≡ Menu

Modeling Early JWST Work on TRAPPIST-1

So much rides on the successful launch and deployment of the James Webb Space Telescope that I never want to take its capabilities for granted. But assuming that we do see JWST safely orbiting the L2 Lagrange point, the massive instrument will stay in alignment with Earth as it moves around the Sun. allowing its sunshield to protect it from sunlight and solar heating.

Thus deployed, JWST may be able to give us information more quickly than we had thought possible about the intriguing system at TRAPPIST-1. In fact, according to new work out of the University of Washington’s Virtual Planetary Laboratory, we might within a single year be able to detect the presence of atmospheres for all seven of the TRAPPIST-1 planets in 10 or fewer transits, if their atmospheres turn out to be cloud-free. Right now, we have no way of knowing whether any of these worlds have atmospheres at all. A thick, global cloud pattern like that of Venus would take longer, perhaps 30 transits, to detect, but is definitely in range.

“There is a big question in the field right now whether these planets even have atmospheres, especially the innermost planets,” says Jacob Lustig-Yaeger, a UW doctoral student who is lead author of the paper on this work. “Once we have confirmed that there are atmospheres, then what can we learn about each planet’s atmosphere — the molecules that make it up?”

Image: New research from UW astronomers models how telescopes such as the James Webb Space Telescope will be able to study the planets of the intriguing TRAPPIST-1 system. Credit: NASA.

Working with Lustig-Yaeger are UW’s Victoria Meadows, principal investigator for the Virtual Planetary Laboratory, and doctoral student Andrew Lincowski. The latter should be a familiar name if you’ve been following TRAPPIST-1 studies, because back in November of 2018 he was lead author on a paper on climate models for this fascinating system (see Modeling Climates at TRAPPIST-1).

We’ll now be hoping to follow up that work with early JWST data. Briefly, Lincowski and team pointed to the extremely hot and bright early history of TRAPPIST-1, a tiny M-dwarf 39 light years out with a radius not much bigger than Jupiter (although with considerably more mass — the star is about 9 percent the mass of the Sun). These early conditions could produce planetary evolution much like Venus, with evaporating oceans and dense, uninhabitable atmospheres. The Lincowski paper, though, did point to TRAPPIST-1 e as a potential ocean world.

These findings were in the context of a system among whose seven transiting worlds are three — e, f and g — that are positioned near or in the habitable zone, where liquid water might exist on the surface. Now we have Lustig-Yaeger and company modeling our early JWST capabilities. The paper finds that beyond the presence of an atmosphere, we may be able to draw further conclusions, particularly with regard to the evolution of what gas envelopes we find.

Although oxygen as a biosignature may not be detectable for the potentially habitable TRAPPIST-1 planets, oxygen as a remnant of pre-main-sequence water loss may be easily detected or ruled out… the 1.06 and 1.27 µm O2-O2 CIA [collisionally-induced absorption] features are key discriminants of a planet that has an oxygen abundance greatly exceeding biogenic oxygen production on Earth and may therefore indicate a planet that has undergone vigorous water photolysis and subsequent loss during the protracted super-luminous pre-main-sequence phase faced by late M dwarfs,,,

Such features could be detected fairly quickly:

… in as few as 7-9, 15, 8, 49-67, 55-82, 79-100, and 62-89 transits of TRAPPIST-1b, c, d, e, f, g, and h, respectively, should they possess such an atmosphere. These quoted number of transits may be sufficient to rule out the existence of oxygen-dominated atmospheres in the TRAPPIST-1 system. Additional evidence of ocean loss could be provided by detection of isotope fractionation, which may also be possible in as few as 11 transits with JWST.

Moreover, the authors find that water detection could help to pare down various evolutionary scenarios on these worlds, particularly for TRAPPIST-1 b, c and d, assuming atmospheres high in oxygen content that have not been completely desiccated by the star’s early history. Thus we are probing planetary evolution, but assessments of habitability are going to be tricky, and it seems clear that we will need to turn such analysis over to future direct-imaging missions.

