The results from the Kepler space observatory have revealed a wealth of planets with small sizes and small orbital separations, referred to as “super-Earths” or “sub-Neptunes”. This thesis will focus on such planets.
Atmospheres of super-Earths/sub-Neptunes are vulnerable to hydrodynamic escape due to the proximity to their host stars. I begin by comparing two mechanisms by which this can occur: photoevaporation, which is caused by high energy radiation from the host star; and core-powered mass-loss, which relies on the stellar bolometric luminosity in conjunction with the cooling luminosity from the planetary core. I construct a robust statistical test that can be performed on future exoplanet surveys in order to determine which of these mechanisms is dominant.
I then go on to show that if one takes the population of exoplanets observed today, one can rewind the clock on atmospheric evolution to infer the planet properties at birth. I perform this with photoevaporation models to place constraints on planetary properties that are currently unobservable with standard techniques. I find that a typical super-Earth/sub-Neptune has a core mass of ∼ 4M⊕, a core composition consistent with that of Earth and an initial atmospheric mass fraction after protoplanetary disc dispersal of ∼ 2%.
This latter result highlights a discrepancy between atmospheric mass-loss and gas accretion models, the latter of which predicts initial atmospheric masses ≳ 10%. I perform sophisticated evolution models of planets immersed in protoplanetary discs that self-consistently model gas accretion and mass-loss induced by rapid disc dispersal. I find that this mechanism efficiently removes atmospheric mass and provides a robust solution to the aforementioned discrepancy. This mass-loss phase induces rapid contraction and advection of high entropy material into the escaping atmosphere. This process, which I find is strongly controlled by the disc dispersal rate, produces a planet that has prematurely cooled, given its age.