≡ Menu

A Second Super-Earth for K2-18

The transiting red dwarf K2-18 is about 111 light years out in the general direction of the constellation Leo, with a mass of 40 percent of Sol’s. A super-Earth, K2-18b, was detected here in 2015 through light curve analysis of data from the reconfigured Kepler K2 mission, and we now have the first measurement of the planet’s mass, drawing on radial velocity data from HARPS. The two planet detection methods in conjunction thus firm up our knowledge of a possible habitable zone planet.

But they also reveal, in the analysis of Ryan Cloutier (University of Toronto) and colleagues, a second super-Earth, K2-18c, which turns out to be non-transiting, and therefore non-coplanar with K2-18b. As we saw yesterday, HARPS (High Accuracy Radial Velocity Planet Searcher), is capable of drilling down to about one meter per second in the analysis of the stellar wobbles that radial velocity methods examine. The current data set gives us another interesting world while reminding us of the capabilities of the ESPRESSO spectrograph I talked about yesterday.

Image: An artist’s conception of planet K2-18b, its star K2-18 and the second planet K2-18c. Credit: Alex Boersma (www.alexboersma.com). Two things I like about this image: Its evocative feel, an imaginative take on a world whose actual conditions are speculative, and the fact that the artist is given credit by the Université de Montréal Institute for research on exoplanets (iREx). Artists should always get credit but all too often do not in press materials.

When it came to K2-18b, the challenge was to figure out whether the planet was a gaseous mini-Neptune or a predominately rocky world. Hence the importance of the radial velocity measurements, giving us in combination with earlier transit data a tentative mass as well as a radius, allowing the planet’s density to be inferred. The researchers believe K2-18b is either a large, rocky planet with a small gaseous atmosphere or a water planet with an ice crust. The range is explained by the uncertainty in the measured mass, which the paper cites at 24%.

It’s interesting to note that this planet’s mass is close to that of another possible habitable zone planet, LHS 1140b, though the dissimilarities are also striking:

K2-18b is of a similar mass to the habitable zone planet LHS 1140b (Dittmann et al. 2017) and receives a comparable level of insolation despite being ∼ 1.6 times larger than LHS 1140b. Analyzing the mass-radius relationship of these small planets over a range of equilibrium temperatures is a critical step towards understanding which of these systems have retained significant atmospheric content thus making them more suitable to extraterrestrial life.

Given the inability to distinguish between the two outcomes, we have to look to future instruments, like the upcoming James Webb Space Telescope. Cloutier points out that JWST should be able to probe the atmosphere of K2-18b to make the call. We keep bumping up against this ceiling as we wait for new instruments to come online. A nearby transiting world, this planet should climb toward the top of the target list for atmospheric analysis. K2-18 is now the second brightest M dwarf with a transiting habitable zone planet behind LHS 1140.

On JWST, by the way, be sure to read the always reliable Lee Billings’ article What Will NASA’s Biggest-Ever Space Telescope Study First? for a look at the inevitable prioritization of observing time and also the potential of the space telescope. A quick bit from the essay:

Unlike Kepler, which simply surveyed a single star-packed field of view for transits, Webb can zoom in on individual transiting worlds for deeper study. Astronomers should be able to use it to detect water vapor, methane, carbon dioxide and other gases in some silhouetted planets’ upper atmospheres by monitoring the starlight streaming through. They could also record a planet’s passage in front of and then behind its star, using the difference between the two observations to crudely measure a world’s temperature, weather patterns and clouds. “Webb is going to be great for exoplanets,” says Kepler project scientist Natalie Batalha, an astronomer at NASA Ames Research Center who leads the most time-intensive ERS program, which will use nearly 80 hours of Webb’s time to study transiting worlds. “It’s just that this is a difficult game to play, because the signals we’re looking for are really tiny. Seen [in transit] from another star, Venus blocks one part per ten-thousandth of the sun’s light, and its atmosphere intercepts one two-hundredth of that. It’s tough to see—you need a big mirror and great instrumentation to do it.”

As to newly detected K2-18c, the researchers looked for but did not find significant transit timing variations (TTV) in the K2 data. The Cloutier paper continues:

The orbit of the newly discovered K2-18c lies interior to that of K2-18b and yet the planet is non-transiting. This implies that the orbital planes of the planets are mutually inclined. In order for K2-18c to not be seen in-transit the planetary system requires a mutual inclination of just ≿ 1.4° which is consistent with the observed distribution of mutually inclined multi-planet systems (Figueira et al. 2012; Fabrycky et al. 2014).

The team’s dynamical simulations show that the oscillation timescale of the planets’ orbital inclinations is in the range of 106 years, meaning that it may be a long time before K2-18c can be seen in a transit from Earth. And the authors believe that finding the second planet through radial velocity data emphasizes the prevalence of Earth- to super-Earth size planets in M-dwarf systems, with multi-planet systems offering potential insights into planet formation processes around M-dwarfs. Comparative planetology here we come.

