≡ Menu

A Strong Case for TRAPPIST-1 Planets

TRAPPIST continues to be my favorite astrophysical acronym. Standing for Transiting Planets and Planetesimals Small Telescope, the acronym flags a robotic instrument at the La Silla Observatory in Chile that is operated by the the Institut d’Astrophysique et Géophysique (University of Liège, Belgium) in cooperation with the Geneva Observatory. The name is a nod to the branch of the Cistercian order of monks called Trappists, whose beer is world-renowned and closely associated with Belgium itself (although also brewed in the Netherlands and a few other countries). A jolly telescope indeed.

You’ll recall TRAPPIST-1 as the far more approachable term for the red dwarf star 2MASS J23062928-0502285, a bit over 39 light years away in the direction of the constellation Aquarius. A 2016 paper in Nature announced three rocky planets orbiting the star, one of which could conceivably be in its habitable zone, where liquid water can exist on the surface. Now we have a helpful follow-up from the 8-meter Gemini South telescope in Chile. A team led by Steve Howell (NASA Ames) has been able to rule out any close stellar companion.

That’s good news for planet hunters because what we have been looking at are fluctuations in the light of this small star (TRAPPIST-1 is only about 8 percent of the Sun’s mass), making the assumption that these were caused by the three planets mentioned above. Howell used the Differential Speckle Survey Instrument (DSSI) at Gemini South to demonstrate that there was no hitherto undiscovered small star complicating an already complicated planetary detection.

“By finding no additional stellar companions in the star’s vicinity we confirm that a family of smallish planets orbit this star,” says Howell. “Using Gemini we can see closer to this star than the orbit of Mercury to our Sun. Gemini with DSSI is unique in being able to do this, bar none.”

Speckle imaging, which is what the DSSI instrument Howell used in this work does, works by taking numerous extremely short exposures, allowing astronomers to combine the images and eliminate distortion caused by the Earth’s atmosphere. We wind up with high-resolution images that duplicate what the same telescope would produce if it were in space.


Image: Artist’s concept of what the view might be like from inside the TRAPPIST-1 exoplanetary system showing three Earth-sized planets in orbit around the low-mass star. This planetary system is located only 40 light years away. Gemini South telescope imaging, the highest resolution images ever taken of the star, revealed no additional stellar companions, providing strong evidence that three small, probably rocky planets orbit this star. Credit: Robert Hurt/JPL/Caltech.

Ongoing work may detect further planets in this system, but for now, what we have are the original three. The inner worlds are in orbits of 1.5 and 2.4 days respectively, both far too hot for liquid water on the surface; in fact, these two would receive four and two times the radiation the Earth does from the Sun respectively. The third planet’s orbital period has proven difficult to constrain, so that all we can say is that it is between 4 and 73 days, though this Gemini Observatory news release pegs the most likely period at 18 days, which would evidently be in the habitable zone. Confirming that would add to the sizzle of the recent Proxima Centauri b discovery.

Thus we continue to learn about TRAPPIST-1, a promising candidate for still more detailed work in coming years. M-dwarfs are small enough that planets in their habitable zone have short orbital periods. That means frequent transits, giving astronomers the chance to analyze their planets’ atmospheres by studying starlight as it is filtered through them. The odds on the Proxima Centauri planet transiting are slight (we’ll soon know more), but in TRAPPIST-1 we have the potential for follow-up with space-borne instruments and the coming generation of extremely large telescopes (ELTs) on the ground, to learn what kind of atmospheres such planets have, and whether we may eventually find biosignatures within them.

The Howell paper is “Speckle Imaging Excludes Low-Mass Companions Orbiting the Exoplanet Host Star TRAPPIST-1,” Astrophysical Journal Letters Vol. 829, No. 1 (abstract). The original work on TRAPPIST-1 is Gillon et al., “Temperate Earth-sized Planets Transiting a Nearby Ultracool Dwarf Star,” published online in Nature 2 May 2016 (abstract).


Comments on this entry are closed.

  • Andrew LePage September 13, 2016, 12:56

    > the most likely period at 18 days, which would evidently be in the habitable zone.

    Actually, an orbital period of 18.2 days (the most probable according to the discovery paper by Gillon et al.) would result in an effective stellar flux for TRAPPIST-1d of something like ~0.15 times that of the Earth, scaling from the various values given in the original discovery paper. With the outer limit of the TRAPPIST-1 HZ having a stellar flux around 0.23, an 18.2-day orbital period would place this exoplanet beyond the HZ.


