Global Navigation Satellite System Reflectometry (GNSS-R) has long been explored for retrieving sea ice properties, but in-situ validation in the central Arctic during the freezing season is rare, limiting its application. The primary objective of this study is to advance the current understanding of multi-polarization GNSS-R remote sensing for sea ice application. This paper presents observations from the full-polarization GNSS-R(FpolGNSSR) prototype during the MOSAiC expedition. The FpolGNSSR, with four polarization channels and high antenna gain (11.3 dB), aims to assess the impact of sea-ice thickness and permittivity on GNSS-R data, with observations from October 2019 to January 2020, the onset period of ice growth. First, the reflectivity is simulated by a four-layer model, and the sensitivity of multi-polarization GNSS-R to sea ice is qualitatively analyzed. Subsequently, a simplified model reveals a linear relationship between reflectivity and ice thickness, with regression showing a correlation of 0.74 (P<0.01). The retrieval error (RMSE) of sea ice thickness retrieval is 0.13 m for first-year ice (0.3–1.0 m thick). Additionally, the imaginary component of sea ice permittivity is estimated, between 0.018 and 0.039. This study provides valuable insights for the GNSS-R community, which could be summarized as: (1) the “H-polarization advantage”, which is first proposed in sea ice GNSS-R, (2) the validity of a simplified model, reinforcing the feasibility of satellite-based sea ice thickness estimations using Left-hand-circular, V, and H polarizations, and (3) a lower apparent permittivity of GNSS-R than previously reported, corresponding to a deeper penetration depth.

Antarctic sea ice is generally considered to be free from surface melt, as opposed to the Arctic sea ice where surface melt is prevalent during summer. In this work, we present recent high-resolution Sentinel-2 images which confirm the presence of melt ponds also on the Antarctic sea ice. We employ an optical melt pond fraction retrieval applied to Sentinel-3 data to investigate occurrences of surface melt in the vicinity of Lützow-Holm Bay where sea ice ponding has been confirmed in situ. Sea ice surface melt has the potential to disrupt passive microwave sea ice concentration product. We focus on an area of land fast ice with confirmed 100% ice concentration to highlight this effect on real data.

Understanding the Antarctic sea ice optical properties is the key to the accurate assessment of the radiative balance of the region. Although high resolution optical satellite data have recently become more available also over sea ice, most Antarctic remote sensing research is dedicated to land ice and glaciers, with sea ice classification studies still lacking. In the context of recent change with the two last years 2022 and 2023 showing record low Antarctic sea ice extent, high resolution sea ice optical properties are key to understanding the Antarctic sea ice evolution. In this work, we present high resolution surface classification based on the remote sensing Sentinel-2 MSI data. We retrieve sea ice albedo, fraction of blue ice and impurities. A case study scene features variable sea ice albedo, where pure sea ice alternates with darker sea ice featuring yellow substance contamination and potential surface melt.

A ground-based ultra-wideband radiometer operating at 540, 900, 1380, and 1740 MHz was used to measure microwave thermal emissions from an Arctic sea ice floe as part of the Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC) Expedition. The instrument was deployed on a drifting ice floe near 86°N, 120°E in leg 1 of the expedition (December 2019) and observed second-year ice (potentially with refrozen melt ponds) that experienced new ice growth at its base over a ten-day period. Measured circularly polarized brightness temperatures were compared with the predictions of a radiative transfer (RT) model for a layered medium consisting of ocean, growing new ice, desalinated remnant second-year ice/refrozen melt pond, and snow layers. Characteristics of the sea ice composition used in the model were determined fromin-situmeasurements. Comparisons of the measured and modeled wideband brightness temperatures showed good agreement consistently over the observation period and for various off-nadir observation angles. The results demonstrate the capabilities of 0.5–2 GHz microwave radiometry for observing sea ice properties and also show the impact of a saline ice layer at the ice bottom on the measured brightness temperature.

Automated solutions for sea ice-type classification from synthetic aperture radar (SAR) imagery offer an opportunity to monitor sea ice, unimpeded by cloud cover or the arctic night. However, there is a common struggle to obtain accurate classifications year round, particularly in the melt and freeze-up seasons. During these seasons, the radar backscatter signal is affected by wet snow cover, obscuring information about underlying ice types. By using additional spatiotemporal contextual data and a combination of convolutional neural networks and a dense conditional random field, we can mitigate these problems and obtain a single classifier that is able to classify accurately at 3.5-m spatial resolution for five different classes of sea ice surface from October to May. During the near year-long drift of the Multidisciplinary Drifting Observatory for the Study of the Arctic Climate (MOSAiC) expedition, we collected satellite scenes of the same patch of Arctic pack ice with X-band SAR with a revisit time of less than a day on average. Combined within situobservations of the local ice properties, this offers up the unprecedented opportunity to perform a detailed and quantitative assessment of the robustness of our classifier for level, deformed, and heavily deformed ice. For these three classes, we can perform accurate classification with a probability >95% and calculate a lower bound for the robustness between 85% and 88%.

