How to Validate JWST Biomarker Claims in 2026: Practical Guide

A JWST molecule detection is not automatically a biosignature. JWST biomarker validation requires a feature to survive competing molecules, independent data reductions, realistic photochemical models, and tests of the planet’s atmospheric architecture.

By that standard, JWST has revealed extraordinary chemistry—but it has not confirmed life beyond Earth.

Key takeaways

• A molecule is evidence of atmospheric chemistry, not biology by default.
• Retrieval results depend on the instrument, model assumptions, and molecules included in the comparison.
• A credible biosignature must survive alternative absorbers, independent reductions, abiotic chemistry, and planetary-context checks.

What JWST can actually detect

JWST covers approximately 0.6–28.8 micrometers, but no single instrument captures the whole atmospheric picture. Each observing mode trades wavelength coverage, resolution, throughput, and susceptibility to systematics.

The exact resolving power varies with wavelength and observing mode, so these are useful operating ranges rather than universal instrument constants.

A spectrum first produces a feature: a change in transit depth near a wavelength where one or more molecules may absorb. A retrieval then estimates which atmospheric compositions best explain that feature. The output is a probability distribution conditioned on the data, temperature structure, cloud model, stellar model, and chemical species included in the calculation.

That last condition is critical. If a retrieval compares “methane” with “no methane” but excludes ethane, other hydrocarbons, haze, stellar contamination, or alternative temperature profiles, its significance describes a restricted model contest. It does not establish that methane is the only possible absorber.

This is JWST retrieval degeneracy in practical terms: different atmospheric compositions can produce nearly indistinguishable spectra, especially when the signal is small or the resolution is limited.

Three JWST case studies

WASP-39 b: sulfur dioxide is a photochemistry lesson

WASP-39 b is a large, inflated gas giant with frequent transits and a strong atmospheric signal. JWST observations spanning roughly 0.6–5.3 micrometers have supported detections of water vapor, carbon dioxide, carbon monoxide, sulfur dioxide, sodium, and potassium.

The sulfur dioxide result is especially instructive. On WASP-39 b, ultraviolet radiation can break apart hydrogen sulfide and drive reactions that produce SO₂. The molecule shows that stellar radiation can push an atmosphere far from thermochemical equilibrium. It does not indicate biology.

Interpreting that chemistry requires the host star’s ultraviolet spectrum, photolysis rates, vertical mixing, temperature-pressure structure, gas-phase reaction networks, and aerosol chemistry. An equilibrium-only model would miss important parts of the atmosphere.

SO₂ on WASP-39 b is evidence of active photochemistry, not evidence of life.

The target also illustrates a less glamorous data problem. A combined analysis can offer broad wavelength coverage while parts of a spectrum—such as saturated regions in NIRSpec/PRISM data—remain less reliable. More wavelength coverage is useful only when the relevant pixels are measured and calibrated well.

K2-18 b: a disputed molecule remains disputed

K2-18 b became a major target after near-infrared observations supported methane and raised questions about carbon dioxide. Later MIRI observations produced a possible mid-infrared signal consistent with dimethyl sulfide (DMS) and/or dimethyl disulfide (DMDS), compounds associated with biological activity on Earth.

A 2025 analysis reported an approximately 3.4-sigma preference for models containing DMS and/or DMDS, with a candidate abundance near 10 parts per million by volume or higher. The result appeared in the context of the ongoing K2-18 b reanalysis debate, following the original 2023 reporting of carbon-bearing molecules and subsequent 2025 studies that revisited the spectrum and model assumptions.

The number sounds precise, but sigma is conditional. It changes when researchers expand the model space. Later analyses and alternative data treatments reported that:

• The DMS/DMDS preference weakened or disappeared under some reductions.
• Ethane and other methyl-bearing hydrocarbons could fit the spectrum comparably well.
• Temperature structures and cloud assumptions changed the inferred abundances.
• An oxygen-poor mini-Neptune atmosphere could explain the observations without requiring a habitable ocean or biology.

A large reanalysis of NIRISS and NIRSpec data examined dozens of data treatments and hundreds of retrievals. It supported methane at roughly 4 sigma within that analysis, but that figure is not a universal consensus measurement. It depends on the selected data, baseline treatment, and competing atmospheric models. The same analysis did not establish reliable evidence for DMS or carbon dioxide.

A compact example shows why model expansion matters:

The observation is still valuable. MIRI opens mid-infrared bands unavailable to most near-infrared observations. But it is not an automatic confirmation channel. Its lower resolution and throughput, along with sensitivity to extraction and binning choices, can make a weak feature harder to interpret.

