≡ Menu

An Exoplanet Model for the Carbon Cycle

Earth’s long-term carbon cycle is significant for life because it keeps carbon in transition, rather than allowing it to accumulate in its entirety in the atmosphere, or become completely absorbed in carbonate rocks. The feedback mechanism works over geological timescales to allow stable temperatures as CO2 cycles between Earth’s mantle and the surface. As a result, we have carbon everywhere. 65,500 billion metric tons stored in rock complements the carbon found in the atmosphere and the oceans, as well as in surface features including vegetation and soil. It’s a long-term cycle that can vary in the short term but be stabilizing over geological time-frames.

The Sun has increased in luminosity substantially since Earth’s formation, but the long-term carbon cycle is thought to be the key to maintaining temperatures on the surface suitable for life. Does it exist on other planets? It’s an open question, as astronomer Mark Oosterloo (University of Groningen, The Netherlands) points out:

“We don’t know if there are any other planets at all with plate tectonics and a carbon cycle. In our solar system, the Earth is the only planet where we have found a carbon cycle. We hope that our model can contribute to the discovery of an exoplanet with a carbon cycle, and therefore, possibly life.”

Image: This diagram of the fast carbon cycle shows the movement of carbon between land, atmosphere, and oceans. Yellow numbers are natural fluxes, and red are human contributions in gigatons of carbon per year. White numbers indicate stored carbon. Credit: Diagram adapted from U.S. DOE, Biological and Environmental Research Information System.

Oosterloo is lead author of a paper that has just appeared in Astronomy & Astrophysics. Working with researchers at SRON (Netherlands Institute for Space Research) and Vrije Universiteit Amsterdam, the scientist has developed a model designed to analyze whether or not a carbon cycle can emerge on an exoplanet, varying its mass, core size and the amount of radioactive isotopes it contains.

Quite a lot goes into determining how the feedback mechanisms of a carbon cycle work, enough so that our problems in observing exoplanets swim into sharp relief — we’re usually limited to mass and radius measurements along with a degree of atmosphere characterization on those worlds where we can deploy methods like transmission spectroscopy, viewing the light of a star as it is filtered by a planet’s atmosphere.

But we’re learning about the composition of the planets slowly but surely. The authors cite Kepler-452b, a world whose interior has been shown to have a larger fraction of rock than Earth, which would affect the chemical structure of the interior. That in turn affects the amount and type of volatiles outgassed into the atmosphere.

To go beyond this, the authors investigate how planetary interiors of different composition affect long-term carbon cycling as enabled by plate tectonics. This involves not just the relative abundances of radioactive isotopes, the size of the planet’s core and its mass, but also into the evolution of CO2 in the atmosphere. The team’s two-component model, connecting mantle convection to the emergence of a long-term carbon cycle, has plate tectonics at its center — in fact, the paper refers to mean plate speed as “the key coupling variable between the two models.”

Among the findings thus far is that the cooling of a planet’s mantle (and its effect on the tectonic plate speed) produces gradually declining CO2 levels over time as outgassing slowly decreases. And note the effects of varying a planet’s internal heating:

A long-term carbon cycle driven by plate tectonics could operate efficiently on planets with amounts of radiogenic heating in their mantles different from Earth. However, planets with their mantles enriched in radioactive isotopes with respect to Earth, may favor the development of warmer climates resulting from a more CO2 rich atmosphere. This is in particular the case for a planet with a higher thorium abundance. In addition, the carbon cycle operates more efficiently on planets rich in radioactive isotopes, motivating the characterization of planetary systems around stars whose atmospheres are rich in thorium or uranium.

A long-term carbon cycle regulated by plate tectonics, say the authors, may not operate on planets with a core mass fraction greater than 0.8 (for reference, Earth’s core mass fraction is 0.32 of the planetary mass; that of Mars is 0.24). Here the value of mass/radius measurements on Earth-sized planets becomes apparent. And it was surprising to me to learn that the effects of carbon cycling can be felt relatively quickly. In the authors’ models, equilibrium is achieved between 100 and 200 million years. In other words, if plate tectonics is operational, an exoplanet does not have to be old to have its atmosphere affected by a long-term carbon cycle.