Reflectometry measurements have been conducted aboard the German research icebreakerPolarsternduring the Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC) expedition. Signals of Global Navigation Satellite Systems (GNSS) were recorded using a dedicated GNSS reflectometry receiver for retrieval of sea-ice reflectivity. The primary goal is reflectometry-based monitoring of sea ice as a part of the Arctic climate study. The dataset presented here covers the expedition’s first leg (late September to mid-December 2019) in the Siberian Sector of the central Arctic (at about 82 ° N to$87~^\circ $N). Daily profiles of reflectivity are retrieved for satellite elevations$< 45^\circ $. In agreement with model prediction, the results show best reflectivity contrast (about 5 dB between compact pack-ice and lower ice concentrations) for observations at left-handed circular polarization and elevation angles of 10°–20°. A daily resolved time series of sea-ice relative permittivity is inverted from the left-handed data. In general, the level of inversion results is at the lower limit of sea-ice values (relative permittivity of 3 and below), potentially indicating an influence of incoherent volume scattering. An occasional increase in the relative permittivity is attributed to the presence of water. Sea-ice profiles show anomalies that are confirmed by enhanced model prediction (slab reflection). A long-term comparison of prediction and retrieved profiles indicates anomalies’ dependence on ice thickness and temperature.

The research vessel ‘Polarstern’, moored to an ice floe, completed a year long drift with Arctic pack ice in the autumn of 2020. During that expedition, named MOSAiC, a comprehensive data set of airborne laser scanner (ALS) and spaceborne X-band SAR images in the area of the research vessel was acquired. With successful fusion of these two measurements, we can extrapolate sea ice features from the ALS data to the entire SAR scene using a convolutional neural network (CNN). From two preliminary scenes of ALS data we are able to show this for classes of sea ice roughness. This will be the basis for more comprehensive research, once the complete data set is available.

In September 2019, the German research icebreaker Polarstern started the largest multidisciplinary Arctic expedition, the MOSAiC (Multidisciplinary drifting Observatory for the Study of Arctic Climate) drift experiment. Being moored to ice floes at high Arctic for a whole year, thus including the winter season, the main goal of the expedition is to better understand and quantify relevant processes within the atmosphere-ice-ocean system that impact the sea ice, ultimately leading to improved climate models. Satellite remote sensing, specially multi-frequency synthetic aperture radar (SAR) plays a major role to achieve this goal. Two major objectives in SAR based remote sensing of sea ice is on the one hand to have a large coverage, and on the other hand to obtain a radar response that carries as much information as possible. A comprehensive set of C- and L- band SAR images were acquired during the course of MOSAiC. In this initial study we evaluate the effects of seasonal changes on C- and L-band backscatter in respect to three different sea ice types, i.e., Young Ice, Smooth Ice and Rough/Deformed Ice along with the performance of sea ice type retrieval of a established algorithm. Areas of deformed, smooth and young sea ice were observed in the vicinity of R/V Polarstern and these areas are included whenever possible in the yearlong time series. For both frequencies a change in all backscatter channels values can be observed during the early melt season. This is first noticeable in the C-band images and later followed by a change in the L-band images, probably caused by their different penetration depth and volume scattering sensitivities.

Sea ice lead area fraction, distribution of lead length and orientation of leads are subject of this study. Leads are classified from dual-band Sentinel-1 SAR data with an automatic supervised learning classification algorithm. Binary maps are combined from scenes acquired within a three-day interval to provide an Arctic-wide composite lead map. Resolution of these binary maps is 80 meters. Based on these binary maps, lead area fraction is calculated on a 12 km grid. Regional maps of lead area fraction for the Beaufort Sea and the Fram Strait are calculated on a 4 km grid. The Hough transform is used to detect linear features on binary lead maps. The lead length distribution and the orientation of leads are calculated for the Fram Strait region and the Beaufort Sea. Most of the detected leads have length below 15 km. Two pronounced peaks on the lead orientation are found at around 50 and 130 degrees.

