🔭 astrophysics

Homogeneous stellar parameters for 717,807 TESS FGK stars using Gaia DR3

This paper presents a homogeneous catalogue of 717,807 TESS FGK stars with precisely derived physical parameters—including mass, radius, temperature, and age—by combining Gaia DR3 data with isochrones, thereby providing a state-of-the-art resource for studying exoplanet demographics and recalculating radii for 278 confirmed and 915 candidate exoplanet hosts.

Original authors: Francesca Waines, Angharad Weeks, Vincent Van Eylen

Published 2026-04-01

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Francesca Waines, Angharad Weeks, Vincent Van Eylen

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the universe as a giant, bustling city. For a long time, astronomers have been trying to map this city, but they've been working with a messy, inconsistent address book. Some neighborhoods had detailed blueprints, while others just had scribbled notes. This made it hard to understand the "residents" of the city: the stars.

This paper is like a massive, high-tech renovation project. The authors, led by Francesca Waines, have created a brand new, perfectly organized address book for over 700,000 stars that are being watched by the TESS space telescope.

Here is the simple breakdown of what they did and why it matters:

1. The Problem: A Messy Address Book

Before this study, if you wanted to know the age, size, or weight of a star, you had to look at different lists. Some lists used one method to guess the age, others used a different method. It was like trying to compare the height of people using one ruler in inches and another in centimeters, without converting them first. This "inconsistency" made it hard to spot patterns in how planets form around these stars.

2. The Solution: The "Universal Translator"

The team used a powerful computer algorithm called BASTA (think of it as a super-smart translator) to look at all these stars at once. They fed it data from the Gaia satellite (which acts like a giant 3D map of the Milky Way) and the TESS telescope (which watches for planets).

Instead of guessing, they compared the stars' light and temperature against a massive library of theoretical star models (called "isochrones"). Imagine you have a photo of a person and a library of photos of people at every age from 1 to 100. You compare the photo to the library to find the exact match. That's essentially what they did for 717,807 stars.

3. The Big Reveal: We Now Know Their Ages

The most exciting part of this new catalog is that it includes ages for almost all these stars.

Why is this hard? You can't just ask a star, "How old are you?" Unlike humans, stars don't have birth certificates.
The Trick: The team looked at how heavy the star is, how hot it is, and what it's made of. By comparing these clues to their library of models, they could estimate the star's age with surprising accuracy.
The Result: They now have a list where every star has a consistent "birth date," a "weight," and a "size."

4. Why Does This Matter? (The Planet Connection)

This is like finally having a clean dataset to study how families are formed.

The "Hot Jupiter" Clue: The team found that giant gas planets (like Jupiter, but closer to their sun) tend to orbit stars that are "richer" in heavy metals. It's like finding that big, fancy houses are always built on land with very fertile soil.
The "Radius Valley": They looked at the sizes of planets and noticed a gap between small rocky planets and larger gas ones. This helps scientists understand how planets grow and shrink over time.
Better Science: Because the data is now "homogeneous" (all measured the same way), scientists can finally run big statistical tests. They can ask questions like, "Do older stars have more planets?" or "Do metal-rich stars have bigger planets?" without worrying that the data is just messy.

5. The "Quality Control" Check

The authors didn't just trust the computer blindly. They compared their new list against other famous lists of stars (like the "NASA Exoplanet Archive" or "SWEET-Cat").

The Verdict: Their numbers matched up very well.
The Improvement: While some older lists were great for a few specific stars, this new list is huge (covering 700,000 stars) and consistent. It's like upgrading from a hand-drawn map of a single street to a GPS satellite image of the entire country.

In a Nutshell

Think of this paper as the Great Unification of Star Data. The authors took a chaotic pile of information about 700,000 stars and turned it into a single, clean, reliable database. Now, when scientists want to study how planets are born and evolve, they have a perfect, consistent set of "parents" (the stars) to study, allowing them to finally understand the family dynamics of our galaxy.

1. Problem Statement

Accurate and homogeneous stellar characterization is fundamental to understanding exoplanet demographics, formation, and evolution. However, existing large-scale catalogues (such as the TESS Input Catalogue, TIC) suffer from inhomogeneity. Parameters in the TIC are derived from a mix of curated sub-samples, spline relations based on photometry, and Stefan-Boltzmann calculations, leading to systematic offsets and a lack of stellar age data. Furthermore, combining smaller, high-precision catalogues (e.g., SWEET-Cat, Berger et al.) is challenging due to differences in model grids, microphysics, and input parameters. There is a critical need for a massive, homogeneous sample of TESS stars with reliable ages to study the influence of host star properties on planetary systems.

2. Methodology

The authors constructed a homogeneous catalogue by applying a consistent isochrone fitting technique to a large sample of TESS stars using Gaia DR3 data.

