With the online press conference re new results from the HARPS spectrograph (High Accuracy Radial Velocity Planet Searcher) now being discussed, I want to pause for a moment before getting into them to mention the ongoing Extreme Solar Systems II conference, which runs until the 17th at quite a venue, Jackson Lake Lodge in Wyoming. The tentative program is available online, with the welcome news of new HARPS and Kepler results and any number of intriguing talks on everything from debris disk imaging around nearby stars to core accretion models.
We’ll doubtless be talking about some of these findings in coming weeks. But for now, on to the HARPS discussion at the Wyoming conference. The take-away quote from today’s news was this, from Michel Mayor (University of Geneva):
“The harvest of discoveries from HARPS has exceeded all expectations and includes an exceptionally rich population of super-Earths and Neptune-type planets hosted by stars very similar to our Sun. And even better — the new results show that the pace of discovery is accelerating.”
Mayor is referring to more than 50 new exoplanets orbiting nearby stars, 16 of which are said to be super-Earths. One of them, HD 85512 b, is estimated to be on the edge of, if not within, its star’s habitable zone. Lisa Kaltenegger (Harvard Smithsonian Center for Astrophysics) points out that this world is the lowest-mass planet yet confirmed by radial velocity methods that is potentially in the habitable zone. It’s another triumph for the HARPS spectrograph, which has been shown to make the detection of planets below two Earth masses possible. More on HD 85512 b in HARPS: Hunting for Nearby Earth-like Planets, a Centauri Dreams story from August.
Image: A team of astronomers has shown that the newly discovered exoplanet HD 85512 b lies at the edge of the habitable zone of its star, where liquid water oceans could potentially exist if the atmosphere of the planet has sufficient cloud cover. This diagram shows the distances of the planets in the Solar System (upper row) in the new HD 85512 system (middle) and in the Gliese 581 system (lower row), from their respective stars (left). The habitable zone is indicated as the blue area. Based on an original diagram by Franck Selsis, Univ. of Bordeaux. Credit: ESO.
HD 85512 b is the second HARPS planet potentially inside the habitable zone, the other being the much discussed Gliese 581 d. And in addition to HARPS, which has found about ? of all exoplanets with masses less than Neptune’s, we should put ESPRESSO on our radar. Echelle SPectrograph for Rocky Exoplanet and Stable Spectroscopic Observations is in its early design stage, but scheduled to begin operations on the ESO Very Large Telescope in 2016. Now we’re really getting down to cases: ESPRESSO will have a radial velocity precision of 0.35 km/h or less. Compare that to the 0.32 km/h that Earth induces on the Sun and you can see why astronomers look forward to using ESPRESSO to discover Earth analogues around nearby stars.
HARPS observations of 376 Sun-like stars are helping us firm up the likelihood of low-mass planets around such stars. The result: About 40 percent of them have at least one planet less massive than Saturn. And the majority of exoplanets of Neptune mass or less appear in systems with multiple planets. The paper is Mayor et al., “The HARPS search for southern extra-solar planets, XXXIV — Occurrence, mass distribution and orbital properties of super-Earths and Neptune-type planets.” Still in preparation, it is to be published in Astronomy & Astrophysics. And see Kaltenegger et al., “A Habitable Planet around HD 85512?” submitted to Astronomy & Astrophysics (preprint).
On worlds with extremely deep oceans, the pressure in the deepest abysses can crack the ocean floor and create tilting plates, one side of which breaks the surface, creating narrow continents on worlds that, if they had Earth type tectonics, would have had no land at all. This has implications for many super-Earths and mini-Neptunes.
It lies on the inner edge of the ‘H’ zone – not good-Venus lies on ours and look what happened to it !!!
