Slashdot Mirror
Optical Telescope Arrays by Amateur Astronomers?
fl_ska asks: "I recently attended a meeting of local amateur astronomers in my area known as The Local Group of Deep Sky Observers. After being blown away by views of the Orion Nebula and such, I got to thinking about the state of modern amateur astronomy. For instance, I recall reading about a project to link multiple optical telescopes so as to approximate the light-gathering capacity of a much larger telescope, a so called 'array of optical telescopes.' With the advent of the Web, it seems like it would be relatively easy to coordinate such a project via some central server which could then process and link images for all to view. I was wondering if there were any amateur astronomers out there who were possibly working on a similar project?"
"Could the same gains that were achieved with grid-computing be found in amateur telescope arrays? What kind of issues would be relevant to the problem of organizing such a project? Also, I once read about a way to correct for atmospheric perturbation by way of creating an artificial point of light in the upper atmosphere and, in real time, analyzing how the atmosphere acted on it. Could such a method be utilized by amateurs and would it detrimentally affect the original data set?"
I'd volunteer to help for a project like this, in whatever way I could contribute. Astronomy is the only science where amateurs can really help make a difference, and so tends to be very interesting to me. Hopefully there are some interested parties out there.....
Life is hard, and the world is cruel
In the grid computing analogy, 100 cheap computers working together can produce the equivalent output of one computer 100 times as powerful. The same cannot be said of amateur telescopes working together. For a simplistic example, adding two images both of signal-to-noise 10:1 will result in an image with best signal-to-noise of 20:1.4, which is not twice as good as two single images. Modern large-scale telescopes are so great that all the amateurs in the world would have trouble stitching together a comparable image. Best for amateurs to concentrate in areas the big boys can't match, and not try to win the signal-to-noise (or deep image) game.
"Eve of Destruction", it's not just for old hippies anymore...
It would be difficult. I think you're talking about interferometry. This was originally developed for radio telescopes, and is harder to do at shorter wavelengths. The Submillimeter Array, working at the shorter submm wavelengths, has just opened on Mauna Kea, although some work has already been done with linking the James Clerk Maxwell Telescope and the Caltech Submillimeter Observatory. At optical wavelengths it gets harder still. An example is the Cambridge Optical Aperture Synthesis Telescope (COAST). There's also the proposed `Ohana project.
A major problem is that you have to preserve the phase information of the light when you combine the signals from the telescopes, so you can't just record images with a CCD (which only gets you the intensity) and then try to handle the rest of it in software.
Essentially this means that you'd have to combine light from the telescopes in real time and keep the path lengths between them accurate to a small fraction of the wavelength you're measuring. You can do this "off-line" at radio frequencies, for example with the Very Long Baseline Array (VLBA) but not at optical frequencies.
So, in summary, the Internet lets amateur observers collaborate in various ways. However, combining their optical telescopes to get the resolving power of a larger telescope (the size of the distributed collection of individual telescopes) through optical interferometry is not one of them.
This post is strictly my own opinion and not necessarily that of my employer.
There is a project in the U.K. by professionals to do this (Cambridge Optical Aperture Synthesis Telescope). This gives a pretty good description of what is necessary for setting up an optical array. Note the combining building, which is where the light path lengths are matched. The main thing is that the images are formed from light that arrives at the telescopes at the same moment. With radio frequencies, the signal can be recorded on tape, along with a time hack, allowing for multiple signals to be combined after the fact. I don't believe there is a way to do that with light.
That "linking" of optical telescopes together is called optical interferometry and the linking usually requires that you know the relative location of the two or more optically identical telescopes to within a wavelength of the EM radiation you're viewing in. For radio telescopes, this is a meter (more or less). For optical telescopes, this requires that the telescope tubes be very close together and usually mounted in a rather strong frame.
There are some very cool optical systems that can make the linking easier by eliminating some of the unknowns in location within the optics. There are also some newer control systems that can make this still easier. But the main optics in both tubes still need to be almost identical. The same model of the same brand isn't quite good enough. They usually need to be matched (fabricated at the same time to have the same characteristics).
