Science Wiki

The carbonate-silicate cycle (also called the long-term inorganic carbon cycle or the geological carbon cycle) is the set of processes that exchange carbon among the atmosphere, the ocean, and the rocks of the lithosphere on timescales of hundreds of thousands to hundreds of millions of years. It links plate tectonics, the water cycle, chemical weathering, and the carbon cycle. Carbon enters the atmosphere and ocean by volcanic and metamorphic degassing, and it leaves them by the chemical weathering of silicate rocks, which converts carbon dioxide to dissolved bicarbonate, followed by the burial of carbonate minerals in marine sediments. Because the rate of silicate weathering is thought to increase with temperature, the cycle is proposed to act as a negative feedback on climate, the "geological thermostat" first described by Walker, Hays, and Kasting in 1981. It is usually considered to be a main reason why the climate of Earth has remained suitable for liquid water over billions of years, in spite of the increase in the luminosity of the Sun (the faint young Sun paradox).

The cycle is distinct from the short-term carbon cycle, in which photosynthesis, respiration, and the exchange of CO₂ with the ocean move carbon among the atmosphere, the biosphere, and the surface ocean over days to centuries. The carbon in the geological cycle is stored mostly in carbonate rocks and in organic carbon in sedimentary rocks, and the reservoirs are very large (about 6 × 10⁷ petagrams of carbon in the carbonate rocks, compared with about 850 petagrams in the atmosphere). The geological cycle sets the long-term baseline of atmospheric CO₂, on which shorter-term variability is superimposed. The strength of the thermostat, and the role of other factors such as tectonic uplift, the lithology of the land, and the biosphere, are still debated.

Sources: degassing

Carbon is returned to the surface from the interior of the Earth by:

• Volcanism and metamorphism. Magmatism at volcanic arcs, mid-ocean ridges, and hotspots, and the thermal breakdown of carbonate rocks in orogenic belts (metamorphic decarbonation), release CO₂. The present total flux of CO₂ from the solid Earth is about 0.3 to 0.4 Gt of CO₂ per year or less, of the order of 0.1 Gt of carbon per year (the range of estimates is wide).
• Return of subducted carbon. Part of the carbon of the sediments and the oceanic crust that are carried down at subduction zones is released in the arc magmas, and part is carried deeper into the mantle. The proportion that is returned is uncertain.

The statement that the mantle "convection" provides "kinetic energy" for the cycle is a loose description. Plate motion is driven by mantle convection and slab pull, and it carries the carbon to the sites of subduction, while the energy for the weathering process comes from the Sun, through the hydrological cycle.

Weathering: the sink

Carbon dioxide dissolves in rainwater and, especially, in soil water, where respiration by roots and microbes raises its concentration far above that of the atmosphere. It forms carbonic acid (H₂CO₃), which attacks silicate minerals. For the idealized silicate wollastonite:

Real silicate rocks contain feldspars, pyroxenes, and other minerals, which give dissolved calcium and magnesium (and sodium and potassium), silica, bicarbonate, and clay minerals. Rivers carry the products to the ocean. In the ocean, calcium and bicarbonate are removed as calcium carbonate by organisms such as coccolithophores, foraminifera, and corals, and by inorganic precipitation:

The net reaction of weathering and carbonate precipitation is the Urey reaction:

so that one mole of CO₂ is consumed for each mole of calcium that is precipitated, because the precipitation releases one of the two moles that the weathering consumed. The weathering of carbonate rocks does not change atmospheric CO₂ over long times, because the carbonate that is later precipitated returns the carbon, so that the thermostat depends on silicate weathering (of calcium and magnesium silicates), and not on carbonate weathering. Sodium and potassium silicates do not lead to a net sink, because their cations are not removed in carbonates. The weathering of the rocks also supplies the cations for the buffering of ocean acidity on timescales of thousands of years, as part of the system that neutralizes the CO₂ that humans emit.

