Habitable Planet Reality Check: The Seven Planets of TRAPPIST-1

When hearing claims about the potential habitability of newly discovered exoplanets, the old adage “if it seems too good to be true, it probably is not true” frequently comes to my mind. Far too often it seems that fantastic claims made by some about an exoplanet’s potential habitability prove to be overblown or even purposeful hype to create click bait. This is one of the reasons I started the Habitable Planet Reality Check series of articles on this web site – I wanted to present an independent (and hopefully more objective) examination of these sometimes dubious claims.

One example that comes to mind is the announcement made on May 2, 2016 of the discovery of three exoplanets found orbiting the nearby red dwarf star known as TRAPPIST-1. Named after the European Southern Observatory’s (ESO’s) TRAPPIST (TRAnsiting Planets and PlanetesImals Small Telescope) telescope which spotted the transits of these exoplanets, ESO’s own press release claimed “Three Potentially Habitable Worlds Found Around Nearby Ultracool Dwarf Star”. That claim just seemed too good to be true and a closer look revealed that it was dubious at best (see “Habitable Planet Reality Check: TRAPPIST-1”). Two of the exoplanets would be expected to be Venus-like rather than Earth-like and the orbit of the third planet was too ambiguous to make any meaningful claim whatsoever.

This diagram compares the sizes of the newly-discovered planets around TRAPPIST-1 with the Galilean moons of Jupiter and the inner Solar System. Click on image to enlarge. (ESO/O. Furtak)

On February 22, 2017 NASA held a press conference to present the results of new observations made using TRAPPIST and other ground-based telescopes as well as NASA’s Spitzer Space Telescope. The tally of exoplanets found orbiting TRAPPIST-1 now swelled from three to seven roughly Earth-size worlds making this the star with the most planets known outside of our own solar system. But what caught everyone’s attention was the claim that three of these exoplanets were potentially habitable. Given the dubious claim made about TRAPPIST-1 ten months earlier, I was naturally skeptical. So… how does this new claim about the planets orbiting TRAPPIST-1 hold up under closer scrutiny?


The Observations

TRAPPIST-1 (also know by the 2MASS catalog designation J23062928-0502285) is a sub-type of red dwarf known an ultracool dwarf star of spectral type M8 located 39.4±1.3 light years away in the constellation of Aquarius with a V-magnitude of only 18.8. It is estimated that this star has a luminosity of 0.00052 times that of the Sun and a radius of 0.12 times. With a mass estimated to be only 0.08 times that of the Sun and a surface temperature of about 2560 K, TRAPPIST-1 is close to being the smallest possible size for a main sequence star – any smaller and it would be a quickly cooling (and dimming) brown dwarf incapable of fusing hydrogen in its core. Determining the ages of such small stars is difficult because they evolve so slowly over their lifetimes of many trillions of years but it is estimated to be in excess of a half a billion years and is probably much more.

This chart indicates the position of TRAPPIST-1 in the constellation Aquarius. Click on image to enlarge. (ESO/IAU and Sky & Telescope)

TRAPPIST-1 is one of about 60 ultracool dwarf stars in the southern skies whose brightness is being regularly monitored by the 0.6-meter telescope now called TRAPPIST-South located at ESO’s facility in La Silla, Chile. One of the objectives of this project, led by the University of Liège in cooperation with the Geneva Observatory, is to monitor the brightness of ultracool dwarf stars with spectral types later than M5 in order to detect any transits from orbiting planets as well as monitor the stars’ high-frequency variability. Stars of this sort are ideal for ground-based planetary transit searches given their small sizes which results in larger decreases in brightness for the transit of a planet of a given size (thus increasing the detectability of any planets) and the tighter orbits of their planetary systems (which increases the odds of a planet’s orbit being aligned by chance to produce an observable transit). TRAPPIST is a prototype for a more ambitious photometric survey called SPECULOOS (Search for Planets EClipsing ULtra-cOOl Stars) which will monitor the brightness of about 500 of the brightest ultracool dwarfs visible in the southern hemisphere using a quartet of one-meter telescopes at the ESO facility in Paranal, Chile starting in December 2017.

