What is the geometric plan for ALMA?

[swampwiz]

I was watching a TV show about these telescopes being transported up to a high plateau in Chile. Looking at the Wiki page, it seems that they are to be arranged in random geometry, although maybe there is some pattern to the arrangement. Does being in some pattern - or being in a random pattern - matter for these arrays?

https://en.wikipedia.org/wiki/Atacama_Large_Millimeter_Array

 

[Zeke137]

The ALMA antennas are now all situated at the high plateau in the Atacama desert in Chile at 5000 meters altitude, and will remain there, unless some major fault occurs in one, in which case it would be transported back down to the Operations Support Facility at 2900 meters altitude.

Each of the 66 antennas sits on a fixed concrete pad, of which there are about 200 arranged in a semi-random pattern over a 16 kilometers wide area. Each pad provides a solid base for the antenna, and provides means by which the antenna can be fixed securely and in a position accurate to within a millimeter or so. The pads also provide connectors for power to enable the antenna to move its' dish, as well as data connectors to enable the received signals to be relayed to the central computers for data analysis.

The antennas can be moved by the transporters to different sets of pads in order to change the configuration and extent of the array as a whole, or of a subset of the array. The reason for doing this is to reconfigure the size and shape of the "virtual dish" formed by the array of antennas, and thereby change the sensitivity and field of view of the array for any particular observing run.

Each observing run uses a subset of the main array of varying size depending on the observation required, so more than one observing run can take place at anyone time, each using a different subset. The more antennas used in an observing run, the more sensitive the results. Placing the antennas closer together provides a wider field of view, while placing them further afield results in a narrower field of view, allowing exquisitely detailed observation of objects of small angular extent.

So, yes, the distribution of the antennas at anyone time does matter.

 

[swampwiz]

So you are saying that the arrangement is intentionally done to be random.

 

[Zeke137]

So you are saying that the arrangement is intentionally done to be random.

The placing of the antenna pads appears to be random, but it's only "semi-random". The actual positioning of the fixed pads was carefully calculated before the site was built to give the best results when sets of antennas are placed to give optimum results when conducting specific observing runs.

Again, the positions of the pads are not actually random, they just appear to be. You should drop the idea of randomness here, since it does not apply.

 

[MikeeMiracle]

Yes it's not random, depending on what they want to image they are moved around. If you want to take a high resolution image of a very small area more of the antenna's are moved towards the center. If you are taking an image of something bigger with less overall resolution you move the antenna's towards the edges. There is no "generic / normal" configuration of their placements, it is all dependant on what it is you are trying to image at that time. This is partly what makes ALMA such an effective platform.

 

[swampwiz]

OK, so it sounds like only a few of these (maybe only 3?) will be used at any time to view a single object, and having so many telescopes allows for there to be multiple observations done simultaneously. Can I presume that the set of telescopes that are to be used to view a single object are at least in a circular pattern?

 

[MikeeMiracle]

Im not sure it's designed to image multiple objects, it takes readings from each antenna and uses software to make in effect one big antenna. The same process that works with regular optical telecopes. For example you may have heard of the "Black Hole Telescope" from the news recently, this was just a bunch of telescopes from around the world and their data was combined to be in effect 1 big telescope that was almost the size of the earth.

 

[Zeke137]

@swampwiz, it needs to be pointed out here that PF is not a question-and-answer site, but rather a place for discussions in general on science topics. Naturally, during the course of a discussion, questions will be raised, mainly in an attempt to clear up points on which one might not be clear.

So, people are encouraged to do their own research, either online or in a library: self-help and self-education are key here. I would encourage you to do some research yourself on the subject of the ALMA observatory, and on radio interferometry in general. Basing a series of questions on what one has gleaned from a TV show is at least a start, but you really should now start making the effort on your own to research such topics.

In this vein, you could Google such topics as:

• ALMA observatory antenna placement,
• ALMA observatory antenna docking pads,
• ALMA observatory observation run,

and so on. Follow links thrown up by the Google searches; do not expect to be spoon-fed.

