lunar surface imaging (was Re: Too much crap in s.s.h)

[Henry Spencer]

In article >,
Julian Bordas > wrote:
>Hey, I've got an idea. How big and or how far apart would three
>Newtonian telescopse need to be to collectively image the Apollo
>landing sites? The three images are computer processed to give the
>impression that there is one huge telescope.

The basic math is not that hard. To a reasonable first approximation,
aperture = distance*wavelength/resolution. At 384,000km in 500nm light,
achieving 1m resolution (enough to show an LM descent stage as a shape
rather than just a spot) requires a telescope about 200m in diameter.
But really convincing pictures would probably require 10cm resolution
or thereabouts, calling for a 2km telescope.

Note that you cannot fake a 200m or 2000m telescope by putting multiple
telescopes that far apart, snapping pictures with them, and then somehow
magically combining those pictures. Sorry, interferometry doesn't work
that way. To combine recordings after the fact, the recordings have to
include phase information, which images *don't*. In fact, there is no
practical way to detect and record phase information for light. Optical
interferometry requires combining the light from the telescopes as it
arrives, so you immediately get phase-difference information. This has
been demonstrated, but has some way to go yet before it's really a routine
technique. It's hard.

And as others have noted, it won't convince the True Believers anyway,
especially if it's a government-funded project. If they don't believe
razor-sharp photos shot on the spot by astronauts with cameras, why should
they believe fuzzy computer-processed images from telescopes?
--
MOST launched 1015 EDT 30 June, separated 1046, | Henry Spencer
first ground-station pass 1651, all nominal! |

[Julian Bordas]

On Sun, 20 Jul 2003 19:03:38 GMT, (Henry Spencer)
wrote:

>In article >,
>Julian Bordas > wrote:
>>Hey, I've got an idea. How big and or how far apart would three
>>Newtonian telescopse need to be to collectively image the Apollo
>>landing sites? The three images are computer processed to give the
>>impression that there is one huge telescope.
>
>The basic math is not that hard. To a reasonable first approximation,
>aperture = distance*wavelength/resolution. At 384,000km in 500nm light,
>achieving 1m resolution (enough to show an LM descent stage as a shape
>rather than just a spot) requires a telescope about 200m in diameter.
>But really convincing pictures would probably require 10cm resolution
>or thereabouts, calling for a 2km telescope.
>
>Note that you cannot fake a 200m or 2000m telescope by putting multiple
>telescopes that far apart, snapping pictures with them, and then somehow
>magically combining those pictures. Sorry, interferometry doesn't work
>that way. To combine recordings after the fact, the recordings have to
>include phase information, which images *don't*. In fact, there is no
>practical way to detect and record phase information for light. Optical
>interferometry requires combining the light from the telescopes as it
>arrives, so you immediately get phase-difference information. This has
>been demonstrated, but has some way to go yet before it's really a routine
>technique. It's hard.
>
>And as others have noted, it won't convince the True Believers anyway,
>especially if it's a government-funded project. If they don't believe
>razor-sharp photos shot on the spot by astronauts with cameras, why should
>they believe fuzzy computer-processed images from telescopes?

Thanks for the reply, and the maths required. I'm not thinking of
this as a means of changing the minds of the True Believers, I was
curious as to how do-able it would be for the enthusiatuc amateur.
Not so easy it seems. The idea is to have ccd devices sending the
data to a computer to work with the phase difference info.
I'd like to take some stunningly good photo's of the moon and mars :-)



Julian Bordas

"It's not like I'm using.
It's just that my body's developed
this severe drug drug deficiency"
Neuromancer

[Henry Spencer]

In article >,
Julian Bordas > wrote:
>...curious as to how do-able it would be for the enthusiatuc amateur.
>Not so easy it seems.

Not easy at all. The telescopes need to have optics connecting them, and
the dimensions of those optical connections -- hundreds of meters long --
need to be stable to within a fraction of a wavelength of light.

>The idea is to have ccd devices sending the
>data to a computer to work with the phase difference info.

Unfortunately, CCDs don't sense phase at all, only intensity.

The only practical way to do phase differences at optical wavelengths is
to actually bring the light from the different scopes together at a common
point; see above. At radio wavelengths, sensing phase is not so hard, and
high-speed recorders with ultra-precise time references (atomic clocks)
can do a good enough job to permit computers to combine the signals later.
Light, alas, is around five orders of magnitude higher in frequency, and
there are no recorders or time references which will work at that speed.

>I'd like to take some stunningly good photo's of the moon and mars :-)

You *can* do better than you would think, using super-resolution
techniques to get sub-pixel resolution. See the article in the Dec 2001
Sky & Telescope.
--
MOST launched 1015 EDT 30 June, separated 1046, | Henry Spencer
first ground-station pass 1651, all nominal! |

[Dave Michelson]

Henry Spencer wrote:
>
> At radio wavelengths, sensing phase is not so hard, and
> high-speed recorders with ultra-precise time references (atomic clocks)
> can do a good enough job to permit computers to combine the signals later.

Correct.

> Light, alas, is around five orders of magnitude higher in frequency, and
> there are no recorders or time references which will work at that speed.

Holograms do a fairly good job of trying, though.

--
Dave Michelson