On balance, we are talking about getting useful results with a fairly low number of transits. JWST’s onboard Near-Infrared Spectrograph will use transmission spectroscopy — where the star’s light passes through a planet’s atmosphere to reveal its spectral ‘fingerprint’ — to detect the presence of an atmosphere via the absorption of CO2. Such analysis can likewise either detect or rule out oxygen-dominated atmospheres, while constraining the extent of water loss through measurements of H2O abundance. All of this provides fodder for other, still evolving observing strategies using the JWST instrument package that can begin the characterization of these compelling worlds.

The paper is Lustig-Yaeger et al., “The Detectability and Characterization of the TRAPPIST-1 Exoplanet Atmospheres with JWST,” Astronomical Journal Vol. 158, No. 1 (21 June 2019). Abstract / preprint. The Lincowski paper referenced above is “Evolved Climates and Observational Discriminants for the TRAPPIST-1 Planetary System,” Astrophysical Journal Vol. 867, No. 1 (1 November 2018). Abstract / Preprint.


Comments on this entry are closed.

  • Dave Moore August 15, 2019, 15:46

    I have been thinking about the atmospheric circulation of steam atmosphere planets since your last set of posts on the Trappist planets, and there is a factor, which doesn’t seem to have been taken into account.

    Let us assume we have a planet with a steam atmosphere, and the top of the atmosphere cools below the boiling point of water and eventually below freezing. Let us also assume that considerable water has been disassociated into Hydrogen and Oxygen with a large Oxygen surplus due to Hydrogen escape.

    An atmosphere of steam has a molecular weight of 18. A dry atmosphere of Oxygen, one cooled below the freezing point of water, has a molecular weight of 32. It would sink like a rock through the steam atmosphere.

    With the exception of H2, He, and CH4 (which could be oxidized to CO2) you cannot have a stable dry atmosphere over a steam atmosphere. You’d get violent convective overturn unless the dry atmosphere is considerably hotter than the steam atmosphere.

    An example, by looking at steam tables and air density tables, to get our atmosphere, dry, the same density as our atmosphere, saturated at 90 deg. C, you have to heat it to 300 deg C.

    This points to a lot of high cloud on the Trappist planets. It also points to planets with steam atmospheres being very efficient at cooling themselves because the transport a lot of heat high into their atmospheres were they can radiate it unimpeded into space.

    I can’t really go much further than that as I’m not an expert on atmospheric circulation.

    Slightly off topic, I did my initial thinking using a steam Venus as a model, but from what I can see Venus never had a steam atmosphere.

    Davis Grinspoon thinks that Venus didn’t suffer from run away greenhouse for its first 2 billion years. I don’t know the details as I haven’t got around to reading the paper yet.

    But modeling a planet like Earth and heating it up, you get more and more violent convection, which helps cool it. The temperature differential between the dry air and saturated air goes up pushing the Tropopause higher, which leads to more water loss.

    CO2 levels are drawn down as carbonate becomes less soluble with increasing ocean temp. (This can be seen on Earth as at the end of its cryogenic periods when it had a warn to hot, high in CO2 atmosphere and there are massive deposits of Calcium Carbonate.)

    So it looks like Venus fought off run-away heating as the sun got hotter until it effectively ran out of water. Only then would have the CO2 built up. (This may affect our estimates of the inner range of the habitable zone.)

    From their densities, the Trappist planets appear to have a steam over ice or ocean over high pressure ice composition with little contact between the lithosphere and the atmosphere, so their chemistry and evolution will be different from Venus’s, but the circulation and evolution of steam atmospheres needs to be taken into account. Generally, the higher the molecular weight of a molecule, the higher the boiling point. Water is unique in that it’s a very light molecule with a high boiling point.

    • Alex Tolley August 15, 2019, 20:09

      Your argument would suggest that an atmosphere would stratify into layers based on teh mol. wt. of the constituent gases. E.g. CO2 at teh base, and H2 at the top. Even our own N2-O2 atmosphere should be richer in O2 at the bottom of the column (excluding the proximity effects of photosynthetic planets). But in practice, mixing due to random motion, turbulence, and winds will ensure a relatively uniform mix of gases.

