five

DynIceData: A gridded ice-water classification dataset at short-time intervals based on observations from multiple satellites over the marginal ice zone

收藏
科学数据银行2023-07-11 更新2026-04-23 收录
下载链接:
https://www.scidb.cn/detail?dataSetId=7ad5407c4fd54c81b8676393eb0a4a2b
下载链接
链接失效反馈
官方服务:
资源简介:
The high-resolution observations to the short-term changes of sea ice are critical for scientific understanding to the ice dynamic pattern, especially in the Arctic, which is also important to the Arctic shipping advisory. Though, single orbit spaceborne sensor provides periodically sea ice observations at tens of meters resolution, the knowledge of short-time interval dynamic changes, e.g., within minutes or hours, are limited. A gridded ice-water classification dataset at short-time intervals by multi spaceborne observations in the Marginal Ice Zone (MIZ), the DynIceData, has been developed with multi spaceborne observations. The DynIceData combines Sentinel-1 Synthetic Aperture Radar (SAR) data with the Gaofen-3 (GF-3) satellite SAR and SDGSAT-1 thermal infrared image to obtain multi-time interval observations in the ice margin regions, with the time scale from minutes to tens of hours interval, covering key areas of the Arctic of the Kara Sea, Beaufort Sea, and Greenland Sea during August 2021-August 2022. Object-oriented segmentation and threshold method were used to obtain the ice-water classification map with Sentinel-1 and GF-3 SAR image pairs, Sentinel-1 SAR and SDGSAT-1 thermal image pairs, the time interval of those image pairs are between 1 minute to 68 hours, at a 10km-grid sample granule with the spatial resolution of 25m for GF-3 SAR data and 30m for SDGSAT-1 thermal data. It was verified that the overall accuracy of classification is above 95.58%. The DynIceData includes 7338 samples, which could be used as reference data for further research on the sea ice rapid change patten at different short-time interval scale, providing data support for knowledge extraction about the thermodynamic and dynamic models of sea ice in combination with environmental factors, and potentially improving the accuracy of sea ice forecasting by the Artificial Intelligence is used.The dataset file includes two folders, which of name are '25m' and '30m'. (1)The '25m' folder includes GF-3 and Sentinel-1 SAR sample pairs at different time intervals. The time range of the sample pairs is from August to December 2021, and the spatial resolution is 25m, mainly distributed in the Kara Sea. There are 5443 files in total, with the data volume of 98.3MB. (2)The '30m' folder includes SDGSAT-1 thermal infrared and Sentinel-1 SAR sample pairs at different time intervals, the time range of the sample pairs is from July to August 2022, the spatial resolution is 30m, mainly distributed in the Greenland Sea and Beaufort Sea. There are 1946 files in total, with the data volume of 17.5MB. The files are the ice-water remote sensing classification samples of the 10km grid in the Arctic sea ice margin, raster data in tif format.
提供机构:
Wanyang Zhong; University of Chinese Academy of Sciences; Aerospace Information Research Institute, Chinese Academy of Sciences; Shuwen Yu; Yubao Qiu; International Research Center of Big Data for Sustainable Development Goals
创建时间:
2023-02-24
二维码
社区交流群
二维码
科研交流群
商业服务