The paper is Cloutier et al., “Characterization of the K2-18 multi-planetary system with HARPS: A habitable zone super-Earth and discovery of a second, warm super-Earth on a non-coplanar orbit,” submitted to Astronomy & Astrophysics (preprint).


{ 15 comments… add one }
  • Tom Mazanec December 7, 2017, 13:32

    With 2018 closing in, what can we expect in Exoplants for the year?

    • andy December 8, 2017, 4:22

      Exoplants would be a huge milestone, no? :-P

  • Andrew LePage December 7, 2017, 16:25

    The current density measurements for K2-18b are consistent with a “predominantly rocky planet with a significant gaseous envelope or an ocean planet with a water mass fraction ≳50%”, according to the preprint’s abstract. In other words, this exoplanet with half the mass of Neptune is likely to have more in common with that ice giant than the Earth. This largely confirms my analysis on the habitability prospects for K2-18b (also known as EPIC 201912552b) I made back in May 2015:


  • Michael Fidler December 7, 2017, 23:23

    Noticed that LHS 1140 does not seem to have a history of flaring and there may be a reason why so many M dwarfs do. Been looking at some videos of comets hitting the sun taken by the many NASA solar spacecraft. What interesting is that they show coronal mass ejection (CME) and flares as the impacts take place but not where the comet hits. This could be why M dwarfs have so many flares and CME because of a large reservoir of comets that orbit and impact on them, since M dwarfs have a smaller surface area then the sun these cometary impacts would cause the large flares observed on them. M dwarfs that do not flare often, may have depleted the cometary reservoir or there Kuiper belt does not have a mechanism that cause them to form a highly elliptical orbits.
    The impacts seem to cause a trigger in the solar magnetic field that releases the distant flare and CME.


    • Michael Fidler December 8, 2017, 22:48

      Which of Kepler’s Stars Flare?


      “This large statistical study led the authors to several interesting conclusions, including:
      1. Roughly 3.5% of Kepler stars in this sample are flaring stars.
      2. 24 new A stars are found to show flaring activity. This is interesting because A stars aren’t thought to have an outer convective zone, which should prevent a magnetic dynamo from operating. Yet these flaring-star detections add to the body of evidence that at least some A stars do show magnetic activity.
      3. Most flaring stars in the sample are main-sequence stars, but 653 giants were found to have flaring activity. As with A stars, it’s unexpected that giant stars would have strong magnetic fields — their increase in size and gradual spin-down over time should result in weakening of the surface fields. Nevertheless, it seems that the flare incidence of giant stars is similar to that of F or G main-sequence stars.
      4. All stellar types appear to have a small fraction of “flare stars” — stars with an especially high rate of flare occurrence.
      5. Rapidly rotating stars are more likely to flare, tend to flare more often, and tend to have stronger flares than slowly rotating stars.”

      So where are all the Red Dwarfs that are burning there planets to the core???

    • Michael Fidler December 8, 2017, 23:02

      If comets cause some of the flaring around red dwarfs, there could be a short period of high activity between long periods of low activity. Comets or asteroids may come in swarms that are caused by large swings in the orbital alignment of celestial bodies in orbit around the M Dwarfs.

      • Michael Fidler December 8, 2017, 23:50

        Case Studies of Exocomets in the System of HD 10180.

        “The aim of our study is to investigate the dynamics of possible comets in the HD 10180 system. This investigation is motivated by the discovery of exocomets in various systems, especially β Pictoris, as well as in at least ten other systems. Detailed theoretical studies about the formation and evolution of star–planet systems indicate that exocomets should be quite common. Further observational results are expected in the foreseeable future, in part due to the availability of the Large Synoptic Survey Telescope. Nonetheless, the Solar System represents the best studied example for comets, thus serving as a prime motivation for investigating comets in HD 10180 as well. HD 10180 is strikingly similar to the Sun. This system contains six confirmed planets and (at least) two additional planets subject to final verification. In our studies, we consider comets of different inclinations and eccentricities and find an array of different outcomes such as encounters with planets, captures, and escapes. Comets with relatively large eccentricities are able to enter the inner region of the system facing early planetary encounters. Stable comets experience long-term evolution of orbital elements, as expected. We also tried to distinguish cometary families akin to our Solar System but no clear distinction between possible families was found. Generally, theoretical and observational studies of exoplanets have a large range of ramifications, involving the origin, structure and evolution of systems as well as the proliferation of water and prebiotic compounds to terrestrial planets, which will increase their chances of being habitable.”

        “Our study shows different kinds of outcomes such as encounters with the planets, captures, escapes, and secular comet–planet resonances. The integration process itself included the four outer planets of the system, i.e., planet e, f, g, and h (see Table 1). Generally, comets with relatively large eccentricities were able to enter the inner region of the system, thus facing early close
        planetary encounters. Owing to our theoretical approach, the number of ejected comets considered in our simulation readily increased over time.”

