scieee AI-readable full text Open interactive document viewer

A New Type of Stochastic Low Frequency Variable Star: the Extreme Helium Stars

Crawford, Courtney

Publisher
Zenodo
DOI
10.5281/zenodo.17430020
Repository
zenodo.org
Language
en

Abstract

The launch of TESS revealed a new type of variability known as stochastic low frequency (SLF) variations, characterized by a smoothly varying red noise in power density space, rather than the discrete peaks associated with coherent modes. SLF variability is common in massive stars such as O & B dwarfs and blue supergiants, which is similar to the granulation component of solar-like oscillators. Its physical origin remains uncertain, with proposed mechanisms including sub-surface convection zones, internal gravity waves, variable stellar winds, and surface granulation. Using TESS light curves, we report for the first time that Extreme Helium (EHe) stars also exhibit SLF variability. The EHe stars are a class of low-mass (<1 Msun) B-type supergiants with atmospheres that are helium-rich (~98% by mass) and incredibly hydrogen-deficient (<1% by mass). The leading theory for their formation is that they are the remnant of a double-white-dwarf merger with total mass less than the Chandrasekhar limit, making them a low-mass counterpart to Type Ia supernovae. In this talk, I will present TESS observations of several EHe stars and our ongoing efforts to characterize their SLF variability using Gaussian Processes, which are well-suited to modeling correlated, stochastic signals. I will compare these results with recent studies of SLF behavior in massive and solar-like stars. Finally, I will discuss whether the variability mechanism in EHe stars aligns more closely with those of hot, massive stars or cool, low-mass stars. Our study will help to illuminate how surface and internal processes operate across different stellar masses and evolutionary histories, and shed light about these rare remnants of double white dwarf mergers.

Full text

A new type of Stochastic Low-Frequency Variable The Extreme Helium Stars Courtney Crawford - USyd Simon Jeffery, May Gade Pedersen, Tim Bedding, Geoff Clayton, Patrick Tisserand, Jamie Soon, Nikita Nikultsev Aerts (2021) Kurtz (2022) Aerts (2021) Kurtz (2022) What are these? PV Tel and R CrB The Hydrogen-deficient Giants • Both are types of giant stars with H-deficiency • < 1% H, ~98% He, ~1% C (by mass) •Masses < 1 M ⊙ PV Tel (EHe) R CrB (HdC) g, T data from A. Philip Monai (2023) • R CrB = Hydrogen-deficient Carbon (HdC) stars • Strong Carbon Molecular features • PV Tel = Extreme Helium (EHe) stars • Strong Helium lines (extremely strong…) The Hydrogen-deficient Stars PV Tel (EHe) R CrB (HdC) He-sdO DO WD O(He) A. Philip Monai (2023), Zhang (2014), Crawford (2025) DWD Merger HdC EHe He-sdO/B O(He) DO WD Suggested in Webbink (1984) The EHe stars (PV Tel variables) • PV Tel is a variable star classification created in 1985 by Walker & Hill using an RV light curve of PV Tel itself. Walker & Hill (1985) The EHe stars (PV Tel variables) • Initial PV Tel classification - Walker & Hill (1985) • In 2008, Jeffery suggested that EHes should be split into two (three) variable classes Jeffery (2008) PV Tel The EHe stars (PV Tel variables) • Initial PV Tel classification - Walker & Hill (1985) • PV Tel I, PV Tel II, BX Cir split - Jeffery (2008) • In 2020, Jeffery published the first TESS light curve of PV Tel itself - highlighting its stochasticity Jeffery (2020) The EHe stars (PV Tel variables) • Initial PV Tel classification - Walker & Hill (1985) • PV Tel I, PV Tel II, BX Cir split - Jeffery (2008) • PV Tel is stochastic - Jeffery (2021) Now, we have significantly more TESS data of all known EHe stars Time (d) Time (d) Time (d) TESS Flux Compared to other SLFs SLF data from Pedersen (2025) Typical Luminosity Spectroscopic Luminosity What causes SLF Variability? • 4 (published) possibilities: • Surface Granulation (as in solar-like oscillations) • Subsurface Convection Zones • Internal Gravity Waves • Variable, clumpy stellar winds What causes SLF Variability? • 4 (published) possibilities: • Surface Granulation (as in solar-like oscillations) • Subsurface Convection Zones • Internal Gravity Waves • Variable, clumpy stellar winds No surface convection zone in models What causes SLF Variability? • 4 (published) possibilities: • Surface Granulation (as in solar-like oscillations) • Subsurface Convection Zones • Internal Gravity Waves • Variable, clumpy stellar winds No surface convection zone in models FeCZ (Cantiello 2009, 2021) consistent with EHe models What causes SLF Variability? • 4 (published) possibilities: • Surface Granulation (as in solar-like oscillations) • Subsurface Convection Zones • Internal Gravity Waves • Variable, clumpy stellar winds No surface convection zone in models No convective cores FeCZ (Cantiello 2009, 2021) consistent with EHe models What causes SLF Variability? • 4 (published) possibilities: • Surface Granulation (as in solar-like oscillations) • Subsurface Convection Zones • Internal Gravity Waves • Variable, clumpy stellar winds No surface convection zone in models No convective cores FeCZ (Cantiello 2009, 2021) consistent with EHe models Possible? Wind variability unclear What causes SLF Variability? • 4 (published) possibilities: • Surface Granulation (as in solar-like oscillations) • Subsurface Convection Zones • Internal Gravity Waves • Variable, clumpy stellar winds No surface convection zone in models No convective cores FeCZ (Cantiello 2009, 2021) consistent with EHe models Possible? Wind variability unclear Could also be something else! Potential Caveat Some SLF targets are isolated Potential Caveat Some SLF targets are isolated Many targets are quite crowded Conclusions/Future Work • Many EHe stars show evidence of stochastic low frequency variation • TESS has continually uncovered more and more of these types of variables • This could potentially help to constrain some of the theoretical models of SLFV • Future work will involve more detailed modelling of the power spectra (GPs, typical fitting) • Lots of EHe stars are in the bulge (s91, s92) which will take more work to get data from • Comparisons with longer baseline historical data may reveal more interesting variability