Image: An artist’s impression of an Earth-like exoplanet. Can we develop the tools to establish the presence of a carbon cycle on such worlds? Credit: NASA.

The significance of this paper in my estimation is that it shows that carbon cycling can work as a regulator of climate for planets with a wide range of masses, core sizes and radioactive isotope abundances. As noted above, there is more efficient carbon cycling on planets with a high level of thorium or uranium in their mantles. That would imply that mapping these elements as found in their host stars would be a useful observational tool in exploring potential habitability of planets in a given stellar system.

The paper is Oosterloo, et al., “The role of planetary interior in the long-term evolution of atmospheric CO2 on Earth-like exoplanets,” Astronomy & Astrophysics Vol. 649, A15 (3 May, 2021). Abstract / Preprint.

{ 22 comments… add one }
  • Josh May 4, 2021, 9:09

    Always a fascinating read, Paul. Nice write-up!

    Though the authors’ paper was focused primarily on abiotic characterizations, I hope they or other teams can use this work as a springboard to look at the question of the role of life in these processes. For instance, could plate tectonics *require* the presence of biomass and activity to maintain over Gy eras?

    The ‘70’s era Gaia hypothesis for long-term planetary habitability could benefit from a modernization pass using this type of approach, since then we’ve discovered life literally everywhere on, above, below, and inside the Earth we look. That doesn’t seem like a subtle detail 🙂

  • Michael Fidler May 4, 2021, 10:25

    They may want to look into this since it also heats the mantle even after radioactive elements are depleted.

    Quartz” crystals at the Earth’s core power its magnetic field.


  • Michael Fidler May 4, 2021, 10:32

    Just maybe we need a good hard giant impact from a huge comet to put enough oxygen into the outer core. Such impacts would also turbocharge plate tectonics.

  • Alex Tolley May 4, 2021, 11:34

    If the model for the carbon budget and recycling is correct, it implies that the biological carbon management (Gaia hypothesis) is a detail compared to the long-term geologic effects.

    I note that figure 5B shows the CO2 in the atmosphere as having a long-term decline as plate tectonics subducts the CO2 (as carbonate?) into the mantle. This continues for the length of the modeling time – 10 gy. The atmospheric CO2 declines with surface temperature within the first 10 my, and by 100 my, the surface temperature is relatively low around 300 K, but the CO2 continues to decline. Figure 7 shows the continued exponential decline of atmospheric CO2 under differing assumptions.

    Life certainly has had an impact on atmospheric CO2, with a rapid decline since the Pre-Cambrian, a sharp decline in teh Carboniferous when plants were rapidly fixing carbon and being buried, resulting in the huge coal seams we exploit today. But life, the randomness of sustained volcanic eruptions, and asteroid impacts seem to be details added to the general dynamics of CO2 recycling.

    The model does seem to imply that without plate tectonics, CO2 cannot be effectively recycled into the mantle, but at this point, it has nothing to say about the requirement for plate tectonics on abiogenesis.

    If I understand the argument on the amounts of radiogenic materials, the model indicates that they do not have much impact on carbon recycling over the long term, an outcome at odds with a recent CD post on planetary internal heating (IIRC), although the magnitude of the differences was not explained.

