≡ Menu

Roman Space Telescope: Planets in the Tens of Thousands

The Nancy Grace Roman Space Telescope is the instrument until recently known as WFIRST (Wide-Field Infrared Survey Telescope), a fact I’ll mention here for the last time just because there are so many articles about WFIRST in the archives. From now on, I’ll just refer to the Roman Space Telescope, or RST. Given our focus on exoplanet research, we should keep in mind that the project’s history has been heavily influenced by concepts for studying dark energy and the expansion history of the cosmos. The exoplanet component has grown, however, into a vital part of the mission, and now includes both gravitational microlensing and transit studies.

We’ve discussed both methods frequently in these pages, so I’ll just note that microlensing relies on the movement of a star and its accompanying planetary system in front of a background star, allowing the detection because of the resultant brightening of the background star’s light. We’re seeing the effects of the warping of spacetime caused by the nearer objects, with the brightening carrying the data on one or more planets around the central star. You can see why a target-rich environment is needed here — such occulations are random and lots of stars are needed to cull out a few. Thus RST looks toward galactic center.

Assuming a successful launch and operations beginning in the mid-2020s, the RST should be a prolific planet-hunter indeed. NASA is now projecting, based on a 2017 paper from Benjamin Montet (now at the University of New South Wales) and colleagues, that the anticipated light curves for millions of stars should turn up as many as 100,000 planets. Montet et al. actually cite as many as 150,000, noting that the WFIRST field is more metal rich than the main Kepler field. All of these systems should have measured parallaxes, though most will be too faint for follow-ups.

Even so, the Montet paper finds ways to confirm some of these:

We find that secondary eclipse depth measurements can be used to confirm as many as 2900 giant planets, which can be detected at distances of > 8 kpc. From these confirmed WFIRST planets, we will be able to measure the variation in the occurrence rate of short-period giant planets. Furthermore, we show that WFIRST is capable of detecting TTVs which can be used to confirm the planetary nature of some systems, especially those with smaller planets.

This is particularly fruitful — consider that the transiting planets that RST finds will in most cases not be found around the same host stars as the planets found through microlensing. The two methods enable separate probes of the same population of planets but at different separations, with transits being most detectable for close-in planets and microlensing being the method to detect planets much further out in the system. We go from star-hugging hot Jupiters to planets beyond 10 AU. Microlensing should also turn up free-floating ‘rogue’ planets.

The Kepler mission studied stars in an area encompassing parts of Lyra, Cygnus and Draco, most of them ranging from 600 to 3,000 light years away. With RST, we will be going from Kepler’s 115 square degree field of view of relatively nearby stars to a 3 square degree field that, because it is toward galactic center, will track up to 200 million stars. The average distance of stars in this field will be in the range of 10,000 light years in what will be the first space-based microlensing survey, looking at planets all a wide range of distance from the host star.

Image: This graphic highlights the search areas of three planet-hunting missions: the upcoming Nancy Grace Roman Space Telescope, the Transiting Exoplanet Survey Satellite (TESS), and the retired Kepler Space Telescope. Astronomers expect Roman to discover roughly 100,000 transiting planets, worlds that periodically dim the light of their stars as they cross in front of them. While other missions, including Kepler’s extended K2 survey (not pictured in this graphic), have unveiled relatively nearby planets, Roman will reveal a wealth of worlds much farther from home. Credit: NASA’s Goddard Space Flight Center.

The synergy between microlensing and transit work is clear. Jennifer Yee, an astrophysicist at the Center for Astrophysics | Harvard & Smithsonian, notes that thousands of transiting planets are going to turn up within the microlensing data. “It’s free science,” says Yee. Benjamin Montet agrees:

“Microlensing events are rare and occur quickly, so you need to look at a lot of stars repeatedly and precisely measure brightness changes to detect them. Those are exactly the same things you need to do to find transiting planets, so by creating a robust microlensing survey, Roman will produce a nice transit survey as well.”

