Retrievals of Sea Surface Current Vectors from Geostationary Satellite Data (Himawari-8/AHI)

An operational sea surface current (SSC) retrieval algorithm was developed using consecutive Himawari-8/AHI data based on a feature tracking method. Comparative analyses were conducted to determine the appropriate input data for the SSC retrieval algorithm. Investigation of the input data revealed some limitations in the use of single-band brightness temperatures caused by atmospheric features under moist conditions, especially in the mid- and low-latitude regions. Because of the motion of atmospheric features, cloud and cloud-contaminated pixels tended to contribute to the overestimation of SSC. To reduce overestimation, sea surface temperature images were used as input data and the feature tracking method was applied to calculate the displacement of the surface current vectors. The estimated currents were subjected to a quality control process to remove erroneous vectors. The accuracy of the retrieved surface currents was assessed by comparing the results with the quality-controlled currents obtained from surface drifters in the full-disk region of Himawari-8/AHI. The results revealed that the estimated current speeds and directions agreed with the drifter-based calculated values—the root-mean-square (bias) errors were 0.35 ms −1 (0.11 ms −1 ) and 33.28° (5.47°), respectively. The estimated current field showed diverse dynamic ocean features, such as a rotating feature around a mesoscale eddy and the characteristic meandering pattern of the Kuroshio Current. Hourly varying surface current fields from geostationary satellite data with high spatio-temporal resolution are expected to augment oceanic and atmospheric applications in real time.