| The accuracy of passive-microwave snow depth retrievals is potentially subject to complex error sources. These error sources can be grouped into three categories: (a) errors due to variations in snow physical properties, (b) errors due to variations in the underlying surface, and (c) errors due to variations in the atmosphere. This study investigates the impacts of these error sources on the microwave signature of snow on sea ice and suggests improvements to the retrieval algorithm. This is done using various tools including a thermal microwave emission model of snow, microwave radiometer measurements from aircraft, radiance measurements from satellite instruments, and field observations of snow, and ice physical properties. The results indicate that the 19/10 GHz combination has the ability to retrieve larger snow depths with the additional benefit of being relatively insensitive to variations in atmospheric conditions. In addition, the 89 GHz channel may be used to identify new snow fall events. Model analysis of measured brightness temperatures suggest that the current snow depth retrieval algorithm is sufficient for areas of smooth first-year sea ice whereas new algorithm coefficients are needed for rough first-year sea ice.; Despite these potential errors, comparisons of satellite derived snow depth distributions with in-situ distributions have shown good agreement on regional scales. Therefore, satellite derived snow depth data are used in a data assimilation approach to investigate how changes in snow depth affect the Southern Ocean sea ice cover. Precipitation rates are varied from 0.0 to a doubling of the present day precipitation. Initially, sea ice volume decreases with increasing precipitation rate multiplying factor (PRMF) due to the insulation effects of a deeper snow layer. The turning point at which the insulation effect becomes balanced by the snow to ice conversion effect ranges from PRMF=0.50 to PRMF=0.75 depending upon the snow thermal conductivity and density. This suggests that with present-day precipitation rates the snow effect on Southern Ocean sea ice is dominated by snow-ice formation rather than the snow's insulation. |
