Analysis the Characteristics of typhoon rainstorm in Guigang based on K-Means Clustering Method and EOF Analysis
Using the observation data and the best track dataset of typhoons from the China Meteorological Administration to analyze the characteristics of rainstorms caused by 65 typhoons affecting Guigang from 2010 to 2023 using the K-means clustering method combined with the Empirical Orthogonal Function(EOF)analysis method.The results indicate that:(1)there are 39 typhoons caused rainstorms,with an average of 2.8 typhoon-induced rainstorms per year,and the concentrated period for these rainstorms is in July and August.The southern part of Gangnan District,the central and western part of Qin Tang District,and the northern mountainous area of Guiping are the areas with more frequent and severe typhoon rainstorms.(2)Based on the K-means clustering analysis,the 65 typhoons were classified into three categories:A,B,and C.Typhoons of categories A and B originated from the northwest Pacific,while category C typhoons mostly came from the South China Sea.Typhoons of categories A and B were primarily characterized by northwestward movement,whereas category C typhoons had complex paths and were active in the Beibu Gulf.Category B typhoons were the strongest,while category C typhoons were the weakest.(3)The analysis of the spatial distribution of the first principal component vectors of the three types of typhoon rainstorms in Guigang showed that the annual rainfall of categories A and B typhoon rainstorms was generally more or less distributed throughout the city,while the annual rainfall variation trend of category C typhoon rainstorms was mainly characterized by more rainfall in the northeastern part of Guiping,the southern part,and the northwestern part of Pingnan,with less rainfall in the remaining areas,or less rainfall in the northeastern part of Guiping,the southern part,and the northwestern part of Pingnan,with more rainfall in the remaining areas.
typhoon rainstormK-means clustering methodEOF analysisspatial and temporal variations
万方数据
维普
