≡ Menu

Exoplanet Prospects at Earth-based Observatories

Although I often write about upcoming space missions that will advance exoplanet research, we’re also seeing a good deal of progress in Earth-based installations. In the Atacama Desert of northern Chile, the Extremely Large Telescope is under construction, with first light planned for 2024. With 256 times the light gathering area of the Hubble instrument, the ELT is clearly going to be a factor in not just exoplanet work but our studies of numerous other astronomical phenomena, from the earliest galaxies in the cosmos to the question of dark energy.

Today we learn that the first six hexagonal segments for the ELT’s main mirror have been cast by the German company SCHOTT at their facility in Mainz, Germany. We’re just at the beginning of the process here, for the primary mirror is to be, at 39 meters, the largest ever made for an optical-infrared telescope. 798 individual segments — each 1.4 meters across and 5 centimeters thick — will go into it, working together as a single gigantic mirror.

Image: The first six hexagonal segments for the main mirror of ESO’s Extremely Large Telescope (ELT) have been successfully cast by the German company SCHOTT at their facility in Mainz. These segments will form parts of the ELT’s 39-metre main mirror, which will have 798 segments in total when completed. The ELT will be the largest optical telescope in the world when it sees first light in 2024. Credit: ESO.

SCHOTT will embark on making a total of 900 segments, 798 for the primary mirror plus a spare set, with production rates when up to speed of about one segment per day. After cooling and a heat treatment sequence, the mirror segment blanks will be ground and polished to a precision of 15 nanometers, with shaping and polishing performed by the French company Safran Reosc, which will mount and test the individual segments.

Binocular Vision

Meanwhile, we also get word of two new instruments that will be mounted on the Large Binocular Telescope (LBT), located on Mount Graham in Arizona. The SHARK instruments (System for coronagraphy with High order Adaptive optics from R to K band) are designed with an explicit exoplanet purpose, to conduct direct imaging of distant worlds.

What makes the SHARK effort intriguing is that it comprises two instruments. SHARK-VIS works in visible light, SHARK-NIR in near-infrared, and on the LBT platform, the two will be operated in parallel, using the two 8.4-meter mirrors of the observatory. Built by an international consortium led by INAF, the Italian National Institute for Astrophysics, the two SHARK instruments will likewise take advantage of the observatory’s adaptive optics system, also developed by INAF. Adaptive optics corrects for distortions caused by turbulence in the Earth’s atmosphere.

Image: Each SHARK will be installed on one side of the LBT Interferometer (LBTI), the green structure seen in the middle of the picture between the two main mirrors of LBT. Credit: SHARK Consortium/INAF.

Notice what’s happening here: The LBT, once equipped with the two SHARK add-ons, becomes the first telescope in the world that can observe exoplanets simultaneously over such a wide range of wavelengths, helping astronomers tease out planets that would otherwise be drowned in the glare of the host star. The installation, which is expected to be completed in 2019, also points the way toward the upcoming Giant Magellan Telescope, which will deploy seven 8.4-meter mirrors on the same mount instead of the LBT’s two. The GMT facility is under construction at the Las Campanas Observatory in Chile’s southern Atacama Desert.

“With SHARK, we will observe exoplanets at unprecedented angular resolution and contrast, so that we will be able to go closer to their host stars than what has been achieved up to now with direct imaging,” says Valentina D’Orazi of the INAF-Osservatorio Astronomico di Padova, instrument scientist for SHARK-NIR. “This will be possible thanks to the use of coronagraphy, which blocks out the light from the central star and highly improves the contrast in the region around the source, thus allowing us to detect the planetary objects we want to study, which otherwise would remain hidden in the star light.”

Clearly we’re moving into an era where Earth-based observatories will be capable of major advances in the exoplanet hunt, complementing the upcoming space missions that will expand the planetary census and begin the analysis of smaller exoplanet atmospheres, particularly those around red dwarf stars. Both the Extremely Large Telescope and the Giant Magellan Telescope could be completed, if current schedules are realistic, by 2025.

