion years from now, as the Sun brightens, the rate of weathering is expected to reduce atmospheric CO₂ to levels that are too low for C₃ photosynthesis, which would end most plant life long before the oceans are lost. The claim that the feedback ensures the planet never reaches a runaway greenhouse or a snowball state is therefore not correct. The climate of the Earth has changed greatly in its history, and the Snowball Earth episodes show that glacial states can occur. The rate of weathering is not tuned only by the temperature but also by changes in tectonics, lithology, and the biosphere, and it may not respond in a simple way.

Timescale of response

The response time of the weathering feedback is of the order of hundreds of thousands of years. The time constant for the removal of a pulse of CO₂ by silicate weathering is often estimated at about 200,000 to 400,000 years, with a range of values depending on the model. If the weathering flux is proportional to the excess CO₂ over the equilibrium, the excess decays as:

with τ ≈ 3 × 10⁵ years. For this value, about 63 percent of the excess has been removed after 300,000 years, and about 14 percent remains after 600,000 years.

On shorter timescales, other processes remove most of the CO₂ that is emitted by humans. The ocean absorbs part of the CO₂ over decades to centuries by dissolution and the carbonate chemistry of seawater, and the neutralization of the acid by the dissolution of seafloor carbonate sediments takes thousands of years. Silicate weathering is the final step, which removes the remaining part over hundreds of thousands of years. A substantial fraction of the CO₂ that is emitted by humans therefore remains in the atmosphere for many millennia.

Comparison with the present flux

The present global CO₂ consumption by silicate weathering is estimated to be of the order of 0.1 to 0.3 gigatons of carbon per year. Human emissions of CO₂ from fossil fuels and cement are about 10 gigatons of carbon per year in the 2020s, which is about 30 to 100 times larger. The earlier statement that emissions exceed the weathering rate by a factor of tens of thousands is not correct as a comparison of the two fluxes. The rate of the human perturbation is, however, much faster than the rate at which the weathering feedback responds. Geological sources, such as volcanoes, release roughly 0.1 to 0.3 gigatons of carbon per year, so that human emissions are 30 to 100 times larger than the natural geological source as well. The weathering flux is expected to increase with the warming, but the increase over the coming centuries is small compared with emissions.

The proposal to accelerate weathering to remove CO₂ is called enhanced weathering or enhanced rock weathering. It involves grinding silicate rocks, usually basalt, or olivine, to increase the surface area and spreading them on land, in particular on agricultural soils, or on beaches and in the ocean. The potential removal of CO₂ is estimated to be of the order of gigatons of CO₂ per year at a large scale, but the estimates have a wide range. Open questions include the rate of dissolution in the field, which is slower than in the laboratory, the energy and emissions that are needed for mining, grinding, and transport, the measurement and verification of the carbon that is removed, the release of trace metals such as nickel and chromium from some rocks, and the effects on soils and ecosystems. Field trials have begun, and the approach is not yet proven at scale. In the ocean, the related idea of ocean alkalinity enhancement raises similar questions.

Silicate weathering is studied with measurements of the chemistry of rivers and of groundwater (concentrations of cations, silica, and alkalinity, with corrections for atmospheric, carbonate, and evaporite inputs), isotopic tracers (such as ⁸⁷Sr/⁸⁶Sr, lithium, and osmium isotopes in marine sediments, and calcium and magnesium isotopes), cosmogenic nuclides to measure erosion rates, laboratory dissolution experiments, and models of the carbon cycle over geological timescales, such as GEOCARB and its successors. Records of atmospheric CO₂ in the geological past come from proxies, such as the stable isotopes of boron in foraminifera, the density of stomata on fossil leaves, and paleosol carbonate, and they have large uncertainties, which is one reason why the strength of the feedback is not well constrained.

• Carbon cycle
• Weathering
• Carbonate-silicate cycle
• Snowball Earth
• Enhanced weathering
• Faint young Sun paradox
• Plate tectonics