        “Studies about exocomets yield a large range of implications, involving the origin, structure and evolution of systems as well as the proliferation of water and prebiotic compounds to
        terrestrial planets, which will increase their chances of being habitable. This latter aspect has been showcased by results from Rosetta (e.g., Capaccioni et al. 2015; Rickman et al. 2015; Bosiek et al. 2016), which indicate the pivotal role of comets regarding prebiotic chemistry and potential exobiology.
        To consider the importance of exocomets for possible life in exosolar systems another possible extension of our future work will include the detailed study of encounters and collisions with planet HD 10180 g located at the outer edge of the system’s habitable
        zone, comparable to the position of Mars in the Solar System. This exosolar planet is Neptune-like; therefore, not being able to host life. However, an exomoon or a Trojan-type object associated with HD 10180 g could still be habitable.”

  • andy December 8, 2017, 4:49

    So on the one side we have the theoreticians saying that the extended luminous pre-main sequence stage, enhanced activity and suchlike should be completely desiccating low-mass habitable zone planets around M dwarfs, while on the other side the observers are finding very volatile-rich planets in these environments. Do we expect there to be some kind of sweet spot between the two extremes, where Earth-like planets with oceans and continents can exist, or is that going to be an extremely small region of parameter space that is unlikely to be realised?

    • Michael Fidler December 8, 2017, 9:31

      Well I have 15 to 20 years left so hopefully will know by then!

    • Mauldred December 8, 2017, 11:40

      Two possibilities: either the models are wrong, or the observations are.
      I’d suspect the models need to be refined and adjusted to match the observations, but better and more precise data thanks to JWST would also help.

    • Nicky December 8, 2017, 17:29

      A planet orbiting a main sequence stage M-dwarf with perfect mass fraction of water (ocean and land both exist) is theoretically possible. For example, if a planet starts with an initial water mass that is not much greater than today’s earth ocean and mantle volatile reservoir, PMST would not completely desiccate the planet in many cases (see Ribas et al., 2016 & Bolmont et al., 2016 for more information). However, in order for these ocean-land planets to exist, water mass fraction has to be lower than 0.5%, which we can’t reach with current technology and can’t be distinguished from regular rocky planets (we have discovered many), and there is only little chance of determining the precise water mass fractions with JWST. Thus, this question will not be answered in the near future, but LUVOIR or HabEX in 2030s or 2040s will help us to constrain the probability of ocean-land planets orbiting M-dwarf.

      • Nicky December 8, 2017, 17:32

        The next generation ground-based telescopes also have the capability of imaging habitable zone planets orbiting M-dwarf.

    • Michal December 13, 2017, 6:28

      Bear in mind, “Earth” and “Neptune” are probably just two points in a spectrum. There may be massive, drenched planets that nevertheless support marine life. The ice at the ocean bottom would be a problem, but comets (and such large planets are certain to attract a lot!) could bring a lot of minerals and organics.

      As for dessication, those studies have been way overblown by pop science press. Mind, those red dwarf planets are supposed to have lost about the same amount of water as our oceans. So they must be deserts right? Wrong, Earth is just 0.02 percents water by mass. A body with an Europa-like composition (25 percent water ice, 75 percent rock) or like this planet (+50 percent water)would lose a miniscule fraction of water from flares, it would barely make a dent! To aliens from a waterworld, Earth would be a dessicated husk not worth including in “potential habitability” surveys. Yet, open bodies of water could exist with just 0.2 percent of Earth’s water http://www.worlddreambank.org/T/THARN.HTM !

      I think the thought of completely dry, dead, red dwarf HZ planets is nonsense for several reasons. First, even if all water (and it only takes very little water to have open lakes) from the planet is boiled away, comets will still bring water from the outer system. Second, a wet planet may barely lose any water at all if it formed beyond the snowline. Third, a desert planet might actually be more protected against runaway greenhouse and get Venuslike insolation without going full greenhouse.

      I also seriously doubt that a world like this will be “ice covered.A world this massive probably has a thick atmosphere and water vapor is a great greenhouse gas. I remember reading that an ocean planet 1 AU from the Sun might go runaway greenhouse so I think HZ for ocean planets should be considered further away.

  • Harry R Ray December 14, 2017, 11:15

    Other exoplanet news on what appears to be A VERY VERY BUSY DAY for exoplanet news IN GENERAL: A ~Jupiter mass planet in a 7.75 hour circular orbit that may have filled its Roche lobe and whose temperature has been detected by the Keck Telescope. “PSR-J2322-2650 – A low luminosity milisecond pulsar with a planetary-mass companion.” by Spiwak et al. This planet probably started out as either a Super-Jupiter or a Brown Dwarf and had most of its atmosphere blown away by the pulsar beam, which appears to oe of signifigantly LOWER strength than STANDARD milisecond pulsar beams dor some reason.

Leave a Comment