Astron. Astrophys. 362, 27-41 (2000)

2. IPS observations

A compact radio source or component seen through the solar wind exhibits scintillations due to electron density fluctuations in the interplanetary medium. The power spectrum of the fluctuations depends on the conditions of the solar wind, and the size and structure of the compact component. Using a suitable model of the solar wind, IPS observations can be used to estimate the angular size of the scintillating component from the power spectrum, and the scintillation index, m, which is defined as S/S. Here S is the rms of the flux density variations and S is the total flux density of the source (e.g. Little & Hewish 1966, 1968; Cohen et al. 1967; Rao et al. 1974). By comparing m_obs for a given source with a point source calibrator, m_cal, we can estimate µ, the fraction of flux density in the scintillating component. Generally the scintillation index, m, decreases with source size and becomes non-detectable when the source size is so large that the signal to noise ratio is poor. This limits the detectable IPS size to about 0."4 at 327 MHz, which we will refer to as the IPS cut-off size. Also for the typical solar wind parameters, and the observing frequency of 327 MHz, the minimum size of the components which can be estimated is about 50 mas (Manoharan & Ananthakrishnan 1990; Gothoskar & Rao 1999). In a typical radio source the compact scintillating components could be the hotspots at the outer edges of the lobes, the nuclear or core components or prominent knots in a jet. If the separation of the compact components, such as the hotspots, is smaller than the IPS cut-off size, the source scintillates as a single source. On the other hand, if the scintillating components in the source are separated by larger than the IPS cut-off size, the components scintillate independently and the parameters estimated from IPS observations are the weighted average values of the different scintillating components.

2.1. The sample of sources

The sample of CSS and GPS sources for the IPS observations have been compiled from a number of papers (Gopal-Krishna et al. 1983; O'Dea et al. 1991; Spoelstra et al. 1985; Savage et al. 1990; Cersosimo et al. 1994; Stanghellini et al. 1990; Peacock & Wall 1982; Sanghera et al. 1995), with the following criteria: (i) declination , the normal observable range of the ORT; (ii) total flux density at 327 MHz, 200 mJy so that the source can be observed with adequate signal to noise ratio; (iii) the ecliptic latitude of the source is so that there is enough scintillating power to estimate the parameters reliably. Many sources in our sample occur in more than one of the above lists and they have been counted only once in our list. Also, more recent estimates of sizes and redshifts have led to a number of objects in our sample being larger than the canonical limit of 20 kpc. These have been retained in the list. A few of the sources which were earlier identified as GPS sources appear to have a flat radio spectrum. These have been marked separately in the table and have not been used in the statistical analyses. Our final sample of largely compact sources consists of 100 objects, 48 of which are GPS sources. In addition, for comparison we have compiled a sample of 19 larger-sized 3C and 4C sources of similar redshift and luminosity. All these sources have 1 Jy, and their linear sizes are larger than about 100 kpc for all but one source.

The basic radio and optical properties of the sample of largely CSS and GPS sources and the sample of larger sources are listed in Table 1 and Table 2 respectively, which are arranged as follows. Columns 1 and 2: IAU name and an alternative name; Column 3: optical identification where G denotes a galaxy, Q a quasar and EF an empty field; Column 4: redshift; Column 5: identification of the object as a GPS source. Some of the sources which were tentatively classified as GPS objects appear to have a flat radio spectrum. These have been indicated as an FSC or flat-spectrum core. Column 6: flux density at 327 MHz in Jy. The flux density values are interpolated from a least-squares fit to the data compiled from the literature, as described in Steppe et al. (1995). A denotes an extrapolated value in which a measurement of the flux density was available at about 400 MHz. Column 7: radio luminosity at 5 GHz in units of W Hz^-1 sr^-1; Column 8: the largest angular size of the source (LAS) in arcsec; for sources with weak extended emission, the LAS is defined by the bright features on opposite sides. Column 9: the projected linear size in kpc; Column 10: structure of radio emission, where T denotes a triple source, MT a misaligned triple where the supplement of the angle formed at the core by the outer hotspots exceed 15^o, D a double-lobed one, CJ for a core-jet structure, S for a single component and Cpx for a complex structure. A CJ source is one where there is one dominant component with a jet-like extension rather than a second distinct lobe or hotspot. The double-lobed and triple sources where the flux density ratio of the outer components exceeds 10 are defined to be asymmetric and are identified in this column as AD, AT or AMT. A flat-spectrum core with extended emission on only one-side is indicated by the symbol OS. Any uncertainty in the information or classification is indicated by a question mark. Column 11: references to radio structure. The key to the references are listed in Table 3.

Table 1. Properties of the sample of largely CSS and GPS sources

Table 1. (continued)
Notes:
0218+357: This source is gravitationaly lensed (O'Dea et al. 1992; Patnaik et al. 1995). The LAS quoted here is the separation between the two lensed images (Polatidis et al. 1995). 0319+121: This spectrum of the source is flat (c.f. Dallacasa et al. 1995; Saikia et al. 1998). The LAS is estimated using the extended emission seen by Murphy et al. (1993).

