J/MNRAS/508/4202 Improved and new RVs for the GALAH+ survey (Zwitter+, 2021)
The GALAH+ survey: a new library of observed stellar spectra improves radial
velocities and hints at motions within M67.
Zwitter T., Kos J., Buder S., Cotar K., Asplund M., Bland-hawthorn J.,
Casey A.R., De Silva G.M., D'orazi V., Freeman K.C., Hayden M.R.,
Lewis G.F., Lin J., Lind K., Martell S.L., Schlesinger K.J., Sharma S.,
Simpson J.D., Stello D., Zucker D.B., Beeson K.L., De Grijs R.,
Nordlander T., Ting Y.-S., Traven G., Vogrincic R., Watson F.,
Wittenmyer R.
<Mon. Not. R. Astron. Soc. 508, 4202-4215 (2021)>
=2021MNRAS.508.4202Z 2021MNRAS.508.4202Z (SIMBAD/NED BibCode)
ADC_Keywords: Surveys ; Stars, standard ; Clusters, open ; Optical ;
Spectroscopy ; Radial velocities
Keywords: methods: data analysis - surveys - stars: fundamental parameters -
stars: kinematics and dynamics - Galaxy: kinematics and dynamics -
open clusters and associations: individual: M67
Abstract:
GALAH+ is a magnitude-limited survey of high-resolution stellar
spectra obtained by the HERMES spectrograph at the Australian
Astronomical Observatory. Its third data release provides reduced
spectra with new derivations of stellar parameters and abundances of
30 chemical elements for 584015 dwarfs and giants, 88 per cent of them
in the Gaia magnitude range 11 < G < 14. Here, we use these improved
values of stellar parameters to build a library of observed spectra
which is useful to study variations of individual spectral lines with
stellar parameters. This and other improvements are used to derive
radial velocities with uncertainties which are generally within
0.1 km/s or ∼25 per cent smaller than in the previous release.
Median differences in radial velocities measured here and by the Gaia DR2
or APOGEE DR16 surveys are smaller than 30 m/s, a larger offset is
present only for Gaia measurements of giant stars. We identify 4483
stars with intrinsically variable velocities and 225 stars for which
the velocity stays constant over => 3 visits spanning more than a
year. The combination of radial velocities from GALAH+ with distances
and sky plane motions from Gaia enables studies of dynamics within
streams and clusters. For example, we estimate that the open cluster
M67 has a total mass of ∼3300 M☉ and its outer parts seem to be
expanding, though astrometry with a larger time-span than currently
available from Gaia eDR3 is needed to judge if the latter result is
real.
Description:
Here, we describe the derivation of RVs with uncertainties typically
smaller than 0.1 km/s though for a dozen-times smaller set than
derived by Gaia DR2 (Gaia Collaboration 2018A&A...616A...1G 2018A&A...616A...1G, Cat. I/345).
The cornerstone are new values of effective temperature,
surface gravity, metallicity, and α-enhancement for 584015
spectra from the third data release of the GALAH+ survey (Buder et al.
2021MNRAS.506..150B 2021MNRAS.506..150B, Cat. J/MNRAS/506/150) which presents also an
unprecedented set of measurements of abundances of 30 chemical
elements ([X/Fe]) for the same stars. Derivation of accurate RVs
builds on a procedure described earlier (Zwitter et al.
2018MNRAS.481..645Z 2018MNRAS.481..645Z), but better parameter values and a number of
procedure improvements now make the uncertainties ∼ 25 per cent
smaller, and allow RVs to be derived for 72 per cent more spectra.
The GALAH+ includes data from ambitious stellar spectroscopic surveys
which use the HERMES spectrograph, we use data reduction pipeline
described in Kos et al. (2018A&A...616A...1G 2018A&A...616A...1G) which gives results of
694459 spectra collected between 2013 November 16 and 2019 February 25
are considered, but with additional requirements on their physical
characterization as reported in the third GALAH+ data release. After
applying criteria on stellar parameters, our final selection contains
579653 spectra for which RVs are determined (i.e see the section 2
Observational data and their reductions).
RVs computations based on observed and synthetic spectra are explained
in the section 3 Library of observed spectral templates. As a first
step we compute RV shifts of each observed spectrum versus the
relevant observed median spectrum. This is done as in Zwitter et al.
2018MNRAS.481..645Z 2018MNRAS.481..645Z, using an iterative process of computing a
weighted average of 20 wavelength intervals along the four
spectrograph arms. Next, we need to compute the RV shift between the
observed median spectrum and a suitable synthetic spectral library. We
use the one of Chiavassa et al. (2018A&A...611A..11C 2018A&A...611A..11C, Cat. J/A+A/611/A11),
which includes 3D convective motions within the stellar atmosphere. It
has been computed using the radiative transfer code Optim3D (Chiavassa
et al. 2009A&A...506.1351C 2009A&A...506.1351C) for the STAGGER grid of 3D radiative
hydrodynamical simulations of stellar convection (Magic et al.
2013A&A...557A..26M 2013A&A...557A..26M, Cat. J/A+A/557/A26). Comparison of observed
median spectra to synthetic ones can yield more than one RV
measurement. The table1.dat reports results of the RV measurement
pipeline, (i.e see the section 4 RV measurement pipeline).
