PASJ: Publ. Astron. Soc. Japan 55, 351-384, 2003 April 25 c\Delta 2003. Astronomical Society of Japan. ATCA Monitoring Observations of 202 Compact Radio Sourcesin Support of the VSOP AGN Survey Steven J. TINGAY Centre for Astrophysics and Supercomputing, Swinburne University of Technology, P.O. Box 218, Hawthorn, VIC 3122, Australia stingay@astro.swin.edu.auDavid L. J AUNCEY, Edward A. KING, Anastasios K. TZIOUMIS, James E. J. LOVELL, Australia Telescope National Facility, P.O. Box 76, Epping, NSW 2121, Australiaand Philip G. EDWARDS Institute of Space and Astronautical Science, Yoshinodai 3-1-1, Sagamihara, Kanagawa 229-8510 (Received 2003 January 14; accepted 2003 February 23) Abstract The Australia Telescope Compact Array (ATCA) has been used in support of the VLBI Space ObservatoryProgramme (VSOP) space VLBI mission, to monitor the total and polarised flux densities of 202 objects that make up 83% of the VSOP all-sky Survey of compact extragalactic radio sources south of \Delta = + 10\Theta . The primary goal ofthe ATCA observations is to provide information on the total and polarised emission from the compact components in these sources, for correlation with parameters obtained from VSOP imaging observations. These data representthe first high resolution, long timescale flux density monitoring observations of a large number of southern compact radio sources. In the future, comparison of the ATCA and VSOP data will be used to investigate relativistic beamingmodels and identify similarities or differences between the major classes of extragalactic radio sources. As an illustration of the scientific value of the ATCA data we undertake a comparison of the properties of the gamma-rayloud and gamma-ray quiet AGN in the southern component of the VSOP Survey sample, finding that in a flatspectrum sub-sample the gamma-ray loud AGN are more variable than the gamma-ray quiet AGN.Key words: galaxies: active -- radio continuum: galaxies -- surveys -- technique: interferometric 1. Introduction The radio astronomy satellite HALCA (the HighlyAdvanced Laboratory for Communications and Astronomy) was launched from the Kagoshima Space Center on 1997February 12 with the purpose of conducting very long baseline interferometry (VLBI) observations in conjunction withground-based radio telescopes for the VLBI Space Observatory Programme (VSOP) (Hirabayashi et al. 1998). The maximumbaseline length achieved between HALCA and ground radio telescopes is approximately 30000 km, \Delta 3 times longer thanthe baselines accessible to ground-based VLBI, providing a corresponding increase in angular and spatial resolution in theHALCA operating bands of 1.6 and 5 GHz. The majority of the VSOP time has been utilised inobservations allocated by an open and competitively peer reviewed proposal process. Hirabayashi et al. (2000a) reviewmany of the new scientific results to come from this General Observing Time; see also the contents of a special issue ofPublications of the Astronomical Society of Japan (volume 52, #6, 2000). Approximately 25% of VSOP observing time hasbeen devoted to a mission-led survey of 402 compact extragalactic radio sources over the entire sky, using modest ground-based resources in conjunction with the HALCA satellite, at a frequency of 5 GHz. The definition of the Survey sam-ple and descriptions of the Survey aims, ground resources, data reduction, and initial results can be found in Fomalontet al. (2000a), Moellenbrock et al. (2000), Lovell et al. (2000), and Hirabayashi et al. (2000b).In support of the VSOP Survey, several ground-based investigations have been undertaken, or can be utilised, tomaximise the scientific value of the space VLBI observations. These investigations include: a pre-launch ground-based VLBIstudy of the Survey sample, to eliminate any sources that would not be detected on the much longer VSOP baselines(Fomalont et al. 2000b); 15 GHz ground-based VLBI observations, which are resolution-matched to the 5 GHz VSOP obser-vations (Kellermann et al. 1998); and long-term flux density monitoring of as many of the VSOP Survey sample objectsas possible. For the Northern Hemisphere component of the VSOP Survey sample, Kovalev, Kovalev, and Nizhelsky (2000)have observed 232 sources with the RATAN-600 radio telescope. For the Southern Hemisphere component of the VSOPSurvey sample, the Australia Telescope Compact Array (Frater et al. 1992 and references therein) has been used for monitor-ing observations in the 8.6, 4.8, 2.5, and 1.4 GHz bands at 16 epochs between 1996 October (approximately 4 months beforeHALCA launch) and 2000 February. VSOP observations of the Survey sample were ongoing after 2000 February.Here we present the data from the 3.5 years of ATCA monitoring for 202 (primarily Southern Hemisphere) sources thatare listed in the VSOP Survey sample. Apart from the eventual comparison to the VSOP Survey data, the ATCA databaseforms a comprehensive and valuable resource in its own right. It provides the first long-term, high resolution monitoring ofa large number of Southern Hemisphere extragalactic radio 2. Observations and Calibration Observations of 202 radio sources from the VSOP Surveysample (Hirabayashi et al. 2000b) were made over a 3.5 year period with the ATCA (table 1) in the 8.6, 4.8, 2.5, and 1.4 GHzbands. The definition of the VSOP Survey sample was initially an iterative process; as the list was selected and observed it wasrefined by the addition of some sources and the removal of others. As a result, we do not have a completely uniform coveragefor all sources. However, for the great majority of sources, data were collected at every epoch of observation. Each observationutilised one of the four ATCA 6 km array configurations. All observations were made using a 128 MHz bandwidth ineach of two Intermediate Frequency (IF) bands, with all linear products (XX, YY, XY, YX) correlated, giving full polarimet-ric sensitivity. During each observing session each source was observed in a series of 5-minute scans, half of which were con-figured with one IF at 8.640 GHz (3 cm) and one at 4.800 GHz (6 cm), the other half configured with one IF at 2.496 GHz(13 cm) and one at 1.384 GHz (20 cm). Typically, at each epoch two or three scans per source per frequency configuration wereobtained, separated by several hours. The approximate angular resolution obtained with an ATCA6 km configuration at each frequency is as follows: 1 .4 GHz\Theta 7\Lambda \Lambda ; 2 .5 GHz \Theta 4.