Global asteroseismology of 19,000 red giants in the TESS Continuous Viewing Zones

This paper presents a comprehensive asteroseismic catalogue of 19,151 red giants in the TESS Continuous Viewing Zones, utilizing seven years of data to achieve high-precision measurements of stellar parameters that significantly expand the known population of oscillating giants and provide valuable uniform data for Galactic Archaeology.

The Gist

Imagine the universe as a giant, silent orchestra. For a long time, we could only see the musicians (the stars) from a distance, guessing what they were playing based on how bright they looked. But recently, we've learned that stars aren't silent; they are constantly vibrating, ringing like giant bells. These vibrations are sound waves trapped inside the star, and by listening to them, we can figure out exactly what the star is made of, how heavy it is, and how old it is. This field is called asteroseismology (star-quakes).

This paper is like a massive new music sheet written by a team of astronomers using NASA's TESS satellite. Here is the story of what they did, explained simply:

1. The Mission: Listening to the "Continuous View"

TESS is a space telescope that scans almost the entire sky, but it usually only looks at a patch of sky for about a month before moving on. However, there are two special spots near the top and bottom of the sky (the "Continuous Viewing Zones" or CVZs) that TESS can stare at for years without blinking.

Think of these zones as VIP boxes at a concert. While other seats only get a 27-minute snippet of the show, the VIP boxes get a 3-year uninterrupted performance. The astronomers decided to use this long, uninterrupted time to listen to the "songs" of red giant stars (which are old, puffy stars like our Sun will become).

2. The Challenge: Finding the Needle in the Haystack

The team looked at over 72,000 stars in these VIP zones. But not all of them were singing. They had to filter out the noise.

• The Visual Check: First, humans looked at the data like detectives looking for a specific pattern in a noisy room. They found about 19,000 stars that were definitely singing.
• The Computer Check: Then, they used a super-fast computer program (called nuSYD) to double-check. It's like having a second pair of eyes that never gets tired.
• The "Blend" Problem: Because TESS's "eyes" (pixels) are a bit blurry, sometimes the light from two stars gets mixed together. If a neighbor star is singing, it might look like your target star is singing. The team had to carefully check the neighborhood to make sure they weren't listening to the wrong star.

3. The Discovery: A Louder, Clearer Song

By listening for three years instead of just one month, they achieved something amazing:

• They heard fainter stars: Before, they could only hear the "loud" (bright) stars. Now, they could hear the "whispers" of distant, dimmer stars.
• They heard higher notes: They found stars vibrating at very high speeds that previous short observations missed.
• The Result: They created a catalog of 19,151 singing red giants. This is an 80% increase in known singing red giants in this part of the sky!

4. Decoding the Stars: The "Star ID Card"

Once they knew which stars were singing, they used the "notes" of the song to build an ID card for each star.

• The Pitch ($\nu_{max}$): How fast the star is vibrating tells us its surface gravity (how heavy the surface feels).
• The Spacing ($\Delta\nu$): The distance between the notes tells us the star's average density.

By combining these musical clues with the star's temperature and color, they could calculate the star's Mass (how heavy it is) and Radius (how big it is) with incredible precision—about as accurate as the best measurements we've ever had from the famous Kepler mission.

5. Sorting the Musicians: The "Old" vs. The "Middle-Aged"

Red giants go through different life stages.

• RGB (Red Giant Branch): These stars are like teenagers; they are burning hydrogen in a shell around a dead core. They are still growing.
• CHeB (Core Helium Burning): These are the "middle-aged" stars that have started burning helium in their core. They have settled down.

The team used a Neural Network (a type of AI) to look at the shape of the sound waves and sort the stars into these two groups. It's like a music critic listening to a song and instantly knowing if the band is in their "rookie" phase or their "veteran" phase.

6. The Big Picture: Mapping the Galaxy's History

Why does this matter? Because mass is a proxy for age.

• Heavy stars burn their fuel fast and die young.
• Light stars burn slowly and live long.

