Exoplanets · Space Codex
K2-18 b
A world 2.4 times Earth's width that gets Earth-like sunlight.
A sub-Neptune at the centre of a biosignature debate
Artist's concept. No telescope has seen K2-18 b as more than a point of light, and most exoplanets not even that. What we know comes from how a planet dims or tugs its star. Any picture of its surface, colour or sky, anywhere, is an illustration, not an observation.
At a glance
Key points
JWST robustly detected methane; a claimed sign of life, dimethyl sulfide, is disputed and not confirmed.
- Methane detected by JWST
- DMS 'sign of life' claim is disputed
- May be a gas-rich mini-Neptune, not an ocean world
The claim
In 2023 a team led by Nikku Madhusudhan reported methane and carbon dioxide and proposed a 'Hycean' world, an ocean under a hydrogen atmosphere, with a tentative hint of dimethyl sulfide, which on Earth is made by life. A 2025 mid-infrared analysis claimed DMS or its relative DMDS at about 3σ.
The pushback
Independent teams reprocessed the same data with dozens of methods. They confirm methane but find no reliable evidence for DMS, and carbon dioxide is uncertain. A gas-rich mini-Neptune with no habitable surface explains the spectrum as well, and a 2026 formation study favours a dry interior.
What we still don't know
Whether K2-18 b has any ocean at all. Methane is robust; the life-related molecules are not established.
About this view
Artist's concept. Interpretation debated: drawn as a hazy, gas-rich sub-Neptune because whether an ocean lies below is disputed, so no ocean is shown. The haze tint is a guess; the orange light follows from its 3,457 K star.
Missions
- 2021–
Infrared observatory at Sun–Earth L2, about 1.5 million km from Earth.
- 2009–2018
Kepler / K2
NASACompletedTransit survey that showed small planets are common.
First-hand · our 3D model
How we render it
K2-18 b is drawn as a hazy, gas-rich sub-Neptune: a deep, low-contrast haze deck under the strongest haze shell of any exoplanet in the model. Whether an ocean lies beneath is debated, so none is drawn. The view opens a little past quarter phase under the 3,457 K star, so part of the lit haze deck shows and the haze also glows along the limb.
No telescope has imaged the surface of any planet shown here, so what you see is an artist's concept: the surface is generated by a shader from noise, not drawn from a map, and its colours and features are illustrative. What is tied to measurement is the starlight and the geometry. The host star's colour is a blackbody at its catalogued temperature, and the planet is tidally locked, with a fixed day side facing the star and a permanent night side.
The host star hangs in the sky at its true apparent size seen from the planet (its radius divided by the orbit), with limb darkening that is stronger in blue and animated granulation. The planet is lit only from that star's direction and in its colour, adapted 40% towards white the way an eye adapts to a single light source. A surface glows only where it is hot enough to glow in visible light, judged from Planck's law at 550 nm. Where the concept has an atmosphere, a ray-marched haze shell adds the limb glow (10 samples at full quality, 6 at reduced, none at minimal).
Related entries
- LHS 1140 bA temperate super-Earth, possibly icy
- James Webb Space TelescopeInfrared observatory at L2
- Exoplanets: the censusPlanets around other stars
Sources
- Madhusudhan et al. 2025, ApJL (MIRI) (opens in a new tab)
- Wogan et al. 2024, ApJL (opens in a new tab)
- Fernández-Rodríguez et al. 2026, A&A (opens in a new tab)
- NASA Exoplanet Archive: K2-18 overview (opens in a new tab)
- Schmidt et al. 2025, AJ (arXiv) (opens in a new tab)
- Luque et al. 2025, A&A (opens in a new tab)
Last reviewed against these sources. Built from public NASA, ESA and JPL data; not affiliated with or endorsed by any agency.