About forty light years from Earth in the constellation Aquarius is the star designated 2MASS J23062928-0502285, which as of today qualifies as perhaps the most interesting ultracool dwarf we’ve yet found. What we learn in a new paper in Nature is that the star, also known as TRAPPIST-1 after the European Southern Observatory’s TRAPPIST telescope at La Silla, is orbited by three planets that are roughly the size of the Earth. We may have a world of astrobiological interest — and conceivably several — orbiting this tiny, faint star.
Image: Comparison between the Sun and the ultracool dwarf star TRAPPIST-1. Credit: ESO.
If we untangle the TRAPPIST acronym, we find that it refers not to an order of monks (famous for their beers) but to the TRAnsiting Planets and PlanetesImals Small Telescope, a 60 cm robotic instrument that is operated from a control room in Liège, Belgium. TRAPPIST homes in on sixty nearby dwarf stars at infrared wavelengths to search for planets. Michaël Gillon, who led the team from the University of Liège that made the recent discovery, nails its significance:
“Why are we trying to detect Earth-like planets around the smallest and coolest stars in the solar neighbourhood? The reason is simple: systems around these tiny stars are the only places where we can detect life on an Earth-sized exoplanet with our current technology. So if we want to find life elsewhere in the Universe, this is where we should start to look.”
In other words, life may exist on many worlds, and many of us believe that it does. But at our current level of equipment and expertise, a star like TRAPPIST-1 is significant because the star is small and dim enough for the atmospheres of Earth-sized planets to be studied. We learn from the Nature paper that two of the planets here have orbital periods of 1.5 days and 2.4 days respectively. The third is not as well characterized, with its orbit in a range from 4.5 to 73 days despite follow-up work with ESO’s 8-metre Very Large Telescope in Chile.
Noting how close the planets are to the host star, Gillon likens the scale of the system to that of Jupiter and its larger moons. These are worlds twenty to one hundred times closer to their star than the Earth is to the Sun, but the two inner planets receive only twice and four times respectively the stellar radiation that the Earth receives from the Sun. That puts them too close to the star to be in the habitable zone, while the outer planet may possibly lie within it.
But bear in mind that at least the inner two planets are probably tidally locked, with one side perpetually facing the star, the other turned away from it. Hence there may be regions near the terminator that receive daylight but maintain relatively cool temperatures. Given that the third planet may turn out to be entirely within the habitable zone, we have a fascinating test case for upcoming attempts to characterize the atmospheres of each of these Earth-sized worlds.
Image: Artist’s impression of the ultracool dwarf star TRAPPIST-1 from close to one of its planets. Credit: ESO.
This ESO news release quotes Julien de Wit, a co-author of the paper on this work, with regard to where we go next:
“Thanks to several giant telescopes currently under construction, including ESO’s E-ELT and the NASA/ESA/CSA James Webb Space Telescope due to launch for 2018, we will soon be able to study the atmospheric composition of these planets and to explore them first for water, then for traces of biological activity. That’s a giant step in the search for life in the Universe.”
The paper is Gillon et al., “Temperate Earth-sized Planets Transiting a Nearby Ultracool Dwarf Star,” published online in Nature 2 May 2016 (abstract). I’ll note too the continuing interest Centauri Dreams reader Harry R. Ray has shown in TRAPPIST 1, and thank him for bringing it to my attention before these first news reports surfaced.
I JUST found out that the Kepler Space Telescope will be ABLE to observe this system in JUST A FEW MONTHS! There seems to be PLENTY OF ROOM for SEVERAL sub-earths in the CONSERVATIVE habitable zone of TRAPPIST-1. I just cant WAIT to find out. Spitzer just got through a 32 hour FOLLOW-UP run to the ORIGINAL 5 hour observation period in late February-early March resulting from the (in)famous …confirming the binary nature of the transit of TRAPPIST-1d… proposal. Maybe the SPITZER team chanced upon a couple sub-earth candidates,too!