An Ancient Planetary System

I just noticed that the team behind PEPSI (Potsdam Echelle Polarimetric and Spectroscopic Instrument) at the Large Binocular Telescope has released three papers analyzing high spectral resolution data from the site. Because I’ve only had the chance to skim the papers, let me just quote the news release on one of these, examining the 10-billion year old system Kepler-444:

…the star “Kepler-444”, hosting five sub-terrestrial planets, was confirmed to be 10.5 billion years old, more than twice the age of our Sun and just a little bit younger than the universe as a whole. The star is also found being poor on metals. The chemical abundance pattern from the PEPSI spectrum indicates an unusually small iron-core mass fraction of 24% for its planets if star and planets were formed together. For comparison, terrestrial planets in the solar system have typically a 30% iron-core mass fraction. “This indicates that planets around metal-poor host stars are less dense than rocky planets of comparable size around more metal-rich host stars like the Sun”, explains Claude “Trey” Mack, project scientist for the Kepler-444 observation.

The paper is Mack et al., “PEPSI deep spectra. III. A chemical analysis of the ancient planet-host star Kepler-444,” in press at Astronomy &Astrophysics (preprint).


Comments on this entry are closed.

  • DJ Kaplan January 9, 2018, 14:32

    Exciting to hear about instruments to go online in the near future!
    Of course, even the most recent, such as ESPRESSO on the VLT array, will take months before they can share verified data from exoplanet observations.

  • Geoffrey Hillend January 9, 2018, 16:58

    I read that star age is determined by radiometric spectroscopy where they look at the decay ratio and chain of radioactive isotopes just like in radiometric dating. Their being metal poor also determines age. Very interesting that metal poor stars indicate metal poor planets which might help to rule life beginning there; one needs a metal core for a magnetic field.

    • Peter Alcivar January 9, 2018, 23:07

      Remember that the astronomical definition of “metal” in a star is not the same as metal in a planet. Metal in a star is elements heavier than helium.

  • P January 9, 2018, 20:16

    The ELT seems like such a massive technological leap to actually get working at anything like the diffraction limit, I wouldnt be surprised if there are delays getting it up and running properly. My $$$ is on the GMT. Not that it doesnt have challenges, but it seems like a more conservative extension of existing capacities. Anyway, good luck to all of these new instuments, I dont doubt we are on the cusp of a revolution greater than the arrival of Hubble and 1st generation adaptive optics.


  • Michael Fidler January 9, 2018, 23:56

    Interesting results coming from the American Astronomical Society (AAS) meeting in National Harbor, Maryland:

    The long and short of it: Iron-rich stars host shorter-period planets.

    “Using SDSS data, they found that stars with higher concentrations of iron tend to host planets that orbit quite close to their host star — often with orbital periods of less than about eight days — while stars with less iron tend to host planets with longer periods that are more distant from their host star”.

    Image: http://www.sdss.org/press-releases/wp-content/uploads/2018/01/exoplanets_sdss_text_sm.jpeg

    And from Keck and Kepler:

    Planets around Other Stars are like Peas in a Pod.

    “Using a statistical analysis, the team found two surprising patterns. They found that exoplanets tend to be the same sizes as their neighbors. If one planet is small, the next planet around that same star is very likely to be small as well, and if one planet is big, the next is likely to be big. They also found that planets orbiting the same star tend to have a regular orbital spacing.

    “The planets in a system tend to be the same size and regularly spaced, like peas in a pod. These patterns would not occur if the planet sizes or spacing were drawn at random,” explains Weiss.

    This discovery has implications for how most planetary systems form. In classic planet formation theory, planets form in the protoplanetary disk that surrounds a newly formed star. The planets might form in compact configurations with similar sizes and a regular orbital spacing, in a manner similar to the newly observed pattern in exoplanetary systems.

    However, in our solar system, the inner planets have surprisingly large spacing and diverse sizes. Abundant evidence in the solar system suggests that Jupiter and Saturn disrupted our system’s early structure, resulting in the four widely-spaced terrestrial planets we have today. That planets in most systems are still similarly sized and regularly spaced suggests that perhaps they have been mostly undisturbed since their formation”.


    The California-Kepler Survey. V. Peas in a Pod: Planets in a Kepler Multi-planet System Are Similar in Size and Regularly Spaced*


    The California-Kepler Survey. IV. Metal-rich Stars Host a Greater Diversity of Planets.


    Lower density, more hydrocarbons and water greater distance from star.
    Sounds like the early norm in the galaxy may be the most habitable. Could it be that the earth is an example of megaengineering of disrupted later solar systems by the early multi-planet peas in a pod type civilizations? So we may be a salvaged solar system???

    • Michael Fidler January 10, 2018, 10:33

      From her website with a very good image:

      Planets Form in Patterns like Peas in a Pod.

