Successful retrieval of surface properties from space is hampered by the presence of atmospheric aerosol particles that contribute significantly to the measured signal. Our ability to obtain reliable information about surface properties depends to a large extent on how well we can account for the influence of aerosols. The problem is complicated by the fact that aerosols often consist of a multicomponent mixture of particles with different chemical compositions and different affinities to water. For example, to predict how the optical properties of such particles change with increasing humidity, we must make assumptions about how the particles grow, change their refractive indices, and mix as a function of humidity. We discuss possible strategies for reliable atmospheric correction over dark as well as bright surfaces. Also, we discuss the role of realistic simulations of the radiative transfer process in the coupled atmosphere-ocean system in the solution of the inverse problem required to retrieve surface properties is also discussed.