Astron. Astrophys. 319, 607-616 (1997)

4. Variability

Experience shows that the CO velocity coverage does not change with time, therefore any change in the ratio R means a change in the SiO line, presumably due to stellar activity.

If the SiO line wings are related to the stellar pulsation activity of Mira stars, one could expect a variation of R with the stellar phase. R Leo is perhaps the most representative case showing extreme SiO wing variations. Cernicharo et al (1994) reported broad red and blue wings (R =1.90) and discrete maser features at large red velocities in different SiO maser lines. We have monitored the SiO v =1 J =2 1 maser emission from this star with the IRAM interferometer (unpublished data). Gathering the interferometer and 30 m data, R varies from 0.96 in April 1991 to 1.53, 1.90, 0.83, 0.86, 0.67, 1.0, 1.08 and 1.0 in June 1992, November 1992, January 1993, May 1993, November 1993, January 1994, May 1994, and June 1995 respectively (the light phase being 0.3, 0.7, 0.2, 0.4, 0.8, 0.4, 0.6, 0.0, and 0.6 respectively).

In 1995 we have reobserved a few objects. The SiO profiles are shown in Fig. 5, together with those obtained in 1994 for SiO and CO. The line profile in all these objects shows an important variation between both periods. The more remarkable cases are o Cet, T Cas, R LMi, and W Hya. o Cet is known to have a bipolar structure for the distribution of the CO emission (Planesas et al, 1990; Planesas, Kenney and Bachiller 1990).

In our SiO spectrum of 1995 the emission reaches the red edge of the CO profile and approaches the blue one. However, in 1994 the SiO emission was narrower and concentrated toward the central velocities. Our 1995 spectrum is much similar to that observed by Wright et al. in 1990. The value of R changes from 0.44 in 1994 (see Table 1) to 0.85 in 1995 (stellar phase = 0.6 and 0.2 respectively). Moreover, the 1995 spectrum shows prominent blue and red wings that were not present in 1994. These wings are not discrete features but form a continuous and strong broad pedestal (see Fig. 5 top panel). The H₂ O VLA map of this source by Bowers and Johnston (1994) shows a line with a width at half power of 2 kms^-1 and a spatial distribution indicating an elongated structure in the NE-SW direction. Other observations in the optical and infrared also indicate an asymmetric structure near the star (Karovska et al 1991, Haniff et al 1992; Danchi et al 1994). These observations could suggest that asymmetric mass loss processes may play a role in the formation of blue and red wings maser emission in SiO.

T Cas shows impressive changes in the SiO profile. In the 1995 observations the line profile exhibits several features in the red part of the CO line. The strongest feature is centered at the terminal velocity and exceeds the CO line by roughly 5 km s^-1. The value of R changes from 0.56 to 1 between both observing periods (stellar phase = 0.9 and 0.0 respectively). In the blue part, the velocity extent looks similar in 1994 and 1995; however, in the latter observations the blue SiO emission is stronger (see bottom-right panel of Fig 5).

R LMi shows in 1995 a double peaked profile with a weak feature near the central velocity of the star. The SiO line profile reaches and exceeds the CO profile in the red. The R value in 1995 is 0.9 (stellar phase = 0.8) while in 1994 it was 0.7 (stellar phase = 0.4).

Contrary to other stars, the SiO line profile in W Hya is narrower in 1995 than in 1994. Nevertheless, R is 1 in 1995 (see Fig. 5). In 1994 this value was 1.3 (see Table 1). The correspondig stellar phases were 0.8 and 0.1 for 1994 and 1995 observations.

Finally, the most impressive change in the SiO line profile was observed in the supergiant µ Cep (see Fig 1). All of the high velocity discrete features detected in 1994 disappeared in 1995. Although the CO line profile is blended with interstellar emission and R is poorly determined, R changes by a factor 2 between both observing runs.

The broad wings reported by Cernicharo et al (1994) in R Leo for SiO v =1 J =2 1 and the discrete features at high velocities for other high v transitions of SiO have been only observed in 1992. Some of the additional observations for the same star are at similar light phases but only moderate wings have been observed. It is worth noting that during the 1992 observations of R Leo, the SiO maser emission had a particular strong maximum (Alcolea et al, in preparation). In other stars we find that R increases after the optical maximum (o Cet, T Cas, R LMi) while in W Hya the situation is the opposite. These results could suggest that in some objects (e.g. R Leo, W Hya) there are changes in spatial structure and velocity field over a timescale of a few months/years not related to the periodic optical pulsation of Mira stars. Observations of H₂ O in W Hya by Reid and Menten (1990) and Bowers and Johnston (1994) also indicate similar results (note that this object shows a prominent red wing in Fig. 2; this wing disappeared in 1995 -see Fig. 5). Although our limited data do not indicate correlation between R and the stellar phase, a more detailed monitoring study of the SiO emission is necessary to determine the effect of stellar phase on the emergent SiO profiles. All other Miras with R 1 had a light phase in January 1994 between 0.5 and 1.0.

© European Southern Observatory (ESO) 1997

Online publication: July 3, 1998
helpdesk@link.springer.de