Journal of Food, Agriculture and Environment

Clustering rainfall stations aiming regional frequency analysis 

Author(s):

Wagner A. Pansera ¹, Benedito M. Gomes ², Marcio A. Vilas Boas ², Eloy Lemos de Mello ²

Recieved Date: 2012-10-18, Accepted Date: 2013-04-26

Abstract:

Reliable rainfall estimations are vital to the successful hydrological planning of a region. For this task, probabilistic models to estimate the frequency and intensity of rainfall are available, but require inputs from the target location. Hydrological monitoring networks, however, do not cover all places of interest and therefore a tool capable of transferring the information from the monitored location to that of interest is necessary. This can be accomplished by index-flood regionalization methodology, which permits to estimate quantiles, evaluate the homogeneity of a region and identify discordant stations. The study focuses on the state of Paraná, Brazil, using monthly rainfall in 227 stations during 31 years (1976-2006). The study reviews various clustering methods, both hierarchical and nonhierarchical, and presents methods to validate the classifications using measures of both discordance for stations belonging to a particular cluster and homogeneity of the obtained regions. Hierarchical methodologies used were: centroid, single, average, complete and Ward linkages. Nonhierarchical methodology used was k-means. Also was utilized a hybrid form between nonhierarchical and hierarchical methodologies. Optimized number of groups was found utilizing the cophenetic correlation coefficient, Davies-Bouldin and Dunn indexes. Each group of optimum repartition was evaluated with respect their homogeneity and outlier stations. The best repartition of rainfall stations was found with hybrid form between Ward and k-means methodologies, because elimination of outlier stations allowed forming homogeneous regions with respect to L-moments. Other point observed was the effect of outlier stations in homogeneity of studied region, as well as, group’s number of stations. Finally, gamma distribution was tested with two or three parameters and showing satisfactory results. 

Keywords:

Heterogeneity measure, discordancy measure, L-moments, regionalization, hierarquical clustering, k-means, hybrid clustering, monthly rainfall, cluster validation, design precipitation

---

Journal: Journal of Food, Agriculture and Environment
Year: 2013
Volume: 11
Issue: 2
Category: Environment
Pages: 877-885

---

Full text for Subscribers

---

Information:

Note to users

The requested document is freely available only to subscribers/registered users with an online subscription to the Journal of Food, Agriculture & Environment. If you have set up a personal subscription to this title please enter your user name and password. All abstracts are available for free.

Article purchasing

If you like to purchase this specific document such as article, review or this journal issue, contact us. Specify the title of the article or review, issue, number, volume and date of the publication. Software and compilation, Science & Technology, all rights reserved. Your use of this website details or service is governed by terms of use. Authors are invited to check from time to time news or information.

---

Purchase this Article:   20 Purchase PDF Order Reprints for 15

---

Share this article :