    Not discussed are the long-term implications for life. Kasting and others have shown that surface temperatures on Earth will remain stable until CO2 is almost depleted from teh atmosphere, after which the increasing solar output will heat the Earth’s surface. Without sufficient atmospheric CO2, photosynthesis fails to work, primary productivity collapses, and with it complex life. Only the bacteria can continue. This will happen on Earth well before 10 gy. It may also have a bearing on life on M_dwarfs. It is often stated that the longer life of M_dwarf stars offers more time for life to evolve once the active flaring has ended, and the slow increase in stellar output brings a planet into the HZ. However, if an earth mass planet follows the model in this paper, the planetary conditions may be more important than the stellar ones. Life may have a limited time to evolve before the atmosphere can no longer support photosynthesis. Such M_dwarf worlds may harbor unicellular life, but the surface will not be covered in photosynthesizing plants and the atmosphere will consequently have almost no free O2.

    It does seem to me that the carbon recycling model presented could be coupled with 3D climate models on planetary habitability to refine the dynamics of long-term climate and the impact on life. There do seem to be some interesting constraints to be applied.

  • Geoffrey Hillend May 4, 2021, 17:10

    The carbon cycle: There are rifts which spread apart where new crust is created and after a long time and distance the crust is subducted beneath the crust into the mantle where it is melted down. This resets the radioactive isotopes in the rock to the time it was subducted. Volcanic eruptions put the carbon dioxide back into the air since the magma comes from the subducted molten crust in the mantle. The rain and oceans take the CO2 out of the air which becomes limestone on the crust or ocean floor and the crust and CO2 are recycled. Alex Tolley mentioned the carboniferous period which is when life began to draw down the carbon dioxide and the continental drift and the supercontinent Pangaea Proxima kept the Earth climate warm. It’s splitting apart helped cool the climate with coast line, more mountain ranges and exposed basalt. Photosynthesis also reduced the carbon dioxide levels to low enough, so we have polar ice caps around 20 million years ago and the ice ages began around 3 million years ago.

    It looks like the carbon cycle is very important for an exoplanet to evolve life. If there were no plate tectonics, then there will not be any carbon cycle and all of the carbon dioxide might be turned into limestone at the bottom of the ocean through a one way, incomplete carbon cycle, the oceans and rain taking the CO2 out of the atmosphere.

    I agree with this paper’s idea that more radio isotopes in the core are important, but I will add that is not the only heat source. There is also the gravitational energy in the core. We can use Venus as an example. Earth has a higher density and more heavier elements than Venus. I assume this is due to the large iron core Earth received from it’s collision with Theia, the giant impact hypothesis. A larger iron core has more heat than a smaller one. Plate tectonics work through convection currents in the Earth’s mantle. The chemical composition of our crust is different than Venus which has less heavier elements than Earth. A speculation of mine is we have to have the right composition of the crust in order for there to be plate tectonics and continental drift. If an exoplanet the exact size of Earth without a Moon still does have plate tectonics, then it is essential for life to survive over a long period of time. We can absolutely assume based on the principles of science and geology, that an exoplanet with a Moon which is nearly exactly the same size as our Earth Moon system must have plate tectonics if in the habitable zone around a G class star. Since an exoplanet around a M dwarf is tidally locked, it can’t have a Moon due to the hill sphere. It still could have a collision with another body an get a large iron core.

    • wdk May 5, 2021, 10:57

      This topic, paper and exposition caught me and my group studying exoplanets at the right moment. We had been focused on detection methods and stellar environments. Now how to examine the swarm of exoplanets that seem to resemble the Earth? Examination of components for sure: crusts, atmospheres, cores, mantles, magnetospheres… But this carbon cycle entry and further deliberation has been just primo!

      I’m still getting my sea legs on varying the interiors. But I note the examination of the role of the Earth’s moon and the absence of such in closely packed systems around red dwarf stars. I suggest that in cases such as Trappist 1 and Kepler 62, the multitude of planets might be doing just as well. Io’s volcanism is not necessarily a direct result of Jupiter, but the perturbation of the other Galilean satellites. Similarly, the planets in these system are not as distant as Earth is to Venus and vice versa, but nearer to the Earth-Moon scale. Plus they are considerably more massive.

