≡ Menu

LHS 3844b: Rocky World’s Atmosphere Probed

These days we have a keen interest in small red dwarf stars (M-dwarfs) not only because they’re ideal for study, with deep transits of worlds in their habitable zones and the prospect of future analysis of their atmospheres, but also because they are so plentiful. Comprising perhaps 80 percent of all stars, they may well be home to the great majority of planets in the galaxy. And while they are common, they’re also long-lived, so that life would have plenty of opportunity to develop.

Now we have word of new work using both the Transiting Exoplanet Survey Satellite (TESS) and the Spitzer Space Telescope. TESS is, of course, a transit hunter, looking for the telltale dips in light from a parent star when a planet passes in front of it. The planet in question is LHS 3844b, about 48.6 light years out, and discovered by TESS in 2018. Follow-up observations in the infrared with Spitzer have detected light from the surface of this newly discovered world, allowing study of its atmosphere and composition. Note: This is not direct imaging; see below for more on the techniques used.

LHS 3844b orbits its star in 11 hours, making it almost certainly tidally locked; i.e., with one side always facing the star. The Spitzer data show that the dayside here reaches 770 degrees Celsius, while the nightside temperature is consistent with 0 Kelvin. In other words, the researchers could detect no heat being transferred from one side to the other, a process we would expect in the presence of an atmosphere.

Heat transfer is a mechanism that could ameliorate the effects of tidal lock, spreading warmth to the dark side and moderating global temperatures, but it takes an atmosphere to do that. We learn, then, that LHS 3844b is an object something like the Moon, or at any rate, a large version of it. Laura Kreidberg (Harvard-Smithsonian Center for Astrophysics), lead author of the paper that appears in Nature, says that this planet “…matches beautifully with our model of a bare rock with no atmosphere.” The scientist continues:

“We’ve got lots of theories about how planetary atmospheres fare around M dwarfs, but we haven’t been able to study them empirically. Now, with LHS 3844b, we have a terrestrial planet outside our solar system where for the first time we can determine observationally that an atmosphere is not present.”

Image: This artist’s illustration depicts the exoplanet LHS 3844b, which is 1.3 times the mass of Earth and orbits an M dwarf star. The planet’s surface may be covered mostly in dark lava rock, with no apparent atmosphere, according to observations by NASA’s Spitzer Space Telescope. Credit: NASA/JPL-Caltech/R. Hurt (IPAC).

This is painstaking analysis indeed, drawing on phase curve data from the planet’s transits. Phase curves are a combination of reflected light and thermal emission from the planet. Unable to resolve the planet from the host star, astronomers must work with their combined light, and observe light variations of exquisite subtlety as planets go through phase changes as they orbit. A phase curve, then, is the time-dependent change in the brightness of a planet as seen from Earth during one orbital period.

Image: Detecting Light from Exoplanet LHS 3844b. Credit: NASA/JPL-Caltech/L. Kreidberg (CfA | Harvard & Smithsonian).

Learning about the atmosphere (or lack thereof) of a small rocky world — LHS 3844b has a radius 1.3 times that of Earth — is therefore something of a coup, and bodes well for future discovery. The authors infer from the planet’s reflectivity (albedo) that it is covered with basalt, much like the mare of the Moon, which is probably an indication of volcanic activity in the distant past. From the paper:

We modeled the emission spectra of several rocky surfaces and compared with the measured planet-to-star flux… We considered multiple geologically plausible planetary surface types, including primary crusts that form from solidification of a magma ocean (ultramafic and feldspathic), secondary crust that forms from volcanic eruptions (basaltic), and a tertiary crust that forms from tectonic re-processing (granitoid). Governed by the reflectivity in the visible and the near-infrared and the emissivity in the mid-infrared, the surface types have distinct emission spectra. The measured planet-to-star flux for LHS 3844b is most consistent with a basaltic composition. Such a surface is comparable to the lunar mare and Mercury, and could result from widespread extrusive volcanism.

But this is a world much larger than the Moon, so what happened to its atmosphere? M-dwarf flare activity is thought to erode early planetary atmospheres, especially given how closely worlds like this orbit their star. The researchers rule out an atmosphere of over 10 bars (Earth’s atmospheric pressure at sea level is about 1 bar), and largely rule out one between 1 and 10 bars. They believe stellar winds and flares are the culprit. Modeling atmospheric escape over time, they assess how an early atmosphere dissolves within a magma ocean during planet formation or photolyzes into hydrogen and oxygen because of the intense bombardment of X-rays and UV from flares.

