In this work we investigate the effects of the environment on the X-ray photoelectron spectra of hydrogen chloride and the chloride ions adsorbed on ice surfaces, as well as of chloride ions in water droplets. In our approach, we combine a density functional theory (DFT) description of the ice surface with that of the halogen species with the recently developed relativistic core- valence separation equation of motion coupled cluster (CVS-EOM-IP-CCSD) via the frozen density embedding formalism (FDE), to determine the K and L_1,2,3 edges of chlorine. Our calculations, which incorporate temperature effects through snapshots from classical molecular dynamics simulations, are shown to reproduce experimental trends in the change of the core binding energies for Cl^- upon moving from a liquid (water droplets) to an interfacial (ice quasi-liquid layer) environment. Our simulations yield water valence band binding energies in good agreement with experiment, and that vary little between the droplets and the ice surface. For the halide core binding energies there is an overall trend of overestimating experimental values, though good agreement between theory and experiment is found for Cl^- in water droplets and on ice. For HCl on the other hand there are significant discrepancies between experimental and calculated core binding energies when we consider structural models which maintain the H-Cl bond more or less intact. An analysis of models that allow for pre-dissociated and dissociated structures suggests that experimentally observed chemical shifts in binding energies between Cl^- and HCl would require that H⁺ (in the form of H₃O⁺) and Cl^- are separated by roughly 4-6 Å.