J/A+A/679/A127 100pc white dwarf spectral classification (Garcia-Zamora+, 2023)
White dwarf Random Forest classification through Gaia spectral coefficients.
Garcia-Zamora E.M., Torres S., Rebassa-Mansergas A.
<Astron. Astrophys. 679, A127 (2023)>
=2023A&A...679A.127G 2023A&A...679A.127G (SIMBAD/NED BibCode)
ADC_Keywords: Stars, white dwarf ; Spectral types ; Optical
Keywords: white dwarfs - stars: atmospheres -
Hertzsprung-Russell and C-M diagrams - catalogs
Abstract:
The third data release of Gaia has provided approximately 220 million
low resolution spectra. Among these, about 100000 correspond to white
dwarfs. The magnitude of this quantity of data precludes the
possibility of performing spectral analysis and type determination by
human inspection. In order to tackle this issue, we explore the
possibility of utilising a machine learning approach, based on a
Random Forest algorithm.
We aim to analyze the viability of the Random Forest algorithm for the
spectral classification of the white dwarf population within 100pc
from the Sun, based on the Hermite coefficients of Gaia spectra.
We utilized the assigned spectral type from the Montreal White Dwarf
Database for training and testing our Random Forest algorithm. Once
validated, our algorithm model is applied to the rest of unclassified
white dwarfs within 100pc. First, we started by classifying the two
major spectral type groups of white dwarfs: hydrogen-rich (DA) and
hydrogen- deficient (non-DA). Next, we explored the possibility of
classifying the various spectral subtypes, including in some cases the
secondary spectral types.
Our Random Forest classification presented a very high recall (>80%)
for DA and DB white dwarfs, and a very high precision (>90%) for DB,
DQ and DZ white dwarfs. As a result we have assigned a spectral type
to 9,446 previously unclassified white dwarfs: 4739 DAs, 76 DBs (60 of
them DBAs), 4437 DCs, 132 DZs and 62 DQs (9 of them DQpec).
Despite the low resolution of Gaia spectra, the Random Forest
algorithm applied to the Gaia spectral coefficients proves to be a
highly valuable tool for spectral classification.
Description:
This catalogue contains the predicted spectral type for all objects in
the 100pc sample, as well as their Gaia DR3 source ID, 2016 position,
position errors, parallaxes, parallax errors, colour and absolute
magnitude.
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
catalog.dat 174 12351 Catalog of predicted spectral type for all
objects in the 100pc sample (table 3)
--------------------------------------------------------------------------------
See also:
I/355 : Gaia DR3 Part 1. Main source (Gaia Collaboration, 2022)
Byte-by-byte Description of file: catalog.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 5 I5 --- Index [0/12350] Object index
7- 25 A19 --- GaiaDR3 Object Gaia DR3 source ID
27- 32 F6.1 yr refEp [2016.0] Position reference epoch
34- 53 F20.16 deg RAdeg Right ascension (ICRS) at Ep=2016.0
55- 66 F12.10 mas e_RAdeg Right ascension error
68- 87 F20.16 deg DEdeg Declination (ICRS) at Ep=2016.0
89-100 F12.10 mas e_DEdeg Declination error
102-120 F19.15 mas plx Parallax
122-132 F11.9 mas e_plx Parallax error
134-147 E14.11 mag BP-RP BP-RP colour
149-166 F18.15 mag GMAG Absolute magnitude (G)
168-174 A7 --- SPPred Predicted spectral type of object
--------------------------------------------------------------------------------
Acknowledgements:
Santiago Torres, santiago.torres(at)upc.edu
(End) Patricia Vannier [CDS] 06-Oct-2023