Evaluation of spatial–temporal characteristics of precipitation using discrete maximal overlap wavelet transform and spatial clustering tools

Abstract In the present study, classical and proposed methods were used to investigate monthly precipitation characteristics of 30 stations in southeastern United States during 1968–2018. Maximal overlap discrete wavelet transform (MODWT) as preprocessing method K-means clustering used. First, time series decomposed into several subseries using MODWT considering db mother wavelet. Then, energy values theses calculated inputs radial basis functions (RBF) methods. The optimum number clusters obtained for considered both was five clusters. order use data input RBF method, correlation evaluated by variogram. Based on results accordance with latitude longitude variations stations, it found that increasing clusters, amount decreased vice versa. silhouette coefficient 0.3 0.8, which indicates better selected area method.