6 Earth-Size planets candidates found by Kepler around M-Dwarf star habitable zones!
the candidates:
KOI 463.01
KOI 1422.02
KOI 947.01
KOI 812.03
KOI 448.02
KOI 1361.01
Near-Infrared Spectroscopy of Low-Mass Kepler Planet-Candidate Host Stars: Effective Temperatures, Metallicities, Masses and Radii
http://arxiv.org/abs/1109.1819
We report stellar parameters for low-mass planet-candidate host stars recently announced by the Kepler Mission. We obtained medium-resolution, K-band spectra of 84 low-mass Kepler Objects of Interest (KOIs). We identified one KOI as a giant; for the remaining dwarfs, we estimated effective temperatures by comparing measurements of K-band regions dominated by H2O opacity with predictions of synthetic spectra for low-mass stars. We measured overall metallicities ([M/H]) using the equivalent widths of Na I and Ca I absorption features and an empirical metallicity relation calibrated with nearby stars. With effective temperatures and metallicities, we estimate the masses and radii of the low-mass KOIs by interpolation onto evolutionary isochrones. The resultant stellar radii are roughly half of the values reported in the Kepler Input Catalogue and, by construction, correlate better with effective temperature. Our results significantly reduce the sizes of the corresponding planet-candidates, with many less than 1 Earth radius. Recalculating the equilibrium temperatures of the planet-candidates from the implied stellar luminosities and masses, and assuming Earth’s albedo and re-radiation fraction, we find that six of the planet-candidates are terrestrial-sized with orbital semi-major axes that lie within the habitable zones of their low-mass host stars. The stellar parameters presented in this letter serve as a resource for further characterization of the planet-candidates.
@Michael,
I agree and not only that : this planet is a lot more massive than Venus and it seems likely to have an atmosphere. My guess is that it is denser than Earth’s and possible more than capable of retaining heat and making things worse.
http://www.dailykos.com/story/2011/09/12/1016229/-Planet-in-habitable-zone,-possibly-with-Earth-like-atmosphere,-discovered-36-light-years-away?via=siderec
Based on Centauri-Dreams
Paul you have made CD the authority
David, you’re too kind! Thank you. What a pleasure to see that story.
To Daniel, thanks, that is a really interesting paper about recalculating the radius of Kepler KOI stars and the ensuing revised planet candidates radius. I hope everybody visiting here will take the time to read it. It raises some questions alright.
Great find Daniel, thanks!
Daniel, the details of the Muirhead et al. paper you pointed to seem solid but the results are a bit troubling. An overestimate of the size of the stars by a factor of two by the KIC authors seems, on the face of it, unlikely. The mentioned discrepancy between observed M dwarf radii from double-lined eclipsing binaries vs. models is only on the order of 10%, as I remember. I would have liked to have seen a direct comparison of their calibration against the known, only roughly 4, eclipsing, double-lined M dwarf systems with really well determined parameters. Those systems are on the average much brighter than most of the KOI targets, so good s/n should be possible (all may not be accessible from Palomar though, as I seem to recall one system is in the south….). It’ll be interesting to see how the KIC authors respond, who are a veritable who’s who of stellar astronomers. My prediction: this paper spends a relatively long time in the referring process.
Well finally this answers the question of whether you can get planetary systems around more than one star in a binary system…
HD 20782 was already known to have a giant planet in an extremely eccentric orbit. Now there is the detection of two planets (12 and 16 Earth masses) around the companion star HD 20781. Nice.
A relevant paper regarding an interesting difference in the range of exoplanets being picked up by HARPS and KEPLER that might be of interest to readers – the former picking up a higher proportion of smaller / higher density planets and vice versa, which the authors suggest may be related to the different methodologies used.
http://arxiv.org/abs/1108.5842
What I found most interesting are the metallicities of the stars these planets orbit. I’d always assumed that terrestrial worlds like Earth needed higher metallicity environments, and that gas giants made from hydrogen and helium could form everywhere. Turns out that it’s the other way around – assuming the bottom-up formation mechanism is dominant, then gas giants need rocky cores of 10-20 Earth masses to accrete onto, and in a low metallicity environment it’s just not possible to build these cores up quickly enough before the gas in the protoplanetary disc is blown away. Lower mass terrestrial planets on the other hand can form at a slower rate from a lower metallicity disc, so they seem to be found around both high and low metalliocity stars. To quote from the paper, Mayor et al that accompanies the new HARPS findings (http://www.eso.org/public/archives/releases/sciencepapers/eso1134/eso1134b.pdf):
“The correlation between the occurrence of giant planets and the metallicity of the host star (i.e. the metallicity of the material in the proto-planetary disc) is a natural outcome of the core accretion planet formation theory. In this paradigm, massive gaseous planets form by runaway gas accretion onto cores
exceeding a critical mass, typically of the order of 10 ? 20M. The gas accretion from the disc goes on until the disc vanishes, typically after a few million years. Hence, the sooner in the evolution of the disk a critical-mass core can form, the larger the amount of gas that will still be available for accretion. A high metallicity (interpreted as a large dust-to-gas ratio in the models) and/or massive discs favors the early growth of such critical cores. Conversely, lower-mass planets that do not accrete significant amount of gas, can grow their cores over a longer timescale and therefore do not depend as critically upon the metallicity.”
