Regionalization of River Basins Using Cluster Ensemble

Abstract

In the wake of global water scarcity, forecasting of water quantity and quality, regionalization of river basins has attracted serious attention of the hydrology researchers. It has become an important area of research to enhance the quality of prediction of yield in river basins. In this paper, we analyzed the data of Godavari basin, and regionalize it using a cluster ensemble method. Cluster Ensemble methods are commonly used to enhance the quality of clustering by combining multiple clustering schemes to produce a more robust scheme delivering similar homogeneous basins. The goal is to identify, analyse and describe hydrologically similar catchments using cluster analysis. Clustering has been done using RCDA cluster ensemble algorithm, which is based on discriminant analysis. The algorithm takes H base clustering schemes each with K clusters, obtained by any clustering method, as input and constructs discriminant function for each one of them. Subsequently, all the data tuples are predicted using H discriminant functions for cluster membership. Tuples with consistent predictions are assigned to the clusters, while tuples with inconsistent predictions are analyzed further and either assigned to clusters or declared as noise. Clustering results of RCDA algorithm have been compared with Best of k-means and Clue cluster ensemble of R software using traditional clustering quality measures. Further, domain knowledge based comparison has also been performed. All the results are encouraging and indicate better regionalization of the Godavari basin data.

Regionalization Methods for Low Flow Estimation in Ungauged Catchments – A Review

“…In hydrology, the most popular approaches to regionalization of watersheds include: methods of residuals pattern approach, multivariate statistics such as cluster analysis, Classification and Regression Tree (CART) models or seasonality of low flow (see: Table 1). Regionalization methods are applied in searching for hydrological similarity between catchments by examining the attributes that describe their geomorphology, land cover, climate, and soils, for low flow characteristics estimation in ungauged catchments (Ahuja, 2012;Vezza et al, 2010). Regionalisation techniques can be applied in estimating low flow characteristics under consideration, in order to obtain a predictive model for low flows in ungauged catchments as preparatory reference for the environmental flows assessment (Vezza et al, 2010).…”

“…al., 2010); small trees are readily interpretable (Laaha and Blöschl, 2006a); Disadvantages: the method consists of having unstable results with the modification of the learning sample (the structure of the tree may change when models are refitted for subsets of the data (Vezza et. al., 2010); big trees are difficult to interpret, there is a lack of smoothness and there are potential problems with overfitting the data; a method for pruning the tree is needed (Laaha and Blöschl, 2006a) Cluster analysis as above as above Vezza et al, 2010;Števková et al, 2012;Laaha and Blöschl, 2006a;Nathan and McMahon, 1992;Kahya and Demirel, 2007;Ahuja, 2012;Cupak, 2013;Cupak, 2017 Advantages: one of the most frequently used methods in hydrology; few clustering methods can be used; the data is grouped in such a way that the data points in one cluster are very similar and objects in different clusters are quite distinct; in partitional clustering the number of clusters need to be assumed; K-means method is effective for grouping large number of data (Rao and Srinivas, 2008) www.acta.urk.edu.pl/pl…”

Using entropy theory to improve the definition of homogeneous regions in the semi-arid region of Brazil

“…Cluster analysis has been successfully used to identify homogeneous regions (Lin and Chen 2006, Rao and Srinivas 2006a, Kahya et al 2008, Ahuja 2012. Cluster analysis is a standard method of statistical multivariate analysis that can divide large and complex datasets into many groups, where members of a group share similar characteristics as far as possible and different groups are dissimilar as much as possible (Rao and Srinivas 2006a).…”