In our 2013 Astronomical Review article, we discussed the statistics of variability for 633 faint spectral type A-F stars observed by the Kepler spacecraft during Quarters 6-13. We found six stars that showed no variability with amplitude 20 ppm or g
reater in the range 0.2 to 24.4 cycles/day, but whose positions in the log g--Teff diagram place them in the delta Sct or gamma Dor pulsation instability regions established from pre-Kepler ground-based observations. Here we present results for 2137 additional stars observed during Quarters 14-17, and find 34 stars that lie within the instability regions. In Paper I, we included a +229 K offset to the Kepler Input Catalog Teff to take into account an average systematic difference between the KIC values and the Teff derived from SDSS color photometry for main-sequence F stars (Pinsonneault et al. 2012). Here we compare the KIC Teff value and the Teff derived from spectroscopy taken by the LAMOST instrument (Molenda-Zakowicz et al. 2013, 2014) for 54 stars common to both samples. We find no trend to support applying the offset, but instead find that a small average temperature decrease relative to the KIC Teff may be more appropriate for the stars in our spectral-type range. If the offset is omitted, only 17 of our 34 `constant stars fall within the instability regions. For the two `constant stars also observed by LAMOST, the LAMOST Teff values are cooler than the KIC Teff by several hundred K, and would move these stars out of the instability regions. It is possible that a more accurate determination of their Teff and log g would move some of the other `constant stars out of the instability regions. However, if average (random) errors in Teff are taken into account, 15 to 52 stars may still persist within the instability regions. Explanations for these `constant stars, both theoretical and observational, remain to be investigated.
As part of the NASA Kepler Guest Observer program, we requested and obtained long-cadence data on about 2700 faint (magnitude 14-16) Kepler stars with effective temperatures and surface gravities that lie near or within the pulsation instability regi
on for main-sequence gamma Doradus and delta Scuti pulsating variables. These variables are of spectral type A-F with masses of 1.4 to 2.5 solar masses. The delta Scuti stars pulsate in radial and non-radial acoustic modes, with periods of a few hours (frequencies around 10 cycles/day), while gamma Doradus variables pulsate in nonradial gravity modes with periods 0.3 to 3 days (frequencies around 1 cycle/day). Here we consider the light curves and Fourier transforms of 633 stars in an unbiased sample observed by Kepler in Quarters 6-13 (June 2010-June 2012). We show the location of these stars in the log surface gravity--effective temperature diagram compared to the instability region limits established from ground-based observations, and taking into account uncertainties and biases in the Kepler Input Catalog T_eff values. While hundreds of variables have been discovered in the Kepler data, about 60% of the stars in our sample do not show any frequencies between 0.2 and 24.4 cycles per day with amplitude above 20 parts per million. We find that six of these apparently constant stars lie within the pulsation instability region. We discuss some possible reasons that these stars do not show photometric variability in the Kepler data. We also comment on the non-constant stars, and on 26 variable-star candidates, many of which also do not lie within the expected instability regions.
The Luminous Blue Variable stars exhibit behavior ranging from light curve `microvariations on timescales of tens of days, to `outbursts accompanied by mass loss of up to 10e-03 solar masses per year, occurring decades apart, to `giant eruptions such
as seen in Eta Carinae ejecting one or more solar masses and recurring on timescales of centuries. Here we review the work of the Los Alamos group since 1993 to investigate pulsations and instabilities in massive stars using linear pulsation models and non-linear hydrodynamic models. The models predict pulsational variability that may be associated with the microvariations. Using a nonlinear pulsation hydrodynamics code with a time-dependent convection treatment, we show that, in some circumstances, the Eddington limit is exceeded periodically in the pulsation driving region of the stellar envelope, accelerating the outer layers, and perhaps initiating mass loss or LBV outbursts. We discuss how pulsations and mass loss may be responsible for the location of the Humphreys-Davidson Limit in the H-R diagram. The `giant eruptions, however, must involve much deeper regions in the stellar core to cause such large amounts of mass to be ejected. We review and suggest some possible explanations, including mixing from gravity modes, secular instabilities, the epsilon mechanism, or the SASI instability as proposed for Type II supernovae. We outline future work and required stellar modeling capabilities to investigate these possibilities.