      I would have thought that a “steam planet” would be subject to so much energy that this mixing might even be quite violent, with fierce storms.

      • Mike Serfas August 17, 2019, 19:14

        Certainly this argument should hold in most circumstances … still, suppose a ‘hot Mars’ around a red dwarf has a chaotic rotation scheme, a fairly large mass of water, and (for now) a north polar basin, with a temperate zone near the boiling point of water. Water rains out of the atmosphere into the polar basin until much is stored as liquid. Then the basin swings out to an equatorial position. Eventually the water boils… sending a massive plume of hot water vapor from a fixed point high into the atmosphere? Could such a phenomenon allow brief periods when water molecules escape the planet intact and turn up in the star’s spectrum?

        • Alex Tolley August 18, 2019, 10:11

          Would a planet change obliquity so fast that it is analogous to placing a saucepan on a lit gas range?

          • Mike Serfas August 18, 2019, 17:35

            I certainly have no idea. Nix and Hydra shift very rapidly, but they are tiny. I’ve read that Mars has chaotic obliquity that may have led to major changes in its atmosphere over time, though apparently that’s still up for argument ( https://www.hou.usra.edu/meetings/lpsc2019/pdf/3276.pdf ), but Mars is much further from its star and other planets/resonances than TRAPPIST-like worlds. With a sufficiently outlandish set of sci-fi premises I imagine you could start by crash landing near a lake verdant with [heat-tolerant] short season annuals, and by the end be riding a wave of steam to the next planet out using only a spacesuit and a hang glider. :)

    • Harold Daughety August 16, 2019, 19:28

      David Moore -See “https://simple.m.wikipedia.org/wiki/Entropy” for an explanation why heavy gasses do not hug the surface of the earth. It’s good for life that it doesn’t; otherwise we would suffocate in a cloud of heavy noble gasses overlaid with layered CO2, nitrogen, oxygen, topped with water vapor. A thin layer of helium might top it off, but solar radiation would probably strip that away. Entropy is one of the primordial facts that makes life possible.

  • Alex Tolley August 15, 2019, 15:50

    So we might get data that teases us further. Enough to determine whether a planet is dry, cloudy, very wet, or even potentially Earth-like. But not enough to determine out-of-equilibria gases to indicate the presence of life. We need even bigger ‘scopes.

    • Ashley Baldwin August 17, 2019, 6:53

      For me Alex, the main issue is as to whether JWST shows that any, some or all of the TRAPPIST-1 planets have sustained any meaningful atmospheres over extended periods . This in the face of a representatively active M dwarf. There is plenty of theory that suggests that all M dwarf planets will have their primordial and secondary atmospheres eroded and stripped. JWST is shown here to be capable of detecting both primordial and secondary atmospheres. Around all 7 planets via NIRSPEC transit spectroscopy, with some assistance for the inner three planets with eclipse mapping via MIRI. That’s a good start and if the results are positive ( for ‘e’ especially ) and JWST can last ten years, it might be capable of identifying a terrestrial atyle atmosphere . Not biosignatures unfortunately but it would certainly be enough to heavily incentivise further more capable and bespoke missions for as good or better targets.

      • Alex Tolley August 17, 2019, 13:59

        I am perhaps not so interested in the Trappist-1 system as others are. There are so many issues with M-dwarfs as life-bearing worlds that I would be more interested in worlds that are around more sun-like stars. Is atmosphere detection by JWST problematic for worlds around these stars? If not, are there comparable analyses for hypothetical planets in the HZ around these stars so that we can target promising worlds?

        • Ashley Baldwin August 18, 2019, 17:32

          It doesn’t get any better than TRAPPIST. JWST observation JD Atmosphere detection/characterisation doesn’t come any more sensitive . Within the limits of its technology , aperture and various observation time constraints. It isn’t able to image the planets direct like HabEX. It can only act as a 6.5m ” light bucket” – feeding star and planetary light to its various IR spectrographs and then sifting out planetary atmospheric characteristics utilising transit spectroscopy and photometry of both primary and secondary planetary eclipses .

