Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 334, 799-804 (1998)

Previous Section Next Section Title Page Table of Contents

2. Sample description, observations and data reduction

In order to determine the synchrotron far-infrared component in 3C 47, 3C 207, and 3C 334, we observed their cm-mm core spectral energy distribution, using the NRAO Very Large Array and the Owens Valley Radio Observatory mm-array. We supplemented these data with VLA Archive and literature data, to examine core variability. Some characteristics 1 of the three quasars are listed in Table 1. The parameter [FORMULA] specifies the fractional core flux density at 5 GHz emitted frequency. It should be noted that double-lobed 3C narrow-line radio galaxies of similar radio power have typical [FORMULA] -values of order 0.01 (e.g., Fernini et al. 1997). As mentioned above, 3C 47 and 3C 334 are reported superluminal objects (Vermeulen et al. 1993, Hough et al. 1992), while 3C 207 is a suspected superluminal object (Hough, 1984). The last entry in Table 1 specifies the measured superluminal component speeds. Images of the global QSR radio morphologies can be found in Bridle et al. (1994 - 3C 47 and 3C 334) and Bogers et al. (1994 - 3C 207). They all show large double-lobed structures and fairly prominent one-sided jets.


Table 1. Basic properties of the quasars. [FORMULA] is the IRAS flux density at 60µm, [FORMULA] is the total radio source power at 178 MHz (calculated adopting a spectral index of 0.7), [FORMULA] is the 5 GHz core radio power, [FORMULA] is the core/total flux density ratio at 5 GHz emitted frequency, calculated adopting radio spectral indices of 0 and 0.7 for the core and extended emission respectively.

2.1. VLA cm observations

VLA A- and A/B-array observations at 4.9 GHz (C-band, 6 cm), 8.4 GHz (X, 3.6 cm), 15 GHz (U, 2 cm), 22 GHz (K, 1.2 cm) and 43 GHz (Q, 7 mm), were conducted 1995 Sept. 5, 6, 7. Two IFs were recorded at 50 MHz bandwidth each. Resolving VLA beams varied from 0.9 to 0.2 arcsec, at C- down to K-band. Q-band resolution was 0.4 arcsec, due the fact that only the six inner telescopes of the array were equipped with Q-band receivers. High resolution observations are necessary in order to isolate core from possible jet emission. Single 3 minute snapshot scans were made at C- and X-band, two scans of 3.5 minutes each at U- and K-band, and three 3.5 minute scans at Q-band. Beam reference pointing at Q-band was employed. Secondary phase and amplitude calibrators were observed before and after each scan.

Primary calibrators were 3C 84 and 3C 286, with appropriate baseline constraints, and inferred/adopted flux densities as listed in Table 2. Both primary calibrators were observed during the run on 3C 207 in order to infer the absolute 3C 84 flux density scale (3C 84 was used as primary calibrator for the 3C 47 observations). Calibration uncertainty is dominated by the uncertainty in the absolute flux density of the primary calibrators, which is [FORMULA]. The array performed well: antenna phase and amplitude calibration appeared stable to within [FORMULA], except for Q-band where these figures were [FORMULA].


Table 2. Input flux densities in Jy of the primary calibrators - see text

Data reduction was carried out in Groningen, using standard AIPS routines. The IF1 data being of inferior quality, analysis was restricted to the IF2 data. Tapering was applied for the (high resolution) high frequency data, aiming for comparable resolution at all bands. All cores were detected, except for 3C 47 in K- and Q-band, and 3C 334 in Q-band. For these we derived an upper limit, using the noise levels. The 3C 47 K-band map was dominated by large phase errors and therefore only a fairly high upper limit could be determined. The noise in all maps. including the calibrators, was on the order of one to a few mJy/beam, except for Q-band, where it was about 6 mJy/beam for 3C 47 and 3C 207. Few Q-band antenna's were available for the 3C 334 observations, yielding a low quality image and high Q-band flux density upper limit. We use the dirty map [FORMULA] level as Q-band upper limit. Gaussian core profiles were fitted to determine the integral flux densities, using the AIPS routine IMFIT. The resulting values are listed in Table 3. The ([FORMULA]) errors combine calibration and fitting uncertainties.


Table 3. VLA core flux density values in mJy from Gaussian modelfits

2.2. OVRO 3mm observations

Aperture synthesis observations of the 100 GHz continuum emission in 3C 47, 3C 207, and 3C 334 were carried out with the Owens Valley Millimeter Array on September 22, 1995, hence nearly simultaneously with the VLA observations. There were six 10.4 m diameter telescopes in the array, each equipped with an SIS receiver cooled to 4K. The receivers were tuned to 100 GHz, and typical system temperatures of 300K (single sideband) were achieved. The array was in a compact configuration with baselines of 15-65m E-W and 15-35m N-S, yielding a typical beam size of [FORMULA] (N-S) [FORMULA] [FORMULA] (E-W). An analog correlator with 1 GHz total bandwidth was used for these continuum mode observations. Nearby quasars were observed at 20 minute intervals to track the phase and short term instrument gain, and Uranus ([FORMULA] =120K) was used for the absolute flux calibration. The data were calibrated using the standard Owens Valley array program MMA (Scoville et al. 1993), while DIFMAP (Shepherd et al. 1994) and AIPS were used for mapping and analysis. The uncertainty in the absolute flux measurement is about 15%, and the positional accuracy of the resulting images is better than [FORMULA].

A 100 GHz nuclear continuum source is detected in all three quasars, unresolved by the synthesized beam (see Table 4). The on-source integration time was between 180 and 340 minutes, resulting in noise sensitivity of 1-2 mJy/beam. 3C 207 is a strong source at 100 GHz, and the synthesized map is limited by dynamic range rather than by thermal noise. Evidence for lobe structures is seen in the visibility plots for all three sources, but limited uv -coverage did not permit mapping of these structures. In 3C 47, two bright hot spots (features A and H in Bridle et al. 1994) are clearly detected, located [FORMULA] away from the core. However, their fluxes are uncertain because they lie near the half-power point of the primary beam. Contamination with arcsec scale jet emission at 100 GHz is negligible, due to the spectral steepness of jet radiation combined with the fact that the elongated synthesized beam will not pick up substantial jet emission, as these are not oriented N-S.


Table 4. OVRO 100 GHz core flux densities (mJy)

2.3. Archive and literature data

Archival VLA data (1982-1983) were obtained from projects by Wardle, Perley, and Ekers. These data were reduced in the same way as our 1995 data. To check for, and exclude, calibration induced variability, we compared the archive short spacing flux densities with our 1995 values. Details on the observations and flux densities are listed in Table 5.


Table 5. Arrays, bands and flux densities in mJy from VLA archive data - see text for details.

In addition to the archive VLA data, we examined literature data from other telescopes, in search for variability during the epoch 1975-1995. These observations are listed in Table 6. Since for a number of observations no precise value for the resolution is given, we listed the telescopes used and whenever possible in which configuration.


Table 6. Literature references, with year of observation. All fluxes are in mJy; column 4 gives the literature reference. References: 1=Fernini et al. (1991); 2=Pooley & Henbest (1974); 3=Barthel et al. (1984); 4=Rudnick et al. (1986); 5=Hough et al. (1992); 6=Bridle et al. (1994); 7=Jenkins et al. (1977); 8=Hintzen et al. (1983); 9=Bogers et al. (1994); 10=Hough (1986).

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1998

Online publication: June 2, 1998