Picture a world squashed into the shape of a football, orbiting so close to its star that iron and magnesium boil off the top of its sky. That’s WASP-121b, and the JWST caught it doing something stranger than its shape. Its dawn and its dusk are running two different atmospheres.
Astronomers led by Cyril Gapp at the Max Planck Institute for Astronomy found that the evening side of this tidally locked exoplanet absorbs more starlight than the morning side. It’s the clearest proof yet of a split that atmospheric models had only predicted.
Here’s the part that grabbed my attention. This isn’t the obvious day vs. night contrast you would expect on a planet with a permanent sunlit face. It’s a difference between two thin sliver-zones that are both, technically, at the same line between day and night. The James Webb Space Telescope pulled them apart anyway. Below I unpack what was measured and how a single telescope reads one planet longitude by longitude.
Key Takeaways
- The dusk side wins. WASP-121b’s evening terminator absorbs more infrared starlight than its morning terminator. That confirmed a long-predicted asymmetry.
- This is dawn vs. dusk, not day vs. night. Both terminators sit on the day-night boundary; the new result separates the two edges of that boundary.
- Eastward winds are the engine. Heat dragged off the permanent dayside piles up on the evening side, puffing the atmosphere out so it intercepts more light.
- Temperatures in the upper atmosphere are high enough to rip H2O molecules apart, so water genuinely thins out where it’s hottest.
- JWST mapped the planet across roughly 30 degrees of rotation during one transit, turning time into longitude.
- One mystery remains: the measured asymmetry is stronger than simulations predict, and silicate mineral clouds are the leading, unconfirmed suspect.
What the James Webb Space Telescope Found on WASP-121 b
The JWST found that the evening edge of WASP-121b is hotter and more swollen than the morning edge. JWST watched infrared starlight filter through the planet’s atmosphere as it crossed in front of its host star, and the absorption came back lopsided. The evening terminator soaked up more light than the morning terminator, which only makes sense if the two sides differ in temperature and chemistry.
To be precise, a tidally locked planet has a permanent dayside and a permanent nightside. The thin ring between them is the terminator, and it has two arcs. The morning arc, where the surface is rotating into daylight, and the evening arc, where it’s rotating out. This study is about those two arcs, dawn and dusk.
How do you isolate one arc of a planet 858 light-years away? You exploit the fact that the planet rotates a little during the few hours it spends crossing its star. The signal shifts as it turns, and that shift maps to position on the globe. So, the takeaway is that JWST didn’t just average the whole planet into one blurry reading; it read the dawn and the dusk separately, and they didn’t match.
Three signals carried the story: overall absorption, carbon monoxide, and water. Each one points the same direction, toward an evening hotter side.
Why the evening side runs hotter than the morning side
The chain is short and physical. WASP-121b’s permanent dayside is a furnace, and the planet’s winds blow eastward, in the direction of its rotation. Those winds scoop heat off the dayside and dump it onto the evening terminator before the gas cools.
Hot gas expands. And as the evening atmosphere swells, it presents a bigger silhouette to the incoming starlight, so it blocks and absorbs more of it during transit. Wind carries heat to the dusk side; the dusk side inflates, the inflated side eats more light; that’s the whole mechanism: The morning side, having spent the night cooling, stays smaller and lets more light slip past.
WASP-121b at a Glance: An Ultra-Hot Jupiter Exoplanet
Before the table, it helps to know why this specific planet was the right case study. WASP-121b is a tidally locked, ultra-hot Jupiter exoplanet parked absurdly close to its star. This makes its day-night contrast extreme and its atmosphere bright and puffy enough to read. The closeness also means it rotates fast during transit, which is exactly the lever the new method needs. So, here are the numbers.
| Specification | Value |
| Dayside temperature | ~2,500 °C (2,770 K) |
| Nightside temperature | ~725 °C (1,000 K) |
| Planet type | Ultra-hot Jupiter (officially named Tylos) |
| Rotation and orbit | ~30 hours, tidally locked |
| Distance to its star | 1.9 stellar diameters (~0.0257 AU) |
| Instrument used | JWST NIRSpec |
| Distance from Earth | ~858 light-years (constellation Puppis) |
| Discovery year | 2015 (WASP survey) |
Source | SciTechDaily, ScienceDaily, NASA Exoplanet Catalog, Nature Astronomy
How a Tidally Locked Planet Creates Permanent Day and Night
Tidal locking is what happens when a planet sits close enough to its star for the star’s gravity to act unevenly across the planet’s body. Over time, that gravitational drag slows the spin until the planet’s rotation period matches its orbital period. One full turn equals one full orbit. The Moon does this to Earth, which is why we only ever see one lunar face.
