北极地区海冰密集度和海冰覆盖范围预测数据（2020年6-9月）

数据是本项目成员自主研发的气候系统模式FGOALS对北极海冰密集度和海冰覆盖范围进行预测的结果。同化技术的正确选取，是北极海冰预测的重要因素，在海冰资料同化技术中，奇异值演化插值卡尔曼滤波（简称SEIK），是发展相对较早但是仍很常用的一种滤波算法，但由于计算所有格点之间的误差协方差，存在虚假的遥相关误差，因此考虑发展局部滤波方法，对海冰密集度和海冰厚度进行同化。本项目将在气候系统模式FGOALS 中，初始化处理欧洲航天局（ESA）CryoSat-2 和Soil Moisture and Ocean Salinity（SMOS）卫星遥感反演的海冰厚度数据。

The dataset consists of predictions of Arctic sea ice concentration and sea ice extent generated by the FGOALS climate system model, which was independently developed by members of this project. Proper selection of data assimilation technologies is a critical factor for Arctic sea ice prediction. Among sea ice data assimilation methods, the Singular Evolutive Interpolated Kalman filter (SEIK for short) is a relatively early-developed but still widely used filtering algorithm. However, since it computes error covariances between all grid points, it introduces spurious teleconnection errors. Therefore, localized filtering methods are being considered for assimilating sea ice concentration and sea ice thickness data. This project will initialize and process sea ice thickness data retrieved via satellite remote sensing from the European Space Agency (ESA)'s CryoSat-2 and Soil Moisture and Ocean Salinity (SMOS) satellites within the FGOALS climate system model.