The formation and distribution of melt ponds have an important influence on the Arctic climate. Therefore, it is necessary to obtain more accurate information on melt ponds on Arctic sea ice by remote sensing. The present large-scale melt pond products, especially the melt pond fraction (MPF), still require verification, and using very high resolution optical satellite remote sensing data is a good way to verify the large-scale retrieval of MPF products. Unlike most MPF algorithms using very high resolution data, the LinearPolar algorithm using Sentinel-2 data considers the albedo of melt ponds unfixed. In this paper, by selecting the best band combination, we applied this algorithm to Landsat 8 (L8) data. Moreover, Sentinel-2 data, as well as support vector machine (SVM) and iterative self-organizing data analysis technique (ISODATA) algorithms, are used as the comparison and verification data. The results show that the recognition accuracy of the LinearPolar algorithm for melt ponds is higher than that of previous algorithms. The overall accuracy and kappa coefficient results achieved by using the LinearPolar algorithm with L8 and Sentinel-2A (S2), the SVM algorithm, and the ISODATA algorithm are 95.38% and 0.88, 94.73% and 0.86, and 92.40%and 0.80, respectively, which are much higher than those of principal component analysis (PCA) and Markus algorithms. The mean MPF (10.0%) obtained from 80 cases from L8 data based on the LinearPolar algorithm is much closer to Sentinel-2 (10.9%) than the Markus (5.0%) and PCA algorithms (4.2%), with a mean MPF difference of only 0.9%, and the correlation coefficients of the two MPFs are as high as 0.95. The overall relative error of the LinearPolar algorithm is 53.5% and 46.4% lower than that of the Markus and PCA algorithms, respectively, and the root mean square error (RMSE) is 30.9% and 27.4% lower than that of the Markus and PCA algorithms, respectively. In the cases without obvious melt ponds, the relative error is reduced more than that of those with obvious melt ponds because the LinearPolar algorithm can identify 100% of dark melt ponds and relatively small melt ponds, and the latter contributes more to the reduction in the relative error of MPF retrieval. With a wider range and longer time series, the MPF from Landsat data are more efficient than those from Sentinel-2 for verifying large-scale MPF products or obtaining long-term monitoring of a fixed area.

We present an algorithm for computing ice drift in the marginal ice zone (MIZ), based on partial shape recognition. With the high spatial resolution of Sentinel-1 and Sentinel-2 images, and the low sensitivity to atmospheric influences of Sentinel-1, a considerable quantity of ice floes is identified using a mathematical morphology method. Hausdorff distance is used to measure the similarity of segmented ice floes. It is tolerant to perturbations and deficiencies of floe shapes, which enhances the density of retrieved sea ice motion vectors. The PHD algorithm can be applied to sequential images from different sensors, and was tested on two combined image mosaics consisting of Sentinel-1 and Sentinel-2 data acquired over the Fram Strait; the PHD algorithm successfully produced pairs of matched ice floes. The matching result has been verified using shape and surface texture similarity of the ice floes. Moreover, the present method can naturally be extended to the problem of multi-source sea ice image registration.

Melt ponds play a crucial role in the melting of Arctic sea ice. Studying the evolution of melt ponds is essential for understanding changes in Arctic sea ice. In this study, we used a revised sea ice model to simulate the evolution of melt ponds along the MOSAiC drift at a resolution of 10 m. A novel melt pond parameterization scheme simulates the movement of meltwater under the influence of gravity over a realistic sea ice topography. We evaluated different melt pond parameterization schemes based on remote sensing observations. The absolute deviation of the maximum pond coverage simulated by the new scheme is within 3%, while differences among parameterization schemes exceed 50%. Errors were found to be primarily due to the calculation of macroscopic meltwater loss, which is related to sea ice surface topography. Previous studies have indicated that sea ice with a lower surface roughness has a larger catchment area, resulting in larger pond coverage during the melt season. This study has identified an opposing mechanism: sea ice with lower surface roughness has a larger catchment area connected to the macroscopic flaws of the sea ice surface, which leads to more macroscopic drainage into the ocean and thereby a decrease in melt pond coverage. Experimental simulations showed that sea ice with 46% higher surface roughness, resulting in 12% less macroscopic drainage, exhibited a 38% higher maximum pond fraction. The presence of macroscopic flaws is related to the fragmentation of sea ice cover. As Arctic sea ice cover becomes increasingly fragmented and mobile, this mechanism will become more significant.

