Modeling atmospheric escape from exoplanets

Atmospheric escape, where exoplanets lose their atmospheres to space, plays a critical role in planetary evolution, potentially stripping gaseous planets of a large portion of their envelopes, or eroding rocky planets' atmospheres, affecting their habitability. Theoretical models of escape often depend on poorly constrained parameters, such as the stellar XUV flux, and observations have mainly focused on a limited set of spectral lines. This thesis aims to advance the field through new modeling approaches, as well as identification of additional observational tracers and promising exoplanet targets. We develop an open-source code called "sunbather", which combines a Parker wind framework with the photoionization code Cloudy. We propose a new Bayesian approach to fitting observations, and show how sunbather can be used in this way to fit helium transit data of various exoplanets. This allows us to constrain their mass-loss rates, which is the parameter of highest interest. We also identify new spectral lines tracing atmospheric escape, by simulating transit spectra across the FUV to NIR range for various planetary systems. Targeting these lines observationally would enable deeper insights into upper-atmosphere composition and dynamics. We then study how atmospheric metallicity affects the escape process. Our simulations show that higher metallicity enhances radiative cooling, reduces mass-loss rates and alters observable spectral features. Finally, we create a database of synthetic spectra for nearly all transiting exoplanets, which can be used to find optimal targets for future observations.