Any period climatic data statistics for successive years is a common real/quasireal time operational task in meteorological service, which should be supported by the corresponding operational system. In order to improve the efficiency, we optimize the statistic flow and propose a logic decomposition process of the original data in this paper. Taking the relational database Oracle as an example, based on the 65year daily data of 86 national weather stations and the 10year daily data of 2313 regional weather stations, the optimized flow is tested. The results show that, based on the flow before the optimization, the query efficiency decreases with the increasing period span. Conversely, based on the flow after the optimization, the query efficiency maintains a high level, regardless of the span of the period. The optimized flow is applied in the AGRometeorology Operation System of Guangdong Province (AGROS) and plays an important role in improving the system efficiency and optimizing the user experiences. Eventually, the popularization and application of the system in Guangdong Province are further promoted. The optimization flow proposed in this paper has important reference value in the related operational system development in other provinces and national meteorological services.