Research on estimating sea ice concentration (SIC) from synthetic aperture radar (SAR) data using convolutional neural networks (CNNs) has been widely reported. However, the presence of speckle noise in dual-polarization SAR signals and confusion at ice-water boundaries complicates accurate density estimation, often leading to significant underestimations of SIC. Recently, diffusion models (DMs) have shown significant success in various remote sensing tasks, demonstrating their potential to address these challenges. However, applying DMs directly to SIC estimation leads to singularity issues, hindering the accuracy of results. Additionally, directly incorporating conditional images can cause denoising models to overlook the differences between conditional and noise information. We introduce a conditional denoising diffusion probabilistic model (DiffSIC) that can fundamentally resolve the singularity problem by reweighting the loss function. We designed a U-shaped architecture that integrates conditional information, time steps, and noise information for SIC estimation. Extensive experiments conducted on the AI4Arctic dataset indicate that the proposed DiffSIC framework achieves a coefficient of determination (R2) of 90.959% and a root mean square error (RMSE) of 8.632%, demonstrating the effectiveness and potential of diffusion models in the task of SIC estimation.

Arctic sea ice leads, which are narrow linear openings between sea ice, are critical to polar climate and ocean–atmosphere interactions. However, detecting multi-width leads accurately under the case of cloud interference remains challenging for thermal infrared (TIR) data. In this article, we propose an improved U-Net-based model–CTMLU-Net–which is designed to extract multi-width leads, including internal bright leads, from FY-3D MERSI-II 250-m TIR imagery under varied cloudy conditions. The model incorporates a dynamic fast Fourier transform (FFT) module to distinguish leads from clouds in the frequency domain and introduces cloud categories into the training labels to improve the lead detection accuracy. A dual-loss module and multi-window brightness temperature anomaly (BTA) inputs further enhance the sensitivity to different lead widths. Under cloudy conditions, CTMLU-Net achieved a precision of 86.82% and a recall of 71.73%, outperforming the traditional BTA method by 39.26% and 23.95%, respectively. For multi-width leads, it achieved a precision of 92.10% and a recall of 87.73%, and for bright leads, the corresponding scores were 90.59% and 84.75%, respectively. Compared to the original U-Net model, the FFT module improved the precision and recall by 4.80% and 17.66% in cloudy conditions, respectively. Compared to single-loss setting, the dual-loss module further enhanced the multi-width lead detection precision and recall by 23.64% and 9.92%. CTMLU-Net was also applied to 7045 scenes of MERSI-II TIR images to generate monthly Arctic lead frequency maps from November 2019 to April 2020. The spatiotemporal patterns of the leads can capture dynamic events, such as the collapse of the Beaufort High in spring 2020. Overall, CTMLU-Net offers a robust and accurate solution for multi-width sea ice lead detection under cloudy conditions.

Leads are almost linear fractures within the ice pack, which are commonly observed in polar regions. In wintertime, leads promote energy flux from the underlying ocean to the atmosphere. Synthetic aperture radar (SAR) can monitor leads at a finer spatial resolution than other spaceborne datasets, regardless of solar illumination and atmospheric conditions. However, the SAR-based lead detection methods proposed to date are restricted to some specific areas, instead of the entire Arctic. In this article, we present a generalized deep learning-based approach for automatic sea ice lead detection (SILDET) in the Arctic wintertime using Sentinel-1 SAR images. The validation results show that SILDET has the capability of detecting open and frozen leads at different stages of development. Compared with the visual interpretation of Sentinel-1 images, the overall detection accuracy is 97.80% and the Kappa coefficient is 0.88. The lead map of a regional study obtained from SILDET was compared to that from a previous SAR-based lead detection method and a lead dataset based on Moderate Resolution Imaging Spectroradiometer (MODIS) data. The lead map was also validated using Sentinel-2 images. The result shows that SILDET can provide a more detailed distribution of leads and a better estimation of lead width and area. SILDET was applied to present the Arctic-wide lead distribution from January to April 2023 with a spatial resolution of 40 m. The Arctic-wide lead width distribution follows a power law with an average exponent of 1.65. The SILDET approach can be expected to provide long-term high-resolution lead distribution records.

