Classification of Annual Temperature of Iran Meteorological Stations Using Fuzzy Cluster Analysis and Kohonen Artificial Neural Networks

Classification of Meteorological stations, causing a large volume of data to be allocated a smaller homogeneous groups, Ease of use in modeling and also can help to spread the information to point to the lack of regional data to Statistics regional statistic. Meteorological stations in the region has a crucial role in the management and effective use of information. In this study, 112 stations were analyzed in order to classify regions with fuzzy cluster analysis and Kohonen artificial neural Networks. Four parameters, namely mean annual temperature, longitude, latitude and elevation are considered as the classification criteria for grouping to obtain the optimal number of groups the lowest value of Davies- Bouldin index were used. Demarton climatic zonation was performed to evaluate the spatial distribution of clusters obtained from various methods. The results showed that the fuzzy clustering technique with the Demarton climatic zones is more consistent.