JWST Directly Images Coldest Exoplanet Ever: Water-Ice Clouds Upend Giant Planet Models

For decades, planetary scientists could not get a direct look at a cold, mature giant planet beyond our solar system. Instruments were not powerful enough to isolate the faint infrared glow of a world sitting in the frigid outer reaches of another star's system. On April 22, 2026, that barrier fell — and the first clear view has already broken the models astronomers spent years building to predict what they would find.

Using the James Webb Space Telescope's MIRI instrument, a team led by Elisabeth Matthews at the Max Planck Institute for Astronomy in Heidelberg, Germany, directly imaged and characterized Epsilon Indi Ab — a super-Jupiter orbiting just 11.9 light-years from Earth. The results, published in The Astrophysical Journal Letters, mark the first time astronomers have obtained direct photometric measurements of a cold, solar-age giant exoplanet. What they found was not what anyone expected.

Nearest Cold Super-Jupiter Sits in Our Cosmic Backyard

Epsilon Indi Ab orbits the orange dwarf star Epsilon Indi A in the southern constellation Indus, a system visible to the naked eye from the Southern Hemisphere. The planet weighs in at 7.6 times the mass of Jupiter while retaining roughly the same diameter — a dense heavyweight that orbits about 30 astronomical units from its host star, comparable to Neptune's distance from our Sun. Its orbital period of approximately 180 Earth-years means it has never been observed completing a single circuit.

What makes this planet exceptional is its temperature. With an effective surface temperature of 275 Kelvin — barely above the freezing point of water — Epsilon Indi Ab sits in a thermal regime that older telescopes like Hubble and Spitzer could not probe. The vast majority of directly imaged exoplanets are young, gravitationally hot objects radiating at temperatures above 1,000 K. Epsilon Indi Ab is a mature, slowly cooling world whose atmosphere resembles the outer planets of our own solar system far more than it resembles the hot Jupiters that dominate exoplanet databases.

Reaching those cold temperatures required JWST's MIRI coronagraph, which blocks the blinding glare of the host star and isolates the planet's own mid-infrared emission. The team compared new observations at 11.3 micrometers — a wavelength band just outside a spectral feature associated with ammonia — against 2024 images taken at 10.6 micrometers. The ratio between the two channels allows researchers to estimate how much ammonia is present in the upper atmosphere.

How Does JWST Detect Water-Ice Clouds on an Exoplanet?

The comparison delivered a surprise. Based on the planet's mass and temperature, models predicted its upper atmosphere would be heavily laden with ammonia gas — mirroring the ammonia-dominated chemistry seen in Jupiter's own cloud layers. Instead, the planet appeared significantly brighter than predictions, and the ammonia signal came back weaker than a clear-sky atmosphere should produce.

The team's best explanation: a thick, patchy blanket of water-ice clouds sitting high in the atmosphere, analogous in form to the high-altitude cirrus clouds in Earth's own atmosphere. These clouds would reflect additional infrared emission upward — making the planet brighter overall — while simultaneously masking the ammonia gas below them, suppressing the signal instruments expected to detect.

"It's a great problem to have, and it speaks to the immense progress we're making thanks to JWST," said James Mang, a co-author of the study at the University of Texas at Austin. "What once seemed impossible to detect is now within reach, allowing us to probe the structure of these atmospheres, including the presence of clouds. This reveals new layers of complexity that our models are now beginning to capture."

Cold-Giant Atmospheric Models Must Be Rebuilt

The discovery carries a direct and uncomfortable implication for theorists: most existing atmospheric models for cold giant planets do not include clouds at all. Clouds are notoriously difficult to simulate — they form at specific pressure levels, vary in patchiness, and interact with atmospheric chemistry in ways that add layers of computational complexity. Many research groups have simplified their models by omitting clouds entirely, treating them as a secondary effect at temperatures like Epsilon Indi Ab's.

The new photometry shows that assumption was wrong. The suppressed ammonia signal is consistent with cloud interference at exactly the temperature range this planet occupies. The paper notes that a small but growing sample of cold giant exoplanets shows a consistent pattern of being fainter than models predict between 3 and 5 micrometers — a signal also consistent with the water-ice cloud hypothesis. Epsilon Indi Ab may not be an outlier; it may be a typical example of a class of planets whose atmospheres standard models have been misrepresenting for years.