So we’ve gone from the relatively close — Kepler with stars at an average distance of 2,000 light years — to the much closer — TESS, with its scans of the entire sky focusing in particular on stars in the range of 150 light years — and now to RST, which backs out all the way to galactic center, a field encompassing stars as far as 26,000 light years away. NASA estimates that three-quarters of the transits RST detects will be gas or ice giants, with most of the rest being planets between four and eight times as massive as Earth, the intriguing ‘mini-Neptunes.’ For its part, microlensing takes us down to rocky planets smaller than Mars and up to gas giant size.

The Montet paper is “Measuring the Galactic Distribution of Transiting Planets with WFIRST,” Publications of the Astronomical Society of the Pacific Vol. 129, No. 974 (24 February 2017). Abstract.

{ 26 comments… add one }
  • Dwight Williams April 5, 2021, 9:15

    I feel obligated to speak in defence of Gaia’s planet-hunting mission component, covering a sphere reaching out at least 10,000 light-years. Or is this thread focused solely upon NASA projects, rather than including other space exploration services’ work?

    • Paul Gilster April 5, 2021, 17:09

      This thread in particular is on the RST. But news from Gaia is always on topic here as well, and you’ll find articles about it in the archives (with more to come).

      • Ashley Baldwin April 7, 2021, 13:52

        Battle of the fallacies versus forensic analysis ?

        If the RST can indeed deliver what amounts to a representative survey of all planets from those of Mars mass within a faction of an AU from a star all the way up to multiple Jupiter mass giants out to many AUs then the numbers will finally be indicative. Versus the current McNamara’s fallacy represented by the close in and large exoplanetary types heavily favoured by Doppler photometry and time series transit spectroscopy. This fallacy being count what can easily be counted and ignore or even dismiss that which can’t . Such as planets smaller than mini Neptune’s or Super ( duper) Earths . I have a hunch that ultimately Earth mass terrestrial planets will turn out to be the most common .

        But in turn I worry about bandying exaggerated 100k figures about too. Years before time . If measurements become targets then they cease to be useful and become fallacious too. This is what happened with Gaia’s projected gas giant astrometry haul which also maxed out at around 100k for a ten year extended mission – before falling back . Alot. But hopefully drawn by the pull of practicality . And reality . .

        Always beware round numbers – especially big ones.

        • Paul Gilster April 7, 2021, 19:27

          Interesting point, and interesting lesson from Gaia. Can you elaborate a bit on the fallback in the astrometry figures? Thanks.

  • DCM April 5, 2021, 12:10

    Does the increasing number of known planets give us cause for joy or more anxiety about invasion…..

    • Michael April 6, 2021, 8:03

      Always anxiety, it keeps our film industry going.

      • DCM April 6, 2021, 13:40

        Excellent observation.
        I wonder when some bizarre kind of guilt is going to possess many — because we may somehow be hurting possible aliens by looking for them even if they don’t know it and we can’t see them.

  • Alex Tolley April 5, 2021, 13:27

    A millennium ago we knew of just 5 extraterrestrial planets, all in out system.

    With the invention of the telescope 400 years ago we were able to find 3 (now 2) more planets in our system but were ignorant whether other stars had planets.

    30 odd years ago we found the first extrasolar planet, and the hunt ratcheted up to find more in the single digits.

    Within the last decade, the Kepler space telescope increased the number of confirmed extrasolar planets to single-digit thousands, around 2 orders more than before. We can no longer give them familiar names at all.

    How long before future telescopes detect near unambiguous biosignatures for thousands of exoplanets in the HZ of their star?

    With RST, the estimate is that the number of planets may increase by another 2 orders to 100’s of thousands, more than a fleet of Starship Enterprises could explore in many generations.

    In the novel 2001: A Space Odyssey, Clarke noted that there were so many habitable (and inhabited) planets in the galaxy, that galactic survey ships could only visit each one once, never to return again. They just left sentinels to keep track of their experiments to develop higher intelligence.

