J/ApJS/258/26 Spectroscopic binaries from LAMOST MRS. I. (Zhang+, 2022)
The spectroscopic binaries from the LAMOST Medium-Resolution Survey.
I. Searching for double-lined spectroscopic binaries with a convolutional
neural network.
Zhang B., Jing Y.-J., Yang F., Wan J.-C., Ji X., Fu J.-N., Liu C.,
Zhang X.-B., Luo F., Tian H., Zhou Y.-T., Wang J.-X., Guo Y.-J., Zong W.,
Xiong J.-P., Li J.
<Astrophys. J. Suppl. Ser., 258, 26 (2022)>
=2022ApJS..258...26Z 2022ApJS..258...26Z
ADC_Keywords: Binaries, spectroscopic; Spectra, optical; Surveys;
Radial velocities; Effective temperatures
Keywords: Astronomy data analysis ; Close binary stars ;
Convolutional neural networks ; Sky surveys ; Spectroscopy ;
Spectroscopic binary stars
Abstract:
We developed a convolutional neural network model to distinguish the
double-lined spectroscopic binaries (SB2s) from others based on
single-exposure medium-resolution spectra (R∼7500). The training set
consists of a large set of mock spectra of single stars and binaries
synthesized based on the MIST stellar evolutionary model and ATLAS9
atmospheric model. Our model reaches a novel theoretic false-positive
rate by adding a proper penalty on the negative sample (e.g., 0.12%
and 0.16% for the blue/red arm when the penalty parameter Λ=16).
Tests show that the performance is as expected and favors FGK-type
main-sequence (MS) binaries with high mass ratio (q≥0.7) and large
radial velocity separation (Δv≥50km/s). Although the real
false-positive rate cannot be estimated reliably, validating on
eclipsing binaries identified from Kepler light curves indicates that
our model predicts low binary probabilities at eclipsing phases (0,
0.5, and 1.0) as expected. The color-magnitude diagram also helps
illustrate its feasibility and capability of identifying FGK MS
binaries from spectra. We conclude that this model is reasonably
reliable and can provide an automatic approach to identify SB2s with
period ≲10days. This work yields a catalog of binary probabilities
for over 5 million spectra of 1 million sources from the LAMOST
medium-resolution survey (MRS) and a catalog of 2198 SB2 candidates
whose physical properties will be analyzed in a follow-up paper. Data
products are made publicly available online, as well as our Github
website.
Description:
We measured the radial velocities (RVs) for all the spectra (S/N>5) in
LAMOST MRS DR8 v1.0 following the method we proposed in
Zhang+ (2021ApJS..256...14Z 2021ApJS..256...14Z). See Section 4.1.
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
table4.dat 519 5893382 The 5 million binary probabilities and RVs
obtained from the LAMOST MRS DR8 (v1.0)
--------------------------------------------------------------------------------
Description of file:
In the FTP section, datafile4.fits.gz is the original table in FITS format.