For WASP-121b, both the spin and the orbit take about 30 hours. But the consequence is very dramatic; one hemisphere faces the star forever, the other faces cold space forever. There’s no sunrise that sweeps across the planet, no day that turns into night. The boundary between light and dark is frozen in place as a fixed ring, the terminator.
That fixed geometry is the only reason this study is possible. Because the dayside and nightside never trade places, the temperature gap between them is enormous but stable. On WASP-121b it runs from around 2,770 K on the day hemisphere to roughly 1,000 K on the night hemisphere, a difference of more than 1,700 degrees. A gap that wide and that steady is what drives the violent winds, and the winds are what make the morning and evening edges diverge in the first place.
How JWST Mapped the Exoplanet Atmosphere Longitude by Longitude
This is the genuinely new contribution, and it’s cleverer than the result itself. A transit isn’t instantaneous. WASP-121b takes a few hours to cross the face of its star, and during that crossing it keeps rotating. By the time it exits, the planet has turned by roughly 30 degrees. Different longitudes rotate into view as the hours pass.
The old approach threw that motion away. Astronomers would stack every measurement from the whole transit into one combined spectrum to beat down the noise. Cleaner signal, but you lose the timing, and with it any sense of which part of the planet you are looking at. It is like averaging every frame of a movie into a single smeared photograph.
Gapp’s team did the opposite. They let the signal change over time and tied each moment to the longitude facing Earth at that instant. The leading edge of the orbit corresponds to the morning terminator; the trailing edge corresponds to the evening one. As the planet turns, JWST’s NIRSpec instrument reads a slightly different slice of atmosphere, and time becomes a coordinate on the globe.
Here’s the more convincing part. The team ran the statistics both ways, and the time-varying model fit the data significantly better than the flat averaged one. When a more complicated model beats a simpler one by a clear statistical margin, the extra complexity is usually capturing something physical. Here that something is the literal shape of the planet’s weather, read off one transit, one longitude at a time. That’s a measurement that didn’t exist a couple of years ago.
Why this beats earlier dawn-dusk studies
To be precise, this isn’t the first time anyone noticed morning and evening differing on a hot exoplanet. An earlier 2024 JWST study confirmed that a tidally locked giant can host distinct eternal-morning and eternal-evening skies, which was itself a milestone.
What’s new in 2026 is the precision. Rather than detecting that a difference exists, James Webb Space Telescope mapped the difference across longitude with rotational resolution, and it found the asymmetry.
Water Is Being Ripped Apart in the Hottest Air
Two molecules did the heavy lifting in the spectra, and the trick is reading them correctly, because they tell opposite-looking stories for different reasons.
Carbon monoxide first. NIRSpec picked up a stronger CO signal toward the end of the transit. More carbon monoxide over there, but that’s wrong. The team attributes the louder signal to temperature, not abundance. Hotter gas radiates and absorbs more emphatically, so the same amount of CO simply shouts louder when it’s warmer. The molecule count did not change; the thermostat did.
Water is the genuine drop. In the hottest regions of the upper atmosphere, conditions get violent enough to break H2O apart into its constituent atoms. The water isn’t hiding or being outshone. It’s being destroyed, dissociated into hydrogen and oxygen by sheer heat.
That difference is important. One signal gets louder without the substance increasing; the other genuinely shrinks because the substance is being torn apart. Conflating the two would imply chemistry that isn’t happening. The fact that water thins out precisely where the evening winds are dumping heat is also an independent confirmation that the dusk side really is the hotter side.
The Mystery: Mineral Clouds the Models Can’t Explain
Here’s where the paper gets refreshingly honest about what it doesn’t know. The researchers built simulations of how heat moves through the upper atmosphere of a gas giant, and the models did reproduce the general direction of the asymmetry. Temperature differences alone get you a hotter, puffier evening side. So far so good.