    Based on the relative probabilities of various orbital solutions in Gillon et al., there is maybe about a 20% chance that TRAPPIST-1d orbits inside the conservatively defined HZ for a synchronously rotating, Earth-size planet. We will have to wait for results from new data from either TRAPPIST or the K2 Campaign 12 (which is scheduled to run from December 15, 2016 to March 4, 2017) to pin down the orbit of this exoplanet.

  • Harry R Ray September 13, 2016, 13:35

    As PREVIOUSLY NOTED in coments on this website, AFTER THE INITIAL DISCOVERY OF THE TWO INNER PLANETS WERE CONFIRMED by other GROUND-BASED telescopes, the discovery team requested TWO observing runs with the SST, one to confirm the “nature” of the TRAPPIST-1d transit signals(there were ONLY two), and the other to “characterize” THE WHOLE SYSTEM! Thse observation runs were for five plus and thirty two plus hours respectively in January and February respectively. The only result of these combined observation was the PREDICTION of the DOUBLE TRANSIT on May 4. No papers on these observations have EVER been SUBMITTED! To me, this means either one of two things. Either there are AMBEGUITIES that must be resolved with MORE OBSERVATIONS, which is not that likely since no more observing time has been requested, or, the results are both VERY COMPLICATED AND VERY INTERETING to the point that the team has held off submitting a paper out of fear of OUTRIGHT REJECTION(as was the case for the 2013 Proxima b paper)as opposed to a rejection with a request for further data. If that is the case, we may have to wait for TESS to get all the perameters for TRAPPIST-1d.

    • Ashley Baldwin September 14, 2016, 15:06

      A fair point and something to be constantly aware of when reading arXiv ( though it’s generally brilliant with publication either already arranged or a formality for most articles ) . The lack of peer review . Thanks for raising the issue.

      There are many reasons for non publication of course such as lack of interest from journals ( which are after all dependent on selling copy to make money like any periodical and thus charge more for advertising ) though this doubtful here for obvious reasons .

      Obtaining points of clarification ahead of peer review is reasonable though. I think we have all begun to take exoplanet discovery as the norm when discovery by transit or RV is excruciatingly hard and arduous work , especially given the increasing magnitude of new discoveries and their enormous implications . We’ve also seen high profile discoveries overturned on subsequent review, an occurrence that has led to many ” RV ” teams ( such as for Proxima b) using simultaneous stellar photometry to exclude spurious signals due to this , as seems to have been the case with “Alpha Centauri b” . That spectre now hovers ominously over every exciting exoplanet finding . The Proxima b signal was apparent in the data almost a year ago but it took till this summer to get sufficient evidence to publish the discovery unequivocally . Extraordinary discoveries require extraordinary evidence and all that.

      The first nearby M dwarf Hab zone planet to undergo detailed atmospheric spectroscopy will be momentous . Far better to get it right prepublication , whatever the delay . The evidence to date has already been sufficient to persuade the Hubble Telescope Imaging committee to grant observation time for transit spectroscopy .

      • Harry R Ray September 18, 2016, 15:44

        So THAT’S what the latest PRD tweet was all about! Hubble MUST be the “major facility” mentioned in the tweet. ALSO: Keep in mind that PRD gave Davenport, Kipping, et al ALL of the parameters of Proxima b WEEKS before the announcement! The CO-OPERATION between ALL OF THESE COMPETING GROUPS has just been SIMPLY AMAZING, and, hopefully; is a portant of things to come! If the Gaia team volunteers support as well, the case might be cracked VERY QUICKLY INDEED!

  • galacsi September 13, 2016, 15:26

    Now, seriously !

  • Moebius September 14, 2016, 3:00

    40 ly away sounds like quite far even for Starshot to get there and give us more information about the surfaces of these worlds. Given a speed of 0.2c that would mean 200 years to get there, 400 years to get the data back to Earth. We might have invented some kind of Warp Drive long before a Starshot-class probe arrives at Trappist-1…

    …Or we might not if the laws of physics don’t allow it. But it is not the end of all hope.

    The other possibility is that we might extend our lifespan so much thanks to progress in nanobiotechnologies, genetics and prothestetics that 400 years wouldn’t seem too long a wait. In that case we will need self-healing probes that can auto-repair any damage and last for centuries in the cold of interstellar space with a near-zero energy consumption.

    That is also a technology that will be very useful for Starshot or any slower-than-light probe in the near future when it will be aimed at closer star systems: http://phys.org/news/2016-09-composites-self-heal-temperatures.html

    • Daniel Suggs September 14, 2016, 21:54

      200 years to get there, but, hopefully only 40 years more to get the data back to Earth. :)

  • Ashley Baldwin September 14, 2016, 13:57

    There was an article on arxiv yesterday describing how the SPHERE planet imager operating on one of the 8.2 m unit telescopes of the VLT could be used both alone and in conjunction with soon to be operational high resolution ESPRESSO spectrograph ( at the Coude focus of all the four VLT telescopes ) to both image and characterise the atmosphere of Proxima b in visible light. Via the High dispersion spectroscopy technique ( a combination of high contrast imaging – 1e7 on SPHERE and high resolution spectroscopy via ESPRESSO) described on this very site last year by Ignas Snellen ( part of the publishing team) in “Here come the Giants ” .

