Testing models of radio source space density evolution with the SUMSS survey

We present the population data as predicted by our space density analysis at the SUMSS frequency of 843 MHz. This data demonstrates the potential of the SUMSS survey to trace in detail the change-over in radio source populations from ‘radio monsters’ at high flux densities to a local starbursting population at the survey limit. The exact form of the observed change-over will be used to refine our models of radio source evolution. Extragalactic radio sources at 843 MHz Our recent analyses (Wall & Jackson 1997 and Jackson 1997) have shown that the powerful radio source populations are well described by a dual-population unification scheme. By adopting an elegantly simple aspect-dependent paradigm we have determined that the two ‘parent’ populations have undergone quite separate cosmic evolution histories with the result that their space density distributions are very different. Whilst our working hypothesis is straightforward, the results from our analysis are found to be comprehensively supported by cosmological tests embodied in radio source count and identification data over a wide radio frequency range (151 MHz – 8.4 GHz). The Sydney University Molonglo sky survey (‘SUMSS’, E. Sadler these proceedings) at 843 MHz can be considered to be an ‘intermediate’ radio frequency survey, lying between the ‘low’ frequency surveys (ν < 200 MHz, e.g. 3C, 6C, 7C) and those at ‘high’ frequency (ν > 2 GHz, e.g. Parkes, 87GB, PMN). The change in radio source types which comprise the source count between the low and high frequencies has been shown to be a natural consequence of the increasing contribution from the preferentially-aligned sources which are Doppler-beamed along or very close to our line-of-sight. Surveys below ∼ 200 MHz are almost completely dominated by extended radio galaxies whilst amongst the brightest radio sources at 5 GHz there are almost equal numbers of compact and extended sources. We have applied our dual-population space-density model to the SUMSS frequency and show the predicted integral population mix in Figure 1. The fraction of bright (S843 MHz > 1 Jy) sources which will appear flat-spectrum (α 5 GHz 843 MHz > -0.5) is determined to be ∼ 10 % from a fit to the well-defined source count at 1.4 GHz. The changing contribution from each source type is a result of the differential evolution of the underlying ‘parent’ populations in the radio luminosity function. 2 C. A. Jackson and J. V.Wall Figure 1 shows that at high flux density (S843 MHz > 0.1 Jy) SUMSS will be dominated by the high-power population: FRII radio galaxies with strong optical emission lines (spectral class ‘A’, Hine & Longair 1979) and quasars, their beamed counterparts. There is also a significant contribution from the lowerpower population: radio sources which have only weak, if any, optical emission lines (i.e. FRIs and low-excitation FRIIs, spectral class ‘B’, Hine & Longair 1979 and Laing et al. 1994). In contrast, at the milli-jansky flux density level the dominant populations are expected to be the starburst and Seyfert galaxies, the low-excitation radio galaxies and their beamed counterparts (BL Lac-type sources). Fig. 1. Predicted population mix at 843 MHz as a function of flux density.