Circumstellar Structures Near  Cas
Inferred from Ultraviolet Continuum
and Si IV Line Variations

This is the fourth report appearing in these newsletters of a optical/UV/X-ray campaign carried out on the prototypical B0.5e star  Cas just over two years ago. The thrust of the first part of our work was to show that the X-rays from this source can be easiest understood as originating from near the surface of the Be star itself, with no hypothetical companion needed (Paper I: Smith, Robinson, & Corbet, ApJ, 503, 877, 1998). In Paper II (Smith, Robinson, & Hatzes 1998 ApJ, 507, 945) we demonstrated from light curves generated from IUE and HST/GHRS time-series specra that a few cool clouds are forced into co-rotation over the star and attenuate far-UV continuum flux during each transit. The work reported below by Smith and R. D. Robinson will be published in the June 10th edition of the Astrophysical Journal. Interested readers may download the paper in an anonymous ftp area (cd to pub/GAMMACAS3) in nobel.stsci.edu. A popular-level article on the background issues leading to this work was published in the May/June, 1998 issue of Astron. Soc. Pacific's Mercury Magazine.

In this study of GHRS data we used difference spectra from our 21+ hour time series to investigate 2% spectral variations within the photospheric and wind Si IV 1394-1403 lines as well as smaller variations from features in neighboring wavelengths at 1382-1386 and 1404-1417. This time series is shown in Figure 1. Note both the diagonal striations, due to "migrating subfeatures" (MSFs), which dominate the top panel and the sharp, generally static absorptions. The lower panel shows an enlargement during Orbits 8 and 12 (out of 13) in the observing sequence.

Fig. 1: Grayscale of 21+ hour time series of  Cas in the spectral region surrounding the UV Si IV doublet (central absorptions are time-varying Discrete Absorption Components, not discussed here).

Using Hubeny model-atmospheres codes and a Kurucz line list, we first computed a grid of cloud opacity for various temperatures in our spectral range. Using this synthetic spectrum, we were able to identify most features as optically thick absorptions due to Fe II, Cr II, and C I lines from `"cool" (T<10,000K) plasma, of Si IV, Si III, S IV, and Ni II lines from "warm" plasma (10,000-18,000K), and of Si IV and Fe V lines from hot plasma (30,000K). Examples of the faint absorptions from cool- and hot- plasma are shown in Figures 2 and 3, respectively.

Fig. 2: Archival GHRS spectrum (compliments of D. Meyer) of the region near the ISM O I line (two sharp absorptions), which also contains several broader features probably arising from circumstellar material occulting  Cas. The lower spectra are modeled fits of opacity spectra from putative intervening clouds with temperatures of 7000K and 8000K. Note the good matches in wavelength.

Fig. 3: Identification of a complicated set of sharp stationary features in the red portion of our Orbit 13 data (solid line). Arrows denote matches for Si IV 1403 (and by dotted line for 1394) at a variety of wavelengths (radial velocities). The horizontal "combs" show identifications of redshifted systems of Si IV lines as well as Fe V lines at 1402.3, 1406.7, 1409.4, and 1415.1. The emission lines at the bottom are from the calculated zero-velocity opacity spectrum for a heated cloud of 34,000K. All identified lines (with one exception noted) are due to Si IV or Fe V.

The variability of the cool- and hot-plasma lines varies in phase with the UV continuum light curve. The warm-plasma line curves are also phase-correlated but lead the curves for hot- and cool- plasma lines by 3-4 hours.

The cool- and warm-plasma lines, except for Si IV, comprise the MSFs. The velocity range for these lines is consistent with limits of ±v sin i, leading us to conclude that they are formed in cloudlets forced into co-rotation around the star (about 1.7R_* above its surface), similar to the clouds which produce the dips in the UV continuum light curve.

In contrast to this behavior, hot-plasma features are seen as narrow absorptions, as exemplified in Figure 3, are "ultra sharp features" (USFs) which have a constant velocity over their 1 to 5 hour lifetimes. The USFs features are seen over a wide velocity range, in some cases having a velocity of at least +1500 km s^-1 (Fig. 3).

Fig. 4: magnetic flaring interaction between star and disk

The cooling and heating of circumstellar plasma near  Cas is inconsistent with radiative processes and suggests that magnetic dissipative processes precipitate plasma instabilities, such as those observed in solar prominences. Yet, some Bp/Ap stars show evidence of co-rotating clouds as well, but not of highly dynamic UV and X-ray activity as we have described. So what makes  Cas a unique Be star in these respects? To answer this we consider that the existence of the stationary USF absorptions, particularly at large positive velocities, is difficult to explain for an isolated star, even if one adds magnetic loops to the picture.

However, suppose we think of the USFs as signatures of mass ejections, similar to solar "coronal mass ejections," only ejected both toward and away from the star. In that case, we are forced to consider magnetic activity arising also from another nearby place. That other place would most likely be the circumstellar disk. Because the disk orbits the star at a different angular rate than the rotation rate, the interaction of field lines from the star and the disk would create powerful stresses, leading to entanglements, reconnections, and the creation of explosive flares and high temperatures. Two cartoon views of the ejection of masses from points near the star's surface (top panel) or in the disk (bottom) are depicted in Figure 4.

While speculative, this picture is not very dissimilar to loop-loop pictures currently envisioned in some T Tauri stars with bi-polar outflows. Such a configuration requires both a magnetic star and a dense disk, so the field interactions would not happen for the vast majority of Be stars, or even at most times for a given Be star. This idea holds the potential of explaining the puzzlement that the X-rays from  Cas are unique both in their high temperature (T 10⁸K) and flaring nature. It also predicts what (rare) conditions are needed for  Cas-analogs to form.