I have also heard that Kepler will be observing a field which includes TRAPPIST-1 for its K2 Campaign 12 (if memory serves). However, since TRAPPIST-1 has a V-mag of 18.8 and a Kp mag that is several tenths lower still, I think the chances of Kepler being able to observe even a 1% decrease in apparent brightness will be very difficult.
NASA seems that doesn’t agree with you:
I was informed about this after I posted my comment. I’m looking forward to seeing the results from Campaign 12!
I hope the spacecraft lasts that long! The K2 mission is a pretty impressive follow-up mission for Kepler, particularly given the circumstances that led to it.
an INDEPENDANT study puts the odds at ONLY TEN PERCENT, more in line with Andrew Le Page’s ORIGINAL ESTIMATE than NASA’s We’ll just have to wait and see!
The REASON fir NASA’s(misguided or NOT: YOU decide)optimism is a NEW SOFTWARE PACKAGE that(APPARENTLY) brings K2’s sensitivity UP TO PAR with the ORIGINAL mission’s(to me, the jury is STILL OUT on that one). HOWEVER: A recent tweet by someone in the KNOW(David Charbonneau) makes me feel a lot more optimistic than I would OTHERWISE be. Some reader with more knowledge on this subject(i.e. who can ask him the RIGHT QUESTIONS) might want to contact him on this!
Note that the Spitzer proposal “Confirming the binary nature of the transit of TRAPPIST-1d” was renamed at some point: the proposal link is now titled “On the eclipsing binary nature of a nearby ultracool dwarf”, implying that by “confirming the binary nature of the transit” they meant that they were confirming that it is an eclipsing binary system not that the transiting object was itself binary.
The inner world should still have some lighting on the darkside due to reflectance from the other planet slightly further out so won’t be completely dark all of the time. It would be interesting to see if we could determine if the planets atmosphere is rotating by looking at the thermal signature but that could be a long shot due to the distance.
Wonderful! Great place to terraform, grow barley, wheat and hop, build robotic breweries and produce a famous new interstellar Trappist beer brand, one beer which will be matured by the natural warm infrared light of a red dwarf! :)
Hmm… Trappist… I think I have an idea for another Fermi Paradox hypothesis ;-)
Another Fermented Paradox hypothesis. ;-)
You mean the aliens are not stupid but actually stupor’ed.
If so we may have to change the Fermi Paradox hypothesis to the Fermi Paralytic hypothesis.
I was thinking more of monk-like silence, not unlike the idea that a civilisation turns inward and pursues meditation or similar practices.
But drunken ETs is something to think of. We might need an extra term in the Drake Equation; f (sober), equivalent to 1-f(drunk) :-)
Also it raises doubts about the size of L :-( . Actually you might have something there, planetary decadence.
It could explain the apparent lack of visiting starships, if ETs are too blithered to navigate out of their own solar system, or to care about it :-(
How distant is Trappist?
40 light years out.
If that third planet exits and it is in the habitable zone I can see 2 advantages to said planet.
The tidal locking, I don’t believe will be as disruptive to world
the size of the earth. Assuming the orbital period of the 3rd theoretical
planet is 4-7 days (which it must be for a world in the HZ) together with
a 800-1200 mb atmosphere and a modest equatorial ocean, you can
expect a fall/summer diurnal pattern, with rain and dry period. As long
as the more temperate regions don’t freeze (or lightly freeze a few cm. it
would certainly be habitable to single celled animals.
The flaring: this ‘SUN’ is barely capable of nuclear fusion.
Does this mean that flaring is very infrequent. Are looking at that
theoretical 13 x Jupiter mass sun ? If so has its behavior modeled?
The Drawback is of course that there can be no photosynthesis as we know
it. On the earth we have machines capable of extracting energy from
a heat – cool cycle, there would be a big evolutionary advantage to such
a ‘trick’ by an organism. More primitive organisms there probably reduce
chemicals for energy.