Burial, subduction, and return

The carbonate that is precipitated sinks to the seafloor or accumulates on shelves, and part of it is buried as limestone. In the modern ocean, a large share of carbonate production is by plankton in the open ocean, and the carbonate that sinks below the carbonate compensation depth (about 4 to 5 km, depending on the ocean basin) dissolves, so that deep-sea carbonate is preserved only above that depth. Before the evolution of calcifying plankton in the Mesozoic, carbonate was deposited mainly on shallow shelves and platforms, and by inorganic precipitation. The statements that the carbonate is compressed "into massive limestone" by the "immense pressure of the ocean" and that subduction "drags the carbonate plates deep into the mantle, melting them", are simplified. Burial and diagenesis turn the sediment into rock, and plate motion carries the sediment and the altered oceanic crust to subduction zones. There, a part of the carbonate is decarbonated and released in the volcanic arcs, part is scraped off or dissolved, and a part reaches the deeper mantle. The cycle is therefore not a perfectly closed loop with a fixed fraction of return.

Temperature and runoff dependence

The rate of silicate weathering increases with temperature, as the rate of chemical reactions increases according to the Arrhenius equation, and with runoff, because water supplies the reactants and removes the products. It also depends on the supply of fresh minerals, on the vegetation and soils, and on the lithology. A simple representation of the weathering flux F_w in terms of the atmospheric partial pressure of CO₂ (_p_CO₂) and temperature T is:

where β is the CO₂ dependence (a value of about 0.3 is often used), and k is the sensitivity to temperature. With the temperature dependence of k of about 0.05 to 0.1 per kelvin, a warming of 5 K increases the flux by exp(0.25 to 0.5), which is about 1.3 to 1.6 times. For a balance between the source F_v and the weathering sink F_w, a change in the volcanic input leads to a change in CO₂ and temperature, until a new balance is reached. For example, if F_v increases by 20 percent, and the flux depends on CO₂ only (β = 0.3, with no temperature dependence), then the equilibrium CO₂ rises by 1.2^(1/0.3) ≈ 1.8 times. If temperature dependence is included, the rise in CO₂ required is smaller, since the warming also increases weathering. This explains how the feedback stabilizes the climate, by limiting but not eliminating the response to changes of the sources.

For k = 0.07 per kelvin, a rise of 3 K increases the weathering rate by exp(0.21) ≈ 1.23 times. Field data from rivers give apparent sensitivities that are often lower, and that differ among catchments, because of the limits by the supply of minerals ("supply-limited" regimes in the mountains and "kinetically limited" regimes in the deeply weathered tropics). The strength of the feedback is therefore uncertain, and the evidence from proxies of geological records is also not definitive.

Operation of the thermostat

If volcanic degassing increases, CO₂ and temperature rise, the hydrological cycle intensifies, and weathering speeds up and removes more CO₂, until the climate returns to near its previous state. If the climate cools, weathering slows, and the continued supply of volcanic CO₂ raises the concentration. The statement that rainfall "violently accelerates" weathering and that weathering "halts" in the cold is an exaggeration. Weathering slows and does not stop, except under extensive ice cover. The response time of the cycle is of the order of hundreds of thousands of years, so that the thermostat is effective over timescales much longer than the short-term climate fluctuations.

Limits

The feedback does not ensure that the climate always avoids extremes. The Neoproterozoic Snowball Earth episodes, when ice may have reached the equator, show that glaciations can occur. In such cases, weathering is reduced under the ice, and volcanic CO₂ accumulates over millions of years until the greenhouse effect melts the ice (the origin of the "cap carbonates" that sit on top of the glacial deposits, which formed in the high-CO₂ and high-temperature aftermath, with intense weathering). Conversely, the thermostat cannot prevent a runaway greenhouse if the solar flux is increased sufficiently, and, in about 1 to 1.5 billion years, the brightening of the Sun is expected to reduce CO₂ by faster weathering to levels that are too low for C₃ photosynthesis, and later the oceans will evaporate. The thermostat also operates alongside other changes of CO₂ on geological timescales, for example the burial of organic carbon and its oxidation, and changes in tectonics, which may reflect long-term changes that are not caused by the feedback.