Let’s see what this means in the grander scale. If terrestrial planets can form around lower metallicity stars, then they could have formed sooner in the history of the Universe, allowing more time for life to develop. Since it’s believed that hot jupiters cause the ejection of smaller planets as they migrate towards their star, and if hot jupiters are more likely to exist the higher the metallicity over time, then the survival of habitable exoplanets may become less frequent the older the Universe gets. I’m not sure if this has been spelled out explicitly in any paper, but the analysis suggests that this is a possible implication. If there is life out there, this could be another argument for why it should be far older than life on Earth. I wonder when, in the history of the Galaxy, was the optimum time for the formation of habitable planets based both on metallicity and the conditions to allow the formation of hot jupiters that could mess things up?
Michael and Enzos concerns are well-founded. However, climate and atmospheric effects are not fully understood, especially in exotic situations, which leaves the possibilities open.
Just to give an example, Venus became an oven because the water evaporated, and the hydrogen was split off by the sun’s rays and swept into space. This led to the formation of more C02 and a runaway greenhouse effect (correct me if I’m wrong here). But if you change the variables, it could lead to a different outcome. HD 85512 b has a less powerful star, and it most likely has more mass and surface gravity than Venus. This could prevent the aforementioned scenario from taking place (less energy from the sun to split hydrogen, and more gravity to hold on to it).
I keep an open mind about exoplanets in general. There could be ice age earths out there, or maybe temperate earths with very high winds across the surface. And of course, we’ve discussed tidally locked earths with a hot side, cold side, and moderate terminator.
The interesting thing about their paper (http://www.eso.org/public/archives/releases/sciencepapers/eso1134/eso1134a.pdf) is that although they do mention Alpha Cen B as a target, they expressly skip it as they will apparently dedicate it a whole paper :
“A detailed discussion specific to Alpha CenB will be presented in a forthcoming paper.”
You’d think that if they hadn’t found anything they probably wouldn’t be saying that eh? Especially in their intro where they describe the possibility of planets around ACenB as tantalizing…
Maybe a joint paper with Fisher’s team? At any rate, given that observations started nearly three years ago, we should know pretty soon – one way or another.
Fascinating times we live in and a truly fascinating HARPS paper!
As the authors are saying, the sample size and present results allow for sufficiently robust statistics and modelling of occurrence of low-mass planets.
A few interesting highlights:
– About 40 % of solar type stars have at least one planet with a mass smaller than Saturn (was already mentioned by Paul).
– More than 50% of solar-type stars harbor at least one planet of any mass and with period up to 100 days.
– As was reported before and confirmed here: occurrence of Super-Earths and Neptune-mass planets (SEN) is not strongly correllated with metallicity, and strongly increasing between 30 and 15 Mearth.
– Most SEN planets are in multi-planet systems.
– Also reported before and confirmed here: occurrence of gas giants is strongly correlated with (increasing) metallicity and with orbital period (i.e. the hot Jupiters are rare, the ‘normal’ Jupiters common).
– About 14% of solar-type stars have a planetary companion more massive than 50 Mearth (i.e. about Saturn) on an orbit with a period shorter than 10 years, 10% if we take a 100 Mearth lower limit.
– From table 1: less than 1% have a hot gas giant (hooray!).
– From table 1: at least 75% of stars have at least one planet with an orbital period less than 10 years.
– From table 1: about 50% of stars have a SEN planet (less than 30 Mearth) in close orbit less than 100 days.
See in particular table 1 and figures 12 and 16. And read chapter 5 on the metallicity correllation (I was drooling while reading it, luckily a few hours off from work).
Important conclusions:
– There is a very clear abundance of these mid-sized super-earths/neptunes (SEN), and also a dominance of these with short orbital periods (over 50% of all stars).
