Posts Tagged ‘Radial Velocities’

This just in…

March 25th, 2010 1 comment

With HAT-P-13c rapidly coming ’round the mountain, there was a very timely update on astro-ph last night. Josh Winn and his collaborators have obtained an additional slew of radial velocities which (1) demonstrate using the Rossiter-McLaughlin effect that the inner planet b’s orbit is likely well aligned with the stellar equator, (2) modify the orbital parameters, including the period of the outer massive planet, and (3) hint at a third body further out in the system.

How do these updates affect the unfolding story?

The Rossiter-McLaughlin measurement gives an estimate of the angle λ = -0.9°±8.5°, which is the angular difference between the sky-projected orbital angular momentum vector and sky-projected stellar spin vector. A non-intuitive mouthful. If we’re viewing the star edge-on, then λ = -0.9° amounts to a determination that the planet’s orbital plane is well-aligned with the star’s equator. (See this post for a discussion of what can happen if the star’s rotation axis is tipped toward the Earth). The good news from the measurement is that it’s a-priori more likely that planets b and c are coplanar — that happy state of affairs which will permit direct measurements of planet b’s interior structure and tidal quality factor. If, on the other hand, the planets b and c have a large mutual inclination, then b’s node will precess, and measurement of a small value for λ will occur only at special, relatively infrequent, times during the secular cycle. A close to co-planar configuration also increases the likelihood that the outer planet can be observed in transit.

With their beefed-up data set of out-of-transit Doppler velocities, Winn and his collaborators are able to get a better characterization of the planetary orbits. The best-fit orbital period and eccentricity of the outer planet are slightly modified when the new data are included. The best-guess center of the transit window for c has “slipped” to April 28, 2010, with a current 1-σ uncertainty of 2 days.

The later date, however, is not an excuse for procrastination! Measuring the TTV for this system is a giant opportunity for the whole ground-based photometric community, and a definitive result will require lots of good measurements of lots of transits starting now (or better yet, last month.) I’ll weigh in in detail on this point, along with the challenge posed by Mr. D very shortly…

Earthbound Planet Search

December 14th, 2009 15 comments

61 Vir b (simulation by J. Langton, Principia College)

61 Vir b (simulation by J. Langton, Principia College)

The ranks of the super Earths and the sub Neptunes continue to grow! In two papers that have been accepted by the Astrophysical Journal, and which will be coming out in tomorrow’s astro-ph mailing, the Earthbound Planet Search team is announcing the detection of very low mass planets orbiting the nearby solar twins 61 Vir and HD 1461. (Link to paper #1, link to paper #2).

The 61 Vir system is particularly compelling. The radial velocity data for this star indicate that at least three planets are present, with an architecture that’s quite a bit more crowded than the Sun’s terrestrial planet zone:


The innermost planet, 61 Vir b, with Msin(i)~5.5 Earth masses, has a radial velocity half-amplitude K=2.15 m/s, which puts it in league with Gl581e (with K=1.9 m/s) and HD40307b (with K=2.0 m.s) as the lowest-amplitude Doppler detections to date.

We’ve adopted the Systemic Console software to analyze the Doppler velocities that are produced by the Earthbound Planet Search. I’ve written a tutorial (link here) that explores the 61 Vir dataset in detail, and shows how the planets are extracted.

parallel observing

December 13th, 2009 3 comments


As the decade draws to a close, it’s hard not to be amazed at the progress that’s been made on every research front related to extrasolar planets.

An area that I think is now ripe for progress comprises coordinated multi-observer checks for transits by super-Earth/sub-Neptune planets. There are now over thirty known extrasolar planets with Msin(i)’s less than that of Gliese 436b (which tips the scales at 23 Earth masses). Of these, only CoRoT-7b has so far been observed to transit, and it’s very probable that the current catalog of low-mass RV-detected planets contains one or more transiting members. Needless to say, it’d be very interesting to locate them.

To my knowledge, the lowest-amplitude transits that have been observed by amateur astronomers have been those by HD 149026b. This anomalously dense Saturn-mass planet induces a photometric transit depth of roughly 0.4%.  State-of-the-art amateur detections show the transit very clearly. Here’s an example (the observer was Luboš Brát of the Czech Republic) taken from the TRESCA database:


The identification of transits by small planets certainly won’t be a picnic. Super-Earths and  sub-Neptunes orbiting G and K stars present targets that are intrinsically much tougher than HD 149026. Unless the parent star is a red dwarf, the expected transit depths will generally be less than 0.1%, and it’ll be extremely difficult for a single small-telescope observer to obtain a definitive result.

