Few distant worlds have drawn as much attention, or as much argument, as K2-18 b. It was found in 2015 when it passed in front of its star, a cool red dwarf about 124 light-years away. It orbits every 33 days inside the star's habitable zone, the range of distances where liquid water could exist. At 2.4 times Earth's width and almost 9 times its mass, it is a sub-Neptune: a planet between Earth and Neptune in size, a type our solar system lacks.
In 2023 a team led by Nikku Madhusudhan used the James Webb Space Telescope (JWST) to study starlight filtering through the planet's atmosphere. They reported methane and carbon dioxide, with little ammonia. They argued this fits a hycean world: a planet with a hydrogen-rich atmosphere over a global water ocean. They also saw a weak hint of dimethyl sulfide (DMS), a gas that on Earth comes from living things, mostly ocean plankton. In 2025 the same group reported a stronger signal, at about 3-sigma confidence, from JWST's mid-infrared instrument. They attributed it to DMS or a related gas, dimethyl disulfide, and said more data and lab work were needed.
Other teams disagree. A reanalysis led by Stephen Schmidt tried 60 ways of processing the 2023 data and found methane, but no reliable evidence for carbon dioxide or DMS. A team led by Rafael Luque combined all of JWST's data and found insufficient evidence for either sulfur gas, because other molecules fit just as well. They estimated about 25 more mid-infrared transits would be needed to settle it. A study led by Renyu Hu, using four new JWST observations, detected methane and carbon dioxide but only marginal hints of DMS, and showed that such gases can form without life in a hydrogen-rich atmosphere.
So methane is the one gas all these groups accept. Carbon dioxide is supported by some analyses and not others, and DMS is a contested claim, not a detection. The planet's interior is debated too. Hu's team favours a water-rich world, either a thick steamy envelope or a thin atmosphere over an ocean. Schmidt's team found the data fit an oxygen-poor mini-Neptune with no ocean at all.
TerraForge's model, working only from mass, orbit and star, classifies K2-18 b as an ice giant at about 324 K. That broadly matches readings that put deep water or high-pressure ice under a thick gas envelope, but the model cannot tell whether an ocean or life is there.