Does TRAPPIST-1e Have an Atmosphere? Constraints from Space Telescope Observations

Among all currently known exoplanets, the planetary system orbiting TRAPPIST-1 has long been regarded as one of the most promising targets for studying Earth-like atmospheres. Within this system, TRAPPIST-1e stands out: its size and mass are close to Earth, and it lies within the star's habitable zone, making it a strong candidate for maintaining liquid water on its surface. However, this possibility depends on a crucial prerequisite—the presence of an atmosphere. Without an atmosphere, stable liquid water cannot exist, and the concept of habitability becomes meaningless. Recent observations using the James Webb Space Telescope (JWST) were designed to address this fundamental yet challenging question.

TRAPPIST-1 system illustration, with the two black dots in the foreground representing planets（Image source：NASA/JPL-Caltech，CC0 1.0 ）

Illustration of TRAPPIST-1e（Image source：NASA/JPL-Caltech，CC0 1.0 ）

The study relies on transit spectroscopy, a method that observes how starlight changes as a planet passes in front of its host star. If the planet has an atmosphere containing molecules such as carbon dioxide or methane, these gases will absorb light at specific wavelengths, leaving identifiable signatures in the spectrum. The research team observed four transits of TRAPPIST-1e, a dataset theoretically sufficient to begin constraining the nature of its atmosphere.

However, the observations revealed an unexpected complication. The spectra from the four transits differed significantly from one another. Such variation cannot be explained by changes in a single planetary atmosphere, pointing instead to the host star as the dominant source of noise. TRAPPIST-1 is an ultracool M-dwarf star with a highly active surface, covered in starspots and bright regions, and prone to flares. When the planet transits across different regions of the stellar surface, the resulting spectra can vary substantially, creating signals that mimic or obscure atmospheric features.

Four separate transit observations of TRAPPIST-1e, with the bottom panel showing all four overlaid. The x-axis represents wavelength, and the y-axis shows the transit depth, indicating how much starlight is blocked by the planet. The first two spectra are more similar to each other, suggesting relatively weaker stellar contamination, while the latter two are clearly more strongly affected by stellar activity（Image source：Glidden A et al. (2025)， CC BY 4.0 ）

Among the four observations, the first two were relatively less affected by stellar activity, while the latter two were heavily contaminated, including one case with a stellar flare occurring during transit. This variability means that the data cannot simply be combined without careful correction, as doing so risks misinterpreting stellar effects as planetary atmospheric signals.

To avoid overinterpretation, the researchers adopted a deliberately conservative approach. Instead of immediately attempting to identify specific atmospheric compositions, they first tested whether the data could be explained by a flat spectrum—one with no significant variation across wavelengths. Such a spectrum would correspond either to a planet with no atmosphere or to one with a high–mean molecular weight atmosphere whose spectral features are too weak to detect. After correcting for stellar contamination, the combined data were indeed consistent with a flat spectrum, although small residual fluctuations remained. These could arise from incomplete stellar correction, instrumental noise, or weak atmospheric signals. At present, the data cannot distinguish between a completely airless planet and one with a spectrally muted atmosphere.

Even the absence of clear spectral features provides meaningful constraints. If TRAPPIST-1e possessed a hydrogen-dominated atmosphere, it would be highly extended, producing strong and easily detectable absorption features. Given the sensitivity of JWST, molecules such as carbon dioxide or methane should have been observed. Their absence effectively rules out hydrogen-rich atmospheres, consistent with earlier results from the Hubble Space Telescope, but now with tighter constraints.

Comparison between the observed data of TRAPPIST-1e and a range of hypothetical atmospheric models. The semi-transparent colored circles represent individual transit observations; the gray circles show the combined result of the first two, relatively cleaner observations; the black circles represent the final spectrum combining all four observations after correcting for stellar contamination; the dark red star symbols indicate earlier observations from the Hubble Space Telescope for comparison; the colored curves correspond to different assumed planetary atmosphere models（Image source：Glidden A et al. (2025)， CC BY 4.0 ）

The study further explored a wide range of atmospheric compositions. Models including even small amounts of hydrogen were found to produce spectral features that should have been detectable. Since such features are not observed, nearly all hydrogen-containing atmospheres can be excluded. In contrast, atmospheres dominated by heavier molecules—particularly nitrogen—remain viable. These atmospheres can produce spectral signatures so weak that they remain undetectable with current data. Similarly, dense carbon dioxide–dominated atmospheres like those of Venus or Mars are not strongly supported under typical pressure conditions, especially if the atmosphere is relatively clear.

Interestingly, some models suggest that a nitrogen-dominated atmosphere with trace amounts of methane could better reproduce subtle variations in the observed spectra. These variations align with wavelengths where methane absorption might occur. However, the signal is extremely weak and far below the threshold required for a definitive detection. It may equally result from residual stellar contamination or instrumental effects. The researchers emphasize that this should be viewed only as a tentative hint, not as evidence of methane or any biological process.

Spectral fitting results for the stellar-contamination-corrected transmission spectrum of TRAPPIST-1e. Top panel: best-fit forward atmospheric models under three different partial pressure combinations of nitrogen (N2) and methane (CH4) (solid dark purple, dashed purple, and dashed cyan lines), compared with a flat spectrum (black dashed line). Bottom panel: results from atmospheric retrieval incorporating a Gaussian process, comparing a centered log-ratio prior (blue line) and a log-uniform prior with a background gas component (gray line), both shown against a flat spectrum (black dashed line). All models are presented at the same spectral resolution as the observations. The figure also marks wavelength regions that may correspond to methane absorption bands（Image source：Glidden A et al. (2025)， CC BY 4.0 ）

To address the entanglement of stellar and planetary signals, the study employed statistical methods that simultaneously model both contributions. These analyses indicate that two scenarios remain equally consistent with the data: a completely airless rocky planet, or a planet enveloped by a high–mean molecular weight atmosphere with minimal spectral features .

The significance of this work lies not in a definitive detection, but in the boundaries it establishes. TRAPPIST-1e is highly unlikely to possess a thick hydrogen atmosphere, and it does not resemble a Venus-like world dominated by dense carbon dioxide under typical conditions. The planet is therefore either a bare rocky body or one covered by a relatively heavy, spectrally quiet atmosphere, possibly rich in nitrogen.

For terrestrial planets orbiting active red dwarf stars, stellar variability emerges as a central obstacle in atmospheric characterization. Future observations—both more numerous and better designed to mitigate stellar contamination—will be essential for determining whether TRAPPIST-1e truly has an atmosphere and, if so, uncovering its composition.

Author: Shui-Ye You

Reference:

Glidden A et al. (2025). JWST-TST DREAMS: Secondary Atmosphere Constraints for the Habitable Zone Planet TRAPPIST-1 e. The Astrophysical Journal Letters.

（Paid content. Unauthorized reproduction or use is prohibited.）