Sample Selection:
- Target: 717,807 TESS FGK dwarfs and early subgiants (0.6 – 1.6 $M_\odot$ ).
- Constraints: Effective temperature ( $T_{eff}$ ) between 4000 K and 6500 K; surface gravity ( $\log g$ ) $\geq$ 3.8.
- Data Sources: Cross-matched with Gaia DR3 GSP-Spec (providing homogeneous $T_{eff}$ , $\log g$ , and [M/H] from RVS spectra) and Gaia photometry ( $G, G_{BP}, G_{RP}$ ) and parallax.
- Quality Control: Applied strict error filters on GSP-Spec parameters and removed stars with large temperature discrepancies between TIC and Gaia or missing parallaxes.
Stellar Characterization (BASTA):
- Algorithm: Used the BAyesian STellar Algorithm (BASTA) to fit pre-computed BaSTI (Bag of Stellar Tracks and Isochrones) isochrones.
- Inputs: Gaia photometry, parallax (corrected for zero-point offsets), GSP-Spec $T_{eff}$ and [M/H].
- Microphysics Strategy: To address mass-dependent physical processes, the authors ran two separate grids:
  1. Overshoot Case: Includes convective overshooting (suitable for higher mass stars).
  2. Diffusion Case: Includes overshooting, atomic diffusion, and mass loss (suitable for lower mass stars).
- Selection Logic: For each star, the best-fit parameters were selected based on the fitted mass:
  - If $M > 1.2 M_\odot$ in both fits $\rightarrow$ Overshoot case.
  - If $M \leq 1.2 M_\odot$ in both fits $\rightarrow$ Diffusion case.
  - If masses fell into different groups $\rightarrow$ The case with the highest log-likelihood.
- Filtering: Stars with invalid fits (e.g., age errors of 20 Myr, NaN values, or extremely young ages with low uncertainty) were removed, leaving the final sample of 717,807 stars.

3. Key Contributions

Massive Homogeneous Catalogue: Produced one of the largest homogeneous catalogues of stellar ages for TESS stars to date, covering 717,807 FGK dwarfs and subgiants.
Comprehensive Parameters: Provided derived values for distance, $\log g$ , [M/H], $T_{eff}$ , radius, mass, and age (a parameter largely missing from the TIC).
Exoplanet Host Identification: Identified 278 confirmed TESS exoplanet hosts and 915 candidates within the sample, recalculating their planet radii using the new homogeneous stellar radii.
Methodological Rigor: Demonstrated a robust approach to handling microphysical processes (diffusion vs. overshoot) by running dual grids and selecting the optimal fit based on the resulting stellar mass, avoiding arbitrary mass cuts that could introduce edge-case errors.

4. Results and Validation

The authors validated their derived parameters against various literature sources, ranging from the heterogeneous TIC to precise asteroseismic samples (Kepler LEGACY, TESS-Keck).

Precision Metrics (Median Absolute Errors):
- Radius: 0.06 $R_\odot$ (4.29% fractional error).
- Mass: 0.05 $M_\odot$ (4.49% fractional error).
- Temperature: 104 K (1.74% fractional error).
- Age: 2.1 Gyr (50.00% fractional error).
Comparisons:
- vs. TIC: Showed significant improvement in precision, particularly for mass (improved by a factor of ~2.5 compared to TIC's 11.12%).
- vs. SWEET-Cat & Berger et al.: Good agreement with median residual scatters of ~3-5% for radius and mass.
- vs. Kepler LEGACY & TESS-Keck: While these smaller, high-precision samples showed lower absolute errors (e.g., 1.09% radius error for LEGACY), the authors' sample achieved comparable precision for overlapping stars (25.60% age error for overlapping LEGACY stars vs. 50% for the full sample).
Exoplanet Analysis:
- Recalculated planet radii for 278 confirmed planets and 915 candidates.
- Confirmed the well-known correlation that Hot Jupiters orbit metal-rich stars, while smaller rocky planets show no such strong trend in this sample.
- The "radius valley" was not clearly resolved, likely due to the precision limits of the radius ratio and the specific sample composition.

5. Significance

This work provides a state-of-the-art resource for the astrophysical community, specifically for:

Exoplanet Demographics: Enabling large-scale statistical studies on how host star properties (mass, age, metallicity) influence planet occurrence rates, architectures, and evolution.
Stellar Physics: Offering a large dataset to test stellar models and isochrones across a wide range of evolutionary states.
Galactic Archaeology: Providing age estimates for a vast number of stars to map the formation history of the Milky Way.

By addressing the heterogeneity of previous catalogues and providing reliable ages for nearly 720,000 stars, this paper removes a major bottleneck in linking exoplanet populations to their stellar environments. The authors note that while age precision (50%) is lower than radius/mass, it represents a significant advancement over the lack of age data in the TIC and is comparable to other large-scale isochrone studies.