Sea ice regulates the overall energy exchange and radiation budget of the Arctic region, and understanding this relationship requires an accurate determination of snow depth. However, methods for deriving snow depth have a large error through the annual winter and early spring periods due to the potential complexity of surface melting during early summer. In this study, we explore the potential of retrieving snow depth during the early summer using optical satellite imagery of the sea-ice cover. Measurements using VIS/IR (visible and infrared) usually feature much higher spatial resolution than L-band satellite data and can provide additional surface melting and leads information; in addition, considering the snow grain size–snow surface temperature interaction, there is co-variability between the observed sea-ice surface broadband albedo using an optical satellite sensor, the sea-ice surface temperature, and the retrieval target of snow depth on the spatial scale of optical imagery samples. We applied a surface classification procedure to optical satellite imagery and introduce an approach to derive snow depth from optical satellite imagery and ice surface temperature data using two solar radiation transfer models: the Delta-Eddington solar radiation model, which is the shortwave radiative scheme of the Los Alamos sea-ice model, and a simplified snow albedo scheme, which is tuned to the observational data of buoys. The snow depth was inversed from the model simulation results using a lookup-table-based method. For comparison with the observational data, using the Delta-Eddington solar radiation model, about 55% of the differences are below 5 cm, and thicker snowpack has a larger bias; using the simplified snow albedo scheme, a mean difference of 4.1 cm between retrieval and measurements was found, with 93% of the differences being smaller than 5 cm. This approach can be applied to optical satellite imagery acquired under clear-sky conditions and can serve as an addition to overcome the limitations of existing methods.

Sea ice surface roughness plays a critical role in snow redistribution and accumulation; however, the spatial scale at which its influence is most effective remains poorly constrained by field observations. During the 15th Chinese National Arctic Research Expedition in 2025, synchronous measurements of ice surface elevation and snow depth were conducted using a lightweight Global Navigation Satellite System (GNSS) Real-Time Kinematic (RTK) system. Observations were collected over multi-year sea ice in the central Arctic north of 85° N, encompassing both level ice and deformed ice regimes. Multi-scale moving-window analysis revealed a strong scale dependence in the statistical relationship between sea ice surface roughness and mean snow depth. The ensemble-mean correlation coefficient reached a maximum at a window radius of approximately 7–10 m (r ≈ 0.56), indicating that this scale dominates the coupling between surface roughness and snow redistribution. At this characteristic scale, rougher ice surfaces generally corresponded to greater snow depths, reflecting the enhanced trapping and retention of wind-blown snow by complex surface topography during wind-driven transport. These findings provide observational constraints for parameterizing snow redistribution processes on sea ice and offer guidance for selecting appropriate spatial resolutions in remote-sensing products aimed at resolving roughness-controlled snow processes.

Seasonal Arctic sea ice prediction is important for understanding Arctic climate change and its impacts. Existing deep-learning models commonly use the multi-year mean climatology of the forecast period as a historical-evolution background. This provides stable seasonal information, but has limited ability to represent interannual differences in sea ice evolution. We propose a dynamic similar-year feature construction method based on sea ice spatial similarity and incorporate it into the SwinUNet forecasting framework. Using sea ice concentration (SIC) at forecast initialization and the Intersection over Union (IoU) of sea ice extent (SIE), the method identifies the most similar historical year since 1979 and extracts its SIC evolution over the forecast period as the historical-evolution feature. Continuous SIC and sea ice thickness (SIT) from the preceding four months are also included, combining recent conditions with historical same-period evolution. Using NSIDC SIC and PIOMAS SIT data from 1979 to 2024, four-month Arctic sea ice forecasts are conducted at pentad intervals. The Similarity scheme is compared with Continuous and Climatology features in SwinUNet, as well as Persistence and ECMWF SEAS5. Similarity generally outperforms Continuous and Climatology. Over the first 12 forecast periods, it reduces MAE by 5.41% and 6.97%, improves ACC by 1.63% and 4.03%, and increases BACC by 0.09% and 0.42%, respectively. Relative to Persistence and SEAS5, the annual-mean MAE decreases by 22.19% and 42.24%, while BACC improves by 4.67% and 8.43%. These results show that dynamic similar-year features can enhance SwinUNet skill without increasing model complexity.