Articles | Volume 20, issue 12
Hydrol. Earth Syst. Sci., 20, 4717–4729, 2016
Hydrol. Earth Syst. Sci., 20, 4717–4729, 2016

Research article 29 Nov 2016

Research article | 29 Nov 2016

Delineation of homogenous regions using hydrological variables predicted by projection pursuit regression

Martin Durocher1, Fateh Chebana2, and Taha B. M. J. Ouarda2,3 Martin Durocher et al.
  • 1Université du Québec à Trois-Rivières, University of Québec, 3351, Blvd. des Forges, C.P. 500, Trois-Rivières, G9A 5H7, Canada
  • 2Institut National de Recherche Scientifique (INRS-ETE), University of Québec, 490 de la Couronne, Québec, G1K 9A9, Canada
  • 3Institute Center for Water Advanced Technology and Environmental Research (iWater), Masdar Institue of Science and Technology, P.O. Box 54224, Abu Dhabi, UAE

Abstract. This study investigates the utilization of hydrological information in regional flood frequency analysis (RFFA) to enforce desired properties for a group of gauged stations. Neighbourhoods are particular types of regions that are centred on target locations. A challenge for using neighbourhoods in RFFA is that hydrological information is not available at target locations and cannot be completely replaced by the available physiographical information. Instead of using the available physiographic characteristics to define the centre of a target location, this study proposes to introduce estimates of reference hydrological variables to ensure a better homogeneity. These reference variables represent nonlinear relations with the site characteristics obtained by projection pursuit regression, a nonparametric regression method. The resulting neighbourhoods are investigated in combination with commonly used regional models: the index-flood model and regression-based models. The complete approach is illustrated in a real-world case study with gauged sites from the southern part of the province of Québec, Canada, and is compared with the traditional approaches such as region of influence and canonical correlation analysis. The evaluation focuses on the neighbourhood properties as well as prediction performances, with special attention devoted to problematic stations. Results show clear improvements in neighbourhood definitions and quantile estimates.

Short summary
For regional flood frequency, it is challenging to identify regions with similar hydrological properties. Therefore, previous works have mainly proposed to use regions with similar physiographical properties. This research proposes instead to nonlinearly predict the desired hydrological properties before using them for delineation. The presented method is applied to a case study in Québec, Canada, and leads to hydrologically relevant regions, while enhancing predictions made inside them.