Improvement of artificial neural networks to predict daily streamflow in a semi-arid area

ABSTRACT The application of artificial neural networks (ANNs) has been widely used recently in streamflow forecasting because of their flexible mathematical structure. However, several researchers have indicated that using ANNs in streamflow forecasting often produces a timing lag between observed and simulated time series. In addition, ANNs under- or overestimate a number of peak flows. In this paper, we proposed three data-processing techniques to improve ANN prediction and deal with its weaknesses. The Wilson-Hilferty transformation (WH) and two methods of baseflow separation (one parameter digital filter, OPDF, and recursive digital filter, RDF) were coupled with ANNs to build three hybrid models: ANN-WH, ANN-OPDF and ANN-RDF. The network behaviour was quantitatively evaluated by examining the differences between model output and observed variables. The results show that even following the guidelines of the Wilson-Hilferty transformation, which significantly reduces the effect of local variations, it was found that the ANN-WH model has shown no significant improvement of peak flow estimation or of timing error. However, combining baseflow with streamflow and rainfall provides important information to ANN models concerning the flow process operating in the aquifer and the watershed systems. The model produced excellent performance in terms of various statistical indices where timing error was totally eradicated and peak flow estimation significantly improved. Editor D. Koutsoyiannis; Associate editor Y. Gyasi-Agyei

[1]  Jose D. Salas,et al.  Streamflow Forecasting Based on Artificial Neural Networks , 2000 .

[2]  M. Imteaz,et al.  Catchment flow estimation using Artificial Neural Networks in the mountainous Euphrates Basin , 2011 .

[3]  Holger R. Maier,et al.  Application of partial mutual information variable selection to ANN forecasting of water quality in water distribution systems , 2008, Environ. Model. Softw..

[4]  Lerma Sanchez,et al.  A neural network approach to a dimensionality reduction problem , 1992 .

[5]  A. Soldati,et al.  River flood forecasting with a neural network model , 1999 .

[6]  R. Hooper,et al.  Chapter 10 – Oxygen and Hydrogen Isotopes in Rainfall-Runoff Studies , 1998 .

[7]  Rodney A. Stewart,et al.  ANN-based residential water end-use demand forecasting model , 2013, Expert Syst. Appl..

[8]  J. Nash,et al.  River flow forecasting through conceptual models part I — A discussion of principles☆ , 1970 .

[9]  Ashu Jain,et al.  Development of effective and efficient rainfall‐runoff models using integration of deterministic, real‐coded genetic algorithms and artificial neural network techniques , 2004 .

[10]  Ashwani Kumar,et al.  Comparative evaluation of numerical model and artificial neural network for simulating groundwater flow in Kathajodi–Surua Inter-basin of Odisha, India , 2013 .

[11]  T. Rientjes,et al.  Constraints of artificial neural networks for rainfall-runoff modelling: trade-offs in hydrological state representation and model evaluation , 2005 .

[12]  Chandranath Chatterjee,et al.  Uncertainty assessment and ensemble flood forecasting using bootstrap based artificial neural networks (BANNs) , 2010 .

[13]  Lyubka Pashova,et al.  Daily sea level forecast at tide gauge Burgas, Bulgaria using artificial neural networks , 2011 .

[14]  K. Chau,et al.  Prediction of rainfall time series using modular artificial neural networks coupled with data-preprocessing techniques , 2010 .

[15]  Vahid Nourani,et al.  An ANN‐based model for spatiotemporal groundwater level forecasting , 2008 .

[16]  Tsung-Yi Pan,et al.  State space neural networks for short term rainfall-runoff forecasting , 2004 .

[17]  P. C. Nayak,et al.  Improving peak flow estimates in artificial neural network river flow models , 2003 .

[18]  K. M. O'Connor.  River flow forecasting , 2005 .

[19]  A. Ahmadi,et al.  Daily suspended sediment load prediction using artificial neural networks and support vector machines , 2013 .

[20]  Ali Akbar Safavi,et al.  A simple neural network model for the determination of aquifer parameters , 2007 .

