I’m curious because I wanted to know if it is technically possible for us to build a telescope that allows us to look far enough to be able to see planets in other star systems as well as we can see those in ours. Are there any absolute upper limits? Or could you spot a human-sized object on a planet 100 light-years away as long as your telescope is giant enough? 100,000 light-years away? I’m aware that we’d be looking into the past.
I assume you mean “as well as we can currently see those in ours”, since any technology that let us get good images of distant planets would also let us get even better images of nearby planets.
There are indeed proposed telescopes, that could be built with current or very-near-future technology, that could produce rudimentary images of exoplanets. There’s nothing proposed that could get pictures nearly as good as what we have now of, say, Jupiter, but I don’t see any fundamental reason why we couldn’t.
I was thinking that maybe not enough light from those planets reaches us for us to be able to make detailed images of them. But I guess that’s not true? With the most sensitive and advanced possible technology, could we see an ant on a planet 100,000 light years away? Or are there just limitations in physics that prevent us from being able to see that far and that well?
Just to emphasize one point that hasn’t been mentioned; I’m no astrophysicist, but I’m pretty sure the telescope would have to be out in space like the Hubble, not on Earth. Atmospheric distortion would be insurmountable at some point if you were trying to build a giant one on Earth.
Earthbound telescopes suffer from looking through the atmosphere, which is not sufficiently uniform nor stable to allow very good resolution. Hubble is “less” of an instrument than many earthbound 'scopes, but produces superior imaging since it is above the atmosphere.
The resolution of a telescope is determined by it’s aperture, which poses weight (big issue for space born telescopes) and mechanical and thermal problems when you try to go too big. You can fake it to a degree with multiple telescopes combining their data to increase the effective aperture to around the separation between the individual telescopes. This is much harder if they are in orbit, as you have to be very precise when you combine the data, and in orbit they will be on a sphere or ovoid, rather than on a plane like earthbound arrays, and at optical wavelengths changes with movements of literally millionths of an inch matter.
So with a field of stationary cooperating telescopes on a planet without an atmosphere the possibilities would be nearly limitless, and the only thing stopping you from being able to see an ant at 100,000 light-years is the size of the field?
Light isn’t really the limit–diffraction is. The only way around the diffraction limit is to build a very large scope, but that scope also collects way more light than you need. So instead you build an array of scopes. In terms of the diffraction limit, they act as a single scope the same diameter as the array, but without one giant mirror covering the whole array.
This technology is called interferometry. It’s used quite widely in the radio spectrum, and in a few places in the optical spectrum (the Keck Observatory at Mauna Kea is one place). But Earth isn’t big or stable enough for an array capable of exoplanet imaging. For that, you want *very *widely spaced scopes. The L4 and L5 Lagrange points (+/- 60 degrees from Earth’s orbital position) might be good candidates for a two-scope system. The tech needs to improve a bit before we’re capable of that, though.
Number of photons available can be a more significant limit than diffraction, depending on how long your exposure time is.
For very distant, dim objects like early galaxies–certainly. The Hubble Ultra Deep Field image required hundreds of hours of exposure time. But for relatively nearby exoplanets it’s not nearly as big an issue. If I remember, I’ll work out tonight how many photons/sec you could expect from one. Think about this, though–a 1/100 s exposure of Mars gives a fairly bright image even on a modest hobbyist scope. A nearby exoplanet might be 1e6 times farther away, so it has 1e-12 the light. You can make that up with 1e6 times the exposure (a few hours) and 1e6 the aperture (a 100 m scope (or equivalent collection) instead of 0.1 m). But for the same angular resolution, you’d need a 100,000 m scope–or use interferometry.
A few hour exposure would, admittedly, blur some detail on a quickly rotating planet. If this is a problem, you can add more aperture until you’re happy.
I was thinking about this idea of seeing an ant 10,000 LY away. Seems to me that if you get that extreme you will run into another problem even if you can get high enough resolution. I can’t see how you could possibility get enough stability even in space (certainly not on earth) to avoid smudging out the image given that the ant is some astronomically small (pun intended) fraction of an arc-second.
What about building an array of scopes on the moon? that would provide a platform for them, at least… maybe make the glass up there too, if there’s sufficient of the right materials.
If we build a really big telescope (and I mean REALLY big), don’t we run into trouble with temporal blurring?
That is, the light collected at the rim of the reflector would be older than the light collected nearer the centre - and we would end up trying to image an object as it appears across a span of time.
Obviously the same sort of problem bedevils smaller telescopes, as the exposure time needs to be longer to collect enough light to form an image.
Not if it’s designed correctly. For instance, a parabolic reflector will have the same path lengths for all points on the surface. Points along the edge are farther from the focus, but because the mirror curves up they’re also closer to the imaged object, and these two factors exactly cancel out (in fact, this is one definition of a parabola). So all the incoming light will have the same age.
A real telescope will be more complicated, and probably have multiple reflectors, but this can be accounted for. In fact it’s critical if you’re going to do any interferometry.
Temporal blurring over long exposures is a problem, though.