Background and justification of selected algorithm

Over the course of the different L-band radiometers such as SMOS and SMAP, several algorithms were proposed and implemented for the retrieval of the SIT. [Tian-Kunze et al., 2014] presented an algorithm which is used today in the ESA SMOS official SIT product. While it is based on a combination of physical models, it provides similar SIT values to the empirical formulation from [Huntemann et al., 2014], although, extending the range of retrievable SIT values up to 1 m of ice thickness for growing sea ice in freeze-up conditions. The algorithm from [Tian-Kunze et al., 2014] uses SMOS incidence angles close to nadir and thus no polarization information. While the algorithm from [Huntemann et al., 2014] uses SMOS incidence angles observations at incidence angles from 40° to 50°, which makes it suitable for an adaption to CIMR with an incidence angle of 52° at L-band.

This algorithm used for CIMR SIT retrieval is using the empirical variant of [Huntemann et al., 2014] and [Paţilea et al., 2019], but also extended to higher ice thicknesses at increased uncertainty. Technically this is achieved by considering a certain background ice thickness with a very low weight. Once the sensitivity of brightness temperatures to thickness decreases in the thick ice case, the relative weight on the background value is increasing. This is done to avoid the retrieval of unrealistically high ice thicknesses without artificially constraining at a fixed boundary of 50 cm like in [Huntemann et al., 2014] and [Paţilea et al., 2019]. Another difference to the original formulation is the use of horizontal and vertical polarization instead of intensity and polarization difference. The motivation is that the horizontal and vertical polarization are directly obtained from the instrument without transformation. For illustration, the intensity and polarization difference are still serving as comparisons for the \(T_{b,v}\) and \(T_{b,h}\) based retrieval. The entire algorithm is described in detail in Sea Ice Thickness Retrieval Algorithm.