The paper states that all of these planets are made up MOSTLY of ICE. Assuming this is correct, AND assuming planet “d”‘s orbital period is the MOST LIKELY VALUE of 18+ days, it is out of Abel Mendes HEC habitable zone, BUT, if it is ALSO tidally locked, one SMALL PART(the stellar point would CONTINUALLY receive earth-like insolation, raising the POSSIBILITY of liquid water THERE, and a Europa-like ice sheet EVERYWHERE ELSE! These kind of planets are called “Eyeball Earths” for obvious reasons, and may support life, but only via panspermia, since it is unlikely that life could have ORIGINATED there!
How lethal is the x-ray radiation emitted by brown dwarfs? Isn’t it substantial enough to make life doubtful or terraforming unlikely?
The paper states that there is very little(but, SOME) flaring going on, because this is a very OLD star that has settled down. FOR THIS REASON, the authors believe that this star is ACTUALLY a main sequence M8 RED Dwarf star, and NOT a Brown Dwarf.
I believe even relatively thin atmosphere (like Earth’s) is opaque to X-rays and gamma rays.
TRAPPIST-1 seems to be a hydrogen-burning star rather than a brown dwarf, but only just. From the paper:
There do appear to be a few flares that increase the brightness of the star by a couple of percent. Inferred rate is around 1/60 per hour.
I looked up the mass of the primary and it is 0.08 sol which makes it about an 80 Jup. mass brown dwarf. The planets are listed as approximately Earth size. If you assume this system is a scaled up Jupiter analog, then linear scaling of Jupiter’s moons by a factor of 80 brings them up the the mass of the Earth.
This would mean that an 8 Jup. mass planet would have Mars mass moons and at 40 Jup. mass brown dwarf would have half Earth mass moons, about the minimum considered habitable.
Of course, given that you have more massive bodies orbiting a more massive primary the tidal heating would be much more extensive. This could mean that smaller mass bodies than the ones that are assumed to be the minimum habitable could have decent atmospheres because they are being replenished through volcanism .
On the other side, it is thought brown dwarfs have a pretty intense radiation belts, which could strip the atmospheres off.
Only spectra will tell.
I have read about infrared chlorophyll as far back as 2008.
That article is no longer available, but I found this :
Regarding : “but the two inner planets receive only twice and four times respectively the stellar radiation that the Earth receives from the Sun”
That would make them both not habitable
A while ago I saw this that puts the transition to mini-neptune at ~1.2 Re:
A study of water loss from planets orbiting ultracool dwarfs, including the case of the TRAPPIST-1 system. The implied water loss for the inner two planets is significant but not necessarily fatal depending on the planets’ initial water content and how much water gets cycled between the interior and the planetary surface.
Bolmont et al. (2016) “Water loss from Earth-sized planets in the habitable zones of ultracool dwarfs: Implications for the planets of TRAPPIST-1“
I have a question I have never seen addressed regarding possible habitable planets around red dwarf stars. Isn’t it true that red dwarfs are the remnants of stars at the end of their lifespans? Just like what will happen to the sun in 5 billion years or so when it expands and engulfs the inner planets and Earth and then eventually cools down to a small red dwarf? From what I’ve read, once stars reach this stage, they will last a long time, far longer than the sun’s age now.
In other words, any planets around such a star that either have life or are hospitable to life, are literally on their second go- round. Any life on them would have to have time to re-evolve, assuming the planet escaped destruction when it’s star expanded. But because red dwarf’s last a long time, once they reach that stage, life may have time to begin again, or maybe even for the first time. Is that the reasoning behind searching for life on such a planet?
Scott, no, you’re thinking of white dwarfs — these are the remnants of stars like the Sun. Red dwarfs are long-lived stars that can last for trillions of years. They are not the remnants of stars at the end of their lifetimes.
Apparently there can be photosynthesis even if the dominant part of the received spectrum is in the infrared.
And organisms such as those thriving at the bottom of deep oceans around thermal vents wouldn’t care much about UV flares and lack of bright visible light.