On the other hand, if a platoon of experienced observers mount a coordinated campaign on a single star, then there’s a possibility that a startlingly good composite light curve might be obtained. In theory, if one were to combine the results from sixteen independent observers, one could obtain a light curve of the equal signal-to-noise as the HD 149026b curve shown above, but for a planet with a transit depth of only 0.1%.

I spent time this weekend making sure that the transit predictions for the known RV-detected low-mass planets are as up-to-date and accurate as possible. I found that HD 7924 is a good candidate star with which to test a coordinated observing strategy. The star harbors a low-mass RV-detected planet was announced earlier this year (discovery paper here):


HD 7924b has Msin(i)~10 Earth Masses, a P=5.3978d orbital period, and a 6.7% a-priori chance of being observable in transit. The (folded) photometry in the discovery paper is of quite high quality, and shows that the star is not photometrically variable. The photometry also indicates that transits with depth greater than 0.05% are probably not occurring. The parent star, HD 7924 is a K-dwarf, with a radius of something like 78% that of the Sun, which means that if the planet is a sub-Neptune it’ll have a central transit depth of order 0.075%, whereas if it is a rocky object, the depth will likely be less than 0.05%. The 1-sigma uncertainty on the time of the transit midpoint is about 0.35 days. The parent star has V=7.2, and with Dec=+76 deg, it’s circumpolar for high-latitude observers (RA=01h 21m).

Here are the next predicted transit midpoints (dates and times are UT):

HJD        Y    M  D  H  M
2455182.04 2009 12 16 12 51
2455187.01 2009 12 21 12 14
2455192.41 2009 12 26 21 48
2455197.81 2010  1  1  7 21
2455203.20 2010  1  6 16 54
2455208.60 2010  1 12  2 28

Because HD 7924b’s period is known to an accuracy of 0.0013 days (2 minutes), participating Northern-hemisphere observers can obtain data during any of the upcoming opportunities. Their light curves, once standardized, can in theory be stacked to obtain increased precision. It would be very interesting to get a sense of the practical limits inherent in such an approach. I think the best way to test the limits is to give the observations a try!

that golden age

December 8th, 2009 6 comments


I’m nostalgic for ’97, when the discovery of a new extrasolar planet was literally front-page news. What’s now cliche was then fully viable poetic sweep. Epicurus and his multitude of worlds. Bruno burning at the stake. In that frame of mind, it’s fascinating to go back and read John Noble Wilford’s extended New York Times piece, written at the moment when the number of known extrasolar planets equaled the number of planets in our own solar system.

Some of the hyperbole still seems fresh, especially with regard to the frequency and diversity of planetary systems:

And the discoveries may be only beginning. One recent study suggested that planets might be lurking around half the Milky Way’s stars. Astronomers have already seen enough to suspect that their definition of planets may have to be broadened considerably to encompass the new reality. As soon as they can detect several planets around a single star, they are almost resigned to finding that the Sun’s family, previously their only example, is anything but typical among planetary systems.

At the recent Porto conference, the Geneva team not only reiterated their claims regarding the frequency of low-mass planets, but actually upped their yield predictions. According to a contact who heard Stephane Udry’s talk, the latest indication from HARPS is that between 38% (at the low end) and 58% (at the high end) of nearby solar-type stars harbor at least one readily detectable M<50 Earth-mass planet. This is quite extraordinary, especially given the fact that were the HARPS GTO survey located 10 parsecs away and observing the Sun, our own solar system (largely in the guise of Jupiter’s decade-long 12-m/s wobble)  would not yet be eliciting any particular cause for remark.

It also looks like planets beyond the snowline are quite common. In yesterday’s astro-ph listing, there’s a nice microlensing detection of a cold Neptune-like planet orbiting a ~0.65 solar mass star with a semi-major axis of at least 3 AU. The microlensing detections to date indicate that Neptune-mass objects are at least three times as common as Jupiter mass objects when orbital periods are greater than five years or so. Microlensing detections are an extremely cost-effective way to build up the statistics of the galactic planetary census during belt-tightening times. Much of the work is done for free by small telescope observers.


Yet another dispatch pointing toward a profusion of planets comes from an article posted last week on astro-ph by Brendan Bowler of the IfA in Hawaii. Work that he’s done with John Johnson and collaborators indicates that the frequency of true gas giant planets orbiting intermediate-mass stars (former A-type stars like Sirius that are now in the process of crossing the Hertzsprung gap) is a hefty 26% within ~3 AU.