The gradient ratios (GRs), defined as the normalized difference between measured vertically (V) or horizontally (H) polarized brightness temperatures (TBs) at two frequencies, have been commonly used to retrieve snow depth on Arctic sea ice from passive microwave radiometer data. In this study, the influences of snow density on the relationship between GR of 6.9 and 18.7 GHz vertically polarized TBs (i.e., GRV(19/7)) and snow depth were investigated through observed data and simulation. The former was based on regression analysis between GRV(19/7) observations from Advanced Microwave Scanning Radiometer 2 (AMSR2) and the altimetric snow depth estimates derived by differencing freeboard heights from ICESat-2 and CryoSat-2 while the latter was based on model simulations from the Microwave Emission Model for Layered Snowpacks (MEMLS). An improved snow depth retrieval algorithm is proposed based on a multilinear regression model with GRV(19/7) from AMSR2 and snow density from the NASA Eulerian Snow On Sea Ice Model (NESOSIM) as predictors, and then validated using three airborne snow depth datasets. The validation results show an overall good accuracy of the improved algorithm with the correlation coefficient (r) ranging from 0.60 to 0.72 and the root mean square error (RMSE) values varying between 6.18 cm and 7.53 cm. The improved algorithm enables daily snow depth estimation on sea ice over the entire Arctic Ocean during the full winter season (October to April). More importantly, it successfully captures the seasonal variation of snow depth which is expected to increase throughout the winter season due to snow accumulation.

Monitoring Arctic sea ice drift is essential for understanding climate change. Currently, large-scale observed sea ice drift datasets primarily rely on single-sensor remote sensing data, which suffer from low spatial resolution or poor spatial continuity. To address these limitations, this study proposes a sea ice drift retrieval framework based on multi-sensor data, utilizing the complementary sea ice drift information derived from passive microwave radiometers and medium-resolution optical sensors. The proposed framework employs the maximum cross-correlation (MCC) based pattern-matching method to obtain sea ice drift fields from coarse-resolution Fengyun-3D (FY-3D) Microwave Radiation Imager (MWRI) data, and an A-KAZE-based feature-tracking method to extract sea ice motion vectors from FY-3D Medium-Resolution Spectral Imager II (MERSI-II) data. Finally, the sea ice drift vectors from different sensors are merged using the Co-Kriging algorithm to obtain the final sea ice drift result. The effectiveness of the proposed framework was assessed by comparing displacements from 166 buoys with the retrieved vectors derived from FY-3D single-sensor and multi-sensor data, as well as an existing sea ice drift product (Ifremer-AMSR2) collected in the Beaufort Sea, the East Siberian Sea, and the Fram Strait. The results demonstrate the proposed framework’s ability to retrieve fine-resolution (i.e., 1 km) and spatially continuous sea ice drift fields in areas where vectors from fine-resolution data can be obtained. The overall mean absolute errors (MAEs) of the merged sea ice motion vectors are 0.76 km/day for speed and 4.53° for angle, exhibiting superior drift accuracy to Ifremer-AMSR2 in areas covered by MERSI-II vectors.