Matthews was direct about what the telescope has changed: "JWST is finally allowing us to study solar-system analog planets in detail. If we were aliens several light-years away looking back at the Sun, JWST would be the first telescope that could allow us to study Jupiter in detail."

She was equally clear about the limits: studying an Earth analog at this level of detail would require substantially more powerful instruments — either a future generation of 30-meter-class ground telescopes or a dedicated space observatory such as the proposed NASA Habitable Worlds Observatory.

Direct Imaging Opens Doors Transmission Spectroscopy Cannot

The methodological significance of the study extends beyond the specific finding about clouds. For most of JWST's history as an exoplanet instrument, the telescope has relied on transit spectroscopy — analyzing how starlight filters through a planet's atmosphere as the planet crosses in front of its star. This technique is powerful, but it selects almost exclusively for planets on close, hot orbits, because only those planets are geometrically likely to pass across their star's face from Earth's perspective.

Epsilon Indi Ab orbits at 30 astronomical units and takes 180 years to complete a single loop. From Earth's viewing angle, it will never transit its star. The only way to study its atmosphere is by suppressing the star's light and collecting the planet's own emission directly — which is precisely what MIRI's coronagraph accomplished. The result validates this approach at the cold-giant tier: what the team demonstrated with a 275-Kelvin world is the same direct-imaging capability that will need to be refined and extended, eventually, to rocky planets in the habitable zones of nearby stars.

Roman Space Telescope Next, Then a Growing Cold-Jupiter Sample

The Matthews team is already applying for additional JWST observation time to study further cold Jupiter analogs. If the suppressed ammonia pattern and the water-ice cloud hypothesis hold across a broader sample, they would establish these features as standard characteristics of the cold-giant class rather than anomalies of a single unusual planet.

A near-term observational opportunity also exists for the reflective component of the water-ice clouds: NASA's Nancy Grace Roman Space Telescope, now targeting early September 2026 for launch, is designed to observe reflected starlight from planets at wide orbital separations and is well suited to detect the highly reflective signature that water-ice clouds would produce.

The long arc — from the first JWST image of Epsilon Indi Ab in 2024 to its detailed atmospheric characterization in 2026 — illustrates how rapidly the field is moving. Each result at the cold-giant tier lays groundwork for the eventual instruments that will attempt the same science on far smaller, far fainter worlds. The coldest exoplanet ever directly imaged has given astronomers both a problem to solve and a method that works. In the search for planets that might resemble our own, those two things together turn out to be exactly what the field needed.

Frequently Asked Questions

What did JWST find on Epsilon Indi Ab?

JWST's MIRI instrument detected less ammonia in the planet's upper atmosphere than existing models predicted, with the planet appearing brighter than expected across mid-infrared wavelengths. The team's best explanation is a thick, patchy layer of water-ice clouds sitting high in the atmosphere, similar to cirrus clouds on Earth, which block the ammonia signal below and reflect additional infrared emission outward.

Why is direct imaging of cold exoplanets so difficult?

Most exoplanet atmosphere studies use transit spectroscopy, which requires a planet to pass in front of its star from Earth's perspective. Cold giant planets orbit far from their stars and rarely align for transits, making them effectively invisible to that technique. Direct imaging avoids this limitation by using a coronagraph to block the star's glare and collect the planet's own faint infrared glow — but this requires the extreme sensitivity and infrared precision that only JWST currently provides.

What does this discovery mean for the search for Earth-like planets?

The Epsilon Indi Ab result validates direct atmospheric characterization at temperatures and orbital distances that transit spectroscopy cannot reach. The same technique, applied with more powerful future instruments such as the proposed NASA Habitable Worlds Observatory or next-generation 30-meter ground telescopes, is the pathway astronomers plan to use to eventually study rocky planets in the habitable zones of nearby stars.

How does water ice form in an exoplanet atmosphere?

At around 275 Kelvin — close to the freezing point of water — water vapor in a planet's upper atmosphere can condense into ice crystals, forming cloud layers analogous to cirrus clouds in Earth's stratosphere. These clouds are highly reflective and can trap molecular signals like ammonia beneath them, making a planet appear brighter overall while suppressing the specific spectral features that instruments use to measure its chemical composition.