– As mentioned: occurrence rate of gas giants is strongly correlated with the host star metallicity.
– As mentioned: occurrence rate of SEN does not exhibit a preference for metal rich host stars. Question remains whether there is an absolute lower metallicity limit for (any) planet formation. There could be a lower limit around dex -0.45 0r -0.50 (about 30 – 35% of solar metallicity).
– There might be (though not proven yet) two metallicity regimes at work here: above about Fe dex 0.15 – 0.20 (about 150% solar metallicity) there is virtually always a gas giant over 40 Mearth. And virtually all stars that only harbor planets less than 30 Mearth have lower metallicity than that (note: this last fact does not work the other way around: stars of lower metallicity may also harbor a gas giant, be it less commonly with decreasing metallicity and rarely below about dex -0.3, about 50% solar metallicity). And finally there is a gap in planet mass distribution: a rarity of planets of mass 30 – 40 Mearth.
James Cameron has already proven there are planets and life in the Alpha Centauri system. And a large supply of unobtanium. So let’s fire up the corporate starships already.
An exoplanet hunting device named ESPRESSO? Not sure whether to be appalled or impressed.
Another fascinating paper and mentioned in this HARPS paper, is the recent “COMBINING KEPLER AND HARPS OCCURRENCE RATES TO INFER THE PERIOD-MASS-RADIUS DISTRIBUTION OF SUPER-EARTHS/SUB-NEPTUNES”, by Wolfgang and Laughlin (August 2011), I also found the reference to it on Laughlin’s systemic site.
It harmonizes the seeming discrepancy in occurrence of super-earths and sub-neptunes, as found by HARPS on the one hand and Kepler on the other (HARPS found this frequency about 2 to 3 times as high as Kepler), “via plausible distributions of planetary compositions”.
The most interesting result is that there are probably two distinct sub-populations of these mid-class planets: (1) dense silicate-iron planets and (2) low-density gas-dominated worlds. This suggesting “multiple formation mechanisms”.
And “the fraction of dense planets decreases with increasing mass” (from about 90% to 10% across the entire mass range). In other words: toward the lower end of the range are mainly the real terrestrial planets (iron-silicate) and toward the high end mainly the gas/liquid subgiants.
This paper also mentions about the same estimate of occurrence of these super-earths/sub-neptunes in close orbit (up to 50 days) around solar type stars that was mentioned at the HARPS press conference: at least 40%.
@andy: “Well finally this answers the question of whether you can get planetary systems around more than one star in a binary system.
HD 20782 was already known to have a giant planet in an extremely eccentric orbit. Now there is the detection of two planets (12 and 16 Earth masses) around the companion star HD 20781.”
Yes, they are both solar type (G9.5, Lum 0.43*solar, and G1.5, Lum 1.23*solar) and almost solar metallicity (slightly lower).
However, how far apart are they in AU (mean, minimum)? In other words, how close binaries? I could not find that myself. Would be interesting to know,
This paper gives a projected separation (in the sky plane) of 9080 AU, about 0.14 light years. So definitely a wide binary, and the orbit is obviously unknown since we haven’t been observing for long enough. Apparently the high eccentricity of HD 20782b cannot be induced by the Kozai mechanism if the object responsible is HD 20781, if the Kozai mechanism (which is a pretty good way of driving objects to high eccentricity) is responsible then there may well be more objects in this system…
Keith Cooper -yes I’ve been commenting for a while that some planetary formation models favour lower metallicity than solar to produce terrestrial planets. One paper by Ida et al stated there could be about 80 terrestrial planets produced per low-metallicity star in a globular cluster (many of which get ejected of course).
As to the effect on the GHZ, yes if you look at the Lineweaver paper and the later Prantzos paper you will see the balance between a minimum metallicity and a hot-Jupiter metallicity is explicitly modelled. However not very well because of the lack of data, so hopefully these models can now be refined extensively.
Pierre Cauchy
I do not think they are going to talk about a planets that they have found there, because if they had found a planet around cen B. They will talk only about that. I think they are going to talk about that it is a little harder to find a planet there because of the other star and that it will take a little more time to find a planet around cen B
I’ve never been much convinced of the GHZ concept. One finding that comes out of simulations of galactic structure and evolution is that spiral arms are pretty good at moving stars around the galaxy – at the leading edge of an arm stars tend to move inwards, at the trailing edge they move outwards. I’d guess this would tend to disrupt the nice neat idea of a metallicity-based GHZ.