40 ly, that’s quite a challenge for a Starshot-like mission. We’d need to reach 0.9c in order to get data within a century.
If we get to the SGFL this star offers a great system to view because it has a low luminous star that reduces glare issues.
While I applaud Paul for taking his usual measured approach to the claims being made about the important discovery of three Earth-size exoplanets orbiting a nearby ultracool dwarf star, the same can not be said for other space-related media outlets. All too often there was the claim being made that these planets are “Earth-like” or (as in the title of ESO’s own press release) “potentially habitable”. A closer look at the facts indicates otherwise:
The inner two planets are much more likely to be Venus-like since they are slow rotators like Venus with effective stellar fluxes greater than that of Venus. The situation with the third planet is more ambiguous with its poorly constrained orbit but, given what we known now, it is more likely this planet orbits beyond the outer edge of the habitable zone. While the discovery of these Earth-size planets is important, it is a real stretch to characterize them as “Earth-like” or “potentially habitable”.
Agreed. What is different here, though, is the likelihood that we’ll be able to study these planetary atmospheres for possible biomarkers. ESO got carried away in the media releases, but TRAPPIST-1 is going to be a useful star for tuning up our techniques.
Actually the BEST prospect for characterization of an atmosphere of an earth-sized(or, if you accept Jingjing Chen’s recent paper as FACT)a BOUNDARY planet BETWEEN terrestrial and Neptunian)planet BEFORE JWST becomes operational, would not be ANY of the three TRAPPIST-1 planets, but; INSTEAD, the recently discovered(by MEarth) Gliese 1132b, because it is slightly warmer than than TRAPPIST-1b at the cloudtops, and thus, puts out MORE energy for an instrument like SST to detect. In fact, as I am writing this, Spitzer is FINISHING a 100 hour campaign to detect sub-earths in the habitable zone of Gliese 1132!
> What is different here, though, is the likelihood that we’ll be able to study these planetary atmospheres for possible biomarkers.
Oh, I certainly agree that these three worlds offer an excellent opportunity to probe the atmospheres of Earth-size exoplanets and stated as such at the end of my review. As for the usefulness of possible biomarkers, I’ve made my feelings on that subject known before ;-)
Andrew, I like your post on drewexmachina. It gives a good analysis missing in most of the media “hype”.
I have quickly run the numbers for the three planets based on transit duration, stellar mass (0.08) and stellar radius (0.114) data provided in the paper.
I get the following numbers from a visual of the light curves (assuming minimal eccentricity).
Planet – Transit – Orb Velocity – AU – Orb Period
.b. – 33.1 mins – 79.9 km/s – .01113 – 1.52 days
.c. – 38.3 mins – 69.1 km/s – .01487 – 2.34 days
.d. – 59.0 mins – 44.8 km/s – .03533 – 8.58 days (Lower bound)
.d. – 70.6 mins – 37.5 km/s – .05046 – 14.64 days (Upper bound)
*Transit durations given in the paper are for the total transit time including ingress and egress of the planet (therefore are higher than that used to calculate orbital velocity).
These compare to the periods in the paper of 1.51 days for planet b and 2.42 days for planet c. That is an error rate of <1% and 3.3% for planets b and c respectively.
I would be reasonably confident that planet d has an orbital period of between 8.6 and 14.6 days, unless planet d has a highly eccentric orbit.
Orbital period solutions for planet d in the paper of 9.101, 10.401, 12.135, 14.561 all fall within this range.
I think there is still a high probability that planet d (under certain atmospheric conditions) could orbit within the habitable zone.