      Io is tidally locked, of course. and it succeeds in refreshing its surface features. It doesn’t have an ocean though and its processes reflect the flow of sodium and sulfur; carbon probably absent. Whether these markers point to genuine plate tectonics and convection, I would leave that open to other comment. Maybe a more violent species if we compare to what has happened at Europa and Ganymede ( features described as the result of convective processes) and remarkable drop off at Callisto. Still, I would guess that red dwarf multi-planet systems similar to the ones above could host carbon cycles.

      Given that, what if we were to move up from the M dwarfs to the K stars?

  • Geoffrey Hillend May 4, 2021, 17:14

    My idea was that Earth has more convection in its mantle than Venus mantle due to Earth’s larger iron core and size so maybe Venus never had plate tectonics. It most likely did have plate tectonics and it would be nice to have a lander on Venus with a seismograph to study it’s crust and core.

  • Geoffrey Hillend May 4, 2021, 18:14

    We can add another contingency and end up with the carbon cycle and conclude that plate tectonics don’t work or are not efficient without the tidal forces of our Moon? This would certainly support a semi rare Earth twin hypothesis that without a Moon, life can’t exist on the long term since without a carbon cycle, an exoplanet atmosphere would run out of carbon dioxide? This paper suggests that the Earth’s plate movement is dependent on a viscosity of the mantle caused by the our Moon’s tidal forces which put a gravitational torque on it. https://www.sciencedirect.com/science/article/pii/S1674987117302037

  • Robin Datta May 5, 2021, 9:01

    Over geologic time the formation of carbonates through the weathering of certain rocks will sequester the carbon from land, sea and air: when the CO₂ in the atmosphere is depleted plants, and ecosystems dependent on them, will perish. This will be well before the sun’s red giant phase. The carbon cycle may seem stable in shorter time frames, but where the right rocks are periodically exposed to surface carbon, it will all be sequestered.

  • Dave Moore May 5, 2021, 13:09

    I think these guys were a little too focused on plate tectonics. The CO2 cycle doesn’t need plate tectonics. Take Mars, a planet without plate tectonics. It still has a higher partial pressure of CO2 in its atmosphere than Earth does.

    Weathering reduces the CO2 level in a Earthlike planet’s atmosphere until the temperature approaches the freezing point of water. At which point, weathering stops and the CO2 draw down stops. If the planet has any sort of internal heat, there will be volcanism, which will build up the CO2 levels in the atmosphere to the point where the average temperature is such that weathering draws down the CO2 at a greater rate than its released into the atmosphere.

    The temperature set point of the planet is determined by when weathering removes CO2 at the rate it is produced.

    A planet with vigorous volcanism will have a higher average temperature to a point where the weathering is vigorous enough to balance it.

    A planet with low volcanism will have a lower average temperature.

    A planet with very little continent, like early Earth will have a higher CO2 level, and higher temperature as the efficiency of weathering is low.

    A planet with little ocean will have a lower set temperature as the weathering efficiency is higher.

    Lower insolation will result in higher CO2 levels to get the rate of weathering up.

    Higher insolation will result in lower CO2 levels as the higher temperatures and weathering will draw down the CO2 to a point where temperatures drop and the weathering is in balance.

    Life drawing down CO2 levels will lower the set temperature so there will be less weathering to compensate for the CO2 draw down.

    As you know from the history of Earth and Mars, this mechanism has a slow response time so tends to go through swings away from the set point.

    • Patient Observer May 6, 2021, 22:44

      Volcanism does emit CO2 but for the concentration to be high, plate tectonics is needed to bring fresh carbonate rock to the “root” I guess of the volcano. And plate tectonics may be needed for widespread volcanism to occur to provide a steady source of buoyant rock.

      For whatever reason, Venus does not have plate tectonics and its level of volcanism seems small to non-existent at the moment from what I have read. Supposedly, internal heat can build up to a point where there is a massive volcanic outbreak flooding the surface with basalt lava that may have last occurred about 500 million years ago per Wikipedia.