Hereafter in the section 5 Objects with variable and with constant
RVs, we distinguished stars with variable RVs to those with constant
RVs. To do so, we first use a 4σ type of criterion for the
detection of RV variability with the probability P in the equations
(1) and (2), hence P > 0.9999366575, we find 4483 stars with variable
RVs. So intrinsically variable stars are listed in the table2.dat. For
stars having a constant RV which can be used as RV standards, we list
in the table3.dat, 225 objects with at least three observations which
span more than a year in time and less than 0.2 km/s in their
individual RV measurements without gravitational correction.
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
table1.dat 111 579653 *All RVs presented in this paper forming a value
added catalogue of RVs of GALAH+ DR3
table2.dat 89 4483 *List of 4483 stars with variable RV that exceed
a 100 km/s
table3.dat 83 225 *List of 225 stars with constant RVs with at
least N => 3 observations that span
{Delat}t => 1yr in time and ΔRV < 0.2 km/s
--------------------------------------------------------------------------------
Note on table1.dat and table2.dat and table3.dat: We note that light suffers
from gravitational redshift as it travels from the stellar surface to the
observer. This effect is substantial, it reaches 0.636 km/s for a solar type
star and is proportional to the ratio of the stellar mass and radius. So stars
of different types would show inconsistent RVs if this effect was neglected.
The implementation of gravitation redshift is the same as described in
Zwitter et al. (2018MNRAS.481..645Z 2018MNRAS.481..645Z), but benefits from better values of the
stellar parameters. Still, one should note that the radius of the star is
difficult to determine accurately. So the final velocities, which take
gravitational redshift into account, are internally consistent but have
substantially larger uncertainties than without taking gravitational redshift
into account. So one should use values corrected for gravitational redshift
if different types of stars are to be compared, such as within a stellar
cluster or in studies of Galactic dynamics. But if the goal is to study RV
variability of a certain star, the values without gravitational redshift
correction are preferred because of the more realistic uncertainties,
(i.e see more details in the section 4 RV measurement pipeline).
--------------------------------------------------------------------------------
See also:
I/345 : Gaia DR2 (Gaia Collaboration, 2018)
J/MNRAS/506/150 : The GALAH+ Survey DR3 (Buder+, 2021)
J/A+A/611/A11 : STAGGER-grid of 3D stellar models. V. (Chiavassa+, 2018)
J/A+A/557/A26 : STAGGER-grid of 3D stellar models. I. (Magic+, 2013)
III/284 : APOGEE-2 data from DR16 (Johnsson+, 2020)
Byte-by-byte Description of file: table1.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 15 I15 --- GALAH GALAH identifier (sobject_id)
17- 32 A16 --- 2MASS 2MASS identifier (star_id)
34- 52 I19 --- GaiaDR2 ? GaiaDR2 source_id (dr2source_id)
55- 62 F8.3 km/s RV ?=- Radial velocity including the
gravitational redshift (RV)
64- 68 F5.3 km/s e_RV ?=- Mean uncertainty of RV (e_RV)
70- 77 F8.3 km/s RVnogr ?=- Radial velocity without the gravitational
redshift correction (RVnogr)
79- 83 F5.3 km/s e_RVnogr ?=- Mean uncertainty of RVnogr (e_RVnogr)
85- 97 F13.7 d MJD ?=- The local modified Julian date (MJD)
99-111 F13.5 d JD ?=- The heliocentric Julian date (JD)
--------------------------------------------------------------------------------
Byte-by-byte Description of file: table2.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 16 A16 --- 2MASS 2MASS identifier (star_id)
18- 36 A19 --- GaiaDR2 GaiaDR2 source_id (dr2source_id)
38 I1 --- N The number of observations (N)
40- 46 F7.3 km/s |RV2-RV1| The highest RV differences between two
measurements without gravitational redshift
correction (|RV2 - RV1|)
48- 57 F10.5 d MJD2-MJD1 The highest MJD difference between two
measurements (t2 - t1)
59- 73 I15 --- GALAH1 GALAH identifier of the first measurement
(sobjectid1)
75- 89 I15 --- GALAH2 GALAH identifier of the second measurement
(sobjectid2)
--------------------------------------------------------------------------------
Byte-by-byte Description of file: table3.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 16 A16 --- 2MASS 2MASS identifier (star_id)
18- 36 I19 --- GaiaDR2 GaiaDR2 source_id (dr2source_id)
38 I1 --- N The number of observations (N)
40- 44 F5.3 km/s |RV2-RV1| The highest RV differences between two
measurements without gravitational redshift
correction (ΔRV)
46- 55 F10.5 d MJD2-MJD1 The highest MJD difference between two
measurements (Δt)
57- 63 F7.3 km/s RV The weighted average RV including the
gravitational redshift correction (RV)
65- 69 F5.3 km/s e_RV Mean uncertainty of RV (e_RV)
71- 77 F7.3 km/s RVnogr The weighted average RV without the
gravitational redshift correction (RVnogr)
79- 83 F5.3 km/s e_RVnogr Mean uncertainty of RVnogr (e_RVnogr)
--------------------------------------------------------------------------------
History:
From electronic version of the journal
(End) Luc Trabelsi [CDS] 22-Aug-2024