An alternate (and much more achievable) plan would be to let your brain trial and error out the differences between two tubes in a binocular telescope. There are a number of websites out there describing some of the more successful efforts to do this. Collimation is again critical (the two tubes had better be pointing in the same direction) and even then you're probably going to get a headache.
But while you're rubbing your temples to deal with the headache, you'll be thinking back to those absolutely amazing views you saw through the binocular eyepieces.
There are some bigger binoculars that blur the difference between binocular and telescope. Oberwerk sells some very nice 100mm binoculars with telescope-style eyepieces for under $1600 and just a pair of 100mm binoculars will only set you back $400. Together, those 100mm tubes gather more light than a 5" refractor and the view through properly collimated binoculars is just plain better than through a single tube (IMNSHO). But before you think about laying out that kind of money, get some decent, inexpensive 70mm binoculars and keep going to those local meetings. Once you get to the point where you know you want more, you'll have a bit more experience and have learned a bit more about where to spend your money.
Regards,
Ross
Posted by Cliff on Monday February 02, @06:32PM
from the distributed-science dept.
fl_ska asks: "I recently attended a meeting of local amateur astronomers in my area known as The Local Group of Deep Sky Observers. After being blown away by views of the Orion Nebula and such, I got to thinking about how smart I am. After one foray into something I know nothing about I decided to use my gigantic brain to help these poor idiots make their little 'hobby' so much better. However, since I think I am a lot smarter than I actually am, I turn to the collective brain power of Slashdot to solve this problem for me. Once a few +5 interesting comments get posted, I will take them to the next meeting and prove to everyone how smart I am. Those stupid idiot amateurs will never know what hit them. Maybe they will immediately vote me to be their new leader for life."
If you just want to improve light-gathering ability, but not resolution, it is theoretically possible, although difficult.
Normally, when taking a long observation, you take a number of CCD exposures (a charge-coupled device is basically a photon counter, and you need to empty and measure it before it overflows), delete any from the sequence that are corrupted (by, e.g. high-energy radiation), and digitially add the remainder together to produce an image that is conceptually the sum of many hours of observation.
You could, in theory, merge images taken from several telescopes, but you'd have CCD grid alignement issues, and it's easier to just either
It takes 2844 15 cm backyard telescopes to equal the light-gathering power of one 8m telescope, and the loss in resolution is phenomenal, so this is not likely to be terribly interesting. (For low-cost, you have to compete with the 10m Hobby-Eberly telescope built for $13.5 million, so do your math accordingly.)
The exciting thing in modern astronomy that involves multiple telescopes is interferometry, which gives you a telescope with an effective resolution equal to that of a telescope as wide as the spacing between telescopes. You don't get the light-gathering ability of a telescope that big, but that's not usually the limit. You do get the focusing ability.
This, however, requires that you can measure not only the numbers, but also the phase of the arriving photons and combine them properly.
The classical focusing mirror does this directly. The intricate mirrors-on-trolleys arrangement beneath the VLT is another way of doing it.
Radio astronomers work at low enough frequencies that they can record all the information they need on tape at two telescopes (with clocks aligned to the nanosecond) and combine them later, but visible light, from 430 to 750 THz, gives more problems:
Thus, optical interferometry currently requires fully optical beam combining; the data is never converted to bits and so it can't be done over the net in the forseeable future.
Placing an artificial point of light in the upper atmosphere may not be feasible, but why not have the ISS take some measurements so we know the properties of light emitted by the points of light that are already there? Then we'd have some known reference points for correcting for the atmosphere's effects.
Give me my freedom, and I'll take care of my own security, thank you.
Other people have pointed out the problems with combining many small optical telescopes to get a deep image, but deep images tend to cover small fields. There's a lot of sky that goes unobserved every night, and moving or transient phenomena are easily missed. Why not coordinate your team to try to cover as much of the sky as possible? This could be great (depending on brightness and telescope size) for near Earth asteroids, optical SETI, (super)novae, optical counterparts to gamma ray bursters, and other things that go bump in the night. Plus the eyes and brains of amateurs are a distributed processing system for picking out the interesting stuff. I hope there's enough amateurs who are tired of taking yet another image of that nebula everyone else has taken images of, that this catches on.
