Improving l-band radiation model and representation of small-scale variability to simulate brightness temperature of sea ice

The L-band microwave has the ability for penetrating snow and sea ice, hence the potential for data retrieval of snow depth and sea ice thickness. The L-band radiometer Soil Moisture and Ocean Salinity has been applied to retrieve sea ice parameters using a L-band radiation model. Here, we improve the L-band radiation model in two ways. First, instead of a single layer, we reformulate the model to consider multiple layers for sea ice and snow to better characterize the radiative effect of vertical temperature and salinity profile within the sea ice and the snow cover. Second, we incorporate small-scale sea ice variability, sea ice leads, which are not distinguishable with the L-band satellite data but have profound impacts on the simulated brightness temperature. Small-scale features of the sea ice cover such as leads can significantly lower the overall brightness temperature (T_B ). The results show that the improved radiation model can correct sea ice type dependent biases of T_B (R² from 0.05 to 0.72) and compensate for the overestimation of T_B when (refrozen) leads are ignored (R² from 0.51 to 0.62). The improved radiation model could serve as the basis of the retrieval of sea ice thickness or snow depth.