The challenges of utilizing the 250-m resolution thermal infrared (TIR) data obtained from the Medium Resolution Spectral Imager-II (MERSI-II) onboard the Chinese Fengyun-3D (FY-3D) satellite are bowtie effect and nonuniform brightness stripe noise. While previous solutions have addressed these issues separately, this article introduced a more integrated two-step image quality enhancement strategy for MERSI-II TIR images. It considered the interactions between the two issues and overcame the excessive or inadequate destriping in existing models due to the ideal stripe-type assumption. Specifically, for the bowtie effect, a rigorous geometric model suitable for MERSI-II was constructed by considering the Earth’s curvature and adjusting the preset image width. For the nonuniform brightness stripe noise, a novel adaptive multiscale frequential (AMSF) algorithm was developed. The multiscale spectral detection effectively captured the anomaly frequency, and the adaptive threshold dynamically adjusted the detection range, profiting in preserving details. The proposed strategy was validated on MERSI-II TIR images, outperforming existing methods in quantitative and qualitative assessments with higher efficiency on both bowtie effect and stripe noise removal. Further experiments conducted on Moderate Resolution Imaging Spectroradiometer (MODIS) data demonstrated the AMSF algorithm’s applicability to different data. In addition, the 250-m MERSI-II FY-3D data can help us understand the rapid variations of Arctic sea ice leads, which are key features within the sea ice. Using the quality-enhanced images to extract the sea ice leads in winter Arctic Baffin Bay improved the overall accuracy from 0.88 to 0.95, thereby providing more accurate and reliable sea ice lead data.

The neutral form drag coefficient is an important parameter when estimating surface turbulent fluxes over Arctic sea ice. The form drag caused by surface features ($C_{\text {dn},\text {fr}}$) dominates the total drag in the winter, but long-term pan-Arctic records of$C_{\text {dn},\text {fr}}$are still lacking for Arctic sea ice. In this study, we first developed an improved surface feature detection algorithm and characterized the surface features (including height and spacing) over Arctic sea ice during the late winter of 2009–2019 using the full-scan laser altimeter data obtained in the Operation IceBridge mission.$C_{\text {dn},\text {fr}}$was then estimated using an existing parameterization scheme. This was followed by applying a satellite-derived backscatter coefficient (${\sigma }_{\text {vv}}^{o}$) to$C_{\text {dn},\text {fr}}$regression model to extrapolate, for the first time,$C_{\text {dn},\text {fr}}$to the pan-Arctic scale for the entire winter season over two decades (from 1999 to 2021). We found that the surface features have a larger height and smaller spacing over multiyear ice (1.15 ± 0.21 and 142 ± 49 m) than over first-year ice (0.90 ± 0.16 and 241 ± 129 m). The monthly mean$C_{\text {dn},\text {fr}}$increases through the winter from$0.2\times 10^{-3}$in November to 0.4–$0.5\times 10^{-3}$in April. The central Arctic has the largest$C_{\text {dn},\text {fr}}$(up to$2\times 10^{-3}$) but experienced a drop of ~50% in the period from 2001/2002 to 2008/2009. The interannual fluctuations in$C_{\text {dn},\text {fr}}$are strongly linked to the variability of sea ice thickness and deformation, and the latter has become increasingly important for$C_{\text {dn},\text {fr}}$since 2009.

The Antarctic blue ice can be classified into wind- and melt-induced based on their origins. They play a different role in the development of surface water systems, the surface energy balance and the infrastructure. Currently, visible light remote sensing is the most effective method for mapping blue ice. However, optical imagery faces difficulties in classifying blue ice accurately, and it is also greatly influenced by weather conditions. Synthetic Aperture Radar (SAR) images have the potential to map blue ice under all weather conditions, but it is difficult to distinguish blue ice from other similar weak microwave reflecting surfaces. In this study, by employing a segmentation method based on band ratios of the Sentinel-2 images, we delineated the geographical distribution of blue ice area (BIA) in the Lambert Glacier Basin. Taking advantage of the disparity in coherence levels between melt-induced and wind-induced blue ice, we performed blue ice classification in the Lambert Glacier Basin using Sentinel-1 images. The proposed method achieves an overall accuracy of 0.91 and F1-score of 0.91 and provides blue ice types with a spatial resolution of 10 m. The total area of blue ice in the Lambert Basin was estimated to be approximately 1.986 × 104km2. Among them, the area of melt-induced blue ice was approximately 1.276 × 104km2, while the wind-induced blue ice covered around 0.710 × 104km2. Melt-induced BIA was predominantly distributed in low-altitude coastal areas and downstream of glaciers, exhibiting higher surface temperatures compared to wind-induced BIA. Wind-induced BIA, on the other hand, was mainly found near nunataks and exposed rocks, displaying higher albedo than melt-induced BIA.