Thus a thick atmosphere is ruled out by the data, while stellar winds could account for further erosion of a thin atmosphere, all leading to the conclusion that LHS 3844b is a bare rock unless a thin atmosphere is replenished over time. Should we assume that hot terrestrial planets orbiting well inside the habitable zone of M-dwarfs are all devoid of atmospheres? Perhaps, but these stars may still be of astrobiological interest:

The results presented here motivate similar studies for less-irradiated planets orbiting small stars. Cooler planets are less susceptible to atmospheric escape and erosion, and may provide a friendlier environment for the evolution of life. In coming years this hypothesis can be tested, thanks to the infrared wavelength coverage of the James Webb Space Telescope and the influx of planet detections expected from current and future surveys.

The paper is Kreidberg et al., “Absence of a thick atmosphere on the terrestrial exoplanet LHS 3844b,” Nature 19 August 2019 (abstract / preprint)

{ 24 comments… add one }
  • Alex Tolley August 20, 2019, 11:01

    0K? The background temperature of the universe is approximately 4K. Any radioactive decay would add to the heat flux. Is there also no conduction or other means to transfer even a smidgeon of heat to the other side of the planet? So a very low temperature is possible, but not absolute zero. That seems extreme.

    • pochimax August 21, 2019, 8:23

      The original paper said:
      “The values correspond to a dayside brightness temperature of 1040 ± 40 K, and a nightside brightness temperature broadly consistent with zero (0 − 710 K at 1σ confidence).”
      It is far away from saying temperature on the nightside is zero.

  • Robert August 20, 2019, 11:51

    Do they use Mercury and Venus transits as calibration studies?

  • Ashley Baldwin August 20, 2019, 12:41

    So now we know what no atmosphere looks like. Transit spectroscopy has already constrained the thick envelopes of gas giants. All is now set up nicely to push the envelope further. Hopefully to “somethings” in between for the TRAPPIST-1 system as the next step. That’s the nature of science – incremental gains. Rather than great leaps. Biosignatures come after that.

    I can’t say I’m surprised to find that a relatively low mass planet orbiting just 0.006 AU out from its star has had its atmosphere stripped. Even for a dim M5 dwarf . Regardless of any additional flare and UV activity, the orbital radius of this planet is just half of even the closest in TRAPPIST-1 planet “b” .

    For a star fifty times more luminous.

  • Bill Spillman August 20, 2019, 12:58

    Just to nitpic a little. It is unlikely that the planet’s darkside is at a temperature of 0 Kelvin. That would put it at absolute zero. A typo perhaps?

    • Paul Gilster August 20, 2019, 13:49

      Not a typo. From the paper:

      “The observed phase variation is symmetric and has a large amplitude, implying a dayside brightness temperature of 1040±40 kelvin and a nightside temperature consistent with zero kelvin (at one standard deviation).”

      • Ron S. August 20, 2019, 14:51

        It may just be sloppy wording. If they detect no emissions the best they can hope to do is set a maximum temperature, which is in the vicinity of the minimum detection threshold. Calling it consistent with 0 K is true but misleading.

        • Paul Gilster August 20, 2019, 15:03

          I think that’s the best explanation as well, Ron.

          • Steve Muise August 20, 2019, 17:07

            One wonders if the data collected doesn’t allow for a higher resolution result and therefore the 0 K measurement is some sort of default result coming not from the data itself but via some combination of relatively low-resolution data fed into their algorithm.

    • Alex Tolley August 20, 2019, 15:00

      a nightside brightness temperature broadly consistent with zero (0 – 710 K at 1 [stddev] confidence).

      Their mean = 355K +/- 1 sigma = 355K, or mean = 0, 1 sigma = 710K ? As the lower bound is fixed, unless they used a transformed scale (e.g. log), the use of a Gaussian distribution is not valid.

      This and my other reasons makes me question the 0K dark side temperature value.

      • Bruce D. Mayfield August 20, 2019, 19:04

        Very poorly worded to call this “consistent with zero K”. I wouldn’t be surprised if some reporter files a story on this claiming “Coldest spot in universe discovered orbiting nearby star!”

        • Ashley Baldwin August 21, 2019, 12:06

          I now finally understand the difference between “proof reading” and “sense checking”.

