【摘要】 北极海冰是地球系统的重要组成部分，调节着大气和海洋之间的能量交换，其中薄冰对海气之间的热量交换影响尤为显著.凭借L波段亮温对海冰探测的灵敏性，土壤湿度与海洋盐度(SMOS)卫星可以获取高精度、日尺度的薄冰厚度信息.现有研究多采用单系点方法进行SMOS海冰厚度反演，但由于不同海域在温度、盐度及海冰生长情况方面存在显著差异，该方法在应用于全北极薄冰厚度反演时会带来一定的偏差.本研究首先通过对海冰亮温与厚度变化的模拟分析，探讨了系点选取对海冰厚度反演精度的影响；进而基于ERA5再分析数据、海冰密集度和亮温数据，构建了多系点选取策略，获取了北极多系点数据集；最后基于该数据集，反演获取了2010—2023年冻结期内的日尺度北极薄冰厚度.通过德国不莱梅大学SMOS海冰厚度产品(UB＿SIT)以及美国冰桥计划海冰厚度(OIB＿SIT)数据，对多系点海冰反演结果 (MTP＿SIT)进行了对比验证.结果表明，MTP＿SIT与UB＿SIT平均偏差约为0.05 m，二者相关性显著. OIB＿SIT验证结果显示，MTP＿SIT的精度约为-0.11±0.21 m，优于UB＿SIT精度(-0.24±0.28 m).本研究提出的多系点海冰厚度反演方法能够实现更为精确的薄冰厚度估计.更多还原

【Abstract】 Arctic sea ice is a key component of the Earth system, regulating energy exchange between the atmosphere and ocean. Thin ice, in particular, dominates heat transfer across the sea-air interface. Exploiting the sensitivity of L-band brightness temperature to thin ice, the Soil Moisture and Ocean Salinity(SMOS) mission enables daily high-precision estimates of thin-ice thickness. However, most existing studies rely on a single-tie point(STP) for SMOS sea-ice thickness retrieval. Spatial variability in temperature, salinity and ice growth conditions across the Arctic sea regions introduces substantial errors when STP is applied to thin sea ice thickness retrieval. This study simulates the brightness temperature variations on sea ice thickness and quantifies the effect of tie-point selection on retrieval accuracy. A multi-tie point(MTP) strategy is developed using ERA5 reanalysis data, sea ice concentration, and brightness temperature to construct an Arctic MTP dataset. Based on this dataset, daily MTP sea ice thickness(MTP＿SIT) is retrieved for the freeze seasons from 2010 to 2023. The resulting MTP＿SIT product is compared with the SMOS sea ice thickness product(UB＿SIT) from university of Bremen in Germany and the sea ice thickness dataset from Operation IceBridge(OIB) in the United States. The results show that the mean deviation between MTP＿SIT and UB＿SIT is approximately 0.05 m with strong correlation. The OIB sea ice thickness validation reveals that the accuracy of MTP＿SIT is about-0.11 ±0.21 m, outperforming UB＿SIT(-0.24 ± 0.28 m). The MTP＿SIT retrieval method developed in this study provides more accurate estimates of thin ice thickness.更多还原