The 0.6-meter TRAPPIST-South robotic telescope at ESO’s La Silla Observatory in Chile was used to spot the transits of the exoplanets orbiting TRAPPIST-1. (E. Jehin/ESO)

On May 2, 2016 the TRAPPIST team led by Michaël Gillon (University of Liège – Belgium) announced the discovery of three Earth-size planets orbiting TRAPPIST-1. The brightness of the star had been monitored at a wavelength of about 0.9 μm about once every 1.2 minutes for a total of 245 hours on 62 nights between September 17 and December 28, 2015 using the TRAPPIST-South telescope. These light curves contained a total of 11 unambiguous transit events each of which decreased the apparent brightness of the parent star by about 1%. Follow up observations were made at visible wavelengths using the two-meter Himalayan Chandra Telescope in India and in the infrared using ESO’s 8.3-meter Very Large Telescope (VLT) in Cerro Paranal as well as the 3.8-meter UKIRT telescope on the summit of Mauna Kea in Hawaii. It was found that nine of the transits could be attributed to a pair of planets, TRAPPIST-1b and c, with orbital periods of 1.51 and 2.42 days.

The interpretation of the other two detected transits was a bit more difficult owing to the noncontinuous nature of the available photometric data. Because the main parameters of the two transits (i.e. duration, depth and impact parameter) were so similar, Gillon et al. favored the explanation that both these transits had been made by a single planet they designated “d”. If one planet were responsible, 11 different orbital periods ranging from as few as 4.5 days to as many as 72.8 days could satisfy the existing observations. Even as the original discovery paper was being prepared for hard-copy publication on May 12, 2016 in Nature, Gillon and his collaborators were already securing additional data about this system.

The situation became even more interesting after the original discovery paper had been submitted and a more detailed analysis of the follow up observations was being performed. Data acquired on the night of December 10/11, 2015 using the HAWK-I infrared imager on ESO’s VLT showed that the observed event not only included a transit of TRAPPIST-1c and another of what had been originally designated “d” blended together, but also evidence of a third previously unobserved transiting planet. This suggested that there were additional transiting planets to be found orbiting this small star and that more observations would be required to sort them out.

This plot shows the varying brightness of the TRAPPIST-1 during an unusual triple transit event observed by the ESO’s VLT on 11 December 2015. Click on image to enlarge. (ESO/M. Gillon et al.)

During February and March 2016, NASA’s Spitzer Space Telescope was used to obtain over 12 hours of photometry at a wavelength of 4.5 μm in the infrared during six possible windows when “d” could have been transiting in an effort to pin down its orbital period. This was followed by an intensive campaign of observations using ground-based instruments. From April 30 to October 11, 2016, the TRAPPIST-South telescope made regular observations of TRAPPIST-1. Combined with earlier, unpublished measurements made between December 29 and 31, 2015, an additional 469 hours of new photometric data were obtained. The newly commissioned 0.6-meter TRAPPIST-North telescope at Oukaïmeden Observatory in Morocco was also used to make measurements of TRAPPIST-1. A collaboration between the University of Liège and Cadi Ayyad University of Marrakesh, TRAPPIST-North secured an additional 202 hours of photometric data between June 1 and October 12.

The TRAPPIST team were not the only ones to make new observations of TRAPPIST-1 during 2016. A team led by Steve Howell (NASA Ames Research Center) used the DSSI (Differential Speckle Survey Instrument) on the 8.1-meter Gemini-South Telescope located on the mountain top in Cerro Pachón, Chile to obtained diffraction limited images of TRAPPIST-1 and its surroundings in a search for low mass companions which had escaped earlier searches and whose presence might alter the interpretation of the photometric transit data. Images acquired on June 22 and 27, 2016 at wavelengths of 692 and 883 nm with a resolution of 27 milliarc seconds showed no hint of additional objects 0.32 to 14.5 AU from TRAPPIST-1. Howell et al. were able to show that TRAPPIST-1 has no small stellar or brown dwarf companions which may be causing their own transit events or otherwise complicating the interpretation of the photometric data.

