Tuesday, April 18, 2017

HD 219134 Scorecard: 5 planets, 2 transiting

Figure 1. Architecture of the mixed-mass planetary system around HD 219134, a nearby K dwarf, as characterized by Gillon et al. 2017 and Johnson et al. 2016. All five planets are shown at their approximate relative sizes. Planets b and c are observed in transit, so their radii are known. The radii of the other planets are based on those of planets with similar masses and measured radii (e.g., planets d and e are assigned the same radii as Neptune and Saturn, respectively). See Table 1.
Fresh from their discovery of four new Earth-size planets transiting TRAPPIST-1 (a minuscule M dwarf in the Sun’s back yard), Michaël Gillon & colleagues have contributed exciting new data to our developing picture of another nearby exoplanetary system: HD 219134. The host star is a metal-rich K3 dwarf located only 6.55 parsecs away (21 light years) in the direction of Cassiopeia. Gillon & colleagues confirm a previous report by Motalebi & colleagues (blogged here) that the innermost planet (b) is visible in transit, while announcing that the second planet (c) also transits. In addition, they present radial velocity masses for both transiting planets (hereafter tranets), as well as a refined radius for planet b. For both objects, these values are consistent with a purely rocky composition.

After the 2015 announcement by Motalebi & colleagues (hereafter M15), two subsequent studies offered additional data and analyses of the HD 219134 system: Vogt & colleagues (hereafter V15, blogged here) and Johnson & colleagues (hereafter J16, blogged here). Many findings from these studies overlap, but others are mutually inconsistent, including results on the precise number of planets and their approximate masses and periods. Nonetheless, all three studies agreed that HD 219134 hosts two or more low-mass planets on hot and warm orbits, plus one gas giant less massive than Saturn outside the system ice line.

In their new study, Gillon & colleagues confine their attention to the inner system, where they identify a total of four low-mass planets. This result contrasts with the three small planets proposed by M15 and the five preferred by V15. Along with their transit findings, Gillon’s group also report the latest radial velocity data on all four planets from continuing observations with the HARPS-N spectrograph. In passing, they affirm the presence of the giant planet but do not comment on the conflicting results of the earlier studies. Table 1 summarizes their findings on planets b through f, supplemented by the findings of J16 on planet e (which J16 called planet h, even though no one has proposed as many as seven planets – i.e., b through h – for this system).

Table 1. Revised system parameters of HD 219134

Tags: Period = orbital period in days; a = semimajor axis in astronomical units (AU); Radius = radius in Earth units; Mass = mass in Earth units; Teq = equilibrium temperature in Kelvin. Data on planet e are based on those for planet h in Johnson et al. 2016. All other data are based on Gillon et al. 2017.

conflicts resolved?

We’ve now seen four successive studies of HD 219134 conducted by three different scientific teams. (I regard Gillon et al. 2017 as the same team as M15, since the two author groups overlap substantially and both report HARPS-N data.) Our picture of planet b is largely unchanged from the results of M15 and V15, except that we have a firmer understanding of its radius. With the new transit data and improved radial velocity data on planet c, we can also be confident of that planet’s mass, radius, and density, with no change in its estimated period. Even planet d retains virtually the same period reported by M15 and V15, although its mass falls between the values they provided.

Although planet f was not included in the analysis of M15, and a potential planet with a similar periodicity was rejected by J16, the object reported by Gillon & colleagues looks a lot like the planet f proposed by V15: its period is virtually the same, while its mass is a bit lower – a minimum of 7.3 Earth masses (7.3 Mea) instead of 8.9 Mea. Notably, the newly published period of planet f appears to place it just outside a 2:1 mean motion resonance with planet d. Similar period ratios have appeared in several compact low-mass systems discovered by the Kepler Telescope.