[21]  Kurt Hornik,et al.  Approximation capabilities of multilayer feedforward networks , 1991, Neural Networks.

[22]  Sajjad Ahmad,et al.  Suspended sediment load prediction of river systems: An artificial neural network approach , 2011 .

[23]  Y. R. Satyaji Rao,et al.  Modelling groundwater levels in an urban coastal aquifer using artificial neural networks , 2008 .

[24]  Bhoop Singh,et al.  Artificial neural network model as a potential alternative for groundwater salinity forecasting , 2011 .

[25]  Kiyoshi Kawaguchi.  A multithreaded software model for backpropagation neural network applications , 2000 .

[26]  C. L. Wu,et al.  Methods to improve neural network performance in daily flows prediction , 2009 .

[27]  L. Benaabidate,et al.  Design flood estimation in ungauged catchments and statistical characterization using principal components analysis: application of Gradex method in Upper Moulouya , 2013 .

[28]  L. Tallaksen.  A review of baseflow recession analysis , 1995 .

[29]  M. Imteaz,et al.  Multiple regression and Artificial Neural Network for long-term rainfall forecasting using large scale climate modes , 2013 .

[30]  Ken-ichi Funahashi,et al.  On the approximate realization of continuous mappings by neural networks , 1989, Neural Networks.

[31]  E. B. Wilson,et al.  The Distribution of Chi-Square. , 1931, Proceedings of the National Academy of Sciences of the United States of America.

[32]  C. L. Wu,et al.  Rainfall–runoff modeling using artificial neural network coupled with singular spectrum analysis , 2011 .

[33]  F. R. Hall.  Base‐Flow Recessions—A Review , 1968 .

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

[35]  Aman Mohammad Kalteh,et al.  Monthly river flow forecasting using artificial neural network and support vector regression models coupled with wavelet transform , 2013, Comput. Geosci..

[36]  Özgür Kisi,et al.  Constructing neural network sediment estimation models using a data-driven algorithm , 2008, Math. Comput. Simul..

[37]  J. Arnold,et al.  AUTOMATED METHODS FOR ESTIMATING BASEFLOW AND GROUND WATER RECHARGE FROM STREAMFLOW RECORDS 1 , 1999 .

[38]  W. Kirby,et al.  Computer‐oriented Wilson‐Hilferty transformation that preserves the first three moments and the lower bound of the Pearson type 3 distribution , 1972 .

[39]  K. P. Sudheer,et al.  A data‐driven algorithm for constructing artificial neural network rainfall‐runoff models , 2002 .

[40]  Necati Kayaalp,et al.  Estimation of the Amount of Suspended Sediment in the Tigris River using Artificial Neural Networks , 2008 .

[41]  K. Eckhardt.  How to construct recursive digital filters for baseflow separation , 2005 .

[42]  Holger R. Maier,et al.  Input determination for neural network models in water resources applications. Part 1—background and methodology , 2005 .

[43]  Saman Razavi,et al.  Long‐lead seasonal rainfall forecasting using time‐delay recurrent neural networks: a case study , 2008 .

[44]  Shih-Ching Wu,et al.  Estimating anisotropic aquifer parameters by artificial neural networks , 2010 .

[45]  Shaozhong Kang,et al.  Integrated neural networks for monthly river flow estimation in arid inland basin of Northwest China , 2012 .

[46]  A. W. Minns,et al.  Artifical neural networks as subsymbolic process descriptors , 1998 .

[47]  Ozgur Kisi,et al.  Modeling discharge–sediment relationship using neural networks with artificial bee colony algorithm , 2012 .

[48]  M. Sklash,et al.  A conceptual model of watershed response to rainfall, developed through the use of oxygen-18 as a natural tracer , 1976 .

[49]  T. McMahon,et al.  Evaluation of automated techniques for base flow and recession analyses , 1990 .

[50]  V. Yevjevich,et al.  Stochastic hydrology and its use in water resources systems simulation and optimization , 1993 .

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