The vertical distribution of atmospheric aerosols is highly complex and exhibits significant spatiotemporal variability, providing essential information for improving the accuracy of aerosol retrievals from satellite remote sensing. This study systematically investigates the vertical distribution characteristics of aerosols based on CALIOP L3 aerosol profile data from 2010 to 2020 using unsupervised clustering methods. By evaluating the clustering performance of three algorithms—GMM, K-means, and spectral clustering—using multiple metrics, the GMM clustering method was selected to classify aerosol profiles into five representative types based on the vertical distribution features of extinction profiles: Low-Pollution Composite Type, High-Pollution Composite Type, Exponential Decay Type, Low-Pollution Uniform Type, and High-Pollution Oscillatory Type. Further analysis was conducted on the spatiotemporal distribution characteristics of these profiles across different seasons and three typical regions: the Tibetan Plateau, Beijing-Tianjin-Hebei, and the Yangtze River Delta. The results reveal that the aerosol profiles obtained through clustering analysis exhibit significant seasonal and regional differences.