Consistent with M15 and J16, Gillon & colleagues implicitly reject V15’s proposed planet g at 94 days. They remain silent on the parameters of the system’s gas giant, to which J16 assign a period of almost 6 years and a semimajor axis of about 3 AU.

mass distribution

The new data suggest that HD 219134 harbors a substantial mass in refractory elements within a space much smaller than the region bounded by Mercury’s orbit in our system (see Table 1). Gillon & colleagues argue that the two innermost planets, b and c, have minimal volatile constituents, with radial velocity data indicating a combined refractory mass of about 9 Mea for this pair. Although we remain ignorant of the radii of the next two planets, f and d, most exoplanets with comparable masses (in the approximate range of 5 to 20 Mea) and comparable thermal environments (cooler than Venus) support substantial envelopes of hydrogen and helium. Existing studies suggest that the bulk mass composition of such planets is generally between 1% and 30% hydrogen/helium (see Lopez & Fortney 2014). Therefore, given their combined radial velocity masses of 23.5 Mea, we can estimate an aggregate mass of 16-22 Mea in refractory elements for this cooler, fatter pair, with hydrogen and other volatiles accounting for the remaining 1-7 Mea. That brings the total refractory mass of the four inner planets into the range of 25-31 Mea.

Altogether, the radial velocity data indicate a total mass of about 33 Mea for the inner system of HD 219134, placing it in roughly the same ballpark as the aggregate mass of the six-planet systems around Kepler-11 (about 30 Mea) and Kepler-20 (about 55 Mea). All these endowments are much richer than the aggregate mass of the four terrestrial planets in our Solar System, which collectively sum to 1.98 Mea. 

stingy spitzer scheduling

The late, lamented Kepler Mission demonstrated the need to collect continuous photometric data over a period of years in order to characterize the inner reaches of any planetary system. The longer the sequence of light curves, the more robust the resulting analysis of potential transits. In light of that history, I was surprised to learn that Gillon’s group was able to observe only two transits each for planets b and c. Their observing sessions with the Spitzer Space Telescope were confined to periods of 6.5 to 7.5 hours centered on the transit window predicted for each planet. As a result, we have no idea whether any of the other planets orbiting HD 219134 can be observed in transit. Definitive findings one way or the other would dramatically improve our understanding of this system’s distinctive architecture and constrain the degree of coplanarity among the orbits of the inner planets

Notably, Spitzer had to monitor TRAPPIST-1 for 20 consecutive days in order to untangle the orbits of its seven known planets, and even that allotment was too brief to obtain a clear picture of the outermost planet, which has a period just shy of 19 days. To illuminate the inner system of HD 219134 would require an observing run in excess of 200 days – evidently an impossibility at present. What we humans need is a whole array of space-based observatories staring year after year at all the most interesting stars in our neighborhood.

architecture and habitability

As I discussed in an earlier post on HD 219134, this system can be characterized as a “rich mixed-mass system,” defined as a planetary system with at least two low-mass planets plus at least one gas giant on a wider orbit. Including our Solar System, about a dozen such configurations are known. They share important similarities. In most of them – including HD 219134 – two or more low-mass planets are observed in transit, while the gas giant tends to be the outermost of the known planets. (The exceptions to the latter generalization are Kepler-87, Kepler-89, and the Solar System, each of which includes a gas giant with a low-mass planet on an exterior orbit.)

In more than half of the known systems –Kepler-48, Kepler-68, Kepler-87, Kepler-167, HD 219134, HD 10180, and our Solar System – we note a gap between the inner system of low-mass planets and the outer gas giant. Except for our own system, we can’t be sure whether any of these apparent gaps is truly empty. Some might be occupied by one or more planets that are misaligned with the others, and thus not visible in transit, or hiding a planet or planets that are too lightweight for radial velocity observations to detect.

In any case, such gaps are significant, because in HD 219134 and several other systems, they correspond to the classical habitable zone. That’s why I’d like to see lots of follow-up observations of HD 219134, accompanied by analyses of the orbital stability of potential Earth-mass planets occupying the space between 0.5 and 1 AU. Three such analyses have already been conducted for the habitable zone of HD 10180, with conflicting results (as blogged here).

beware of outsiders

Equally significant is the mechanism responsible for creating and maintaining these orbital gaps. The difficulty of identifying such a mechanism is underscored by the case of our Solar System, where we still have no widely endorsed scenario to explain why mass is severely depleted between the orbits of Earth and Jupiter, and altogether absent between the orbits of Mars and the Main Belt asteroids. Jupiter is probably involved, but the details are elusive (see Batygin & Laughlin 2015, Raymond et al. 2016). 

