Spring Snow Depth Over Pan-Arctic Sea Ice Derived from AMSR2 Using Machine Learning Methods (2013–2023)

Description This dataset offers daily estimates of snow depth over pan-Arctic sea ice during the spring seasons (March–April) from 2013 to 2023, utilizing data from the Advanced Microwave Scanning Radiometer-2 (AMSR2). The snow depth retrieval employs an innovative machine learning (ML) framework that integrates K-Nearest Neighbors (KNN), Extremely Randomized Trees (Extra Trees), and TabNet models. The Dataset includes NetCDF files containing daily snow depth fields for the entire study period, alongside corresponding snow depth maps and uncertainty maps (Map). Snow depth values are provided only for regions where sea ice concentration exceeds 50%, ensuring data quality. Additionally, seven-day moving average results are included to help mitigate the effects of melt-refreeze-thaw cycles, although this reduces the temporal resolution of the data. We also provide videos that visually illustrate the daily temporal variation of snow depth during the spring months of March and April, along with a day-to-day comparison of algorithmic uncertainty estimates across different ML models (Video). For any inquiries, please contact [e-mail]. Data Files Filename: YYYYMMDD [Year-Month-Day] AMSR2_Merged: Merged snow depth estimates based on the fusion of KNN, Extra Trees, and TabNet models. AMSR2_Merged_Unc: Uncertainty in the merged snow depth estimates, calculated as the weighted standard deviation. AMSR2_KNN: Snow depth estimates derived from the KNN model. AMSR2_ExtraTrees: Snow depth estimates derived from the Extra Trees model. AMSR2_TabNet: Snow depth estimates derived from the TabNet model. AMSR2_Merged_7DayAvg: Snow depth estimates from the merged model with a seven-day moving average applied. AMSR2_Merged_Unc_7DayAvg: Uncertainty in the merged snow depth estimates, calculated as the weighted standard deviation with a seven-day moving average applied. AMSR2_KNN_7DayAvg: Snow depth estimates derived from the KNN model with a seven-day moving average applied. AMSR2_ExtraTrees_7DayAvg: Snow depth estimates derived from the Extra Trees model with a seven-day moving average applied. AMSR2_TabNet_7DayAvg: Snow depth estimates derived from the TabNet model with a seven-day moving average applied. Platform(s): GCOM-W satellite Sensor(s): Advanced Microwave Scanning Radiometer 2 (AMSR2) Data Format(s): NetCDF Temporal Coverage: March 1, 2013 to April 30, 2023 Temporal Resolution: Daily/Weekly Spatial Resolution: 25 km × 25 km Spatial Reference System(s): NSIDC Sea Ice Polar Stereographic North (EPSG:3411) Spatial Coverage: North: 90°N, South: 0°N, East: 180°E, West: 180°W Example of Data Reading Using MATLAB >> ncdisp('20130301.nc')Source: 20130301.ncFormat: netcdf4_classicGlobal Attributes: description = 'This dataset provides daily estimates of snow depth over pan-Arctic sea ice for the spring seasons (March–April) from 2013 to 2023, utilizing data from the Advanced Microwave Scanning Radiometer-2 (AMSR2). The snow depth retrieval is based on a novel machine learning (ML) framework that combines K-Nearest Neighbors (KNN), Extremely Randomized Trees (Extra Trees), and TabNet models. The OIB-MEDIAN (2013–2015) serves as the benchmark for data fusion constraints.'Dimensions: x = 304 y = 448Variables: lon Size: 304x448 Dimensions: x,y Datatype: double Attributes: Long_name = 'Longitude: NSIDC Sea Ice Polar Stereographic North (EPSG:3411)' Units = 'degrees_east' lat Size: 304x448 Dimensions: x,y Datatype: double Attributes: Long_name = 'Latitude: NSIDC Sea Ice Polar Stereographic North (EPSG:3411)' Units = 'degrees_north' AMSR2_Merged Size: 304x448 Dimensions: x,y Datatype: double Attributes: Long_name = 'Merged snow depth estimates based on the fusion of KNN, Extra Trees, and TabNet models' Units = 'centimeter' AMSR2_Merged_Unc Size: 304x448 Dimensions: x,y Datatype: double Attributes: Long_name = 'Uncertainty in the merged snow depth estimates, calculated as the weighted standard deviation' Units = 'centimeter' AMSR2_KNN Size: 304x448 Dimensions: x,y Datatype: double Attributes: Long_name = 'Snow depth estimates derived from the KNN model' Units = 'centimeter' AMSR2_ExtraTrees Size: 304x448 Dimensions: x,y Datatype: double Attributes: Long_name = 'Snow depth estimates derived from the ExtraTrees model' Units = 'centimeter' AMSR2_TabNet Size: 304x448 Dimensions: x,y Datatype: double Attributes: Long_name = 'Snow depth estimates derived from the TabNet model' Units = 'centimeter' AMSR2_Merged_7DayAvg Size: 304x448 Dimensions: x,y Datatype: double Attributes: Long_name = 'Snow depth estimates from the merged model with a seven-day moving average applied' Units = 'centimeter' AMSR2_Merged_Unc_7DayAvg Size: 304x448 Dimensions: x,y Datatype: double Attributes: Long_name = 'Uncertainty in the merged snow depth estimates, calculated as the weighted standard deviation with a seven-day moving average applied' Units = 'centimeter' AMSR2_KNN_7DayAvg Size: 304x448 Dimensions: x,y Datatype: double Attributes: Long_name = 'Snow depth estimates derived from the KNN model with a seven-day moving average applied' Units = 'centimeter' AMSR2_ExtraTrees_7DayAvg Size: 304x448 Dimensions: x,y Datatype: double Attributes: Long_name = 'Snow depth estimates derived from the ExtraTrees model with a seven-day moving average applied' Units = 'centimeter' AMSR2_TabNet_7DayAvg Size: 304x448 Dimensions: x,y Datatype: double Attributes: Long_name = 'Snow depth estimates derived from the TabNet model with a seven-day moving average applied' Units = 'centimeter'