Snow on sea ice plays an important role in the polar climate system and accurate snow depth (SD) information on sea ice is necessary for satellite estimates of sea ice thickness from both radar and laser altimeters. In this study, three independent SD products have been generated from the Advanced Microwave Scanning Radiometer 2 (AMSR2) data using different algorithms which are trained separately based on three different reference data sets, including the Ice Mass Balance Buoy (IMB) measured snow depth, the Operation IceBridge (OIB) airborne SD measurements and the monthly altimetric snow depth (ASD) product derived from CryoSat-2 (CS2) and ICESat-2 (IS2). An in-situ validation based on SD measurements from OIB and IMB and a triple collocation (TC) evaluation are both conducted to assess the accuracy of these SD products. Furthermore, a merging scheme based on error variance estimates obtained from TC analysis has been tested for merging these three SD products into a single data set. Results indicate that TC can provide information about the error characteristics of each product which is complementary to in-situ validation. Meanwhile, the merged SD product is superior to its input parent products and achieves an overall good accuracy with the correlation (r) and root mean square error (RMSE) values being 0.70 and 6.06 cm when validating using OIB data, and 0.75 and 9.12 cm when validating using IMB data. This study demonstrates the great potential of the TC method in validating and merging snow depth estimates on sea ice.

The objective of this letter is to extend the commonly used gradient ratio (GR) method for Arctic sea ice snow depth estimation by combining brightness temperatures from the Soil Moisture Active Passive (SMAP) and the Advanced Microwave Scanning Radiometer 2 (AMSR2). The L band (1.4 GHz) channel from SMAP together with higher frequencies (i.e., 6.9, 10.7, 18.7, and 36.5 GHz) from AMSR2 were used to calculate GRs, which were then used to derive empirical snow depth retrieval algorithms based on 5 years of Operation IceBridge (OIB) snow depth measurements acquired on Arctic sea ice during springtime. Results show that the gradient ratio GR(1/19) at vertical polarization is suitable for snow depth estimation over both first-year ice (FYI) and multi-year ice (MYI) and could achieve the best performance with correlationrand root mean square distance (RMSD) values being -0.80 and 5.95 cm, respectively. More importantly, there exists a one-to-one relationship between snow depth and GR(1/19) independent of sea ice types, which is an advantage of the GR(1/19) over the previous GR(7/19).

Lead is an important feature of the Arctic ice cover, with possible contents of thin ice /or open water. In this letter, we present an algorithm for lead detection based on brightness temperature observations from a single thermal infrared channel of MERSI-II onboard the Chinese FY-3D satellite. Lead contents is classified into open water and thin ice with support information from Sentinel-1 SAR data. Results are evaluated based on visual interpretation of MERSI-II TIR (Thermal infrared) and Sentinel-2 NIR (Near Infrared) data. The accuracy is found to be 85.6% for lead detection and 67% and 52% for thin ice and open water within the lead, respectively.

