GOES-16 SST: cloud correction and gap filled for observation of upwelling in the MAB

This cloud correction method applied to GOES-16 hourly SST to cloud correct and retaining upwelling pixels in the MAB. The Quality Filter (DQF) provided by GOES consistently removed coastal upwelling pixels along the coast of New Jersey and Delaware. This Spike Filter method was applied to hourly GOES SST for the summer of 2019 and increased SST coverage in the MAB coastal region by ~15% and maintained an acceptable accuracy when validated to regional buoy SST. This includes a 1. rate of SST change 2. minimum SST threshold 3. comparison to recent SST and 4. SST bias correction. This cloud corrected SST was then gap-filled using DINEOF and compared to other available SST products.

本云校正方法应用于GOES-16地球静止运行环境卫星（Geostationary Operational Environmental Satellite 16, GOES-16）的逐小时海表温度（Sea Surface Temperature, SST）数据，以完成云校正并保留中大西洋湾（Middle Atlantic Bight, MAB）内的上升流像素。GOES提供的质量过滤器（Quality Filter, DQF）会持续移除新泽西州与特拉华州沿岸的近岸上升流像素。本尖峰过滤方法被应用于2019年夏季的GOES逐小时SST数据，使MAB近岸海域的SST覆盖范围提升约15%，且在与区域浮标SST数据进行验证时仍保持可接受的精度。该尖峰过滤方法涵盖以下四项判断逻辑：1. SST变化速率；2. SST最低阈值；3. 与近期SST数据的对比；4. SST偏差校正。随后使用数据插值经验正交函数法（Data Interpolating Empirical Orthogonal Functions, DINEOF）对经云校正后的SST数据完成间隙填充，并与其他可用SST产品开展对比分析。