\Lambda \Lambda 5; 4.8GHz \Theta 2\Lambda \Lambda ; 8.6 GHz \Theta 1\Lambda \Lambda .As a measure of the completeness of this program, we have monitored 83% of the VSOP Survey sources (Hirabayashiet al. 2000b) south of declination + 10\Theta (202 /243). Thiscorresponds to 84% of the VSOP Survey sources south of + 10\Theta chosen to be observed as part of the VSOP Survey(147 /175), based on the 5 GHz pre-launch VLBA survey(Fomalont et al. 2000b). Data reduction was undertaken in the MIRIAD package(Sault et al. 1995) using tasks specifically designed for ATCA data. Calibration of the data was achieved by a short observa-tion of PKS 1934-638, the primary calibrator for the ATCA (Reynolds 1994),1 which is known to have a stable flux den-sity at these frequencies and extremely low linear polarisation, well below the detection limit of our observations. Thissource is used to calibrate the bandpass, flux density scale, the polarisation leakages, and the XY gains using the MIRIADtasks MFCAL and GPCAL. The calibration tables were then transferred to each individual source and the task UVAVERwas used to apply the calibration to the data. The data for each source were then self-calibrated in phase using SELFCALand a point source model, to partially correct the phases for antenna-based errors.As an additional calibration, we analysed the data for 10 sources which are known to be unresolved and relatively sta-ble in flux density, mainly GHz-peaked spectrum or compact steep spectrum sources similar to PKS 1934-638. We firstnormalised the light curves at each frequency for each source by the mean flux density (over the 3.5 years) appropriate tothat source and frequency. We then co-added all 10 normalised light curves at each frequency. In the case of randomly 1 \Delta http://www.atnf.csiro.au/observers/memos/d96783~1.pdf\Theta . Table 1. Epochs of ATCA observations. Epoch MJD Duration (hr)(observation midpoint) 1996 Oct. 12/13/14 50369.0 47 1996 Dec. 13/14 50430.5 25 1997 Feb. 08/09 50487.5 39 1997 May 26/27 50594.5 32 1997 Jul. 05/06 50634.5 24 1997 Sep. 05/06/07 50797.0 58 1997 Nov. 18/19 50770.5 31 1997 Dec. 28/29 50810.5 24 1998 Feb. 15/16/17 50860.0 54 1998 May 24/25 50957.5 31 1998 Aug. 05/06 51030.5 30 1998 Oct. 02/03 51088.5 25 1998 Nov. 30/Dec.01 51147.5 24 1999 Feb. 13/14 51222.5 24 1999 Jun. 01/02 51330.5 24 2000 Feb. 24/25 51598.5 23 distributed errors at each epoch, these averages should tend tounity. If a systematic error exists at any given epoch then the deviation from unity at that epoch will give an estimate of theerror in the flux density scale at that epoch. At all epochs and all frequencies the systematic errors on the flux density scalewere less than 5% and most corrections were less than 2%. These corrections have been applied to the data.The MIRIAD task UVFLUX was used to fit a point source to the data for each object at each frequency, using only the dataon baselines between antennas one and six (the 6 km maximum baseline of the ATCA). In UVFLUX a scalar average of thetotal intensity amplitude was used in the fit to the point source model. A scalar average is appropriate in this case since thetotal intensity measurements from each 10 second correlator integration period are in the high signal-to-noise regime andthe scalar average will not suffer from structure-phase decorrelation of the amplitudes for any sources that have a vary-ing structure-phase over the course of the observations. On the other hand, estimates of the polarised intensities (Stokes Qand U ) require a vector average since the individual 10 secondintegrations from the correlator are dominated by noise in many cases. A scalar average in this case would cause an over-estimate of the polarised flux density due to the positive noise bias. A vector average is far more appropriate in this case.However, if any sources have significantly extended polarised emission, structure-phase decorrelation of the amplitudes willcause a slight underestimate of the polarised flux density, and therefore the percentage polarisations.UVFLUX reports the sum of the squared deviations from the best-fit amplitude (for both vector and scalar averages) andthe number of visibilities used in the fit, which can be used to derive an error on the best-fit point source flux densities (I , Q, and U ). Measured in this way, the error includes boththermal noise and source structure contributions (discussed further below). No. 2] ATCA Observations of Compact Radio Sources 353 Table 2. Approximate flux densities and spectral indices for the 17 sources dropped from the ATCA monitoring. Source I_nu Flag 8.6 4.8 2.5 1.4 GHz alpha J0126-0123 0.06 0.10 0.15 0.30 -0.87 e J0134-3629 0.04 0.04 0.11 0.21 0.00 e J0322-3712 0.02 0.04 0.09 0.32 -1.11 e J0455-3006 0.04 0.11 0.36 0.92 -1.63 e J0620-3711 0.04 0.07 0.15 0.40 -0.90 e J0648-3957 0.02 0.06 0.30 0.44 -1.77 e J1048-4114 0.41 0.45 0.65 1.16 -0.19 c J1145-4836 0.35 0.51 0.90 1.74 -0.61 e J1146-3328 0.31 0.30 0.47 0.83 0.05 c J1242-0446 0.29 0.61 1.48 3.02 -1.20 e J1530-4231 0.06 0.25 0.73 1.90 -2.30 e J1644-7715 0.30 0.24 0.27 0.41 0.36 c J1651+0459 0.13 0.42 1.75 5.12 -1.89 e J1702-7741 0.21 0.38 0.67 0.96 -0.95 e J1720-0058 0.25 0.50 1.25 3.70 -1.11 e J2341-5816 0.26 0.54 1.10 2.04 -1.18 e J2358-6054 0.10 0.37 1.10 2.18 -2.10 e By considering only the data on the 6 km baseline in thisway, we have used the ATCA as an angular size filter, isolating the compact core of each source and filtering out as muchextended emission seen on short baselines as possible. This was done because sparsely sampled data of the type collectedfrom these observations cannot be used to estimate the total flux density (core plus extended) very well. In addition, theshortest baseline available in the 6 km configurations varies substantially in length with the configuration, making compar-isons from epoch to epoch and between sources difficult, when based on the short baseline data. Also, the shorter baselinesat the lower frequencies suffer from confusion. However the core flux density is better defined on the 6 km baseline at allfrequencies, independent of configuration, since the baseline length changes only very little between different 6 km config-urations, making it most useful for comparisons of the type we wish to undertake. 3. Results A number of sources were observed for the first fewepochs, then dropped from the ATCA monitoring, since they showed flux densities below the VSOP detection limit.Only one of these 17 sources had been assigned a VSOP observation code, J2341-5816 (vs10j). The sources droppedwere: J0126-0123, J0134-3629, J0322-3712, J0455-3006, J0620-3711, J0648-3957, J1048-4114, J1145-4836,J1146-3328, J1242-0446, J1530-4231, J1644-7715, J1651 + 0459, J1702-7741, J1720-0058, J2341-5816,and J2358-6054. Table 2 contains a brief summary of the approximate mean flux densities (in Jy) and spectral indicesmeasured from the available ATCA data for these 17 sources. The flag column in table 2 is defined and discussed below.Table 3 lists parameters for the 185 remaining sources observed at most epochs on the 6 km baseline for the four ATCA bands, including: \Lambda I \Xi , the mean flux density in Jy; M, the totalflux density variability index; \Lambda m\Xi , the mean linearly polarisedpercentage; and \Lambda \Lambda \Xi , the mean spectral index between 4.8 and8.6 GHz ( I \Pi \Theta \Lambda ). The variability index, M, is defined as theRMS variation from the mean flux density, divided by the mean flux density. It should be noted that over the period of these ob-servations, significant radio frequency interference (RFI) was present in the 2.5 GHz band at the ATCA, and to a lesser extentin the 1.4 GHz band. The RFI in the 2.5 GHz data has been removed as much as possible by careful editing but some of thescatter in the visibility amplitudes for some sources at some times at this frequency may be due to residual RFI in the data.Finally, table 3 (and table 2) contains a flag which gives an indication of the amount and type of extended emission foreach source. These flags were assigned based on a comparison of the data from the 6 km baseline only to the data on all base-lines, including the shortest baselines available for any given observation. As discussed above these flags will not give anaccurate assessment of the amount and type of extended structure. However, this comparison is useful at least in determiningif significant extended structure exists. The flag designations are as follows:* c: highly compact structure, little