Luminous blue variables and related high-mass evolved stars in M31 and their surprising environments

Luminous Blue Variables (LBVs) are erratically eruptive massive stars near the Eddington limit generally thought to be the descendants of blue supergiants and the progenitors of Wolf-Rayet stars. Recent observations suggest that this standard path is not necessarily true for all LBVs, including the intrinsically bright ones. In order to understand the evolutionary positions of these objects and their relation to other high mass-losing stars in the upper H-R diagram, we would like to know their bolometric luminosities, effective temperatures, current masses, progenitor masses, ages, and abundances. With the use of 3.5-m class telescopes (WIYN and ARC) we are able to roughly constrain all but the abundances for these LBV-type stars in M31. We began the study by developing a method for identifying candidate LBVs in nearby galaxies through deep narrow-band images. The method was used to make a target list of such objects in the NE half of M31. We obtained follow-up spectroscopy that confirmed the success of our search criteria. We obtained high angular resolution broad band and Halpha images of the environments of the first five LBV-type objects and two previously identified M31 LBVs using the WIYN 3.5-m telescope. The spectral energy distribution (SED) of K895 exhibits near infrared excess emission similar to that observed in B[e] supergiants. The calibrated WIYN photometry was used to measure color-magnitude diagrams for the resolved luminous stars. We charted the locations of HII regions and HI holes as indicators of stellar ages in the areas around our program objects. We used the Q-method, after a rough empirical validation of its applicability, to constrain the luminosities and effective temperatures of our program objects. We constrained the present masses and examined the ages predicted for our program objects by the Meynet et al. (1994) stellar evolution models. Surprisingly, the majority of the LBV-type objects were found to be surrounded by stars that appeared to be older than the LBV-type objects themselves. The age discrepancies between LBV or LBV-type object and its environment might be explained if the LBVs or candidate LBVs are runaways, are products of discrete massive star forming events, have somehow lived longer than theoretically expected, or are exotic binaries such as Thorne-Zytkow objects or mass-transferring close binaries. (Abstract shortened by UMI.)