Rafik: I ASSUME that the authors of the paper did a SIMILAR ANALYSIS, but came up with the 18+ day orbit as the MOST LIKELY ONE! Maybe they KNOW that the orbit is eccentric, but had submitted the paper BEFORE they were absolutely sure. Keep in mind that they submitted TWO SST proposals BEFORE the discovery paper was even accepted: The apparently OVERHYPED “…confirming the binary nature of the transit of TRAPPIST-1d…” 5+ hour one in January, and another FOLLOW-UP one of 30+ hours in February. My guess is that they now KNOW the EXACT orbit of planet d and it is probably an 18+ day orbit with a high eccentricity whose value they are REASOBABLY sure of. A NEW paper on this issue may have been ALREADY submitted.
The paper states they did a global MCMC analysis to come up with the orbital parameters. “It shows that our data favour (relative probability > 10%) a circular orbit and an orbital period between 10.4 and 36.4 days, the most likely period being 18.4 days.”
In Table 3 they normalised the probabilities. “The likelihoods shown are normalized to the most likely solution (circular orbit – P=18.204-
If they did a “visual” estimate of the light curve for planet d they have not mentioned it. One of the problems in doing a visual is that the data does not well define the light curve. That is why I have put an upper and lower bound in my earlier comment.
In my opinion it is possible, but unlikely, that the two data points outside my upper and lower bounds may be correct (8.09 and 18.204 days). However, a 36.4 or 72.8 day orbit for planet d does not concord with the transit data, unless planet d has a highly eccentric orbit.
If the authors have additional data to better define the orbit of planet d, other than the two transits 72.82 days apart, they should make that public. To determine eccentricity they would need more transits and there is no indication they have any further transit data for planet d.
My guess is that this will happen VERY SOON, once they have COMPLETELY REDUCED the NEW Spitzer data! If they do it as an UPDATE to the original paper(which is what I hope they do)it could be JUST A FEW DAYS before we find out. My best guess is that they will wait and submit a NEW PAPER with JUST the Spitzer data. In that case, it will probably be a couple og months before we know for sure. In the meanwhile, HST has JUST finished an observing run in which the MAIN EVENT was observations of a DOUBLE TRANSIT of the “b” and “c” planets to see if their atmospheres are evaporating a la Gliese 436b! This double transit supposedly took place at 9:10 UT on “Star Wars” day. HOW APPROPRIATE!
If you look at the orbital velocities you will notice they are getting higher to get the same solar flux, that increases the kinetic energy of an impact. Double the velocity and the energy quadruples, that means more extinction likely events, low luminous stars can be more dangerous than larger stars in many ways.
Wow I missed the memo on infrared Photosynthesis. Remarkable finding.
Elaborating on the possible 3rd planet:
One might think that such an old red dwarf, might have given rise to life
already, and I am sure all type of new scopes will be trained on if it is
discovered in the HZ of the star. But fret not if it a negative result.
Since stars get hotter as they age, it is possible that the 3rd planet is only
recently becoming warm enough to keep most of it’s surface unfrozen.
Since the inverse square law applies here in terms of where such an orbit
should lie in an orbit that take 6 or so days to complete.
I think there would be additional heating to due to gravitational interactions between this 3rd planet and the known 2nd planet and the primary. That
would replenish the any atmosphere from any infrequent flare events.
Unfortunately odds are that this theoretical 3rd planet is a large terrestrial 1.5-2.5 RE and quite a bit farther from it’s primary or it might have been detected outright, or it might be large moon sized, in which case it would be hard to confirm it’s existence.
I’m becoming skeptical about the chances for life on any world that lacks a geodynamo. We have close-up examples of Venus and Mars, and we don’t see them swimming with life.
> We have close-up examples of Venus and Mars, and we don’t see them swimming with life
The reason Venus is lifeless has nothing to do with its lack of a magnetic field and everything to do with its high stellar flux of about twice that of the Earth. Detailed modeling shows that such a high flux removes the atmospheric cold trap resulting in permanent global water loss which seizes up important geophysical processes like the carbonate-silicate cycle rendering the planet sterile. Earth-size planets can retain substantial atmospheres and conceivably be habitable even without a power magnetic field (although for smaller planets like Mars, such a field is vital).