evidence for structureon baselines less than 6 km in length; * e: significantly extended structure at a majority offrequencies and on baselines less than 6 km in length; * l: evidence for extended structure mainly at 1.4 GHz or2.5 GHz, with the source appearing to be more compact at the higher frequencies. Limited (u, v) sampling means that these structure flags areapproximate only. It is likely that the c sources, when observed with a full imaging observation, will reveal at least someextended structure at some frequencies. Likewise, the majority of the e sources are likely to still be core dominated. Finally, 354 S. J. Tingay et al. [Vol. 55, Table 3. Mean flux densities \Delta I\Theta \Theta , variability indices M\Theta , mean percentage linear polarisations \Delta m\Theta \Theta , mean spectral indices \Delta \Lambda \Theta , and structure flags (seetext for more details) for the 185 (= 202 - 17) VSOP Survey sources monitored with the ATCA. Source \Lambda I\Theta \Xi M\Theta \Lambda m\Theta \Xi \Lambda \Lambda \Xi Flag 8.6 4.8 2.5 1.4 8.6 4.8 2.5 1.4 8.6 4.8 2.5 1.4 J0006-0623 2.64 2.53 2.16 2.00 0.12 0.08 0.04 0.02 4.80 2.51 4.60 5.00 0.07 lJ0059 + 0006 1.01 1.37 1.83 2.37 0.04 0.02 0.03 0.01 4.82 7.25 5.44 3.56 -0 .52 cJ0106-4034 5.02 3.56 1.64 0.80 0.10 0.06 0.09 0.11 2.11 1.34 1.17 < 0.50 0.57 cJ0108 + 0135 2.19 2.14 2.19 2.36 0.23 0.11 0.04 0.03 1.38 1.98 2.79 3.65 0.01 c J0115-0127 0.88 0.97 1.05 0.89 0.08 0.09 0.06 0.07 6.08 6.32 2.99 1.93 -0.16 e J0116-1136 0.82 0.87 1.07 1.25 0.11 0.10 0.09 0.09 1.24 2.18 3.95 4.20 -0.10 lJ0119 + 0829 0.75 1.15 1.73 2.45 0.02 0.02 0.03 0.01 1.34 1.60 < 0.50 < 0.50 -0.72 cJ0120-2701 0.84 0.87 0.84 0.86 0.05 0.05 0.06 0.06 6.45 6.07 5.30 5.35 -0 .04 eJ0121 + 0422 1.41 1.37 1.17 1.02 0.07 0.08 0.10 0.07 1.78 1.99 1.99 1.14 0.05 c J0125-0005 1.64 1.54 1.55 1.51 0.02 0.05 0.04 0.04 2.86 3.21 2.72 3.17 0.11 e J0149 + 0555 1.27 1.33 1.10 0.82 0.05 0.03 0.03 0.05 3.94 3.89 2.44 1.23 -0.07 cJ0153-3310 0.66 0.88 1.07 1.15 0.06 0.04 0.02 0.02 1.70 1.35 0.79 1.75 -0 .49 eJ0155-4048 0.78 1.13 1.56 1.98 0.02 0.01 0.02 0.01 0.94 0.82 < 0.50 < 0.50 -0.63 cJ0202-7620 0.18 0.18 0.25 0.42 0.14 0.19 0.30 0.37 3.72 4.56 5.49 5.23 -0 .02 eJ0204-1701 1.09 1.16 1.18 1.17 0.07 0.04 0.04 0.04 1.13 3.85 4.44 3.16 -0 .10 c J0210-5101 2.83 3.01 3.25 3.40 0.14 0.08 0.08 0.06 1.65 1.43 1.57 1.84 -0.12 eJ0217 + 0144 1.54 1.26 0.92 0.72 0.36 0.37 0.34 0.20 1.81 0.95 1.38 0.98 0.30 c J0224 + 0659 0.67 0.57 0.42 0.37 0.15 0.08 0.04 0.10 1.21 2.16 3.66 4.04 0.24 lJ0239 + 0416 0.58 0.69 0.78 0.79 0.14 0.13 0.06 0.04 4.60 4.23 2.52 2.49 -0 .33 cJ0240-2309 1.70 2.88 4.71 6.04 0.05 0.03 0.02 0.01 3.42 3.47 2.33 0.76 -0 .90 c J0241-0815 2.63 2.47 1.46 0.99 0.14 0.09 0.11 0.08 < 0.50 < 0.50 < 0.50 0.59 0.10 cJ0253-5441 1.11 0.91 0.75 0.74 0.13 0.10 0.08 0.04 3.25 3.15 1.81 4.56 0.34 l J0303-6211 1.98 2.27 2.44 2.43 0.02 0.02 0.04 0.04 2.18 2.26 2.18 2.17 -0.23 cJ0309-6058 1.25 1.26 1.10 0.98 0.05 0.08 0.09 0.06 2.27 2.21 2.81 3.14 -0 .01 cJ0312 + 0133 0.32 0.35 0.37 0.39 0.10 0.08 0.12 0.16 3.85 4.24 5.12 5.55 -0 .16 c J0339-0146 2.44 2.31 1.99 1.95 0.11 0.09 0.08 0.07 1.88 2.87 4.51 5.70 0.09 lJ0348-2749 1.71 1.58 1.14 0.90 0.24 0.17 0.19 0.18 2.50 2.03 3.10 3.64 0.13 e J0403-3605 2.06 1.82 1.41 1.01 0.24 0.16 0.06 0.05 1.79 0.94 1.27 < 0.50 0.17 cJ0405-1308 1.45 2.30 3.01 3.86 0.15 0.05 0.03 0.01 2.57 2.40 2.23 1.78 -0 .79 eJ0406-3826 1.99 1.93 1.48 1.04 0.28 0.24 0.17 0.09 1.53 1.19 1.11 1.66 0.03 c J0414 + 0534 0.34 0.68 1.20 1.86 0.07 0.03 0.05 0.03 0.95 < 0.50 < 0.50 < 0.50 -1.19 lJ0423-0120 3.17 2.79 2.25 2.00 0.09 0.08 0.09 0.09 2.08 1.86 2.23 2.12 0.23 c J0424-3756 1.67 1.38 0.83 0.50 0.08 0.07 0.14 0.10 1.30 0.86 1.40 1.63 0.32 cJ0428-3756 1.41 1.43 1.24 1.09 0.22 0.21 0.16 0.12 5.75 5.11 4.16 4.26 -0 .01 cJ0433 + 0521 4.11 4.14 3.65 3.25 0.18 0.15 0.09 0.04 2.89 2.74 4.18 2.19 0.01 l J0437-1844 0.80 0.95 0.92 0.65 0.05 0.05 0.04 0.08 0.63 0.80 0.82 0.80 -0.29 lJ0440-4333 3.36 3.19 3.50 4.53 0.10 0.10 0.05 0.02 1.70 0.61 < 0.50 < 0.50 0.09 eJ0442-0017 1.37 1.68 1.92 1.76 0.11 0.10 0.11 0.09 1.25 1.99 1.04 0.87 -0 .32 eJ0450-8101 1.67 1.33 0.98 0.87 0.25 0.21 0.15 0.07 2.04 1.67 2.34 2.99 0.39 c J0453-2807 1.51 1.72 2.08 2.37 0.09 0.05 0.03 0.02 4.25 4.73 4.88 0.76 -0.22 c J0457-2324 2.32 2.04 1.50 1.23 0.16 0.15 0.15 0.13 1.93 1.44 1.41 1.93 0.23 lJ0459 + 0229 0.96 1.43 1.96 1.96 0.05 0.06 0.07 0.14 0.71 < 0.50 < 0.50 < 0.50 -0.67 cJ0501-0159 1.28 1.37 1.45 1.59 0.10 0.12 0.11 0.11 0.95 1.14 1.71 0.76 -0 .11 eJ0503 + 0203 1.51 2.05 2.44 2.21 0.02 0.02 0.03 0.04 < 0.50 < 0.50 < 0.50 < 0.50 -0.52 cJ0509 + 0541 0.55 0.58 0.54 0.50 0.09 0.11 0.11 0.08 2.23 2.34 3.08 1.20 -0 .04 c J0519-4546 1.17 1.24 2.28 4.04 0.05 0.13 0.10 0.09 < 0.50 4.48 9.82 9.97 -0.07 eJ0522-3627 2.65 2.47 3.20 4.71 0.15 0.20 0.22 0.26 1.59 2.96 2.65 1.36 0.16 e J0525-4557 1.51 1.77 1.77 1.71 0.13 0.11 0.13 0.15 1.53 1.37 1.17 0.79 -0.27 e No. 2] ATCA Observations of Compact Radio Sources 355 Table 3. (Continued.) Source \Lambda I\Theta \Xi M\Theta \Lambda m\Theta \Xi \Lambda \Lambda \Xi Flag 8.6 4.8 2.5 1.4 8.6 4.8 2.5 1.4 8.6 4.8 2.5 1.4 J0532 + 0732 1.78 1.79 1.98 2.38 0.16 0.09 0.08 0.06 0.93 0.79 1.05 0.84 -0.04 lJ0538-4405 3.93 3.48 3.14 2.89 0.35 0.30 0.24 0.22 1.38 1.28 1.33 1.15 0.15 l J0539-2839 1.13 0.95 0.67 0.67 0.34 0.34 0.23 0.18 0.95 1.14 2.41 1.76 0.31 cJ0541-0541 0.90 1.01 1.10 1.16 0.16 0.15 0.10 0.08 2.25 3.13 4.13 4.76 -0 .20 lJ0604-3155 0.12 0.44 1.24 2.08 0.38 0.25 0.06 0.04 25.06 8.08 2.77 0.69 -2 .16 eJ0607-0834 2.20 2.32 2.37 2.20 0.17 0.15 0.10 0.07 1.32 1.23 2.45 3.39 -0 .10 lJ0609-1542 6.53 5.04 3.66 3.06 0.22 0.26 0.24 0.13 1.44 0.97 1.86 1.76 0.45 c J0616-3456 0.77 1.33 2.03 2.76 0.04 0.02 0.03 0.01 1.83 0.61 < 0.50 < 0.50 -0.93 cJ0627-3529 0.76 0.86 0.95 1.20 0.08 0.08 0.11 0.06 1.61 2.77 2.39 2.18 -0 .22 eJ0627-0553 2.50 4.92 9.00 13.23 0.04 0.03 0.03 0.01 5.45 6.53 13.96 6.81 -1 .15 eJ0635-7516 5.92 6.31 5.16 4.07 0.06 0.02 0.07 0.10 1.12 1.18 2.29 4.34 -0 .11 eJ0644-3459 0.64 0.85 0.81 0.67 0.14 0.09 0.04 0.03 2.05 2.03 0.59 1.26 -0 .48 c J0739 + 0137 1.65 1.70 1.80 2.03 0.13 0.09 0.05 0.03 5.35 6.82 7.87 8.53 -0.05 cJ0743-6726 1.04 1.16 1.40 1.59 0.04 0.09 0.18 0.31 6.99 8.08 7.86 5.44 -0 .20 eJ0745-0044 2.12 2.05 1.27 0.70 0.10 0.04 0.06 0.05 0.84 0.61 0.57 0.59 0.05 c J0808-0751 2.36 2.04 1.69 1.65 0.16 0.15 0.17 0.12 3.13 2.66 1.91 2.09 0.26 cJ0811 + 0146 0.90 0.85 0.73 0.62 0.19 0.18 0.18 0.15 7.64 8.09 7.64 5.41 0.10 c J0820-1258 0.45 0.49 0.56 0.76 0.14 0.12 0.09 0.08 4.03 5.04 5.95 4.86 -0.16 eJ0825 + 0309 1.40 1.31 1.24 1.19 0.24 0.14 0.08 0.05 3.97 3.63 3.81 4.77 0.07 c J0831 + 0429 1.01 0.94 0.86 0.82 0.11 0.09 0.07 0.06 3.16 2.45 4.10 4.70 0.10 cJ0900-2808 3.02 3.18 2.18 1.72 0.11 0.08 0.10 0.06 2.22 0.52 1.56 < 0.50 -0.10 cJ0921-2618 1.62 1.44 1.28 1.26 0.09 0.08 0.06 0.06 1.69 3.05 2.41 0.67 0.20 c J1018-3144 0.82 1.41 2.36 3.52 0.03 0.02 0.03 0.02 1.12 < 0.50 < 0.50 < 0.50 -0.92 cJ1035-2011 1.36 1.01 0.71 0.80 0.06 0.11 0.06 0.04 0.87 3.03 6.01 5.77 0.52 l J1037-2934 1.61 1.52 1.26 1.02 0.21 0.19 0.20 0.21 3.72 3.10 3.05 2.41 0.10 cJ1041 + 0610 1.72 1.77 1.64 1.31 0.06 0.02 0.03 0.02 1.59 2.79 2.65 2.37 -0 .05 eJ1041 + 0242 0.27 0.47 0.57 1.35 0.10 0.22 0.29 0.22 5.93 4.13 1.00 < 0.50 -0.83 l J1051-3138 0.53 0.62 0.71 0.72 0.11 0.08 0.09 0.15 2.25 3.64 3.51 3.13 -0.28 cJ1058 + 0133 3.30 3.05 2.90 3.07 0.24 0.13 0.04 0.03 2.18 2.33 2.16 2.22 0.09 e J1058-8003 2.22 2.03 1.28 0.77 0.17 0.20 0.24 0.22 5.92 3.14 1.09 0.96 0.16 cJ1107-4449 2.51 2.63 2.42 2.44 