          The TRAPPIST system is currently the nearest known transiting planetary system, containing terrestrial sized and hab zone planets too. The most accessible of their type to JWST imaging. JWST will of course observe many other transiting planets but most of these will be close in gas giants with the expanded dense atmospheric envelopes most amenable to this technique – Unless TESS can find something better over the next two years.

          The closer in planets of the TRAPPIST system obviously have more frequent and deeper transits . They will offer the largest number of transits within JWST’s primary five year mission. However it can’t see every transit of every planet due to other mission commitments and the fact that the TRAPPIST-1 system sits close to the plane of the obscuring Milky Way ( which ironically aided its imaging in a Kepler K2 observation window) .

          JWST will be able to see about a quarter of the total planetary transits of each of the planets over five years ( 81 for “e”). Combining ( “binning”) these to build atmospheric spectra with the mnimum acceptable signal to noise ratio, SNR of between 5 and 10. Looking for the least thick/dense atmospheric envelopes possible. Theoretically better than for gas or ice giants but sadly without the sensitivity to reach down to an Earth style atmosphere. It’s the limits of the observations outcomes that are described and discussed in the article. If JWST lasts for the hoped for ten years, these limits will lessen, but never to biosignature level .

          Not least because of the high sensitivity required to detect the necessary low concentrations of methane out of equilibrium with a low bar O2 atmosphere and accompanied with H2O, and trace amounts of CO2 .

          One for the future..

  • Geoffrey Hillend August 15, 2019, 20:41

    @Dave Moore, there are other factors. The size of the planet which determines the escape velocity. A planet with a larger size has a higher escape velocity, so it retains an atmosphere better than one with a lower escape velocity and smaller size. Also the temperature which depends on the distance from the Star.

    What happened with Venus is it’s oceans evaporated into the atmosphere once the boiling point of water is reached. This could have happened even below the boiling point since higher temperatures evaporate more sea water and more water vapor traps more heat. Water vapor is a green house gas but less potent than Co2.

    When the oceans evaporate all the water is in the air, there is a big green house effect and a huge atmosphere which always keeps the temperature on the surface very high as on Venus for an example. The temperature on Venus got too high so any rain falling down to the surface always got evaporated before it reached the surface to cool it off. The water is lost through the solar stripping of it after the ultra violet splits the water into it’s constituents of hydrogen and oxygen high in Venus atmosphere. It depends on how much water the planet has and it’s surface temperature.

    A planet that did not have a lot of water and atmosphere would still be hot near the ground and cold in the upper atmosphere. A thin atmosphere can’t have a big green house effect even if the surface is hot. I would not be surprised to see most of the TRAPPIST-1 planets with atmospheres since Venus has one. With Earth, the water molecules are mostly trapped in the troposphere with a kind of heat sink with convection keeping the temperature uniform there, but that could change if the Earth moved out of the life belt closer towards the Sun and the surface temperature gets higher. This will happen in Earth’s future as the Sun gets hotter but not until over one hundred million years. in the future.

    It will be interesting to see if any atmosphere have been retained by the TRAPPIST-1 planets and their composition will tell something about their atmosphere’s evolution.

  • Geoffrey Hillend August 15, 2019, 21:03

    I made a mistake. There is a cold trap in the troposphere not a heat sink.

    • Ashley Baldwin August 17, 2019, 7:01

      I thought it was the stratosphere that acted as the ‘cold ( and equally as crucially – dry) trap’? It was when Venus heated up the point where it’s stratosphere became saturated that UV dissociation of water started a moist greenhouse that then became runaway greenhouse . Which created modern day Venus. A fate that awaits future Earth.

  • ljk August 16, 2019, 9:12

    Why Exoplanets Matter: In Conversation With Sara Seager

    MIT astrophysicist and MacArthur fellow Sara Seager on her tenacious search for a true Earth twin.

    By Adolfo Plasencia

    t the time when the first reports of exoplanets — planets that orbit stars other than our Sun — began appearing, Sara Seager, an astrophysicist and planetary scientist at MIT, was busy earning a doctorate in astronomy at Harvard University.

