Scientists Uncover Flaws in Common Models for Studying Exoplanet Atmospheres

New research reveals that simple computer models widely used to study exoplanet atmospheres often produce inaccurate predictions about their temperatures and chemical compositions. A team of scientists, led by Dr. Yassin Jaziri from the Laboratory for Atmospheres, Environments, and Space Observations (LATMOS) in Paris, demonstrated that more advanced three-dimensional (3D) models provide significantly better insights into the true nature of these distant worlds.

By using a 3D Global Climate Model (GCM), the team analyzed the atmosphere of GJ 1214 b, an exoplanet located about 48 light-years away. Their work revealed major differences between the results of simple 1D models and the more sophisticated 3D approach. For example, the 3D model identified a clear chemical signature of carbon dioxide (CO₂) that the 1D models failed to detect. This was because the 1D models only consider a narrow region around the planet’s edge and cannot account for variations between the hot day side and cooler night side.

Figure 1 : 3D temperature representation of the atmosphere of the exoplanet GJ 1214 b, obtained from a GCM model by Charnay et al., 2015. The hottest point, shown in red, exceeds 665.4 K, which corresponds to 392.3°C or 738.1°F.

These findings highlight the limitations of traditional 1D models, which have long been used to estimate the molecular makeup and temperature profiles of exoplanets. While these simpler models are effective for detecting certain molecules, their results can differ significantly from one another due to the uncertainties between models. Therefore, it is preferable to cross-compare the results of multiple 1D models rather than relying on a single 1D model for accurate predictions.

Exoplanet atmospheres are not uniform. They have complex, three-dimensional structures with significant temperature and chemical variations across their surfaces and layers. Understanding these features is essential for learning about a planet’s climate, its formation, and whether it could support life.

To investigate these complexities, telescopes like the James Webb Space Telescope (JWST) collect light that passes through a planet’s atmosphere during transit events, when the planet crosses in front of its star. By analyzing this light, scientists can create spectra that reveal key details about the atmosphere’s molecular composition and temperature structure.

Dr. Jaziri’s team used their GCM to produce synthetic spectra for GJ 1214 b, simulating what telescopes like JWST would observe. Their analysis showed that the 3D model accurately captured how molecules like CO₂ vary across the planet, while the 1D model failed to do so. The team also compared results from multiple 1D models, finding that their predictions were inconsistent and often divergent.

Looking ahead, the research suggests that as telescopes become more powerful, scientists will need to adopt more sophisticated models to match the complexity of the data. With upcoming missions like the ARIEL telescope, these improved methods could provide unprecedented insights into the atmospheres of planets beyond our solar system, offering clues about their potential for habitability.

Figure 2 : Top panel: Synthetic spectra produced by 1D and 3D numerical models simulating observations by the JWST telescope of the exoplanet GJ 1214 b, at infrared wavelengths between approximately 1 and 12 microns. The y-axis represents the "transit depth," which is the ratio of the flux measured by the telescope when the planet passes in front of its star to the flux of the star alone (without the planet). Bottom panel: The residuals—representing the difference between the spectra generated by the 3D and 1D models—are shown in units of parts per million (ppm), which quantify the variation in intensity between the two models. The residuals highlight key differences: for instance, the CO₂ chemical signature between 4 and 5 microns appears only in the 3D model. This can be explained by the fact that CO₂ is particularly abundant on the hot day side of the planet, which is captured in the 3D model. In contrast, the 1D model here simulates the spectrum only along the planet's limb—thin region around the edge of the planet. This limitation highlights the fact that the 1D model does not account for the thermal or chemical variations between the day and night sides of the planet.

This article made use of the following publications:

Jaziri et al. 2024, A&A, 684, A25

Charnay et al. 2015, ApJL, 813, L1