This dataset utilizes multi-source remote sensing data to generate information on Arctic glacier area, thickness, ice volume, mass balance, movement velocity, and surface temperature since 1980. A total of 16 representative glaciers within the study area were selected as research subjects, distributed as shown in Figure 1. The dataset was produced as follows: Based on Landsat data, visual interpretation methods were employed to generate five glacier inventories for the Arctic region covering the years 1985, 1995, 2005, 2015, and 2024, from which corresponding glacier areas were extracted. Using two publicly available glacier thickness and mass balance datasets, thickness and mass balance information for the target glaciers during the study period was obtained through mean value processing. Mass balance data were sourced from published literature and model reconstructions. Glacier movement velocities for the target glaciers from 1985 to 2022 were extracted using ITS_LIVE velocity data. Monthly temperatures for the target glaciers from 1985 to 2024 were determined using ERA5 reanalysis data. Glacier surface temperature ranges were delineated based on glacier inventories, with the average temperature within each range representing the surface temperature.
(1) Glacier area data is derived from Landsat satellite series data, obtained from the https://earthexplorer.usgs.gov/ (USGS) and the Geospatial Data Cloud (http://www.gscloud.cn/). This satellite series carries sensors including the Multispectral Scanner (MSS), Thematic Mapper (TM), Enhanced Thematic Mapper (ETM+), and Operational Land Imager (OLI), with spatial resolutions ranging from 15 to 100 meters and a temporal resolution of 16 days.
(2)Glacier thickness and volume data: Glacier thickness data are sourced from Millan et al. (2022) global glacier surface velocity and ice thickness dataset.
(3) Glacier mass balance data are sourced from field observations, literature collection, and existing datasets. In this dataset, data for the Novaya Zemlya Archipelago, PARRISH, and WYKEHAM GLACIER SOUTH glaciers are derived from literature, as follows: ① Ciracì E, Velicogna I, Sutterley T C. Mass balance of Novaya Zemlya Archipelago, Russian High Arctic, using time-variable gravity from GRACE and altimetry data from ICESat and CryoSat-2 [J]. Remote Sensing, 2018, 10(11): 1817. ② Millan R, Mouginot J, Rignot E. Mass budget of the glaciers and ice caps of the Queen Elizabeth Islands, Canada, from 1991 to 2015 [J]. Environmental Research Letters, 2017, 12(2): 024016. Data for thirteen glaciers—McCall, West Gulkana, Eklutna, Wolverine, Alexander, Lemon Creek, Taku, Yuri, Andrei, MITTIVAKKAT, GEITLANDSJOKULL, AUSTRE BROEGGERBREEN, and ENGABREEN—are sourced from existing datasets. The specific sources are as follows: The existing datasets are from the Fluctuations of Glaciers (FoG) Database of the World Glacier Monitoring Service. This database is a standardized international collection of information on glacier conditions and changes (length, area, volume, mass), based on in-situ measurements, remote sensing, and reconstructions.
(4) Glacier velocity data originate from the ITS_LIVE (Inter-mission Time Series of Land Ice Velocity and Elevation) dataset, extracted from Landsat 4, 5, 7, and 8 satellite imagery. This dataset covers all land ice areas exceeding 5 km² in size, spanning the period from 1985 to 2022. This dataset offers resolutions of 120m and 240m. This dataset utilizes the 120m resolution velocity data generated from optical satellite imagery. Data retrieval website: (https://its-live.jpl.nasa.gov/).
(5) Glacier temperature data is sourced from the European Centre for Medium-Range Weather Forecasts (ECMWF) Fifth Generation Atmospheric Reanalysis (ERA5), a comprehensive reanalysis dataset. This dataset utilizes atmospheric reanalysis products to obtain air temperature data. Data acquisition website: (https://cds.climate.copernicus.eu/#!/home)
(1) Glacier Area. The ratio threshold method and snow cover index method based on optical remote sensing imagery are currently the most commonly used techniques for glacier boundary extraction. Given that some glacial areas in this study are covered by surface moraines and high-quality imagery is insufficient, we used the RGIV7 glacier catalog as a reference to accurately quantify glacier boundaries. This involved combining multiple feature parameters to obtain boundary ranges, followed by manual revision of the results in ArcGIS software. Finally, glacier area changes in this region were studied based on the manually revised results. During manual revision, snow cover and shadow effects should be minimized to ensure smooth glacier boundaries without jagged edges. Vector lines should closely follow actual glacier margins, avoiding large jumps at segment turning points.
(2) Glacier thickness and volume information. Glacier thickness and volume information: Glacier ice thickness estimation is based on the Shallow-Ice Approximation (SIA) model, which accounts for surface movement and bedding slip. The relationship between surface velocity and bedding velocity is calculated using a formula that incorporates ice density, gravitational acceleration, and the elevation difference between the ice surface and bedding. Bedding velocity is estimated by introducing a proportionality factor related to surface velocity. Ice thickness calculations account for the effects of velocity, ice surface slope, and density, ultimately providing estimated ice thickness values for a glacier network with connectivity less than 2 across all glaciers. Based on the acquired glacier thickness data, corresponding glacier volume is derived by multiplying thickness by area.
(3)The glacier mass balance is based on data collected from literature and does not involve processing methods.
(4) Glacier velocity acquisition. Based on IT_LIVE data and glacier flow line data extraction, the flow velocity of each glacier is extracted using the raster clipping tool in ArcGIS software, and its average value is calculated.
(5) Glacier temperature acquisition. Based on ERA5 data and glacier boundary data extraction, the temperature of each glacier is extracted using the raster clipping tool in ArcGIS software, and its average value is calculated.
(1) Glacier area uncertainty: This dataset only accounts for errors caused by the spatial resolution of Landsat remote sensing imagery. The calculation formula is as follows: ε = N*A. Here, ε represents the glacier area error (km²); N denotes the glacier perimeter length; A is the length of half a pixel (15 m).
(2) Model validation involves using field data to determine whether discrepancies between the model and measured data stem from overfitting. To this end, the research team removed 60% of the thickness data from the Alpine region for modeling and compared the results with the removed data. Results showed a discrepancy of −16 ± 51 meters between the model and measured data, indicating no overfitting. Error analysis across different ice thickness ranges revealed that ice layers thicker than 100 meters exhibited errors between 25% and 35%, while those thinner than 100 meters showed errors exceeding 50%.
(3) Mass balance data quality control follows the same standards as the data source;
(4) Glacier velocity: Refer to the ITS_LIVE data documentation (http://its-live-data.jpl.nasa.gov.s3.amazonaws.com/documentation/ITS_LIVE-Regional-Glacier-and-Ice-Sheet-Surface-Velocities.pdf) for relevant data quality and error calculations.
(5) For glacier temperature, related data quality, and error calculations, refer to (https://confluence.ecmwf.int/display/CKB/ERA5%3A+data+documentation)
| # | number | name | type |
| 1 | 2020YFA0608501 | Research on Arctic Terrestrial Environmental Change and Its Effects | National key R & D plan |
This work is licensed under
CC BY 4.0 (Creative Commons Attribution 4.0 International License).
CSTR:11738.11.NCDC.ARCTIC-CHANGE.DB7131.2026
10.12072/ncdc.arctic-change.db7131.2026
