Support vector machine―an alternative to artificial neuron network for water quality forecasting in an agricultural nonpoint source polluted river?

Water quality forecasting in agricultural drainage river basins is difficult because of the complicated nonpoint source (NPS) pollution transport processes and river self-purification processes involved in highly nonlinear problems. Artificial neural network (ANN) and support vector model (SVM) were developed to predict total nitrogen (TN) and total phosphorus (TP) concentrations for any location of the river polluted by agricultural NPS pollution in eastern China. River flow, water temperature, flow travel time, rainfall, dissolved oxygen, and upstream TN or TP concentrations were selected as initial inputs of the two models. Monthly, bimonthly, and trimonthly datasets were selected to train the two models, respectively, and the same monthly dataset which had not been used for training was chosen to test the models in order to compare their generalization performance. Trial and error analysis and genetic algorisms (GA) were employed to optimize the parameters of ANN and SVM models, respectively. The results indicated that the proposed SVM models performed better generalization ability due to avoiding the occurrence of overtraining and optimizing fewer parameters based on structural risk minimization (SRM) principle. Furthermore, both TN and TP SVM models trained by trimonthly datasets achieved greater forecasting accuracy than corresponding ANN models. Thus, SVM models will be a powerful alternative method because it is an efficient and economic tool to accurately predict water quality with low risk. The sensitivity analyses of two models indicated that decreasing upstream input concentrations during the dry season and NPS emission along the reach during average or flood season should be an effective way to improve Changle River water quality. If the necessary water quality and hydrology data and even trimonthly data are available, the SVM methodology developed here can easily be applied to other NPS-polluted rivers.

[1]  Jiabao Zhang,et al.  Agricultural diffuse pollution from fertilisers and pesticides in China , 1999 .

[2]  D. Conley,et al.  Transformation of particle-bound phosphorus at the land-sea interface , 1995 .

[3]  R. Young,et al.  AGNPS: A nonpoint-source pollution model for evaluating agricultural watersheds , 1989 .

[4]  A. Belousov,et al.  Applicational aspects of support vector machines , 2002 .

[5]  Dingjiang Chen,et al.  Spatio-temporal variations of nitrogen in an agricultural watershed in eastern China: catchment export, stream attenuation and discharge. , 2011, Environmental pollution.

[6]  Nassim Al-Abed,et al.  Hydrological Modeling of Zarqa River Basin – Jordan Using the Hydrological Simulation Program – FORTRAN (HSPF) Model , 2008 .

[7]  J. Barko,et al.  Water Quality Impacts of Mechanical Shredding of Aquatic Macrophytes , 2002 .

[8]  H. Stabel,et al.  Phosphorus adsorption to riverine suspended matter: Implications for the P-budget of Lake Constance , 1985 .

[9]  Nurettin Acir.  A support vector machine classifier algorithm based on a perturbation method and its application to ECG beat recognition systems , 2006, Expert Syst. Appl..

[10]  K. P. Singh,et al.  Support vector machines in water quality management. , 2011, Analytica chimica acta.

[11]  P. Chambers,et al.  The interaction between water movement, sediment dynamics and submersed macrophytes , 2001, Hydrobiologia.

[12]  J. Arnold,et al.  SWAT2000: current capabilities and research opportunities in applied watershed modelling , 2005 .

[13]  O. Pietiläinen,et al.  Trends of phosphorus, nitrogen and chlorophyll a concentrations in Finnish rivers and lakes in 1975-2000. , 2003, The Science of the total environment.

[14]  A. V.DavidSánchez,et al.  Advanced support vector machines and kernel methods , 2003, Neurocomputing.

[15]  J. Basu,et al.  Application of artificial neural network for prediction of Pb(II) adsorption characteristics , 2013, Environmental Science and Pollution Research.

[16]  Cheng-Lung Huang,et al.  A GA-based feature selection and parameters optimizationfor support vector machines , 2006, Expert Syst. Appl..

[17]  S. Lek,et al.  Predicting stream nitrogen concentration from watershed features using neural networks , 1999 .

[18]  Deva K. Borah,et al.  WATERSHED-SCALE HYDROLOGIC AND NONPOINT-SOURCE POLLUTION MODELS: REVIEW OF APPLICATIONS , 2004 .

[19]  Mark Beale,et al.  Neural Network Toolbox™ User's Guide , 2015 .

[20]  Davor Antanasijević,et al.  Modelling of dissolved oxygen content using artificial neural networks: Danube River, North Serbia, case study , 2013, Environmental Science and Pollution Research.

[21]  Gwo-Fong Lin,et al.  Effective forecasting of hourly typhoon rainfall using support vector machines , 2009 .

[22]  Bernhard Schölkopf,et al.  A tutorial on support vector regression , 2004, Stat. Comput..

[23]  Dingjiang Chen,et al.  Artificial neural network modelling of concentrations of nitrogen, phosphorus and dissolved oxygen in a non‐point source polluted river in Zhejiang Province, southeast China , 2009 .

[24]  Alexander J. Smola,et al.  Support Vector Method for Function Approximation, Regression Estimation and Signal Processing , 1996, NIPS.

[25]  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..

[26]  Xiao Ma,et al.  Assessment and analysis of non-point source nitrogen and phosphorus loads in the Three Gorges Reservoir Area of Hubei Province, China. , 2011, The Science of the total environment.

[27]  Mohamed Cheriet,et al.  Model selection for the LS-SVM. Application to handwriting recognition , 2009, Pattern Recognit..

[28]  M. Gevrey,et al.  Review and comparison of methods to study the contribution of variables in artificial neural network models , 2003 .

[29]  Alexander H. Elliott,et al.  Estimating the sources and transport of nutrients in the Waikato River Basin, New Zealand , 2002 .

[30]  R. Hecht-Nielsen,et al.  Theory of the Back Propagation Neural Network , 1989 .

[31]  Shaogang Gong,et al.  Facial expression recognition based on Local Binary Patterns: A comprehensive study , 2009, Image Vis. Comput..

[32]  Meiyi Zhang,et al.  Impact of suspended inorganic particles on phosphorus cycling in the Yellow River (China). , 2013, Environmental science & technology.

[33]  Lutgarde M. C. Buydens,et al.  Using support vector machines for time series prediction , 2003 .

[34]  E. Doğan,et al.  Modeling biological oxygen demand of the Melen River in Turkey using an artificial neural network technique. , 2009, Journal of environmental management.

[35]  Yi Wang,et al.  Monthly water quality forecasting and uncertainty assessment via bootstrapped wavelet neural networks under missing data for Harbin, China , 2013, Environmental Science and Pollution Research.

[36]  U. Selig,et al.  Dissolved and particulate phosphorus forms in a eutrophic shallow lake , 2002, Aquatic Sciences.

[37]  Muttucumaru Sivakumar,et al.  Prediction of urban stormwater quality using artificial neural networks , 2009, Environ. Model. Softw..

[38]  Mac McKee,et al.  Applicability of statistical learning algorithms in groundwater quality modeling , 2005 .

[39]  Jan-Tai Kuo,et al.  USING ARTIFICIAL NEURAL NETWORK FOR RESERVOIR EUTROPHICATION PREDICTION , 2007 .

[40]  Babak Nadjar Araabi,et al.  Uncertainty analysis of developed ANN and ANFIS models in prediction of carbon monoxide daily concentration , 2010 .

[41]  Yannis Dimopoulos,et al.  Use of some sensitivity criteria for choosing networks with good generalization ability , 1995, Neural Processing Letters.