By measuring the mass of these 19,000 stars and looking at where they are in the galaxy (how high they are above the galactic plane), the team found a pattern:

• Stars closer to the galactic "floor" (the plane) are generally younger and heavier.
• Stars floating higher up are older, lighter, and have been "kicked" up there over billions of years by the gravity of the galaxy.

This is like looking at a pile of leaves. The fresh, green leaves are at the bottom, and the dry, old leaves are scattered higher up. By studying these stars, astronomers are essentially doing Galactic Archaeology, digging up the history of how our Milky Way formed and evolved.

Summary

In short, this paper is a massive, high-definition audio recording of the universe. By listening to the vibrations of 19,000 old stars for three years straight, the astronomers have created a precise map of the Milky Way's past, proving that with enough patience and the right tools, we can hear the history of our galaxy singing.

Technical Summary

1. Problem Statement

While space missions like Kepler and CoRoT revolutionized asteroseismology, they were limited to specific, small fields of view. The Transiting Exoplanet Survey Satellite (TESS) offers nearly all-sky coverage but typically provides short observation baselines (27 days per sector), which limits its ability to detect oscillations in faint red giants or those with high frequencies of maximum power ($\nu_{\text{max}}$). However, TESS has two Continuous Viewing Zones (CVZs) at the ecliptic poles (Northern and Southern) where observation baselines extend to 2–3 years. The challenge is to leverage these long baselines to create a large, homogeneous catalogue of oscillating red giants to study Galactic structure and stellar evolution, overcoming issues related to noise, blending, and the calibration of asteroseismic scaling relations.

2. Methodology

The authors analyzed TESS data from Sectors 1–87 (Years 1–7) covering the Northern (NCVZ) and Southern (SCVZ) Continuous Viewing Zones.

• Sample Selection:
  • Selected stars from the TESS Input Catalog (TIC v8.2) within the CVZs with ecliptic latitudes $>78^\circ$ or $<-78^\circ$.
  • Restricted the sample to stars with TESS magnitude ($T_{\text{mag}}$) $< 13.5$ to ensure detectability.
  • Applied color-magnitude cuts ($1 < G_{BP} - G_{RP} < 2$ and $-3 < M_G < 4$) to isolate red giants, resulting in an initial sample of 72,647 stars.
• Data Processing:
  • Utilized TESS-SPOC light curves (preferred for lower noise) and QLP light curves where SPOC was unavailable.
  • Applied high-pass filtering (Gaussian kernel, $\sigma=5$ days) and $5\sigma$ clipping to remove outliers.
  • Generated power density spectra using Lomb–Scargle periodograms up to the Nyquist frequency of 283.4 $\mu$Hz.
• Oscillation Detection:
  • Visual Inspection: Manually classified 72,647 power spectra as "yes," "maybe," or "no" based on Gaussian-shaped power excess.
  • nuSYD Pipeline: Applied a computationally efficient method to estimate $\nu_{\text{max}}$ using a correlation between Gaia absolute magnitude ($M_G$) and $\nu_{\text{max}}$ derived from Kepler data.
  • Signal-to-Noise (SNR) Verification: Calculated empirical SNR and detection probability ($p_{\text{det}}$) using the method of Chaplin et al. (2011). Only stars with $p_{\text{det}} > 0.99$ were retained.
• Blending Analysis:
  • Addressed TESS's large pixel size (21.6 arcsec) by calculating Shape-Based Distances (SBD) between target and neighbor power spectra.
  • Flagged 813 stars (5.1%) as potential blends but retained them with a specific flag rather than discarding them.
• Parameter Estimation:
  • Global Parameters: Used the pySYD pipeline to measure $\nu_{\text{max}}$ and the large frequency separation ($\Delta\nu$) via autocorrelation.
  • Reliability Vetting: Employed a Convolutional Neural Network (CNN) trained on Kepler data to verify $\Delta\nu$ measurements by checking for vertical ridges in collapsed échelle diagrams.
  • Evolutionary Classification: Used a CNN to classify stars as Red Giant Branch (RGB) or Core Helium Burning (CHeB) based on folded power spectra.
  • Stellar Parameters: Derived mass, radius, and surface gravity using asteroseismic scaling relations.
    • Applied corrections for the large frequency separation ($f_{\Delta\nu}$) using the asfgrid tool.
    • Assumed $f_{\nu_{\text{max}}} = 1$ due to the lack of a robust theoretical framework for metallicity-dependent corrections.
    • Combined seismic data with spectroscopic parameters ($T_{\text{eff}}$, [M/H]) from Yu et al. (2023) and Gaia XP spectra (Andrae et al. 2023).

