Why are we building larger land-based telescopes instead of launching larger ones into space?

  • This question is a follow-up to Do bigger telescopes equal better results?

    How much bigger does a ground-based mirror have to be to match what a space-based one can do? I guess I'm asking primarily for visible light, but I'm interested in general too.

    I guess on the ground, you're safe from micrometeorites, so it will probably last longer. At what point does it become cheaper to build a telescope on the moon or something?

    But in space, you don't have clouds to block your view (well actually you do, but those are big clumps of dust) or airplanes photo bombing you.

  • Mark Olson

    Mark Olson Correct answer

    4 years ago

    It's cheaper.

    (1) With adaptive optics you can get 0.1 arc second resolution on the ground (admittedly only on a mountain top with particularly good air flow, but still!). This eliminates one of the major advantages of space until you get above several meters mirror diameter.

    (2) Rocket fairings are the shrouds which protect payloads during the supersonic atmospherics speeds reached during launch. A 5 meter fairing is about the largest that can be flown, which limits the size of the one-piece mirrors which can be launched. (The Dreaded Webb Telescope's mirror is in pieces which will assemble themselves in space -- a very scary and very expensive piece of design.)

    (3) Servicing a telescope on the top of Mauna Kea or in the high Chilean Andes is a difficult and expensive process. Servicing a telescope in orbit makes that look like small change. (Cost comparable to the cost of building a new giant scope on Earth.) And in-orbit servicing can't even be done with current technology except in low earth orbit.

    (4) While high resolution is one frontier in astronomy, going deep is another, and going deep requires big mirrors. A 30 meter mirror on Earth gathers much more light than a 5 meter mirror in space. The giant terrestrial telescopes simply do a better job of being light buckets for spectroscopy than anything we can yet put into space.

    The bottom line is that with the development of adaptive optics, space-based telescopes of currently buildable and launchable size lost their main advantage over ground-based telescopes. And since they're 10x to 100x the cost, they are simply not worth building for many purposes.

    Space based telescopes still hold a significant edge in parts of the spectrum blocked by the atmosphere such as UV and IR (Webb), and for certain tasks involving long-term high accuracy photometry (Kepler) and astrometry (Gaia). But for general purpose use, the balance seems firmly on the ground's side for large telescopes.

    This will change if space flight becomes cheaper -- the SpaceX BFR, for example, with its 9 meter fairing and dramatically lower launch costs, offers great hope for space telescopes.

    Perhaps add that adaptive optics doesn't really work at visible wavelengths; only near-IR. A space-based telescope is always going to give better angular resolution at visible wavelengths.

    WRT being better light buckets, you can keep a space-based telescope pointed at the same spot for a long time without much difficulty, e.g. the Hubble deep fields.

    @jamessqf: True, but you're using up many, many hours of time on a very expensive instrument. It's much more efficient to build a bigger telescope on Earth for fewer dollars. Taking multiple exposures of the same spot and adding them work on Earth as well as in space -- most of the sky is occulted by the Earth at some point int he Hubble's orbit.

    @RobJeffries AO works darn well in the visible, tho' I'll concede that the blue end gets tough to handle.

    Past a certain point, doesn't deep also require good IR capabilities? I don't think the JWST design was picked out of a hat.

    @Donald.McLean It depends on what you want to do. The Webb is designed to look 'way back, but a big telescope operating in the visible can do spectroscopy of faint objects out past z=2, and that covers a *very* large number of interesting objects. The critical thing is that Webb is racing past $9 *billion* dollars, whole the Thirty Meter Telescope is estimated at $1.4 billion. The Webb's not called "the telescope that ate astronomy" for nothing!

    @CarlWitthoft Which large telescopes currently have AO systems that give diffraction-limited imaging at <700nm? I think the performance degrades very significantly below this. Maybe SAXO on the VLT?

    @RobJeffries ESO does or will, I think. I left a certain company named "Adaptive Optics Associates" about 8 years ago, so I don't know what the final performance specs are on the systems said company was installing on a few multi-telescope installations.

    @CarlWitthoft I was hoping for something a little more concrete to justify your (now 3-times) up-voted statement. Certainly 8-years ago there was no visible AO system on a big telescope. However, I extend the same invite to the up-voters. PALM-3000 on the Palomar 5-m? Does this work shortward of 600nm?

    @Mark Olson -- you can't do *rest-frame optical* spectroscopy of objects at $z > 2$ from the ground the way you can in space, because different spectral features are redshifted in and out of narrow atmospheric windows. You can't do rest-frame near-IR spectroscopy at all. And of course you can't do a whole range of things for local objects in the mid-IR from the ground. (Why was the Spitzer Space Telescope, with only a 1m mirror, even launched? It certainly wasn't for superior angular resolution.)

    @jamesqf Actually, neither space- nor ground-based telescopes expose for much more than ~30 minutes. Rather, multiple exposures are combined to a single image. The HUDF, for instance, used 800 exposures, each of 1200 seconds. The reason is both that bright objects will cause pixels to saturate if exposed for too long, and that the probability cosmic rays ruining a nice image increases with exposure time. But with several shorter exposures, CRs are eliminated by taking the median of many exposures.

    @Mark Olson: I really can't see that most of the sky would be occulted from Hubble. Some of it, yes, but there should be areas to the north & south that are not. And of course you can always point it somewhere else - there's always some area of sky that's visible, while an Earth-based telescope is useable less than half the time. Then you have the option of parking it somewhere else than LEO...

    @jamesqf The context was the supposed problem that Earth-based telescopes wouldn't permit ultra-long exposures. My point was that for an LEO telescope, the same is true: Most of the sky is occulted by the Earth at some point in the telescope's orbit. (Obviously most of the sky is also *visible* at some point in the orbit.)

License under CC-BY-SA with attribution

Content dated before 7/24/2021 11:53 AM