Clustering of heterogeneous precipitation fields for the assessment and possible improvement of lumped neural network models for streamflow forecasts

Abstract. This work addresses the issue of better considering the heterogeneity of precipitation fields within lumped rainfall-runoff models where only areal mean precipitation is usually used as an input. A method using a Kohonen neural network is proposed for the clustering of precipitation fields. The evaluation and improvement of the performance of a lumped rainfall-runoff model for one-day ahead predictions is then established based on this clustering. Multilayer perceptron neural networks are employed as lumped rainfall-runoff models. The Bas-en-Basset watershed in France, which is equipped with 23 rain gauges with data for a 21-year period, is employed as the application case. The results demonstrate the relevance of the proposed clustering method, which produces groups of precipitation fields that are in agreement with the global climatological features affecting the region, as well as with the topographic constraints of the watershed (i.e., orography). The strengths and weaknesses of the rainfall-runoff models are highlighted by the analysis of their performance vis-a-vis the clustering of precipitation fields. The results also show the capability of multilayer perceptron neural networks to account for the heterogeneity of precipitation, even when built as lumped rainfall-runoff models.

[1]  George Cybenko,et al.  Approximation by superpositions of a sigmoidal function , 1992, Math. Control. Signals Syst..

[2]  Leo Breiman,et al.  Using Iterated Bagging to Debias Regressions , 2001, Machine Learning.

[3]  M. Goldstein,et al.  Multivariate Analysis: Methods and Applications , 1984 .

[4]  R. Benzi,et al.  CHARACTERIZATION OF TEMPERATURE AND PRECIPITATION FIELDS OVER SARDINIA WITH PRINCIPAL COMPONENT ANALYSIS AND SINGULAR SPECTRUM ANALYSIS , 1997 .

[5]  null null,et al.  Artificial Neural Networks in Hydrology. II: Hydrologic Applications , 2000 .

[6]  Charles A. Doswell,et al.  A Diagnostic Study of Three Heavy Precipitation Episodes in the Western Mediterranean Region , 1998 .

[7]  Shie-Yui Liong,et al.  Advance flood forecasting for flood stricken Bangladesh with a fuzzy reasoning method (Copies of English Papers by the Center Staff Published in the Fiscal Year of 1999) , 2000 .

[8]  V. Singh,et al.  Mathematical Modeling of Watershed Hydrology , 2002 .

[9]  François Anctil,et al.  Impact of the length of observed records on the performance of ANN and of conceptual parsimonious rainfall-runoff forecasting models , 2004, Environ. Model. Softw..

[10]  Martin T. Hagan,et al.  Gauss-Newton approximation to Bayesian learning , 1997, Proceedings of International Conference on Neural Networks (ICNN'97).

[11]  R. Schnur,et al.  A case study of statistical downscaling in Australia using weather classification by recursive partitioning , 1998 .

[12]  P. Naden.  Spatial variability in flood estimation for large catchments: the exploitation of channel network structure , 1992 .

[13]  Bernard Bobée,et al.  Prévision hydrologique par réseaux de neurones artificiels : état de l'art , 1999 .

[14]  E. Toth,et al.  Comparison of short-term rainfall prediction models for real-time flood forecasting , 2000 .

[15]  P. Kitanidis,et al.  Real‐time forecasting with a conceptual hydrologic model: 2. Applications and results , 1980 .

[16]  Nicolas Lauzon.  Water resources data quality assessment and description of natural processes using artificial intelligence techniques , 2003 .

[17]  Holger R. Maier,et al.  Neural networks for the prediction and forecasting of water resource variables: a review of modelling issues and applications , 2000, Environ. Model. Softw..

[18]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[19]  Juan B. Valdés,et al.  On the influence of the spatial distribution of rainfall on storm runoff , 1979 .

[20]  Mohammad Bagher Menhaj,et al.  Training feedforward networks with the Marquardt algorithm , 1994, IEEE Trans. Neural Networks.

[21]  Holger R. Maier,et al.  Optimal division of data for neural network models in water resources applications , 2002 .

[22]  N. Null.  Artificial Neural Networks in Hydrology. I: Preliminary Concepts , 2000 .

[23]  James P. Hughes,et al.  A stochastic approach for assessing the effect of changes in synoptic circulation patterns on gauge precipitation , 1993 .

[24]  David J. C. MacKay,et al.  Bayesian Interpolation , 1992, Neural Computation.

[25]  Alex J. Cannon,et al.  Downscaling recent streamflow conditions in British Columbia, Canada using ensemble neural network models , 2002 .

[26]  François Anctil,et al.  A soil moisture index as an auxiliary ANN input for stream flow forecasting , 2004 .

[27]  Armando Brath,et al.  Neural networks and non-parametric methods for improving real-time flood forecasting through conceptual hydrological models , 2002 .

[28]  E. Bazile,et al.  The 12–13 November 1999 flash flood in southern France , 2001 .

[29]  A. W. Minns,et al.  The classification of hydrologically homogeneous regions , 1999 .

[30]  Yonghong Tan,et al.  Neural-network-based d-step-ahead predictors for nonlinear systems with time delay , 1999 .

[31]  Lucien Duckstein,et al.  Relationship Between Monthly Atmospheric Circulation Patterns and Precipitation: Fuzzy Logic and Regression Approaches , 1996 .

[32]  D. A. Woolhiser,et al.  Impact of small-scale spatial rainfall variability on runoff modeling , 1995 .

[33]  Leo Breiman,et al.  Bagging Predictors , 1996, Machine Learning.

[34]  S. Pryor,et al.  Downscaling temperature and precipitation: a comparison of regression‐based methods and artificial neural networks , 2001 .

[35]  Lucien Duckstein,et al.  Knowledge Based Classification of Circulation Patterns for Stochastic Precipitation Modeling , 1994 .

[36]  D. R. Dawdy,et al.  Effect of rainfall variability on streamflow simulation , 1969 .

[37]  James P. Hughes,et al.  Stochastic characterization of regional circulation patterns for climate model diagnosis and estimation of local precipitation , 1995 .

[38]  R Govindaraju,et al.  ARTIFICIAL NEURAL NETWORKS IN HYDROLOGY: II, HYDROLOGIC APPLICATIONS , 2000 .

[39]  T O Siew-Yan-Yu,et al.  Régionalisation du régime des précipitations dans la région des Bois-francs et de l'Estrie par l'analyse en composantes principales , 1998 .

[40]  S. Sénési,et al.  The Vaison-La-Romaine Flash Flood: Mesoscale Analysis and Predictability Issues , 1996 .

[41]  H. Storch,et al.  The Analog Method as a Simple Statistical Downscaling Technique: Comparison with More Complicated Methods , 1999 .

[42]  Teuvo Kohonen,et al.  The self-organizing map , 1990 .

[43]  A. Bárdossy,et al.  SPACE-TIME MODEL FOR DAILY RAINFALL USING ATMOSPHERIC CIRCULATION PATTERNS , 1992 .

[44]  S. Sievers,et al.  Dioxin, dioxin-like PCBS and organotin compounds in the river Elbe and the Hamburg harbour: identification of sources , 1998 .

[45]  François Anctil,et al.  ANN output updating of lumped conceptual rainfall/runoff forecasting models , 2003 .