Big-Data Rush: Mining Time-Domain Data of Local Stellar Populations
Monday, 30 March 2020 9 a.m. — 10 a.m. MST
AURA Lecture Hall
With several large-scale sky surveys ongoing and recently completed, the era of Big Data has dawned on optical time-domain astronomy. Abundant archival data provide an opportunity to perform systematic studies of stellar variability in nearby galaxies, and thereby yield constraints on the effect of environmental factors, such as metallicity. I will present results from mining the data from the Palomar Transient Factory survey for photometric variability of luminous stars in the Andromeda Galaxy. The enormous volume and rate of data bring their own challenges. Correspondingly, we have witnessed significant progress in the development of infrastructure and analysis tools to tackle those challenges. For real-time analysis, a triage of the overwhelming time-domain alert data by brokers (e.g., ANTARES) allows timely discovery and characterization of interesting short-lived astrophysical events. As part of the broker, an efficient and effective algorithm for selecting rare and novel events is crucial in the LSST era. I will briefly describe such an algorithm designed in preparation for LSST alerts.