Data Fusion to estimate sea-ice permittivity: a GNSS processor for 1-year MOSAiC data

The retrieval of Earth surface parameter using GNSS reflectometry techniques has become a valuable source for Earth observation. Typical parameters can be found over the open ocean (sea state, oceanwind), over land (soil moisture, inundation areas) or over sea ice (for example its extent andconcentration). The compactness of passive GNSS receiver instrumentation is a crucial advantage forthe versatile application scenarios of GNSS reflectometry techniques. We demonstrate here the estimation sea-ice permittivity based on the fusion GNSS and ancillary data. In the given scenario, GNSS observations were performed on the German research icebreaker Polarstern during its one year drift with the Arctic sea ice as part of the MOSAiC expedition (Multidisciplinary drifting Observatory for the Study of Arctic Climate). A dedicated GORS type (GNSS OccultationReflectometry Scatterometry) receiver was used with three antenna links attached: up-looking master link with right-handed polarization and two side-looking slave links (dual-polarization, leftand right-handed). Coherent samples (in-phase and quadrature) of the respective links are provided by the receiver. The processing steps comprise, at first, the separation of the GNSS multipath signal into direct and reflected contributions using the right- and left-handed slave-link samples. Two steps of data fusion follow, first, combining the separated signal power estimates to obtain reflectivity time series and, second, adding geo-reference to the obtained time series defining specular point and elevation angle. The geo-referencing involves: standard point position data of the GORS receiver and broadcast orbit data of the GNSS satellites (available at the IGS). Additionally, attitude data from the ship's inertial navigation system is used for event masking to assure satellite visibility and account for shadowing of the ship structure. Sea-ice permittivity is finally inverted from the referenced and masked reflectivity time series. For this purpose, the data fusion scheme is extended by ancillary sea ice concentration data acquired on the ship according to the ASSIST protocol (Arctic Ship-based Sea Ice Standardization Tool). The GNSS data processor, presented here, is focused on reflectometry considering the challenges of a ship-based setup (multipath signals and ship's attitude changes). Currently, the processor is enhanced to GNSS remote sensing concept that also monitors ionosphericimpact on the MOSAiC GNSS data record