0.07 0.08 0.06 0.10 2.10 2.39 2.97 2.28 -0 .07 cJ1118-4634 0.89 1.01 1.07 1.56 0.15 0.15 0.13 0.11 5.46 5.50 4.58 1.46 -0 .24 e J1127-1857 1.87 1.49 1.02 0.70 0.37 0.37 0.34 0.27 1.73 2.42 2.23 1.89 0.45 cJ1146-2447 1.00 1.41 1.68 1.44 0.03 0.04 0.03 0.04 1.58 0.60 1.15 1.60 -0 .58 cJ1147-3812 2.31 2.08 1.61 1.40 0.25 0.23 0.18 0.23 2.52 2.09 2.12 1.96 0.18 c J1150-0023 1.21 1.60 2.14 2.62 0.04 0.03 0.03 0.02 2.37 2.88 4.29 4.64 -0.47 lJ1205-2634 0.48 0.45 0.48 0.63 0.07 0.13 0.18 0.26 2.19 2.21 6.31 4.45 0.14 e J1209-2406 0.59 0.56 0.51 0.46 0.11 0.11 0.07 0.07 1.50 1.12 1.15 1.95 0.08 cJ1215-1731 2.54 1.97 1.42 1.45 0.06 0.07 0.07 0.04 1.84 1.14 0.62 1.17 0.44 e J1218-4600 1.23 2.08 3.36 4.78 0.02 0.03 0.03 0.01 6.22 2.23 0.65 < 0.50 -0.88 eJ1224 + 0330 1.06 1.12 1.19 1.29 0.06 0.04 0.04 0.03 1.42 1.33 1.29 1.33 -0 .09 lJ1229 + 0203 32.53 31.95 34.66 35.82 0.14 0.07 0.07 0.13 4.41 3.78 1.50 1.04 0.01 e J1232-0224 0.79 0.71 0.63 0.72 0.05 0.12 0.06 0.21 2.01 2.99 2.84 2.80 0.19 eJ1239-1023 1.26 1.30 1.37 1.40 0.06 0.04 0.03 0.03 2.97 3.86 3.12 3.97 -0 .05 cJ1246-0730 0.87 0.85 0.71 0.56 0.14 0.09 0.04 0.08 3.45 2.58 2.38 3.91 0.03 c J1246-2547 1.70 1.38 1.07 1.02 0.36 0.26 0.12 0.07 2.22 3.05 3.45 3.31 0.30 lJ1256-0547 22.29 14.71 9.68 7.64 0.12 0.19 0.09 0.04 2.05 4.48 3.47 2.72 0.75 e J1257-3155 1.47 1.62 1.41 1.10 0.12 0.08 0.07 0.06 3.19 2.83 3.87 5.55 -0.17 e 356 S. J. Tingay et al. [Vol. 55, Table 3. (Continued.) Source \Lambda I\Theta \Xi M\Theta \Lambda m\Theta \Xi \Lambda \Lambda \Xi Flag 8.6 4.8 2.5 1.4 8.6 4.8 2.5 1.4 8.6 4.8 2.5 1.4 J1316-3338 1.39 1.25 1.15 1.13 0.19 0.20 0.18 0.14 1.75 2.51 3.13 2.97 0.17 eJ1325-4301 6.79 5.88 5.84 4.52 0.06 0.08 0.08 0.11 < 0.50 < 0.50 0.67 1.57 0.26 eJ1337-1257 4.66 3.55 2.54 1.97 0.30 0.24 0.14 0.17 2.47 3.34 2.30 3.15 0.41 l J1351-1449 0.50 0.75 1.05 1.21 0.03 0.05 0.03 0.02 0.78 < 0.50 < 0.50 < 0.50 -0.67 c J1357-1744 0.86 1.16 1.41 1.42 0.03 0.03 0.02 0.04 4.64 4.92 3.67 1.07 -0.50 lJ1419-1928 0.46 0.47 0.41 0.49 0.08 0.07 0.16 0.15 3.44 2.44 2.42 4.19 -0 .03 eJ1424-4913 3.54 5.06 6.61 8.06 0.06 0.03 0.03 0.02 5.59 3.22 1.61 1.74 -0 .61 eJ1427-4206 3.87 3.35 3.20 3.53 0.19 0.14 0.09 0.05 2.30 2.71 2.21 1.63 0.23 e J1435-4821 0.64 1.03 1.52 2.05 0.02 0.02 0.03 0.02 2.35 1.01 < 0.50 < 0.50 -0.81 c J1445 + 0958 0.60 1.07 1.80 2.41 0.02 0.03 0.03 0.01 2.09 2.02 1.06 0.63 -0.98 cJ1454-3747 2.17 2.01 1.52 1.03 0.27 0.25 0.13 0.09 3.34 2.10 1.99 3.55 0.10 l J1501-3918 0.66 1.17 1.93 2.76 0.02 0.02 0.03 0.02 0.61 < 0.50 < 0.50 < 0.50 -0.97 cJ1507-1652 2.52 2.58 2.59 2.63 0.11 0.06 0.03 0.05 1.75 0.77 0.87 1.23 -0 .04 lJ1512-0905 1.76 1.86 1.95 2.19 0.24 0.12 0.06 0.10 2.75 2.87 4.01 3.75 -0 .13 e J1513-1012 1.28 1.06 0.72 0.59 0.12 0.08 0.07 0.06 1.40 1.29 2.50 4.46 0.32 lJ1516 + 0015 1.09 0.96 0.76 0.70 0.05 0.03 0.03 0.03 0.54 1.00 1.82 2.56 0.21 l J1517-2422 2.73 2.77 2.47 2.12 0.12 0.10 0.11 0.13 3.58 4.05 3.42 5.38 -0.02 lJ1522-2730 1.68 1.74 1.47 1.08 0.13 0.15 0.21 0.24 2.98 2.45 2.17 2.65 -0 .05 lJ1526-1351 0.75 1.20 1.88 2.66 0.03 0.02 0.02 0.02 2.43 2.56 0.68 < 0.50 -0.80 e J1534 + 0131 0.79 0.91 1.06 1.18 0.15 0.09 0.06 0.05 1.03 1.72 1.96 < 0.50 -0.26 lJ1543-0757 0.69 1.00 1.33 1.54 0.02 0.02 0.03 0.05 < 0.50 < 0.50 < 0.50 < 0.50 -0.61 cJ1546 + 0026 0.67 0.94 1.32 1.67 0.07 0.05 0.04 0.05 < 0.50 < 0.50 < 0.50 < 0.50 -0.59 lJ1549 + 0237 3.14 2.44 1.43 0.86 0.30 0.18 0.18 0.21 2.20 1.70 1.63 3.31 0.37 c J1550 + 0527 3.12 3.14 2.58 2.21 0.08 0.06 0.06 0.03 3.34 2.24 1.10 1.28 -0.02 c J1556-7914 2.41 3.73 4.82 5.35 0.03 0.02 0.03 0.02 < 0.50 < 0.50 < 0.50 < 0.50 -0.74 cJ1617-7717 2.65 3.07 3.42 3.67 0.05 0.06 0.03 0.03 2.21 3.55 3.40 2.39 -0 .25 lJ1624-6809 1.21 1.69 1.82 1.41 0.04 0.04 0.02 0.03 < 0.50 < 0.50 0.58 0.53 -0.56 cJ1625-2527 1.97 1.76 1.60 1.68 0.36 0.26 0.21 0.15 1.28 1.18 1.55 1.73 0.17 e J1626-2951 2.63 2.38 2.09 2.03 0.12 0.14 0.07 0.09 3.59 4.57 6.73 8.10 0.17 l J1642-0621 0.69 0.70 0.72 0.75 0.38 0.43 0.36 0.34 2.75 2.91 2.65 1.98 -0.05 cJ1647-6438 0.69 0.80 0.87 0.84 0.28 0.28 0.32 0.20 5.65 4.93 5.08 5.13 -0 .23 cJ1658 + 0741 2.13 1.93 1.52 1.32 0.06 0.07 0.12 0.10 3.05 5.36 3.76 4.37 0.18 c J1658 + 0515 0.98 1.25 1.37 1.53 0.04 0.04 0.04 0.04 6.31 5.68 4.33 1.11 -0.42 eJ1658-0739 1.13 1.32 1.02 0.75 0.03 0.03 0.06 0.05 1.09 < 0.50 0.60 0.65 -0.26 e J1723-6500 3.67 4.64 4.48 3.54 0.02 0.02 0.04 0.03 < 0.50 < 0.50 < 0.50 < 0.50 -0.40 cJ1726-6427 0.56 1.26 2.59 3.96 0.01 0.02 0.03 0.02 0.85 < 0.50 < 0.50 < 0.50 -1.37 cJ1733-1304 7.80 7.01 5.59 5.20 0.38 0.26 0.10 0.08 3.69 2.93 2.83 2.47 0.08 e J1743-0350 5.22 4.43 3.01 1.75 0.10 0.11 0.25 0.16 1.37 0.96 0.76 0.63 0.29 cJ1744-5144 2.46 3.88 5.59 6.94 0.01 0.02 0.02 0.03 < 0.50 < 0.50 < 0.50 < 0.50 -0.77 c J1745-0753 1.09 1.19 1.19 1.08 0.26 0.23 0.14 0.09 1.93 1.56 2.38 1.94 -0.16 eJ1751 + 0939 3.44 2.41 1.50 1.04 0.31 0.28 0.32 0.29 1.73 1.41 2.50 2.01 0.61 c J1804 + 0101 0.70 0.77 0.82 0.81 0.04 0.07 0.04 0.06 2.50 2.30 1.54 0.65 -0.17 cJ1809-4552 0.97 0.84 0.73 0.71 0.06 0.04 0.05 0.03 0.87 1.95 2.78 2.47 0.25 e J1819-5521 0.76 0.84 0.87 0.90 0.08 0.10 0.09 0.11 1.92 1.85 1.96 1.08 -0.17 c J1833-2103 2.53 5.87 8.70 9.87 0.39 0.24 0.13 0.06 1.60 < 0.50 < 0.50 < 0.50 -1.39 eJ1837-7108 2.42 2.39 2.02 1.49 0.08 0.06 0.06 0.06 2.29 1.77 0.88 1.54 0.03 c J1911-2006 2.62 2.74 2.69 2.57 0.25 0.21 0.19 0.12 1.06 1.27 1.48 2.14 -0.08 cJ1912-8010 0.81 0.93 1.01 1.12 0.09 0.08 0.04 0.03 2.87 2.73 3.59 3.71 -0 .25 c No. 2] ATCA Observations of Compact Radio Sources 357 Table 3. (Continued.) Source \Lambda I\Theta \Xi M\Theta \Lambda m\Theta \Xi \Lambda \Lambda \Xi Flag 8.6 4.8 2.5 1.4 8.6 4.8 2.5 1.4 8.6 4.8 2.5 1.4 J1924-2914 14.61 13.87 12.92 12.48 0.19 0.16 0.09 0.05 2.55 2.82 3.54 3.79 0.07 l J1932-4536 0.71 0.74 0.76 0.79 0.02 0.08 0.10 0.09 3.04 3.17 3.32 4.20 -0.06 lJ1937-3958 0.95 0.98 0.96 0.93 0.09 0.07 0.05 0.05 1.95 3.41 2.94 3.04 -0 .08 cJ1939-6342 2.82 5.84 11.31 14.98 0.02 0.01 0.01 0.01 < 0.50 < 0.50 < 0.50 < 0.50 -1.24 cJ1940-6907 0.67 0.90 1.26 1.54 0.09 0.09 0.02 0.03 1.71 2.01 0.95 1.06 -0 .49 lJ1949-1957 0.90 1.05 1.19 1.40 0.09 0.04 0.04 0.05 2.23 4.37 1.96 < 0.50 -0.27 c J1957-3845 3.30 3.21 2.46 1.53 0.30 0.26 0.16 0.11 2.73 3.09 1.99 2.65 0.05 cJ2000-1748 1.70 1.35 0.91 0.67 0.46 0.39 0.35 0.32 1.45 1.32 1.33 1.27 0.38 c J2003-3251 0.73 0.92 0.70 0.40 0.04 0.04 0.04 0.05 0.81 < 0.50 0.73 1.01 -0.40 eJ2009-4849 1.18 1.29 1.27 1.26 0.09 0.09 0.08 0.11 3.49 3.68 4.19 3.27 -0 .13 cJ2101 + 0341 0.75 0.71 0.60 0.51 0.08 0.09 0.08 0.11 2.73 2.83 1.26 0.76 0.07 c J2109-4110 2.95 2.63 2.01 1.70 0.07 0.03 0.04 0.03 3.46 2.70 2.00 3.49 0.19 cJ2110-1020 0.73 0.98 1.19 1.11 0.03 0.02 0.03 0.07 1.70 1.09 0.59 0.87 -0 .51 cJ2123 + 0535 1.28 0.95 0.68 0.62 0.36 0.32 0.26 0.15 1.32 1.17 1.37 1.99 0.54 c J2129-1538 1.11 1.18 0.96 0.56 0.02 0.02 0.03 0.04 < 0.50 < 0.50 < 0.50 0.75 -0.11 lJ2131-1207 2.71 2.55 1.93 1.61 0.10 0.04 0.06 0.05 0.92 1.20 0.83 1.65 0.09 c J2136 + 0041 8.32 9.47 7.32 3.6 0.02 0.02 0.04 0.05 2.10 0.60 < 0.50 < 0.50 -0.22 cJ2147 + 0929 0.83 0.85 0.88 0.91 0.12 0.12 0.12 0.08 2.11 2.14 1.01 0.66 -0 .03 cJ2148 + 0657 7.35 6.33 3.48 2.61 0.05 0.05 0.04 0.02 < 0.50 < 0.50 < 0.50 < 0.50 0.25 cJ2151-3027 1.43 1.72 1.57 1.19 0.08 0.05 0.04 0.03 5.10 4.09 1.57 1.67 -0 .30 cJ2152-7807 0.87 1.15 0.99 0.69 0.02 0.03 0.04 0.05 < 0.50 < 0.50 < 0.50 0.56 -0.47 l J2158-1501 2.04 2.07 2.21 2.42 0.09 0.09 0.09 0.08 6.60 6.77 7.13 5.06 -0.03 eJ2206-1835 3.08 4.12 5.20 6.12 0.02 0.02 0.03 0.02 4.34 2.67 < 0.50 < 0.50 -0.49 eJ2207-5346 1.22 1.19 1.13 1.18 0.15 