3. Key Contributions

• Catalogue Creation: Produced a homogeneous catalogue of 19,151 oscillating red giants in the TESS CVZs.
• Increased Sensitivity: Identified an 80% increase in the number of known oscillating red giants at $T_{\text{mag}} > 8$ compared to previous studies, demonstrating the power of long-baseline TESS data.
• High Precision: Achieved typical precisions of 1.5% for $\nu_{\text{max}}$ and 1.0% for $\Delta\nu$, comparable to 4-year Kepler data.
• Gold Sample: Defined a high-quality "gold sample" of 5,226 stars with reliable $\Delta\nu$, clear evolutionary classification, and spectroscopic parameters, achieving median precisions of 7.2% in mass, 2.6% in radius, and 0.01 dex in $\log g$.
• Methodological Validation: Validated the use of TESS long-baseline data for detecting high-$\nu_{\text{max}}$ (less evolved) and faint (distant) stars, expanding the reach of asteroseismology.

4. Key Results

• Stellar Parameters:
  • The median mass of the sample is $1.17 M_\odot$ (corrected) and $1.23 M_\odot$ (uncorrected).
  • Seismic radii show excellent agreement with Gaia-derived radii, with a median offset of only 0.7% for RGB stars and 0.3% for CHeB stars, significantly smaller than previous reports of ~4% offsets.
• Evolutionary Features:
  • Successfully identified the RGB bump and the Zero Age Helium Burning (ZAHeB) edge in the mass-radius and Kiel diagrams.
  • Observed a deficit of low-mass RGB stars in the CVZ compared to Kepler, likely due to selection effects.
  • Identified a population of CHeB stars with $\log g < 2$ (previously unseen in Kepler), which appear to be in a short-lived helium-flash phase or misclassified.
• Galactic Archaeology:
  • Analyzed the sample in Galactocentric coordinates, revealing clear trends:
    • Mass-Age-Height: Stellar mass decreases with increasing vertical distance ($Z$) from the Galactic plane, consistent with older stars being dynamically heated to higher altitudes.
    • Metallicity: Metallicity decreases with increasing $Z$.
    • Kinematics: The Toomre diagram shows a gradient in mass with total velocity, distinguishing thin disk, thick disk, and halo populations. The halo population contains significantly fewer CHeB stars.

5. Significance

This work represents a major step forward in Galactic Archaeology. By providing a large, uniform sample of red giants with precise seismic masses and radii across the TESS CVZs, the authors have created a powerful tool for:

1. Mapping the Milky Way: The sample allows for the study of stellar populations at high Galactic latitudes and distances previously inaccessible to Kepler.
2. Calibrating Spectroscopy: The high-precision seismic $\log g$ values serve as a benchmark to calibrate spectroscopic pipelines (e.g., APOGEE, GALAH).
3. Testing Stellar Evolution: The clear delineation of the RGB bump and ZAHeB edge provides critical constraints for stellar evolution models.
4. Future Surveys: The success of this methodology validates the use of TESS long-baseline data for future large-scale asteroseismic studies, bridging the gap between Kepler and future missions like PLATO.

The catalogue is publicly available and includes flags for blending, evolutionary state, and data reliability, making it a foundational resource for the astrophysical community.