Meanwhile, the TRAPPIST team and their collaborators continued to gather additional photometric data from a range of telescopes across the globe in the hopes of figuring out the TRAPPIST-1 system. The 3.8-meter UKIRT telescope in Hawaii was used again to obtain an additional 25 hours of J-band photometry on six nights between June 24 and August 1, 2016. The 4.2-meter William Herschel Telescope in La Palma on the Canary Islands took 26 hours of photometric data on three consecutive nights from August 23 to 25. The robotic 2-meter Liverpool Telescope at the Roque de los Muchachos Observatory also in La Palma was used to secure a total of 50 hours of photometric data between June and October. Finally the 1-meter telescope at the South African Astronomical Observatory (SAAO) in Sutherland, South Africa was used for another 11 hours of observation on the nights of June 18/19, 21/22 and July 2/3.

NASA’s Spitzer Space Telescope was used to provide precision photometry of TRAPPIST-1 in an effort to better characterize this system. (NASA)

As all these photometric data were being analyzed, it was realized that the two transits which had been identified as belonging to “d” in the original discovery paper by Gillon et al. could not have been made by a single object since no possible orbital solution could fit all of the original and new photometric data. There were at least four and very likely more exoplanets orbiting TRAPPIST-1 but the non-continuous nature of the existing multi-site photometric data set made it difficult to figure out what exactly was going on. What was needed was a stretch of continuous photometric observations long enough to observe at least a couple of transits from the longest period exoplanet that might be there. The TRAPPIST team was able to secure the use of NASA’s Spitzer Space Telescope once again to provide these vital photometric data with a virtually continuous observation run from September 19 to October 10. With a total of 518 hours of Spitzer photometric data available, it was now possible for Gillion et al. to sort out the TRAPPIST-1 system.

This diagram shows the changing brightness of TRAPPIST-1 over a period of 20 days in September and October 2016 as measured by NASA’s Spitzer Space Telescope and various ground instruments. The dips in brightness caused by transiting exoplanets are clearly seen. Click on image to enlarge. (ESO/M. Gillon et al.)

The Planets Found

With the 1,333 hours of new photometric data now available, the total number of transit events observed went from just 11 in the original discovery paper in 2016 to 92 for the latest paper by Gillon et al.. All but one of those transits have been attributed to a total six exoplanets orbiting TRAPPIST-1. TRAPPIST-1b and c were confirmed and four more planets with orbital periods ranging from 4 to 12 days were also found designated TRAPPIST-1d through g  (with the “d” designation from the 2016 discovery paper being reassigned after it was found that the original observations were in fact two different exoplanets). The sole remaining “orphan” transit was observed by Spitzer but has not been found in the ground-based data. As a result, the orbital parameters are not known with any certainty but it is estimated that its orbital period might be around 20 days.

This plot of Spitzer photometric data shows how the brightness of the star TRAPPIST-1 changes during transits of each of its planets. Click on image to enlarge. (ESO/Gillon et al.)

In order to derive the orbital and physical parameters of these exoplanets, Gillion et al. used only the Spitzer photometry since it was of far superior quality compared to the ground-based photometry. The orbit parameters for the seven planets of TRAPPIST-1 are summarized below in Table 1. Included in this table are the effective stellar flux values (i.e. the amount of energy each exoplanet receives from its parent star) with the Earth being defined as 1. All values have been taken from the new paper by Gillon et al..


Table 1: Orbit Parameters
Planet P (days) a (10-3 AU) Seff (Earth=1)
b 1.511 11.1 4.3
c 2.422 15.2 2.3
d 4.050 21.4 1.14
e 6.100 28.2 0.66
f 9.207 37 0.38
g 12.353 45 0.26
h ~20? ~63? ~0.1?


From the depths of the transit events measured most precisely in the Spitzer photometry and knowing the size of the star TRAPPIST-1, the radii of the planets in the system can be determined. All of them appear to be approximately Earth size or slightly smaller providing additional evidence that such worlds are indeed fairly common (see “Prevalence of Earth-Size Planets Around Sun-Like Stars”). The radii of the seven exoplanets orbiting TRAPPIST-1 and their associated measurement uncertainties are listed below in Table 2.


