Dépôt numérique

A Nonlinear Approach to Regional Flood Frequency Analysis Using Projection Pursuit Regression.

Durocher, Martin, Chebana, Fateh ORCID: https://orcid.org/0000-0002-3329-8179 et Ouarda, Taha B. M. J. ORCID: https://orcid.org/0000-0002-0969-063X (2015). A Nonlinear Approach to Regional Flood Frequency Analysis Using Projection Pursuit Regression. Journal of Hydrometeorology , vol. 16 , nº 4. p. 1561-1574. DOI: 10.1175/JHM-D-14-0227.1.

Ce document n'est pas hébergé sur EspaceINRS.


This paper presents an approach for Regional Flood Frequency Analysis (RFFA) in the presence of nonlinearity and problematic stations, which requires adapted methodologies. To this end, we propose the Projection Pursuit Regression (PPR). The latter is a family of regression models that applies smooth functions on intermediate predictors to fit complex patterns. The PPR approach can be seen as a hybrid method between the Generalized Additive Model (GAM) and the Artificial Neural Network (ANN), which combines the advantages of both methods. Indeed, the PPR approach has the structure of a GAM to describe nonlinear relations between hydrological variables and other basin characteristics. On the other hand, PPR can consider interactions between basin characteristics to improve the predictive capabilities in a similar way to ANN, but simpler. The methodology developed in the present study is applied to a case study represented by hydrometric stations from Southern Quebec, Canada. It is shown that flood quantiles are mostly associated to a dominant intermediate predictor, which provides a parsimonious representation of the nonlinearity in the flood generating processes. The model performance is compared to eight other methods available in the literature for the same dataset, including GAM and ANN. When using the same basin characteristics, the results indicate that the simpler structure of PPR does not affect the global performance and that PPR is competitive with the best existing methods in RFFA. Particular attention is also given to the performance resulting from the choice of the basin characteristics and the presence of problematic stations.

Type de document: Article
Mots-clés libres: artificial neural network; environmental protection; flood frequency; nonlinearity; regression analysis
Centre: Centre Eau Terre Environnement
Date de dépôt: 27 avr. 2018 20:21
Dernière modification: 21 févr. 2022 17:29
URI: https://espace.inrs.ca/id/eprint/4366

Actions (Identification requise)

Modifier la notice Modifier la notice