The question of orbital gaps and missing mass in this subset of system architectures is subsumed by a more general concern regarding the role of longer-period gas giants in systems with multiple low-mass planets on hot or warm orbits. Several recent studies have addressed the stability of compact multiplanet systems in the presence of an outer giant, whether seen or unseen (Hands & Alexander 2016, Hansen 2017, Becker & Adams 2017, Huang et al. 2017, Jontof-Hutter et al. 2017, Read et al. 2017). The results suggest that “friendly” giants, meaning those permitting the survival of several small, closely spaced planets, need to be cool in more ways than one. Not only must they follow orbits well-separated from the inner planets, in regions where insolation and equilibrium temperatures are lower; they must also be dynamically cool, with minimal eccentricities (deviations from circularity) and inclinations (deviations from coplanarity).

The most delicately balanced configuration – and thus the one most easily upset – contains several planets continually observable in transit. This balance can be maintained only a) if no gas giant is present or b) if the giant is either precisely aligned with the inner ensemble or very widely separated from it. Higher values of eccentricity, inclination, and mass, as well as lower values of semimajor axis, will lead to perturbations of the inner planets. In a perturbed regime, their inclinations might oscillate, such that transits periodically cease for certain planets and then resume after an interval, or become permanently misaligned, such that only one planet, or none at all, is observed in transit. In cases of extreme excitation of inclinations and eccentricities, some or all of the inner planets would be lost altogether.

For example, Becker & Adams found that, among 18 Kepler systems with at least 4 transiting planets each, potential gas giant companions would have to maintain semimajor axes of 10 AU or more to avoid perturbing the inner planets. They also made more specific predictions for several interesting systems featured in previous blog posts: WASP-47, Kepler-11, Kepler-62, Kepler-90, and Kepler-20.

The inner system of WASP-47 consists of a gas giant flanked by two low-mass planets inside a semimajor axis of 0.10 AU; the outer system contains another gas giant at 1.36 AU. Becker & Adams concluded that the orbit of the outer giant must be approximately coplanar with those of the inner tranets, or else they would become mutually misaligned and no longer be observable in transit. For Kepler-11, Kepler-62, and Kepler-20, their conclusions were even more restrictive: none of these systems could harbor an additional planet of 30 Mea or more between 1 and 30 AU without upsetting the clockwork orbits of the inner tranets. (Notably, Jontof-Hutter & colleagues reached a less restrictive conclusion for Kepler-11, ruling out any slightly inclined Jupiter-mass planets within 3 AU.) For Kepler-20, a total of six low-mass planets are known, but only five are observed in transit, given the misalignment of planet g. Becker & Adams propose that an undetected gas giant on a cool orbit might be responsible for this configuration.  

In an analogous study, Read & colleagues investigated the stability of systems containing short-period tranets and non-transiting giants at larger semimajor axes, including two rich mixed-mass systems, Kepler-48 and Kepler-68. For Kepler-48, they concluded that the outer giant must be closely aligned with the inner system, whereas for Kepler-68, available data provided no strong constraints on inclination.

The upshot of this group of studies is that compact inner systems of low-mass planets can typically tolerate cool giants only when the latter are well separated and well aligned. Accordingly, Brad Hansen included an evocative short title for his recent article on the stability problem: Beware of Outsiders. Once we have more data on HD 219134, comparable analyses should be able to constrain the alignment of planet e.

tranets in near space

HD 219134 now warrants a throng of superlatives: it’s the nearest star with a tranet of any description, the nearest Sun-like star with a tranet, the nearest with more than one tranet, the nearest with a terrestrial tranet, and the nearest with more than one terrestrial tranet. In order of increasing distance from the Sun, its closest rivals are Gliese 436 at 10.2 parsecs (one hot tranet more massive than Neptune), Gliese 1132 at 12 parsecs (one hot terrestrial tranet), TRAPPIST 1 at 12.1 parsecs (seven terrestrial tranets), 55 Cancri at 12.3 parsecs (one very hot, massive terrestrial tranet), LHS 1140 at 12.47 parsecs (one temperate terrestrial tranet), Gliese 1214 at 13 parsecs (one hot, puffy tranet about half the mass of Uranus), HD 189733 at 19.2 parsecs (one transiting Hot Jupiter), HD 97658 at 21 parsecs (one hot tranet more massive than Neptune), Gliese 3470 at 29 parsecs (one hot Uranus-mass tranet), HAT-P-11 at 38 parsecs (one hot tranet more massive than Neptune), and Kepler-42 at 39 parsecs (three warm terrestrial tranets in the nearest Kepler system). All the more distant tranets are either Hot Jupiters or low-mass planets discovered by space-based telescopes (Kepler and CoRoT).

