Calibration of the JAGB method for the Magellanic Clouds and Milky Way from Gaia DR3, considering the role of oxygen-rich AGB stars

TL;DR

This study calibrates the JAGB distance measurement method for the Magellanic Clouds and Milky Way using Gaia DR3 data, highlighting the impact of oxygen-rich AGB star contamination and metallicity effects on the method's accuracy.

Contribution

It provides an empirical calibration of the JAGB method considering O-rich star contamination and metallicity variations across different galaxies.

Findings

LFs of SMC and LMC JAGB stars differ but mean magnitudes agree within 0.02 mag.

Oxygen-rich star contamination is significant in the Milky Way sample, especially at lower (J-Ks)0 values.

The absolute calibration of the JAGB method depends on metallicity, contrary to previous claims.

Abstract

The JAGB method is a new way of measuring distances with use of AGB stars that are situated in a selected region in a J versus J-Ks CMD, using the fact that the absolute J magnitude is (nearly) constant. It is implicitly assumed in the method that the selected stars are carbon-rich AGB stars. However, as the sample selected to determine M_J is purely colour based there can also be contamination by oxygen-rich AGB stars. As the ratio of C to O-rich stars is known to depend on metallicity and initial mass, the star formation history and age-metallicity relation in a galaxy should influence the value of M_J. The aim of this paper is to look at mixed samples of O- and C-rich stars for the LMC, the SMC and Milky Way (MW), using the Gaia catalogue of long period variables as basis. We report the mean and median magnitudes and the results of fitting Gaussian and Lorentzian profiles to the luminosity function (LF) using different colour and magnitudes cuts. For the SMC and LMC we confirm the previous results in the literature. The LFs of the SMC and LMC JAGB stars are clearly different, yet the mean magnitude inside a selection box can be argued to agree at the 0.02~mag level. The analysis of the MW sample is less straightforward. The contamination by O-rich stars is substantial for a classical lower limit of (J-Ks)0= 1.3, and becomes less than 10% only for (J-Ks)0= 1.5. The sample of AGB stars is much smaller than for the MCs for two reasons. Nearby AGB stars tend to be absent as they saturate in the 2MASS catalogue, and the parallax errors of AGB stars tend to be larger compared to non-AGB stars. The mean and median magnitudes are fainter than for the MC samples by about 0.4~mag which is not predicted by theory. We do not confirm the claim in the literature that the absolute calibration of the JAGB method is independent of metallicity up to solar metallicity.