Spectral classification of the 100 pc white dwarf population from Gaia-DR3 and the virtual observatory

The third data release of Gaia has provided low-resolution spectra for similar to 100 000 white dwarfs (WDs) that, together with the excellent photometry and astrometry, represent an unrivalled benchmark for the study of this population. In this work, we first built a highly complete volume-limited sample consisting in 12 718 WDs within 100 pc from the Sun. The use of Virtual Observatory Spectral energy distribution Analyzer tool allowed us to perform an automated fitting of their spectral energy distributions to different atmospheric models. In particular, the use of spectrally derived Javalambre-Physics of the Accelerating Universe Astrophysical Survey photometry from Gaia spectra led to the classification of DA and non-DA WDs with an accuracy >90 per cent, tested in already spectroscopically labelled objects. The excellent performance achieved was extended to practically the whole population of WDs with effective temperatures above 5500 K. Our results show that while the A branch of the Gaia WD Hertzsprung-Russell diagram is practically populated by DA WDs, the B branch is largely formed by non-DAs (65 per cent). The remaining 35 per cent of DAs within the B branch implies a second peak at similar to 0.8 M-circle dot in the DA mass distribution. Additionally, the Q branch and its extension to lower temperatures can be observed for both DA and non-DA objects due to core crystallization. Finally, we derived a detailed spectral evolution function, which confirms a slow increase of the fraction of non-DAs as the effective temperature decreases down to 10 500 K, where it reaches a maximum of 36 per cent and then decreases for lower temperatures down to similar to 31 per cent.

Automated summary

By analyzing low-resolution spectra, researchers have established a sample of 12,718 white dwarfs within 100 light-years from the Sun. The use of the Virtual Observatory Spectral energy distribution Analyzer tool allowed for the automated fitting of their spectral energy distributions and the successful classification of DA and non-DA white dwarfs with an accuracy over 90%. Results show that the proportion of non-DA white dwarfs increases as the effective temperature decreases.