Thanks to radio polarization observations, the global properties of interstellar magnetic fields in external galaxies and the field structures on pc and sub-pc sizes in the Milky Way are reasonably well known. However, the processes connecting the features at large and small scales are not understood because the angular resolution in external galaxies is too low with present-day radio telescopes. Most of the existing polarization data are observed in wide frequency bands and hence suffer from very low spectral resolution, which causes depolarization by gradients of Faraday rotation or by different Faraday rotation components within the beam or along the line of sight. Modern radio telescopes are (or will be) equipped with multichannel polarimeters, allowing application of RM Synthesis (section 2.4) and resolving Faraday components along the line of sight. This method is going to revolutionize radio polarization observations.
New and planned telescopes will widen the range of observable magnetic phenomena. The importance of polarimetry for the planned giant optical telescopes still needs to be established, while huge progress is expected in the radio range. The PLANCK satellite and several balloon instruments (PILOT, BLAST-pol) will improve the sensitivity of polarimetry in the submillimeter range at arcminute resolution. The Atacama Large Millimetre Array (ALMA) will provide greatly improved sensitivity at arcsecond resolution for detailed imaging diffuse polarized emission from dust grains and for detection of the Zeeman effect in molecular clouds. High-resolution, deep observations at high frequencies (≥ 5 GHz), where Faraday effects are small, require a major increase in sensitivity for continuum observations which will be achieved by the Jansky Very Large Array (JVLA) and the planned Square Kilometre Array (SKA). The detailed structure of the magnetic fields in the ISM of galaxies and in galaxy halos will be observed, giving direct insight into the interaction between magnetic fields and the various gas components. High angular resolution is also needed to distinguish between regular and anisotropic (sheared) fields and to test various models of the interaction between spiral shocks and magnetic fields. The power spectra of turbulent magnetic fields could be measured down to small scales. The SKA will also allow to measure the Zeeman effect in much weaker magnetic fields in the Milky Way and in nearby galaxies.
The SKA will detect synchrotron emission from Milky Way-type galaxies at redshifts of z ≤ 1.5 (Fig. 51) and their polarized emission to z ≤ 0.5 (assuming 10% polarization). Bright starburst galaxies could be observed at larger redshifts, but are not expected to host ordered or regular fields. Total synchrotron emission, signature of total magnetic fields, could be detected with the SKA out to large redshifts for starburst galaxies, depending on luminosity and magnetic field strength (Fig. 51). However, for fields weaker than 3.25 µG (1 + z)2, energy loss of cosmic-ray electrons is dominated by the inverse Compton effect with photons of the cosmic microwave background, so that their energy is transferred mostly to the X-ray and not to the radio domain. On the other hand, for strong fields the energy range of electrons emitting in the GHz range shifts to low energies, where ionization and bremsstrahlung losses become dominant. The mere detection of synchrotron emission from galaxies at high redshifts will constrain the range of allowed magnetic field strengths.
Figure 51. Total synchrotron emission at 1.4 GHz as a function of redshift z, total magnetic field strength B and total infrared luminosity LIR. The 5 detection limits for 10 h and 100 h integration time with the SKA are also shown (Murphy 2009).
Dynamo theory predicts timescales of amplification and coherent ordering of magnetic fields in galaxies (section 2.6). Based on models describing the formation and evolution of dwarf and disk galaxies, the probable evolution of turbulent and regular magnetic fields can be tested observationally (Arshakian et al. 2009):
The detections of total synchrotron emission in starburst galaxies at z ≤ 4 and of RMs from intervening galaxies at z ≤ 2 (section 4.2) are consistent with dynamo theory. Observed field patterns are so far in agreement with the predictions of the - dynamo (sections 4.4 and 4.7). Progress is needed in numerical MHD simulations. Crucial tests of dynamo action will be possible in young galaxies. Detection of regular fields at z ≥ 3 would call for a faster process than the dynamo. On the other hand, the failure to detect global coherent field patterns in galaxies z ≤ 1 would indicate that the time needed for field ordering is even longer than the - dynamo theory predicts, or that this theory is not applicable. If bisymmetric spiral (BSS) magnetic patterns turn out to dominate, in contrast to nearby galaxies, this would indicate that the fields could be primordial or intergalactic fields which are twisted and amplified by differential rotation.
