J/ApJ/808/16 The Cannon: a new approach to determine abundances (Ness+, 2015)
The Cannon: a data-driven approach to stellar label determination.
Ness M., Hogg D.W., Rix H.-W., Ho A.Y.Q., Zasowski G.
<Astrophys. J., 808, 16 (2015)>
=2015ApJ...808...16N 2015ApJ...808...16N (SIMBAD/NED BibCode)
ADC_Keywords: Abundances, [Fe/H] ; Spectra, infrared
Keywords: methods: data analysis; methods: statistical; stars: abundances;
stars: fundamental parameters; surveys; techniques: spectroscopic
Abstract:
New spectroscopic surveys offer the promise of stellar parameters and
abundances ("stellar labels") for hundreds of thousands of stars; this
poses a formidable spectral modeling challenge. In many cases, there
is a subset of reference objects for which the stellar labels are
known with high(er) fidelity. We take advantage of this with The
Cannon, a new data-driven approach for determining stellar labels from
spectroscopic data. The Cannon learns from the "known" labels of
reference stars how the continuum-normalized spectra depend on these
labels by fitting a flexible model at each wavelength; then, The
Cannon uses this model to derive labels for the remaining survey
stars. We illustrate The Cannon by training the model on only 542
stars in 19 clusters as reference objects, with Teff, logg, and [Fe/H]
as the labels, and then applying it to the spectra of 55000 stars from
APOGEE DR10. The Cannon is very accurate. Its stellar labels compare
well to the stars for which APOGEE pipeline (ASPCAP) labels are
provided in DR10, with rms differences that are basically identical to
the stated ASPCAP uncertainties. Beyond the reference labels, The
Cannon makes no use of stellar models nor any line-list, but needs a
set of reference objects that span label-space. The Cannon performs
well at lower signal-to-noise, as it delivers comparably good labels
even at one-ninth the APOGEE observing time. We discuss the
limitations of The Cannon and its future potential, particularly, to
bring different spectroscopic surveys onto a consistent scale of
stellar labels.
Description:
Here, we use the Apache Point Observatory Galactic Evolution
Experiment (APOGEE) DR10 data (S. R. Majewski et al.
2015arXiv150905420M 2015arXiv150905420M) to illustrate and showcase The Cannon. APOGEE,
part of the SDSS-III, is a high resolution (R∼22500), high signal to
noise (S/N∼100), H-band (15200-16900Å) spectroscopic survey of
primarily red giant stars spanning the bulge, disk, and halo of the
Milky Way. We use here spectra from the set of 55000 stars that were
released as part of the SDSS Data Release 10 (DR10) (Ahn et al.
2014ApJS..211...17A 2014ApJS..211...17A).
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
table1.dat 136 55007 Table of stellar labels
--------------------------------------------------------------------------------
See also:
III/272 : RAVE 4th data release (Kordopatis+, 2013)
II/246 : 2MASS All-Sky Catalog of Point Sources (Cutri+ 2003)
J/ApJ/794/125 : IN-SYNC. I. APOGEE stellar parameters (Cottaar+, 2014)
J/AJ/148/54 : The Hypatia Catalog ((Hinkel+, 2014)
J/A+A/564/A133 : Gaia FGK benchmark stars: metallicity (Jofre+, 2014)
J/AJ/146/133 : Stellar parameters from SDSS-III APOGEE DR10 (Meszaros+, 2013)
J/AJ/142/193 : RAVE stellar elemental abundances (DR1) (Boeche+, 2011)
J/A+A/531/A165 : MILES atmospheric parameters (Prugniel+, 2011)
J/A+A/525/A71 : Atmospheric parameters for 1273 stars (Wu+, 2011)
J/A+AS/141/371 : Low-mass stars evolutionary tracks&isochrones (Girardi+, 2000)
J/ApJ/527/879 : Gravities for late-type stars (Allende Prieto+, 1999)
J/A+AS/133/221 : Library of ELODIE spectra (F5-K7 stars) (Soubiran+ 1998)
http://www.sdss3.org/ : SDSS-III home page
Byte-by-byte Description of file: table1.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 14 A14 --- --- [apogee.n.s.s3.]
15- 18 I4 --- AP ? Apoge version (4412 or 4413)
19 A1 --- --- [.]
20- 37 A18 --- ID Steller identifier (1)
39- 51 F13.8 K Teff [3140/5722] Effective temperature
53- 57 F5.2 [cm/s2] log(g) [-1.4/5.2] Log surface gravity
59- 63 F5.2 [-] [Fe/H] [-3/3.4] Metallicity
65- 70 F6.2 K e_Teff [1/285] The 1σ uncertainty in Teff
72- 76 F5.3 [cm/s2] e_log(g) [0.003/0.6] The 1σ uncertainty in log(g)
78- 82 F5.3 [-] e_[Fe/H] [0.001/0.4] The 1σ uncertainty in [Fe/H]
84- 89 F6.2 --- Chi2 [0.01/810] The χ2 value
91- 98 F8.1 km/s Vscat [0/522068] Velocity scatter
100-110 I11 --- TFlag The APOGEE_TARGET2 flag
111-112 A2 --- --- [.0]
114-124 I11 --- AFlag The APOGEE_ASPCAPFLAG flag (2)
125-126 A2 --- --- [.0]
128 A1 --- CFlag [CS] Commissioning flag (3)
130-134 F5.3 --- Dref [0.002/1] Distance between test star and
nearest reference object (4)
136 I1 --- EFlag [0/1] Extrapolation flag (5)
--------------------------------------------------------------------------------
Note (1): "HHMMSSss+DDMMSSs" for 2MASS (J2000) or
"APHHMMSSss+DDMMSSs" for APOGEE
Note (2): http://www.sdss3.org/dr10/algorithms/bitmask_apogee_aspcapflag.php
has documentation details for the APOGEE_ASPCAPFLAG flag.
ASPCAP = APOGEE Stellar Parameters and Chemical Abundances Pipeline
Note (3): We also apply The Cannon to the commissioning data and caution the
reader about the fidelity of these results, as the Line Spread
Function (LSF) of the commissioning data are different from the main
survey and consequently different from the reference dataset of stars
in the training step. Those objects are flagged as commissioning data.
Note (4): Normalised to the maximum distance, so values lie between 0 and 1
(see Equation 9).
Note (5): Indicates if the test star lies outside of the label-space of the
reference objects (set to 1 if so).
--------------------------------------------------------------------------------
History:
From electronic version of the journal
(End) Prepared by [AAS], Emmanuelle Perret [CDS] 12-Nov-2015