Abstract. Sea-ice thickness is a key component of the Arctic climate system, but yet a comprehensive assessment across observations and numerical models is still missing. Previous studies have either compared only a small ensemble of sea-ice thickness products, focused on a short time scale, or both. We use an ensemble of 23 harmonised large-scale satellite, model, reanalysis and multi-product data ranging partly from 1995 to 2024, but mostly from 2010 to 2023. The products are compared against reference data derived from upward-looking sonar measurements in the Beaufort Sea. We find that biases are typically in the range of −0.2 m to 0.3 m, root mean square deviations are usually between 0.25 m and 0.5 m, and correlation coefficients mostly fall between 0.7 and 0.85, although larger deviations occur in some cases. Satellite and multi-product data mostly have lower biases and lower root mean square deviations (RMSD), but similar correlation coefficients compared to models and reanalyses. We examine the reliability of the uncertainties stated by the providers of twelve products and find that, while individual products tend to state uncertainties that are either too small or too large relative to their actual difference towards reference data, the ensemble as a whole shows comparable magnitudes of uncertainties and difference towards reference data. Subsequently, we do a pairwise comparison between decadal averages of large-scale products. We find biases largely between 0.2 and 0.4 m, RMSDs largely between 0.4 and 0.9 m and correlation coefficients largely between 0.5 and 0.8. Our study concludes with a time-series analysis of sea-ice thickness for each category (model/reanalysis, satellite, multi-product) in November and March for 2010–2023 and 1995–2023, with the second period being limited to the region south of 81.5° N. For the first period, we find no significant trend in any category for both months. For the second period, we find that sea-ice thickness has declined by roughly 0.5–0.6 m in November and 0.3–0.4 m in March, with stronger trends for models/reanalyses and multi-product data than for satellite products.

Abstract. Scattering and absorption from air bubbles, voids, and brine pockets significantly affect radar and microwave radiometer measurements of sea ice and light propagation through sea ice. However, there are only a limited number of in situ measurements of the size of natural sea ice inclusion and its distribution. Here, we used a high-resolution data set of 90-mm wide and 1-mm thin ice slices of various types of Arctic sea ice to estimate the autocorrelation length and density of the inclusions. The data set was collected during the MOSAiC International Arctic Drift Expedition 2019–20 during the winter months of January and February 2020. Thin ice slices from new ice, first-year ice, level second-year ice, second-year hummocks, and a refrozen melt pond were collected and analyzed. The 50-percentile autocorrelation lengths derived (Lobs), a measure of the size and distribution of inclusions, for new ice and brine ice brine pockets of the first-year have mean values between 0.11 and 0.30 mm and a vertical anisotropy ratio (Lz) and horizontal (Lx) of 1.7–3.0 with respect to the ice surface. The exponential model was fitted to the observed autocorrelation function with a delay between 0–2 Lobs. The exponential correlation lengths derived (Lexp) correspond to the 50 percentile Lobs for the ordinary (horizontal) image samples (Lexpx) and the transposed (vertical) images (Lexpz). For new and first-year ice with varying salinity and brine pocket image density, the autocorrelation length is a very robust measure of the size and distribution of brine pockets. For new ice, we find a Lobsx or Lexpx of 0.18 mm and with a Lz / Lx anisotropy of 2 and for first-year ice, we find a Lobs or Lexp of 0.16 mm with a Lz / Lx anisotropy of 3. The samples from the hummock and the refrozen melt pond had 50 percentile Lobsx and Lobsz air bubble autocorrelation lengths varying in the range [0.22, 0.73] mm and [0.22, 0.74] mm and [0.17, 0.56] mm and [0.17, 0.79] mm, respectively. The very consistent Lobs for new and first-year ice can be used to constrain sea ice microwave emission and scattering models.

Abstract. Radiometric measurements at L-band (1.4 GHz), collected in the Canadian Arctic in 2024, are used to study which type of model best reproduces the observations. While incoherent radiative transfer models are standard for sea ice thickness retrievals, they neglect phase interference effects. However, the observations analyzed here can only be explained when interference phenomena are explicitly included, requiring a coherent approach. To reduce uncertainties and ensure the robustness of the models, an optimal estimation method is used to determine snow and sea ice parameters consistent with the measured brightness temperatures and in situ measurements. The results show that the coherent model reproduces the observations substantially better than the incoherent formulation, yielding less than half the total cost with respect to the in situ measurements and being approximately 30,000 times more likely to explain the observations. These findings highlight the relevance of coherence effects at L-band, which are commonly neglected, at least in the context of local in situ measurements.