0.07 0.06 0.12 1.59 2.34 4.43 7.50 0.03 e J2218-0335 1.71 2.08 2.23 1.83 0.07 0.08 0.04 0.07 3.07 2.06 1.37 < 0.50 -0.34 lJ2225-0457 4.53 4.34 4.75 5.99 0.17 0.09 0.04 0.03 3.37 4.85 5.09 5.40 0.06 e J2229-0832 0.75 0.89 1.01 0.97 0.18 0.22 0.08 0.05 1.95 2.30 2.21 1.78 -0.27 cJ2235-4835 1.03 0.97 1.01 1.08 0.12 0.09 0.09 0.11 1.85 1.86 2.25 2.70 0.09 l J2239-5701 0.83 0.72 0.53 0.40 0.06 0.05 0.02 0.04 6.75 4.39 5.18 6.12 0.24 cJ2243-2544 0.60 0.69 0.72 0.78 0.04 0.02 0.04 0.04 3.50 3.35 4.58 6.10 -0 .22 lJ2246-1206 2.38 2.18 2.01 1.85 0.16 0.14 0.06 0.04 1.41 0.95 0.98 0.92 0.13 c J2257-3627 1.03 1.28 1.37 1.24 0.05 0.04 0.02 0.03 0.52 < 0.50 < 0.50 < 0.50 -0.38 cJ2258-2758 5.67 3.37 1.62 1.04 0.09 0.10 0.09 0.03 2.46 0.92 1.26 1.31 0.89 l J2320 + 0513 1.13 1.09 0.77 0.53 0.19 0.20 0.18 0.13 1.02 1.10 1.43 1.37 0.12 cJ2329-4730 1.49 1.63 1.95 2.32 0.08 0.05 0.02 0.02 4.03 5.15 5.19 5.55 -0 .16 lJ2331-1556 1.21 1.30 1.31 1.30 0.10 0.07 0.04 0.03 2.27 1.03 0.57 0.95 -0 .14 l J2336-5236 1.38 1.63 1.83 1.97 0.02 0.02 0.02 0.02 < 0.50 < 0.50 < 0.50 < 0.50 -0.29 lJ2346 + 0930 1.39 1.57 1.65 1.77 0.02 0.02 0.03 0.03 1.35 1.82 2.30 3.33 -0 .21 lJ2357-1125 1.37 1.53 1.57 1.48 0.05 0.03 0.03 0.02 3.30 3.03 2.74 1.91 -0 .19 cJ2357-5311 1.55 1.49 1.54 1.61 0.06 0.03 0.03 0.04 1.98 3.08 3.42 2.68 0.07 c J2358-1020 1.04 1.11 0.95 0.72 0.42 0.34 0.17 0.06 3.22 2.84 2.10 2.20 -0.18 c the sources with an l designation may have extended steepspectrum emission or may suffer from confusion within the primary beam at 1.4 GHz, a common occurrence at the ATCA.However, the c sources have a higher degree of core dominance than the e sources.Of the 185 sources in table 3, 93 (50.3%) have a c flag, 49 (26.5%) have an e flag, and 43 (23.2%) have an l flag. Incomparison, for the 17 sources in table 2, which were dropped from the monitoring program after the first few epochs, 14 hadan e flag and 3 had a c flag. This is not surprising: source selection for the VSOP Survey was initially based on the total(single dish) flux density, and those sources with significant contributions to the total flux density in extended componentswill not have a sufficient core flux density to be detected on the baselines to the HALCA satellite.It is possible to gain a more quantitative understanding of 358 S. J. Tingay et al. [Vol. 55, these flags for those sources which also appear in an ATCAsurvey of 461 sources south of \Delta = -20\Theta (Lovell 1997; Lovellet al., in preparation). Lovell (1997) undertook a series of snapshots at 4.8 and 8.6 GHz and was able to image most of the 461sources. In addition, the fraction of the flux density contained in an unresolved core, F , was calculated for all sources. Sixty-seven of the 185 sources are also included in the Lovell survey: of these 34 have a c flag, 17 e, and 16 l, in proportion with theoverall distribution for the 185 sources. All 34 c sources had a "core fraction" at 4.8 GHz, F4.8, of 0.98 or more. As expected,the core fractions of the e and l sources were more widely distributed: 9 of the 17 e sources had an F4.8 of less than or equalto 0.95, with 6 e sources having F4.8 >= 0.98. Of the l sources,3 had F4.8 <= 0.95 and 7 had F4.8 >= 0.98. The observationsof Lovell were undertaken with ATCA 6 km array configurations between 1993 and 1995 and so variability of the core fluxbetween then and the monitoring here will affect the comparison of F4.8 and the source flags. Also, it should be kept in mindthat the source flags are based upon inspection of data at all frequencies, not just 4.8 GHz. However, the comparison confirmsthat the sources assigned a c flag do have the overwhelming majority of their flux density confined to an unresolved corecomponent. To check the consistency of our results with the results ofother workers we have made a comparison of our 8.6 GHz flux densities, percentage polarisations, and polarisationposition angles for 9 sources near the equator with data from the University of Michigan Radio Astronomy Observatory(UMRAO) single dish monitoring program (H. D. Aller et al. 2002, in preparation) at 8.0 GHz. In all cases, the agreementbetween the ATCA and UMRAO data is very good. The 9 sources are: J0339-0146, J0423-0120, J0607-0834,J1037-2934, J1229 + 0203, J1256-0547, J1337-1257, J1733-1304, and J1924-2914. In some cases, an offset in to-tal flux density is seen between the University of Michigan data and the ATCA data. The great majority of this offset is due toextended emission detected by the University of Michigan single dish observations but resolved out on the ATCA 6 km base-line. A small percentage of the offset may be due to spectral index effects arising from the difference in frequency used for theUMRAO observations (8.0 GHz) and the ATCA measurements (8.6 GHz). However, for the highly compact, flat spectrumsources among the 9 sources, the agreement is extremely good. A further check can be made with the monitoring programundertaken by Kovalev et al. (2000). Of 258 sources with \Delta > -30\Theta monitored with the RATAN-600 radio telescope ateight epochs between 1997 and 2000, 96 were included in the 185 sources monitored here. The RATAN-600 monitoring wascarried out at 22, 11, 7.7, 3.9, 2.3, and 1.0 GHz, and an estimate of the 5.0 GHz flux density was obtained by interpolating be-tween the 3.9 and 7.7 GHz values. Denoting the mean 4.8 GHz flux density measured from the ATCA monitoring as \Lambda I ATCA4.8 \Xi and the mean 5.0 GHz flux density interpolated from RATAN600 monitoring as \Lambda I RATAN5.0 \Xi , we calculate the fractional differ-ence between the two,\Delta \Lambda I ATCA4.8 \Xi - \Lambda I RATAN5.0 \Xi \Theta \Lambda \Lambda I ATCA4.8 \Xi , (1) and plot the resulting distribution in figure 1. Three sourceswith large fractional differences are apparent -- J1205-2634, Fig. 1. Fractional difference in flux density between the ATCA4.8 GHz and RATAN-600 (interpolated) 5.0 GHz data. See text for details. J1419-1928, and J1625-2527 -- and these will be consid-ered in more detail below. It is apparent from figure 1 that the RATAN-600 \Lambda I RATAN5.0 \Xi is on average slightly larger thanthe ATCA \Lambda I ATCA4.8 \Xi . Fitting a Gaussian to the distribution (ex-cluding the three sources above), the distribution for the 93 remaining sources has a mean of -0.05 and standard deviation of0.10. For the 47 c sources the corresponding values are -0 .02and 0.10, whereas for the 45 e or l sources the values are -0 .09and 0.10, indicating that the main difference can be ascribed to the better sensitivity of RATAN-600 to extended emission.It is notable that the three sources with large fractional differences have e flags. Both J1205-2634 and J1419-1928 haveATCA (6 km) flux densities comparable to the 2 M \Xi correlatedflux densities observed in the VLBA pre-launch survey observations (Fomalont et al. 2000b), both of which are significantlyless than the RATAN-600 flux density. Clearly, for these two sources a large extended component is resolved out by the 6 kmATCA baseline, leaving a compact core. Further evidence that this is the case for J1205-2634 comes from the F4.8 of 0.64(Lovell 1997). In contrast, J1625-2527 has a relatively large M at 4.8 GHz of 0.26, and a relatively high F4.8 of 0.95 (Lovell1997), and so the difference in average flux density may be due to significant source variability.A comparison of the difference in flux density (as opposed to fractional difference in flux density) with the Kovalev et al. datareveals 6 sources that differ by