• Tectonics and uplift. Mountains expose fresh rock, so uplift can increase weathering. The "uplift-weathering hypothesis" proposes that the uplift of the Himalaya and the Tibetan Plateau in the Cenozoic increased the global weathering flux and lowered CO₂, contributing to cooling and the Pleistocene glaciations. It is debated, since the cations in the rivers of the Himalaya come in part from carbonate weathering, which does not give a net CO₂ sink, and since the degassing and burial also changed.
• Lithology. Basalts weather much faster than granites, and a disproportionate share of CO₂ consumption comes from basaltic terrains, for example in volcanic islands and large igneous provinces. The emplacement of large igneous provinces has been linked to changes in climate, with an initial warming from degassing and a later cooling from weathering.
• Biology. Plants and soil microbes increase weathering by raising soil CO₂, by producing organic acids, and by exposing fresh minerals. The rise of vascular land plants in the Devonian is thought to have increased weathering, but the size of the effect is uncertain.
• Sea level and continent configuration. The area of land at tropical latitudes, where weathering is faster, and the extent of the shelf seas, which affect the burial of carbonate, vary with the positions of the continents.

Natural fluxes

The present natural fluxes of the cycle are small compared with those of the short-term cycle. The volcanic degassing flux is of the order of 0.1 Gt of carbon per year (about 0.3 to 0.4 Gt of CO₂ per year), and the silicate weathering sink is of the same order (about 0.1 to 0.3 Gt of carbon per year in some estimates), in a long-term balance. Human emissions from fossil fuel and cement are about 10 Gt of carbon per year (about 37 Gt of CO₂ per year) in the early 2020s, which is about 30 to 100 times larger than the geological source and the weathering sink, so that the rate of emission is not matched by a response of the geological cycle on human timescales. The earlier statement that the cycle moves "approximately 0.1 billion metric tons of carbon per year" and humans inject 10 billion is correct in magnitude, with a ratio of about 100, which is two orders of magnitude. The conclusion that the cycle is "functionally useless" on human timescales is correct in the sense that it does not remove a significant amount of CO₂ within centuries, and the statement that the thermostat is "functioning exactly as dictated" is partly a simplification, since the cycle responds to the warming only slowly.

Response time

If the weathering flux responds in proportion to the excess CO₂, the excess decays exponentially:

with a time constant τ of about 200,000 to 400,000 years in many estimates. For τ = 3 × 10⁵ years, 63 percent of the excess is removed after 300,000 years. A large part of the emitted CO₂ therefore stays in the atmosphere for many millennia. On shorter timescales, other processes remove most of the emitted CO₂. The ocean absorbs part of it in decades to centuries, and the neutralization by the dissolution of carbonate sediments of the seafloor takes thousands of years. The silicate weathering is the final step.

Proxies of past CO₂ (such as boron isotopes in foraminifera, stomata of fossil leaves, and paleosol carbonates) and models of the long-term carbon cycle (such as GEOCARB and its successors) suggest that atmospheric CO₂ was much higher in the early Phanerozoic (several thousand ppm in the Cambrian, with large uncertainties) and that it declined, with variations, to the low values of the Pleistocene. Factors proposed for the decline include the rise of vascular plants, the uplift of mountains, and changes in the volcanic degassing. Large perturbations, such as the emplacement of the Siberian Traps at the end of the Permian, the Central Atlantic Magmatic Province at the end of the Triassic, and the Paleocene-Eocene Thermal Maximum about 56 million years ago, show a release of carbon, a warming, and then a recovery over hundreds of thousands of years, which is in line with the timescale of the weathering feedback.

The cycle is studied by the chemistry of rivers and groundwater (the fluxes of dissolved cations and alkalinity), by isotopic tracers (including ⁸⁷Sr/⁸⁶Sr, lithium, osmium, calcium, and magnesium isotopes), by carbon isotopes in carbonates and organic matter, by cosmogenic nuclides for the rates of erosion, by laboratory dissolution experiments, and by numerical models of the carbon cycle and of the climate. The enhanced weathering of rocks, by spreading crushed silicate rock, has been proposed as a method of CO₂ removal, but it is still at an early stage.

• Silicate weathering
• Carbon cycle
• Snowball Earth
• Faint young Sun paradox
• Plate tectonics
• Enhanced weathering
• Paleocene-Eocene Thermal Maximum