The distribution feature and forecasting of low-temperature and sustained-rain finishing period in Guangxi

In this paper, the distribution feature of low-temperature and sustained-rain finishing period in Guangxi is analyzed by using EOF method, it shows that eigenvectors and its corresponding time-coefficient series well represent the time changes and the spatial distribution. The two main eigenvectors characterizing the distribution obviously have 2-3 years or quasi 4 years periodic oscillation. By using mean generating function model made of coefficient-time series of main distribution pattern, the result showed good for the experimental forecast.