Improving Earth-like planet detection in radial velocity using deep learning
arXiv:2405.13247 · doi:10.1051/0004-6361/202450022
Abstract
Many novel methods have been proposed to mitigate stellar activity for exoplanet detection as the presence of stellar activity in radial velocity (RV) measurements is the current major limitation. Unlike traditional methods that model stellar activity in the RV domain, more methods are moving in the direction of disentangling stellar activity at the spectral level. The goal of this paper is to present a novel convolutional neural network-based algorithm that efficiently models stellar activity signals at the spectral level, enhancing the detection of Earth-like planets. We trained a convolutional neural network to build the correlation between the change in the spectral line profile and the corresponding RV, full width at half maximum (FWHM) and bisector span (BIS) values derived from the classical cross-correlation function. This algorithm has been tested on three intensively observed stars: Alpha Centauri B (HD128621), Tau ceti (HD10700), and the Sun. By injecting simulated planetary signals at the spectral level, we demonstrate that our machine learning algorithm can achieve, for HD128621 and HD10700, a detection threshold of 0.5 m/s in semi-amplitude for planets with periods ranging from 10 to 300 days. This threshold would correspond to the detection of a 4 in the habitable zone of those stars. On the HARPS-N solar dataset, our algorithm is even more efficient at mitigating stellar activity signals and can reach a threshold of 0.2 m/s, which would correspond to a 2.2 planet on the orbit of the Earth. To the best of our knowledge, it is the first time that such low detection thresholds are reported for the Sun, but also for other stars, and therefore this highlights the efficiency of our convolutional neural network-based algorithm at mitigating stellar activity in RV measurements.
Accepted for publication in A&A
References in corpus (42)
- The PLATO 2.0 Mission
- The CORALIE survey for southern extra-solar planets VII. Two short-period Saturnian companions to HD108147 and HD168746
- A Gaussian process framework for modelling stellar activity signals in radial velocity data
- Identifying Exoplanets with Deep Learning: A Five Planet Resonant Chain around Kepler-80 and an Eighth Planet around Kepler-90
- Using the Sun to estimate Earth-like planets detection capabilities II. Impact of plages
- Planetary detection limits taking into account stellar noise. I. Observational strategies to reduce stellar oscillation and granulation effects
- SOAP 2.0: A tool to estimate the photometric and radial velocity variations induced by stellar spots and plages
- Measuring precise radial velocities on individual spectral lines. I. Validation of the method and application to mitigate stellar activity
- The Sun as a planet-host star: Proxies from SDO images for HARPS radial-velocity variations
- Three years of Sun-as-a-star radial-velocity observations on the approach to solar minimum
- Three Years of HARPS-N High-Resolution Spectroscopy and Precise Radial Velocity Data for the Sun
- HARPS-N observes the Sun as a star
- Measuring precise radial velocities on individual spectral lines : II. Dependance of stellar activity signal on line depth
- The IAG solar flux atlas: Accurate wavelengths and absolute convective blueshift in standard solar spectra
- pyaneti II: A multidimensional Gaussian process approach to analysing spectroscopic time-series
- Separating planetary reflex Doppler shifts from stellar variability in the wavelength domain
- YARARA: Significant improvement of RV precision through post-processing of spectral time-series
- Identifying Exoplanets with Deep Learning II: Two New Super-Earths Uncovered by a Neural Network in K2 Data
- Variability of stellar granulation and convective blueshift with spectral type and magnetic activity. I. K and G main sequence stars
- 'Ultimate' Information Content in Solar and Stellar Spectra: Photospheric line asymmetries and wavelength shifts
- RASSINE: Interactive tool for normalising stellar spectra I. Description and performance of the code
- Insights on the Spectral Signatures of Stellar Activity and Planets from PCA
- Unsigned magnetic flux as a proxy for radial-velocity variations in Sun-like stars
- Overfitting Affects the Reliability of Radial Velocity Mass Estimates of the V1298 Tau Planets
- Efficient modeling of correlated noise. III. Scalable methods for jointly modeling several observables' time series with Gaussian processes
- The Changing Face of Centauri B: Probing plage and stellar activity in K-dwarfs
- Measuring precise radial velocities on individual spectral lines. III. Dependence of stellar activity signal on line formation temperature
- Quasi-periodic Gaussian Processes for stellar activity: from physical to kernel parameters
- Convective blueshift strengths of 810 F to M solar-type stars
- YARARA V2: Reaching sub m/s precision over a decade using PCA on line-by-line RVs
- Stellar activity correction using PCA decomposition of shells
- Stellar signal components seen in HARPS and HARPS-N solar radial velocities
- Detection Limits of Low-mass, Long-period Exoplanets Using Gaussian Processes Applied to HARPS-N Solar RVs
- Auto-correlation functions of astrophysical processes, and their relation to Gaussian processes; Application to radial velocities of different starspot configurations
- Deriving stellar inclination of slow rotators using stellar activity
- SOAP-GPU: Efficient Spectral Modelling of Stellar Activity Using Graphical Processing Units
- : a data-driven approach to correct for systematics in RV data -- Application to SPIRou data of the planet-hosting M dwarf GJ 251
- A Machine Learning approach for correcting radial velocities using physical observables
- USuRPER: Unit-Sphere Representation PERiodogram for full spectra
- Spectral Line Depth Variability in Radial Velocity Spectra
- New Periodograms Separating Orbital Radial Velocities and Spectral Shape Variation
- ExoplANNET: A deep learning algorithm to detect and identify planetary signals in radial velocity data