Does anybody know if HD 85512 b is expected to be tidally locked?
Speaking of planets in binary systems, is it true that the Kepler mission is also capable of finding planets that orbit around two very closely orbiting stars? I am referring to stars like CM Draconis– eclipsing binaries that orbit each other with a period of a day or two.
@bigdan201 ,
you make some very good points, particularly about the weaker UV light and he reduced H2O photolysis.
I think that the best there hope is for a very reflective atmosphere.
However, even innocuous gases like nitrogen can be greenhouse gases at higher densities :
http://www.astrobio.net/index.php?option=com_retrospection&task=detail&id=1762
I’m much more hopeful for super-earths out of the cold side of the canonical habitable zone as I expect a thick atmosphere to take care of that.
Steinn Sigurðsson of Penn State (one of my alma maters, go Nittany Lions) is live-blogging this meeting at his excellent blog Dynamics of Cats: http://scienceblogs.com/catdynamics/.
Astronomy journalist (and amateur astronomer) Govert Schilling is tweeting the meeting (not as informative, given the media) at GovertTweets.
@Andy interesting paper you linked to, but I find its results pretty dubious as there’s no evidence that the mass density of spiral arms is higher than that of the regions between the spiral arms. The luminosity is higher of course but that’s bc of all the HII regions around the newly formed high mass systems. People have been modelling spiral galaxies for a pretty long time and it’s possible everyone has gotten it wrong except these guys, but it seems fairly unlikely.
Talking about the GHZ;
As kzb pointed out in the recent discussion thread on “SETI: Let the Search Continue”, the paper “A Model of Habitability Within the Milky Way Galaxy”, by Gowanlock et al July 2011, presents a new model of the Galactic Habitable Zone (GHZ), based on supernova occurrence and individual stellar modeling, and gives a new estimate of the number of habitable planets in the MW.
Remarkably, and deviating from Lineweaver and others, the GHZ in this model is considered to be much larger than in previous models.
Also remarkable is the fact that the number of habitable planets is estimated much higher than the 50 – 200 million of previous estimates: 1.2% of all stars are estimated to have a habitable planet, which would imply some 2 billion for our MW.
One reason for this much higher estimate is the mentioned much larger GHZ and the higher density of habitable planets toward the inner MW as well as above and below the (thin) disc. Whereas Lineweaver and others consider the GHZ as an annular ring in the galactic disk from (6)7 – 9(10) kpc from the galactic centre, this model takes the entire galactic disk from 2.5 – 15 kpc. “The GHZ is thought to be affected by an inner boundary that is determined by hazards to planetary biospheres (supernova explosions!), and an outer boundary set by the minimum amount of metallicity required for planet formation”.
Another reason is that this model includes all main sequence stars, not just the solar type stars. Very high metallicity and a possibly related Hot Jupiter effect is considered a much lower danger than for instance by Lineweaver.
Yet another reason is that habitable planets in this model can have a mass from 0.1 – 10 Mearth, a rather great range (I would rather suggest something from 0.3 – 3 Me or so). And the HZ in our own solar system is considered to be from 0.8 – 1.5 AU, again rather optimistic, especially on the low (hot!) side. Most researchers would put this lower limit at about 0.95 AU.
Finally, the estimate also includes tidally locked planets, in fact about 75% of all habitable planets are tidally locked in this estimate.
Most remarkable, the “the greatest number of habitable planets exist in the inner Galaxy in all of our models. More specifically, 50% of the habitable planets lie at R < 4.1 kpc and R < 4.4 kpc”, because of greater stellar density and higher metallicity. This is a region that most earlier models of the GHZ would exclude.
Paul: maybe you could delete my first of two comments on the HARPS paper (Ronald September 13, 2011 at 9:29; starting with ‘Fascinating times we live in and a truly fascinating HARPS paper!’), because it came through distorted and was made redundant by the second, succesful attempts at 9:50 (starting with the same sentence).
And the short comment of 9:53 (Hi Paul, etc.) was purely a notice to you and is also redundant for this thread I think.
