Mapping the Milky Way: Data-miners, Modelers, Observers, Unite!

Mapping the Milky Way: Data-miners, Modelers, Observers, Unite!

Recently, the big data analytics technique of mapping the distributions of very large numbers of individual stars in spatial-kinematic-metallicity-age space has emerged as an invaluable tool for creating models of galaxy structure and evolution. Studies of the structure and dynamics of our Milky Way galaxym in particularm advance a pathway to understanding the details of galaxy evolution in general. These big data analytics “near-field cosmology” studies in turn place constraints on models of universal evolution and may further refine Λ-CDM as the dominant cosmological model.

Variable stars, identified using big data analytics, are particularly useful tracers of galactic structure. Their utility derives from well-defined luminosities for main sequence pulsators and simple relations correlating observable parameters, such as period, amplitude of pulsation and metallicity, with evolutionary parameters such as luminosity for evolved pulsators. Evolved variables are sufficiently bright to be seen beyond the halo number density steepening at ~30 kpc for a survey with limiting magnitude V ≈ 18. Additional corrections to their absolute magnitude due to both period and metallicity make these standard candles with a small scatter. These characteristics combine to make evolved pulsators useful tracers of the old, population II, structure of our galactic halo. Metallicity estimates of detected stars additionally constrain galaxy accretion models that predict a difference in chemical composition between the inner (old) and outer (recent) halo. Using big data analytics, we can map the distributions of these stars in spatial-kinematic-metallicity-age space to achieve our scientific objectives.

To more fully compare simulations of hierarchical galaxy formation to our local environment, we will need to follow filaments of substructure across the entire sky, requiring an increase in the depth and breadth of the known stars' populations. Big data analytics studies of sources in three major new databases of time-resolved photometric observations are being data-mined to study these valuable periodic variable sources.