Defining Homogeneous Regions for Streamflow Processes in Turkey Using a K-Means Clustering Method
- 237 Downloads
The major problem in ungauged basins for planning and management of water resources projects is to estimate the flood magnitudes and frequencies. The identification of hydrologically homogeneous regions is one of the most important steps of regional frequency analysis. In this study, K-Means clustering method is applied to classify the maximum annual flows and identify the hydrologically homogeneous groups. For this aim, the annual maximum river flows, coefficient of variation and skewness of annual maximum river flows, latitude and longitude at 117 stations operated by the General Directorate of Electrical Power Resources Survey and Development Administration throughout Turkey are used. The optimal number of groups was determined as seven. Regional homogeneity test based on L-moments method is applied to check homogeneity of these seven regions identified by clustering analysis. The results show that regions defined by K-Means method can be used for regional flood frequency analysis. According to the results, K-Means method is recommended to identify the hydrologically homogeneous regions for regional frequency analysis.
KeywordsCluster analysis K-Means Annual maximum flow Homogeneous region
Unable to display preview. Download preview PDF.
- 7.Andrade E.M.: Regionalization of small watersheds in arid and semiarid regions: Cluster and Andrews’ curve approaches. Eng. Agric. Jaboticabal 18(4), 39–52 (1999)Google Scholar
- 12.Soltani S., Modarres R.: Classification of spatio -temporal pattern of rainfall in Iran using a hierarchical and divisive cluster analysis. J. Spat. Hydrol. 6(2), 1–12 (2006)Google Scholar
- 14.Demirel, M.C.: Cluster analysis of streamflow data over Turkey. Master of Science Thesis. İstanbul Technical University, İstanbul, p. 119 (2004)Google Scholar
- 15.Turan, A.: Türkiye akarsu verimlerinin küme analizi ile sınıflandırılması. Sakarya Üniversitesi, Fen Bilimleri Enstitüsü Yüksek Lisans Tezi, 155s (2005)Google Scholar
- 16.Demirel, M.C.; Mariano, A.J.; Kahya, E.: Performing K-Means analysis to drought principal components of Turkish rivers. Proceedings of 27th AGU Hydrology Days, Fort Collins, Colorado, 145–151 (2007)Google Scholar
- 17.Kahya, E.; Demirel, M.C.; Piechota, T.C.: Spatial grouping of annual streamflow patterns in Turkey. Proceedings of 27th AGU Hydrology Days, Fort Collins, Colorado, 169–176 (2007)Google Scholar
- 18.Kahya, E.; Demirel, M.C.: A Comparison of low-flow clustering methods: Streamflow grouping. J. Eng. Appl. Sci. 2(3), 524–530 (2007)Google Scholar
- 20.Kahya E., Demirel M.C., Bég A.O.: Hydrologic homogeneous regions using monthly streamflow in Turkey. Earth Sci. Res. J. 12(2), 181–193 (2008)Google Scholar
- 21.Yerdelen C., Karimi Y., Kahya E.: Frequency analysis of mean monthly streamflow in Çoruh Basin, Turkey. Fresenius Environ. Bullet. 19((7), 1300–1311 (2010)Google Scholar
- 22.Demirel, M.C.; Booij, M.J.; Kahya, E.: Validation of an ANN prediction model using a multi-station cluster analysis. ASCE J. Hydrol. Eng. doi:10.1061/(ASCE)HE.1943-5584.0000426 (in press)
- 23.Hosking, J.R.M.; Wallis, J.R.: Regional frequency analysis: An approach based on L-moments. Cambridge University Press, Cambridge (1997)Google Scholar