Regional flood frequency analysis using residual kriging in physiographical space

• DOI: 10.1002/hyp.7631
• Publication Date: 2010-03-25T04:12:26Z
• Authors: M. Kamali Nezhad; K. Chokmani; T. B. M. J. Ouarda; M. Barbet; P. Bruneau
• Keywords: Quantile; Variogram; Spatial Dependence

Abstract In this article, an approach using residual kriging (RK) in physiographical space is proposed for regional flood frequency analysis. The constructed physiographical/climatic characteristics of gauging basins by means canonical correlation analysis (CCA). This a modified version the original method, based on ordinary (OK). It intended to handle effectively any possible spatial trends within hydrological variables over space. approach, trend first quantified and removed from variable quadratic regression. OK therefore applied regression values. final estimated value specific quantile at ungauged station sum estimate kriged residual. To evaluate performance cross‐validation procedure applied. Results method indicate that RK CCA leads more efficient estimates quantiles when compared straightforward regression‐based estimator. Copyright © 2010 John Wiley & Sons, Ltd.