Research Interest

Photometry of Variable Stars
Long Term Monitoring of Mira Variable Stars
Search for Microvariability in Mira Stars
Monitoring Post-AGB Stars for Variability

For many years I was performing narrow-band, near-infrared photometry of Mira stars. Mira stars are large amplitude, long period pulsating variable stars whose spectra and light curves exhibit peculiarities and cycle to cycle differences. Knowledge of the atmospheric structure of these stars is of fundamental importance to theories of stellar atmospheres and stellar evolution. Along with numerous ETSU undergraduates and summer REU students (see photo below) we carried out a photometric study of several Mira variables using a set of three narrow-band interference filters to monitor an infrared continuum magnitude, a color index, and TiO spectral bands for these stars [see figure below]. Since the light curves of Miras do not repeat well from pulsation cycle to pulsation cycle, continuous monitoring is needed. The narrow-band filter set allows us to determine a light curve similar to a bolometric curve and at the same time determine the spectral type. Our goal was to determine the causes of the differences in the shapes, amplitudes, and phasing of the light curves which may be attributed to changes in temperature, diameter, and/or spectroscopic effects. Students helped to obtain, reduce, and analyze data to produce more complete light curves with parallel spectral type information.

Definition of TiO index. The sketch shows the reduced magnitudes of a typical M star in filters A, B, and C plotted against the filter wavelength; the magnitude in filter A is depressed due to TiO absorbtion. The index is defined as the diffrence (in magnitude units) between the measured flux in filter A and the value predicted from the linear fit to the magnitude points of the continuum filters B and C. (reproduced from Wing, R. F. 1992, J. AAVSO, 21, 47.)

The long-term Mira monitoring program indicated some unusual short-term behavior in a few instances in the light curves of a couple of stars.  The literature indicated a strong possibility of large amplitude short-term variability in a large group of Miras so  beginning in the summer of 2001, a program to monitor several select stars for microvariability in B,V, R, and I filters was begun.  A significant event was detected that first summer in RR Boo (see figure) but over 2 years of intensive monitoring of RR Boo and 5 other stars have not yielded any other significant microvariability events.  The observations do support intrinsic short-term variations of several hundredths of a magnitude (mostly at longer wavelengths) in all the stars, but large amplitude events appear to be rare and unpredictable so far.

 

Recently a new program to monitor Post-AGB stars for variability was begun in collaboration with Dr. Beverly Smith.  In the models for stellar evolution, the transition from star to planetary nebula is still poorly understood.  At the end of the asymptotic giant branch (AGB) stage of stellar evolution, the mass loss rate has risen to such high values, that virtually the entire hydrogen-rich envelope is expelled from the central star. The mass loss ends and by definition the star now enters its post-AGB (or proto-planetary nebula) phase. The central star will evolve to higher temperatures, while the luminosity remains more or less constant. When the effective temperature of the AGB remnant becomes high enough, the radiation will ionize the expanded circumstellar material and this is the birth of a Planetary Nebula (PN). However, only a certain fraction of AGB stars are expected to become a PN.  If the circumstellar material is pushed too far out from the central star or mass loss rates are low or inconsistent, a PN may not form.  We plan to search for variability in a select group of post-AGB stars and a few other unique infrared sources of interest.  The objects will be monitored using both the SARA Observatory and the ETSU Harry D. Powell Observatory (for the brighter objects).  We will observe the stars in as many wavelength bands as possible, but at least in BVR and I bands at intervals of one to two weeks.