4 ms·

NASA should over the next 100 years build a hyper-telescope, so that people can begin to actually see these planets in detail before before a probe can be sent.

NASA should over the next 100 years build a hyper-telescope, so that people can begin to actually see these planets in detail before before a probe can be sent. Even with current technology, a hyper-telescope could be self-assembled in space, built from billions of 3-d printed & smoothened mirror segments, all aligned to half wavelength of visible light. We need to have a 1000px x 1000px resolution of earth-like planets out to 100 light years away, which would be several thousand miles in diameter.

dredmorbius 10y ago
That's all but impossible. To optically resolve an Alpha Centauri planet of Earth size at Hubble resolutions of Pluto (about 680 pixels square) you'd need a 28 km wide scope (or array). https://www.quora.com/How-big-would-a-telescope-have-to-be-to-see-Kepler-452b https://www.quora.com/How-big-would-a-telescope-have-to-be-t... Rather than seeing shapes of objects, detecting their signatures -- light occlusion, spectroscopy, thermal and radio signatures -- is far more useful. Several of those would reveal chemistry and radiation indicative of life and/or intelligence. Much xenoplanetary analysis isn't done through optical telescopes, per se, but through spectrographic analysis, which tells us what compounds and elements are present (or are blocking / reradiating) starlight.
- DenisM 10y ago
  Nothing wrong with a 28 km array though, if we can put it in space. Besides, what about dynamic aperture? I know very little about it, I only heard that a moving camera effectively expands it aperture by the distance traveled within the duration of the exposure.
  - dredmorbius 10y ago
    You're going to want _both_ light gathering _and_ range. Keep in mind, a planet is probably going to be moving a lot over the course of a few hours. A 24 hour time-lapse of the Earth will give you a lot of pretty colours. But not much by way of a useful image. You're trying to see something tiny and dim a very long way away, next to something which whilst also tiny (relative to the Universe) is far, far, far brighter. A long-duration moving-apature exposure of a remote planet next to a dim star seems to me not to be an ideal candidate for high-quality imaging. I'm not sure a large-baseline, large-collector array would do much better, but at least it could narrow the time interval somewhat. (Though I think you're still basically going to end up with a glared-out, underexposed, blurred pixel.)

dredmorbius 10y ago

That's all but impossible. To optically resolve an Alpha Centauri planet of Earth size at Hubble resolutions of Pluto (about 680 pixels square) you'd need a 28 km wide scope (or array). https://www.quora.com/How-big-would-a-telescope-have-to-be-to-see-Kepler-452b https://www.quora.com/How-big-would-a-telescope-have-to-be-t... Rather than seeing shapes of objects, detecting their signatures -- light occlusion, spectroscopy, thermal and radio signatures -- is far more useful. Several of those would reveal chemistry and radiation indicative of life and/or intelligence. Much xenoplanetary analysis isn't done through optical telescopes, per se, but through spectrographic analysis, which tells us what compounds and elements are present (or are blocking / reradiating) starlight.

DenisM 10y ago
Nothing wrong with a 28 km array though, if we can put it in space. Besides, what about dynamic aperture? I know very little about it, I only heard that a moving camera effectively expands it aperture by the distance traveled within the duration of the exposure.
- dredmorbius 10y ago
  You're going to want _both_ light gathering _and_ range. Keep in mind, a planet is probably going to be moving a lot over the course of a few hours. A 24 hour time-lapse of the Earth will give you a lot of pretty colours. But not much by way of a useful image. You're trying to see something tiny and dim a very long way away, next to something which whilst also tiny (relative to the Universe) is far, far, far brighter. A long-duration moving-apature exposure of a remote planet next to a dim star seems to me not to be an ideal candidate for high-quality imaging. I'm not sure a large-baseline, large-collector array would do much better, but at least it could narrow the time interval somewhat. (Though I think you're still basically going to end up with a glared-out, underexposed, blurred pixel.)