more than 1 Jy. For 5 sources \Lambda I ATCA4.8 \Xi - \Lambda I RATAN5.0 \Xi < -1. Four of these (J1229 + 0203,J1256-0547, J1733-1304, and J1924-2914) were among the 9 sources used in the comparison with UMRAO data describedabove, with the remaining source being J1625-2527, mentioned in the preceding paragraph. The source with \Lambda I ATCA4.8 \Xi -\Lambda I RATAN5.0 \Xi > 1 is J0609-1542. This source has a c flag and(like J1625-2527) a relatively high variability index, and so the discrepancy between \Lambda I ATCA4.8 \Xi and \Lambda I RATAN5.0 \Xi can be ascribedto source variability in this case. The ATCA primary flux density calibrator PKS 1934-638(J1939-6342 in table 3), was, aside from the calibration scans mentioned earlier, included in the general monitoring pro-gram (and not used in the initial calibration). These data thus give a consistency check, additional to that given by the No. 2] ATCA Observations of Compact Radio Sources 359 comparisons described above. Given a perfect calibration andabsence of RFI, we would expect J1939-6342 to have zero linear polarisation and zero variability index. From table 3 asmall apparent variability and linear polarisations up to 0.3% are seen at all frequencies. The small variability indices at thehigh frequencies are likely due to variations in instrumental factors that are not fully calibrated out using the proceduresdescribed above. The small but non-zero linear polarisations measured for J1939-6342 (< 0.3%) are also likely to be dueto variations in instrumental parameters. We thus take a conservative lower limit on the percentage polarisation that we canreliably measure from our data to be 0.5%. In table 3, this limit is quoted for sources that showed average linear polarisationsbelow 0.5%. Figure 2 contains plots of all the data used to derive thequantities for the 185 sources in table 3. The 4 different frequencies in figure 2 are designated as follows: 8.6GHz \Theta openstar symbols; 4 .8GHz \Theta open circle symbols; 2.5GHz \Theta opentriangle symbols; and 1 .4 GHz \Theta open square symbols. The8.0 GHz data from the UMRAO database are also plotted as solid squares for the 9 comparison sources described above.Error bars are shown on the flux densities in figure 2. For sources listed as compact (c) in table 3 the error estimateson the fluxes discussed above should be close to the thermal noise estimate since the sources can be described well as pointsources. However, for sources listed as extended (e), the contribution from source structure will change the distributionof visibility amplitudes around the best-fit point source flux density in a non-uniform way, from source to source (and ingeneral from epoch to epoch). The sources with the most structure on the 6 km baseline will have the greatest errorin the best-fit point source flux density. One example of this is J1833-2103 (PKS 1830-211), a gravitational lenswith a 1\Lambda \Lambda separation between compact lensed images (Jauncey et al. 1991). The variation in total flux density and polarisationpercentage, at 8.6 GHz in particular, is due in part to the effects mentioned above and not to real source variation.In addition, the necessity to observe with scans separated by several hours introduces another (generally small) source ofuncertainty in the flux density measurements. The prevalence of intra-day variability (IDV) caused by refractive interstellarscintillation has only recently become fully appreciated. Based on the first epoch results of the MASIV survey (J. E. J. Lovellet al., in preparation), it is likely that at any given time \Delta 10% of the sources in the monitoring program will be showing IDVabove a level of 2%. For the percentage polarisation and polarisation positionangles plotted in figure 2, the errors have been determined by propagating the errors on the Stokes I , Q, and U fluxes.The error in the polarisation position angle has been estimated following Rayner (2000).In addition, the flux density scale as defined by PKS 1934-638 is thought to be consistent with NorthernHemisphere flux density scales at the 1 to 2% level and approximately 5% uncertain with respect to the absolute flux densityscale (Reynolds 1994). The 5% uncertainty is not shown in the error bars in figure 2, for clarity.Figure 3 shows the overall properties of the 185 sources at 4.8 GHz, the distributions of flux density, variability index, percentage linear polarisation, and spectral index between 4.8and 8.6 GHz. Not shown on these plots are three sources with flux densities greater than 10 Jy (J1229 + 0203 = 3C 273,J1256-0547 = 3C 279, and J1924-2914 = PKS 1921-293) and one source with percentage linear polarisation greater than10% (J0519-4546= Pictor A). The expectation that source variability is stronger at higherfrequency is borne out by the values in table 3. The mean and median values of M are 0.08 and 0.06 at 1.4 GHz, 0.09 and0.06 at 2.5 GHz, 0.11 and 0.08 at 4.8 GHz, and 0.12 and 0.09 at 8.6 GHz. The number of sources with M >= 0.30 is 4, 7, 8,and 19 at 1.4, 2.5, 4.8, and 8.6 GHz, respectively. 4. Discussion The main purpose of the observations described here is toprovide information on the flux density, polarisation properties, spectral index, and variability properties of the VSOP Surveysample sources in the Southern Hemisphere. When data analysis for the VSOP Survey is complete, a detailed statisticalanalysis of the space VLBI properties will be possible, incorporating the results of the supporting ATCA observations.This analysis will be presented in future publications. Currently, as the VSOP Survey space VLBI observations arebeing reduced, the ATCA monitoring data are proving invaluable in checking the consistency of the calibration of the spaceVLBI observations. Not only do the ATCA data have scientific value to the VSOP mission, they are of great value in theanalysis of the space VLBI data themselves. However, the data presented here, quite apart from the VSOPspace VLBI mission, represents the first high resolution, long time-scale monitoring of a large number of compact extra-galactic radio sources in the Southern Hemisphere at centimetre wavelengths. As such, this database is a valuable resource inits own right and can be used to investigate the properties of different classes of active galactic nuclei (AGN), define SouthernHemisphere samples of bright radio sources, or provide information for detailed studies of individual objects.A further significant use of the ATCA database will be as a source of information on compact radio sources foruse as calibrators for Southern Hemisphere radio telescopes. The presence or absence of extended emission, the degreeof variability, and the total and polarised flux densities are all important factors to consider when choosing a calibrationsource. The ATCA VSOP database complements existing databases in this respect.In illustration of the scientific usefulness of the ATCA observations, here we examine differences between the gamma-rayloud and quiet populations of radio loud AGN in the southern component of the VSOP Survey sample. 4.1. The Properties of Gamma-Ray Loud and Quiet RadioLoud AGN The Energetic Gamma-Ray Experiment Telescope (EGRET)aboard the Compton Gamma-Ray Observatory has revolutionised our view of the gamma-ray universe by identifyingmany AGN as sources of greater than 100 MeV gamma-rays (Hartman et al. 1999; Mattox et al. 2001). Much theoreticaland observational effort has been expended over the last decade 360 S. J. Tingay et al. [Vol. 55, Fig. 2. ATCA light curves at 1.4, 2.5, 4.8, and 8.6 GHz for the monitored sources. No. 2] ATCA Observations of Compact Radio Sources 361 Fig. 2. (Continued.) 