  • Robin Datta August 20, 2019, 13:41

    If this situation is typical of most stars and most “habitable zone” planets, it may go quite a way towards explaining the absence of detected life in the universe.

    • Steve Muise August 20, 2019, 17:17

      Exactly my thoughts as well. The long life of these stars implies a longer amount of time within which flares can strip the atmospheres from any planets orbiting close enough. I would imagine the probability of finding an extent atmosphere on a planet orbiting close to am M-dwarf to be fairly low…although I’d love to be wrong about this I have a hard time envisioning a set of circumstances where such a planet could avoid having its atmosphere at least partially stripped by flare activity over these time spans.

      I’m not sure there’s really anything habitable about the habitable zone around this class of stars.

  • wdl August 20, 2019, 14:10

    A very clear presentation and picture of something we do not want to see or hear. At the very least, it shows that it can happen. How commonplace this is remains to be seen, so to speak. But based on initial conditions such as orbital radius and mass, if which some are volatile mixes, I can’t see this simply as already settled for all M stars.

  • Mike Serfas August 20, 2019, 15:48

    To me this looks like “good news” for habitability by Earthlike life. The planet was far inside the nominal habitable zone, yet habitable temperatures lie right at the middle of that broad one-sigma interval for the nightside temperature. The atmosphere is estimated probably to be less than 0.1 bar if it contains CO2, by a simple model, but this doesn’t solidly exclude a breathable fractional-bar oxygen atmosphere over much of the planet.

    More importantly, just as the Moon offers a 20km range of altitudes, so might this planet. If the low spot were on the nightside, that allows for a respectable atmosphere in a limited area with relatively little circulation to the dayside (perhaps just enough to keep things warm and provide a heat engine to power a chemical ecosystem). True, this is unlikely — but the alternative result was that we would be looking at a smooth white Venus without any refuge in sight.

    Unless, that is, there is some theoretical reason why a tidally locked planet would tend to move a deep basin to a different location?

  • Geoffrey Hillend August 20, 2019, 15:50

    LHS 3844b is a black body radiator, so thermal emission spectroscopy was used to find out it’s temperature of 770 centigrade. 770 C equals 1418 degrees fahrenheit, hotter than the surface temperature of Mercury and Venus. I am not an expert on ratios, but LHS 3844b might get more light than Mercury. Besides mass, temperature is the other factor used to determine atmospheric pressure and how long a planet can retain an atmosphere and volatiles. At 1418 F all types of atmospheric gas atoms and molecules are propelled past the escape velocity of 1.3 Earth masses which is LHS 3844b, so it’s lack of atmosphere should not be a surprise.

  • Geoffrey Hillend August 20, 2019, 16:21

    The atmospheric escape is also determined by the type of gas. Hydrogen and Helium are lighter, so they move faster than heavier gases when heated and escape easily. Co2 is heavier but at high temperature all gases will escape given a large amount of thermal energy in addition to solar wind stripping.

  • Laintal August 20, 2019, 17:05

    “These days we have a keen interest in small red dwarf stars (M-dwarfs)”

    Me too thats why I have been enjoying reading your posts on the subject Paul.

    A very interesting article, there could be a exosphere from the steller wind, volcanic activity and impacts this could result in a frozen polar cap on the night side of this planet.

    Now I’m off to read the Nature paper

  • Geoffrey Hillend August 21, 2019, 23:04

    To be more precise the temperature is based on Jeans escape. Each gas has a different particle speed at fixed temperature since every gas has a different molecular weight. The particle speed of a gas has to be a fraction or one fifth of the escape velocity, so the heavier gases remain longer than the light ones. Pluto would not have any atmosphere if it was in the life belt or much closer since the gravity and escape velocity are low but the temperature is low at Pluto’s distance so it can keep an atmosphere. The same is true about Titan in the life belt would lose it’s atmosphere due to higher temperature and low escape velocity.

    Temperature changes that If the planet is too close to the star, so it does not mater what the gas molecular weight is. All gases will be accelerated well above one fifth the escape velocity of 1.3 masses.

  • Laintal August 22, 2019, 5:43

    Erosion of an exoplanetary atmosphere caused by stellar winds

  • Geoffrey Hillend August 22, 2019, 15:47

    It is always both the stellar winds and Jeans escape since it is easier for the solar wind to strip an atmosphere from a planet with a low escape velocity like Mars for example. The Earth has a magnetic field, so we don’t have to worry about sputtering which is the solar wind stripping of an atmosphere.

Leave a Comment