Table 2: Physical Parameters
Planet RP (Earth=1) MP (Earth=1) ρP (Earth=1)
b 1.086±0.035 0.85±0.72 0.66±0.56
c 1.056±0.035 1.38±0.61 1.17±0.53
d 0.772±0.030 0.41±0.27 0.89±0.60
e 0.918±0.039 0.62±0.58 0.80±0.76
f 1.045±0.038 0.68±0.18 0.60±0.17
g 1.127±0.040 1.34±0.88 0.94±0.63
h 0.755±0.034 ? ?


One thing that is immediately evident about this system is the close spacing of the planets’ orbits. In fact, they are quite close to being small integer ratios of each other: approximately 2:3:5:8:12:16 for the inner six planets whose orbits are best determined. This strongly suggests that these planets originally formed farther from TRAPPIST-1 and migrated inwards jostling each other into the tight arrangement we see today.

This diagram shows a comparison of the properties of the newly discovered planets of TRAPPIST-1 with the inner planets of our solar system. Click on image to enlarge. (NASA)

Because of the packed nature of this planetary system with its orbits near resonance, it would be expected that they would strongly interact with each other gravitationally producing variations in the transit timings. Gillon et al. performed an analysis of these transit timing variations (TTV) derived from all of the photometry data and found them to be on the order of tens of seconds to more than a half an hour – more than sufficient to estimate the masses of the inner six planets and constrain their orbital eccentricities. The masses found by Gillon et al. and the resulting bulk densities of these planets calculated using the radii derived from the transits are listed above in Table 2. Interestingly, the mass ratio of TRAPPIST-1 to its individual planets is about the same order of magnitude as the mass ratio of Jupiter to its individual Galilean moons suggesting that they may have formed by similar processes.

The masses derived from this initial TTV analysis are still quite uncertain with the resulting densities values even more so. All that can be stated with any certainty is that all these exoplanets are approximately Earth-mass objects and that most have densities that do not exclude the possibility that they have a primarily rocky composition much like the Earth and the other terrestrial planets of our solar system. The one exception is TRAPPIST-1f whose mass derived from TTV analysis has a relative error of “only” ~26%. The resulting density that is 0.60±0.17 of Earth’s strongly favors a volatile-rich composition.

This plot of planet mass versus radius shows the the inner six exoplanets of the TRAPPIST-1 system in addition to other selected planets with the error in mass and radius indicated. Theoretical curves for planets of various compositions are shown with an Earth-like composition in black. Click on image to enlarge. (M. Gillon et al.)

Two different methods were used by Gillon et al. to check the stability of the TRAPPIST-1 system. The results suggest that it might be unstable over relatively short time scales. However, the uncertainties in the orbital eccentricity and especially the masses are still too large to draw any strong conclusions. In addition, stabilizing tidal effects which should be important in a system like this have yet to be included in the models. Predictions about the stability of this packed system will have to wait for better information. Since only upper limits for the orbital eccentricity have been derived, it is possible that tidal heating will be important sources of internal heat especially for the inner planets (since tidal heating scales inversely with the 15/2 power of orbital radius, all else being equal). However, the eccentricity does seem to be low enough that it is likely that all these planets will be synchronous rotators. Only at higher eccentricities do planets tend to settle into a super-synchronous rotation state like Mercury which rotates three times for every two orbits it makes around the Sun.

There have been concerns about the ability of planets orbiting small red dwarfs in general and especially TRAPPIST-1 in particular to hold onto their water and other volatiles due to excessive luminosity during the star’s early life and subsequent excessive chromospheric activity such as intense flares. A team led by Peter Wheatley (University of Warwick – UK) used ESA’s XMM-Newton X-ray space observatory to observe TRAPPIST-1. They found that the X-ray luminosity of TRAPPIST-1 is similar to the Sun’s despite the former’s diminutive size and low surface temperature. Wheatley et al. note that such a flux of X-rays and extreme ultraviolet radiation on the close-orbiting planets of the TRAPPIST-1 system would significantly alter their primary and secondary atmospheres.

