J/AJ/158/147 Spectrophotometric parallaxes with linear models (Hogg+, 2019)
Spectrophotometric parallaxes with linear models: accurate distances
for luminous red-giant stars.
Hogg D.W., Eilers A.-C., Rix H.-W.
<Astron. J., 158, 147-147 (2019)>
=2019AJ....158..147H 2019AJ....158..147H (SIMBAD/NED BibCode)
ADC_Keywords: Milky Way ; Stars, giant ; Parallaxes, trigonometric ; Models ;
Parallaxes, spectroscopic
Keywords: catalogs - Galaxy: disk - methods: statistical - stars: distances -
surveys - techniques: spectroscopic
Abstract:
With contemporary infrared spectroscopic surveys like APO Galactic
Evolution Experiment (APOGEE), red-giant stars can be observed to
distances and extinctions at which Gaia parallaxes are not highly
informative. Yet the combination of effective temperature, surface gravity,
composition, and age-all accessible through spectroscopy - determines a
giant's luminosity. Therefore spectroscopy plus photometry should enable
precise spectrophotometric distance estimates. Here we use the overlap
of APOGEE, Gaia, the Two Micron All Sky Survey (2MASS), and the Wide-field
Infrared Survey Explorer (WISE) to train a data-driven model to predict
parallaxes for red-giant branch stars with 0<logg=<2.2 (more luminous than
the red clump). We employ (the exponentiation of) a linear function of
APOGEE spectral pixel intensities and multiband photometry to predict
parallax spectrophotometrically. The model training involves no logarithms
or inverses of the Gaia parallaxes, and needs no cut on the Gaia parallax
signal-to-noise ratio. It includes an L1 regularization to zero out
the contributions of uninformative pixels. The training is performed with
leave-out subsamples such that no star's astrometry is used even indirectly
in its spectrophotometric parallax estimate. The model implicitly performs
a reddening and extinction correction in its parallax prediction, without
any explicit dust model. We assign to each star in the sample a new
spectrophotometric parallax estimate; these parallaxes have uncertainties
of less than 15%, depending on data quality, which is more precise than
the Gaia parallax for the vast majority of targets, and certainly any stars
more than a few kiloparsec distance. We obtain 10% distance estimates
out to heliocentric distances of 20 kpc, and make global maps of the
Milky Way's disk.
Description:
We take the APOGEE (Allende Prieto et al. 2008AN....329.1018A 2008AN....329.1018A;
Wilson et al. 2010SPIE.7735E..1CW; Majewski et al. 2017AJ....154...94M 2017AJ....154...94M)
spectral data from SDSS-IV (Blanton et al. 2017AJ....154...28B 2017AJ....154...28B) DR14
(Abolfathi et al. 2018ApJS..235...42A 2018ApJS..235...42A). Because we want to make a purely
linear model, which has very little capacity, we restrict our consideration
to a small region in stellar parameter space. We cut the APOGEE data down
to the range of 0<logg<2.2, which isolates stars that are more luminous
than the red clump. The APOGEE pipeline (Garcia Perez et al. 2016,
J/AJ/151/144) values of surface gravity logg (which we use) have
uncertainties but this cut leads to a clean sample of luminous red giants,
and it is a cut that is only on spectral properties of the stars (and
not photometry nor astrometry).
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
table1.dat 51 44784 The generated spectrophotometric parallaxes
and their uncertainties
--------------------------------------------------------------------------------
See also:
II/246 : 2MASS All-Sky Catalog of Point Sources (Cutri+ 2003)
VII/233 : The 2MASS Extended sources (IPAC/UMass, 2003-2006)
I/337 : Gaia DR1 (Gaia Collaboration, 2016)
I/345 : Gaia DR2 (Gaia Collaboration, 2018)
I/347 : Distances to 1.33 billion stars in Gaia DR2 (Bailer-Jones+, 2018)
J/AJ/151/144 : ASPCAP weights for the 15 APOGEE chemical elements
(Garcia+, 2016)
Byte-by-byte Description of file: table1.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 18 A18 --- 2MASS 2MASS identifier (2MHHMMSSss+DDMMSSs) (2MASS_ID)
20- 26 F7.4 mas Plx [-4.8958/14.4807]? Gaia parallax (Gaia_parallax)
28- 33 F6.4 mas e_Plx [0.0106/1.022]? Uncertainty in Plx
(Gaiaparallaxerr)
35- 40 F6.4 mas Plxsp [0.0153/2.0165] Spectrophotometric parallax
estimate (spec_parallax)
42- 47 F6.4 mas e_Plxsp [0.0015/5.2148] Uncertainty in Plxsp
(specparallaxerr)
49 I1 --- Set [0/1]? Training set (training_set) (1)
51 A1 --- Sample [AB] Sample (sample) (2)
--------------------------------------------------------------------------------
Note (1): Training set as follows:
1 = Parent sample and training set star;
0 = Parent sample star.
Note (2): Every parent sample star gets a randomly assigned binary label (A or
B). This is used for two-fold validation and jackknife. In short, we will use
the model trained on the training set A data to assign spectrophotometric
parallaxes to the parent sample B data and use the model trained on the
training set B data to assign spectrophotometric parallaxes to the parent
sample A data.
--------------------------------------------------------------------------------
History:
From electronic version of the journal
(End) Tiphaine Pouvreau [CDS] 21-Nov-2019