Forum Springer Astron. Astrophys.
Forum Whats New Search Orders

Astron. Astrophys. 349, 169-176 (1999)

Previous Section Next Section Title Page Table of Contents

5. Line blanketing in RW Hya

So far Rayleigh scattering has been observed in the symbiotic systems EG And (Vogel 1991), SY Mus (Pereira et al. 1995), BF Cyg (Gonzalez-Riestra et al. 1990), and RW Hya (this paper). Only in the case of EG And is Rayleigh scattering sufficient to interpret the continuum attenuation. In the other cases there is clear evidence for additional absorption. This additional attenuation has been accounted for by a wavelength independent term of unknown origin (Gonzalez-Riestra et al. 1990, Pereira et al. 1995). Pereira et al. (1995) suggest, that the additional attenuation is produced by Thomson scattering by free electrons in the nebula. However, according to Schmid (1995), high electron densities of [FORMULA] in the ionized region would be needed. Here, we put forward an alternative explanation: line blanketing due to mostly Fe II transitions. This "iron-curtain" has first been identified by Shore & Aufdenberg (1993) as the origin of the distorted emission lines and anomalous emission line ratios observed in symbiotic binaries. The iron curtain has also been observed in the spectrum of PU Vul (Schmutz et al. 1993). The iron absorption line forest is not resolved in IUE low resolution spectra, and could thus be the cause of the almost wavelength independent attenuation, observed during eclipse in SY Mus. There are no IUE high resolution spectra of SY Mus during ingress or egress.

As RW Hya and SY Mus show very similar observational characteristics (Schmutz et al. 1994, Schild et al. 1996) we analyze HST GHRS spectra of RW Hya to investigate the origin of the additional continuum attenuation in SY Mus.

In Fig. 7 we show an extract of [FORMULA] of the HST spectra of RW Hya taken at egress phase. In this wavelength region, the continuum is dominated by the hot star. It is distorted by numerous Fe II absorption lines. These lines are only seen during the ingress and egress phases and they have their origin in the extended atmosphere or wind of the red giant. These lines require high spectral resolution to be resolved, and it is therefore impossible to measure correctly the absorption due to Rayleigh scattering with IUE low resolution spectra. IUE final archive low resolution short wavelength spectra taken through the large aperture have a resolution of [FORMULA]. In the presence of numerous absorption lines, this leads to a significant underestimation of the stellar continuum. In Fig. 7 we have marked the continuum level that would be seen in an IUE low resolution spectrum, which is at about half of the true continuum value.

[FIGURE] Fig. 7. Observed GHRS medium resolution spectrum of RW Hya at phase [FORMULA] (thick line) compared with a simple "absorption only" radiation transfer simulation (thin line). Most absorption features are due to Fe II. The dotted line gives the continuum as it would be seen in a IUE low resolution spectrum. Flux in units of [FORMULA] erg cm-2 s- 1 Å-1.

We have fitted HST spectra of RW Hya with a simple plane-parallel "absorption only" radiation transfer model where the level populations of the transitions are calculated in LTE. The input parameters were: a radiation temperature of [FORMULA], an electron temperature of [FORMULA], micro-turbulence of [FORMULA], an ion density of [FORMULA], solar iron abundance and the thickness of the absorbing layer corresponding to [FORMULA]. The electron temperature is in the range predicted by the models of Schwank et al. (1997). The resulting fit is shown in Fig. 7 and Fig. 8. With our model, we qualitatively reproduce the majority of the absorption lines.

[FIGURE] Fig. 8. HST low resolution spectra of RW Hya at [FORMULA] (solid) together with the "absorption only" radiation transfer simulation (dotted). Flux in units of [FORMULA] erg cm-2 s- 1 Å-1.

We now consider the effect of spectral resolution on the derived column density of neutral hydrogen. Based on the HST medium resolution spectra observed at [FORMULA], we find [FORMULA]. A Rayleigh scattered Planck spectrum together with the measured continuum points is shown in Fig. 9. If we convolve the HST spectrum with the IUE resolution, then we fit the spectrum with [FORMULA], together with a wavelength independent attenuation factor [FORMULA]. Thus the column density of neutral hydrogen derived from an IUE low resolution spectrum, would be underestimated by a factor [FORMULA].

[FIGURE] Fig. 9. HST low resolution spectra of RW Hya at [FORMULA], [FORMULA] and [FORMULA] (top to bottom). We also mark the continuum values derived from the HST medium resolution spectra taken at [FORMULA] and [FORMULA] (filled squares). Also shown is a Planck function belonging to [FORMULA], with and without Rayleigh scattering on [FORMULA]. Flux in units of [FORMULA] erg cm-2 s- 1 Å-1.

Previous Section Next Section Title Page Table of Contents

© European Southern Observatory (ESO) 1999

Online publication: August 25, 1999