3 Radio data and radio spectra of the 1-Jy sample

In this section we present new 10.5-GHz radio data of 18 of the 26 sources in the 1-Jy sample of GRGs. We have measured total and lobe flux densities of all sources at several frequencies between 325 MHz and 10.5 GHz. Further, we present the radio spectra of the sources and their radio lobes.

3.1 The 10.5-GHz observations

Eight sources in our sample were already observed at 10.5 GHz with the 100-m Effelsberg telescope (Klein et al. 1994; Saripalli et al. 1996; Mack et al. 1997). We have observed the 18 remaining sources with the same instrument and achieving a similar sensitivity of $\sigma_{\rm rms}\!\approx\!1$ mJy beam^-1.

The observations have been carried out in multiple observing sessions between December 1995 and April 1998, using the 4-horn receiver operating at a central frequency of 10.45 GHz and employing a bandwidth of 300 MHz. The beam size at this frequency is $69\hbox{$^{\prime\prime}$ }$(FWHM). For a detailed description of the basic observational technique and data reduction procedure we refer to the paper by Gregorini et al. (1992). The calibration of the flux density scale has been achieved by mapping 3C286 and 3C295, with the flux density scale adopted from Baars et al. (1977). Field sizes and map centers are compiled in Table 2. Mapping was performed by scanning the telescope in azimuth with a speed of $40\hbox{$^\prime$ }$/min, with a scan separation of $20\hbox{$^{\prime\prime}$ }$ in elevation and using all four horns. Difference maps have been computed from all horns to efficiently suppress atmospheric disturbances of the signal. Following the usual restoration technique of the differential maps (Emerson et al. 1979) the maps were transformed into a right ascension - declination system. All maps, with the exception of B0309+411, have been CLEANed to remove sidelobes, applying the algorithm described by Klein & Mack (1995). During the observations of B0309+411, snow was collected on the dish of the telescope. Since this has affected the beam-shape, cleaning was not performed on this source. Also, the snow led to problems with calibrating the data. Therefore the flux density of B0309+411 is estimated to be only correct to 20%. The individual coverages (Table 2) of each source have been averaged to give the final Stokes' I, Q and U maps. Contour plots of the radio maps have been presented in Appendix A.

  

Table 2: Observing log of the 10.5-GHz observations. Column 1 gives the name of the source. Columns 2 and 3 give the B1950.0 coordinates of the center of the map. Column 4 gives the number of coverages used in the final maps (i.e. after removing those which suffered from bad atmospheric conditions). The number of coverages used in the I, Q and U maps are the same. Column 5 gives the size of the area covered. Column 6 gives the observing dates and the number of coverages on that date in the format "number of coverages:month/year''. Columns 7 and 8 give the RMS-noise in the total power maps and in the $Q,\,U$ maps

$$\begin{tabular}{l l l c r@{$\times$}l l l c} \hline \hline \\... ...5:04/98; 7:08/98 & 0.9 & 0.3 \\ \hline \hline\\ \end{tabular}$$

3.2 Other radio data

3.2.1 325-MHz data

We have measured the 325-MHz flux densities and morphological parameters on the radio maps of the WENSS survey. The mosaicing technique used in the observations for WENSS (see Rengelink et al. 1997 for a description) results in a highly uniform coverage of the (u,v)-plane, with baselines as short as $\sim 40\lambda$. WENSS is therefore potentially well suited to obtain accurate flux densities of extended structures. However, since the (u,v)-plane is not as well sampled as with continuous observations, in case of complicated and very extended source structures WENSS cannot map all source components reliably. This is particularly notable in sources such as DA240 or NGC6251. To illustrate this problem we present in Fig. 1 a map of the source DA240 as it appears in the WENSS and as it is published by Mack et al. (1997) from a complete 12-hr WSRT synthesis observation. In the latter case, not only the noise level is much lower, but also the fainter extended radio structures are much better reproduced. Since Mack et al. have published maps resulting from full synthesis WSRT observations for the four spatially largest objects in our sample (NGC315, DA240, 3C236, NGC6251), which will suffer most from this effect, we have measured the flux densities using their maps.

Figure 1: Two contour plots of the giant radio source DA240 at a frequency of 325 MHz. The upper plot is from the WENSS survey. Contours are drawn at $15\times (-1,1,2,4,8,16,32,64,128)$ mJy beam-1. The lower plot is from a 325-MHz WSRT observation by Mack et al. (1997). Contours are at the same levels as in the top plot

3.2.2 1.4-GHz data

All sources have also been observed at 1.4 GHz in the NVSS survey (Condon et al. 1998). The observations for the NVSS survey were done in a "snap-shot'' mode, using the VLA in its D-configuration with baselines down to $\sim 170\lambda$ only. Both these aspects seriously degrade the sensitivity for structures above $10\hbox{$^\prime$ }- 15\hbox{$^\prime$ }$ in angular size. We therefore only present 1.4-GHz flux densities from the NVSS survey for sources smaller than $10\hbox {$^\prime $ }$.