    How long before we have the capability to send probes to all the planets with strong biosignatures to monitor them? Within the next millennium?

    • Robin Datta April 6, 2021, 2:27

      Some say that we can access heretofore unacknowledged intelligence with
      perhaps a shovel or even a trowel.

      • Andrei April 9, 2021, 14:48

        When we who work with the biological sciences talk about ‘intelligence’ we’re referring to the ability of the specimen to communicate or the problem solving ability.
        This does not mean that such a species actually got the ability of reasoning, even less consider to start writing block buster novels.
        My own research have shown that a number of vertebrate species species with very small brains actually can have social systems and communication – which was new, but of course previously known from ants, bees etc.
        As this was new we did indeed label this as a level of intelligence. But that still a far cry from them starting to write literature or become tool users. It’s entirely instinct driven. But we use the term to describe a characteristic that is beneficial for the species in question.

    • Hamilton1 April 6, 2021, 6:36

      But we already know that none of those planets have taken control of this planet. So whatever we discover by our monitoring, is of comparatively lesser importance. The Fermi paradox and all that…

  • Wojciech Jacyk April 5, 2021, 16:22

    Just learned about proposed Pandora mission:a SmallSat telescope to characterize exoplanet atmospheres.


    • Paul Gilster April 5, 2021, 17:07

      More on this one is coming up. Pandora is quite interesting!

    • Andrei April 9, 2021, 14:55

      Lets hope the mission does not open the Pithos of Pandora (or box as it’s popularised to have been.)
      Joking, someone claimed the name is taken from the film Avatar – though the female in the classical story was said to be a very curious person.

  • James M Essig April 5, 2021, 16:32

    This is great news. While many folks conjecture on hyperspatial travel, or an infinite universe, we lose track of what is in our cosmic light-cone. Telescopes like the Roman Space Telescope will continue to shed light on the awesome scope of our cosmic light-cone which likely is just a tiny portion of our universe.

    Within our current cosmic light cone, there are an estimated 10 EXP 24 stars. This number is equivalent to the number of grains of fine-grain table sugar that would cover the entire United States one hundred meters deep. The number of planets is estimated to be roughly ten times greater or about the number of fine grains of table sugar that would cover the entire United States one thousand meters deep. The number of planetary moons is estimated to be about ten times greater or equal to the number of grains of fine-grain table sugar that would cover the entire United States ten thousand meters deep. All these worlds are available for us to explore over the next ten billions years or so. We need nothing more than moderately relativistic spacecraft to do so. However, the closer we approach light speed, the lesser mission time duration is needed for the crew.

  • Geoffrey Hillend April 5, 2021, 19:04

    I was thinking the same thing as Alex Tolley. Wondering what is the range limit of our spectroscopy and how many light years can we increase it in the near future especially if we don’t find any biosignatures locally. I still have to think about a star shade of the JWST. Even if one only lasts a year or two.

    • Ashley Baldwin April 6, 2021, 14:58

      Lots of reasons that isn’t going to happen I’m afraid . Not least because of the JWST’s monster overspend . The telescope wouldn’t need much adapting to work with a star shade. Though not mooted as costing a fortune – yet more cost ( for an immature technology too ) and JWST does not a happy match make.

      The formation flying required to enable a telescope star-shade combo is still unproven and waiting on the PROBA sunshade mission to prove itself in action ( all be it over tens of metres as opposed to tens of thousands of kilometres) .

      Then there is the star-shade itself which is only at a low maturity level and has amongst others , problems associated with solar glint and precision tooling of its external anti refraction petals . The EXO-S Probe class concept explored star-shade costs and found even the 27m shade needed for just a 1.1m telescope costing an optimistic $640 million. What price the 50m shade for JWST?

      The best ( outside ) bet might be a WFIRST star-shade rendezvous mission with a 34m shade described as part of the EXO-S concept. A big if though.

