SRC

Variability of extragalactic radio sources from the results of multifrequency
monitoring at RATAN-600

S.O. Kiikov, M.G. Mingaliev, V.A. Stolyarov, M.S. Stupalov}

SAO RAS
December 11, 2002}{December 17, 2002

Abstract

The results of the observations of 71 extragalactic radio sources
using the RATAN-600 radio telescope are presented. The observations were
conducted in eight sets from 1995 to
1996 at the wavelengths 1.38, 2.7, 3.9, 7.6, 13 and 31 cm. Time variations
of the radio emission flux density of the sources were obtained. Variation
parameters of the sources were studied.
\keywords{radio continuum: galaxies --- galaxies: optical identification
--- radio sources: variability}

Introduction

Observations of 71 extragalactic radio sources, 6 of which are reference
sources, were carried out at the RATAN-600 radio telescope during the years
1995--1996. The principle goal of these observations was to study the flux
density variability of the radio emission of the sources.
In this paper we present the results of investigation of the obtained
observational data.

\section{Observations}

The observations were carried out in the northern sector of the RATAN-600
radio telescope. The declination range was from $-24\degr$ to $44\degr$. The
sources were observed in eight series at the wavelengths 2.7, 3.9, 7.6, 13 and
31 cm, and in some series at 1.38 cm. Each source was observed 5--8 times
during
a single observational set. The accuracy of the flux density measurement was
determined as
a standard error of $N$ observations of each source during a set.

The flux densities of the radio sources were referred to those of sources
which are believed nonvariable and
were used as reference sources for variability investigation.
The flux density of six sources that were used as the calibrators
are listed in Table 1.

The parameters of the detectors used in the northern sector were given in
the papers by Nizhelskij (1996) and Berlin et al. (1997), and the
characteristics
of the beam pattern were reported by Botashev et al. (1998).

Results

The flux densities in Jy and their errors are listed in Table 2 for each
source in each set at the wavelengths of 2.7, 3.9, 7.6, 13 and 31 cm.
Besides,
Fig.1 displays the time variations of flux densities.

To study variability of extragalactic source radiation, the following
values,
the same as those used in the papers by Seielstad et al. (1983), Kesteven et
al. (1976), and Fanti et al. (1979), were employed:

\caption{Adopted calibrator source flux densities}
Source & $S_{1.38}$ & $S_{2.7}$ & $S_{3.9}$ & $S_{7.6}$ & $S_{13}$ & $S_{31}$ \\
1245-197 & 0.63 & 1.24 & 1.75 & 3.0  & 4.0  & 6.3
0624-058 & 1.15 & 2.76 & 4.16 & 8.11 & 12.8 & 24.1
0518+165 & 1.15 & 2.28 & 2.91 & 4.04 & 6.38 & 10.1
1328+307 & 2.49 & 4.22 & 5.53 & 8.57 & 11.5 & 17.2
0134+329 & 1.24 & 2.5  & 3.63 & 6.88 & 10.9 & 21.9
2105+420 & 5.5  & 6.02 & 6.33 & 5.0  & 2.6  & --

....

where
$S_i$ and $\sigma _i$ are the flux densities and their errors presented in
Table 2, respectively, and $n$ is the number of series for which data are
available. $V$ is the relative amplitude of variability.

The computed parameters (1) --- (5),  the number of degrees of freedom
$df$, and $\chi^2$-probability  $P$ are presented in Table 3. A source
considered to be variable at the given wavelength $\lambda$ if $P > 0.985$.
Variability of a radio source in Table 3 is marked by $v$ (variable), and
in the case of non-variable source the symbol $n$ (non-variable) is used
instead.

Discussion

According to the variability criterion used above, out of 65 radio sources
studied in this paper, 47 (72\%) are variable at the wavelength 2.7 cm, 47
(72\%) at 3.9 cm, 53 (82\%) at 7.6 cm, 26 (40\%) at 13 cm, and 38 (58\%)
at 31 cm. 17 (26\%) objects show variability simultaneously at all wavelengths.
As to the variability amplitude, it is observed to rise with
increasing frequency in some part of variable objects (6 sources); in the
rest of the cases the amplitude decreases initially, then it rises and vice
versa. It should be noted that the number of variable sources at some
wavelengths is, generally speaking, underestimated, since some sources for
which the $\chi^2$-probability is close to but less than 0.985 may also
be variable.

To investigate the relation between the variability of the sources and other
parameters, optical identifications and spectral indices are used.

It follows from the optical identifications of the radio sources under study
that 29 are quasars, 15 are BL\,Lacertae objects, 15 are galaxies, and 6 are
empty field objects
(i.e. optically unidentified objects). 7 (24\%) quasars, 9 (60\%)
BL\,Lacertae objects,
and 1  (17\%) optically unidentified object are variable at all the
wavelengths.
The data obtained show that the BL\,Lacertae objects have a significantly
higher probability of being variable than the quasars and other radio sources.
The relationship between the amplitude of variability of these objects and the
redshift is rather complex. If one considers the linear root-mean-square
approximation, it can be seen that with increasing redshift of quasars the
flux variability amplitude rises at all wavelengths. On the contrary,
it decreases for galaxies and BL\,Lacertae objects. The decrease in
amplitude can, for instance, be explained by cosmological `` time dilatation''.
It means that in the rest frame of the observer the time grows by a factor of
($1+z$) and, therefore, only a part of the light curve of extragalactic source
is observed. Depending on the frequency of radiation, the variability amplitude
in a linear root-mean-square approximation rises with a frequency increase
for 16 quasars, 10 BL Lacertae objects, 6
galaxies and 2 empty field objects. Among the investigated extragalactic
radio sources the largest median amplitudes of variability are observed for
the  BL Lacertae objects. Optically unidentified objects at the wavelength 31 cm
have comparable amplitudes.