3.2.3 4.85-GHz data

Five GRGs in our sample (NGC315, DA240, 3C236, NGC6251 and B1358+295) have been observed at 4.75 GHz with the 100-m Effelsberg telescope (Parma et al. 1996; Mack et al. 1997). The beam size for these observations is $\sim\!2\hbox{$.\mkern-4mu^\prime$ }5$ (FWHM). For sources without 5-GHz Effelsberg data we have used the 4.85-GHz Greenbank survey (GB6; Gregory et al. 1996) to measure flux densities. This survey has mapped the northern sky up to a declination of $+75\hbox {$^\circ $ }$, using a beam size of $3\hbox{$.\mkern-4mu^\prime$ }2 \times 3\hbox{$.\mkern-4mu^\prime$ }7$ (FWHM). For sources above $+75\hbox {$^\circ $ }$ declination, we have no 5-GHz flux densities. The flux densities in the source catalogue of the GB6 survey were obtained by fitting Gaussians to the observed sources, which, since all GRGs in our sample are larger than the beam size of the GB6 survey, gives unreliable results in our case. Therefore, we have retrieved the original digital FITS-format images, using the SkyView database, and measured the flux densities directly from the maps. Background offsets in the maps have been determined by measuring the mean flux density in an area directly surrounding the source, carefully omitting any significant sources in this area. Discrete sources which overlap with the radio structure of the GRGs have been identified in the higher resolution NVSS and 10.5-GHz Effelsberg radio maps. Their contribution to the measured 4.85-GHz flux density has been subtracted by estimating their 4.85-GHz flux density using a power-law interpolation of their 1.4 and 10.5-GHz flux densities.

   
3.3 Radio spectra

We have measured the total integrated flux densities, $S_{\rm int}$, at 325 MHz, 4.8 GHz and 10.5 GHz. We have also measured the flux densities of the lobes separately at 325 MHz, 1.4 MHz (for sources below $10\hbox {$^\prime $ }$ in angular size), 4.8 GHz (for sources above $10\hbox {$^\prime $ }$ in angular size and declination below $+75\hbox {$^\circ $ }$) and 10.5 GHz. Only in the case of B0309+411 we have not measured the lobe flux densities at 10.5 GHz due to the strongly dominating radio core at that frequency. The flux densities have been tabulated in Table 3.

The radio spectra of the sources with more than two flux density measurements are plotted in Fig. 2, based on the flux densities from Table 3. We have used separate signs for the total integrated flux densities and those of the two lobes. Not all sources have been plotted here; similar radio spectra of the sources B0055+300 (NGC315), B1003+351 (3C236), B0745+560 (DA240) and B1637+826 (NGC6251) can be found in Mack et al. (1997); for the source B1358+305 we refer to the paper by Parma et al. (1996). The source B2147+815 has not been plotted since we only have data at two frequencies for this source (see Table 3).

For several sources the spectrum of the total integrated emission clearly steepens towards higher frequencies (e.g. B0157+405, B0648+733, B0945+734 and B1312+698). This is usually a sign of spectral ageing of the radiating particles in the source. In other cases the spectrum appears to flatten (e.g. B0309+411, B1626+518). Since all these sources have bright radio cores at 10.5 GHz, this must be the result of the radio core having a flat, or inverted, spectrum.

Figure 2: Radio spectra of the GRGs and their radio lobes. For each source we have plotted the total integrated flux density (indicated by "T''), and the integrated flux density of the two lobes (indicated by "E'', "W'', etc.) where available. The name of the source is indicated above each panel. See the text for more details

  

Table 3: Flux densities of the GRGs and their components. Column 1 gives the name of the source. Column 2 indicates which side of the source is named A and B in this table ("N'' stands for north, etc.). Columns 3 to 5 give the integrated flux density, S, at 325 MHz of the whole source and of each of the two sides of the source. Columns 6 through 8 give the flux densities at 1.4 GHz for sources with an angular size below $10\hbox {$^\prime $ }$. Columns 9 through 11 give the flux densities at 4.85 GHz for sources with declination below $+75\hbox {$^\circ $ }$. For sources larger than $10\hbox {$^\prime $ }$, also the flux densities of the lobes have been measured. Columns 12 through 14 give the flux densities at 10.5 GHz