Sorry for the inconvenience. Kind regards, excellent website as always and always right on top of the latest news, keep up the good work!
Andy
There IS a metallicity gradient, decreasing with galactic radius. The Lineweaver paper (2004) had the GHZ from 7 to 9kpc, but later papers by Prantzos (and the other recent one) said that basically there could be habitable planets at almost any radius.
Thanks for summarising the paper so well Ronald, that whet my appetite and I read it last night.
Perhaps I’m jumping the gun, but I now feel that we can take it as read that planetary systems around stars are pretty much universal.
I’m also wondering if the sun is actually past the optimum metallicity for terrestrial planet population. According to the paper, multiple super-Earths with periods of under 100 days are commonplace with stars of lower metallicity than solar. Yet here, Earth is the largest terrestrial planet.
Also, lower metallicity stars do not have massive gas giants such as Jupiter. Maybe Jupiters are bad news for retaining terrestrial planets, even if they are not a “hot” Jupiter.
@kzb: “I’m also wondering if the sun is actually past the optimum metallicity for terrestrial planet population. According to the paper, multiple super-Earths with periods of under 100 days are commonplace with stars of lower metallicity than solar”.
Well, I I am not so sure about that: assuming that you refer to the HARPS paper, it also shows clearly, as I also put in my summary of it, that for solar type stars, and particularly those with less than 1.5 times solar metallicity, the most common configuration seems to be super-earths and (sub)Neptunes in close orbit, at least 50% of them have them within 100 day orbital period.
It remains to be seen whether there are more terrestrial planets farther out in such systems.
Anyway, this kind of system, consisting of several super-earths and (sub)Neptunes in the inner system seems to be the norm in our MW galaxy.
It may be correllated with a certain metallicity range for solar type stars, but that remains to be seen.
kzb
without a jupiter mass planet in the outer solar system. Is it not a higher chance that the terrestrial planet will not have water and will not protect the terrestrial planets for giant meteorites.
@Ronald
It remains to be seen whether there are more terrestrial planets farther out in such systems.
True enough, but it’s very tempting to extrapolate outwards. If there are so many of these planets close-in, is it unreasonable to imagine more of them further out in the same system?
It may be correllated with a certain metallicity range for solar type stars, but that remains to be seen.
Have they not demonstrated this already, albeit imperfectly? They are saying this pattern correlates with lower metallicity and the possession of massive gas giants (including hot Jupiters) correlates with high metallicity, and the two patterns seem to be exclusive.
@coolstar: Actually in the “traditional” density-wave theory for the spiral arms, the arms do correspond to regions of greater density (hence the term “density wave”).
@kzb: yes there’s a metallicity gradient, but if the stars tend to get moved around you will get higher metallicity stars at outer radii even if the main population at such radii is lower metallicity (the gradient is flattened out but not entirely reversed). Even so, this still leads to habitable stars at a wide range of radii.
Michael Enzo etc: It should be possible to create a new scale for ” Habitable Planet Potential ” to include 1) The size and temperature of the star ( the current sole criteria for determining the boundaries of the “habitable zone” ) 2) the orbital characteristics of the Planet and 3) the Greenhouse index of the planet, which is the ability of the prospective planet to retain heat. This green house index would be primarily a matter of size, density and the ability to retain an atmosphere, as well as composition ( for example “metalicity”) Clearly if Mars were larger and had a deeper atmosphere it would likely be warmer, if Venus were Mars-like in size in composition it would be likely cooler. This could also be extended for moons of planets. Thus big planets on the outer edge of what is now referred to as the “habitable zone” would be favored over the same large sized planet closer to the primary. The habitable zone is a nice concept for a star where we have no idea of the planets it may have but is immediately out of date once we start to get data on the planet (s) characteristics as well. We live in exciting times!
Indeed, that is interesting. I always imagined that exoplanets with deep oceans would have ribbon-like archipelagos of land across their surface. This would make a stronger case for that.
While I stick by what I said, I agree with this. Ice ages seem like less of a threat to habitability than a runaway greenhouse effect, based on our limited observation.
@kzb: “It may be correllated with a certain metallicity range for solar type stars, but that remains to be seen.