      I may be totally wrong but is liquid water necessary for the formation of granite and lighter rocks vital for the the creation of continents? Without continents, weathering of rock and formation of limestone cannot occur to act as a CO2 sink. If Venus had oceans, perhaps it would have had plate tectonics and dinosaurs would be roaming the jungles (just wishful thinking). Perhaps it had oceans a few hundred million years ago according to some scientists.

      • Dave Moore May 8, 2021, 13:31

        A few points here:
        If there is a lower level of CO2 emissions, then the planet will be colder to a point where sequestration via weathering will match emissions.

        I think the over all level of volcanism is proportional in some way to the internal heat of a planet (see Io.) That been said, you may be right in that most of it occurs via massive flood basalts, which would lead to wild climate swings, if there were not plate tectonics. Note basalt lava is high in Calcium, so weathered lava would provide a good CO2 sink.

        Also of note is that plate tectonics leads to a bimodal distribution of elevations: ocean floor, continental surface. Non tectonic planets: Venus, Mars, have rough unimodal elevation distributions. If Earth had that elevation distribution, its oceans would cover a far higher percentage of the planet, meaning good pole to equator oceanic circulation moderating temperatures, and a lower level of CO2 sequestration.

        Also, note that there are certain distinctive geological markers to tectonic plate movement, and as far as geologists can tell, tectonics didn’t get going until 3.2 billion years ago, which means that Earth got by with no plate tectonics for a billion years.

  • Geoffrey Hillend May 5, 2021, 16:28

    Since Earth’s volcanism is not going to stop within the next one hundred million years, plants will not remove all of the carbon dioxide and life will flourish. After 100 million years, the increase in the Sun brightness will become a problem long before the red giant phase. Also don’t forget the Urey reaction. It’s not just the removal of CO2 from the air by basalt rocks, but mostly by the rain which takes the CO2 out of the air. When CO2 combines with water it becomes carbonic acid which combines with calcium silicate to make calcium carbonate (limestone) and silica(quartz) on land. The calcium carbonate is transported to the ocean and the limestone builds up on the bottom of the sea instead of in the air.

    I left out an important part of the process. When there is too much carbon dioxide in the air, there are higher temperatures which causes more evaporation of sea water and more water vapor and clouds in the atmosphere. The result is more rainy days or more rain which eventually lowers the CO2. Eventually more carbon dioxide is expelled by volcanoes which raises the carbon dioxide levels and temperature back up again and it starts to rain more lowering the CO2, so the whole process starts over like a cycle which is why it is called the carbon cycle.
    With the snow ball Earth period, everything was frozen, so it took a very large amount of carbon dioxide in the air and a lot of volcanism over a long period of time to melt the ice. The CO2 levels were much higher at that time than today maybe more than one percent to melt the world wide frozen oceans, but what followed what a very hot period due to the high CO2 levels. The rain and life eventually lowered the CO2 and temperature.

    Also if all the carbon dioxide is removed all the oceans would freeze so we need some CO2 which is what happened with the snow ball Earth periods, the last one around 635 600 million years ago before there was any CO2 producing life like humans, animals, etc.

    Mars has a lot of CO2 because due to its low escape velocity and weak gravity, it only holds onto the heavier gases like CO2 due to Jeans escape.

    I forgot the reason why the Earth’s climate became cooler due to the breaking up of the super continent into the continents we see today. I think the more equal distribution of land and mountains on the Earth results in a cooling over a more wider area because it forms more boundaries, the wind over mountains and forced adiabatic lapse rate of rising, cooling air over coastline, hills and mountains.

  • Alex Tolley May 5, 2021, 18:26

    Also if all the carbon dioxide is removed all the oceans would freeze so we need some CO2 which is what happened with the snow ball Earth periods, the last one around 635 600 million years ago before there was any CO2 producing life like humans, animals, etc.