If we were ants living on a Rubik's cube, differential geometry would be a little more confusing.
Unfortunately, most large-scale parallel processing tasks aren't so simple to break down. But a lot of them can be reduced to ones where you don't need terrible amounts of synchronization between nodes, or can use approximation algorithms to mitigate it or whatever. Second Life apparently runs its weather simulation using a grid running on the client systems, where each user's computer just computes the weather for the nearby "physical" area.
There are 10 kinds of people: ones who understand ternary, ones who don't, and ones who think this joke is about binary
Like the subject says. Optical astronomy, and building good telescopes in particular, is extremely difficult. Let's go though it step by step.
The primary mirror. The biggest night-time telescope (which is not yet in operation) has a 10.4-meter diameter primary, which is formed from 36 hexagonal segments. Search Google for "Gran Telescopio Canarias". Your amateur telescope has about a 30-cm aperture. You'll need 1200 of those alone to just capture the same amount of light. Here's the second point.
Primary mirrors are big so they collect a lot of light. This is good, because it means you don't have to expose for 10 years to actually see some faint object. The second reason for making a primary large, is because the theoretical resolution you can get with your telescope goes down as the size of the primary goes up. Bigger mirror equals better theoretical resolution.
You may have heard of interferometry as a means of getting a high resolution by using lots of small telescopes. It's used extensively in radio astronomy. It's been tried in optical astronomy. ESO's VLT at Paranal Observatory is, as far as I know, the most advanced with it's VLTI instrument. I'm not sure, but I think they've got it working at infrared wavelengths now.
Interferometry becomes more difficult as the wavelength gets smaller. Infrared sits around 1000 nanometers. Optical is around half that. For radio, it's about a centimeter. You'll need to know the distances between your telescopes to a fraction of the wavelength accurately. Also, you can't combine the signals later. Radio astronomy can, because they can easily record both amplitude and phase information of their signal. Then, using custom hardware (DSPs), the signals of a number of telescopes are combined. A CCD only records amplitude information. You'll have to combine the light in real-time. Is that hard? You'd better believe it.
Last point I'll make: seeing. Seeing is the reason why the Hubble makes such nice pictures. It's above the Earth's atmosphere, thus it's view is not disturbed by the same cruft you can see above a road on a hot day. You, on the other hand, are just about as low as you can get. Telescopes aren't just built anyplace. Extensive testing is done to select those sites that have the best seeing. Typical sites are high (think above 2.5km above sea level), near a large, quiet body of water (since it stabalizes the air temperature) and in areas that don't have a stratospheric jet stream. Did I mention clouds? Oh, and in remote areas with little light polution from cities. You'll be observing in less than ideal conditions, giving you a much reduced resolution.
So what about adaptive optics? A bright object, say star close by, with no resolvable extent (e.g., not a galaxy, supernova remnent, you name it), allows the use of AO. The AO needs this bright object, because it needs to adjust the mirror in real time (seconds, at the very most), and it can't wait an hour or so to actually see something. Usually, if you're observing a faint object near a bright one, you'll lock the AO on the bright one. But what if you don't have a bright object near your faint object of interest? The solution it to shoot a rather high power laser up in to the sky. It'll form a bright dot somewhere high in the atmosphere. Aim it close to your faint object, and presto: you've got yourself an artificial star. Search Google for "telescope laser" and you'll find a few nice images. AO is child's play compared to interferometry. That doesn't mean just anyone can do it, though.
Call me a pessimist, but I don't see how any group of amateurs can hope to achieve the quality of the images recorded by professional observatories.
Alfred
Actually, the gains for resolution wouldn't be worth it ... but for COVERAGE they would be. Imagine doing a full-sky survey with 8" scopes every few months or even weeks.