An artist’s impression of ESA’s XMM-Newton X-ray observatory which was used to measure the X-ray and extreme ultraviolet flux of TRAPPIST-1. (ESA)

In order to address this issue, a group led by Emeline Bolmont (University of Namur – Belgium) used a 1D radiation-hydrodynamic simulation to model the details of water loss on planets orbiting red dwarfs. They specifically examined the case for TRAPPIST-1b and c and found that they may have lost the equivalent of 15 Earth-oceans worth of water. Fortunately these effects decrease quickly with distance with Bolmont et al. finding that less than an Earth-ocean’s worth of water can be lost in parts of the habitable zone and beyond, depending on the mass of the planet and other details assumed. The fact that it appears these exoplanets migrated from farther out also means the timing of that migration will be a factor (in addition to enhancing the planets’ initial volatile inventory while beyond the “snow line”). The low density of TRAPPIST-1f, which is a factor of two or three more distant than the inner two planets, suggests that maybe it has retained much of its water and holds out hope that its neighbors may have done the same as well.

NASA’s Hubble Space Telescope has already been used to look for extended hydrogen-rich atmospheres for TRAPPIST-1b and c. (NASA)

Observations made by a team led by Julien de Wit (MIT) using NASA’s Hubble Space Telescope have already helped to constrain the compositions of the atmospheres of the inner two exoplanets orbiting TRAPPIST-1 – the same ones which would be most affected by water loss. The Wide Field Camera-3 on Hubble was used by de Wit et al. to take a series of spectra of TRAPPIST-1 on May 4, 2016 when the two inner planets were simultaneously transiting the star. While Hubble is not capable of detecting atmospheric constituents like water vapor, carbon dioxide and methane on these planets, it should be capable of detecting hydrogen in a hot, extended atmosphere – the sort of atmosphere expected if these exoplanets were volatile-rich mini-Neptunes. De Wit et al. failed to detect any hydrogen effectively ruling out the possibility that TRAPPIST-1b and c have cloud-free atmospheres dominated by hydrogen. Still possible are other scenarios such as a cloud-free water vapor-rich atmosphere or a Venus-like atmosphere but larger telescopes like NASA’s James Web Space Telescope (JWST) will be required to test those possibilities. The other planets in the system are currently being observed by Hubble to similarly constrain the compositions of any atmospheres present.


Potential Habitability

A thorough assessment of the habitability of any extrasolar planet would require a lot of detailed data on the properties of that planet, its atmosphere, its spin state and so on. Unfortunately, at this very early stage, the only information typically available to scientists about extrasolar planets is basic orbit parameters, a rough measure of its size and/or mass and some important properties of its sun. Combined with theoretical extrapolations of the factors that keep the Earth habitable (not to mention why our neighbors are not), the best we can hope to do at this time is to compare the known properties of extrasolar planets to our current understanding of planetary habitability to determine if an extrasolar planet is “potentially habitable”. And by “habitable”, I mean in an Earth-like sense where the surface conditions allow for the existence of liquid water on the planet’s surface – one of the presumed prerequisites for the development of life as we know it. While there may be other worlds that might possess environments that could support life, these would not be Earth-like habitable worlds of the sort being considered here.

One of the important criteria which we can use to determine if a planet is potentially habitable is the amount of energy it receives from its sun known as the effective stellar flux or Seff. According to the work by Kopparapu et al. (2013, 2014) on the limits of the habitable zone (HZ) based on detailed climate and geophysical modeling, the outer limit of the HZ is conservatively defined as corresponding to the maximum greenhouse limit of a CO2-rich atmosphere where the addition of any more of this greenhouse gas would not increase a planet’s surface temperature any further. For a star like TRAPPIST-1 with a surface temperature of 2559±50 K, this conservative outer limit for the HZ has an Seff of 0.22 corresponding to a orbital semimajor axis of 0.048 AU or an orbital period of 13.8 days. This Seff value for the outer limit of the HZ is lower than the 0.36 for an Earth-size planet orbiting a more Sun-like star because ultracool dwarf stars emit so much of their energy in the infrared part of the spectrum where atmospheric absorption is important.

The inner limit of the HZ is conservatively defined by Kopparapu et al. (2013, 2014) by the runaway greenhouse limit where a planet’s temperature would soar even with no CO2 present and lose all of its water in a geologically brief time in the process. For an Earth-size planet orbiting TRAPPIST-1, this happens at an Seff value of 0.91 which corresponds to a distance of 0.024 AU or an orbital period of 4.8 days. Once again, this Seff value for the inner edge of the HZ is lower than the 1.11 for a Sun-like star because TRAPPIST-1 radiates so much of its energy in the infrared.