Among the 12 transiting systems located within 40 parsecs (130 light years), only one includes a transiting gas giant on a hot orbit (HD 189733), a reminder that the well-known species of Hot Jupiters is actually quite rare. More than half of these nearby transiting systems center on M dwarfs, while only two systems with rocky tranets orbit Sun-like stars (HD 219134 and 55 Cancri). Coincidentally or not, these two are also the only mixed-mass systems in the group.

Given so many superlatives and distinctions, it’s safe to predict that we’ll be hearing more news from HD 219134 for years to come.

Batygin K, Laughlin G. (2015) Jupiter’s decisive role in the inner Solar System’s early evolution. Proceedings of the National Academy of Sciences 112, 4214-4217. Abstract: 2015PNAS..112.4214B
Becker JC, Adams FC. (2017) Effects of unseen additional planetary perturbers on compact extrasolar planetary systems. Monthly Notices of the Royal Astronomical Society 468, 549-563. Abstract: 2017MNRAS.468..549B
Gillon M, Demory B-O, Van Grootel V, Motalebi F, Lovis C, Collier Cameron A, Charbonneau D, Latham D, Molinari E, Pepe FA, Ségransan D, Sasselov D, Udry S, Mayor M, Micela G, Piotto G, Sozzetti A. (2017) Two massive rocky planets transiting a K-dwarf 6.5 parsecs away. Nature Astronomy 1, 56. Abstract: 2017NatAs...1E..56G
Hands TO, Alexander RD. (2016) There might be giants: unseen Jupiter-mass planets as sculptors of tightly packed planetary systems. Monthly Notices of the Royal Astronomical Society 456, 4121-4127.
Hansen B. (2017) Perturbation of compact planetary systems by distant giant planets. Monthly Notices of the Royal Astronomical Society, 1531-1560. Abstract: 2017MNRAS.tmp..186H
Huang CX, Petrovich C, Deibert E. (2017) Dynamically hot Super-Earths from outer giant planet scattering. Astronomical Journal 153, 210.
Johnson MC, Endl M, Cochran WD, Meschiari S, Robertson P, MacQueen PJ, Brugamyer EJ, Caldwell C, Hatzes AP, Ramírez I, Wittenmyer RA. (2016) A 12-year activity cycle for the nearby planet host star HD 219134. Astrophysical Journal 821, 74. Abstract: 2016ApJ...821...74J
Jontof-Hutter D, Weaver BP, Ford EB, Lissauer JJ, Fabrycky DC. (2017) Outer architecture of Kepler-11: Constraints from coplanarity. In press. Abstract: 2017arXiv170308829J
Motalebi F, Udry S, Gillon M, Lovis C, Ségransan D, Buchhave LA, Demory BO, Malavolta L, Dressing CD, Sasselov D, Rice K, Charbonneau D, Collier Cameron A, Latham D, Molinari E, Pepe FA, Affer L, Bonomo AS, Cosentino R, Dumusque X, .Figueira P, Fiorenzano A, Gettel S, Harutunyan A, Haywood RD, Johnson J, Lopez E, Lopez-Morales M, Mayor M, Micela G, Mortier A, Nascimbeni V, Philips D, Piotto G, Pollacco D, Queloz D, Sozzetti A Venderburg A, Watson CA. (2015) The HARPS-N Rocky Planet Search I. HD 219134 b: A transiting rocky planet in a multi-planet system at 6.5 pc from the Sun. Astronomy & Astrophysics 584, A72. Abstract: 2015A&A...584A..72M
Raymond SN, Izidoro A, Bitsch B, Jacobson SA. (2016) Did Jupiter’s core form in the innermost parts of the Sun’s protoplanetary disk? Monthly Notices of the Royal Astronomical Society 458, 2962-2972. Abstract: 2016MNRAS.458.2962R
Read MJ, Wyatt MC, Triaud AH. (2017) Transit probabilities in secularly evolving planetary systems. Monthly Notices of the Royal Astronomical Society. In press. Abstract: 2017arXiv170310046R
Vogt SS, Burt J, Meschiari S, Butler RP, Henry GW, Wang S, Holden B, Gapp C, Hanson R, Arriagada P, Keiser S, Teske J, Laughlin G. (2015) A six-planet system orbiting HD 219134. Astrophysical Journal 814, 12. Abstract: 2015ApJ...814...12V