🔭 astrophysics

Most Hot Jupiters Were Cool Giant Planets for More Than 1 Gyr

By analyzing the Galactic velocity dispersions of hot Jupiter subpopulations to infer their relative ages, the study concludes that at least 40% to 70% of hot Jupiters arrived at their current orbits via late-time high-eccentricity migration occurring more than 1.5 Gyr after system formation, rather than forming in situ or through early disk migration.

Original authors: Stephen P. Schmidt, Kevin C. Schlaufman

Published 2026-01-22

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Stephen P. Schmidt, Kevin C. Schlaufman

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

The Big Mystery: How Did "Hot Jupiters" Get So Close to Their Stars?

Imagine our Solar System. Jupiter is a giant planet, but it orbits far away from the Sun, where it's cold. Now, imagine a "Hot Jupiter"—a planet just as massive as Jupiter, but it orbits so close to its star that it's scorching hot.

For decades, astronomers have been arguing about how these Hot Jupiters got there. There are two main theories:

The "Early Commuter" Theory: They formed far away and migrated inward very quickly (within a few million years) while the star was still a baby, riding on a disk of gas and dust.
The "Late Arriver" Theory: They formed far away, stayed cool for a long time, and then got kicked inward by a chaotic event (like a gravitational fight with another planet) billions of years later.

This paper tries to solve the mystery by acting like a cosmic detective, looking at the "age" of these planets to see which theory is right.

The Detective's Tool: The Galactic Speedometer

The problem is that it's very hard to tell exactly how old a single star is. However, the authors found a clever workaround using speed.

Think of the Milky Way galaxy as a busy highway.

Young cars (stars) tend to drive smoothly in the fast lane, staying close to the center. They don't swerve much.
Old cars (stars) have been driving for billions of years. They've hit bumps, been bumped by other cars, and their paths have become "wobbly." They move faster and more erratically.

In astronomy, this "wobble" is called velocity dispersion. The authors realized: If a group of Hot Jupiters is moving slowly and smoothly, they are likely young. If they are moving fast and erratically, they are likely old.

The Experiment: Sorting the Planets by "Seat Location"

The researchers took a huge list of Hot Jupiters and sorted them into three groups based on how close they are to their stars (their orbital period):

Inside-Peak: The very closest, hottest ones.
Near-Peak: The ones right in the middle of the most common distance.
Outside-Peak: The ones that are a bit further out (but still "hot" compared to Jupiter).

They measured the "wobble" (speed) of the stars hosting these planets to figure out their ages.

The Surprise Discovery

The results were like finding that the people sitting in the middle of a theater are older than the people sitting in the back row.

The "Outside-Peak" Group (The Back Row): These planets were found to be younger (about 2.2 to 2.4 billion years old).
The "Near-Peak" and "Inside-Peak" Groups (The Middle and Front Rows): These planets were found to be older (about 3.1 to 3.3 billion years old).

Wait, that's backwards! If the "Early Commuter" theory were true, all these planets should be roughly the same age. If the "Late Arriver" theory were the only thing happening, the closest ones should be the youngest (because they just arrived).

But the data showed the opposite: The planets closest to the star are actually the oldest.

The Solution: A Mix of Two Stories

The authors propose a "Three-Act Play" to explain this:

Act 1: The Early Settlers (The Outside-Peak): Some Hot Jupiters formed early and moved in quickly (via disk migration). These are the "Outside-Peak" group. They are the youngest because they haven't been around as long as the others, and they haven't been pulled in any closer by tides yet.
Act 2: The Late Arrivals (The Near-Peak): A huge chunk of Hot Jupiters (at least 40%, maybe up to 70%) formed far away and stayed there for over 1.5 billion years. They were "cool giant planets" for a long time. Then, a chaotic event (like a planet fight) kicked them inward. Because they arrived late, they are older than the early settlers.
Act 3: The Tidal Pull (The Inside-Peak): Once those "Late Arrivals" got close to the star, the star's gravity acted like a giant magnet, slowly pulling them even closer over billions of years. This is why the "Inside-Peak" group is the oldest of all—they have been in the "hot seat" the longest, slowly spiraling inward.

The Bottom Line

The paper concludes that most Hot Jupiters we see today were actually cool, distant giants for more than a billion years.

They didn't just form and immediately burn up. Instead, they spent a long time as "regular" giant planets before a late-stage cosmic event sent them on a one-way ticket to the star, where they slowly got pulled in even tighter over the next few billion years.

In short: Most Hot Jupiters are not "new arrivals" that just got there; they are "veterans" that have been on a very slow, billion-year journey to their current hot spots.

Technical Summary: "Most Hot Jupiters Were Cool Giant Planets for More Than 1 Gyr"

Problem Statement
The origin of hot Jupiters (giant planets with orbital periods $P_{\text{orb}} < 10$ days) remains a central unresolved issue in exoplanet astrophysics. Two primary formation pathways are proposed:

Early-time mechanisms: In situ formation or disk migration, which place hot Jupiters in their observed locations within a few million years (Myr) of system formation.
Late-time mechanisms: High-eccentricity migration driven by planet–planet scattering or Kozai-Lidov oscillations, followed by tidal damping. These processes typically require $\gtrsim 1$ Gyr to deliver planets to their final orbits.