As mentioned earlier, it is likely that these planets are synchronous rotators with the same side perpetually facing their sun. Detailed climate modeling over the last two decades now shows that synchronous rotation is probably not the impediment to habitability as it was once thought. In fact, it has been shown that slow or synchronous rotation can actually result in an increase of the Seff for the inner edge of the HZ. According to the recent work by Yang et al., the inner edge of the HZ for a synchronous rotator orbiting a star like TRAPPIST-1 would have an Seff of 1.44 corresponding to an orbital distance of 0.019 AU or an orbital period of 3.4 days.

This diagram compares the orbits of the newly-discovered planets around TRAPPIST-1 with the Galilean moons of Jupiter and the inner Solar System. Click on image to enlarge. (ESO/O. Furtak)

Looking at the exoplanets now known in this system, TRAPPIST-1b and c are highly unlikely to be habitable. Their Seff values of 4.3 and 2.3 are higher than the 1.9 value for Venus, which is most definitely not a habitable planet. With their Venus-like sizes, Venus-like rotation states, lack of a hydrogen atmosphere and Seff values in excess of Venus’, these are most likely to be non-habitable, Venus-like worlds contrary to the claims made by some when the discovery of these planets was announced in May 2016.

The situation with TRAPPIST-1d is a bit more promising. With a radius of 0.77 RE, it is smaller than the Earth and unlikely to be a volatile-rich mini-Neptune with a deep atmosphere dominated by hydrogen with little prospect of being habitable in the conventional sense. Future observations by Hubble could provide observational evidence to support this view. With an Seff of 1.14, TRAPPIST-1d would seem to be comfortably inside the HZ for a slow or synchronous rotator. However, as Gillon et al. mention in their new paper (where they have adopted the more conservative definition of the HZ), more recent work by Kopparapu et al. (2016) has shown that Coriolis effects for synchronous rotators with short orbital periods will alter the global circulation pattern in a way which affects cloud formation on the dayside – clouds which help to reflect away much of the energy the planet receives from its sun. With an orbital period (and presumably a period of rotation) of just four days, TRAPPIST-1d would likely be unable to maintain sufficient cloud cover on its day side to keep from experiencing a runaway greenhouse effect. While it is certainly worthy of detailed study, it would seem that the chances that TRAPPIST-1d is potentially habitable are not very promising and Gillon et al. do not categorize this find as “potentially habitable”.

This diagram shows the relative sizes of the orbits of the seven planets orbiting TRAPPIST-1. The shaded area shows the extent of the habitable zone (HZ) with alternative boundaries indicated by dashed lines. Click on image to enlarge. (ESO/M. Gillon et al.)

The situation with TRAPPIST-1e is substantially better and it has been identified by Gillon et al. as being potentially habitable. With an Seff of 0.66, this exoplanet is comfortably inside the conservatively defined HZ of TRAPPIST-1. With a radius of 0.91 RE, it is only slightly smaller than Earth and is not expected to be a mini-Neptune. Observations being made by Hubble could help to eliminate this possibility as will better density estimates to help constrain this exoplanet’s bulk composition. If it were not for the unresolved issues associated with orbiting so close to an ultracool dwarf star, TRAPPIST-1e could be considered one of the best candidates currently known for being a potentially habitable exoplanet. Undoubtedly, detailed climate modelling of this world will help to determine the range of water and other volatile content values which could yield a habitable world much as is being done for our Earth-size neighbor, Proxima Centauri b (see “Habitable Planet Reality Check: Proxima Centauri b”).

The next planet out, TRAPPIST-1f, was also identified as being potentially habitable by Gillon et al.. Its Seff value of 0.38 is comparable to that of Mars but, since so much of the energy emitted by TRAPPIST-1 is in the infrared, it is still comfortably inside the conservatively defined HZ for this star. Despite its Earth-like size of 1.05 RE, the low density of TRAPPIST-1f is a cause of some concern since it implies that it is a volatile-rich world. It could be that this is a mini-Neptune with a deep hydrogen-rich atmosphere overlaying layers of high temperature/pressure phases of ice rendering it non-habitable. It might also be more of an ocean world with an atmosphere a few times denser than the Earth’s and rich in CO2 capping a deep ocean of liquid water. Hubble observation might help to eliminate the former possibility although observations by JWST and other future instruments will be required to explore the latter.