An embarrassment of riches? Certainly, the outsize planetary frequency means that the cutting-edge of the planet-detection effort will be shifting toward the Sun’s nearest stellar neighbors, as these are the stars that offer by far the best opportunities for follow-up with space-based assets such as HST, Spitzer, JWST et al.

As competition for ground-based large-telescope RV confirmation of run-of-the-mill planet transit candidates orbiting dim stars heats up, the threshold magnitude (at a given bandpass) at which stars become largely too faint to bother with will grow increasingly bright. We’re talking twelve. Maybe nine. Pont et al., in their discovery paper for OGLE-TR-182b refer to this threshold as the “Twilight Zone” of transit surveys:

The confirmation follow-up process for OGLE-TR-182 necessitated more than ten hours of FLAMES/VLT time for the radial velocity orbit, plus a comparable amount of FORS/VLT time for the transit lightcurve. In addition, several unsuccessful attempts were made to recover the transit timing in 2007 with the OGLE telescope, and 7 hours of UVES/VLT were devoted to measuring the spectroscopic parameters of the primary. This represents a very large amount of observational resources, and can be considered near the upper limit of what can reasonably be invested to identify a transiting planet.

Lobbying for Alpha Cen

August 29th, 2009 10 comments

Philippe Thebault sent me a link to an article on the Alpha Centauri planet search published earlier this month in the Frankfurter Allgemeine Zeitung. The text is in German, but the Google translator does a passable job of getting the gist across.

I got my first inkling of the Geneva Planet Search’s Alpha Centauri campaign through Lee Billings’ article in Seed Magazine. (See this post). In the Frankfurter Allgemeine article, Francesco Pepe gives further details — Alpha Cen B is one out of ten stars that are receiving special scrutiny for terrestrial planets at HARPS. They are getting one observation every two weeks, meaning that the star is being hit roughly one out of every two of their planet search nights:

“Allerdings müssen wir uns Harps mit anderen Gruppen teilen”, sagt er. Zudem ist Alpha Centauri B nur einer von zehn Sternen, die sie auf erdähnliche Planeten absuchen wollen. “Aber alle zwei Wochen schauen wir damit auf Alpha Centauri, und das Gerät ist sehr effizient.”

This quote implies that my speculations regarding the Geneva team’s data collection rate on Alpha Cen B were somewhat overheated. Instead of getting 100 ultra-high-precision HARPS velocities per year, it looks like a more realistic estimate of their current rate is 25 velocities per year. Since signal-to-noise increases as the root of the number of observations, this means that the minimum mass threshold for Alpha Cen Bb at any given time is approximately doubled relative to my estimates at the beginning of the Summer. Instead of arriving at 2.5 Earth masses in the habitable zone a bit more than a year from now, they’ll be at roughly 5 Earth masses.

Now nobody likes backseat drivers. As the saying goes, “theorists know the way, but they can’t drive”, and theorists have had a particularly dismal record in predicting nearly everything exoplanetary.

But nevertheless, I’m urging a factor-of-four increase to that data rate on Alpha Cen B. I would advocate two fully p-mode averaged velocities per night, 50 nights per year. I know that because Alpha Cen B is so bright, the duty cycle isn’t great. I know that there are a whole panoply of other interesting systems calling for time. It is indeed a gamble, but from the big-picture point of view, there’s a hugely nonlinear payoff in finding a potentially habitable planet around Alpha Centauri in comparison to any other star.

During the next few months, it’s inevitable that one of the numerous Super-Earths that have been turning up in the radial velocity surveys will be announced to be observable in transit (see, e.g. this post). When that occurs, we’ll effectively have had our last first look at a truly new category of planet — the logarithmic mass interval between Earth an Uranus is currently by far the largest among the 70-odd planets that have accurately determined radii. My own guess is that the emerging population of super-Earths will be better described as a population of sub-Neptunes. That is, the transit depths will indicate compositions that are largely water.

So if 5-Earth mass planets turn out to be primarily water-based rather than rock-based, it’s (in my mind) an argument in favor of cranking up the data rate on Alpha Cen B. There were no structurally substantial quantities of water in the Alpha Cen planet-forming environment. If we’re seeing sub-Neptunes rather than super-Earths in the HD 40307, Gliese 581, et al. systems, then the odds are heightened that any planets orbiting Alpha Cen B are less than 2 Earth masses. There’s no payoff in tuning your Alpha Cen B strategy for sub-Neptunes. Finding truly terrestrial-mass planets will require paying full freight.