    Aerobic respiration by unicellular plants and animals, as well as bacteria, is a dominant cause of O2 depletion and CO2 production in the oceans. If the snowball earth did occur, it was unlikely due to low CO2 levels unless volcanism ended or weathering increased. It could have been due to other causes, like methane reduction or volcanism cooling the Earth with sulfate emissions, probably a combination of causes.

    I don’t know that there is any hard evidence the CO2 levels were greatly raised to end the glaciation, compared to before and during that glaciation.

    • Dave Moore May 6, 2021, 12:56

      |I don’t know that there is any hard evidence the CO2 levels were greatly raised to end the glaciation, compared to before and during that glaciation.|

      On top of of the shallow ocean glacial sediments of cryogenic periods lies a layer of CaCO3 about 10-30 ft thick. This is the sequestration of a massive amount of CO2 in a short time.

      • Alex Tolley May 6, 2021, 21:13

        Thanks. Do you have a reference for that I can read?

        • Dave Moore May 8, 2021, 12:56

          Scientific America, Jan. 2000, Snowball Earth. It’s a little dated, but this article gives a good overview of Earth’s cryogenic periods and their aftermath. P71-72 covers the cap carbonates.

      • Alex Tolley May 6, 2021, 21:25

        In addition:
        If the oceans cool, doesn’t CaCO3 precipitate out? So carbon is sequestrated out from the large ocean reservoir (50x > atmosphere carbon). But what impact does this have on atmospheric CO2 levels that are responsible for surface temperatures? I understand that glaciation increases albedo that causes positive feedback cooling the Earth further. Reducing CO2 respiration by bacteria might certainly occur by ocean cooling reducing metabolic rates, and I can certainly understand that any glaciation that covers the oceans acts as a trap for that CO2 preventing it from reaching the atmosphere.

        It would be interesting if any reference you can provide for the CaCO3 sediment includes some information/reference on concurrent atmospheric CO2 levels during this extraordinary period.

  • Patient Observer May 6, 2021, 12:45

    I will continue my lonely quest of reminding anyone who cares that 50% or more of heat generated by nuclear processes originates at the Earth’s core in the form of nuclear fission (or fusion?) in a natural nuclear reactor:

    Here is the argument for fission:

    Others say the nuclear reactions are fusion:

    If just seems odd that such a potentially important finding are ignored in discussion regarding Earth’s internal heat and its source(s). This situation is vaguely reminiscent of the resistance to continental drift. The evidence was there but many scientists ignored it for whatever reason.

    • Ron S. May 6, 2021, 23:02

      It isn’t a nuclear reactor. It’s fission of atoms distributed throughout the Earth’s interior. It has been know about for a long time. The total heat from the interior has also long since been quantified but it has been difficult to quantify the fraction due to fission. That fraction is now better measured, and that consequently improves the estimate for the residual heat due to planet formation (gravitational compression).

      Temperature, pressure and density of suitable matter in the Earth’s interior is orders of magnitude lower than that required for fusion, other than statistically very rare quantum reactions.

  • Geoffrey Hillend May 6, 2021, 16:40

    There is of evidence of the snowball Earth periodhttps://en.wikipedia.org/wiki/Snowball_Earth The evidence is C13/C12 carbon isotope ratios in the glaciogenic rocks. See Carbon isotope ratios, footnote 36 . Some of the evidence is arguable. I don’t think the oceans froze solid, but advocate the slushball Earth where only the surface of the oceans were frozen . Wikipedia says the CO2 levels rose to 13 percent before the glaciers melted.

    As far as the atmospheric science is concerned, Methane only hangs around for 9 years, but it is a 28 percent more effective greenhouse gas than carbon dioxide, but CO2 hangs around for much longer. It takes fifty thousand years for the carbon cycle to remove roughly 100 parts per million of carbon dioxide from our atmosphere.

  • Geoffrey Hillend May 6, 2021, 16:42

    Excuse me for the mistake. Methane is a 28 times more efficient greenhouse gas than CO2, not 28 percent.

Leave a Comment