The responsible description of DMS on K2-18 b is therefore a contested candidate interpretation, not a confirmed molecular detection and certainly not a discovery of life.

LP 791-18 c and TRAPPIST-1: context changes the meaning

LP 791-18 c offers a useful comparison. JWST/NIRSpec PRISM observations covering approximately 0.7–5.4 micrometers indicated methane, strong haze scattering, and no clearly discernible carbon dioxide absorption. The atmosphere was estimated to have roughly 246–415 times solar metallicity, with a carbon dioxide-to-methane ratio below 0.07 at 2 sigma.

The important point is not the exact abundance estimate. Methane appears in a high-metallicity, hazy sub-Neptune atmosphere without requiring biology. The same molecule can mean something very different on a small rocky planet with a thin, oxidizing atmosphere.

TRAPPIST-1 presents the opposite measurement problem. Its planets are roughly Earth-sized, but the atmospheric signals are tiny. Star spots, faculae, flares, and time-variable surface regions can imprint spectral patterns that resemble planetary absorption.

JWST observations of TRAPPIST-1 planets b, c, d, and e have not established a robust secondary atmosphere for the most relevant rocky targets. Thick hydrogen atmospheres have been ruled out for d and e, while observations of b and c found no evidence for thick atmospheres. Stronger conclusions may require many more transits.

A flat transmission spectrum does not automatically mean “no atmosphere.” It can indicate a thin atmosphere, a high-mean-molecular-weight atmosphere, clouds, haze, stellar contamination, or simply insufficient signal.

A practical validation workflow

The six commonly discussed validation levels are best treated as one analyst workflow rather than six isolated hurdles.

1. Confirm the feature. Check whether the deviation appears at the expected wavelength and survives reasonable choices of binning, baseline, detector treatment, and extraction. At this stage, call it a feature—not a molecule.

2. Compare a complete retrieval basis. Test the proposed molecule against all chemically plausible absorbers, including hydrocarbons, haze, clouds, and stellar-contamination terms. Report the evidence difference between models, not just the best-fit abundance.

3. Seek independent consistency. Compare instruments, observing visits, reductions, and extraction methods. Agreement is meaningful only when the analyses do not share the same hidden calibration assumption.

4. Check chemistry against the host star. Photochemical models should use the star’s ultraviolet and X-ray environment and include vertical mixing, escape, aerosols, and realistic temperature profiles. A gas that is stable in one stellar environment may be short-lived in another.

5. Establish planetary architecture. A methane-rich rocky planet, hydrogen-rich mini-Neptune, high-metallicity sub-Neptune, and cloudy ocean-world candidate have different atmospheric chemistry and different plausible abiotic sources.

6. Test the biological explanation. A biosignature requires a combination of gases that is difficult to maintain through known nonbiological pathways. Methane becomes more informative alongside a chemically incompatible oxidant, for example, but geology, photochemistry, escape, and surface-atmosphere exchange still need to be modeled.

Non-detections are part of this process. The absence of carbon dioxide on LP 791-18 c, the lack of thick atmospheres on some TRAPPIST-1 planets, and failed attempts to recover particular molecules can eliminate entire atmospheric scenarios.

For practical analysis, researchers should report preferred abundances alongside upper limits, run injection-recovery tests, preserve multiple reductions, and state exactly which molecules and nuisance parameters were allowed. Those details often matter more than an attractive headline sigma value.

FAQ

Has JWST detected a biosignature?

No JWST observation has confirmed life or produced a biosignature that has survived all major alternative explanations. JWST has detected atmospheric molecules and photochemical products, but those findings are not equivalent to detecting biology.

Is methane evidence of life on an exoplanet?

Not by itself. Methane can come from biology, serpentinization, volcanic or interior processes, photochemistry, and hydrogen-rich atmospheres. Its significance depends on the surrounding gases and whether abiotic sources can sustain the measured abundance.

Is DMS on K2-18 b confirmed?

No. DMS and DMDS were reported as possible explanations for a MIRI spectrum, but later analyses found that alternative data treatments and molecules such as ethane and other hydrocarbons could fit the observations. The claim remains disputed.

What does sulfur dioxide on WASP-39 b mean?

It demonstrates active stellar-radiation-driven photochemistry. SO₂ is scientifically important because it shows that atmospheric disequilibrium can arise without biology; it is not, by itself, a sign of life.