The Arctic is significantly affected by climate change, as evidenced by the constant decline of sea ice since the beginning of satellite observations. One driver of this transformation are melt ponds - pools of water that form as a result of melting sea ice during summer. Due to their darker color, they increase the absorption of incoming sunlight and accelerate ice melt. Accurate determination of melt pond extent and characteristics is considered a main factor in reducing uncertainty in Arctic climate models and sea ice concentration retrievals, but precise large scale observations are not available. Most knowledge to date is based on in-situ measurements, which are restricted to small areas. Satellite retrievals offer Arctic-wide coverage on a regular basis but lack resolution. To validate satellite measurements and allow observation at a moderate scale, helicopter-borne images are used. This ongoing work exploits a new dataset of helicopter-borne thermal infrared (TIR) imagery for melt pond retrieval. The derivation of geophysical parameters requires effective segmentation of different surface classes, which is challenged by temporally and spatially varying surface temperatures. We adapt and fine-tune AutoSAM, a prompt-free Segment Anything (SAM)-based segmentation tool that was introduced by Xinrong Hu et al. for medical imagery (Hu, X., Xu, X., & Shi, Y. (2023). How to Efficiently Adapt Large Segmentation Model (SAM) to Medical Images. arXiv preprint arXiv:2306.13731). Initial results with a limited number of annotated images indicate promising outcomes in the generalization of AutoSAM to cases that are rare in the training set, compared to U-Net and PSP-Net approaches. Beyond the scope of this project, this could serve as an example of how to use SAM as a segmentation tool for the remote sensing domain, which is typically hampered by the lack of labeled training data. Code and first results are provided at https://github.com/marlens123/autoSAM_pond_segmentation.

The retrieval of sea ice thickness using L-band passive remote sensing requires robust models for emission from sea ice. In this work, measurements obtained from surface-based radiometers during the MOSAiC expedition are assessed with the Burke, Wilheit and SMRT radiative transfer models. These models encompass distinct methodologies: radiative transfer with/without wave coherence effects, and with/without scattering. Before running these emission models, the sea ice growth is simulated using the Cumulative Freezing Degree Days (CFDD) model to further compute the evolution of the ice structure during each period. Ice coring profiles done near the instruments are used to obtain the initial state of the computation, along with Digital Thermistor Chain (DTC) data to derive the sea ice temperature during the analyzed periods. The results suggest that the coherent approach used in the Wilheit model results in a better agreement with the horizontal polarization of the in situ measured brightness temperature. The Burke and SMRT incoherent models offer a more robust fit for the vertical component. These models are almost equivalent since the scattering considered in SMRT can be safely neglected at this low frequency, but the Burke model misses an important contribution from the snow layer above sea ice. The results also suggest that a more realistic permittivity falls between the spheres and random needles formulations, with potential for refinement, particularly for L-band applications, through future field measurements.

Observations of sea ice surface temperature provide crucial information for studying Arctic climate, particularly during winter. We examined 1 m resolution surface temperature maps from 35 helicopter flights between October 2, 2019, and April 23, 2020, recorded during the Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC). The seasonal cycle of the average surface temperature spanned from 265.6 K on October 2, 2019, to 231.8 K on January 28, 2020. The surface temperature was affected by atmospheric changes and varied across scales. Leads in sea ice (cracks of open water) were of particular interest because they allow greater heat exchange between ocean and atmosphere than thick, snow-covered ice. Leads were classified by a temperature threshold. The lead area fraction varied between 0% and 4% with higher variability on the local (5–10 km) than regional scale (20–40 km). On the regional scale, it remained stable at 0–1% until mid-January, increasing afterward to 4%. Variability in the lead area is caused by sea ice dynamics (opening and closing of leads), as well as thermodynamics with ice growth (lead closing). We identified lead orientation distributions, which varied between different flights but mostly showed one prominent orientation peak. The lead width distribution followed a power law with a negative exponent of 2.63, which is in the range of exponents identified in other studies, demonstrating the comparability to other data sets and extending the existing power law relationship to smaller scales down to 3 m. The appearance of many more narrow leads than wide leads is important, as narrow leads are not resolved by current thermal infrared satellite observations. Such small-scale lead statistics are essential for Arctic climate investigations because the ocean–atmosphere heat exchange does not scale linearly with lead width and is larger for narrower leads.