In the early nineteenth century, the detection of stellar parallax was a problem fully equivalent in both scientific excitement and prestige to the modern-day detection of the first potentially habitable extrasolar planet. I think it’s worth noting that the prize of discovery of the first stellar parallax went not to the eminently capable (but overly cautious and slow-moving) observer who accumulated data on the best star in the sky, but rather to an observer who focused on a rather obscure star in the constellation Cygnus.

Here’s a link to the article, “Thomas Henderson and Alpha Centauri” by Brian Warner of the University of Cape Town.


June 24th, 2009 Comments off

Higher resolution version.

The opportunity to travel is a splendid benefit of being an astronomer. During this week and last, I’ve been to a whirlwind of European destinations.

Stockholm was the first port of call. For someone whose life is lived at 36.974 degrees North, it is surreal to arrive in the late evening to find the Sun still well above the horizon. Night never really falls. As the hours slip through midnight, the sky merely drifts through gradations of twilight. We don’t yet have addresses for terrestrial planets beyond our solar system, but it’s certain that the galaxy is full of them. We have as yet no clues to the alien geologies, landscapes, biospheres, but the spin axes of planets tend to be tilted. The quality of midsummer twilight in the high latitudes is a phenomena shared by worlds throughout the galaxy.

I gave two talks at the Alba Nova University Center, which hosts a collaboration between astronomers, physicists and biologists, and which is mostly located in a vast award-winning building by architect Henning Larsen. The astronomy offices are arrayed along a hallway that curves for nearly a hundred meters along the top floor. Running above the doorways is a continuous printout of the solar spectrum.

All told, it contains millions of resolution elements, and an absolutely bewildering forest of absorption lines.

Even on closest inspection, each angstrom of the spectrum is smooth and full of detail.

The juxtaposition of the micro and the macro readings is dramatic. The printout also drove home the utterly tiny scale of the Doppler shifts that must be measured in order to detect planets via the radial velocity technique. A large planet such as Tau Boo b generates a radial velocity half-amplitude of 500 m/s, which corresponds to moving (and slightly stretching or compressing) the entire hundred-meter-long diagram up or down the hallway by a few dots of printer resolution. The shift caused by Gliese 581 e, on the other hand, would require a microscope to detect.


April 21st, 2009 4 comments

Woke up this morning to the startling news that the Geneva team has added an Msin(i)=1.9 Earth mass planet to the Gliese 581 system! The preprint (Mayor et al. 2009) is available from, and will appear in Astronomy and Astrophysics. With a radial velocity half-amplitude, K=1.85 m/s, Gl 581e is the lowest-mass planet detected to date.

“The orbital period of the new planet “e” is quite close to pi days. i would mark down a score of -1 for competing planet hunters, whose signals-to-noise are accumulating in proportion to the root of the number of measurements.” said Greg Laughlin, an astronomer at the University of California, Santa Cruz.

In addition to the detection of the new 1.9 Earth-mass planet in the system, the period of Gliese 581d has been revised (to great habitability fanfare) from 84 days to 66 days. Indeed, the new, shorter period raises the habitability value of Gliese 581d from about 0.5 cents to nearly one penny.

As often happens, a strong hint of the new planet was lurking unnoticed in the previously published radial velocity data, and it’s especially interesting to look at the details in this particular case to see how the period revision came about. Let’s work with the 50 radial velocities published by Udry et al. 2007.

The two strongest periodicities in the system come from planets b and c. Removing these planets with the assumption of circular orbits leaves a residuals periodogram that has its strongest peak at 84 days:

The 66 day periodicity is lagging in second place with 66% of the power. Nevertheless, both periodicities provide significant improvement to the fit. An 84-day planet has K=2.67 m/s, and leaves an RMS of 1.43 m/s to the three planet fit:

A best fit 66-day planet has a slightly higher K=2.77 m/s, but leaves an RMS of 1.72 m/s. The chi-square is also higher: 5.10 as compared to 3.65. In the 2007 data, the 84-d planet thus looked quite secure. With hindsight, though, one notices that the phase coverage in the 66-day fit is better than for the 84-day fit. As more data was obtained, it became clear that the 84-day period was an alias of the true 66-day periodicity. Fair enough — RVs are expensive to obtain, and revisions of this sort are an inevitable product of progress.

