Lower frequency microwave radar observations can help to measure the properties of snow cover on land by providing information about the soil-snow boundary condition. In this theoretical study, we examine the sensitivities of microwave radar measurements to soil and snow characteristics, and we compare a simple model with previously published data. Depending on the surface roughness, copolarized ratios or single polarization time ratios of radar backscattering may be affected only by the incidence angle and the dielectric contrast at the soil-snow boundary. These measurements can reduce the number of unknowns in any corresponding higher-frequency observations that are also affected by the lower boundary condition. The copolarized ratio is sensitive, first of all, to the snow density, which offers the possibility of measuring this parameter directly if a separate measurement of the soil temperature from a passive microwave system is also available. The thermal insulation provided by snow cover can have a powerful effect on the soil-snow boundary by altering the soil temperature and therefore changing the dielectric contrast. Because the microwave response of snow cover is so sensitive to the conditions of the underlying frozen soil, it is important for future ground truth campaigns that measure snow conditions to also collect data on the temperature, water content, and texture of the soil.