See also:
B/sb9 : SB9: 9th Catalogue of Spectroscopic Binary Orbits (Pourbaix+ 2004-2014)
I/337 : Gaia DR1 (Gaia Collaboration, 2016)
V/156 : LAMOST DR7 catalogs (Luo+, 2019)
I/355 : Gaia DR3 Part 1. Main source (Gaia Collaboration, 2022)
I/357 : Gaia DR3 Part 3. Non-single stars (Gaia Collaboration, 2022)
J/A+A/444/643 : Candidate spectroscopic binaries in SDSS (Pourbaix+, 2005)
J/A+A/465/257 : Radial velocity in multiple systems (Tokovinin+, 2007)
J/AJ/140/184 : RAVE double-lined spectroscopic binaries (Matijevic+, 2010)
J/ApJS/190/1 : A survey of stellar families (Raghavan+, 2010)
J/PASP/126/914 : Kepler eclipsing binary stars. V. (Conroy+, 2014)
J/AJ/151/68 : Kepler Mission. VII. Eclipsing binaries in DR3 (Kirk+, 2016)
J/A+A/608/A95 : GES: multi-line spectroscopic binary candidates (Merle+, 2017)
J/ApJS/228/24 : GALAH semi-automated classification scheme (Traven+, 2017)
J/ApJS/235/5 : EA-type eclipsing binaries observed by LAMOST (Qian+, 2018)
J/ApJS/245/34 : Abundances for 6 million stars from LAMOST DR5 (Xiang+, 2019)
J/AJ/158/155 : SB candidates from the RAVE & Gaia DR2 surveys (Birko+, 2019)
J/other/RAA/19.64 : LAMOST sp. binaries & variable stars (Qian+, 2019)
J/MNRAS/496/1355 : Accurate SB2 radial velocities (Halbwachs+, 2020)
J/A+A/635/A155 : Gaia-ESO Survey SB1 catalogue (Merle+, 2020)
J/A+A/638/A145 : GALAH survey. FGK binary stars (Traven+, 2020)
J/ApJ/891/23 : Stellar abundances from LAMOST MRS (SPCAnet) (Wang+, 2020)
J/ApJS/249/31 : Short period sp. & EBs (LPSEB) from LAMOST & PTF (Yang+, 2020)
J/ApJS/246/9 : Stellar parameters of LAMOST stars using SLAM (Zhang+, 2020)
J/ApJS/251/15 : LAMOST-Kepler/K2 survey (LK-MRS) first year obs. (Zong+, 2020)
J/other/RAA/21.292 : LAMOST Time-Domain survey, first results (Wang+, 2021)
Byte-by-byte Description of file: table4.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 9 I9 --- ObsID LAMOST observational ID
11- 18 I8 min LMJM Local Modified Julian Minute
20- 32 F13.5 d BJDmid Barycentric Julian Date of the middle of
exposure
34- 38 I5 d LMJD Local Modified Julian Day
(only used for spectral naming)
40- 59 A20 --- planid Plan ID
61- 62 I2 --- spid Spectrograph ID
64- 66 I3 --- fiber Fiber ID (fiberid)
68- 77 F10.6 deg RAdeg Right ascension (J2000) (ra)
79- 87 F9.6 deg DEdeg Declination (J2000) (dec)
89- 98 F10.6 --- SNRB ?=- S/N ratio of the blue arm (snr_B)
100- 109 F10.6 --- SNRR ?=- S/N ratio of the red arm (snr_R)
111- 120 F10.4 km/s RVobsB ?=- RV measured from blue arm spectra
(rvobsB)
122- 130 F9.4 km/s e_RVobsB ?=- RVobsB uncertainty (rvobserr_B)
132- 141 F10.4 km/s RVabsB ?=- RV calibrated to Gaia eDR3 (rvabsB)
143- 151 F9.4 km/s e_RVabsB ?=- Total error of absolute RV
(rverrabs_B)
153- 157 I5 K TeffB ?=- Effective temperature of the best
template (rvteffB)
159- 167 F9.6 --- CCFmaxB ?=- CCF max value (ccfmax_B)
169- 178 F10.4 km/s RVobsR ?=- RV measured from red arm spectra
(rvobsR)
180- 188 F9.4 km/s e_RVobsR ?=- RVobsR uncertainty (rvobserr_R)
190- 199 F10.4 km/s RVabsR ?=- RV calibrated to Gaia eDR3 (rvabsR)