An artist’s impression of an Earth-like exoplanet orbiting TRAPPIST-1. (ESO/M. Kornmesser/N. Risinger)

But before too much is read into the apparent low density of this world, it should be remembered that TTV-derived masses are notorious for changing by rather large amounts as new data become available. It may turn out that the uncertainties in the mass and density have been underestimated and this is actually a denser rocky world like the Earth. But even if the low density of TRAPPIST-1f is confirmed and it is found to be unlikely to be potentially habitable, it nevertheless strongly suggests that small planets orbiting small red dwarfs can hang onto substantial amounts of their water and other volatiles contrary to some of the predictions that have been made. This would markedly improve the habitability prospects of many red dwarf planets. For now, TRAPPIST-1f is a reasonable candidate for being potentially habitable – definitely better than TRAPPIST-1d but maybe not as good as e.

The third of their discoveries identified by Gillon et al. as being potentially habitable is TRAPPIST-1g. With a radius of 1.12 RE, it is unlikely to be a mini-Neptune but its currently ill-defined density as well as the fact that the smaller and closer TRAPPIST-1f may be volatile-rich makes it impossible to exclude the possibility. Future observations will be required to determine the nature of this exoplanet. With a Seff of 0.26, TRAPPIST-1g is towards the outer edge but still comfortably inside the HZ for such a cool star. Once again, the claim made by Gillon et al. that this is a potentially habitable exoplanet it a reasonable one given what we currently know about this world. The final planet in this system, TRAPPIST-1h, still has an ill-defined orbit but it seems quite likely that it is well outside of the HZ.

An artist’s impression of a cold world in the outer reaches of the TRAPPIST-1 system. (ESO/M. Kornmesser)


Summary and the Future

While the claim made during NASA’s February 22, 2017 news conference that three of the newly discovered exoplanets found orbiting the ultracool dwarf star, TRAPPIST-1, are potentially habitable seemed too good to be true, that claim does indeed appear to be reasonable given what we now know about them. TRAPPIST-1e, f and g are all approximately Earth-size exoplanets which orbit comfortably inside of the more conservative definition of the habitable zone (HZ).

While the comparatively small radii of these exoplanets would suggest that they should be rocky terrestrial planets like the Earth with good prospects for being habitable, an initial mass value for TRAPPIST-1f derived from an analysis of transit timing variations (TTV) yields a low density suggesting a volatile-rich bulk composition instead. Whether this means TRAPPIST-1f is a mini-Neptune with a deep hot atmosphere with little prospect of being habitable or a potentially habitable ocean world with a thinner, temperate atmosphere covering a deep ocean of liquid water or something else entirely remains to be seen. IF future observations confirm this exoplanet’s low density and volatile-rich bulk composition, it suggests that the worst-case predictions of how red dwarfs can strip a world of its water and other volatiles are not always true. Instead, it means that TRAPPIST-1e and g may have held onto a substantial portion of their original volatile inventories and that the number of other potentially habitable red dwarf planets may have done the same as well (see “Top Five Known Potentially Habitable Planets”).

With so many transiting exoplanets orbiting a nearby star, TRAPPIST-1 is a perfect target for future observation – observations that promise to tell us much about the properties of not only about these exoplanets in particular, but the exoplanets of all red dwarfs. Seven exoplanets of approximately the same size and mass orbiting the same star but at differing distances is a natural testing ground for theories on exoplanet formation, evolution and their potential habitability. And we won’t have long to wait for those new observations.

An artist’s impression of a size comparison of the planets of the TRAPPIST-1 system. (NASA/R. Hurt/T. Pyle)

The Hubble Space Telescope is reported to be already making spectral observations of these transiting exoplanets to help constrain the extent of any hydrogen-dominated atmosphere these worlds may have. We will have to wait for the launch of the James Webb Space Telescope to study the composition of these exoplanets atmosphere and see if more interesting gases are present but TRAPPIST-1 has already been identified as being a high priority target early in the mission.

