Preprint: JWST spectra of K2-18b still consistent with Hycean ocean‑world models

A new arXiv preprint says current James Webb Space Telescope transmission spectra of K2-18b are still compatible with a physically self-consistent Hycean interpretation — a hydrogen-rich atmosphere above a liquid ocean — and do not require extra molecules such as dimethyl sulfide, or DMS. The paper, posted May 18 by Takuya Fujisawa, Masashi Shimada, Tatsuya Yoshida and Kiyoshi Kuramoto, is a preprint and has not been peer-reviewed.

The researchers tested whether self-consistent Hycean atmospheres could match existing Webb observations of the planet. According to the abstract, they combined one-dimensional photochemical modeling, radiative-convective equilibrium calculations and forward modeling of transmission spectra. Their setup assumes an atmosphere made mainly of hydrogen, methane and water vapor above a liquid ocean, then compares those models with JWST NIRISS and NIRSpec spectra spanning 0.8 to 5.2 micrometers. The authors report that Hycean models with roughly a 1-bar hydrogen envelope, percent-level methane and carbon monoxide, and carbon dioxide at about 10^-3 to 10^-2 can reproduce the spectra without invoking DMS or other additional species.

Just as important is what the paper does not claim. It does not say K2-18b has been shown to be habitable, or that an ocean has been detected, or that rival explanations have been ruled out. In the authors’ words: “While mini-Neptune scenarios remain viable, our results show that Hycean configurations are likewise consistent with the data, and current CO and CO2 constraints alone are not yet sufficient to rule out Hycean interpretations of K2-18b.” In other words, the study argues that current Webb data still do not settle the long-running debate over what kind of world K2-18b really is.

K2-18b is a temperate sub-Neptune, or possibly super-Earth, about 124 light-years away with an orbit of about 33 days, according to NASA background material. It has drawn unusual attention because it sits in a category between rocky planets and gas-rich worlds, where internal structure and surface conditions are hard to pin down from remote observations alone. “Hycean” is a term proposed in 2021 for planets with hydrogen-rich atmospheres over global oceans, making them a distinct candidate class in discussions of potentially habitable environments beyond Earth.

The new preprint adds to that debate by focusing on whether a Hycean scenario can be made physically self-consistent, rather than simply mathematically fitted to the data. The authors say their photochemical simulations naturally drive carbon monoxide to around 1% to 2% mixing ratios. They also write that liquid oceans can exist across a wide range of temperatures and pressures in their calculations, as long as the atmosphere avoids a runaway greenhouse state. The abstract further says that percent-level methane in an atmosphere with about 1 bar of hydrogen would require replenishment from the planet’s interior over billion-year timescales, and that the resulting vertical gradients can produce a relatively flat, methane-dominated transmission spectrum.

That matters because K2-18b has become one of the Webb era’s most closely watched exoplanets. Earlier JWST studies reported methane and carbon dioxide in its atmosphere, along with a tentative inference of DMS, a molecule that on Earth is associated with life. But follow-up analyses in 2025 argued that the DMS evidence was not robust and was highly sensitive to data reduction and modeling choices.

The new paper does not revive the case for DMS. Instead, it makes a narrower point: current Webb spectra can still be explained by a Hycean world without adding DMS or other extra molecules. For now, that means the observations alone still do not resolve whether K2-18b is a Hycean ocean world, a mini-Neptune, or something in between.