In the course of conducting such searches online today, I came across the following:

• an ALMA observing and science primer which is a general, and gentle, introduction to the observatory and how it operates, and
• the ALMA Technical Handbook, which is a more technical and detailed study of the physics and math behind the operation of ALMA;

you may find these helpful and illuminating.

Can I presume that the set of telescopes that are to be used to view a single object are at least in a circular pattern?

Why would you wish to presume this? Do you imagine that radio interferometers need to have their individual dishes to be placed precisely on the circumferences of circles or squares? Quick answer is that they don't: it's sufficient that their positions are known to a high degree of accuracy.

Here's the VLA in New Mexico:

and here's the ALMA array in Chile:

and here a plan of some of the antenna pads at ALMA:

which displays antenna pads on a logarithmic radial scale: nonetheless, you'd be sorely challenged to find a group of more than one or two which sit on an arbitrary circle.

 

[Zeke137]

it sounds like only a few of these (maybe only 3?) will be used at any time to view a single object

The ALMA array can observe more than one target simultaneously, but only under certain circumstances. Many more than three will be used for most observing runs, except when conducting a "Total Power" observation with typically one dish to complement observations made with multiple dishes. See the Technical Handbook for more details.

 

[swampwiz]

OK, so it appears that as long as the telescopes are not collinear, that's all that matters. One would presume that a 120-degree Y-shape would be optimal, but it's not necessary.

 

[Zeke137]

Presumptions tend to lead one astray and prevent one from learning in an objective manner.

it appears that as long as the telescopes are not collinear, that's all that matters

Do you have a reputable source where this is shown to be true?

One would presume that a 120-degree Y-shape would be optimal, but it's not necessary

Again, do you have a reputable source where this is shown to be true?

Can I presume that the set of telescopes that are to be used to view a single object are at least in a circular pattern?

What leads you to you presume any of this?

Rule #1 in science: learn science the way it actually is, not the way one would wish or presume it to be. That is a skill to be learned.

 

[swampwiz]

Presumptions tend to lead one astray and prevent one from learning in an objective manner. Do you have a reputable source where this is shown to be true?
Again, do you have a reputable source where this is shown to be true?
What leads you to you presume any of this?

Rule #1 in science: learn science the way it actually is, not the way one would wish or presume it to be. That is a skill to be learned.

OK, since you you know so much more than me about this, what are your musings about my original question (and arrays in general)?

 

[Zeke137]

since you you know so much more than me about this

If that is at all true, it's because a lot of my time has been dedicated to educating myself in matters scientific. The same can also be said of many here in PF.

what are your musings about my original question?

So, you asked whether the arrangement of the individual dishes in these arrays mattered, i.e. whether they needed to be in some pattern, or whether they needed to be in a random pattern?

The answer is: it depends. It depends on the individual array, on the radio wavelengths and type of antennas used, and on the design parameters of the array. The important point to remember is this: that the positions of the individual dishes in an array are known to a high degree of precision, and that the timing of signals or wave-fronts impinging on the dishes is managed to microsecond precision or better. If the computer and software hooked up to the array can only manage regular arrays, then so be it. If they're more sophisticated, as tends to be the case these days, then a random (or "semi-random") arrangement can work.

Another point which needs to be addressed, in regard to the ALMA array, is that it consists actually of two separate arrays:

ALMA consists of a giant array of 12-m antennas (the 12-m array), with baselines up to 16 km, and an additional compact array of 7-m and 12-m antennas to greatly enhance ALMA's ability to image extended targets, located on the Chajnantor plateau at 5000m altitude. The antennas can be placed in different locations on the plateau in order to form arrays with different distributions of baseline lengths. More extended arrays will give higher spatial resolution, more compact arrays give better sensitivity for extended sources. In addition to the array of 12-m antennas, there is the Atacama Compact Array (ACA), consisting of twelve 7-m antennas and four 12-m antennas. This array will mostly stay in a fixed configuration and is used to image large scale structures that are not well sampled by the ALMA 12-m array.

