Observations of the 22 GHz maser emission in the direction of RT Vir were made
with the RT-22 radio telescope of the Radio Astronomical Station of the Lebedev
Physical Institute in Pushchino, Russia. A receiver with low noise amplifiers at
the feed and a 96-channel spectrometer of filter-bank type were used. The
spectral resolution in radial velocity was 
. An antenna
temperature of 1 K from a point source with unpolarized emission corresponds to
a flux of 25 Jy. A detailed description of the instruments and the observational
method can be found in Sorochenko et al. (1985)
and Lekht et al. (1995).
The obtained data were reduced and analyzed at the National Institute of Astrophysics, Optics and Electronics, Tonantzintla, Puebla, Mexico.
profiles and the variation of the integrated flux
In Fig. 1 (click here) we show the maser profiles for RT Vir observed from 1986 to
1996. Some of the profiles obtained during the first half of 1986 were published
earlier by Berulis et al. (1987), but we included them in our paper to have the
total catalog of the profiles for 1986.
Figure 1:
a-k) Catalog of the maser spectra of RT Vir obtained
during 1986-1993. Flux interval shown on the vertical axis is 200 Jy
The maser emission of RT Vir is concentrated in three velocity ranges
of the spectrum: at velocities 
, between 15.5 and 
,
and at velocities 
. In the central interval, near the radial
velocity of the star, emission has not frequently been detected. The
time variation of the integrated flux of the overall spectrum is shown in
Fig. 2 (click here).
Figure 3 (click here) shows the integrated flux for the three above-mentioned ranges of the
spectrum. The integrated fluxes for the epoch 1984 December to 1985 December
were deduced from Berulis et al. (1987). The dashed lines denote the smoothed
curves of the long-term variations of the integrated flux. Since the maximum of
the flare of 1985 is very high (
for the integrated flux at
velocities ), its corresponding value is given in
Fig. 3 (click here)
for convenience.
Figure 2:
Time variations of the flux integrated over all the
spectrum. The dashed line shows the smoothed curve, which excludes the influence
of fast variations
Figure 3:
Time variations of the flux integrated over different spectral
intervals. The dashed line shows the presence of long-term variations of the
emission (the characteristic time of the variability is about 5-7 years)
Beginning from 1988 the emission at 
was very weak and
appeared only occasionally. For this reason, the observations at these velocities
were not carried out from 1989 May to 1990 November. However, at the end of 1990
December intense emission at 
was observed.
A rough estimate of the time of this flare can be deduced from the following
facts: 1) from 1988 the emission at was very weak and
appeared only from time to time; 2) the maximum flux of this feature was
observed in 1990 March; 3) in previous years the increase of the flux was faster
than the decrease; 4) according to the smoothed curve (Fig. 3 (click here)) the maxima and
the minima of the integrated flux of the principal spectral groups
(
and ) are in opposite phase.
On the basis of the above facts, we may assume that the rapid enhancement of the
emission at velocities
began no earlier than the middle of
1990. The probable increase of the integrated flux for the time interval from
1989 May up to 1990 November are presented by dotted lines in Figs. 2 (click here) and 3 (click here).
The dashed line in Fig. 2 (click here) for which fast variations of the integrated flux are
excluded has a complex form. At the beginning, in the time interval 1985-1988,
the curve is approximated by a horizontal line at the level of about 
.
The intense flare and the deep minimum relative to this level are remarkable. Afterwards, the behaviour of the curve may be presented by a waveform curve with a characteristic time variation of about three years.
Such behaviour of the integrated flux of the total spectra does not necessarily
reflect the physical processes that took place in the maser of RT Vir,
since it is a superposition of three curves, each of which reflects the variations
of the integrated flux in one of three ranges of the spectrum (Fig. 3 (click here)). At the
side intervals of the spectra ( and )
where emission appears more frequently, quasi-periodic variations are observed.
The time interval between two consecutive maxima or minima in both cases is
about 6 years. For the mentioned spectral intervals the variations of the
integrated flux are in opposite phase, i.e. there is anticorrelation of the
fluxes (this has been observed in 1986, 1992 and 1993-1995).
Thus, the superposition of the curves with more or less the same period, but in different phase may lead to a curve with twice shorter period (1989-1995) or slow variations will be smoothed and only fast components will be left (1985-1988).
The time variations of the centroid velocity for the three specified spectral intervals are shown in Fig. 4 (click here).
Figure 4:
Time dependence of the centroid velocity of different spectral
intervals
Since the emission was very weak and appeared only from time to time in the second spectral interval from 1989 July to 1992 March and in the third one from 1988 January to 1990 July, the computation of the centroid velocity for these periods was not possible. This is the reason for the discontinuities in the curves in Fig. 4 (click here).
At velocities
emission was seen all the time,
and so the curve is smoothed with no sudden jumps during the period of our
observations. The smoothed curve of Fig. 4 (click here) (dashed lines) is correlated with the
analogous curve of the integrated flux (Fig. 3 (click here));
i.e. the extreme values of these two curves coincide in time.
In order to determine the velocity where the maser in RT Vir is more
active, we deduced the average profile for the period 1984 December - 1996 March
(Fig. 5 (click here)). It justifies the division of each profile into three parts at velocities
15.5 and . Average profiles were also deduced for five different
time intervals (Fig. 6 (click here)) and for each year. The latter are not presented here,
but are used below.
Figure 5:
Average
spectra for all the period of our observations
(1984 December - 1996 March)
Figure 6:
a and b) Average spectra deduced for five time periods
of the evolution of RT Vir maser emission: a) 1984 December -
1985 October, b) 1985 December - 1987 March, 1987 April - 1990 November,
1990 December - 1992 December and 1993 January - 1996 March
The amplitude and the velocity of the more prominent features of the annual
average profile were measured (Fig. 7 (click here)). The circle size in
Fig. 7 (click here) is proportional
to the feature intensity. As may be seen in Fig. 7 (click here), the circles are not randomly
distributed, and are located along four tracks. This means that the maser
activity in RT Vir is manifested in four spectral ranges with average velocities
of 10.7, 12.5, 23.5 and 
respectively. A minimal velocity drift
of 
was observed at about 
, and a maximal drift
of 
was observed at about 
.
The strongest maser emission flares have taken place in the second and
fourth spectral ranges (12.5 and ).
Figure 7:
Velocity drift of the principal spectral features of higher
maser activity in RT Vir, obtained from the annual average spectra.
The size of the circle depends on the intensity of the features
Besides the average profiles we plotted the superposed profiles for the same periods (Figs. 8 (click here)a and b). The most frequent superposition of lines represents the most probable profile. Such spectra obtained for each of the five time intervals well illustrate the existence of anticorrelation of the integrated fluxes of two spectral groups (Fig. 9 (click here)). The full superposition of all profiles obtained from 1984 December to 1996 March is given in Fig. 8 (click here)c.
Figure 8:
a-c) Superpositon of the profiles for RT Vir at different
periods: a) 1984 December - 1985 October, b) 1985 December - 1987
March, 1987 April - 1990 November, 1990 December - 1992 December and 1993
January - 1996 March, c) all the period
Figure 9:
The more probable profiles