Validation of Sea Surface Temperature Derived From Himawari-8 by JAXA

Abstract

Hourly sea surface temperature (SST) retrieved from Himawari-8 by the Japan Aerospace Exploration Agency (H8-SST/JAXA, latest version 1.2) is becoming an important data source for data merging as well as for resolving diurnal variation (DV). However, the spatial and temporal variation of the errors for the full disk is still unclear. In this article, two years of H8-SSTs/JAXA are validated against in situ measurements from iQuam2. In general, H8-SSTs/JAXA shows small biases between −0.11 and −0.03 K with root mean square error (RMSE) between 0.58 and 0.73 K. The spatial distributions of the errors reveal the following patterns: 1) a small median bias close to 0.1 K and RMSE of 0.4-0.6 K comparing to drifters are found at satellite zenith angle (SZA) 0°-35°; 2) negative biases (∼−0.3 K) are detected at SZA s 35°-58°; and 3) larger positive biases exceeding 0.3 K are also found along the viewing boundaries. The temporal variations of the errors show that 1) there is no prominent seasonal variation; 2) the amplitude of the DV of the errors is only ∼0.1 K for the statistical of all matchups, and 3) the maximum errors appears in the morning rather than in the noon. The statistics will be used in future work for DV analysis and merging purposes.

Evaluation and Improvement of FY-4A/AGRI Sea Surface Temperature Data

“…The cooling effect is significant after 15:00 LT. These features are similar to the study of Tu et al [27]. The negative value of SATSST CDF minus in situ SST indicates that the satellite-derived SST is lower than in situ SST.…”

“…These are much larger than the accuracy requirements of 0.5 ℃ to 0.8 ℃ for infrared radiometers onboard the geostationary satellite [4]. This result is worse than the evaluation of SST retrieval from similar geostationary satellites such as Himawari-8 [27] and geostationary operational environmental satellites 16 [40].…”

“…However, these in situ SSTs are very heterogeneous in depth for coming from a variety of observation platforms. The research shows that there are larger differences between the validation results from various types of in situ SST [27], [38]. The difference of observation data from different platforms is not considered during the error evaluation and bias correction of FY-4A/AGRI SST.…”

Inversion of sea surface temperature in South China Sea based on machine learning method

“…Machine learning algorithms including SVR, ANN, and LightGBM are employed to perform sea surface temperature (SST) inversion in the South China Sea in 2022, distinguishing between clear-sky and all-region cases based on the presence or absence of cloud cover. The accuracy of the inversion results for each algorithm is evaluated using evaluation indicators, and the test results are as follows: (1) The correlation coefficient of the LightGBM algorithm under the all-region condition is 0.94, with a mean absolute error (MAE) of 0.2335℃ and a mean squared error (MSE) of 0.1803℃, outperforming the other two algorithms; (2) The inversion accuracy of the LightGBM algorithm under the allregion condition is higher than that under the clear-sky condition, indicating its ability to effectively avoid the interference of cloud cover. By comparing the accuracy of the Himawari-8 L2-level product with the inversion results of the three algorithms and analyzing the scatter plots, it is evident that the machine learning algorithms achieve higher inversion accuracy compared to the Himawari-8 L2-level product.…”

“…Sea surface temperature (SST) is an essential component of many physical, biological, and chemical processes in the Earth system. It is also a major source of energy for oceanic convection and plays a crucial role in determining the intensity and internal structure of tropical cyclones [1][2] . Research indicates that SST anomalies can lead to extreme low temperatures [3] and Interannual variations in precipitation [4] by influencing atmospheric circulation.…”