Using Self-Organizing Maps for Alluvial Fan Classification

Abstract

Abstract. Morphometry research used geography information system (GIS) to explore alluvial fan morphology for determination of fans location and fast and accurate analysis. The aim of this study is to classify alluvial fans formed by highgradient braided streams and torrents that discharge into the Oshtorankook altitudes in the Lorestan province, Iran. The morphology of the fans and their watershed is quantitatively described through estimated morphometric parameters. For relationships between geomorphological features of the fans and their drainage basins, self-organizing maps (SOM) were used. In SOM, according to both qualitative data and morphometric variables, the clustering tendency of alluvial fans was investigated using 15 alluvial fans parameters. The results of the analysis showed that several morphologically different fan types were recognized based on their geomorphological characteristics in the study area. A strong positive relationship was found between the drainage basin area and size of the fan with a simple power function. In addition, the relationship between fan slope and drainage area was found to be negative and moderately strong with a simple power function.