If polarized emission of galaxies is too weak to be detected, the method of RM grids towards background QSOs could still be applied to measure the strength and structure of regular fields. The accuracy is determined by the polarized flux of the background QSO which could be much higher than that of the intervening galaxy. A reliable model for the structure of the magnetic field of nearby galaxies needs many RM values, hence a sufficiently large number density of polarized background sources, calling for high sensitivity. Faraday rotation in the direction of QSOs could even be measured in galaxies at distances near to those of young QSOs (z ≥ 5). The RM values are reduced by the redshift dilution factor of (1 + z)-2 , so that high RM accuracy is needed.
The SKA will be able to detect 1 µJy sources and measure about 104 RMs per square degree at 1.4 GHz within 12 h integration time. The SKA Magnetism Key Science Project plans to observe an all-sky RM grid with 1 h integration per field (Gaensler et al. 2004) which should contain about 20 000 RMs from pulsars in the Milky Way with a mean spacing of 30' (Fig. 52) and several 100 extragalactic pulsars. At least 107 RMs from compact polarized extragalactic sources at a mean spacing of about 1.5' are expected, about 10000 in the area around M31 (Fig. 53). This fundamental survey will be used to model the structure and strength of the magnetic fields in the foreground, i.e. in the Milky Way, in intervening galaxies, and in the intergalactic medium. A pilot all-sky survey called POSSUM with the Australian SKA Precursor (ASKAP) is planned. MeerKAT, the South African SKA precursor, and APERTIF, the Dutch SKA pathfinder telescope, will have a higher sensitivity but a smaller field of view and will concentrate on measuring RM grids centered on individual objects.
Figure 52. Simulation of pulsars in the Milky Way that will be detected with the SKA (blue), compared to about 2000 pulsars known today (yellow) (from Jim Cordes, Cornell University). Graphics: Sterne und Weltraum.
Figure 53. Simulation of RMs towards background sources (white points) in the region of M31 observable with the SKA within 1 h integration time. Optical emission from M31 is shown in red, diffuse total radio continuum intensity in blue and diffuse polarized intensity in green (from Bryan Gaensler, Sydney University).
Progress is also expected at low radio frequencies. Present-day measurements of galactic magnetic fields by synchrotron emission are limited by the lifetime and diffusion length of the cosmic-ray electrons which illuminate the fields. With typical diffusion lengths of only 1 kpc away from the acceleration sites in star-forming regions, the size of galaxies at centimeter wavelengths is not much larger than that in the optical or infrared spectral ranges. There is indication that magnetic fields probably extend much further into the intergalactic space (section 4.7). The Low Frequency Array (LOFAR), and the Murchison Widefield Array (MWA), the Long Wavelength Array (LWA) and the low-frequency part of the planned SKA will be suitable instruments to search for extended synchrotron radiation at the lowest possible levels in outer galaxy disks and halos and investigate the transition to intergalactic space. While most of the disk is depolarized at low frequencies, polarization should be detectable from the outer regions. Faraday rotation in the Earth's ion osphere and in the Milky Way foreground is strong and need to be corrected for.
The filaments of the local Cosmic Web may contain intergalactic magnetic fields, possibly enhanced by IGM shocks, and this field may be detectable by direct observation of total synchrotron emission or by Faraday rotation towards background sources. For fields of 10-8 - 10-7 G with 1 Mpc coherence length and 10-5 cm-3 electron density, |RM| of 0.1-1 rad m-2 are expected. An overall intergalactic field is much weaker and may only become evident as increased |RM| towards QSOs at redshifts of z > 3 by averaging over a large number of sources. As the Faraday rotation angle increases with 2, searches for low |RM| should preferably be done at low frequencies.
In summary, the SKA and its pathfinders (EVLA, LOFAR, LWA, MWA, APERTIF) and precursors (ASKAP, MeerKAT) will measure the structure and strength of the magnetic fields in the Milky Way, in intervening galaxies, and possibly in the intergalactic medium. Looking back into time, the future telescopes could shed light on the origin and evolution of cosmic magnetic fields. The observational methods are:
Fundamental questions are waiting to be answered: