This paper describes a quasi-physical method (the Q-method) for determining the sea surface temperatures (SSTs). The Q-method is a coefficient-based technique developed for processing the multiband infrared (IR) data of the geostationary Himawari-8 satellite. We applied the Q-method to the split-window data from the Second-generation Global Imager (SGLI) onboard the Global Change Observation Mission-Climate (GCOM-C) satellite. A comparison of the determined SGLI SSTs and buoy data shows a bias with a robust standard deviation of −0.097 K and 0.28 K in the daytime and −0.18 K and 0.28 K at night, respectively. Meanwhile, high biases of nearly −0.5 K were calculated for SSTs at and around 305 K. A residuals analysis suggests that the high negative bias is caused by insufficient information on the atmospheric correction brought by split-window data. This paper discusses the physical and mathematical background of the Q-method and compares it with another coefficient-based physical scheme.

本文介绍了一种用于确定海表温度（SST）的准物理方法（Q方法）。Q方法是一种基于系数的技术，用于处理对地静止的Himawari-8卫星的多波段红外（IR）数据。我们将Q方法应用于全球变化观测任务气候（GCOM-C）卫星上第二代全球成像仪（SGLI）的分割窗口数据。对确定的SGLI SST和浮标数据的比较显示，偏差在白天为-0.097 K和0.28 K，在夜间为-0.18 K和0.28 K，具有鲁棒的标准偏差。同时，对于305 K及其附近的SST，计算出了近-0.5 K的高偏差。残差分析表明，高的负偏差是由分割窗口数据带来的有关大气校正的信息不足所引起的。