≡ Menu

The Era of Planet Gathering

We’ve looked at a couple of exoplanet issues this week that bear further comment. The first is that different detection methods can be usefully combined to cover different scenarios. If radial velocity works best with larger planets closer to their star, direct imaging takes us deep into the outer planetary system. We saw yesterday how both imaging and radial velocity could be used to probe subgiant stars. We routinely use RV as a check on transiting planet candidates. And gravitational microlensing can find planets at a wide range of separation from their primary.

I think microlensing has plenty to teach us, though I’m sensitive to the criticism voiced in comments here that we’re dealing with non-repeating events when we have a microlensing detection. Centauri Dreams reader coolstar has also noted that distance may be a factor, questioning whether some of the resources by way of telescope hardware that we’re putting into microlensing studies wouldn’t be better employed looking at nearby worlds. After all, a new paper by Timothy Morton and Jonathan Swift makes the case that small planets around M-dwarfs are even more plentiful than recent studies reported here have shown. Maybe we should be putting more of an effort into finding these through transit studies?

If different detection methods each have their uses, radically different classes of telescope can likewise probe for planets. John Johnson (Caltech) heads up Project Minerva, a telescope array dedicated to Earth-like planets around nearby stars. The plan is to use small-aperture robotic telescopes atop Palomar Mountain to perform photometry and high-resolution spectroscopy, making it, according to Caltech’s site on the project, “the first U.S. observatory dedicated to exoplanetary science capable of both precise radial velocimetry and transit studies.” Following an initial design review, the project has now seen delivery of the enclosure for its telescopes.


Image: Minerva’s hardware. Minerva will be an array of small-aperture robotic telescopes to be built atop Palomar Mountain outfitted for both photometry and high-resolution spectroscopy. It will be the first U.S. observatory dedicated to exoplanetary science capable of both precise radial velocimetry and transit studies. The multi-telescope concept will be implemented to either observe separate targets or a single target with a larger effective aperture. The flexibility of the observatory will maximize scientific potential and also provide ample opportunities for education and public outreach. The design and implementation of MINERVA will be carried out by postdoctoral and student researchers at Caltech. Credit: Caltech/John Johnson.

TESS, the Transiting Exoplanet Survey Satellite, stands as a logical follow-on to the Kepler mission because it takes the hunt to nearby stars even as Kepler’s candidate totals push on toward 3000. But as we wait for new instruments to come online, including the upcoming James Webb Space Telescope and ground-based systems like the European Extremely Large Telescope, it’s great to see grad students like Johnson’s Exolab crew coming up with solutions like Minerva. It’s an inexpensive effort that could gain real traction, and as Johnson told one reporter, “If we lived in an ideal world, we wouldn’t do Minerva because we’d have money from our funding agencies.” But then, who ever said it was an ideal world?

Minerva had a poster devoted to it at the most recent meeting of the American Astronomical Society. The plan is to use an array of four instruments, each with a 0.7-meter mirror, that will over a period of three years scan stars within about 75 light years of the Earth. To get a look at it, see Unblinking Baby Telescopes Will Hunt Exoplanets for Cheap, a recent piece that explains the project this way:

The Minerva telescopes will look for a slight wobble that indicates a planet is tugging gravitationally on those stars. Using this data, [graduate student Kristina] Hogstrom said that her team expects to find around a dozen new local planets, most of them two to three times the size of Earth and a few of them orbiting in the habitable zone where liquid water could exist. The project will also find Earth-sized planets and perhaps smaller, but these will likely orbit too close to their parent star to host life.

What I like about this, and I can see why Centauri Dreams readers find it interesting, is that it’s a dedicated exoplanet project that, Kepler-like, stares at a designated number of stars. The article quotes Johnson as saying that astronomers aren’t really in planet ‘hunting’ mode these days so much as engaged in routine planet ‘gathering,’ which tells you how far we’ve come in this remarkable enterprise in the few short years since 1995, when the announcement of the first planet discovered orbiting a main sequence star was made by Michel Mayor and Didier Queloz, to be quickly confirmed by Geoff Marcy and Paul Butler.

So here we are with off-the-shelf hardware and a custom-built spectrometer, running up a tab of about $3.5 million as compared to the $600 million that the relatively cheap space-based Kepler cost. We’re probably less than a year away from the datastream beginning to flow at the Minerva site. Modest funding and limited resources used imaginatively in arrays of small dedicated telescopes may be forced upon us by circumstance, but we’re learning that good science can flow from such projects and private institutions can find ways to fund them.


Comments on this entry are closed.

  • CatharSeamus April 11, 2013, 11:46

    I’ve always wondered if it is possible to make a distributed network of telescopes run by hobbyists, standarizing on some mountings, and controlled by a network director, such that a high-definition image is gathered from many images of poor resolution. Something like interferometry would probably be impossible in such distributed way and also being tolerant to failure or uncertainty in the nodes. No idea if statistical analysis of such images can make a better image with a resolution that scales with the number of nodes.

