Use of a Distance Measure for the Comparison of Unit Hydrographs: Application to the Stream Gauge Network Optimization

Abstract

This paper reports a case study aimed at optimizing the stream gauge network of the Hantan River Basin in Korea. The dimensionless synthetic unit hydrographs at all potential stream gauge locations were derived and compared by using the distance measure of entropy theory. Based on cluster analysis of the transinformations among stream gauge stations, the entire basin could be divided into four regions with similar subbasin responses. One stream gauge per region would be sufficient to measure the basin response to rainfall input. The optimal set of stream gauges was also derived by applying the concept of maximum total information, where a total of seven stream gauges was found to be needed for the entire Hantan River Basin. However, both applications produced very similar shapes of subbasin clustering over the Hantan River Basin. In particular, most subbasins along the main stream were included in the same group in both cases. Most subbasins in a specific tributary were also found to be included in the same cluster.