December 06, 2011
Despite intensive research, the origin of galactic magnetic fields is still unknown. One assumes, however, that they are built up by dynamo processes in which mechanical energy is converted into magnetic energy. Similar processes occur in the interior of the earth, the sun and in the broadest sense, in the gadgets that power bicycle lights through peddling. By revealing the magnetic field structure throughout the Milky Way, the new map provides important insights into the machinery of galactic dynamos.
One way to measure cosmic magnetic fields, which has been known for over 150 years, makes use of an effect known as Faraday rotation. When polarized light passes through a magnetized medium, the plane of polarization rotates. The amount of rotation depends, among other things, on the strength and direction of the magnetic field. Therefore, observing such rotation allows one to investigate the properties of the intervening magnetic fields.
To measure the magnetic field of our own galaxy, radio astronomers observe the polarized light from distant radio sources, which passes through the Milky Way on its way to the Earth. The amount of rotation due to the Faraday effect can be deduced by measuring the polarization of the source at several frequencies.
Each such measurement can only provide information about a single path through the Galaxy. To get a complete picture of the magnetic fields in the Milky Way from Faraday rotation measurements, one must observe many sources distributed across the entire sky. A large international collaboration of radio astronomers have provided data from 26 different projects to give a total of 41,330 individual measurements. On average, the complete catalogue contains approximately one radio source per square degree of sky.
Even with so much data, coverage of the sky is still rather sparse. There remain large regions, especially in the southern sky, where so far only relatively few measurements have been made. Therefore, to obtain a realistic map of the entire sky, one must interpolate between the existing data points. Here, two difficulties arise. First, the respective measurement accuracies vary greatly, and more precise measurements should have a greater influence. Also, the extent to which a single measurement point can provide reliable information about its surrounding environment is not known. This information must therefore be directly inferred from the data itself.
In addition, there is another problem. The measurement uncertainties are themselves uncertain owing to the highly complex measurement process. It so happens that the actual measurement error for a small but significant portion of the data can be more than ten times as large as those indicated by the astronomers. The perceived accuracy of these outliers can strongly distort the resulting map if one does not correct for this effect.