    Anyone knows research in this regard?

  • Wojciech J April 11, 2013, 14:44

    In all information we shouldn’t forget about less visible space programs, according to this article China and Russia are also starting to get interest in Exoplanet reasearch(hmm I wonder if desire to find another Earth with biosignatures could spark a “telescope race”, similar astronomy related things happened in XIX century between powers)
    “China revealed that it is planning to build the SAT, the Satellite Astrometry Telescope. Not much is known about its design, but indications are that it will be capable of exoplanet searches, possibly using an interferometer. In Russia, Roscosmos has published their outline for space projects for the coming decade. Funding for the Russian space program is erratic, so it is difficult to determine the reality of this outline. However, the Osiris (Astrometria) project is still on their to-do list. It is an ambitious undertaking, considering their track record, but Osiris is proposed to be a high-accuracy interferometric astrometry craft, perhaps resembling NASA’s Space Interferometry Mission (SIM). Its current launch date is 2018, but this may be optimistic considering the current chaos in Russia’s deep space program.”
    Considering the scale of the newest Chinese radio telescope, the Chinese could still surprise us…

  • Wojciech J April 11, 2013, 14:48

    Here is a scarce info on proposed Chinese project that I was able to find on the net

  • CatharSeamus April 11, 2013, 15:12

    Probably we could do it with radio and microwave frequencies, but i doubt we will be able to measure phase directly in infrared or optical channels in the foreseeable future.

    Does the radio and microwave spectrum gave enough information that is still relevant for RV and direct imaging approaches of exoplanet search?

  • FrankH April 11, 2013, 15:17

    One advantage of doing planet searches from Earth with small telescopes is that the site requirements for photometry and spectroscopy are less stringent than for high quality imaging. As long as you can gather enough light to get a significant signal over noise (to see a transit dip that may be much less than 1%), you’re almost there. You want a site with a steady atmosphere (to keep the star from bouncing around the spectroscope slit or smearing an image) and somewhat dark skies (less important if you’re observing in the IR). Adaptive optics can correct for a less than steady air column at a price, so even that’s not that big of a technical issue.
    This opens up a huge number of observing sites around the world; many college telescopes are probably up to the task. The main issue is funding, not technology.

    @ CatharSeamus – optical interferometry is pretty much impossible if you don’t have line of sight between the telescopes, since you have to gather and combine the light from all the telescopes while keeping their separations within very stringent limits. It’s a lot easier to do in microwave or radio wavelengths.
    If you want a large optical interferometer, you have to put it in space.

  • coolstar April 11, 2013, 16:32

    @CatharSeamus Because the wavelengths are so much shorter in the optical than in the radio (factor of about a billion), interferometry is MUCH more difficult in the optical. Still, there have been successes, the CHARA array at Mt. Wilson and the NPOI array on Anderson Mesa (Lowell Obs) outside of Flagstaff come to mind in the states.

  • madscientist April 11, 2013, 18:37

    At $3.5 million that’s not too bad, considering the four Planewave CDK700’s they look like they’ll be using are only going to cost them $740,000 (at $185,000 per scope). The rest of the money will go into the spectroscope and the infrastructure for the scopes (site purchase, domes, cabling, computers etc). Most will probably go into the spectroscope, since it will be custom built and needs to be of high accuracy.

    Mind you, if they happen to have a “spare” CDK700 lying around after all their observatory construction, I’ll be happy to take one off their hands!!! :) :P

  • Eniac April 11, 2013, 20:39


    No idea if statistical analysis of such images can make a better image with a resolution that scales with the number of nodes.

    Anyone knows research in this regard?

    I think what you are looking for is intensity interferometry. It is an old technique that was long used to infer stellar diameters, but then fell into disuse. It has recently been dragged back out with the potential of doing just exactly the kind of imaging you allude to, courtesy modern information processing capabilities. See here:


    and references therein.

  • CatharSeamus April 12, 2013, 15:52

    for millimeter range and above wavelengths, we can sample wave phase with appropiate electronics. This could enable software interferometry, so, line-of-sight would not be needed, only an internet connection.

  • Sedjak April 15, 2013, 15:02

    It would seem that an interferometry array of even smaller telescopes, e.g. 20 cm, would be quicker to deploy and less costly than using these relatively large .7 meter telescopes since we would be focusing on nearby relatively bright stars anyway. Not knowing the relationship of the primary diameter with regard to the intricacies of doing astrometry, it seems intuitive enough that a long enough baseline in an appropriate formation could give precise astrometrics relative to stars that are already characterized. Am I missing something here?