The Sentinel-1 SAR Extra Wide swath mode with its 40 meters pixel size allows accurate lead mapping in Arctic sea ice areas. We present an improved winter lead detection algorithm based on a modified U-Net convolutional neural network. We introduce a preprocessing procedure that balances normalized radar cross section (NRCS) values between sub-swaths. This results in more consistent Sentinel-1 SAR images as input for the classifier, which is especially important for the cross-polarization HV channel. In turn, image consistency is essential for automatic image semantic segmentation. We develop a new lead detection algorithm and compare it with a previously published version. We produce pan-Arctic, 40 meters resolution lead maps and present pan-Arctic lead area fraction on a 12 km grid for January 2019. Within different Arctic regions, the lead area fraction is stable and varies only within the expected natural variability during the one month studied here. The average Arctic-wide lead area fraction is 2.44 % with 0.25 % standard deviation during January 2019. The improved lead detection algorithm leads to a better lead discrimination and extends the applicability area to the entire Arctic.

This paper presents a practical step-by-step approach to Frequency Modulated Continuous Wave (FMCW) radar nonlinearity correction (deconvolution), utilizing surface-based Ku- and Ka-band radar data collected over nilas ice within a newly-opened sea ice lead during the 2019/2020 MOSAiC expedition. Two performance metrics are introduced to evaluate deconvolution effectiveness: the spurious free dynamic range (SFDR), which quantifies sidelobe suppression, and the leading edge width (LEW), which quantifies the improvement in surface return clarity. The impact of deconvolution waveforms on different survey dates, radar polarizations, and surface types is examined using echograms and quantitative metrics. Deconvolution results in a maximum SFDR increase of 28 dB, with a maximum 3 dB decline in deconvolution performance observed over an 8-day period and a maximum decline of 15 dB observed over a 71-day period. The LEW values indicate that the effectiveness of deconvolution in enhancing interface clarity depends on the combination of pre-deconvolution sidelobe shape, prominence of the surface return, the influence of snowpack returns, as well as a time-dependent reduction in deconvolution performance. Deconvolution significantly improves surface return clarity for cross-polarized radar data, where weak surface returns are obscured by returns from within the snowpack. The results demonstrate that deconvolution performance is most effective shortly after deconvolution waveform characterization. Therefore, it is recommended to perform at least weekly calibrations using a large metal sheet and ideally calibration before/after data collection to ensure optimal deconvolution performance and effective sidelobe suppression.

Wind-driven redistribution of snow on sea ice alters its topography and microstructure, yet the impact of these processes on radar signatures is poorly understood. Here, we examine the effects of snow redistribution over Arctic sea ice on radar waveforms and backscatter signatures obtained from a surface-based, fully polarimetric Ka- and Ku-band radar at incidence angles between 0 degrees (nadir) and 50 degrees. Two wind events in November 2019 during the Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC) expedition are evaluated. During both events, changes in Ka- and Ku-band radar waveforms and backscatter coefficients at nadir are observed, coincident with surface topography changes measured by a terrestrial laser scanner. At both frequencies, redistribution caused snow densification at the surface and the uppermost layers, increasing the scattering at the air-snow interface at nadir and its prevalence as the dominant radar scattering surface. The waveform data also detected the presence of previous air-snow interfaces, buried beneath newly deposited snow. The additional scattering from previous air-snow interfaces could therefore affect the range retrieved from Ka- and Ku-band satellite altimeters. With increasing incidence angles, the relative scattering contribution of the air-snow interface decreases, and the snow-sea ice interface scattering increases. Relative to pre-wind event conditions, azimuthally averaged backscatter at nadir during the wind events increases by up to 8 dB (Ka-band) and 5 dB (Ku-band). Results show substantial backscatter variability within the scan area at all incidence angles and polarizations, in response to increasing wind speed and changes in wind direction. Our results show that snow redistribution and wind compaction need to be accounted for to interpret airborne and satellite radar measurements of snow-covered sea ice.

