Abstract
In arid regions, the analysis of wadi morphometric parameters and their interrelationships are fundamental for describing the topographic, geologic evolution, structures and hydrologic potential efficiency of the watershed. Unfortunately, the spatial characteristics and the analyses of these variables are not taken into consideration in an efficient way, especially in arid regions. In this paper, digital elevation model, geographic information system and multivariate statistical techniques are integrated for the identification and the assessment of main morphometric parameters. For this purpose, Q and R modes of cluster analyses and principal component analysis (PCA) are applied to ten different-size watersheds in the western region of Saudi Arabia utilizing 18 morphometric descriptors (variables). The results show that the R-mode cluster analysis classifies the variables into three groups, whereas the Q-mode cluster analysis classifies watersheds according to the similarities in their major variables such as area, perimeter, total stream length and peak discharge. The first three components of PCA accounted for 86% of the total variance in the data and show more details concerning the parameter loadings and the degree of variable significance.