In the residuals to the fit with the 84-day planet, planet e is present, but it’s masked by a spurious periodicity at 3.45 days,

whereas in the residuals to the fit with the 66-day planet, planet-e is in the #1 spot — not yet significant, but certainly more tantalizing…

scenario two

March 22nd, 2009 3 comments

Several readers pointed out that the terrestrial planet valuation formula breaks down dramatically for Venus. Point taken! I’m not sure though, that a top-dollar Venus necessarily points to a flaw. The valuations are a quantitative measure of potential for a planet to be habitable, given only bulk physical properties currently measurable across light years of space. One is still faced with the quandry of whether to invest in to finding out whether a given planet measures up. If Venus were sheathed in water clouds rather than sulfur dioxide clouds, it would quite possibly achieve its potential as a quadrillion-dollar world.

At any rate, given its sky-high atmospheric D/H ratio, it’s not inconceivable that Venus was both habitable and inhabited, at least by microbes, in the distant past. Under the constraint of a zero-sum budget for solar system exploration, I would agitate for spending more exploring Venus and less exploring Mars.

It’s admittedly gauche to price planets like baseball cards. But it’s also true that taxpayer money, big money, well over a billion dollars of real money, is being spent to find planets, and astronomy has long since departed the ivory tower. We know from direct observation that an excitable media is more than eager to paint habitability-lottery losers in neon shades of blue and green. A middling $158.32 best-yet on a scale that will soon be registering million-dollar worlds underscores the importance of keeping the powder dry.

Which brings up scenario number two for how the first million-dollar detection (and indeed the first hundred-million dollar detection) could arise. It’s extremely likely that the first planets with genuine potential habitability will be detected from the ground. It’s also a good bet that these planets will arise from the same technique that’s produced the overwhelming majority of the big-ticket planet detections to date: Doppler radial velocity. If I were pressed to guess the particular star, I’d choose HD 40307. And if I were pressed to guess the time frame? Sometime within the next year.

The Mayor et al. (2009) HD 40307 paper rewards careful study, and indeed, may end up being as illuminating for what it reveals as for what it doesn’t reveal. In the paper, the evidence for the now-famous planets “b” (Msin(i)=4.2 M_Earth, P=4.3d), “c” (Msin(i)=6.8 M_Earth, P=9.6d), and “d” (Msin(i)=9.2 M_Earth, P=20.5d) is presented in the form of phase-folded plots of the radial velocities, and a periodogram of the velocities prior to any fitting. That all three planets are clearly visible in the raw periodogram is in itself quite remarkable. The orbits are close to circular, the system has been observed for many periods, and the signals (despite the small half-amplitudes) are unambiguous:

The actual radial velocities, however, are not included in the paper, and would-be Dexterers are thwarted by the fact that the only plots showing the full data set are phase-folded. The journal version of the paper reports that the velocities are available at: , but the link is still empty…

In lieu of access to the actual data, we have carte blanche to engage in irresponsible (yet technically accurate) speculations to get a sense for what further secrets the HD 40307 system might harbor. Let’s construct a Monte-Carlo data set. An optimally habitable ten million-dollar planet in the HD 40307 system has a mass of ~2.3 Earth masses, an orbital period of 141 days, and induces a K=0.35 m/s radial velocity half-amplitude. We can make a model system that includes such a planet along with the three known planets (noting that the Mayor et al. 2009 paper contains an error for K_d in Table 2). We can generate a synthetic radial velocity data set by perturbing the four-planet model with the reported 0.32 m/s instrumental measurement error and 0.75 m/s of Gaussian stellar jitter, and observing at 135 randomly spaced times within a span of 1628 days.

We can put the resulting data set into the Systemic Console. Removing the 20-day planet gives a residuals periodogram that clearly shows the 9.6d and 4.3d planets, along with an alias peak at ~2 days. As with the actual periodogram in the Mayor paper, there’s nothing particularly interesting at 141 days. That is, there’s no sign of the ten million-dollar world that was baked into the data.

Remarkably, however, when the 9.6d and 4.3d planets are fitted and removed, the periodogram peak for the 141d planet is quite prominent. It’ll be very interesting to see if anything like this is present in the actual data set when it goes online:

It’s straightforward to recover the 141d planet in the orbital fit. Removing the three known inner planets and phase-folding the data at the period of the 141d planet shows what its current (as of last June) signature would look like:

A real planet with these properties would thus be right on the edge of announceability. HD 40307, furthermore, is by no means the only quiet Mv~7 K dwarf in the local galactic neighborhood…