[Previous] | [Session 5] | [Next]
E.M. Halbedel (Corralitos Observatory)
Be stars are known to be photometrically active on a range of time scales from hours to decades. The most common and largest amplitude variations are those which have cycle times of years to decades. Occasional abrupt brightening and fading events have been observed to interrupt these generally smooth light curves. The brightenings are almost flare-like in nature, the fadings eclipse-like. Such events were searched for amongst the 403 stars of the author's long-term Be star BV photometric monitoring program. In addition, Hipparcos satellite data for brighter Be stars were examined.
In general, it was found these these flares and descents are rare occurences, with flares being slightly more numerous than descents. Some stars do undergo both, but this is very uncommon. Almost half undergo more than one event, with multiple descents occurring more frequently. Times between multiple events are usually <1000 days. Both flares and descents are generally of small range (<.2 V mags), though stars can show much larger changes. Both types of events occur most frequently on the descending portions of the long-term light curves. The majority of both flares and descents cause no B-V color changes. In cases where changes do occur, opposite directions are more common (i.e., brighter V, redder B-V); same direction changes are very rare. Though both types of events are concentrated in spectral subtypes B0-3, >1/3 of descents occur in stars later than B4. For both flares and descents, the larger range events belong to lower v sin i stars, though v sin i does not predict either which type is shown, whether a color change will accompany it, or where on the long-term light curve the event happens. V amplitudes are largest for flares that occur on the most stable portion of the long-term curve, while for descents it is on the ascending branch. Opposite direction color changes are most frequent in stars with the largest V amplitudes. Flare ranges are larger for earlier spectral subclasses, but there is no such relation for descents. Cycle time between events is not related to either v sini or spectrum subclass. No estimate of duration of the events is possible at this time due to observational frequency --- a best guess would be a month or less.