You might also like to read up on the some of the other radio arrays around the world:

• the 7-dish Merlin Array in the UK (set up in the 1970's and still going strong), or
• the SKA (Square Kilometer Array, to be made up of thousands of separate dishes and other antennas - in construction in Australia and South Africa), or
• the LOFAR (Low-Frequency Array) in Europe, which consists of several large arrays of omnidirectional antennas (no dishes), 20,000 in all, or
• the IPS (Interplanetary Scintillation Array ), set up in 1967 by Antony Hewish, consisting originally of several thousand dipole antennas in a phased array, and which proved instrumental (pun intended) in the discovery of pulsars, or
• the CHIME (Canadian Hydrogen Intensity Mapping Experiment ), an array of semi-cylinders populated with 1024 dual-polarization radio receivers,

to get an idea of how diverse radio astronomy arrays can be.

So, since it was you who asked the original question, I would ask you whether the posts here have answered your question?

 

[swampwiz]

So you are saying that the randomness/order of the arrangement depends on the wavelength? Or perhaps the noise at a wavelength? There must be some systematic way to make this determination.

 

[Zeke137]

Arrays of non-dish antennas are often ordered; dish arrays can sometimes be ordered, sometimes not. Non-dish arrays tend to be built for wavelengths too long (i.e frequencies too low) to be accommodated by dishes.

There must be some systematic way to make this determination

Indeed, there must be.

Now, be a good fellow and go off and do some research on your own. Self-education is the key here: one also needs to realize that a two-minute attention span, useful in skimming the Internet, is of no use in scientific education.

 

[swampwiz]

Im not sure it's designed to image multiple objects, it takes readings from each antenna and uses software to make in effect one big antenna. The same process that works with regular optical telecopes. For example you may have heard of the "Black Hole Telescope" from the news recently, this was just a bunch of telescopes from around the world and their data was combined to be in effect 1 big telescope that was almost the size of the earth.

OK, I can grok how the calculations can be done to process the data from existing telescopes, and the fact, obviously, that once a telescope is sited, it is astronomically [pun intended] more difficult to move that telescope to some better location than to just use it as is. But certainly, there must be some optimum shape - or sets of shapes - that are better than others.

 

[Zeke137]

there must be some optimum shape

The optimum shape of the ALMA, SKA, LOFAR, CHIME etc. arrays is the one they have. Considering the amount of money poured into these observatories, you can count on it. Thinking otherwise is to consider the designers to be fools. And they amn't fools.

 

[swampwiz]

The optimum shape of the ALMA, SKA, LOFAR, CHIME etc. arrays is the one they have. Considering the amount of money poured into these observatories, you can count on it. Thinking otherwise is to consider the designers to be fools. And they amn't fools.

Well, the anthropic principle is very good at explaining complex solutions. I'd just like to have some insight into the mathematics/physics that make it so.

 

[Zeke137]

I'd just like to have some insight into the mathematics/physics that make it so.

You might like to refer back to one of the links I provided in post #8 above, and which I'll provide again here: the ALMA Technical Handbook which, as it's name implies, is a technical manual for scientists intending to use the ALMA observatory in particular, and also acts as a technical guide to multi-dish radio interferometry in general.

Topics include:

• array components
• principles and concepts of interferometry
• receivers
• correlators
• spectral setups
• imaging with ALMA
• observing modes

and so on. There's a fair bit of math and physics in there, and references to further reading, so as a general guide to the topic, it's not bad at all.

And then you can use your favourite online search engine to carry your enquiries further. Or sign up to a degree course in the subject...

 

[Zeke137]

the anthropic principle is very good at explaining complex solutions

By "complex solutions" I take it you mean complex situations? Well, the anthropic principle (AP) is not a set of rigorous physical laws which could be used to explain things, but merely a philosophical viewpoint which was

formulated as a response to a series of observations that the laws of nature and parameters of the universe take on values that are consistent with conditions for life as we know it rather than a set of values that would not be consistent with life on Earth. [Wiki]

As such it doesn't explain anything at all (to do so would be an exercise in tautology), but simply provides a different perspective on the universe as a whole, and on the existence of life in the universe.