Rho CrB b and HD 156846 b - Inclination constraints

From the EPE

ρ CrB b:

Quote :

11 Jan 2011: Reffert et al. 2011 give a mass range derived from astrometry 100 - 199.6 MJup at the 3 σ level .

HD 156846 b:

Quote :

11 Jan 2011: Reffert et al. 2011 give a mass range derived from astrometry 10.45 - 660.9 MJup at the 3 σ level .
The errors on inclination are at the 3 σ level (Reffert et al. 2011)

Both are now in the "Unconfirmed/Disproven" section.

Could be both late M stars ??

Seems to be a bunch of updates on EPE related to Reffert et al. 2011, which is available if you have an Astronomy and Astrophysics subscription (I don't) from their "forthcoming" section.

Planets affected:

6 Lyncis b: 2.2 – 128.7 M_J
γ¹ Leonis b: 9.2 – 130.9 M_J
γ Cephei Ab: 5.0 – 26.9 M_J
HD 106252 b: 6.9 – 68.9 M_J
HD 114783 b: 1.5 – 32.5 M_J
HD 142022 Ab: 4.6 – 102.2 M_J
HD 168443 c: 18 – 39.7 M_J
HD 181720 b: 6.1 – 217.9 M_J
HD 87883 b: 2.1 – 21.3 M_J
ι Draconis b: 8.82 – 19.8 M_J

And in the unconfirmed list:

HD 131664 b: 36.3 – 139.7 M_J
HD 156846 b: 10.45 - 660.9 M_J
HD 190228 b: 5.93 – 147.2 M_J
HD 43848 b: 24.6 – 130.7 M_J
ρ Coronae Borealis b: 100 – 199.6 M_J

(Errors at the 3σ level)

I don't think all these "low-mass" planet may have actually so much mass.

Well the paper is using re-reduced Hipparcos data. In 2001 the previous version of the Hipparcos catalogue was used to conclude that the majority of planet candidates analysed were in nearly face-on orbits, though the significance of that result was later questioned.

For starters, the best-fit inclination of γ Cephei Ab (5.7°) would suggest that the planet is severely misaligned with the binary orbit, despite stability analyses suggesting that the system must be close to coplanar...

So wouldn't be early to conclude all these exoplanets are stars? Especially after astrometry hasn't proved so conclusive to ascertain the presence of extrasolar planets.

It doesn't help that Hipparcos wasn't really designed to do this kind of thing - primary mission was parallax and proper motion studies which do not require observations on such a frequent timescale as tracing out planetary orbits.

Mass constraints on substellar companion candidates from the re-reduced Hipparcos intermediate astrometric data: Nine confirmed planets and two confirmed brown dwarfs
http://arxiv.org/abs/1101.2227

Quote :

We search for the astrometric signatures of planets and brown dwarfs known from radial velocity surveys in the improved Hipparcos intermediate astrometric data provided by van Leeuwen (2007a). Our aim is to put more significant constraints on the inclination and the longitude of the ascending node than was possible before, resulting in unambiguous companion masses. We fitted the astrometric orbits of 310 substellar companions around 258 stars to the Hipparcos intermediate astrometric data. Even though the astrometric signatures of the companions cannot be detected in most cases, the Hipparcos data still provide lower limits on the inclination for all but 67 of the investigated companions, which translates into upper limits on the masses of the unseen companions. For nine companions the derived upper mass limit lies in the planetary and for 75 companions in the brown dwarf mass regime, proving the substellar nature of those objects. Two of those objects have minimum masses also in the brown dwarf regime and are thus proven to be brown dwarfs. The confirmed planets are the ones around Pollux (beta Gem b), epsilon Eri b, epsilon Ret b, mu Ara b, upsilon And c and d, 47 UMa b, HD 10647 b and HD 147513 b. The confirmed brown dwarfs are HD 137510 b and HD 168443 c. In 20 cases, the astrometric signature of the substellar companion was detected in the Hipparcos data. Of these 20 companions, three are confirmed as planets or lightweight brown dwarfs (HD 87833 b, iota Dra b, and gamma Cep b), two as brown dwarfs (HD 106252 b and HD 168443 b), and four are low-mass stars (BD -04 782 b, HD 112758 b, rho CrB b, and HD169822 b). Of the others, many are either brown dwarfs or very low mass stars. For epsilon Eri, we derive a solution which is very similar to the one obtained using Hubble Space Telescope data.

look like this are not very good news for exoplanets, i was read the paper and seems that most of the objects, can be brown dwarf or low mass stars.

so where is the room for exoplanets in it's all? i mean if there is many brown dwarf star and low mass stars like show the data.

1 statistically isn't good for exoplanets because this can prove that it's more low mass stars and brown Dwarf than planets.

2 if most of stars have a brown dwarf or low mass star,most of this systems,should have instability for planetary orbits especially for small planets,except if the planets are in a far away orbit of the binary system.

I don't know, if i'm wrong in my thoughts, but this new datas seems not good for many exoplanets

All radial velocity planets have only a lower limit to their masses. Their true mass is their RV-derived minimum mass divided by the sine of the inclination of the orbit. The inclination is unknown until astrometry resolves the stellar motion. Until then, they are only candidates. This is what we're seeing here, the confirmation or rejection of planets with astrometric constraints.

The more massive the companion, the greater the astrometric signal is, and the easier it is to detect and reject as a brown dwarf.