      • Wojciech Jacyk April 6, 2021, 16:45

        “The formation flying required to enable a telescope star-shade combo is still unproven”
        I keep wondering if the technology used for precision formation flying of massive drone swarms like the ones companies use in China can’t be refined further to help with this. In this video you can see a formation of over 3000 drones taking various complicated and moving shapes.


  • kzb April 5, 2021, 19:49

    To me, the most interesting aspect of this project is the rogue planet detection. There is huge uncertainty in how many of these there are, particularly in the sub-Jupiter mass range. Will the RST give us some better statistics on the occurrence rate?
    On the orbiting planet detection, I think the true scientific value is detection of wide-orbit planets by microlensing. There are a lot of close-packed systems known already, what is lacking in the current stats are systems similar to our own.

  • Patient Observer April 5, 2021, 21:41

    Adding thousand of planets to the database will certainly help place limits on the general characteristics of planetary systems. All good. However, if the main curiosity driver is the discovery of alien intelligence, would it not be worth a few billion dollars or so (much less than the LHC by comparison) to develop a space probe that could chase down Oumuamua? It would be a long shot (ha ha) but the scientific/cultural impact would be incalculable. Even if it turns out to be a frozen chunk of gas, the technology would open up many new possibilities in space exploration.

    Such a probe would be multiple orders of magnitude less demanding than the contemplated probe to Proxima Centauri. One-hundredth the velocity, one ten-thousandths the energy for the same size probe.

  • Robin Datta April 5, 2021, 22:50

    Perhaps time to revisit the Drake equation…
    Kipping’s ice cream truck filled with Cool Worlds:

    Drake Equation Deep Dive Part 1

    Drake Equation Deep Dive Part 2

    • ljk April 6, 2021, 11:03

      And here is Carl Sagan’s take on the Drake Equation from his famous Cosmos series:


      One million alien civilizations, ten thousand, ten, or none? That is why we need to actively search for them.

  • ljk April 6, 2021, 11:16

    APRIL 5, 2021

    Size of raindrops can help identify potentially habitable planets outside our solar system

    by Leah Burrows, Harvard John A. Paulson School of Engineering and Applied Sciences

    One day, humankind may step foot on another habitable planet. That planet may look very different from Earth, but one thing will feel familiar—the rain.

    In a recent paper, Harvard researchers found that raindrops are remarkably similar across different planetary environments, even planets as drastically different as Earth and Jupiter. Understanding the behavior of raindrops on other planets is key to not only revealing the ancient climate on planets like Mars but identifying potentially habitable planets outside our solar system.

    “The lifecycle of clouds is really important when we think about planet habitability,” said Kaitlyn Loftus, a graduate student in the Department of Earth and Planetary Sciences and lead author of the paper. “But clouds and precipitation are really complicated and too complex to model completely. We’re looking for simpler ways to understand how clouds evolve, and a first step is whether cloud droplets evaporate in the atmosphere or make it to the surface as rain.”

    “The humble raindrop is a vital component of the precipitation cycle for all planets,” said Robin Wordsworth, Associate Professor of Environmental Science and Engineering at the Harvard John A. Paulson School of Engineering and Applied Sciences (SEAS) and senior author of the paper. “If we understand how individual raindrops behave, we can better represent rainfall in complex climate models.”

    Full article here:


    • Alex Tolley April 6, 2021, 14:30

      But too much rain might not be conducive to human occupation: The Long Rain.

      [Of the 2 dramatizations, I preferred the Rod Steiger version as part of the movie “The Illustrated Man” rather than The Ray Bradbury Theater version with Marc Singer.]

      The real Venus proved more of a hell than the older, prehistoric planet model.

  • Mike Serfas April 6, 2021, 19:35

    With an exoplanet hunt starting toward the Galactic Center, it seems like a good time to ask whether predictions of alien civilizations being more likely in the direction of the Center have any merit. We’ve seen things like this a long time: https://curiosmos.com/study-reveals-that-complex-alien-civilizations-likely-exist-near-milky-ways-galactic-center/ But are these arguments widely accepted, or a shot in the dark?

Leave a Comment