Marilyn R. Meade (U. Wisconsin-Madison) and Karen S. Bjorkman (U. Toledo)
Spectropolarimetric monitoring of a number of types of stars has been ongoing for the past 14 years using the Halfwave Polarimeter (HPOL) mounted on the dedicated 0.9m telescope at the University of Wisconsin Pine Bluff Observatory (PBO). The HPOL instrument is described in more detail by Wolff, Nordsieck, & Nook (1996)[postscript]. One of the science programs included in the HPOL observational effort has been monitoring of the spectropolarimetric characteristics and variability of Be stars. Originally done as a ground-based support program for the Wisconsin Ultraviolet Photo-Polarimeter Experiment (WUPPE), the HPOL monitoring program has developed into a large database of excellent quality spectropolarimetric data on a large number of Be stars.
From 1989-1994, HPOL used a dual Reticon array detector, which provided spectral coverage from about 3400-7600Å, with a spectral resolution for both spectroscopy and polarimetric measurements of about 25Å. In 1995, the Reticon was replaced by a CCD detector, which extended the wavelength coverage to 3400-10500Å and improved the spectral resolution to about 12Å. The changeover from one detector to the other thus marked a logical breakpoint for analysis of the data set. During the 1989-1994 period, 297 observations of 61 different Be stars were obtained as part of the monitoring program. We are developing an ATLAS of these observations. The CCD observations will be presented later in future volumes of the atlas.
A full analysis of the data will be included in a paper to be published elsewhere. Here we report some general results regarding the polarimetric variability observed in the sample. Since not all stars were observed the same number of times, we only considered variability for those stars observed 3 or more times during the 1989-94 survey period.
We find that 56% (20 of 36) of the Be stars observed 3 or more times from 1989-94 show significantly variable polarization at the level of 0.1% changes (note that the inclusion of preliminary results from the continuing CCD survey indicates that the total percentage will be even higher). The timescales for these changes range from as short as night-to-night to as long as several months. Several of the stars showed evidence for polarimetric "outbursts" during the time period covered by the observations. Most notable of these was the Be star Pi Aqr, which showed several large polarimetric outbursts during this period. The behavior of the wavelength-dependence of the observed polarization is also strongly variable with the overall level of polarization.
The level of polarimetric variability, together with the observed time scales and the variable wavelength-dependence of the polarization, places constraints on any models for producing such variability. In particular, it must be recognized that many proposed variability mechanisms for Be stars, which originally intended only to explain spectroscopic variations, will also have significant predicted polarimetric consequences, and these can be tested with a large-scale survey such as this one.
Atlas PlansWe plan to publish the atlas of Be stars both in print version (as a journal paper) and electronically via this web site, and the complete data set will be made available to the community as FITS-format files that can be downloaded by users via ftp.
Data representations offered on our web-based version of the atlas include, for each star, the following: summaries of all the observations, including date, time, and JD of each, plus observation details such as exposure length; plots of polarization vs. time, position angle vs. time, and Stokes Q vs U for all observations of a particular star; and a "quick-look" plot showing an overview of all the observational data for a particular star. (See example-Phi Per)
For each individual observation of a given star, the following are available: both graphical (plots) and tabular representations of each observation, including polarization measurements and flux; UBVR filter polarimetry equivalent data for each observation, for use in comparison with data from the literature; H-alpha equivalent width measurements from the spectral data; a detailed combined plot of flux, polarization, and position angle vs. wavelength. (See example-Phi Per 900904 observation)
AcknowledgementsWe thank Ken Nordsieck for providing access to the HPOL instrument and for assistance with data calibration. We are also indebted to the many observers and data reducers at PBO. This work has been supported under NASA contract NAS5-26777 with the University of Wisconsin, and by NASA grants NAG5-3248 and NAG5-8054 to the University of Toledo.
This page is maintained by
Marilyn Meade .