Gaia-DR2 has provided an unprecedented number of white dwarf candidates of ourGalaxy. In particular, it is estimated thatGaia-DR2 has observed nearly 400 000 ofthese objects and close to 18 000 up to 100 pc from the Sun. This large quantity ofdata requires a thorough analysis in order to uncover their main Galactic popula-tion properties, in particular the thin and thick disk and halo components. Takingadvantage of recent developments in artificial intelligence techniques, we make useof a detailed Random Forest algorithm to analyse an 8-dimensional space (equato-rial coordinates, parallax, proper motion components and photometric magnitudes)of accurate data provided byGaia-DR2 within 100 pc from the Sun. With the aid ofa thorough and robust population synthesis code we simulated the different compo-nents of the Galactic white dwarf population to optimize the information extractedfrom the algorithm for disentangling the different population components. The algo-rithm is first tested in a known simulated sample achieving an accuracyof 85.3%. Ourmethodology is thoroughly compared to standard methods based on kinematic criteriademonstrating that our algorithm substantially improves previous approaches. Oncetrained, the algorithm is then applied to theGaia-DR2 100 pc white dwarf sample,identifying 12 227 thin disk, 1410 thick disk and 95 halo white dwarf candidates, whichrepresent a proportion of 74:25:1, respectively. Hence, the numerical spatial densitiesare (3.6±0.4)×10-3pc-3, (1.2±0.4)×10-3pc-3and (4.8±0.4)×10-5pc-3forthe thin disk, thick disk and halo components, respectively. The populations thus ob-tained represent the most complete and volume-limited samples to date of the differentcomponents of the Galactic white dwarf population.

Peer Reviewed