A re-classification of Cepheids in the Gaia Data Release 2

TL;DR

This paper reviews and re-classifies Gaia DR2 Cepheids, derives precise Period-Luminosity and Period-Wesenheit relations, and investigates their implications for the cosmic distance scale and Hubble constant estimation.

Contribution

It provides a new classification of Gaia DR2 Cepheids and derives improved PL/PW relations, including an analysis of metallicity dependence and zero point calibration.

Findings

Re-classified Gaia DR2 Cepheids into classical, anomalous, and Type II categories.

Derived new precise PL/PW relations for Magellanic Cloud and Galactic Cepheids.

Identified a potential underestimation of Gaia DR2 parallax zero point by ~0.07 mas.

Abstract

Classical Cepheids are the most important primary indicators for the extragalactic distance scale. Establishing the precise zero points of their Period-Luminosity and Period-Wesenheit (PL/PW) relations has profound consequences on the estimate of $\rm H_0$. Type II Cepheids are also important distance indicator and tracers of old stellar populations. The recent Data Release 2 (DR2) of the {\it Gaia} Spacecraft includes photometry and parallaxes for thousands of classical and type II cepheids. We aim at reviewing the classification of {\it Gaia} DR2 Cepheids and to derive precise PL/PW for Magellanic Cloud (MCs) and Galactic Cepheids. Information from the literature and the {\it Gaia} astrometry and photometry are adopted to assign DR2 Galactic Cepheids to the classes: Classical, Anomalous and Type II Cepheids. We re-classify the DR2 Galactic Cepheids and derive new precise PL/PW relations in the {\it Gaia} passbands for the MCs and Milky Way Cepheids. We investigated for the first time the dependence on metallicity of the $PW$ relation for Classical Cepheids in the {\it Gaia} bands, finding non-conclusive results. According to our analysis, the zero point of the {\it Gaia} DR2 parallaxes as estimated from Classical and Type II Cepheids seems to be likely underestimated by $\sim$0.07 mas, in full agreement with recent literature. The next {\it Gaia} data releases are expected to fix this zero point offset to eventually allow a determination of $\rm H_0$ to less than 1\%.