Classification of Periodic Variable Stars from TESS

Xinyi Gao, Xiaodian Chen, Shu Wang, Jifeng Liu · 2024

Read Classification of Periodic Variable Stars from TESS by Xinyi Gao, Xiaodian Chen, Shu Wang, Jifeng Liu (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-06175.txt.

Introduction

Xinyi Gao and co-authors sort variable stars in TESS data in a 2024 astronomy paper. The space telescope's all-sky stare catches stars that pulse, spin, or eclipse, and the haul has grown faster than human classifiers can label it.

Using two-minute cadence data from the mission's first sixty-seven sectors, the team catalogued more than seventy-two thousand periodic variables. Nineteen features, spanning period, physical traits, and light-curve shape, fed a random forest that sorted them into twelve subtypes. Most of the stars were newly classified, including tens of thousands of rotating stars. Checks against other catalogs showed high purity for pulsators and eclipsing pairs.

Catalog record

CREATORXinyi Gao, Xiaodian Chen, Shu Wang, Jifeng Liu
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE63 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2412-06175

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Classification of Periodic Variable Stars from TESS Xinyi Gao, Xiaodian Chen, Shu Wang, Jifeng Liu Abstract The number of known periodic variable stars has increased rapidly in recent years. As an all-sky transit survey, the Transiting Exoplanet Survey Satellite (TESS) plays an important role in detecting low-amplitude variable stars. Using 2-minute cadence data from the first 67 sectors of TESS, we find 72,505 periodic variable stars. We used 19 parameters including period, physical parameters, and light curve (LC) parameters to classify periodic variable stars into 12 sub-types using random forest method. Pulsating variable stars and eclipsing binaries are distinguished mainly by period, LC parameters and physical parameters. GCAS, ROT, UV, YSO are distinguished mainly by period and physical parameters. Compared to previously published catalogs, 63,106 periodic variable stars (87.0 % \% ) are newly classified, including 13 Cepheids, 27 RR Lyrae stars, ∼ \sim 4,600 δ \delta Scuti variable stars, ∼ \sim 1,600 eclipsing binaries, ∼ \sim 34,000 rotational variable stars, and about 23,000 other types of variable stars. The purity of eclipsing binaries and pulsation variable stars ranges from 94.2 % \% to 99.4 % \% when compared to variable star catalogs of Gaia DR3 and ZTF DR2. The purity of ROT is relatively low at 83.3 % \% . The increasing number of variables stars is helpful to investigate the structure of the Milky Way, stellar physics, and chromospheric activity. I Introduction Periodic variable stars have been studied for centuries. Periodic variable stars consist mainly of pulsation stars, eclipsing binary systems and rotational stars that exhibit regular or semi-regular luminosity variations. The periods of variable stars are usually associated with their physical properties, such as mass, luminosity, radius, and ages [ 13 ] . Variable stars have long been used to provide important insights into stellar structure and evolution [ 45 , 24 , 86 ] . Variable stars are not only important objects for constraining stellar evolution, but also important objects for studying the Milky Way structure [ 12 , 69 ] . The period–luminosity relations (PLRs) of classical Cepheids and RR Lyrae, for example, were used in the cosmological distance scale [ 23 , 15 , 61 ] . The study of variable stars dates back hundreds of years, but in the 1990s the development of CCDs led to significant advances in variable star search. A large number of large-scale surveys, including ground-based surveys and space-based surveys, have also contributed to the development of variable star discovery. The All Sky Automated Survey (ASAS) made the first variable star search covering almost the entire sky [ 51 ] . ASAS found 3,126 variable stars in the V band throughout the southern hemisphere and 11,509 variable stars brighter than V=15 mag throughout the northern hemisphere [ 52 , 53 ] . The Optical Gravitational Lensing Experiment (OGLE) has been used to study the variable stars for more than two decades [ 79 ] . It discovered 900,000 variable stars and led to significant advances in variable star research [ 78 , 80 ] . [ 8 ] extracted variability periods for 16,880 stars in the Next Generation Transit Survey (NGTS). The Gaia second Data Release (DR2) mapped all-sky RR Lyrae stars and Cepheids and discovered 50,570 new variable stars [ 17 ] . [ 14 ] searched and classified 781,602 periodic variable stars from Zwicky Transient Facility (ZTF) DR2. The All-Sky Automated Survey for Supernovae (ASAS-SN) obtained 687,695 variable stars throughout the sky, including 238,752 new discoveries [ 66 , 32 , 33 , 34 , 35 , 36 , 16 ] . In Gaia DR3, 12.4 million variable stars in the Milky Way and nearby galaxies were classified by supervised machine learning into 22 variability types [ 62 ] . As space telescopes, Kepler and Transiting Exoplanet Survey Satellite (TESS) play an important role in variable source searches for high photometry precision. The Kepler spacecraft, launched into an Earth-trailing orbit, focused almost continuously on a single region of the sky for four years, monitoring the brightness fluctuations of approximately 200,000 stars [ 4 , 76 ] . These data have facilitated the study of stellar variability in an unprecedented way. [ 50 ] found 947 variable stars in the Kepler field of view. [ 54 ] found 1,879 eclipsing binaries in the first Kepler data release, and [ 70 ] found 2,165 eclipsing binaries in the second Kepler data release. [ 3 ] provided an overview of stellar variability in the first quarter data from the Kepler mission. Following the failure of the Kepler mission reaction wheels, the K2 mission was initiated and observed both the northern and southern sky over two years, covering ten times more sky area than the original Kepler mission [ 31 ] . However, K2 experiences higher instrument noise due to increased instability of the spacecraft. TESS aims to discover planets by observing transits across bright and nearby stars, and an all-sky survey approach was adopted for TESS [ 59 , 72 ] . [ 72 ] and [ 73 ] described the TESS Input Catalogs (TIC). [ 34 ] identified \sim 11,700 variables, including \sim 7,000 new discoveries. [ 29 ] studied stellar flares for the 24,809 stars observed with a 2 minute cadence during the first two months of the TESS mission and identified 1,228 flaring stars. [ 84 ] measured the pulsation amplitude and period for 3,213 long period variable stars. [ 64 ] found 1,807 variable stars in full frame images. [ 28 ] presented 2,241 exoplanet candidates identified with data from TESS during its 2 yr Prime Mission. [ 2 ] found 506 variable objects. In the northern TESS continuous viewing zone, 3,025 A-F variable stars brighter than 11 mag were found, of which 1,813 objects were classified [ 68 ] . [ 65 ] identified 1,403 variable stars and [ 22 ] found over 46,000 periodic objects with high confidence. Specific types of variable stars have also been discovered in TESS observations, e.g. more than 20,000 binaries and 307 Ia supernovae [ 19 , 21 , 27 , 30 , 55 ] . [ 22 ] presented a stellar variability catalog using the 2 minute cadence photometry obtained during the TESS 2yr Prime Mission. There is still a dearth of work on the careful classification of large samples of variable stars in the TESS. In this work, we search for periodic variable stars in 67 sectors of the TESS and use the machine learning method and classify 70,100 periodic variable stars into 12 sub-types. The data are described in Section II . In Section III , we show how to identify periodic variable stars. We classify them using a machine learning method and show how the parameters work in classification in Section IV . Section V shows the variable star catalog, the newly classified variable stars, and a consistency comparison with the two previous catalogs. We describe and compare the properties of each subtype of variable star in detail in Section VI . Section VII summarizes our conclusions. II TESS data As an astrophysics exploration mission, TESS is placed in a highly elliptical 13.7-day orbit around the earth and uses four wide-field optical CCD cameras to

Read the complete text