In this paper, Zieba and colleagues report observation results from the James Webb Space Telescope (JWST) on the exoplanet TRAPPIST-1 c, utilizing the MIRI instrument. The results of these observations rule out thick carbon dioxide atmospheres, and could otherwise suggest the absence of an atmosphere.

Several planets orbit the dwarf star TRAPPIST-1, some of which would be located inside its habitable zone. Recent observations on one of these planets, TRAPPIST-1 b, showed that it’s probably a bare rock lacking an atmosphere. Here, Zieba and colleagues report new observations from JWST that targeted another planet in the same system, TRAPPIST-1 c, using the MIRI mid-infrared instrument.

As a reference, the TRAPPIST-1 system is shown on Figure 1 below. The distance between the star and its seven Earth-sized planets is much smaller than in our own planetary system. But given the fact that M-dwarf stars are much colder than Sun-like stars, the habitable zone is also smaller, which implies that three of the planets (e, f and g) are still located in the circumstellar zone where water could be present in its liquid state.

We still know very little about the composition of terrestrial exoplanet atmospheres. Atmospheric composition depends on many factors, such as the initial inventory of volatiles, outgassing resulting from volcanism, and atmospheric escape—a process to which planets around M dwarf stars, such as TRAPPIST-1, are particularly vulnerable because of the hostile space weather around these objects.

Previous observations were made on similar rocky planets, such as LHS 3844 b, GJ 1252 b and TRAPPIST-1 b. These observations revealed temperatures that suggested the absence of heat redistribution and CO\(_2\) absorption. In other words, it is unlikely that these exoplanets could harbor life as we know it.

This time, Zieba and colleagues investigated four eclipses of TRAPPIST-1 c that happened in late 2022 using the MIRI instrument. Light curves of TRAPPIST-1 were extracted, then fitted onto an eclipse model with a range of parameters (Figure 2).

From the data, the authors derived a brightness temperature of 380K for TRAPPIST-1 c. They note that TRAPPIST-1 b was found to have a temperature of 503K, and that the temperatures of TRAPPIST-1 planets are about 500K cooler than other small rocky planets observed so far—which is obviously a good sign when searching for life.

Figure 3 comes from the paper’s Supplementary Material, and shows the various small exoplanets for which we have measured infrared emission. The vertical axis shows exoplanet temperatures, normalized by the maximum equilibrium temperature (i.e., if no atmosphere was present), while the horizontal axis shows planetary radii in Earth radius units. We can see that the two planets in the TRAPPIST-1 system have radii that are close to that of Earth, and that the one observed by Zieba and colleagues indeed has a temperature significantly lower, compared to its maximum equilibrium temperature, than the ones of other previously observed exoplanets. In other words, this indicates that we are getting closer to detecting planets with similar parameters to Earth’s, including the presence of an atmosphere.

Relatedly, Zieba et al. note that TRAPPIST-1 c is the first exoplanet with measured thermal emission (380K) comparable to that of the inner planets of the Solar System (440K for Mercury and 227K for Venus). The measured temperature of TRAPPIST-1 c is also intermediate between two limiting cases: zero heat distribution (a fully absorptive bare rock, which would yield 430K), and global heat redistribution (as expected for a thick atmosphere, yielding 340K). To determine what kind of atmosphere would be compatible with the data they proceeded to compare those with several models.

To further explore which atmosphere are consistent with the observations of TRAPPIST-1 c, Zieba and colleagues compared them against several atmosphere models. They first varied surface pressure and O\(_2\)/CO\(_2\) composition (Figure 4). With the measured eclipse depth, we can however rule out thick atmospheres (darkened tiles in the figure).

Similarly, Zieba and colleagues compared the observations with models of Venus-like atmospheres, as the insolation of TRAPPIST-1 c is just 8% greater than that of Venus, making it possible for both planets to have similar chemistries. Further analysis however showed that the observations however disfavoured these models as well.

Lastly, the authors compared JWST’s observations of the measured flux against bare-rock models for a variety of surface composition, and found that all bare-rock surface models were consistent with the data. Put differently, TRAPPIST-1 c probably has a thin atmosphere—or no atmosphere at all.

In one final investigation on the eclipse data of TRAPPIST-1 c, Zieba et al. compared the measured flux against a grid of atmospheric evolution models to further determine the formation history of the planetary system. Taking as parameters a range of initial water inventories and extreme ultraviolet saturation fractions for the host star, their analysis reveals that TRAPPIST-1 c probably formed with a low initial water abundance.

This last result also suggests that rocky planets around M-dwarfs, such as TRAPPIST-1, may form with a smaller volatile inventory, or experience more atmospheric loss than planets around Sun-like stars. Zieba and colleagues conclude their analysis by suggesting that further assessments are necessary to determine whether this is a typical outcome for planets around M-dwarf stars.