Out of 65 radio sources that have been investigated, 32 have flat spectra
(i.e. $\alpha>-0.5$), for 8 the spectra are steep ($\alpha  \le -0.5$), and
for 25 objects the spectral indices are $>-0.5$ in some ranges, and $ \le -0.5$
in others, i.e. in some regions the spectrum is flat, in others it is steep.
When investigating the relation of the flux density of radio emission of
extragalactic objects and the frequency, a computation of spectral indices
$\alpha _{0.97}^{2.3}$, $\alpha _{2.3}^{3.9}$, $\alpha_{3.9}^{7.7}$
and  $\alpha _{7.7}^{11.1}$ was carried out with allowance made for the
relationship $S = C\nu^\alpha $, where $C = const$, and the spectra were
also plotted. As it is known, flat spectra are characteristic of compact
structures while the steep ones are characteristic of extended regions.
Optical identification shows that most sources with flat spectra are
BL\,Lacertae objects,
13 (87\%) out of 15; they are followed by quasars, 15 out of 29; then follow
galaxies,  3 out of 15; and 1 out of 6 is an empty field source. As far as
the  steep spectra are concerned, they are observed for 3 quasars, 2 empty field
objects and 3 galaxies. The sources with variable indices (i.e. for the
spectra of which in some intervals $\alpha>-0.5$ and in some
$\alpha\le-0.5$) include 11 quasars, 9 galaxies, 3 empty field objects and
2 BL\,Lacertae  objects.

Among 17 objects variable at all wavelengths, 12 have a flat spectrum, out
of which 7 are BL\,Lacertae objects and 5 are quasars, the spectrum of one
empty field object
is steep, while  four sources (2 BL\,Lacertae objects and 2 quasars) have
the spectra
with a variable index. The spectrum appearance is largely associated with
the variability of extragalactic radio sources. Among these objects an
unusual
optically unidentified radio source 0312+149 should be noted, which by the
spectrum shape is optically thin and variable at the same time. For some
sources
no variability is observed at all the wavelengths among which there are the
quasar 1633+382 with a flat spectrum and a large redshift, $z \approx 1.8$,
the galaxy 0941-080 with a steep spectrum, and one optically unidentified
object  1133+432 with a variable
spectral index. The non-variability of the compact object 1633+382 can be
explained by the fact that its nucleus is not active. As to the
non-variability
of the galaxy 0941-080, this is quite naturally since according to its
spectrum
it is an extended object. Concerning the optically unidentified object
1133+432 showing no variability, one can say judging by the spectrum that it
consists of a compact structure, the nucleus of which is not active, and of
an
extended region. The spectra of more than half of the radio sources, the
variability of which is observed at three and four wavelengths, i.e. in the
greater part of the range, are flat. Thus, it follows from the results
obtained that the variability is basically characteristic of extragalactic
radio sources with flat spectra. This can be readily explained by the fact
that the flat spectrum objects are generally compact structures, during the
activity period of which radiation variability is observed.

Conclusion

So, on the basis of our investigations of 65 extragalactic radio sources,
it follows that:

1)
At all wavelengths simultaneously (2.7, 3.9, 7.6, 13 and 31 cm) 17 (26\%)
objects are variable. Out of them 9 are BL\,Lacertae objects (60\% from 15),
7 are quasars (24\% from 29) and 1 is an optically unidentified object (17\% from
6).

2) Variability is basically a characteristic of the radio sources with
the flat spectra. Among 17 variable objects mentioned above 12 (71\%) ones have
a flat spectrum. The spectra of more than a half radio sources, variable at
3 and 4 wavelengths, are flat.

3)In the linear root-mean-square approximation the variability amplitude grows
with the frequency increase for 34 radio sources.

4)The largest median  amplitudes of radio emission variability are observed in
BL\,Lacertae~ objects  (except the amplitude in~ optically
unidentified~ ob- \newpage \noindent jects at the
wavelength 31 cm), and their value by $1.5\div 3$ times and even more
exceeds the
amplitudes of other objects.

5) Variability amplitude at all wavelengths rises  with growing redshift in the
linear root-mean-square approximation for quasars, and for galaxies and
BL\,Lacertae objects it falls.

Bibliography

Berlin A.B., Maksyashev A.A., Nizhelskij N.A., et al., 1997, Proceed.
of 27 Radioastron. Conf., St.Petersburg, {\bf 3}, 115

Botashev A.M., Gorshkov A.G., Konnikova V.K., Mingaliev M.G., 1998, Preprint
SAO RAS, Nizhnij Arkhyz, {\bf 132}

Fanti R., Ficarra A., Mantovani F., Padrielli L., Weiler K., 1979, \aas,
{\bf 36}, 359

Kesteven M.J.L., Bridle A.H., Brandie G.W., 1976, \aj, {\bf 84}, 61

Nizhelskij N.A., 1996, SAO Report, 57

Seielstad G.A., Pearson T.J., Readhead A.C.S.,  1983, \pasp, {\bf 95},
842