Progress and challenges of ground-based cloud remote sensing

Clouds play essential roles to the Earth's energy balance and hydrological cycle, accurate cloud properties are important for understanding the atmospheric physical processes, and improving the weather and climate model simulation. Ground-based cloud remote sensing is one method to obtain cloud properties and validate the satellite remote sensing. It has been developed first since 1980s and a variety of ground-based retrieval algorithms have been proposed. The ground-based remote sensing includes both active and passive remote sensing, and can be used for obtaining both cloud macro- and micro-physical properties. For cloud macrophysical properties, the retrievals can be simply divided into three types based on the purposes, which are cloud detection or cloud amount observation method, cloud boundary identification method, and cloud phase determination method. For cloud microphysical properties, the ground-based cloud retrieval algorithm can be generally classified into two types, the optical retrieval algorithm and the empirical parameterization algorithm. Each retrieval algorithm has its merits and disadvantages. This study provides an overview of existing ground-based remote sensing and retrieval algorithms. The cloud properties from different retrieval algorithms could have significant discrepancies, which are even larger than the uncertainties of cloud properties indicated by the retrieval algorithm. The large discrepancies imply that there are still grand challenges in the ground-based cloud retrievals, which have also been summarized and proposed in this study.