201- 209 F9.4 km/s e_RVabsR ?=- Total error of absolute RV
(rverrabs_R)
211- 215 I5 K TeffR ?=- Effective temperature of the best
template (rvteffR)
217- 225 F9.6 --- CCFmaxR ?=- CCF max value (ccfmax_R)
227- 234 F8.6 --- PbpB8-0 ?=- PqΛ=8 with q=0 for the blue
arm (pbB8_0)
236- 243 F8.6 --- PbpB8-16 ?=- PqΛ=8 with q=16 for the blue
arm (pbB8_16)
245- 252 F8.6 --- PbpB8-50 ?=- PqΛ=8 with q=50 for the blue
arm (pbB8_50)
254- 261 F8.6 --- PbpB8-84 ?=- PqΛ=8 with q=84 for the blue
arm (pbB8_84)
263- 270 F8.6 --- PbpB8-100 ?=- PqΛ=8 with q=100 for the blue
arm (pbB8_100)
272- 279 F8.6 --- PbpB16-0 ?=- PqΛ=16 with q=0 for the blue
arm (pbB16_0)
281- 288 F8.6 --- PbpB16-16 ?=- PqΛ=16 with q=16 for the blue
arm (pbB16_16)
290- 297 F8.6 --- PbpB16-50 ?=- PqΛ=16 with q=50 for the blue
arm (pbB16_50)
299- 306 F8.6 --- PbpB16-84 ?=- PqΛ=16 with q=84 for the blue
arm (pbB16_84)
308- 315 F8.6 --- PbpB16-100 ?=- PqΛ=16 with q=100 for the
blue arm (pbB16_100)
317- 324 F8.6 --- PbpB32-0 ?=- PqΛ=32 with q=0 for the blue
arm (pbB32_0)
326- 333 F8.6 --- PbpB32-16 ?=- PqΛ=32 with q=16 for the blue
arm (pbB32_16)
335- 342 F8.6 --- PbpB32-50 ?=- PqΛ=32 with q=50 for the blue
arm (pbB32_50)
344- 351 F8.6 --- PbpB32-84 ?=- PqΛ=32 with q=84 for the blue
arm (pbB32_84)
353- 360 F8.6 --- PbpB32-100 ?=- PqΛ=32 with q=100 for the
blue arm (pbB32_100)
362 I1 --- GoodB [0/1] True for good blue spectra (S/N>5)
(pbflagB)
364- 367 I4 --- NbadB [0/9999] Number of bad pixels for the blue
arm (npixbad_B)
369- 376 F8.6 --- PbpR8-0 ?=- PqΛ=8 with q=0 for the red
arm (pbR8_0)
378- 385 F8.6 --- PbpR8-16 ?=- PqΛ=8 with q=16 for the red
arm (pbR8_16)
387- 394 F8.6 --- PbpR8-50 ?=- PqΛ=8 with q=50 for the red
arm (pbR8_50)
396- 403 F8.6 --- PbpR8-84 ?=- PqΛ=8 with q=84 for the red
arm (pbR8_84)
405- 412 F8.6 --- PbpR8-100 ?=- PqΛ=8 with q=100 for the red
arm (pbR8_100)
414- 421 F8.6 --- PbpR16-0 ?=- PqΛ=16 with q=0 for the red
arm (pbR16_0)
423- 430 F8.6 --- PbpR16-16 ?=- PqΛ=16 with q=16 for the red
arm (pbR16_16)
432- 439 F8.6 --- PbpR16-50 ?=- PqΛ=16 with q=50 for the red
arm (pbR16_50)
441- 448 F8.6 --- PbpR16-84 ?=- PqΛ=16 with q=84 for the red
arm (pbR16_84)
450- 457 F8.6 --- PbpR16-100 ?=- PqΛ=16 with q=100 for the red
arm (pbR16_100)
459- 466 F8.6 --- PbpR32-0 ?=- PqΛ=32 with q=0 for the red
arm (pbR32_0)
468- 475 F8.6 --- PbpR32-16 ?=- PqΛ=32 with q=16 for the red
arm (pbR32_16)
477- 484 F8.6 --- PbpR32-50 ?=- PqΛ=32 with q=50 for the red
arm (pbR32_50)
486- 493 F8.6 --- PbpR32-84 ?=- PqΛ=32 with q=84 for the red
arm (pbR32_84)
495- 502 F8.6 --- PbpR32-100 ?=- PqΛ=32 with q=100 for the red
arm (pbR32_100)
504 I1 --- GoodR [0/1] True for good red spectra (S/N>5)
(pbflagR)
506- 509 I4 --- NbadR [0/9999] Number of bad pixels for the red
arm (npixbad_R)
511- 519 F9.6 mag Gbp-Grp [-0.75/7.7]?=- Intrinsic color
(GBP-GRP)0 (bprp0)
--------------------------------------------------------------------------------
Acknowledgements:
Bo Zhang [bozhang at nao.cas.cn]
History:
From electronic version of the journal
(End) Emmanuelle Perret [CDS] 03-Aug-2022