Premium
Classification of precipitation series using fuzzy cluster method
Author(s) -
Dikbas Fatih,
Firat Mahmut,
Koc A. Cem,
Gungor Mahmud
Publication year - 2012
Publication title -
international journal of climatology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.58
H-Index - 166
eISSN - 1097-0088
pISSN - 0899-8418
DOI - 10.1002/joc.2350
Subject(s) - homogeneity (statistics) , homogeneous , cluster (spacecraft) , fuzzy logic , precipitation , context (archaeology) , series (stratigraphy) , environmental science , mathematics , computer science , statistics , meteorology , geography , geology , artificial intelligence , paleontology , archaeology , combinatorics , programming language
The identification of hydrologically homogeneous regions is one of the most important steps of regional frequency analysis. The hydrologically homogeneous regions should be determined using cluster analysis instead of the geographically close regions or stations. In this study, fuzzy cluster method (Fuzzy C‐Means: FCM) is applied to classify the precipitation series and identify the hydrologically homogeneous groups. The choice of appropriate cluster method and the variables that will be used according to the data of the basin is also very important. In the context of this study, total precipitation data of stations operated by National Meteorology Works (DMI) in Turkish basins for cluster analysis are used. The optimal number of groups is determined as six, based on different performance evaluation indexes. Regional homogeneity tests based on L‐moments method are applied to check homogeneity of these six regions identified by cluster analysis. Regional homogeneity test results show that regions defined by FCM method are sufficiently homogeneous for regional frequency analysis. According to the results, FCM method is recommended for classifying the precipitation series and for identifying the hydrologically homogenous regions. Copyright © 2011 Royal Meteorological Society