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 (e.g. Haywood et al. 2014). Instead, we systematically remove
activity signals using only changes to the average shape of spectral lines, and
no information about when the observations were collected. We trained our
machine learning models on both simulated data (generated with the SOAP 2.0
software; Dumusque et al. 2014) and observations of the Sun from the HARPS-N
Solar Telescope (Dumusque et al. 2015; Phillips et al. 2016; Collier Cameron et
al. 2019). We find that these techniques can predict and remove stellar
activity from both simulated data (improving RV scatter from 82 cm/s to 3 cm/s)
and from more than 600 real observations taken nearly daily over three years
with the HARPS-N Solar Telescope (improving the RV scatter from 1.47 m/s to
0.78 m/s, a factor of ~ 1.9 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
Identifying Exoplanets with Deep Learning. IV. Removing Stellar Activity Signals from Radial Velocity Measurements Using Neural Networks
%0 Generic
%1 debeurs2020identifying
%A de Beurs, Zoe L.
%A Vanderburg, Andrew
%A Shallue, Christopher J.
%A Dumusque, Xavier
%A Cameron, Andrew Collier
%A Buchhave, Lars A.
%A Cosentino, Rosario
%A Ghedina, Adriano
%A Haywood, Raphaëlle D.
%A Langellier, Nicholas
%A Latham, David W.
%A López-Morales, Mercedes
%A Mayor, Michel
%A Micela, Giusi
%A Milbourne, Timothy W.
%A Mortier, Annelies
%A Molinari, Emilio
%A Pepe, Francesco
%A Phillips, David F.
%A Pinamonti, Matteo
%A Piotto, Giampaolo
%A Rice, Ken
%A Sasselov, Dimitar
%A Sozzetti, Alessandro
%A Udry, Stéphane
%A Watson, Christopher A.
%D 2020
%K activity exoplanet techniques
%T Identifying Exoplanets with Deep Learning. IV. Removing Stellar Activity
Signals from Radial Velocity Measurements Using Neural Networks
%U http://arxiv.org/abs/2011.00003
%X 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 (e.g. Haywood et al. 2014). Instead, we systematically remove
activity signals using only changes to the average shape of spectral lines, and
no information about when the observations were collected. We trained our
machine learning models on both simulated data (generated with the SOAP 2.0
software; Dumusque et al. 2014) and observations of the Sun from the HARPS-N
Solar Telescope (Dumusque et al. 2015; Phillips et al. 2016; Collier Cameron et
al. 2019). We find that these techniques can predict and remove stellar
activity from both simulated data (improving RV scatter from 82 cm/s to 3 cm/s)
and from more than 600 real observations taken nearly daily over three years
with the HARPS-N Solar Telescope (improving the RV scatter from 1.47 m/s to
0.78 m/s, a factor of ~ 1.9 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.
@misc{debeurs2020identifying,
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 (e.g. Haywood et al. 2014). Instead, we systematically remove
activity signals using only changes to the average shape of spectral lines, and
no information about when the observations were collected. We trained our
machine learning models on both simulated data (generated with the SOAP 2.0
software; Dumusque et al. 2014) and observations of the Sun from the HARPS-N
Solar Telescope (Dumusque et al. 2015; Phillips et al. 2016; Collier Cameron et
al. 2019). We find that these techniques can predict and remove stellar
activity from both simulated data (improving RV scatter from 82 cm/s to 3 cm/s)
and from more than 600 real observations taken nearly daily over three years
with the HARPS-N Solar Telescope (improving the RV scatter from 1.47 m/s to
0.78 m/s, a factor of ~ 1.9 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.},
added-at = {2020-11-03T14:42:56.000+0100},
author = {de Beurs, Zoe L. and Vanderburg, Andrew and Shallue, Christopher J. and Dumusque, Xavier and Cameron, Andrew Collier and Buchhave, Lars A. and Cosentino, Rosario and Ghedina, Adriano and Haywood, Raphaëlle D. and Langellier, Nicholas and Latham, David W. and López-Morales, Mercedes and Mayor, Michel and Micela, Giusi and Milbourne, Timothy W. and Mortier, Annelies and Molinari, Emilio and Pepe, Francesco and Phillips, David F. and Pinamonti, Matteo and Piotto, Giampaolo and Rice, Ken and Sasselov, Dimitar and Sozzetti, Alessandro and Udry, Stéphane and Watson, Christopher A.},
biburl = {https://www.bibsonomy.org/bibtex/29a02da3c6d27069c274147ebd909eecb/superjenwinters},
description = {Identifying Exoplanets with Deep Learning. IV. Removing Stellar Activity Signals from Radial Velocity Measurements Using Neural Networks},
interhash = {ef1134e6346c2ea398a033b06cb1351a},
intrahash = {9a02da3c6d27069c274147ebd909eecb},
keywords = {activity exoplanet techniques},
note = {cite arxiv:2011.00003Comment: 26 pages, 12 figures, Submitted to the Astronomical Journal},
timestamp = {2020-11-03T14:42:56.000+0100},
title = {Identifying Exoplanets with Deep Learning. IV. Removing Stellar Activity
Signals from Radial Velocity Measurements Using Neural Networks},
url = {http://arxiv.org/abs/2011.00003},
year = 2020
}