Abstract Comparing helicopter‐borne surface temperature maps in winter and optical orthomosaics in summer from the year‐long Multidisciplinary drifting Observatory for the Study of Arctic Climate expedition, we find a strong geometric correlation between warm anomalies in winter and melt pond location the following summer. Warm anomalies are associated with thinner snow and ice, that is, surface depression and refrozen leads, that allow for water accumulation during melt. Warm surface temperature anomalies in January were 0.3–2.5 K warmer on sea ice that later formed melt ponds. A one‐dimensional steady‐state thermodynamic model shows that the observed surface temperature differences are in line with the observed ice thickness and snow depth. We demonstrate the potential of seasonal prediction of summer melt pond location and coverage from winter surface temperature observations. A threshold‐based classification achieves a correct classification for 41% of the melt ponds.

Surface temperature is crucial in studying the Arctic climate, particularly during winter. We examine 1 m resolution surface temperature maps of 35 helicopter flights between 02 October 2019 and 23 April 2020, recorded during the Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC). The seasonal cycle of the average surface temperature spans from 265.6 K on 02 October 2019 to 231.8 K on 28 January 2020. The surface temperature is affected by atmospheric changes and also varies across scales. Furthermore, we concentrate on leads in sea ice because they allow for greater heat exchange between ocean and atmosphere than thick, snow-covered ice. Leads, which appear considerably warmer than sea ice, are classified by a temperature threshold. The local scale (5–10 km) lead area fraction varies between 0% and 4% with a higher variability than on a regional scale (20–40 km), where leads cover a more stable fraction of 0-1% until mid-January when it increases to 4%. The variability in the lead area is caused by sea ice dynamics (opening and closing of leads), as well as thermodynamics with ice growth (lead closing). To understand better the ice rheology throughout the winter, we identify lead orientation distributions. We find that the orientation varies between different flights but the distribution mostly shows one prominent orientation peak. Thus, we are not able to determine predominant intersection angles, which would need two modes in the orientation distribution. The lead width distribution follows a power law with a negative exponent of 2.63, which agrees with literature values, proves the comparability to other datasets, and extends the existing relationship to the smaller scales, as observed here. The appearance of many more small leads compared to wider leads is important since they only occur on the sub-footprint scale of thermal infrared satellite data. Sub-satellite-footprint lead statistics are essential for Arctic-climate investigations because the ocean-atmosphere heat exchange does not scale linearly with lead area fraction and is larger for smaller leads.

Monitoring surface and atmospheric parameters-like water vapor-is challenging in the Arctic, despite the daily Arctic-wide coverage of spaceborne microwave radiometer data. This is mainly due to the difficulties in characterizing the sea ice surface emission: sea ice and snow microwave emission is high and highly variable. There are very few data sets combining relevant in situ measurements with co-located remote sensing data, which further complicates the development of accurate retrieval algorithms. Here, we present a multi-parameter retrieval based on the inversion of a forward model for both, atmosphere and surface, for non-melting conditions. The model consists of a layered microwave emission model of snow and ice. Since snow scattering and emission effects, as well as temperature gradients, are taken into account, a high variability in brightness temperatures can be simulated. For ocean regions and the atmosphere existing parameterized forward models are used. By using optimal estimation, the forward model can be inverted allowing for the simultaneous and consistent retrieval of nine variables: integrated water vapor, liquid water path, sea ice concentration, multi-year ice fraction, snow depth, snow-ice interface temperature and snow-air interface temperature as well as sea-surface temperature and wind speed (over open ocean). In addition, the method provides retrieval uncertainty estimates for each retrieved parameter. To evaluate the forward model as well as the retrieval, we use the extensive data sets acquired during the year-long Arctic expedition Multidisciplinary drifting Observatory for the Study of Arctic Climate (2019-2020) as a reference.