    Her groundbreaking research on extrasolar planets and their atmospheres has since earned her a prestigious Macarthur fellowship, an election to the National Academy of Sciences, and a spot in Time magazine’s list of the 25 ‘most influential people in space and astrophysics.’

    She also serves as Deputy Director of Science on TESS (Transiting Exoplanet Survey Satellite), an MIT-led NASA mission whose objective is to find planets outside our solar system, including those that could support life.

    Full interview here:


    To quote:

    My personal opinion about life that could traverse the galaxy, if we are now talking about life that could come to Earth, or in the future, if we’re able to travel to a distant star system, is that it probably has to be nonbiological because space is very harmful for people. We can barely survive on Earth, if you think about it, and Earth is a very safe, well-designed place for us, or rather we are adapted to our environment. So I think for us initially as human beings to find life elsewhere, it’s bound to be biological, since that’s all we can see; it’s all we know how to do. But if we ever think of traveling through the galaxy or of alien life coming here, then I believe on a personal level that it will be nonbiological.

  • Paul Wilson August 16, 2019, 13:12

    I hate the endless delays with JWST. Also uncomfortable about parking it where it can’t (at present) be “fixed” in space if something bad is wrong. Beyond that, I would not argue with anyone here about the importance of examining the atmospheres of red dwarf exoplanets in the hopes of finding a Life-viable one.
    That stipulated, I hope in the meantime TESS finds a “nice”, stable G star that could host planets that are not tide locked, melted, or bombarded by frequent solar flares. Ha, also a gravity of 1.0 earth mass would be sweet. Is that asking for too much?!

    • Ashley Baldwin August 17, 2019, 7:22

      With an predominant observation run of 27.4 days, TESS is unlikely to find any potentially habitable planet around a G dwarf. Even with maximal overlap of observation segments ( around the ecliptic poles – specifically to match the continuos field of view for JWST) the longest run in one calendar year is 351 days . Three transits for a clear proof of a planet, gives an maximum orbital period of 116 days . Which is early M dwarf or late K dwarf habzone territory . It may be that the recent two years extension to TESS will repeat the two hemisphere observation approach of the original mission. This would effectively double the maximal observation period to 702 days and the absolute maximal orbital period of any planet to 233 days – which would stretch to an early K dwarf like 40 Eridani A or Alpha Centauri B ( though neither star lies within this observation window ) .

      I suppose improved detection algorithms might allow qualification of a planet via just two transits ( with additional proof via CHEOPS follow up) in which case we could theoretically find a planet with a 351 day period – which would make it to the G dwarf hab zone . Would need to lie within the 4 % of the sky covered by the longest observation period though.

      Wishful thinking?

  • Geoffrey Hillend August 16, 2019, 15:53

    I would add one more thing to the 1.0 mass Earth twin, a Moon comparable to the size of our own, to give the planet as fast rotation, and a magnetic field. If it doesn’t have a Moon, it would be interesting to see the composition of it’s atmosphere to see if there has been any solar wind stripping.
    I hate the delays to the JWST also, but they have to make sure everything works correctly, otherwise the mission could fail. If some part of the JWST breaks or fails in the future, then NASA should accept the challenge of sending astronauts to fix the JWST since their goal of interplanetary travel includes greater distances than the JWST. I like the idea of an interplanetary spacecraft with VASIMR propulsion or at least some kind of roomy, large spacecraft that could make the journey for astronauts to the JWST fast and routine in case something breaks. I guess they want to build it so it lasts a long time since it is so far away they cant give it upgrades and maintenance.

