Automated Classification of a Mysterious Population of Weak CN Stars in the Andromeda Galaxy

Stellar evolution has been researched for decades, yet certain phases of the stellar life cycle remain poorly studied. Recent analyses of spectroscopic and photometric data of stars in the Andromeda Galaxy (M31) from the Spectroscopic Landscape of Andromeda's Stellar Halo (SPLASH) and the Panchromatic Hubble Andromeda Treasury (PHAT) surveys have revealed an unusual population of "weak CN" stars in the galaxy's disk that may represent one such poorly studied phase. Based on their positions in color-magnitude space, members of this previously unknown population appear to be associated with the He-burning phase of relatively massive stars. Their defining feature is a weak double-peaked spectral absorption line at around 8000Å indicating the presence of the CN molecule. This same double-peaked feature appears more strongly in the spectra of carbon stars, leading us to hypothesize that an association exists between the two populations. Using data from SPLASH and PHAT, we sought to investigate this probable association by developing an algorithm capable of objectively classifying stars as "carbon," "weak CN," or "other" based on spectral features. The algorithm employed a multi-metric comparison performed against a training set of visually classified stars to make automated classifications. Stars the algorithm classified as "weak CN" were subsequently plotted in color-magnitude diagrams (CMDs) to photometrically compare the training set with the algorithm's spectroscopically-oriented classifications. Results obtained from this algorithm and the CMDs have been instrumental in confirming the significance of weak CN stars as a relatively well-defined stage of stellar evolution and will ultimately contribute to a greater understanding of the physical properties of both carbon and weak CN stars. This research was funded in part by the National Science Foundation and National Aeronautics and Space Administration/Space Telescope Science Institute. High school students Alexandra Masegian and Arya Maheshwari participated in this research under the auspices of the Science Internship Program at the University of California Santa Cruz.