PCRF: An Independent Cloud Detection Framework for FY-4B GIIRS Hyperspectral Infrared Sounder Using PCA Reconstruction and Random Forest

Accurate cloud detection is essential for assimilating hyperspectral infrared observations from the Geostationary Interferometric Infrared Sounder (GIIRS) onboard FY-4B. This study presents a novel independent cloud detection framework termed PCRF, which integrates Principal Component Analysis (PCA) reconstruction of clear-sky spectra with an optimized Random Forest (RF) classifier. Unlike operational methods that rely on external imager products (e.g., AGRI), PCRF exploits the intrinsic spectral characteristics of GIIRS longwave channels to identify cloud-contaminated pixels through reconstruction error analysis. Fourier-based spectral features are further introduced to enhance cloud microphysical representation.Validated against MODIS cloud masks, PCRF achieves 91.3% accuracy over ocean and 86.7% over land, outperforming the operational AGRI cloud product by 2.1% and 6.6%, respectively.Independent validation using July 2024 data confirms the model's temporal stability and spatial generalizability. Spatial comparisons reveal that PCRF provides sharper cloud boundaries and better detection of fragmented clear-sky regions, especially over complex surfaces. These results highlight PCRF's potential as a robust, imager-independent solution for cloud detection in hyperspectral infrared missions, with direct benefits for NWP data assimilation and satellite product quality control.