🔭 astrophysics

Exoplanets in ancient stellar populations: occurrence constraints and hot-Jupiter candidates in the Galactic halo

This study utilizes Gaia and TESS data to search for short-period transiting planets around ancient Galactic halo stars, finding no confirmed detections and establishing stringent upper limits on hot-Jupiter occurrence rates that are significantly lower than those in the Galactic disk, thereby suggesting that close-in giant planets are rare in old, metal-poor environments regardless of their in-situ or accreted origins.

Original authors: Dolev Bashi, Michelle Kunimoto, Kevin K. Hardegree-Ullman, Tianjun Gan, Sharon X. Wang, Zhen Yuan

Published 2026-05-20

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Dolev Bashi, Michelle Kunimoto, Kevin K. Hardegree-Ullman, Tianjun Gan, Sharon X. Wang, Zhen Yuan

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 Milky Way galaxy as a massive, bustling city. Most of the stars we know and love—like our Sun—live in the "downtown" area, known as the Galactic Disc. This neighborhood is young, rich in heavy elements (like gold and iron, which astronomers call "metals"), and it's full of families with planets.

But there's also the "Galactic Halo," which is like the ancient, dusty outskirts of the city. These stars are the city's oldest residents. They formed billions of years ago, back when the universe was still a construction site, and they are very poor in heavy metals.

The Big Question
For a long time, astronomers wondered: Do these ancient, metal-poor stars in the outskirts have planets? Specifically, do they have "Hot Jupiters"—giant gas planets that orbit incredibly close to their stars, like a giant rollercoaster ride?

In the metal-rich downtown, these planets are common (about 1% of stars have them). But in the metal-poor outskirts, the rules might be different. Without enough heavy metals to build the rocky cores of giant planets, maybe these ancient stars are just lonely.

The Detective Work
This paper is like a massive detective story. The authors used two powerful tools:

Gaia: A space telescope that acts like a high-precision GPS, mapping the speed and direction of stars to identify which ones belong to the ancient Halo.
TESS: A space camera that watches stars to see if they "wobble" or dim slightly, which happens when a planet passes in front of them.

They looked at over 11,000 of these ancient stars, searching for planets that orbit in less than 10 days.

The Findings: A Quiet Neighborhood
The results were surprisingly quiet.

The Search: They scanned the light curves of thousands of stars.
The Candidates: They found two potential "Hot Jupiter" candidates.
- One was orbiting a star in the "in-situ" group (stars born right here in the Milky Way).
- The other was orbiting a star in the "accreted" group (stars that were stolen from a smaller, crashed galaxy long ago).
The Catch: The second candidate was a "grazing" planet. Imagine a planet that only barely clips the edge of the star, like a car skimming the curb. Because of this weird angle, it's hard to be sure it's a planet and not a glitch or a binary star system. So, for the final math, the scientists mostly counted just the one solid candidate.

The Verdict
The paper concludes that Hot Jupiters are extremely rare in the Galactic Halo.

In the metal-rich Disc, about 1% of stars have these planets.
In the metal-poor Halo, the rate is likely less than 0.14%.

To put it in perspective: If the Disc is a city where every 100 houses has a giant trampoline in the front yard, the Halo is a neighborhood where you might have to visit 1,000 houses just to find one.

Why Does This Matter?
The study also split the Halo stars into two groups: those born here ("in-situ") and those stolen from other galaxies ("accreted"). Surprisingly, both groups had the same lack of planets. This suggests that the lack of planets isn't about where the star came from, but about what the star is made of.

It seems that in the early universe, when there wasn't enough "dust" (metals) to build giant planet cores, the process simply didn't work well. Whether the star was born in the Milky Way or a dwarf galaxy, if it was metal-poor, it likely stayed planet-free.

The Bottom Line
This paper confirms that the ancient, metal-poor outskirts of our galaxy are a barren landscape for giant, close-in planets. It reinforces the idea that you need a rich supply of heavy metals to build the massive planets we see so often in our own cosmic neighborhood. The universe has a "metal requirement" for making Hot Jupiters, and the ancient halo stars just didn't have the ingredients.

Technical Summary: Exoplanets in Ancient Stellar Populations

Problem Statement
The Galactic halo preserves a record of the Milky Way's earliest assembly, comprising both in-situ stars formed during the Galaxy's infancy and stars accreted from dwarf galaxies. These populations are characteristically old and metal-poor. While the Galactic thin disc is rich in exoplanets, particularly around solar-metallicity stars, the occurrence of planets in the metal-poor halo remains poorly constrained. Previous searches have yielded few detections, leaving it unclear whether planet formation is suppressed in low-metallicity environments, whether the halo's high binary fraction inhibits planet formation, or if the lack of detections is merely a result of small-number statistics. Furthermore, it is unknown whether planet formation proceeds differently in the in-situ proto-Milky Way compared to the dwarf galaxy progenitors that were later accreted.