As a result of global warming, multiyear ice (MYI) is being replaced by first-year ice (FYI) in the Arctic. Microwave scatterometers in the Ku-band and C-band can provide daily observations of sea ice type. However, their comparative capabilities in mapping ice type have not been thoroughly evaluated. We present a systematic intercomparison of the backscatter signature in VV polarization (${\sigma }_{\mathrm {vv}}^{\mathrm {o}}$) and the sea ice classification from three scatterometer systems using the same ice classification approach. The systems are the Ku-band quick scatterometer (QSCAT) and the newly launched Chinese rotating fan-beam scatterometer (RFSCAT) and the C-band advanced scatterometer (ASCAT). Three freezing seasons are used, i.e., 2007/08 and 2008/09 for the QSCAT/ASCAT comparison and 2019/20 for the RFSCAT/ASCAT comparison. With reference to ASCAT,${\sigma }_{\mathrm {vv}}^{\mathrm {o}}$bias between QSCAT and RFSCAT results from their different incidence angles. A continuous declining trend of${\sigma }_{\mathrm {vv}}^{\mathrm {o}}$from MYI and FYI is observed during winter, with a greater difference between MYI and FYI in the Ku-band. The MYI and FYI extent derived from QSCAT/RFSCAT is highly consistent with that derived from ASCAT, with a difference less than 7% and 3% for MYI and FYI, respectively. The overall accuracy (OA) is around 77% and 80% for the RFSCAT results and ASCAT results, respectively, compared with Sentinel-1 SAR images. The classification results show high consistency (81%–89%) with ice charts from the Canadian Ice Service. The incorporation of${\mathrm {Tb}}_{36\mathrm {h}}$from AMSR-E/AMSR2 improves the OA of the classification when using ASCAT or RFSCAT by 7%–11%.

Ice/snow surface temperature (I/SST) is an essential parameter in many research fields such as the climate change, energy, and matter balance of the South pole regions. Recently, many algorithms have been developed for various satellite observations to derive the I/SST. However, rare studies focus on accurate I/SST retrieval from the observations of Chinese MEdium Resolution Spectral Imager II (MERSI-II) instrument onboard the FY-3D satellite with the spatial resolution of 250 m and temporal resolution of about five days, which just bridges the specifications of the Aqua-MODerate-resolution Imaging Spectroradiometer (MODIS) (1000-m pixel size and 0.5-day revisit cycle) and Landsat-Thermal InfraRed Sensor (TIRS) (100-m pixel size and 16-day revisit cycle) instruments. In this study, a new method with correction of striping noise and consideration of angular emissivity effect is developed for the MERSI-II data to accurately retrieve the I/SST. The performance of the proposed method is assessed by using both MODIS product and ground I/SST measurements. The results show that the FY-3D I/SST retrieval accuracy is comparable to the MODIS product, with a discrepancy of < 1.6 K. Ground-based validation reveals that the proposed method could be used to accurately retrieve the I/SST with a root-mean-square error (RMSE) of < 1.5 K. Overall, this study proposes a method for accurate I/SST retrieval from the FY-3D MERSI-II data with the pixel size of 250 m, implying the possibilities in improving the spatio-temporal resolutions of the current I/SST products. Moreover, the proposed method is also helpful to improve our understandings of the polar regions.

Temporal and spatial variation of sea ice type in the Arctic is an indicator of regional and global change. Arctic sea ice can be classified into two major categories: multiyear ice (MYI) and first-year ice. In this paper, classification method based on machine learning is established and applied to produce daily sea ice classification data set during the winter (November-April) from 2002 to 2017 using active microwave data from QuikSCAT and Advanced Scatterometer as well as passive microwave data from Advanced Microwave Scanning Radiometer for EOS, Special Sensor Microwave Imager/Sounder, and Advanced Microwave Scanning Radiometer 2 radiometer. First, the open water area is flagged out using brightness temperature (Tb) from the passive microwave sensor. Then, K-means algorithm is applied to identify the clusters of the two ice types in the Tb/backscatter parameter space and finally assign pixels to each class. Two optimization methods based on the movement of MYI and marginal ice zone are used to correct the misclassification of MYI. The results have shown a decrease of MYI in winter from 2002 to 2017, especially in 2008 and 2013 with a remarkable recovery in 2014. The classifications are consistent with results by visual interpretation from synthetic aperture radar images in the Canadian Arctic Archipelago with overall classification accuracy over 93%. Comparison with classifications from previous studies and products shows that our method could reflect more differences in MYI declining trend interannually and less anomalous fluctuations in certain years.