    Both instruments would require significant but not extensive modification . In SPHERE’s case either temporarily for a lesser outcome or permanently as a more potent device , a change that will occur ultimately anyway in its successor EPICS on the E-ELT though not till 2029 at the earliest .

    SPHERE is a combination of a spectrograph , polarimeter and high performance adaptive optics system . For Proxima b with an 8.2m telescope and a planetary IWA of 37 mas it would first use its polarimeter to locate Proxima b and then map out its orbit ( and orbital plane ) . The polarimeter essentially blocks out any non polarised starlight in favour of more polarised light reflected off the planet thus helping isolate it. ( assuming a planetary albedo of 0.4- the estimation of which is described in detail in the article )

    Once this is done a specifically placed fibre optic targeted on the planet at the mirror feeds direct to ESPRESSO ( or via the Coude focus though thus would be harder to do) . This spectrograph operates only in the visible spectrum as it is optimised for RV exoplanet discovery which is currently favoured for this approach ( and indeed discovered Proxima b 0n another telescope at la Silla ) . This would usually be limiting as atmospheric spectra reveal more in the near and mid IR , but as telescope resolution has wavelength as the numerator, the longer IR wavelengths are beyond the resolution of the 8.2 m VLT telescope which JUST has another resolution to image Proxima b in visible light operating as near the diffraction limit as the telescope allows ( 2 lambda/ D, ) aperture .

    This still leaves the circa 380 – 750 nm range of ESPRESSO however which includes several key bio signature ( and non bio signature) absorption lines as discussed last week by Rory Barnes. Certainly enough to suggest if Proxima b has an atmosphere , is desiccated and Venus like or even more Earth like. ( O2,O4,H2O ,CH4) .

    The standard resolution of ESPRESSO is an impressive 150000 with the option of an ultra high mode of 220000 which would be essential given the tight nature of Proxima Centauri’s absorption spectrum. The VLT and ESPRESSO will be unique till the ELTs come on line, in being 10m class telescopes with operational access to a high resolution spectrograph – a fact that leads to this potentially exciting option.

    This has huge potential implications and could be done over a manageable number of observations nights ( in favourable conditions ) years ahead of the ELTs and in far greater detail than JWST. The authors emphasise the need for upgrades to SPHERE particularly and the extreme technical difficulties involved . None of which are insurmountable though and the cost of instrumentation upgrades would be millions or a few tens of millions which is tiny in relation to the cost of building a bespoke Space telescope to do this ,which is unlikely for decades. A huge potential fillip for the ESA given the ESO’s pivotal role in the Proxima discovery and aided by Proxima’s proximity to Earth ,Proxima b’s large size in relation to its small star .

    The instrumentation could also be used to characterise several other known nearby M dwarf planets if larger than Proxima b particularly to reduce the contrast between star and planet to SPHERE’s 1e7.

  • Harry R Ray January 26, 2017, 14:41

    Geert Barentsen just tweeted the following: “Kepler spacecraft status: all systems green. Collecting great…TRAPPIST-1 data as part of K2 campaign 12 right now…” This MOST LIKELY MEANS that TRAPPIST-1 is NOT too red for kepler to detect(remember, astronomers gave Kepler only a 10%chance of even being able to OBSERVE the star). Thos also puts pressure on Guillon et al to PUBLISH their Spitzer data AS SOON AS POSSIBLE so they are not SCOOPED by K2!

  • Harry R Ray February 3, 2017, 12:00

    In an UNPRESEDENTED move, K2 has directed the IMMEDIATE RELEASE of ALL CAMPAIGN 12 DATA as soon as it has been downloaded inearly March! This means that Guillen et al will NO LONGER HAVE EXCLUSIVE RIGHTS to TRAPPIST-1 DATA! It will be available to ALL ASTRONOMERS AT ONCE and the race to CONFIRM ADDITIONAL PLANETS(IF they exist) in the system will be ON(unless Guillon et al get their Spitzer data published BEFOREHAND)! Either way, we should know a lot more about this system VERY SOON!

    • Alex Tolley February 3, 2017, 12:45

      Watch out Harry, you are starting to emulate Trump: “UNPRESEDENTED” should be “unprecedented”. ;)