A Machine Learning‐Based Observation Operator for FY‐4B GIIRS Brightness Temperatures Considering the Uncertainty of Label Data

• Published in: Journal of geophysical research. Machine learning and computation Vol. 2; no. 1
• Main Authors: Li, Yonghui, Han, Wei, Duan, Wansuo, Li, Zeting, Li, Hao
• Format: Journal Article
• Language: English
• Published: Wiley 01.03.2025
• ISSN: 2993-5210; 2993-5210
• DOI: 10.1029/2024JH000449

The increasing volume of satellite data, particularly hyperspectral infrared data, combined with the real‐time monitoring requirements of numerical weather prediction (NWP) systems, has heightened the demand for computational efficiency and accuracy in radiative transfer models (RTM). Machine learning (ML) offers a promising approach, and numerous studies on ML‐based RTM have emerged recently. However, existing ML‐based RTMs for hyperspectral infrared were not end‐to‐end. Moreover, since the label data do not represent truth, models trained with loss functions like mean squared error (MSE) or mean absolute error (MAE) fail to account for its uncertainty. This limitation can lead to suboptimal model parameters, as training may assign higher weights to labels with larger errors. This study construct an end‐to‐end ML‐based RTM focused on clear sky conditions over the ocean for the FengYun‐4B satellite (FY‐4B) Geostationary Interferometric Infrared Sounder (GIIRS), using maximum likelihood estimation (MLE) and MSE for training, respectively. MLE accounts for the uncertainty of labels. The results indicate both models achieve high accuracy, with mean errors within 0.1 K (K) and standard deviation (STD) of errors within 0.04 K compared to the labels. The model trained with MLE exhibits a mean error closer to 0 and a STD similar to the error STD of labels, suggesting better parameter configurations to reflect the actual error distribution of the labels. Additionally, the temperature and water vapor Jacobian computed by both models are comparable to those obtained from RTTOV, highlighting their potential for application as observational operator in satellite data assimilation for hyperspectral infrared sounder. Plain Language Summary Satellite hyperspectral infrared data accounts for a significant portion of satellite data and is increasing annually. Moreover, numerical weather prediction requires real‐time monitoring of satellites, leading to increasingly higher demands for computational efficiency and accuracy in radiative transfer models. This article develops a machine learning‐based radiative transfer model specifically for hyperspectral infrared sounder. The model is trained using both maximum likelihood estimation and mean squared error. The former considers the uncertainty of the labels and both achieve high accuracy, with MLE exhibiting even higher accuracy and better parameter configurations. Additionally, by calculating the gradients of brightness temperature with respect to atmospheric profiles, that is, the inputs, it is found that the accuracy is similar to that of the traditional model RTTOV, demonstrating its potential for application as observational operator in satellite data assimilation for hyperspectral infrared sounder. Key Points An end‐to‐end machine learning‐based radiative transfer model for clear sky has been developed for FY‐4B GIIRS The model, trained using a likelihood‐based loss function that weights each channel differently, captures the uncertainty of labels The Jacobian matrix calculated by the constructed model resembles the results from the radiative transfer model RTTOV