Science desk
< April 11	<< Mar \| April \| May >>	April 13 >

Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is a transcluded archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

April 12

What is the output of a radio telescope observation?

Does a typical radio telescope have a single antenna or an array of antennas? Can it produce a pixelated 2D "image" of the object it is pointing at, like an optical telescope? Or does it operate like an ordinary radio receiver with a single antenna, and therefore produce a single time-varying signal characteristic of the object it is pointed at? AxelBoldt (talk) 00:42, 12 April 2019 (UTC)[reply]

"...Typical" radio telescope?

Here's "What Are Radio Telescopes?", from NRAO.

In one sentence, "...Special software designed by radio astronomers and software engineers then assembles the data to create maps of radio objects in the sky."

Any specific facility might be able to provide raw data products as time-series signals, or as "processed" data products - perhaps as ordinary digital picture files, or as specialized data-types for scientific software. Each facility might publish a lot of specialized software that is unique to that particular scientific facility, whose single purpose is to do some kind of data-processing specifically for the radio telescope and its dedicated scientific mission. If a scientist had a special need to get a different type or format of data, they'd work with the facilities to design electronics or software to get the job done.

Nimur (talk) 01:21, 12 April 2019 (UTC)[reply]

Thank you for the link. Do I interpret it correctly if I say that a single dish has a single antenna and operates like a single radio receiver (albeit recording at multiple frequencies simultaneously), and not like an optical telescope that outputs lots of image pixels? But if we have an array of such dishes, wire them together and correlate their outputs, we can compute such images? AxelBoldt (talk) 02:20, 12 April 2019 (UTC)[reply]

It is fantastically more complicated due to not only the reasons Nimur mentions below, but also because these measurements are made over a period of time and while the Earth is translating and rotating relative to the target. For the black hole image, this paper, part four in the series, describes the math behind turning the raw data into images. Someguy1221 (talk) 04:45, 12 April 2019 (UTC)[reply]

That is a generally accurate simple summary of how it works, for a "normal," "simple" radio dish. But, keep in mind that modern radioastronomy is really complicated - the technology involves some of the most specialized and sophisticated physics, electronics, and computer software on our planet (...and beyond); so any simplification will necessarily gloss over the important details.

Some telescopes measure amplitude; some measure radio frequency; some measure polarization; some measure doppler shifts; some radio telescopes are actually RADARs that emit a beam and listen for returned echos; some dishes have multiple different receiving "horns" (see our article on parabolic antenna for the basic anatomy); some dishes have a single horn but a complex beam pattern, and thus are measuring more than one point in the sky at a single instant (so extra work is needed to make the received signal work like a "pixel" for the purposes of imaging anything); some radio telescopes were specifically built to composite with other telescopes to form a synthetic aperture or a long baseline for interferometry; other facilities were built individually, but were later modified to work cooperatively; and so on.

If you're interested in the facilities that were recently in the news, the website is: Event Horizon Telescope, "a planet-scale array of eight ground-based radio telescopes forged through international collaboration." You can read about their very diverse facilities in great detail; and if you want the real technical details, start at the summary Focus on EHT publication (written for a professional science audience); and the six additional detail papers they cite. Paper II gives details of the array instrumentation, and Paper III explains the output data and how it was processed. Paper IV explains how the scientists turned that data into a picture. This required a lot of difficult math and physics. In particular, they used numerical inversion (and variations on that theme) to model and fit to the observed signals.

Nimur (talk) 03:46, 12 April 2019 (UTC)[reply]

Oh, and they published a lot of their numerical methods, in Python and FORTRAN-90, at astrosmili and eht-imaging ! Even a quick glance will illuminate that these software tools are not "general-purpose" - the algorithms and even the data formats are extraordinarily specific to this scientific program using this specific instrument array. Nimur (talk) 04:30, 12 April 2019 (UTC)[reply]

Short answer: one dish has one feed horn producing one time-varying signal, (though it may be possible to swap in different feeds for different frequency bands). Creation of a 2D image requires scanning, either mechanically, for low resolution (including by rotation of the earth), or by a phased-array of dishes for high resolution. Caveat: my experience of radio telescopes is extremely limited. catslash (talk) 00:16, 15 April 2019 (UTC)[reply]