362 S. J. Tingay et al. [Vol. 55, Fig. 2. (Continued.) No. 2] ATCA Observations of Compact Radio Sources 363 Fig. 2. (Continued.) 364 S. J. Tingay et al. [Vol. 55, Fig. 2. (Continued.) No. 2] ATCA Observations of Compact Radio Sources 365 Fig. 2. (Continued.) 366 S. J. Tingay et al. [Vol. 55, Fig. 2. (Continued.) No. 2] ATCA Observations of Compact Radio Sources 367 Fig. 2. (Continued.) 368 S. J. Tingay et al. [Vol. 55, Fig. 2. (Continued.) No. 2] ATCA Observations of Compact Radio Sources 369 Fig. 2. (Continued.) 370 S. J. Tingay et al. [Vol. 55, Fig. 2. (Continued.) No. 2] ATCA Observations of Compact Radio Sources 371 Fig. 2. (Continued.) 372 S. J. Tingay et al. [Vol. 55, Fig. 2. (Continued.) No. 2] ATCA Observations of Compact Radio Sources 373 Fig. 2. (Continued.) 374 S. J. Tingay et al. [Vol. 55, Fig. 2. (Continued.) No. 2] ATCA Observations of Compact Radio Sources 375 Fig. 2. (Continued.) 376 S. J. Tingay et al. [Vol. 55, Fig. 2. (Continued.) No. 2] ATCA Observations of Compact Radio Sources 377 Fig. 2. (Continued.) 378 S. J. Tingay et al. [Vol. 55, Fig. 2. (Continued.) No. 2] ATCA Observations of Compact Radio Sources 379 Fig. 2. (Continued.) 380 S. J. Tingay et al. [Vol. 55, Fig. 2. (Continued.) No. 2] ATCA Observations of Compact Radio Sources 381 Fig. 3. Histograms of 4.8 GHz flux density, variability index, linearly polarised percentage, and 4.8 to 8.6 GHz spectral index for the 185 (= 202 - 17)sources monitored with the ATCA. in understanding the physical mechanisms that generate thegamma-rays in the jets of AGN, close to the proposed massive black-hole and accretion disk systems. There is a large over-lap in interest between gamma-ray observations of AGN and space VLBI observations as both are concerned with the AGNjets on very small spatial scales. Therefore, we examine and compare the properties of the gamma-ray loud and gamma-rayquiet AGN in our flux density monitoring sample. In the declination range covered by our ATCA monitor-ing observations, EGRET has positively identified 36 AGN as sources of greater than 100 MeV gamma-rays. Of these36 sources, 27 are part of the VSOP Survey list (Hirabayashi et al. 2000b).Mattox et al. (1997) and Mattox et al. (2001) have asserted that gamma-ray loud AGN can be identified with bright flat-spectrum radio sources. Based on this assertion we have selected a flat-spectrum sub-sample from our monitoring ob-servations of 202 sources, all sources with average 4.8 GHz flux density greater than 1 Jy and average 4.8 to 8.6 GHz spec-tral index greater than -0 .5. This resulted in a list of 112sources. Of these 112 sources, 18 appear as positive identifications in the third EGRET catalog (Hartman et al. 1999).The remaining 18 gamma-ray sources in the declination range \Delta < + 10\Theta can be categorised as follows: four sources ap-pear in the VSOP survey list but were not included in the radio monitoring, an incompleteness of 15% (4/27) that closelymatches the overall incompleteness of our observations compared to the VSOP Survey sample (17%); four sources werepart of our monitoring program but failed the 1 Jy flux density criterion above, although the spectral index criterion was met;one source, the gravitational lens J1833-2103 failed the spectral index criterion although the flux density criterion was met;finally, 9 sources did not appear in the VSOP Survey list and therefore were not part of our monitoring observations.Data collected from the literature and unpublished observations for all 9 of the gamma-ray AGN that are not part of theVSOP Survey list give flux densities and spectral indices that qualify them as bright, flat-spectrum radio sources; all fulfillthe spectral criterion but most have flux densities that lie just below the 1 Jy limit of the VSOP Survey. Of these objects, thesource with the weakest flux density is PKS 2155-304 at approximately 0.3 Jy, still relatively bright (see Mattox et al. 2001for more details of this EGRET identification). We restrict ourselves now to an analysis of the 112 bright,flat-spectrum sources which have been selected from our monitoring observations, of which 18 are gamma-ray loud AGN.Figures 4 and 5 show comparisons of the 18 gamma-ray loud AGN with the 94 gamma-ray quiet AGN. The follow-ing properties are compared: 4.8 GHz flux density; 4.8 GHz variability index; 4.8 GHz percentage linear polarisation; and 382 S. J. Tingay et al. [Vol. 55, Fig. 4. Comparison of the 4.8 GHz flux density distributions (left) and the 4.8 GHz variability index distributions (right) of the gamma-ray loud andgamma-ray quiet radio sources in a 1 Jy, flat-spectrum sub-sample of the sources monitored with the ATCA. 4.8-8.6 GHz spectral index. Figure 4 shows the comparisonof the 4.8 GHz flux density and the 4.8 GHz variability index. (Sources with flux densities greater than 10 Jy -- two gamma-ray loud and one gamma-ray quiet -- are not shown in the flux density plot, however, these sources are included in thestatistical analysis below.) Figure 5 shows the comparison of the 4.8 GHz percentage linear polarisation and the 4.8-8.6 GHzspectral index. It is apparent that the gamma-ray loud AGN have, onaverage, higher 4.8 GHz flux densities and higher 4.8 GHz variability indices than the gamma-ray quiet AGN. Thereappear to be no differences between the percentage polarisation distributions or the spectral index distributions. This is con-firmed by the standard Kolmolgorov-Smirnov two-sided test. The chance that the distributions of 4.8 GHz flux density forthe gamma-ray loud and quiet AGN have been drawn from the same population is 1%. The two distributions are differ-ent at the 99% confidence level. Likewise, the distributions of 4.8 GHz variability index are different at the 96% confidencelevel. Both of these results are at a level higher than 95% confidence and are taken to be significant.However, the distributions of 4.8 GHz percentage linear polarisation are not significantly different, with the confidencelevel far below 95%. Likewise, the 4.8-8.6 GHz spectral index distributions are not significantly different. The significant difference in the flux density distributionshas been noted many times before by a variety of investigators using independent samples (Impey 1996; Moellenbrocket al. 1996; Mattox et al. 1997; Zhou et al. 1997; Tingay et al. 1998) and appears a very solid and robust result; AGNthat are bright in gamma-rays are at least temporarily in a high state of radio emission. This may be due to enhanced beam-ing or increased intrinsic luminosity. Likewise, the lack of a difference in the spectral index distributions at centimeterwavelengths has previously been noted by Tingay et al. (1998). We note that the current study has the advantage over that ofTingay et al. (1998) in that the flux density measurements used to derive the spectral indices were made simultaneously.The comparison of variability index distributions for gamma-ray loud and quiet AGN shows that the gamma-rayloud AGN are more variable at 4.8 GHz than gamma-ray quiet AGN. It is known that variability is somewhat correlated withspectral index in samples of extragalactic radio sources, the trend being that the sources with flatter spectra tend to bemore variable, since they are dominated by compact emission regions. This effect can be seen in the VSOP sample by plot-ting