But the problem is magnitude. The observed asymmetry came out stronger than the simulations could generate. Something extra is happening, most likely on the morning side, that the temperature-only models are missing.
The leading suspect is clouds, but not the kind you would recognize. On a world this hot, clouds wouldn’t be water vapor. Earlier work has suggested WASP-121b may host clouds made of minerals such as silicates, essentially vaporized rock condensing high in the atmosphere. Such clouds could block infrared radiation rising from the hotter layers below, making the morning side read as cooler than it physically is and widening the gap between dawn and dusk.
When the team crudely added cloud effects to their model, the fit improved. Modeling how clouds form, condense, and evaporate in an atmosphere changing this fast is genuinely hard, and most current models, including this one, don’t fully include that physics.
Why This Discovery Matters for Exoplanet Science
The significance isn’t the single planet. It’s the technique, and the door it opens.
A few concrete reasons this lands:
- It enables 3D atmospheric mapping. Reading a planet longitude by longitude turns a flat average into something closer to a globe with weather, which is a different category of data than a single combined spectrum.
- It’s repeatable. The method needs only a planet that rotates enough during transit and an atmosphere bright enough to read. It’s not a one-off trick that worked by luck on one target.
- More targets are already queued. The team has identified other ultra-hot gas giants with the right temperatures and spin rates for the same treatment.
Run this across a sample of extreme worlds, and you can start comparing their 3D structures, asking whether the dawn-dusk split is universal or whether it varies with temperature, gravity, or chemistry. That comparative angle is where space science usually makes its real jumps, by turning one weird object into a population you can reason about.
One thing I want to rule out clearly: none of this has anything to do with habitability. WASP-121b is a hellscape where rock vaporizes and metals stream off into space. It is a physics laboratory, not a candidate for life, and anyone framing it that way is selling something.
Final Thoughts
Strip it down and three things stand out. WASP-121b’s evening terminator is hotter and more bloated than its morning terminator, a dawn-versus-dusk asymmetry that models predicted but had never been pinned this precisely. JWST nailed it by reading the planet longitude by longitude across a single transit, converting the planet’s own rotation into a mapping tool. And one stubborn gap remains, an asymmetry too strong for the simulations, with mineral clouds as the prime but unproven suspect.
What I will be watching is the queue. The same method is already pointed at other ultra-hot Jupiters, and the moment a second and third planet get the longitude-by-longitude treatment, we stop talking about a curiosity and start talking about exoplanet weather as a comparative field. That is the part worth following.
FAQs
The evening side absorbs more of the starlight than the morning side of WASP-121b’s atmosphere, JWST revealed. In June 2026, it was published in Nature Astronomy; this dawn-dusk asymmetry validated predicted temperature and chemistry disparities between the two sides of the planet’s permanent day-night boundary for the first time.
Yes. WASP-121b is tidally locked, which means that its rotation period is roughly the same as its orbit period (30 hours), with one hemisphere permanently facing its star and the other permanently in darkness. This is a fixed configuration which establishes a persistent terminator zone and the extreme temperature difference that powers violent winds in the zone.
Eastward winds carry heat off the scorching permanent dayside and pile it onto the evening terminator before the gas cools. That heat makes the evening atmosphere expand, giving it a larger cross-section that intercepts and absorbs more starlight than the cooler, more compact morning side.
An ultra-hot Jupiter is an exoplanet that is a gas giant orbiting very close to its parent star, whose dayside temperature is often more than 2,000 K, hot enough to evaporate metals and break up water molecules in the upper atmosphere of the planet. WASP-121b has a dayside temperature of about 2,770 K, hot enough to vaporize metals and break up water molecules in the upper atmosphere of the planet.
The dayside of No. WASP-121b is among the most extreme of all the worlds known, with rock turned to vapor and metals such as iron and magnesium escaping into space. It has nothing to do with habitation and is just a site for studying atmospheric physics as a laboratory environment.
WASP-121b rotates roughly 30 degrees while crossing its star during transit. Instead of averaging the whole transit into one reading, astronomers let the signal vary over time and tied each moment to the longitude facing Earth, turning the planet’s rotation into a longitude-by-longitude map of its atmosphere.
