Machine-learned Identification of RR Lyrae Stars from Sparse, Multi-band Data: The PS1 Sample

Abstract

RR Lyrae stars may be the best practical tracers of Galactic halo (sub-)structure and kinematics. The PanSTARRS1 (PS1) 3π survey offers multi-band, multi-epoch, precise photometry across much of the sky, but a robust identification of RR Lyrae stars in this data set poses a challenge, given PS1's sparse, asynchronous multi-band light curves ( 12 epochs in each of five bands, taken over a 4.5-year period). We present a novel template fitting technique that uses well-defined and physically motivated multi-band light curves of RR Lyrae stars, and demonstrate that we get accurate period estimates, precise to 2 sec in > 80% of cases. We augment these light curve fits with other features from photometric time-series and provide them to progressively more detailed machine-learned classification models. From these models we are able to select the widest (3/4 of the sky) and deepest (reaching 120 kpc) sample of RR Lyrae stars to date. The PS1 sample of ∼ 45, 000 RRab stars is pure (90%), and complete (80% at 80 kpc) at high galactic latitudes. It also provides distances precise to 3%, measured with newly derived period-luminosity relations for optical/near-infrared PS1 bands. With the addition of proper motions from Gaia and radial velocity measurements from multi-object spectroscopic surveys, we expect the PS1 sample of RR Lyrae stars to become the premier source for studying the structure, kinematics, and the gravitational potential of the Galactic halo. The techniques presented in this study should translate well to other sparse, multi-band data sets, such as those produced by the Dark Energy Survey and the upcoming Large Synoptic Survey Telescope Galactic plane sub-survey.

The Profile of the Galactic Halo from Pan-STARRS1 3π RR Lyrae

“…In this section, we describe the pertinent properties of the PS1 3π survey and the RR Lyrae light curves obtained, and we recapitulate briefly the process of selecting the likely RRab, as laid out in Sesar et al (2017). We also briefly characterize the obtained candidate sample.…”

“…Our analysis is based on a sample of highly likely RRab stars, as selected by Sesar et al (2017) from the Pan-STARRS1 3π survey. In this section, we describe the pertinent properties of the PS1 3π survey and the RR Lyrae light curves obtained, and we recapitulate briefly the process of selecting the likely RRab, as laid out in Sesar et al (2017).…”

“…Starting with a sample of more than 1.1×10 9 PS1 3π sources, Hernitschek et al (2016) and Sesar et al (2017) subsequently selected a sample of 44,403 likely RRab stars, of which ∼17,500 are at R gc 20 kpc, by applying machinelearning techniques based on light-curve characteristics. RRab stars are the most common type of RR Lyrae, making up ∼91% of all observed RR Lyrae (Smith 2004) and displaying the steep rises in brightness typical of RR Lyrae.…”

“…The predicted density of stars is then simply the product of the underlying density distribution with the selection function, suggesting that one constrains this underlying density by forward modeling of the observations. The RRab candidates from Sesar et al (2017) were selected uniformly from the set of objects in the PS1 3π survey in the area and apparent magnitude range available for this survey. The selection completeness and purity are uniform over a wide range of apparent magnitude up to a flux-averaged r-band magnitude of 20 mag , which is described later on in Equation (12).…”

“…Hernitschek et al (2016) and Sesar et al (2017) exploited the PS1 survey to create a sample of RRab that reaches far into the outer halo, which is very large (44,403 RRab), with known high purity and completeness. The details of the machine-learning techniques that were used to select this sample and the assessment of its purity and completeness as a function of distance are described in Hernitschek et al (2016) and Sesar et al (2017).…”