According to Yang et al. (2014), a planet with an Earthlike atmosphere in Venus’s orbit with Venus’s rotation rate could remain habitable, while a fast-rotating planet would not. Perhaps the runaway greenhouse transition occurred early on in the planet’s history when it had faster rotation.
This probably doesn’t help the TRAPPIST-1 planets, because they are so close to the star that they are in a fast-rotating regime: the transition is around ~10 days according to Kopparapu et al. (2016). This results in the clouds that would otherwise build up on the day side being smeared around the planet, decreasing the albedo and resulting in a runaway greenhouse transition at lower incident flux.
> According to Yang et al. (2014), a planet with an Earthlike atmosphere in Venus’s orbit with Venus’s rotation rate could remain habitable, while a fast-rotating planet would not.
I agree that this would seem to be true for a Sun-like star with an effective surface temperature of 5780 K (where the max stellar flux for a habitable Earth-size slow rotator is ~2.23 times that of the Earth) but, according to Equation 2 in Yang et al., for TRAPPIST-1 with a temperature of only 2700 K the onset of a runaway greenhouse takes place at a much lower stellar flux of ~1.47 times that of the Earth. In all probability, TRAPPIST-1b and c with their Venus-like sizes, Venus-like rotation states and stellar fluxes exceeding that of Venus would be expected to be decidedly non-habitable Venus-like planets as opposed to potentially habitable Earth-like planets as some are claiming.
Actually Venus’s condition is more about runaway greenhouse than stellar flux.
Venus has a very thick atmosphere yet has no internal magnetic field, the mass of a planet has a huge impact on the atmospheric mass. Mars has lost most of its atmosphere because of its low mass 0.1 E’s not because has no global magnetic field.
Still amazes me that we are capable of making discoveries like this system of tiny planets 40 light years away; after all, it was only about a generation ago when we’re did not know of ANY planets around MS stars! Kudos to the team who made this find. I also think the existence of this compact system bodes well for the Pale Red Dot campaign, as we now have evidence that approximately earth-sized world’s can exist around stars the size of Proxima Centauri. Just out of curiosity, what spectral type is the TRAPPIST-1 star?
> Just out of curiosity, what spectral type is the TRAPPIST-1 star?
TRAPPIST-1 is a spectral type M8 (compared to Proxima Centauri which is about 40% more massive and over three times brighter with a spectral type of M5V)
According to the paper, the spectral type is M8.0±0.5.
Here is an update on the search for planets around Proxima Centauri with the MOST telescope.
There is an interesting graph at 3.11 that shows Kepler 42 and the Trappist planets plotted in size versus orbital period.
If Proxima has transiting planets they may be detectable with MOST and the Columbia team are now crunching the numbers for over one months worth of photometry.
Earth’s atmosphereic pressure was LESS THAN HALF PRESENT VALUE 2.7 billion years ago! This could only mean that it was comprised mostly of Hydrogen back then! Some process then must have chemically bound the Hydrogen to the surface whre it was transported UNDERGROUND via plate tectonics, and then REPLACED by MASSIVE AMOUNTS of heavier atoms, making the atmosphere considerably DENSER! Could a SIMILAR process occur around planets like those orbiting TRAPPIST-1, and; if so, what would the consequences be?
From today’s arXiv: de Wit et al. (2016) “A combined transmission spectrum of the Earth-sized exoplanets TRAPPIST-1 b and c“. Cloud-free hydrogen-rich atmospheres appear to be ruled out.
I am TRUELY AMAZED at how CLOSE to the ECLIPTIC TRAPPIST-1 is! It is ALSO SMACK DAB IN THE MIDDLE of the constellation, and NOT an outlier. It appears to be pouring right out of Aquarius’ water jug! How IRONIC it would be if ALL THREE(OR MORE) PLANETS turn out to be “water worlds” as Gillon et al suspect!
I finally found what I was looking for. The star TRAPPIST- 1 does not emit enough white light for even HARPS to do Radial Velocity technique measurements.