Better measurements of the masses of these exoplanets are sorely needed to help pin down their densities more accurately and constrain their bulk compositions. The reflex motion of TRAPPIST-1 as its planets revolve around it should be as large as a few meters per second. Unfortunately, the star is too dim at visible wavelengths for the current generation of precision radial velocity instruments to make these measurements with the required accuracy. But future instruments operating in the infrared where TRAPPIST-1 is brighter may be able to do so.

For the time being, that leaves masses derived from TTV analysis. Fortunately NASA’s Kepler spacecraft observed a star field that includes TRAPPIST-1 as part of Campaign 12 of its extended K2 mission. With a virtually continuous photometric data set running from December 15, 2016 to March 4, 2017, the current total of 92 observed transit events should swell to about 220 or more. A TTV analysis of this greatly expanded data set should allow a much better determination of these exoplanet’s masses and densities. The Kepler data should also help to pin down the orbital properties of TRAPPIST-1h with a few more transits to add to the single existing event currently in hand. It is expected that more observations by the TRAPPIST telescopes and other ground-based instruments will be made as well.

There is the real possibility that even more exoplanets will be spotted, although given the packed nature of the planets in this system, it is unlikely that any more can be squeezed into the HZ than those already identified. Guaranteed that we will be seeing many more results published about this extraordinary planetary system over the course of 2017 and especially beyond.


Follow Drew Ex Machina on Facebook.


Related Video

Here is a video of the full NASA press conference on February 22, 2017 where the discovery of seven exoplanets found orbiting TRAPPIST-1 was announced.



Related Reading

“The (Potentially) Habitable Worlds of TRAPPIST-1”, Centauri Dreams, February 27, 2017 [Post]

“Habitable Planet Reality Check: TRAPPIST-1”, Drew Ex Machina, May 3, 2016 [Post]

“Top Five Known Potentially Habitable Planets”, Drew Ex Machina, December 13, 2016 [Post]


General References

E. Bolmont et al., ”Water loss from terrestrial planets orbiting ultracool dwarfs: implications for the planets of TRAPPIST-1”, Monthly Notices of the Royal Astronomical Society, Vol. 464, No. 3, pp. 3728-3741, January 2017

Julien de Wit et al., ”A combined transmission spectrum of the Earth-sized exoplanets TRAPPIST-1 b and c”, Nature, Vol. 537, No. 7618, pp. 69-72, September 2016

Michaël Gillon et al., “Temperate Earth-sized Planets Transiting a Nearby Ultracool Dwarf Star”, Nature, Vol. 533, pp. 221-224, May 12, 2016

Michaël Gillon et al., “Seven temperate terrestrial planets around the nearby ultracool dwarf star TRAPPIST-1”, Nature, Vol 542., pp. 456-460, February 23, 2017 [Preprint]

Steve B. Howell et al., “Speckle Imaging Excludes Low-mass Companions Orbiting the Exoplanet Host Star TRAPPIST-1”, The Astrophysical Journal Letters, Vol. 829, No. 1, Article id. L2, September 2016

R.K. Kopparapu et al., “Habitable zones around main-sequence stars: new estimates”, The Astrophysical Journal, Vol. 765, No. 2, Article ID. 131, March 10, 2013

Ravi Kumar Kopparapu et al., “Habitable zones around main-sequence stars: dependence on planetary mass”, The Astrophysical Journal Letters, Vol. 787, No. 2, Article ID. L29, June 1, 2014

Ravi Kumar Kopparapu et al., “The Inner Edge of the Habitable Zone for Synchronously Rotating Planets around Low-mass Stars Using General Circulation Models”, The Astrophysical Journal, Vol. 819, No. 1, Article ID. 84, March 2016

Peter J. Wheatley et al., “Strong XUV irradiation of the Earth-sized exoplanets orbiting the ultracool dwarf TRAPPIST-1”, Monthly Notices of the Royal Astronomical Society: Letters, Vol. 465, No. 1, pp. L74-L78, February 2017

Jun Yang et al., “Strong Dependence of the Inner Edge of the Habitable Zone on Planetary Rotation Rate”, The Astrophysical Journal Letters, Vol. 787, No. 1, Article id. L2, May 2014