Mixed statistical and data mining analysis of river flow and catchment properties at regional scale

Abstract This paper presents an application of the L-moments and L-moment ratio diagrams (LMRD) to analysis hydrological data at regional (country) scale. Existing research focuses on two main areas analysis: statistical using LMRD regression analysis. Further mixes both approaches applying L-moments. Another direction is clustering climatic physiographic catchment properties its validation LMRD. However, plots can be separately used as domain. It proposed decompose features into some classes, than present these results Such constitute source for clustering. Obtained clusters are then validated against k-means performed in diagram Results show that improved with mining algorithms. combination delivers a new perspective interpretation results. shown domain consistent K-means anther argument showing considered very powerful informative tool hydrologists enabling comparison basis respect various properties. The method consisting daily river flow from 290 gauges covering entire Poland.