    • Ashley Baldwin August 18, 2019, 4:09

      VASIMR continues to be dogged by development problems. Managing the large magnetic fields it produces as a byproduct along with an awful lot of waste heat not least. And devoloping the huge solar arrays necessary to power it,

      It should be remembered that even the Hubbke servicing missions in low Earth orbit by the space shuttle were considered risky. However, I guess plugging a Bigelow B330 habitation module onto ORION would create the outline of meaningful “deep space” tug . A very basic outline only though. But that’s just the ( very) beginning . JWST’s SEL2 orbit is nearly a million miles away . Transfer there would thus involve travelling considerably further. Four times further than the Moon . Any manned servicing mission would be out of Earth’s magnetosphere for a month or likely considerably longer . Even ignoring all the other (many) risks of deep space, or indeed manned space flight all round , still leaves it exposed to the intermittent but high risk of deadly solar CMEs to say nothing of continual bombardment from deadly cosmic rays . To the best of my knowledge there are currently no well developed ways of countering these very serious threats . Physical barriers, water or polythene lined safe havens have been mooted as the most realistic – but not yet devoloped very much.( a fact forgotten or ignored by all those advocating much longer still manned Mars missions) .

      A better bet would be to design post JWST telescopes to be 1/ Less complicated and more robust – with far less than 168 moving parts for starters ! 2/ To be robotically serviceable .

      The best bet though, and far easier – far cheaper and far safer – would be to attach the ion thruster to the telescope instead . Thus allowing it to return periodically as required to cis lunar space , or even low Earth orbit.

      • Alex Tolley August 18, 2019, 10:25

        exposed to the intermittent but high risk of deadly solar CMEs to say nothing of continual bombardment from deadly cosmic rays . To the best of my knowledge there are currently no well developed ways of countering these very serious threats . Physical barriers, water or polythene lined safe havens have been mooted as the most realistic – but not yet devoloped very much.( a fact forgotten or ignored by all those advocating much longer still manned Mars missions) .

        Zubrin has pretty much told his followers that the radiation risks for the trip to Mars and even living on the Martian surface are quite tolerable. I had a recent “argument” with one such follower on FB recently where high background radiation in inhabited places on Earth was offered as proof that the radiation risk was hugely overblown. Zubrin made a farcical calculation of the radiation risk on the Martian surface in his book: How to Live on Mars that seems to be one datum for this handwaving away the risk. If a potential Mars colonist wants to take that risk, fine, but this risk should not be diminished as almost inconsequential.

        I agree with you that it would be better to move the JWST back to LEO for local repair, rather than sending a crew out to the JWST’s orbit. It isn’t as if the telescope is fixed in its orbit like a telescope on Mauna Kea. Far too often we seem to use Earth-based experience for space operations.

  • Geoffrey Hillend August 16, 2019, 16:05

    What I meant to say about exoplanets is that it depends on how much water and atmosphere the planet has at the beginning right after the Star is born. A lot of water will make a pressure cooked planet if it is outside the life belt in front of it or too close to the Star.

    We don’t see anything because we don’t yet have the capability to see the evidence of life outside our solar system which is why we are building larger telescopes like the JWST, and European Extremely Large telescope, so the general collective opinion of life will change in the future.

  • Geoffrey Hillend August 18, 2019, 18:25

    It’s a good idea to tow the JWST back if some problems develop. I can now see why they want to get it right the first time and make sure everything works efficiently and lasts a long time.

    One thing I like about VASIMR is it does not emit any radioactivity like those old fashioned nuclear thermal rockets. VASIMR could be used to tow the JWST.

  • wdk August 19, 2019, 22:02

    About twenty or so years ago, in anticipation of this sort of thing had looked at some literature in circulation then about what bands to look for for application Alpha Centauri:

    CH4 bands (1.2, 4 microns) in the atmosphere of Gl 229B have already been used to estimate its mass, T*eff and surface gravity. Detection of absorption lines at 7, 9.5 or 15 microns respectively for H2O, O3 and CO2 in even fainter prospective objects could give direct comparisons with Earth and evidence of extra-solar terrestrial planets.

    I’m probably behind the curve on this in many ways. But first I had to find the candidate lines and then to see if JWST has a capability to look at them. No doubt someone already knows. But what caught my eye is that there is an absence of discussion of ozone (O3) and I don’t see a reason yet to rule it out as a marker. Unless that’s what the O2 estimates are actually based on. We are talking stars that have high energy storms; so even though they are redder black bodies, they should have some capability to generate ozone in an oxygen atmosphere, right? Are some of these lines below the wavelength window or is it the likely intensity?