Thousands of planets orbit their stars closer than Mercury orbits the Sun, and most of them are "sub-Neptunes"—worlds two to four times Earth's size with thick, volatile-rich atmospheres whose compositions remain unknown. This matters because astronomers cannot tell whether a sub-Neptune's atmosphere today reflects how it formed or how it has changed over billions of years. Extreme heat from the host star drives powerful hydrodynamic outflows that strip away atmospheric gases. Lighter species like hydrogen escape faster than heavier ones, a process called fractionation that alters the atmosphere's chemical fingerprint—for example, its carbon-to-oxygen ratio. Without models that track this compositional evolution, interpreting spectra from telescopes like JWST and the future ARIEL mission will be ambiguous. This project will build the first multi-species atmospheric escape models for the entire sub-Neptune population, simulating how their compositions change from birth to the present day. The work is fundamental science: it addresses a core gap in exoplanet theory rather than any immediate practical application. But similar modelling of atmospheric escape has already explained how the Solar System's terrestrial planets lost their primordial atmospheres, and understanding sub-Neptune evolution could eventually help identify which exoplanets might have conditions suitable for life.
View original technical description
Planets closer to their host star than Mercury are known to be abundant, with most stars hosting at least one. In this population, “sub-Neptune” planets (radii between ~2-4x the Earth’s) are the most abundant, yet they are the least understood. Their densities imply they host a large atmosphere dominated by volatiles; however, modelling degeneracies mean we cannot infer their compositions from measurements of their mass and radius alone. Observations of their atmospheres by HST/JWST and in the future by the ELTs and ARIEL are already opening up the possibility of constraining their composition. However, the vast majority of the exoplanets where we can and will study their atmospheres are billions of years old. Thus, understanding evolutionary processes that can change a planet's atmospheric composition is critical. Due to the proximity to their host stars, the extreme heating of these planets' volatile atmospheres can drive mass-loss via powerful hydrodynamic outflows. These outflows are theorised to drive the bulk properties of the close-in exoplanet population, sculpting it into the population we see today after billions of years of evolution. However, these outflows can cause the atmospheric composition to evolve. Despite these outflows being hydrodynamic in nature (where the average collisional mean free path is shorter than the atmospheric scale height), more frequent collisions between lighter species, such as hydrogen, mean heavier species may not always be perfectly coupled to the outflow. This ``fractionation’’ process is known to have shaped the composition of the Solar System terrestrial planets early in their lives; however, due to the complexity that arises from the extreme heating and ionization in close-in planets, it remains poorly explored. Exploratory works have demonstrated that atmospheric escape not only drives the evolution of the close-in exoplanets' bulk properties, it also drives their compositional evolution, where quantities like the C/O ratio are expected to evolve. Using multi-species atmospheric escape models, we will simulate the detailed composition of escaping atmospheres for the population of sub-Neptunes for the first time. By combining these species-dependent mass-loss rates into an evolutionary model for the planet, we can follow an exoplanet's compositional evolution from birth to when we observe it today. Without our modelling, interpreting exoplanet spectra will be fraught with difficulty as it will be unclear whether trends in composition between different planets and across exoplanetary systems are imprints of formation or evolutionary processes. Our work will directly address STFC’s challenges B3 and B5.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know