Methodology
The authors conducted a search for short-period transiting planets ( $1 < P < 10$ days) around kinematically selected halo dwarfs using data from Gaia DR3 and TESS.

Sample Selection: A sample of 20,836 halo dwarf stars was constructed from Gaia DR3 astrometry and radial velocities. Selection criteria included high-quality 6D phase-space information, dwarf classification via a Kiel diagram ( $T_{\text{eff}}$ vs. $\log g$ ), and a total Galactocentric velocity threshold ( $V_{\text{tot}} > 220$ km s $^{-1}$ ) to isolate halo kinematics.
Population Classification: The sample was divided into in-situ and accreted subpopulations using the $E-L_z$ (energy vs. angular momentum) plane, adopting the kinematic classification of Belokurov & Kravtsov (2023) recalibrated to the McMillan (2017) Galactic potential.
Transit Search: Light curves from the TESS Quick-Look Pipeline (QLP) covering Sectors 1–93 were analyzed. The search utilized a Box-Fitting Least Squares (BLS) algorithm to identify signals with Signal-to-Noise Ratios (SNR) $> 9$ and at least three transits.
Vetting and Modeling: Candidates underwent automated vetting using LEO-Vetter to distinguish planets from false positives (e.g., eclipsing binaries). Surviving candidates were modeled using Bayesian transit fitting (via exoplanet and pymc) to derive precise parameters, including impact parameters ( $b$ ) and radii.
Occurrence Rate Calculation: Detection efficiency was characterized via injection-recovery tests (injecting $\sim$ 224,000 synthetic transits). Occurrence rates were calculated using a Bayesian methodology (Hsu et al. 2018) modeling the planet population as a Poisson point process, accounting for geometric transit probabilities and pipeline completeness.

Key Contributions

Candidate Identification: The study identified two Hot-Jupiter (HJ) candidates ( $R_p > 8 R_\oplus$ , $P < 10$ days). One candidate (TIC-421991589.01) resides in the in-situ halo, while the other (TIC-263176654.01) is in the accreted halo.
Metallicity Constraints: If confirmed, the accreted candidate would orbit the most metal-poor HJ host known to date ([Fe/H] $\approx -0.93$ ). The in-situ candidate orbits a star with [Fe/H] $\approx -0.29$ .
Subpopulation Comparison: This work presents the first occurrence-rate comparison between in-situ and accreted halo populations within a homogeneous, kinematically selected sample.
Forward Modeling: The authors compared observed detections against a forward model assuming Kepler-like occurrence rates to quantify the suppression of planets in the halo.

Results

Detection Status: Of the two candidates, TIC-421991589.01 has a non-grazing transit ( $b < 0.9$ ) and is included in occurrence calculations. TIC-263176654.01 is highly grazing ( $b \approx 1.14$ ) and is excluded from the occurrence analysis due to the high likelihood of it being an eclipsing binary.
Occurrence Rates:
- Full Halo: Assuming the non-grazing candidate is a planet, the HJ occurrence rate is $0.13^{+0.12}_{-0.07}\%$ . If the candidate is a false positive, the $1\sigma$ upper limit is $< 0.14\%$ .
- In-situ Halo: The inferred rate is $0.17^{+0.17}_{-0.10}\%$ (or $< 0.19\%$ assuming no detections).
- Accreted Halo: Based on zero confirmed detections in this subsample, the upper limit is $< 0.56\%$ .
Comparison to Disc: These rates are significantly lower than the $\sim 0.5\% - 0.6\%$ HJ occurrence rates measured for Sun-like stars in the Galactic disc.
Forward Model Discrepancy: A forward model assuming Kepler-like abundances predicts $10 \pm 3$ detections in the sample, whereas at most one was observed. This indicates a substantial suppression of close-in giant planets in the halo.
In-situ vs. Accreted: No significant difference in occurrence was found between the in-situ and accreted populations, suggesting that the factors suppressing close-in planets operate similarly in both environments.

Significance
The paper claims that close-in giant planets are intrinsically rare in the old, metal-poor Galactic halo, with occurrence rates well below those of the Galactic disc. The results support the core accretion theory, which predicts that giant planet formation is disfavored at low metallicities due to insufficient dust surface density for rapid core growth. The lack of a significant difference between in-situ and accreted populations implies that the suppression mechanisms (likely related to metallicity and early protoplanetary disc conditions) were effective in both the proto-Milky Way and the accreted dwarf galaxy progenitors. The study establishes robust upper limits and provides the first tentative occurrence rate based on a candidate detection for halo stars, sharpening the constraints on planet formation in the earliest epochs of the Galaxy.

More like this