Previous studies have yielded conflicting evidence. Some suggest high-eccentricity migration is the dominant channel based on spin-orbit misalignments, while others argue for disk migration based on the lack of perturbing companions in many systems. Furthermore, previous attempts to constrain formation timescales using Galactic velocity dispersions (a proxy for stellar age) have been limited by the inability to precisely date individual host stars and the difficulty of distinguishing between subpopulations with different formation histories.

Methodology
The authors propose a statistical test based on the hypothesis that different formation mechanisms produce distinct age distributions for hot Jupiters located at different positions relative to the peak of the orbital period distribution.

Sample Selection: The study utilizes a sample of 503 transiting hot Jupiters ( $0.1 < M_p < 10 M_{\text{Jup}}$ , $P_{\text{orb}} < 10$ d) from the NASA Exoplanet Archive (as of July 2025). The sample is restricted to transiting systems to avoid the intrinsic age bias of Doppler surveys (which preferentially target older, quieter stars). Systems discovered by Kepler are excluded to ensure the analysis is calibrated to the solar neighborhood age–velocity dispersion relation rather than the Kepler field relation.
Subpopulation Definition: The sample is divided into three subpopulations based on orbital period relative to a debiased peak ( $P_{\text{peak}} = 3.92$ $P_{peak} = 3.92$ d), calculated by weighting periods by the inverse of the scaled semimajor axis ( $a/R_*$ $a / R_{*}$ ) to correct for transit probability biases:
- Inside-peak: $P_{\text{orb}} < 3.259$ d.
- Near-peak: $3.259 \text{ d} < P_{\text{orb}} < 4.545$ d.
- Outside-peak: $P_{\text{orb}} > 4.545$ d.
Age Calibration via Kinematics: Since precise ages for individual host stars are unavailable, the authors use Galactic velocity dispersion ( $\sigma$ ) as a proxy for age. They construct a calibrated solar neighborhood age–velocity dispersion relation using a sample of 47,390 subgiants with precise ages (from D. M. Nataf et al. 2024) and a metallicity distribution matched to the hot Jupiter hosts ( $-0.06 < [\text{Fe/H}] < 0.36$ ).
Forward Modeling: To quantify the relative contributions of formation mechanisms, the authors develop a parametric forward model simulating the evolution of hot Jupiter populations. The model includes three components:
- Early-time, uniformly-distributing mechanism (disk migration/in situ).
- Early-time, peak-populating mechanism.
- Late-time, peak-populating mechanism (high-eccentricity migration) with a time delay ( $\Delta\tau$ ).
  The model incorporates tidal evolution using the J. Leconte et al. (2010) framework and varies the stellar tidal quality factor ( $Q'_*$ ) and the fraction of late-time arrivals ( $f_{\text{LP}}$ ).

Key Results

Age Ordering: The analysis reveals a distinct age hierarchy among the subpopulations:
- Outside-peak: Characteristic mean age $\tau \approx 2.20\text{--}2.36$ Gyr.
- Near-peak: Characteristic mean age $\tau \approx 3.11\text{--}3.27$ Gyr.
- Inside-peak: Characteristic mean age $\tau \approx 3.16\text{--}3.30$ Gyr.
  The outside-peak subpopulation is approximately 750 Myr younger than the near- and inside-peak subpopulations. The inside-peak population is slightly older than the near-peak population.
Mechanism Constraints: This age ordering supports a scenario where an early-time mechanism produces a uniform distribution, while a late-time mechanism preferentially populates the orbital period peak. Subsequent tidal evolution removes the oldest members of the inside-peak population (via inspiral), leaving a slightly older characteristic mean age compared to the near-peak population, while the outside-peak population remains largely unaffected by tides.
Quantitative Limits: The forward modeling indicates that to reproduce the observed age offsets and velocity dispersions, the stellar tidal quality factor must be in the range $10^{6.5} \lesssim Q'_* \lesssim 10^7$ . Under these conditions, the model requires that at least 40% but no more than 70% of the hot Jupiter population formed via a late-time, peak-populating mechanism with a characteristic delay of $\Delta\tau \gtrsim 1.5$ Gyr.
Validation: The velocity dispersion of the Kepler-discovered outside-peak subpopulation is consistent with matched control samples of stars without hot Jupiters, confirming that these systems formed early and survived tidal evolution without being engulfed.

Significance and Claims
The paper claims to resolve the "oldest problem in exoplanet astrophysics" by demonstrating that the hot Jupiter population is not monolithic in its formation history. The authors conclude that:

Most hot Jupiters near the orbital period peak are "late arrivals." They spent more than 1.5 Gyr as "cool giant planets" on wider orbits before migrating inward via high-eccentricity mechanisms.
The outside-peak population represents the "survivors" of early formation. These systems likely formed via disk migration or in situ processes and have not been significantly altered by tidal inspiral over the main-sequence lifetime of their host stars.
Tidal evolution is a critical sculpting force. The difference in age between the inside-peak and near-peak populations is attributed to the efficient tidal destruction of the oldest, closest-in systems formed by the late-time mechanism.

The authors emphasize that these conclusions are made possible by the TESS mission's ability to detect hot Jupiters around evolved stars and in the outside-peak regime, which were largely invisible to previous ground-based surveys. They argue that previous demographic studies dominated by ground-based data may have been biased toward older systems and should be reevaluated. The paper does not claim to identify the specific dynamical triggers (e.g., specific companion types) for the late-time migration but establishes the timescale and prevalence of the phenomenon.