Identification of homogeneous regions for regional frequency analysis using the self-organizing map

In this paper, the self-organizing map (SOM) is applied to identify the homogeneous regions for regional frequency analysis. First, the algorithm and structure of the SOM are presented. Then the experimental design is applied to test the cluster accuracy of the SOM, the K-means method and Ward's method. These three clustering methods are tested on experimental data sets where the amount of cluster dispersion and the cluster membership are controlled and known. Among the three clustering methods, the results show that the SOM determines the cluster membership more accurately than the K-means method and Ward's method. Finally, the SOM is applied to actual rainfall data in Taiwan to identify homogeneous regions for regional frequency analysis. A two-dimensional map indicates that the rain gauges can be grouped into eight clusters. A heterogeneity test indicates that the eight regions are sufficiently homogeneous. Moreover, the results show that the SOM can identify the homogeneous regions more accurately as compared to the other two clustering methods. Because of unsupervised learning, the SOM does not require the knowledge of corresponding output for comparison purposes. In addition, the SOM is more robust than the traditional clustering methods. Therefore, the SOM is recommended as an alternative to the identification of homogeneous regions for regional frequency analysis.