FY-4A/AGRI L1 Full-Disk Data and K-means based Voronoi Spatial Partitioning Code for EPCS Analysis

This deposit contains two components supporting the study “Horizontal and Vertical Structure of Cloud Effective Radius in Eastward-Propagating Convective Systems” (Paper # 2026GL124653, Geophysical Research Letters). FY-4A/AGRI L1 Full-Disk DataRaw or preprocessed Level‑1 full‑disk observations from the FY‑4A Geostationary Meteorological Satellite’s Advanced Geosynchronous Radiation Imager (AGRI), covering the Eastward‑Propagating Convective Systems (EPCS) analyzed in the paper. The data include multispectral channels used to retrieve cloud effective radius and other cloud properties. In addition, we provide retrieval results for selected case studies (e.g., cloud effective radius, cloud top properties, and other derived products), which are included as supplementary files within this deposit. K‑means based Voronoi Spatial Partitioning Algorithm (Code)A custom Python (or MATLAB/other – adjust as needed) implementation that combines K‑means clustering with Voronoi tessellation to adaptively determine optimal cluster numbers for spatial segmentation of cloud fields. This algorithm was used to partition the convective systems into homogeneous regions for structural analysis.