Quality Assessment of Cloud Detection and Cloud Top Height Products from FY-3D/MERSI-II in Polar Regions

The Medium Resolution Spectral Imager II (MERSI-II) is one of the primary payloads on the Fengyun-3D (FY-3D) satellite, capable of dynamically monitoring important atmospheric and environmental parameters such as cloud characteristics and aerosols. This study aims to evaluate the cloud detection and cloud top height (CTH) retrieval accuracy of FY-3D/MERSI-II in polar regions (Arctic and Antarctic) against the CALIOP (Cloud–Aerosol Lidar with Orthogonal Polarization) cloud products. The results demonstrate that MERSI-II-derived cloud coverages in the Arctic are 7.2% lower than CALIOP observations, while they are overestimated (underestimated) in Antarctic in summer (other seasons). Seasonal variations in cloud detection accuracy are notable, with Arctic accuracy reaching 90.4% in summer but declining to 51.3% in winter. The detection accuracy is consistently lower in Antarctic than that in Arctic, though MERSI-II shows significantly better detection capability over water surfaces than snow/ice-covered areas. For CTH retrievals, correlation coefficients between MERSI-II and CALIOP reached 0.70 (Arctic) and 0.71 (Antarctic), with systematic underestimations averaging −0.46 km (Arctic) and −1.52 km (Antarctic). Larger deviations occur in multi-layer clouds and ice cloud conditions. Notably, MERSI-II tends to overestimate (underestimate) CTH for low-level (high-level) clouds and demonstrates superior retrieval accuracy for water clouds than ice clouds. These findings can provide robust data support for satellite algorithm optimization and polar environment monitoring.