Very cool. They use “distortions” of the radio signal by the interstellar medium between us and the pulsar to deduce the details smaller than can be resolved by direct interferometric observations. I don’t actually understand the nature of these distortions, but I’m guessing it’s something like how “seeing” in our atmosphere distorts optical images?
Here’s a great quote from their announcement: 50 picoarcsecond resolution is “equivalent to being able to see the double-helix structure of our genes from the Moon”
http://www.caastro.org/news/2014-astrometry
(P.S. the video Brian linked to below is not useful. The announcement I linked to is much more meaningful)
#picoarcseconds
Originally shared by Brian Koberlein
Pico Arcseconds
One of the advantages of radio astronomy is that you can connect observations from radio telescopes thousands of miles apart. Done in the right way, this creates a radio interferometer that effectively makes a virtual telescope as big as the separation (baseline) of the individual telescopes. The bigger your telescope (or virtual telescope), the finer the detail of your image. When we talk about the detail level of an astronomical image, we usually talk about the angle of separation between two distinctly resolvable points. So a resolution of a tenth of a degree would mean you could resolve two points of light (such as stars) separated by at least that angle.
In modern astronomy our resolution is much better than degrees, and we usually measure things in terms of arcseconds. An arcsecond is 1/3600 of a degree. It comes from dividing a degree into 60 minutes of arc, and each minute into 60 seconds of arc. Yes, the terms stem from their historical relation to time measurements. Beyond arcseconds, we simply divide them into parts by base 10. So a thousandth of an arcsecond is a milliarcsecond and so on.
Given the baseline of radio telescope interferometers, the upper resolution is typically on the order of milliarcseconds. But now a new paper in the Monthly Notices of the Royal Astronomical Society has resolved a radio pulsar to picoarcseconds. That is, a trillionth of an arcsecond, or about a separation of 5 kilometers 2000 light years away.
The method they used is quite clever. As the pulsar rotates, it radiates radio energy in beams from the magnetic poles. These radio beams sweep through the surrounding interstellar media. As the radio beams travel through the interstellar media in the general region of the pulsar, they are distorted. By measuring the motion of the pulsar itself, the team was able to use the distorted radio signals from the radio beams to create an astronomical interferometer. But instead of having a baseline of a few thousand kilometers, the effective baseline is about 5 AU, or the distance from the Sun to Jupiter. With such a precise resolution the team found that the emissions of the pulsar’s radio waves are closer to the surface of the pulsar than we had thought. This could help us understand how pulsars generate such strong radio signals.
This trick will only work with radio bright objects like pulsars, but it does demonstrate how clever observations can produce results that are far better than we ever thought possible.
Paper: Ue-Li Pen, et al. 50 picoarcsec astrometry of pulsar emission. MNRAS (May 01, 2014) 440 (1): L36-L40. doi: 10.1093/mnrasl/slu010