Look a little more closely at the paper. Only a few planet candidates are found to have true masses outside the planetary range. Only a handful out of the 310 stars they observed. For all others, they did not detect the astrometric motion, but are able to place constraints (i.e. "the planet must be less massive than m or we would have detected it"). In the case of nine, the upper limits are in the planetary range, confirming them as planets.

For the other ~290 planets they targeted, it's still unknown if they are planets or not.

Daniel wrote:

I don't know, if i'm wrong in my thoughts, but this new datas seems not good for many exoplanets

It's not really good or bad. We've known that some of the planet candidates will turn out to have a true mass greater than the planetary range. We also know that there's plenty that will be found to truly be planets. This paper, while certainly interesting, isn't terribly surprising as far as its results.

But I doubt about the precision of astrometric method. It hasn't proved (so far) able to detect extrasolar planets, so how could these estimates be really coherent? Moreover the data are "re-rededuced", is the quality of these data good enough? Uhm...

Also for inclinations assumed, usually masses vary in much smaller order. For example a third body in an eclipsing binary system with estimated minimum mass of 0.3 Solar may vary, according to inclination 0° up to 90° between 0.1 and 0.6 Solar masses, not exceeding six or ten times the estimate. Here exoplanets seems to have true masses exceeding 600-1000%. Even radial velocities would have appeared quite different in presence of such massive objects.

But it's those planets with true masses 600 - 1000% of the minimum mass that stick out so well in astrometry because of their larger astrometric signal. It's like hot Jupiters. They aren't particularly common, but we're finding them a lot because they make their presence obvious.

Edit:

Edasich wrote:

For example a third body in an eclipsing binary system with estimated minimum mass of 0.3 Solar may vary, according to inclination 0° up to 90° between 0.1 and 0.6 Solar masses

The true mass will never be less than the minimum mass.

M_true = M_min / sin i.
When i is 90°, M_true = M_min (and the system transits).
As i tends toward 0° (or 180°), M_true tends toward infinity (and the system is face-on).

Edasich wrote:

Moreover the data are "re-rededuced", is the quality of these data good enough? Uhm...

By "re-reduced" what it means is that Floor van Leeuwen went back to the original Hipparcos data and used a better model to more carefully take into account various systematic effects. This resulted in the second version of the Hipparcos catalogue, and it has been statistically shown that the new version is better than the old one. So this analysis is using better quality results than previous attempts to obtain orbits of planetary companions from Hipparcos data.

Edasich wrote:

Also for inclinations assumed, usually masses vary in much smaller order. For example a third body in an eclipsing binary system with estimated minimum mass of 0.3 Solar may vary, according to inclination 0° up to 90° between 0.1 and 0.6 Solar masses, not exceeding six or ten times the estimate.

Not sure what you are talking about here. In actuality for a single-lined spectroscopic binary (a class which includes the radial velocity exoplanets) the quantity we can get is the mass function:

f(m) = PK³(1 - e²)^3/2 / (2πG) = (m₂ sin i)³ / (m₁ + m₂)²

(The usually-quoted quantity m₂*sin(i) is an approximation in the case that the mass of the secondary is small when compared to the primary - things start to diverge from this approximation as the secondary mass increases.)

As the inclination tends to zero, the mass of the secondary increases without limit. This is the mass/inclination degeneracy. Of course infinite mass is not physical, so there is usually some cutoff at high mass. It is not realistic that any of the known radial-velocity candidates are in fact supermassive black holes in very nearly face-on orbits.

More on the case of γ Cephei Ab. For the orbital elements I am using Neuhäuser et al. (2007), with planetary inclination and ascending node from Reffert et al. (2011). In fact they constrained their orbits to the earlier parameters in Butler et al. (2006), but there isn't too much difference and with the later elements you get the advantage of knowing the binary orbit is modelled correctly.

Reffert et al. (2011) give two possible solutions for γ Cephei Ab, the first has (i, Ω) = (5.7°, 37.5°), the other has (173.1°, 356.1°). They do not derive the true mass for the second solution but according to the parameters given it should be about 14.7 Jupiter masses.

First solution is retrograde with respect to the binary orbit, with mutual inclination 114°. Evolution over 100,000 years shows it undergoes fairly regular Kozai cycles with length ~4,000 years, during which the eccentricity reaches a maximum of ~0.86. Semimajor axis variations (max(a)-min(a))/mean(a) were 0.7%.

Second solution is prograde with mutual inclination 54°. The maximum eccentricity is ~0.66, though the irregular timing variations in the Kozai cycles may suggest instability. The semimajor axis were 0.9% over 100,000 years.

Either way if the solution is correct it implies that we are looking at γ Cephei during a rare time when the planetary orbit is at low eccentricity. If true, there probably aren't any other planets orbiting the primary star as they would have been disrupted when γ Cephei Ab was at high eccentricity.

It would be such a tragedy if what would otherwise be the first currently-accepted-to-be-legitimate exoplanet turned out to be a brown dwarf.

Well mass range is below the desert at ~30 Jupiter masses or so, so could well be planetary. Will be amusing to see whether the formation models could manage to produce such a massive object in an inclined orbit in such a close binary... last I heard they could barely manage the job at the minimal planetary mass.

HD 168443 c has been dropped into the Extrasolar Planets Encyclopaedia "unconfirmed" list on the grounds of poorly-constrained mass.