variability index against spectral index. Since we have shown that there is no significant difference between the spec-tral index distributions of the gamma-ray loud and gammaray quiet AGN in our flat spectrum sub-sample, we do not No. 2] ATCA Observations of Compact Radio Sources 383 Fig. 5. Comparison of the 4.8 GHz percentage linear polarisation distributions (left) and the 4.8-8.6 GHz spectral index distributions (right) of thegamma-ray loud and gamma-ray quiet radio sources in a 1 Jy, flat-spectrum sub-sample of the sources monitored with the ATCA. expect the significant difference in variability index betweenthe gamma-ray loud and gamma-ray quiet sources to be due to spectral index effects. This is confirmed by analysing thoseflat-spectrum sources in the spectral index range -0 .33 to+ 0 .76 (the range over which gamma-ray loud sources areobserved). Of these sources, 17 /18 of the gamma-ray loudsources have variability indices greater than 0.07, whereas only 53/86 of the gamma-ray quiet sources have variability indicesgreater than 0.07. Using a 2 * 2 contingency table and Yates' correction for continuity, we calculate the probability that thegamma-ray loud and gamma-ray quiet sources are drawn from the same population, with respect to their variability indices, tobe < 2.5%. Therefore, in general, the gamma-ray loud AGNhave significantly higher variability indices than the gammaray quiet AGN, as found by the Kolmolgorov-Smirnov testsdescribed above. The variability index as defined in section 3 is sensitive tothe range over which the flux density of the source varies, relative to the mean flux density, over the course of the ob-servations, but is not sensitive to the time-scale of the variations. Previously, L"ahteenm"aki et al. (1997) and L"ahteenm"akiand Valtaoja (1999) found that a comparison of gamma-ray loud and gamma-ray quiet AGN based on variability brightnesstemperatures (from 22 and 37 GHz data) show that the brightest gamma-ray sources have a tendency for high variability brightness temperature. In this case, the variability bright-ness temperature is sensitive to both the variability amplitude and the variability time-scale. Also, Aller, Aller, andHughes (1996), from 14.5 GHz data of the UMRAO monitoring program, find that gamma-ray loud AGN have a tendencyfor larger amplitude variations than gamma-ray quiet AGN, over a 4 year period.Therefore, in three independent studies, of three different samples, at three different radio frequencies, gamma-rayloud AGN have been shown to be more variable than gamma-ray quiet AGN. Further, Aller, Aller, and Hughes(1996), L"ahteenm"aki et al. (1997), and L"ahteenm"aki and Valtaoja (1999) find that the gamma-ray flares occur mainlyduring the rising phase of radio flares. The apparent causal link between radio variability and gamma-ray variability hasbeen interpreted by L"ahteenm"aki and Valtaoja (1999) as strong evidence for synchrotron self-Compton processes that producegamma-ray emission from the same physical volume as the radio synchrotron emission. 5. Summary We have presented the results of 3.5 years of ATCA observa-tions aimed at monitoring the Southern Hemisphere component of the VSOP space VLBI AGN Survey sample in the frequency 384 S. J. Tingay et al. range 1.4-8.6 GHz. These data have been analysed to givebasic parameters of the sources, to be used in future detailed statistical analyses of the VSOP Survey data. Cross-checksagainst UMRAO and RATAN-600 monitoring, and a previous ATCA imaging survey, have been made to confirm the integrityof the data. As well as contributing to the scientific goals of the VSOP Survey, the ATCA data have been useful in checking thecalibration of the VSOP Survey observations. Similarly, these data will also be useful for calibration of Southern Hemisphereradio telescopes. In illustration of the scientific usefulness of the ATCA datathemselves, we have used them to compare the properties of gamma-ray loud and quiet AGN in the southern component ofthe VSOP Survey sample. The main result from this comparison is that in a flat-spectrum sample, gamma-ray loud AGNhave higher variability indices than the gamma-ray quiet AGN. The Australia Telescope is funded by the Australian Commonwealth Government for operation as a national facilitymanaged by CSIRO. We gratefully acknowledge the VSOP Project, which is led by the Japanese Institute of Space andAstronautical Science in cooperation with many organisations and radio telescopes around the world. Part of this workwas undertaken at the Jet Propulsion Laboratory, California Institute of Technology, under contract with the NationalAeronautics and Space Administration while SJT held an NRC / NASA-JPL Research Associateship. This research hasmade use of data from the University of Michigan Radio Astronomy Observatory which is supported by funds from theUniversity of Michigan. This research has made use of the NASA / IPAC Extragalactic Database (NED) which is oper-ated by the Jet Propulsion Laboratory, California Institute of Technology, under contract with the National Aeronautics andSpace Administration, USA. References Aller, M. F., Aller, H. D., & Hughes, P. A. 1996, in IAU Symposium175, Extragalactic Radio Sources, ed. R. D. Ekers, C. Fanti, & L. Padrielli (Dordrecht: Kluwer), 283Fomalont, E., et al., 2000a, in Astrophysical Phenomena Revealed by Space VLBI, ed. H. Hirabayashi, P. G. Edwards, & D. W. Murphy(Sagamihara: ISAS), 167 Fomalont, E. B., Frey, S., Paragi, Z., Gurvits, L. I., Scott, W. K.,Taylor, A. R., Edwards, P. G., & Hirabayshi, H. 2000b, ApJS, 131, 95Frater, R. H., Brooks, J. W., & Whiteoak, J. B. 1992, J. Electr. Electron. Eng. Aust., 12, 103.Hartman, R. C., et al. 1999, ApJS, 123, 79 Hirabayashi, H., et al. 1998, Science, 281, 1825Hirabayashi, H., et al. 2000a, PASJ, 52, 955 Hirabayashi, H., et al. 2000b, PASJ, 52, 997Impey, C. 1996, AJ, 112, 2667 Jauncey, D. L., et al. 1991, Nature, 352, 132Kellermann, K. I., Vermeulen, R. C., Zensus, J. A., & Cohen, M. H. 1998, AJ, 115, 1295Kovalev, Y. A., Kovalev, Y. Y., & Nizhelsky, N. A. 2000, PASJ, 52, 1027L"ahteenm"aki, A., Ter"asranta, H., Wiik, K., & Valtaoja, E. 1997, in AIP Conf. Proc. 410, Proc. Fourth Compton Symp., ed. C. D. Dermer,M. S. Strickman, & J. D. Kurfess (New York: AIP), 1452 L"ahteenm"aki, A., & Valtaoja, E. 1999, in ASP Conf. Ser. 159, BL LacPhenomenon, ed. L. O. Takalo & A. Sillanp"a"a (San Francisco: ASP), 213Lovell, J. E. J. 1997, PhD Thesis, University of Tasmania Lovell, J. E. J., et al. 2000, in Astrophysical Phenomena Revealed bySpace VLBI, ed. H. Hirabayashi, P. G. Edwards, & D. W. Murphy (Sagamihara: ISAS), 183Mattox, J. R., Hartman, R. C, & Reimer, O. 2001, ApJS, 135, 155 Mattox, J. R., Wagner, S. J., Malkan, M., McGlynn, T. A., Schachter,J. F., Grove, J. E., Johnson, W. N., & Kurfess, J. D. 1997, ApJ, 476, 692Moellenbrock, G. A., et al. 1996, AJ, 111, 2174 Moellenbrock, G. A., et al. 2000, in Astrophysical PhenomenaRevealed by Space VLBI, ed. H. Hirabayashi, P. G. Edwards, & D. W. Murphy (Sagamihara: ISAS), 177Rayner, D. 2000, PhD Thesis, University of Tasmania Reynolds, J. 1994, Australia Telescope Technical Memo AT/39.3/040Sault, R. J., Teuben, P. J., & Wright, M. C. H. 1995, in ASP Conf. Ser. 77, Astronomical Data Analysis Software and Systems IV,ed. R. A. Shaw, H. E. Payne, & J. J. E. Hayes (San Francisco: ASP), 433Tingay, S. J., Murphy, D. W., & Edwards, P. G. 1998, ApJ, 500, 673 Zhou, Y. Y., Lu, Y. J., Wang, T. G., Yu, K. N., & Young, E. C. M.1997, ApJ, 484, L47