Satellite-based atmospheric remote sensing aims at deriving the properties of trace gases, aerosols and clouds, as well as surface parameters from the measured top-of-atmosphere spectral radiance and reflectance. This requires, besides high quality spectra, an accurate modelling of the radiative transfer of solar radiation through the atmosphere to the sensor (forward model) and methods to derive the constituent properties from the measured top-of-atmosphere spectra (inversion methods). Many trace gases have structured absorption spectra in the UV-VIS spectral range serving as the starting point for determining their abundance by applying Differential Optical Absorption Spectroscopy (DOAS) or similar methods. In the UV-VIS-NIR and SWIR spectral regions the solar radiation is strongly scattered by clouds and aerosols. Therefore the presence of clouds and aerosol particles and their properties can also be inferred from the outgoing radiance measured by space-based instruments. Contrary to the forward model, the inversion methods allow to derive characteristics of the atmospheric state based on the measured quantities. A common product of the inversion of satellite measurements in limb, nadir or occultation geometry are total columns or height-resolved profiles of trace gas concentrations and aerosol parameters. Retrieving trace gas amounts in the troposphere constitutes a specific challenge. SCIAMACHY’s unique limb/nadir matching capability provides access to tropospheric columns by combining total columns obtained from nadir geometry with simultaneously measured stratospheric columns obtained from limb geometry.