Astron. Astrophys. 345, 925-935 (1999)

5. Conclusion

Within the uncertainties encountered in the observations as discussed, Figs. 1, 2 & 4 and Tables 1-3 show acceptable comparisons between the predicted and observed integrated H₂O spectra and FIR continua. The error bars for observed SWS line fluxes are nevertheless high by a factor 2-3; and it is interesting to perform new observations in this wavelength region. Some CO and CII lines are also found in the H₂O spectra but their investigation is outside the scope of this paper. Observations I show a spectrum about 30-40% less intense compared to Observations II's where new or stronger lines appear and some components of a blend become clearly visible. This light variation due to stellar pulsations of the Mira variable between 300 days corresponds to a phase variation from 0.5, at minimum luminosity, to 0.2 where the luminosity was falling down to its mean value. A best fit to the overall spectra were obtained with the dilution factor for the FIR field scaled by a factor 0.61 and 1, respectively. This FIR field is mainly dominated by dust. Photodissociation of H₂O by the interstellar UV field does not affect significantly the FIR H₂O line fluxes, since the effect operates in the outer region ( cm) where H₂O excitation temperatures become weak. The derived H₂O abundance and density, the FIR continuum, and the kinetic temperature vary as functions of radius across the circumstellar envelope of R Cas. The derived mass-loss rate is 3.4   . The deduced total (ortho and para) H₂O abundance is 1.1 and the kinetic temperature is 1300 K at the inner radius of the envelope, and they decrease with distance from the star. Compared to W Hya, another O-rich AGB star, the initial H₂O abundance of R Cas is about one order of magnitude lower while the mass-loss rate is nearly the same; the values found for W Hya (1.2 and 6   ; see Barlow et al. 1996) were determined, however, by using a different model. Similar mass-loss rates ( ) have been derived for a number of LPV Miras using CO radio observations (e.g. Knapp et al. 1982, Loup et al. 1993, Kahane & Jura 1994, Gonzalez-Alfonso et al. 1998), and optical KI /NaI scattering (e.g. Guilain & Mauron 1996).

© European Southern Observatory (ESO) 1999

Online publication: April 28, 1999
helpdesk.link@springer.de