Have they not demonstrated this already, albeit imperfectly? They are saying this pattern correlates with lower metallicity and the possession of massive gas giants (including hot Jupiters) correlates with high metallicity, and the two patterns seem to be exclusive.”
Well, yes and no: with regard to this issue the HARPS paper shows a few things more or less clearly (it took me some studying to get this clear), see in particular chapter 5 and figures 16 and 17, noting that it is important to realize that this study chracterized planetary systems only by their most massive planet, in other words, a system with a gas giant may also have smaller planets detected, but not vice versa:
– Gas giants: “The occurrence rate of giant gaseous planets strongly correlates with the host star metallicity. (…) a robust and well-defined relationship between the frequency of gaseous giant planets and the metallicity of their host star”.
Beyond a certain metallicity (dex 0.15 – 0.20, roughly 1.5*solar) there is *always* a gas giant planet present. In fact, in all but 2 or 3 cases (which may even be due to observational angle, because not the real planet mass is measured but m sin i) this is already the case beyond dex 0.05, roughly 1.1*solar. This is shown by the near-empty block top-left in fig. 17. Note that I am not saying that below this metallicity level there are no gas giants, see next.
– Planetary systems without any gas giant (hence only smaller planets) almost only exist at metallicity levels below a certain limit, the same level as above. However, I am not saying that below this level there are *only* systems without gas giants, on the contrary, see next.
– The previous two points are not mutually exclusive. Below this mentioned metallicity level, there are systems with gas giants and systems without them. In other words, there is a large metallicity range with overlap of the two kinds of planetary systems. From the paper we can infer that there are virtually no gas giants below dex -0.3, roughly 0.5*solar. But below this level planets become scarce anyway and not far below this at about dex -0.5, roughly 0.3*solar, planets cease to exist.
– Because of the very strong metallicity-gas giant correlation, it is therefore safe to suggest that between the lower metallicity limit for gas giants (dex -0.3) and higher limit (dex 0.15) there is an increasing occurrance of systems with gas giants and hence a decreasing occurrence of systems without them (only super-earths/(sub)Neptunes). Below the lower level: never. Beyond the higher level: always.
@henk
without a jupiter mass planet in the outer solar system. Is it not a higher chance that the terrestrial planet will not have water and will not protect the terrestrial planets for giant meteorites.
I don’t think it is settled whether Jupiter is a NET protector against asteroid/comet impacts or not. The balance of opinion seems to be swaying in favour of Jupiter causing MORE impacts, not less.
It’s also thought that most of Earth’s water came from cometary impacts, so the absence of a Jupiter might well lead to less water being delivered. But on the other hand, all these planets discovered so far are super-Earths and lots of people are hypothesising they are water worlds, i.e they have TOO MUCH water. So perhaps the absence of a Jupiter is actually beneficial to super-Earths, preventing them from being water worlds.
‘Never’ and ‘always’ in my previous comment refer to occurrence of gas giants, of course.
@Andy “density wave” is actually a misnomer, what that theory really implies is a pressure wave, so that the local density can be a VERY, VERY tiny amount greater, just great enough to start giant molecular clouds collapsing to make stars. While the average density of a spiral arm (total mass divided by volume) is not measurably different from that between the spiral arms. So, not nearly large enough to funnel stars around the galaxy.
@kzb: interesting point about the (dis)advantages of a Jupiter like gas giant and water delivery.
I wonder to what extent comets have delivered water on the young earth. It seems that the amount of water in the earth mantle is very large, I don’t know exactly but I have heard estimates from 5 to 20 times all the earth’s oceans. It actually plays an important part in the earth plate tectonics, as a kind of lubricant (Venus’ absence of plate tectonics may be due to lack of water).
And since water is a very common substance I wonder how much, if any, of this water was ‘original’.
Ronald
In Section 5 it says:
…The occurrence rate of giant gaseous planets strongly correlates
with the host star metallicity……For lower-mass planets, already after the very few first detections (Udry et al. 2006) suggested the absence of correlation between the host star metallicity and the presence of low-mass
planets. This first claim was later confirmed….
It then goes on to say the current study will refine this conclusion.
In the conclusions section 7 it says
Opposite to GGP (Giant Gas Planets), the occurrence
rate of SEN (Super Earth & Neptunes) does not exhibit a preference for metal
rich host stars. The difference between both populations of
planets is striking…..