Table 2. Properties of the sample of larger sources

Table 3. Key to the references for Table 1 and Table 2

2.2. Observations and analyses

We have observed our sample of sources with the ORT at 327 MHz in the phase-switched mode with a bandwidth of 4 MHz. The intensity variations of a source are sampled at a rate of 20ms. Each source was observed at a number of solar elongations, , the observations at each elongation lasting about 15-20 minutes. Our observational procedure is similar to the one described by Manoharan & Ananthakrishnan (1990) and Gothoskar & Rao (1999). We also monitored the off source region simultaneously in the total-power mode, to look for any interference, and edited the data accordingly. Since strong scattering occurs when about 12^o, observations were not made for lower values of . Periods of large solar activity can also modify the average properties of the solar wind and hence the observed power spectrum. Data during such periods were also edited since we are primarily interested in the properties of the sources (Manoharan 1993, 1997). For each source, the average power spectrum of the scintillations was obtained at each elongation. From this spectrum, we estimate the scintillation index, m. The typical error in the scintillation index is usually less than about 5 per cent. Variations in the interplanetary medium also leads to a scatter in the estimated value of µ. The percentage uncertainty in the fraction of scintillating flux density, µ, are listed in Table 4.

Table 4. Observed parameters of largely CSS, GPS sources

Table 4. (continued)

An additional source of uncertainty in µ is the scintillations produced by the confusing sources present in the beam. The confusion limit of the ORT is about 1.5 Jy. However, only a fraction of these sources have compact components which are likely to contribute to the scintillations. At low frequencies the scintillating features would be almost entirely the hotspots at the outer edges of the beams in the high-luminosity FRII sources. Estimating the fraction of FRII radio sources in the B2 sample (Allington-Smith 1982) to be about 60 per cent, and assuming the fraction of flux density in the scintillating hotspots to be about 50 per cent, yields a conservative estimate of the scintillation confusion to be about 0.2 Jy. This affects significantly the interpretation of the scintillation visibility µ of the weaker sources. The values for the weak sources with S₃₂₇ 0.5 Jy have been listed in the tables, but have not been used further in the discussions.

The variation of scintillation index, m, with solar elongation for a point source is given by m. The index is obtained by a least-squares fit to the observed scintillation index of the IPS calibrator, at various elongations. The compact radio source 1148-001 is the standard IPS calibrator used at the ORT (cf. Venugopal et al. 1985), and observations of this source during 1994-95 have been used to calibrate our observed scintillation indices. The observed scintillation index for a given source can be smaller than the calibrator at the same elongation if either only a fraction µ of the total flux density is in the scintillating component, , or the scintillating component is extended. We apply a correction to , for the finite size of the scintillating component using the model of solar wind (Manoharan & Ananthakrishnan 1990; Gothoskar & Rao 1999). This correction could not be applied to 12 of the 91 sources with S₃₂₇ 0.5 Jy in the sample of largely CSS and GPS radio sources (Table 1), and 6 of the 19 larger sources (Table 2) since we could not estimate the size of the component reliably from our observed power spectrum.

The component sizes are estimated from the best fits to the observed power spectrum after trying different values of component sizes and parameters of the standard solar wind model (Manoharan & Ananthakrishnan 1990; Manoharan et al. 1994; Gothoskar & Rao 1999). The free parameters in this model are the solar wind velocity, random velocity component in the scintillating medium, axial ratio of the irregularities in the medium, the power-law index of the power spectrum of the turbulence of the medium, and the size of the scintillating source. The model fitting was done only for sources observed with a high signal to noise ratio. Even though a 5 mas difference can be distinguised from the power spectrum, the typical inner-scale of the interplanetary medium corresponds to an angular scale of 50 mas, which is the typical error in the estimated angular sizes.

If the structure of the scintillating component is simple, then it might be possible to have an idea of the two-dimensional structure of the source if the direction of the solar wind covers a wide range of position angles across the source. For example, in the case of an elliptical gaussian brightness distribution, a variation of the estimated size from a minimum to a maximum should be observed if the solar wind covers a range of position angles during the different observations of the source. There are 19 sources showing variation of the estimated size with the position angle of the direction of motion of the solar wind. Assuming the structure to be elliptical, we have estimated the parameters of 4 sources where we had enough data to represent the ellipse.

The results of our analyses are presented in Table 4, which is arranged as follows: Column 1: IAU name; Column 2: number of observations used to estimate the size of the scintillating component; Column 3: estimated size of the scintillating component. For the four sources whose two-dimensional structure has been determined assuming it to be elliptical, the major and minor axes and the position angle of the major axis are quoted. The sources which exhibit variation in size, but without enough data to determine the position angle of an elliptical structure unambiguously, the maximum and minimum size are quoted. For the remaining sources, the average size is listed. Column 4: the total number of IPS observations; Column 5: the fraction of flux density of the scintillating component, µ, and the percentage error in µ.

© European Southern Observatory (ESO) 2000

Online publication: October 30, 19100
helpdesk.link@springer.de