According to the characteristics of various meteorological data types, large amount of information and high precision, the traditional relational database system has such problems as load saturation, unsatisfactory read and write performance and so on in storage processing and data reading. Based on the researches on cloud data storage technology and the characteristics of meteorological data, the meteorological data storage and processing method based on the cloud data storage platform is designed. Based on the structural and semi/unstructured characteristics of meteorological data, the structured meteorological data is stored in a cloud relational database. The NoSQL object database is used to store the semi/unstructured meteorological data. Moreover, the block compression technology is applied to meteorological data storage and transmission. The results show that the proposed method has good storage and transmission efficiency and can meet the requirements in storage query and processing speed of largescale meteorological data in the operational application.