How to understand a cosmic elephant

As an Indian story tells it, four blind men went to investigate an elephant. One felt its tusk, and one its tail; another felt its leg, while the fourth ran his hands over its side. And so they came to different conclusions about what an elephant was.

The four blind men and the elephant. Source: Wikimedia Commons
The four blind men and the elephant. Source: Wikimedia Commons

Astronomers have had a similar problem. Using various radio telescopes, they've been making catalogues of cosmic radio sources for about sixty years. Each catalogue has been made at a single radio wavelength, as radio telescopes have been able to observe at only one wavelength at a time. But to understand, say, a galaxy, astronomers need to have information about it at several radio wavelengths—and at other wavelengths of the electromagnetic spectrum too: infrared, optical, ultraviolet and so on.

Unfortunately, the positions of objects in the different catalogues don't match up exactly. The surveys generating the catalogues also differ in factors such as resolution and sensitivity.  So, what's the best way to identify the sources in different catalogues that are counterparts of each other?

Several approaches have been used in the past. Now Jamie Farnes (CAASTRO/University of Sydney) and his collaborators have, for the first time, applied a particularly efficient 'nearest-neighbour' search (a 'K-dimensional tree'), which returns all the possible matching sources within a pre-defined radius of a reference source.  Using this technique, they have integrated the data from dozens of previous radio-source surveys carried out by telescopes in Australia and other countries over the last sixty years. The result is a catalogue of more than 25,000 radio sources—galaxies and quasars—at frequencies from 0.4 to 100 GHz. It's literally a rainbow of radio data.

This work blazes a trail for other researchers who want to create large multi-frequency datasets.

Locating 'cosmic magnets'

But there's more. Using their new catalogue, the researchers have solved a problem about where intriguing 'magnets in space' actually lie.

The surveys that have been incorporated into the catalogue didn't measure just the total strength of the radio waves from the distant galaxies. Instead, they measured another aspect of the waves, its polarisation.

Polarisation can be thought of as the direction electromagnetic waves 'vibrate'. Light can be polarised: some sunglasses filter out light polarised in one direction while letting through other light, and polarisation is used to show 3D movies. Like light, radio waves can be polarised.

Magnetic fields in space alter the polarisation of radio waves (they rotate the direction in which it's polarised). So, astronomers can use polarisation measurements to determine the strength and direction of these magnetic fields. But it hasn't been clear if the fields are intrinsic to the galaxies or quasars emitting the radio waves, or if they're much closer to Earth—in intervening gas clouds, for instance.

Jamie Farnes and his colleagues have now answered that question. Using their new multi-frequency catalogue, they've been able to show that the magnetic field is usually related to the galaxy or quasar itself. What's more, they can discern the different effects of the core of the galaxy or quasar, and of its radio-emitting 'lobes'.

A radio image of the galaxy Cygnus A, showing its core (the bright central dot) and large diffuse lobes. Image courtesy of NRAO/AUI

A radio image of the galaxy Cygnus A, showing its core (the bright central dot) and large diffuse lobes. Image courtesy of NRAO/AUI

New telescopes, such as the international Square Kilometre Array (SKA) and CSIRO's Australian SKA Pathfinder (ASKAP), will be able to extend this work by surveying many more sources (ASKAP's POSSUM project, for instance, will capture perhaps as many as 100,000), at many frequencies simultaneously.

Publication details

Farnes, J.S.; Gaensler, B.M. and Carretti, E. "A broadband polarization catalog of extragalactic radio sources." ApJS, 212, 15 (2014)