Accurate water vapour measurements are crucial for a better understanding of Arctic climate change and the global hydrological cycle. In the sparsely observed polar regions, passive microwave satellite-borne sensors allow obtaining the vertically integrated water vapour content of the atmosphere (Total Water Vapour, TWV) with whole Arctic or Antarctic coverage, respectively. However, the available retrievals work only over certain surface types. Over open ocean, microwave imagers like SSM/I, AMSR-2 or AMSR-E provide daily precipitable water. Over ice and land surfaces, another method based on data of the microwave humidity sounders AMSU-B and MHS allows TWV retrieval. Here, we aim to merge these complementary TWV retrievals, creating an Arctic-wide daily dataset of 50 km resolution with seamless coverage from the high Arctic to mid latitudes from 2002 to date.

Synthetic aperture radar is used to identify and monitor oil spills. Separation from oil spill look-alikes is an important part of a fully automatic oil spill detection scheme. Here we investigate the polarimetric signatures for oil spills and newly formed sea ice (a well-known look-alike) in fully polarimetric Radarsat-2 satellite scenes. Using the fully polarimetric scenes we calculate four different parameters, co-polarization ratio, polarization difference, scattering entropy, and mean alpha angle. Three pairs of satellite scenes with comparable incidence angles are used. We observe that a combination of the co-polarization ratio and the polarization difference enables us to delineate the spills from their surrounding and also to discriminate the oil spills from the newly formed sea ice. The scattering entropy and the alpha values provide additional information about the scattering mechanisms of sea ice and oil spills.

A helicopter-based Doppler scatterometer (Multi$^{3}$Scat) is described. It allows simultaneous measurements of the surface radar backscatter at five different frequencies at co- and cross-polarization at incidence angles of 20$^{\circ}$–65$^{\circ}$from an altitude of 30–300 m. Video and infrared (IR) cameras simultaneously sense the surface in the scatterometers' footprint. The Multi$^{3}$Scat is calibrated using measurements carried out over corner reflectors. The stability of the Multi$^{3}$Scat's signal is found to be, on average, better than 0.5 dB. Typical signal-to-noise-ratio values for sigma-0 range between 10 and 20 dB for cross-polarization and between 15 and 25 dB for copolarization over snow and ice surfaces. The potential of the Multi$^{3}$Scat to acquire multifrequency multipolarization radar backscatter data and coincident video and IR temperature observations at different incidence angles over remote terrain such as the Arctic Ocean or the Alps is demonstrated.

Enhanced-resolution QuikSCAT/SeaWinds (QSer) data recently entered the daily ice chart operation of the national ice services. Algorithms have been developed to extract four important sea ice parameters from this data over the whole Arctic: sea ice edge, type, concentration, and drift. This paper will summarize the different algorithms with a more detailed presentation of the sea ice concentration (IC) algorithm that has not been previously published. The sea ice edge can be detected to IC as low as 10%. Sea ice types can be roughly separated by a single threshold of 12 dB in the horizontal polarization. The IC algorithm gives reasonable qualitative results, separating into three classes: high, medium, and low ICs. It resolves even some characteristic ice features in the marginal ice zone and dynamic areas like the Fram Strait. However, it is very empirical and quantitatively not reliable. Sea ice drift can be determined with an accuracy of about 2.6 cm/s for a 48-h drift. Operating since 1999, QS is an important global data set for climate research, and two crucial applications how these sea ice products can be used for climate research are presented: the seasonal evolution of the sea ice cover and the export of sea ice volume through Fram Strait.

Recent progress in spatial resolution enhancement of sea ice concentrations obtained by microwave remote sensing has been stimulated by two new developments: First, the new sensors AMSR (Advanced Microwave Scanning Radiometer) on MIDORI-II and AMSR-E on AQUA offer horizontal resolutions of 6x4 km at 89 GHz. This is nearly three times the resolution of the standard sensor SSM/I at 85 GHz (15x13 km). The sampling distance at the high frequencies is 12.5 km at SSM/I and 5 km at the AMSR-E instrument. Second, a new algorithm enables the estimation of sea ice concentrations from the channels near 90 GHz, despite the enhanced atmospheric influence in these channels. This allows to fully exploit their horizontal resolution which is two to three times finer than the one of the channels near 19 and 37 GHz. These frequencies are used by the most widespread algorithms for sea ice retrieval, the NASA Team and Bootstrap algorithms. These two developments are combined to determine operationally sea ice concentration maps. The used ASI (Artist Sea Ice) algorithm combines a model for retrieving the sea ice concentration from SSM/I 85 GHz data proposed by Svendsen et al. (1) with an ocean mask derived from the 18-, 23-, and 37-GHz AMSR-E data using two weather filters and the Bootstrap Algorithm. The AMSR-E sea ice concentration data are projected into grids of sampling sizes down to 3 km. Hemispherical and regional maps are provided daily at www. iup.physik.uni-bremen.de.