J/AJ/164/49 Identifying Exoplanets with Deep Learning. IV. (de Beurs+, 2022)
Identifying Exoplanets with Deep Learning.
IV. Removing Stellar Activity Signals from Radial Velocity Measurements Using
Neural Networks.
de Beurs Z.L., Vanderburg A., Shallue C.J., Dumusque X., Cameron A.C.,
Leet C., Buchhave L.A., Cosentino R., Ghedina A., Haywood R.D.,
Langellier N., Latham D.W., Lopez-Morales M., Mayor M., Micela G.,
Milbourne T.W., Mortier A., Molinari E., Pepe F., Phillips D.F.,
Pinamonti M., Piotto G., Rice K., Sasselov D., Sozzetti A., Udry S.,
Watson C.A.
<Astron. J., 164, 49 (2022)>
=2022AJ....164...49D 2022AJ....164...49D
ADC_Keywords: Solar system; Sun; Spectra, optical; Radial velocities
Keywords: Exoplanet astronomy ; Radial velocity ; Convolutional neural
networks
Abstract:
Exoplanet detection with precise radial velocity (RV) observations is
currently limited by spurious RV signals introduced by stellar
activity. We show that machine-learning techniques such as linear
regression and neural networks can effectively remove the activity
signals (due to starspots/faculae) from RV observations. Previous
efforts focused on carefully filtering out activity signals in time
using modeling techniques like Gaussian process regression. Instead,
we systematically remove activity signals using only changes to the
average shape of spectral lines, and use no timing information. We
trained our machine-learning models on both simulated data (generated
with the SOAP 2.0 software) and observations of the Sun from the
HARPS-N Solar Telescope. We find that these techniques can predict and
remove stellar activity both from simulated data (improving RV scatter
from 82 to 3cm/s) and from more than 600 real observations taken
nearly daily over 3yr with the HARPS-N Solar Telescope (improving the
RV scatter from 1.753 to 1.039m/s, a factor of ∼1.7 improvement). In
the future, these or similar techniques could remove activity signals
from observations of stars outside our solar system and eventually
help detect habitable-zone Earth-mass exoplanets around Sun-like
stars.
Description:
The HARPS-N solar data set consists of 528 days of solar observations
from the HARPS-N Solar Telescope spanning 3yr (2015 July-2018 July).
Commissioned at the Telescopio Nazionale Galileo (TNG), the HARPS-N
spectrograph is a vacuum-enclosed cross-dispersed echelle spectrograph
that has temperature and pressure stabilization. HARPS-N spans the
wavelength range from 383 to 693nm and has a resolving power of
R=115000.
File Summary:
--------------------------------------------------------------------------------
FileName Lrecl Records Explanations
--------------------------------------------------------------------------------
ReadMe 80 . This file
fig10.dat 62 528 HARPS-N solar telescope raw and model predicted
radial velocity over time
--------------------------------------------------------------------------------
See also:
J/A+A/474/293 : Radial velocities of GJ 674 (Bonfils+, 2007)
J/A+A/529/A75 : Limb-darkening coefficients (Claret+, 2011)
J/A+A/535/A55 : 4 stars with long-period planets (Dumusque+, 2011)
J/A+A/614/A133 : Radial velocities of HD215152 (Delisle+, 2018)
J/MNRAS/476/3661 : Morphology of SDSS galaxies (Dominguez+, 2018)
J/AJ/157/169 : Identifying exoplanets deep learning in K2 (Dattilo+, 2019)
J/ApJ/874/107 : HARPSN solar radial velocities & activity (Milbourne+, 2019)
J/AJ/158/25 : Automated triage and vetting of TESS candidates (Yu+, 2019)
J/A+A/642/A133 : V830 Tau VI light curves and RV curves (Damasso+, 2020)
J/A+A/633/A53 : TESS planet candidates classification (Osborn+, 2020)
J/A+A/639/A77 : Proxima Cen RV, FWHM and fluxes (Suarez Mascareno+, 2020)
J/A+A/648/A103 : Thorium line list (Dumusque+, 2021)
J/AJ/163/171 : The EXPRES Stellar Signals Project. II. RVel (Zhao+, 2022)
Byte-by-byte Description of file: fig10.dat
--------------------------------------------------------------------------------
Bytes Format Units Label Explanations
--------------------------------------------------------------------------------
1- 20 F20.17 m/s RVelraw [-4.26/5.6] Raw radial velocity from HARPS-N (1)
22- 42 F21.18 m/s RVelp [-4.29/5.57] Model-predicted stellar activity (1)
44- 53 F10.2 d BJD [2457223/2458316] Barycentric Julian date of
observation
55- 62 F8.2 d BJDo [57223/58316] Offset BJD-2400000 (as plotted)
--------------------------------------------------------------------------------
Note (1): The stellar activity corrected RVel plotted in the third panel of
Figure 10 are obtained by subtracting the PreRV from RawRV.
--------------------------------------------------------------------------------
History:
From electronic version of the journal
References:
Shallue et al. Paper I : 2018AJ....155...94S 2018AJ....155...94S
Dattilo et al. Paper II : 2019AJ....157..169D 2019AJ....157..169D Cat. J/AJ/157/169
Yu et al. Paper III : 2019AJ....158...25Y 2019AJ....158...25Y Cat. J/AJ/158/25
(End) Prepared by [AAS], Coralie Fix [CDS], 25-Oct-2022