Recovery of TESS Stellar Rotation Periods Using Deep Learning

Abstract

We used a convolutional neural network to infer stellar rotation periods from a set of synthetic light curves simulated with realistic spot evolution patterns. We convolved these simulated light curves with real TESS light curves containing minimal intrinsic astrophysical variability to allow the network to learn TESS systematics and estimate rotation periods despite them. In addition to periods, we predict uncertainties via heteroskedastic regression to estimate the credibility of the period predictions. In the most credible half of the test data, we recover 10%-accurate periods for 46% of the targets, and 20%-accurate periods for 69% of the targets. Using our trained network, we successfully recover periods of real stars with literature rotation measurements, even past the 13.7-day limit generally encountered by TESS rotation searches using conventional period-finding techniques. Our method also demonstrates resistance to half-period aliases. We present the neural network and simulated training data, and introduce the software butterpy used to synthesize the light curves using realistic star spot evolution.

18 pages, 7 figures, 3 tables. Submitted to AAS Journals. For a brief video explaining this paper, see https://youtu.be/JNmMzHshyY0 . The code to simulate star spot evolution and light curves, butterpy, is available at https://github.com/zclaytor/butterpy

References in corpus (21)

• Array Programming with NumPy
• The Transiting Exoplanet Survey Satellite
• The All-Sky Automated Survey for Supernovae (ASAS-SN) Light Curve Server v1.0
• Measuring the rotation period distribution of field M-dwarfs with Kepler
• Fast Automated Analysis of Strong Gravitational Lenses with Convolutional Neural Networks
• Inferring probabilistic stellar rotation periods using Gaussian processes
• The Half Million Quasars (HMQ) Catalogue
• Testing the recovery of stellar rotation signals from Kepler light curves using a blind hare-and-hounds exercise
• Flare Statistics for Young Stars from a Convolutional Neural Network Analysis of $\text{TESS}$ Data
• Surface rotation of Kepler red giant stars
• Stellar rotation periods from K2 Campaigns 0-18 -- Evidence for rotation period bimodality and simultaneous variability decrease
• Variability Properties of 4 Million Sources in the $TESS$ Input Catalog Observed with the Kilodegree Extremely Little Telescope Survey
• Chemical Evolution in the Milky Way: Rotation-based ages for APOGEE-Kepler cool dwarf stars
• Evidence for Metallicity-Dependant Spin Evolution in the Kepler field
• Rotating Stars from Kepler Observed with Gaia DR1
• Systematics-insensitive Periodogram for finding periods in TESS observations of long-period rotators
• The RAdial Velocity Experiment (RAVE): Parameterisation of RAVE spectra based on convolutional neural networks
• Astraea: Predicting Long Rotation Periods with 27-Day Light Curves
• Asteroseismic constraints on active latitudes of solar-type stars: HD173701 has active bands at higher latitudes than the Sun
• Starspot rotation rates vs. activity cycle phase: Butterfly diagrams of Kepler stars are unlike the Sun's
• Data-driven derivation of stellar properties from photometric time series data using convolutional neural networks

Cited by in corpus (18)

• Rotation Distributions around the Kraft Break with TESS and Kepler: The Influences of Age, Metallicity, and Binarity
• The HD 260655 system: Two rocky worlds transiting a bright M dwarf at 10 pc
• The TIME Table: Rotation and Ages of Cool Exoplanet Host Stars
• Stellar Cycles in Fully Convective Stars and a New Interpretation of Dynamo Evolution
• TOI-4010: A System of Three Large Short-Period Planets With a Massive Long-Period Companion
• Stellar spectral-type (mass) dependence of the dearth of close-in planets around fast-rotating stars. Architecture of Kepler confirmed single-exoplanet systems compared to star-planet evolution models
• Closeby Habitable Exoplanet Survey (CHES). I. Astrometric Noise and Planetary Detection Efficiency due to Stellar Spots and Faculae
• Identifying Light-curve Signals with a Deep Learning Based Object Detection Algorithm. II. A General Light Curve Classification Framework
• The unusual M-dwarf Warm Jupiter TOI-1899~b: Refinement of orbital and planetary parameters
• Measuring Long Stellar Rotation Periods (>10 days) from TESS FFI Light Curves is Possible: An Investigation Using TESS and ZTF
• Predicting Stellar Rotation Periods Using XGBoost
• Theoretical and Observational Evidence for Coriolis Effects in Coronal Magnetic Fields Via Direct Current Driven Flaring Events
• The post-disk (or primordial) spin distribution of M dwarf stars
• 200,000+ Deep Learning-inferred Periods of Stellar Variability from the All-Sky Automated Survey for Supernovae
• The TESS-Keck Survey. XI. Mass Measurements for Four Transiting sub-Neptunes orbiting K dwarf TOI-1246
• Characterisation of starspot structure and differential rotation of Kepler-411
• SpinSpotter: An Automated Algorithm for Identifying Stellar Rotation Periods With Autocorrelation Analysis
• Activity and Rotation of Nearby Field M Dwarfs in the TESS Southern Continuous Viewing Zone