Three Hot Jupiters Are Reshaping Their Orbits—but Tidal Theory Only Explains Two

Every time a hot Jupiter crosses the face of its star, it blocks a tiny fraction of starlight. The dip lasts hours, but the exact midpoint can shift by seconds over years. That is the signal astronomers hunt when they ask a deceptively simple question: Is the planet's year getting shorter?

A new preprint led by A. C. Kutluay of Keele University and Ankara University has chased that question for 20 hot Jupiters—gas giants that orbit their stars in less than about 10 days. The team combined 42 new ground-based transit observations with data from NASA's TESS spacecraft, the Exoplanet Transit Database, and earlier studies, assembling 2,930 measured transit times. They then asked whether each planet's orbit is steady, shrinking, or growing, and compared those answers with computer models of tides inside the host stars.

The result is not a simple tale of planets spiraling to their doom. Most of the 20 systems show no convincing change. Only three—WASP-12 b, TrES-1 b, and WASP-121 b—have timing data that favor a changing orbital period. Two of those stories fit tidal theory. One does not.

A Stopwatch Made of Starlight

A transiting planet works like a celestial clock. If its orbit is constant, each transit arrives at a predictable time. If the orbit shrinks or grows, those arrival times drift earlier or later by tiny amounts. These shifts are called transit timing variations. The changes in orbital period are tiny—often just milliseconds per year—and they accumulate into a measurable timing signal only after many years of observations.

Hot Jupiters are ideal clocks for this work. Their gravity raises a tidal bulge in the star, and the star's gravity raises one in the planet. Energy is lost as the bodies flex, and angular momentum moves between the planet's orbit and the star's spin. Depending on which way the transfer goes, the planet can spiral inward or outward.

The efficiency of that process is often described by a number called the modified stellar tidal quality factor, written Q-star-prime. Smaller values mean stronger tidal dissipation. But Q-star-prime is not a fixed constant of nature; it depends on the star's internal structure, rotation, and evolutionary stage.

Most Hot Jupiters Keep a Steady Beat

For the majority of the sample, a constant orbital period was the best fit. That does not mean tides are absent. It means the current data cannot yet detect their slow work. Some systems had quadratic fits suggesting small changes, but the statistical evidence was weak. The authors also tested whether apsidal precession—a slow turning of an eccentric orbit—could explain the timing shifts. For the systems they tested, they found no statistically significant support for that alternative; for the rest, previous work had ruled it out or found circular orbits.

WASP-12 b: A Decay That Keeps Getting Sharper

WASP-12 b is the veteran of this field. It was the first exoplanet reported to show tidal orbital decay, and new data have kept confirming it. The new analysis updates its decay rate to -29.4 ± 4.0 milliseconds per year. In other words, the planet's year is getting about 29 milliseconds shorter each Earth year. The corresponding tidal quality factor is 1.72 ± 0.18 × 10^5, a low value that indicates efficient dissipation.

That number fits theoretical expectations only under certain conditions. The star may be a subgiant, evolved off the main sequence, where internal gravity waves can break near the core and drain orbital energy. Alternatively, the star may be on the main sequence but possess a strong magnetic field that converts gravity waves into magnetic waves. The timing data alone cannot yet decide between those pictures.

The TrES-1 b Puzzle: A Decay That Theory Cannot Match

TrES-1 b offers a sharper mystery. Its transit times favor a shrinking orbit at -14.9 ± 0.6 milliseconds per year, using more than two decades of observations. If interpreted as stellar tidal dissipation, it implies Q-star-prime = 570 ± 60.

That is a problem. Theoretical models for this star predict a value around a million or higher—the paper's own stellar models give about 2.6 million—hundreds to thousands of times larger and therefore far weaker. In the paper's own stellar models, the planet is not expected to be massive enough to break internal gravity waves in the star's radiative core. The star also rotates too slowly for inertial waves to be excited in the way seen in some other systems. So the observed decay, while statistically favored, sits uneasily with standard tidal theory.

The authors consider possible alternatives, including magnetic activity and unseen companions, but do not claim to have solved the puzzle. Stellar activity can shift apparent transit times, and small eccentricities can be difficult to pin down from radial velocity data. For now, TrES-1 b is a caution: a timing signal can be real in the data and still not fully understood physically.

WASP-121 b Is Moving the Other Way

WASP-121 b is the counterexample. Its transit times suggest the orbital period is increasing by 15.1 ± 0.8 milliseconds per year. Here, theory and observation line up. The host star rotates faster than the planet orbits, with a rotation period of about 1.13 days compared with the planet's 1.27-day orbit. In that configuration, tidal waves in the star's convective envelope—inertial waves—can transfer angular momentum to the orbit and push the planet outward.

The paper's models predict efficient inertial wave dissipation for WASP-121 b, making outward migration plausible. Unlike WASP-12 b and TrES-1 b, the sign of the period change prevents a clean lower limit on Q-star-prime, so the result is more about direction than precise tidal efficiency.

What Transit Timing Still Cannot Settle

This work is a preprint and has not completed peer review. Its conclusions rest on combining many data sets of varying quality, and transit timing remains vulnerable to starspots, stellar pulsations, and subtle instrument effects. The authors discarded noisy light curves and outliers, but they cannot remove every astrophysical contaminant.

The broader picture is also uneven. In about half the sample, theoretical models predict that gravity waves should break in the star and drive orbital decay. Yet the timing data do not show it. The authors note that some of those predictions depend on whether the fully damped gravity-wave regime actually applies, and that magnetic wave conversion, uncertain ages, or different stellar structures could change the picture.

Doomed Worlds, and a Better Clock Ahead

The paper adds one more unsettling note. A separate angular-momentum check finds all 20 systems are 'Darwin unstable': they cannot settle into a long-lived tidal equilibrium. In three systems the total angular momentum is above the critical threshold, but the orbital angular momentum is too small relative to the star's spin; the other 17 fall below the critical threshold. Given enough time, the planets are expected to spiral inward until they reach the Roche limit—the tidal destruction radius—or are engulfed by their stars. Stellar magnetic braking would only complicate that fate.

For now, the most important result is not a verdict on tides. It is a sharper map of where the clocks agree with theory and where they do not. TESS is expected to observe at least half of these systems again in 2026 or 2027, and the PLATO mission is expected to launch before the end of 2026, opening another long baseline for transit timing. Each new transit adds a tick to a clock that runs for years. In those tiny drifts, astronomers are learning how stars and planets exchange momentum—and why some worlds fall while others drift away.