Warm air intrusions over Arctic sea ice can change the snow and ice surface conditions rapidly and can alter sea ice concentration (SIC) estimates derived from satellite-based microwave radiometry without altering the true SIC. Here we focus on two warm moist air intrusions during the Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC) expedition that reached the research vessel Polarstern in mid-April 2020. After the events, SIC deviations between different satellite products, including climate data records, were observed to increase. Especially, an underestimation of SIC for algorithms based on polarization difference was found. To examine the causes of this underestimation, we used the extensive MOSAiC snow and ice measurements to model computationally the brightness temperatures of the surface on a local scale. We further investigated the brightness temperatures observed by ground-based radiometers at frequencies 6.9 GHz, 19 GHz, and 89 GHz. We show that the drop in the retrieved SIC of some satellite products can be attributed to large-scale surface glazing, that is, the formation of a thin ice crust at the top of the snowpack, caused by the warming events. Another mechanism affecting satellite products, which are mainly based on gradient ratios of brightness temperatures, is the interplay of the changed temperature gradient in the snow with snow metamorphism. From the two analyzed climate data record products, we found that one was less affected by the warming events. The low frequency channels at 6.9 GHz were less sensitive to these snow surface changes, which could be exploited in future to obtain more accurate retrievals of sea ice concentration. Strong warm air intrusions are expected to become more frequent in future and thus their influence on SIC algorithms will increase. In order to provide consistent SIC datasets, their sensitivity to warm air intrusions needs to be addressed.

Abstract Snow depth on sea ice is an Essential Climate Variable and a major source of uncertainty in satellite altimetry‐derived sea ice thickness. During winter of the MOSAiC Expedition, the “KuKa” dual‐frequency, fully polarized Ku‐ and Ka‐band radar was deployed in “stare” nadir‐looking mode to investigate the possibility of combining these two frequencies to retrieve snow depth. Three approaches were investigated: dual‐frequency, dual‐polarization and waveform shape, and compared to independent snow depth measurements. Novel dual‐polarization approaches yielded r2 values up to 0.77. Mean snow depths agreed within 1 cm, even for data sub‐banded to CryoSat‐2 SIRAL and SARAL AltiKa bandwidths. Snow depths from co‐polarized dual‐frequency approaches were at least a factor of four too small and had a r2 0.15 or lower. r2 for waveform shape techniques reached 0.72 but depths were underestimated. Snow depth retrievals using polarimetric information or waveform shape may therefore be possible from airborne/satellite radar altimeters.

The sea ice surface temperature is important to understand the Arctic winter heat budget. We conducted 35 helicopter flights with an infrared camera in winter 2019/2020 during the Multidisciplinary Drifting Observatory for the Study of Arctic Climate (MOSAiC) expedition. The flights were performed from a local, 5 to 10 km scale up to a regional, 20 to 40 km scale. The infrared camera recorded thermal infrared brightness temperatures, which we converted to surface temperatures. More than 150000 images from all flights can be investigated individually. As an advanced data product, we created surface temperature maps for every flight with a 1 m resolution. We corrected image gradients, applied an ice drift correction, georeferenced all pixels, and corrected the surface temperature by its natural temporal drift, which results in time-fixed surface temperature maps for a consistent analysis of one flight. The temporal and spatial variability of sea ice characteristics is an important contribution to an increased understanding of the Arctic heat budget and, in particular, for the validation of satellite products.

Arctic sea ice is changing rapidly. Its retreat significantly impacts Arctic heat fluxes, ocean currents, and ecology, warranting the continuous monitoring and tracking of changes to sea ice extent and thickness. L-band (1.4 GHz) microwave radiometry can measure sea ice thickness for thin ice ≤1 m, depending on salinity and temperature. The sensitivity to thin ice makes L-band measurements complementary to radar altimetry which can measure the thickness of thick ice with reasonable accuracy. During the Multidisciplinary drifting Observatory for the Study of Arctic Climate (MOSAiC) expedition, we deployed the mobile ARIEL L-band radiometer on the sea ice floe next to research vessel Polarstern to study the sensitivity of the L-band to different sea ice parameters (e.g., snow and ice thickness, ice salinity, ice and snow temperature), with the aim to help improve/validate current microwave emission models. Our results show that ARIEL is sensitive to different types of surfaces (ice, leads, and melt ponds) and to ice thickness up to 70 cm when the salinity of the sea ice is low. The measurements can be reproduced with the Burke emission model when in situ snow and ice measurements for the autumn transects were used as model input. The correlation coefficient for modeled Burke brightness temperature (BT) versus ARIEL measurements was approximately 0.8. The discrepancy between the measurements and the model is about 5%, depending on the transects analyzed. No explicit dependence on snow depth was detected. We present a qualitative analysis for thin ice observations on leads. We have demonstrated that the ARIEL radiometer is an excellent field instrument for quantifying the sensitivity of L-band radiometry to ice and snow parameters, leading to insights that can enhance sea ice thickness retrievals from L-band radiometer satellites (such as Soil Moisture Ocean Salinity (SMOS) and Soil Moisture Active Passive (SMAP)) and improve estimates of Arctic sea-ice thickness changes on a larger scale.