So those were very much the messages I took from the paper.
kzb
it is good for a super earth, but have super earth not a stronger gravity. There can be good places for life. Are they also good places to come as humans, because without good technologie we can not walk on that planet. unless they are not that much bigger. That is one of the reasons that we want to find a earth mass planet,but than we need afcourse now how to travel to it. It is so sad that the stars are so far away from us. We can only look to those exo planet. It is like Galileo look to the moons of jupiter
also people only look at 3 types of planets earths, venus or mars type worlds
maybe some worlds are very different to the ones we now of.
henk
I’m not sure it is sad, because what is being found here is that lower mass planetary systems are commonplace. The smaller planets found so far are all super-Earths, that is not to say smaller worlds further out will not be found, in fact I think the prospects for that are now greater than imagined. It looks highly possible to me that lower-metallicity stars typically have a glut of terrestrial planets, far more than in our system. Anyway time will tell.
Yes we would not find life on these superEarths at all comfortable. But that is not to say that advanced life and civilisations are not possible on them, far from it.
henk & kzb:
Gravity increases linearly with mass, but decreases quadratically with distance. Radius approximately goes with the cube root of mass. Overall, surface gravity should therefore increase roughly with the cube root of mass. This would make Earth-like planets of up to about 4 Earth masses still relatively comfortable for us, considering 1.5 g as “comfortable” once you get used to it.
Even much larger planets could qualify if they are less dense, perhaps lacking a large iron core. Just the kind of thing you might find in low-metallicity systems. Less dense planets of the same mass will be bigger AND have less gravity, a double bonus, if you will. Neptune, for example, has a surface gravity of 1.14 g. If it had a surface, and weren’t quite so chilly, it could be a very comfortable place to live.
Apparently some mice subjected to increased gravity lived longer than those in normal gravity (See the “Habitable Planets for Man” book recently mentioned here). It would sure be an excellent environment to build some muscles (and raise mice).
Indeed @ Eniac. Those facts are very encouraging for potential habitability, since surface gravity is one of the major issues.
@kzb: yes, two populations of planets, however, with overlap, as you can see in chapter 5 and figures 16 and 17.
Above a certain metallicity (roughly 1.5 times solar) there is *always* one or more giant gas planets.
Vice versa, planetary systems with excusively smaller planets *only* occur below a certain metallicity (the same roughly 1.5 times solar).
However, that does not mean that below that threshold metallicity level there are always only smaller planets.
Below this level, there may or may not be, gas giants, the occurrence of them increasing with increasing metallicity. See fig. 17 below the dex 0.15 metallicity line.
See in particular the first part of chapter 5 and figure 17.
Note that the authors are also stating:
“It is however not clear whether we observe a discontinuity in the-host star
metallicity distributions with an increasing planetary mass. On Fig.17, the ([Fe/H], m2 sin i) diagram can be used to set a limit between the two regimes of host star metallicity (if such a limit proves to be meaningful!)”.
To make myself absolutely clear, these are two different statements:
1) Planetary systems with only smaller planets always occur below metallicity level x.
This statement is correct, from the HARPS paper.
2) Below metallicity level x there are only planetary systems with just smaller planets.
This statement is wrong.
I liked the way that Jkittle used the idea of “Habitable Planet Potential”, but I think it could have been extended even further. It was great to acknowledge the planetary orbital characteristics and likely atmospheric structure in the equation, but what we are most interested in is how likely life there is to lead to higher forms. If we even just attempt to add significance to this equation it would allow others to see how obviously wrong the rare Earth argument is.
To do so we would have to work with probabilities of probabilities, and this is so tricky that it is easy to loose your way, but even giving a rough estimate along the lines of “Smith and Jones estimated that only 1% of stars harbour planets that are more suitable for higher life to evolve on than it was on Earth” is enough to emphasise the bloody obvious. It IS possible for a planet to be better disposed for that purpose than Earth. Furthermore IF higher life suitable planets are very rare, THEN most such development of sentience will not happen on the very best possible such candidates, but on the much larger category on the next echelon down (even though we know that it is the paucity of this next category that actually creates the problem). Just listing all factors that Earth has yet look potentially atypical and then using this to sieve is just nutty.