Astron. Astrophys. 329, 809-820 (1998)

1. Introduction

We have undertaken a detailed radio study of a sample of 15 radio galaxies at z 2 (Athreya 1996; Athreya et al. 1997a, b) from the newly defined MRC/1Jy complete sample of extragalactic radio sources (McCarthy et al. 1996; Kapahi et al. 1997a, b). One of the primary aims of defining and studying the MRC/1Jy sample was to discover a large and unbiased sample of high redshift radio galaxies (HRRG) for further study. In addition to their many interesting properties, the radio sources in these galaxies are useful as probes of the environment at high redshifts; it is hoped that studies of these objects will considerably improve our understanding of the earliest epochs of galaxy formation.

Our radio observations of the 15 galaxies provide the first extensive data-set to study the general properties of the radio galaxy population at z 2.0. The discovery of steep spectrum radio cores in a majority of these galaxies has been discussed in detail by Athreya et al. (1997a). Results from this study on other aspects including morphology, energetics, spectra, etc. will be discussed in subsequent papers.

We discuss in this paper the large RMs observed in the sample of radio galaxies at z 2, including an intrinsic (redshift corrected) RM of 6000 rad m^-2 in the galaxy 1138-262 at z = 2.17. We also discuss the various mechanisms which have been suggested for generating and aligning magnetic fields and examine these in the light of the much smaller age of the Universe at those redshifts and hence the shorter time available for these processes to operate.

The Rotation measure (RM), which is the slope of the straight line fit to the Polarisation position angle (PPA) versus (: wavelength), may be related to the thermal (as against relativistic) electron density n_e, magnetic field B and the size L of the screen responsible for the Faraday rotation by

where, the line element ds is along the path of the radiation. A more practical relationship to estimate average values of the physical parameters (equivalent to assuming uniform electron density and magnetic field; in case of field reversal(s) within the screen, the values obtained from this relationship would be lower limits to the actual values) is

where RM is in radian m^-2, n_e in cm^-3, L in kpc and B , the parallel component of the magnetic field in micro gauss ( G). The RM is an algebraically additive quantity and all intervening Faraday screens contribute to the RM of a background source. The observed RM is given by , where RM is the intrinsic RM of the screen at redshift z and the sum is over all Faraday screens (F) along the line of sight to the radio source. Studies of astrophysical magnetic fields are often based on RM observations though RM provides only an indirect estimate of the magnetic field since it requires independent estimates of the electron density, path length and the redshift(s) of the Faraday screen(s) which are not always available.

RMs of low redshift radio galaxies: Most radio galaxies at low redshift show observed RM of 30 rad m^-2 (Tabara & Inoue 1980; Leahy et al. 1980). Much of this RM is believed to arise in the interstellar medium (ISM) of our Galaxy (Simard-Normandin et al. 1981; Leahy 1987). However not all the RM is due to the ISM in our galaxy and, in particular, a small fraction of radio galaxies have very large RMs originating close to the source. Of the hundreds of low redshift radio galaxies (z 1) studied, about 15-20 sources have RM 700 rad m^-2 (e.g. Taylor et al. 1992, 1994). The largest RM known is 20,000 rad m^-2 in 3C 295 (Perley & Taylor 1991). Such radio sources are either compact (sub-galactic) in size or are found in x-ray clusters with cooling flows. An excess of RM has been observed in sources located behind the centres of dense clusters indicating that the clusters are responsible for it (Kim et al. 1991). The association of large RM sources with cooling-flow clusters (Taylor et al. 1994) has led to the belief that cooling-flows are in some way responsible for setting up the deep Faraday screens.

RMs of radio galaxies at z 1: The polarisation properties of a sample of 7 sources from the 3CR at 0.6 z 1.2 and Galactic latitude 20 were studied by Pedelty et al. (1989). The median RM_obs for the sample, considering only the highest value in each radio lobe, was 19 rad m^-2, the highest value for the sample being 86 rad m^-2. A more pertinent quantity for comparison with the present study is the mean RM_obs value for each lobe (i.e. the average of the different RM values measured within the resolved radio lobe); the sample has a median of 12 rad m^-2 with the highest being 60 rad m^-2. The median RM_i for the sample, assuming a Faraday screen close to the radio source, is 40 rad m^-2 while the highest value is 197 rad m^-2.

RMs in intervening galaxies: Several studies have dealt with the RMs of distant core-dominated quasars (Welter et al. 1984; Oren & Wolfe 1995). The absorption-line systems seen in the optical spectra of distant quasars may introduce a Faraday rotation in the background radio quasar. Unfortunately, the compact quasars sample only one line of sight through the intervening systems and it is very difficult to separate the RM contributions from our Galactic ISM, the absorption-line systems, the circum-quasar material and the RM intrinsic to the quasar core.

© European Southern Observatory (ESO) 1998

Online publication: December 16, 1997
helpdesk.link@springer.de