      “I just discovered that, among 355 NASA Kepler planetary systems, planets in the same system tend to have similar sizes and regular orbital spacing. For example, if one planet in the system is twice the radius of the Earth, the other planets in that same system are likely to be about two Earth radii as well. In short, they look like peas in a pod. A figure illustrating this pattern in systems with four or more transiting planets is shown below”.



      This image conveys the orbital configurations of the California Kepler Survey systems with at least four transiting planets. Each row corresponds to one planetary system (name on y-axis) and shows the planet orbital distances in Astronomical Units (x-axis). The point sizes are proportional to the planet sizes, and the point colors correspond to the estimated planet surface temperatures (see key and color bar). The systems are ordered by stellar mass, which is listed to the right of each system. The inner solar system (“SOL”) is included for comparison. In many systems, the sizes and separations of the planets provide a good prediction of the sizes and separations of the rest of the planets.

      • Ronald January 11, 2018, 10:07

        Fascinating material, some conclusions from the paper:

        1. Planets within a given system are more similar in size than planets drawn randomly from the collection of multi-planet systems.

        2. Stellar mass is a not strongly correlated with planet sizes. Thus, some property of the proto-planetary disk other than stellar mass influences the final planet sizes (!).

        3. When adjacent planets in a multi-planet system are not similar in size, the inner planet is smaller in 65% of cases.

        4. Planets have a regular geometric spacing, with a typical ratio of 1.5 between the semi-major axes of adjacent planets.

        5. Planets tend to be about 20 (10-30) mutual Hill radii apart, rarely closer than 10 mutual Hill radii. Their dynamical packing is similar to the solar system.

        The regular spacing of planets might be evidence that planets in systems with multiple transiting planets have been relatively undisturbed since their formation. Perhaps these systems formed in situ, at orbital distances determined by their feeding zones, and grew to their present sizes based on the available material.

    • Michael Fidler January 11, 2018, 0:06

      Scientific American

      You Live In a Strange Solar System.

      Astronomers found that other star systems tend to host similarly sized exoplanets—far different from ours. Christopher Intagliata reports.

      Full Transcript
      The more astronomers study the heavens, the more they realize: our solar system is weird.

      “There are a few things that make the solar system kind of strange.” Lauren Weiss, an astrophysicist at the University of Montreal. “One of which is we have a giant planet. Only about 10 percent of sunlike stars have a giant planet. And there are probably even fewer that have two giant planets.”

      In addition to giant Jupiter and lesser giant Saturn, we have tiny Mercury—just a bit bigger than Earth’s moon.

      So if we’re weird, what does a typical solar system look like? Weiss and her team trained their telescopes on 355 star systems known to host a handful of small exoplanets. And they found that most of the planets within individual star systems tended to be similar in size.

      “So if I’m a planet, and I’m, say, two times the size of Earth, my neighbor, the next planet over, is also likely to be two times the size of Earth, give or take a little bit.”

      And they were strung out at similar distances from each other too…like peas in a pod, she says. Compared to that orderly array, our system looks more like, “Uh let’s see, if I stick with food…I don’t know…like a whole Thanksgiving dinner or something?”

      The results are in The Astronomical Journal. [Lauren M. Weiss et al, The California-Kepler Survey. V. Peas in a Pod: Planets in a Kepler Multi-planet System Are Similar in Size and Regularly Spaced]

      As for hunting for habitable worlds: “If we’re trying to find an Earth-sized planet in the habitable zone”—not too close to the star but not too far away either—“and we find an Earth-sized planet closer in, it might be worthwhile to continue searching for planets around that star.”

      Because there might just be a few more peas in the pod.

      —Christopher Intagliata

      [The above text is a transcript of this podcast.

    • Ronald January 11, 2018, 10:55

      And the paper IV about metallicity and planet sizes is also very interesting, especially in combination with the other paper V (about similar sized planets in the same system and at regular spacing).

      The occurrence of planets with larger sizes or shorter orbital periods is correlated with metallicity, and this correlation steepens with decreasing orbital period and increasing planet size.

      In summary, for orbital periods < 100 days (hot/warm), the default planetary system contains either no planets detectable by Kepler (such as our Solar System) or a system of one or more super-Earths/gas dwarfs. The latter seems to be correlated with higher metallicity. High metallicity is also correlated with the occurrence of hot/warm gas giants and Neptunes.

      Hot Jupiters are strongly correlated with metallicity, though rare (only 0.6% of Sun-like stars).

      For warm terrestrial planets/super-Earths (P 10 -100 days, R 1-1.7 Re), there is no correlation. Terrestrial planets/super-Earths may be a population of planets that form with high efficiency, even in metal-poor disks.

  • Harry R Ray January 10, 2018, 10:52

    The Hypatia Stone has just stolen Oumuamua’s thunder BIG TIME!!! The most recent experiments on this object previously classifies as a comet fragment have reveled TWO STARTLING CONCLUSIONS! ONE: The chemical mixture bears NO SIMILARITY AT ALL to solar comets, and the carbon to silicon ratios best fit an object that formed around a carbon rich star! TWO: There is a VERY GOOD CHANCE that the formation of the stone PREDATED the formation of the solar system by a CONSIDERABLE AMOUNT OF TIME!

    • ljk January 10, 2018, 13:55

      Here are the details:


      To quote:

      “What we do know is that Hypatia was formed in a cold environment, probably at temperatures below that of liquid nitrogen on Earth (-196 Celsius). In our solar system it would have been way further out than the asteroid belt between Mars and Jupiter, where most meteorites come from. Comets come mainly from the Kuiper Belt, beyond the orbit of Neptune and about 40 times as far away from the sun as we are. Some come from the Oort Cloud, even further out. We know very little about the chemical compositions of space objects out there. So our next question will dig further into where Hypatia came from,” says Kramers.

      The little pebble from the Libyan Desert Glass strewn field in south-west Egypt presents a tantalizing piece for an extraterrestrial puzzle that is getting ever more complex.

  • Ronald January 11, 2018, 7:03

    Great developments. I wonder, however, if and to what extent these earth-based giant telescopes can substitute for space-based telescopes, even with adaptive optics: how about IR, isn’t that largely absorbed by Earth’s atmosphere?
    It would be great, if space-based telescopes could be made largely redundant, since the earth-based ones are much cheaper (especially maintenance and upgrading) and can be made much larger.
    Question: could we, theoretically at least, do most or all research on small terrestrial planets (in particular with regard to spectroscopy and bio-signatures) with earth-based telescopes?

    • Alex Tolley January 11, 2018, 15:12

      IR, UV, gamma-rays, X-rays all really need space-based telescopes. I’m old enough to remember when high altitude balloons and rockets provided early data in these regimes. A friend at university was on a team doing gamma-ray astronomy with high altitude balloons. Today he would no doubt be working on building a Cubesat to do similar work.

      Economics and capability determine where experiments should be done. If we had cheap access to space, then the economics might start to favor large space based telescopes.

  • ljk January 11, 2018, 14:57
  • ljk January 11, 2018, 15:37


    Citizen Scientists Discover Five-Planet System

    Caltech staff scientist Jessie Christiansen is a founder of a citizen-scientist project called Exoplanet Explorers


  • Harry R Ray January 12, 2018, 10:53

    ANOTHER binary planet discovered, This time in the Orion Nebula! Just an abstract at the AAAS meeting in Washington DC this week, so details are sketchy at this time, and thus; no way to test my theory that ALL binary planet components are of EQUAL MASS to their partners. STAY TUNED!

  • Harry R Ray January 24, 2018, 11:02

    ExTrA is now FULLY OPERATIONAL and making its first observations1 ExTrA fills in the gap between Mearth, which observes M0 to M4.5 stars, and TRAPPIST-SPECULOOS, which observes M6 to M10(and MAYBE some MAIN SEQUENCE L stars. This is HUGE, because the observatory ExTrA is housed in is La Silla, which can EASILY observe Proxima Centauri, for which there have been a couple of RECENT claims of POSSIBLE Earth-sized planet transit CANDIDATES(BOTH Proxima B AND a slightly smaller and closer orbiting planet). ExTrA should be able to prove the WORTHINESS of these candidates WITHIN A YEAR, WELL BEFORE ANY analysis of ANY TESS observations.

  • Harry R Ray January 24, 2018, 11:32

    An ill wind INDEED! CoRoT 2b’s winds blow east to west instead of west to east! Hunh???

  • ljk January 25, 2018, 12:40

    Upcoming Telescopes Should be Able to Detect Mountains and Other Landscapes on Extrasolar Planets


    “Finding Mountains with Molehills: The Detectability of Exotopography” here:


  • Harry R Ray February 2, 2018, 10:59

    “Probing Planets in Extragalactic Galaxies using Quasar Microlensing.” by Xinyu Dai, Eduardo Guerras. TWO THOUSAND NEW PLANETS detected in the host galaxy of quasar RXJ 1131-1231 3.8 billion light-years away?