Nina Maaß

Improved retrieval of sea ice thickness from SMOS and CryoSat-2

We investigate the potential of a synergetic combination of data from ESAs SMOS and CryoSat-2 mission for sea ice thickness retrieval. SMOS and CryoSat-2 provide complementary information because of their different spatio-temporal sampling and resolution, and because of the complementary uncertainty due to the fundamental difference of the radiometric and altimetric measurement principle. The main limitations of the ice thickness retrieval depend on the emission e-folding depth and the vertical resolution of the effective radar pulse-length, respectively. It is shown that the combination of SMOS and CryoSat-2 considerably reduces the uncertainty with respect to the products derived from the single sensors. The RMS error is reduced from 76 to 66 cm and the squared correlation coefficient increases from 0.47 to 0.61 in comparison to validation data of NASAs Operation IceBridge campaign, 2013. Furthermore, we demonstrate the applicability of the Optimal Interpolation method for the generation of a combined product based on weekly CryoSat-2 averages.

Remote sensing of sea ice thickness using SMOS data

In this study, we investigate how ice conditions influence sea ice thickness retrieval from Soil Moisture and Ocean Salinity (SMOS) data. Additionally, we validate sea ice thicknesses retrieved in the Baltic Sea, and we examine whether it is possible to retrieve snow thickness over thick multi-year ice from SMOS data. The European Space Agency’s SMOS mission is the first passive microwave radiometer that measures at a frequency of 1.4 GHz in the L-band. Using L-band, the large penetration depth in sea ice allows us to extract information on the thickness of thin sea ice, as opposed to other microwave satellites. The accuracy of the retrieval and the maximum retrievable ice thickness depend on the ice conditions, which are mainly characterised by ice temperature Tice and salinity Sice. For typical Arctic conditions, ice thickness is retrievable up to about 50 cm. Here, we use a radiation model to investigate the sensitivity of the ice thickness retrieval to ice temperature and salinity. According to the model, an ice temperature change of 1◦C and an ice thickness change of 1 – 3 cm cause approximately the same signal in the observed brightness temperature (for Sice= 8 g/kg and ice thicknesses up to 45 cm). For typical salinities of Arctic thin ice, an ice salinity change of 1 g/kg has the same impact on brightness temperature as an ice thickness change of 0.5 – 6 cm (for Tice= -7 ◦C). Thus, the brightness temperature development above growing thin sea ice mainly indicates the ice thickness variation, and not the variations of ice temperature and salinity. We investigate the impact of a snow layer on the ice thickness retrieval in L-band. The self-emittance of snow causes brightness temperatures to be higher for snow-covered than for snow-free ice. For typical Arctic conditions, brightness temperatures increase by 10 K at nadir view. At higher incidence angles, horizontal polarisation is even more affected, while vertical polarisation is almost unaffected. In accordance with our theoretical investigations, the root mean square deviation between simulated and observed brightness temperatures at horizontal polarisation decrease from 20.0 K, if the snow layer is neglected, to 4.4 K, if the snow layer is included in the simulations. Since dry snow is almost transparent in L-band, the brightness temperature of snow-covered sea ice is dependent on snow thickness only if the thermal insulation of snow is accounted for. Due to the stronger insulation by a thicker snow layer, brightness temperatures increase with increasing snow thickness over thick multi-year ice for cold Arctic conditions. This temperature effect allows us to retrieve snow thickness over thick sea ice. For the best simulation scenario and snow thicknesses up to 35 cm, the average snow thickness retrieved from horizontally polarised SMOS brightness temperatures agrees within 0.3 cm with the average snow thickness measured during the IceBridge flight campaign in the Arctic in spring 2012. The corresponding root mean square deviation is 5.7 cm and the correlation coefficient is r= 0.58. We observe brightness temperatures over ice to increase by more than 20 K during the ice growth season in the Bay of Bothnia. We show that this brightness temperature increase is caused by increasing ice thicknesses, and not by changing ice conditions, which is the basis for a retrieval of ice thickness from SMOS. For validation with electromagnetic induction (EM) measurements in the Baltic Sea in March, 2011, the average SMOS and EM ice thicknesses agree within 0.5 cm. The root mean square deviation for the ice thicknesses of, on average 40 cm, is 7.7 cm.