Sea ice concentration algorithms using brightness temperatures ( $T_{B}$ ) from satellite microwave radiometers are used to compute sea ice concentration ( $c_{\text{ice}}$ ), sea ice extent, and generate sea ice climate data records. Therefore, it is important to minimize the sensitivity of $c_{\text{ice}}$ estimates to geophysical noise caused by snow/sea ice thermal microwave emission signature variations, and presence of WV and clouds in the atmosphere and/or near-surface winds. In this study, we investigate the effect of geophysical noise leading to systematic $c_{\text{ice}}$ biases and affecting $c_{\text{ice}}$ standard deviations (STD) using simulated top of the atmosphere $T_{B}$ s over open water and 100% sea ice. We consider three case studies for the Arctic and the Antarctic and eight different $c_{\text{ice}}$ algorithms, representing different families of algorithms based on the selection of channels and methodologies. Our simulations show that, over open water and low $c_{\text{ice}}$ , algorithms using gradients between V-polarized 19-GHz and 37-GHz $T_{B}$ s show the lowest sensitivity to the geophysical noise, while the algorithms exclusively using near-90-GHz channels have by far the highest sensitivity. Over sea ice, the atmosphere plays a much smaller role than over open water, and the $c_{\text{ice}}$ STD for all algorithms is smaller than over open water. The hybrid and low-frequency (6 GHz) algorithms have the lowest sensitivity to noise over sea ice, while the polarization type of algorithms has the highest noise levels.

The Weddell Sea is known to feature large openings in its winter sea ice field, otherwise known as open-ocean polynyas. An area within the Weddell Sea region that has repeatedly featured open-ocean polynyas in the past is that which encompasses the Maud Rise seamount. Within this area, after 40 years of intermittent, smaller openings, a larger, more persistent polynya appeared in early September 2017​​​​​​​ and remained open for approximately 80 d until spring ice melt. In this study we present proof that polynya-favorable activity in the Maud Rise area is taking place more frequently and on a larger scale than previously assumed. By investigating thin (< 50 cm) apparent sea ice thickness (ASIT) retrieved from the satellite microwave sensors Soil Moisture and Ocean Salinity (SMOS) and Soil Moisture Active Passive (SMAP), we find an anomaly of thin sea ice spanning an area comparable to the polynya of 2017 over Maud Rise which occurred in September 2018. In this paper, we look at sea ice above Maud Rise in August and September of 2017 and 2018 as well as all years from 2010 until 2020 in an 11-year time series. Using fifth-generation ECMWF Reanalysis (ERA5) surface wind reanalysis data, we corroborate previous findings (e.g., Campbell et al., 2019; Francis et al., 2019; Wilson et al., 2019) on the strong impact that storm activity can have on sea ice above Maud Rise and help consolidate the theory that the evolution of the Weddell Sea polynya is controlled by local atmospheric as well as oceanographic variability. Based on the results presented, we propose that the Weddell Sea polynya, rather than being a binary phenomenon with one principal cause, is a dynamic process caused by various different preconditioning factors that must occur simultaneously for it to appear and persist. Moreover, we show that rather than an abrupt stop to anomalous activity above Maud Rise in 2017, the very next year shows signs of polynya-favorable activity that, although insufficient to open the polynya, were present in the region. This phenomenon, as we have shown in the 11-year SMOS record, was not unique to 2018 and was also identified in 2010, 2013 and 2014. It is demonstrated that L-band microwave radiometry from the SMOS and SMAP satellites can provide additional useful information, which helps to better understand dynamic sea ice processes like polynya events when compared to the use of satellite sea ice concentration products alone.