Estimation of daily suspended sediments using support vector machines

Abstract The use of support vector machines—a new regression procedure in water resources—was investigated for predicting suspended sediment concentration/load in rivers. The method was applied to the observed streamflow and suspended sediment data of two rivers in the USA, which have already been used in earlier studies using soft computing techniques. The estimated suspended sediment values were found to be in good agreement with the observed ones. Negative sediment estimates, which were encountered in the soft computing calculations, are not produced by this method. The results indicate that this approach may give better performance than those described in the literature using different methodologies.

[1]  H. K. Cigizoglu,et al.  ESTIMATION AND FORECASTING OF DAILY SUSPENDED SEDIMENT DATA BY MULTI-LAYER PERCEPTRONS , 2004 .

[2]  O. Ks.  Multi-layer perceptrons with Levenberg-Marquardt training algorithm for suspended sediment concentration prediction and estimation , 2004 .

[3]  Gokmen Tayfur,et al.  Artificial neural networks for estimating daily total suspended sediment in natural streams , 2006 .

[4]  N. Asselman.  Fitting and interpretation of sediment rating curves , 2000 .

[5]  D. Walling.  Assessing the accuracy of suspended sediment rating curves for a small basin , 1977 .

[6]  Ozgur Kisi,et al.  Methods to improve the neural network performance in suspended sediment estimation , 2006 .

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

[8]  R. Ferguson.  River Loads Underestimated by Rating Curves , 1986 .

[9]  Junbin Gao,et al.  A Probabilistic Framework for SVM Regression and Error Bar Estimation , 2002, Machine Learning.

[10]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

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

[12]  B. E. Peters-Kümmerly.  Untersuchungen über Zusammensetzung und Transport von Schwebstoffen in einigen Schweizer Flüssen , 1973 .

[13]  Alexander J. Smola,et al.  Regression estimation with support vector learning machines , 1996 .

[14]  G. Tayfur.  Artificial neural networks for sheet sediment transport , 2002 .

[15]  Robert M. Summers,et al.  The validity of a simple statistical model for estimating fluvial constituent loads: An Empirical study involving nutrient loads entering Chesapeake Bay , 1992 .

[16]  Ozgur Kisi,et al.  Suspended sediment estimation using neuro-fuzzy and neural network approaches/Estimation des matières en suspension par des approches neurofloues et à base de réseau de neurones , 2005 .

[17]  R. Ferguson.  Accuracy and precision of methods for estimating river loads , 1987 .

[18]  Ozgur Kisi,et al.  River suspended sediment modelling using a fuzzy logic approach , 2006 .

[19]  Annie Poulin,et al.  Selecting a calculation method to estimate sediment and nutrient loads in streams: Application to the Beaurivage River (Quebec, Canada) , 2005 .

[20]  Nitin Muttil,et al.  Discharge Rating Curve Extension – A New Approach , 2005 .

[21]  A. Durgunoğlu,et al.  Developing accurate and reliable stream sediment yields , 1989 .

[22]  MohammadSajjad Khan,et al.  Application of Support Vector Machine in Lake Water Level Prediction , 2006 .

[23]  V. Singh,et al.  Fuzzy logic algorithm for runoff-induced sediment transport from bare soil surfaces , 2003 .

[24]  Özgür Kişi,et al.  Daily suspended sediment modelling using a fuzzy differential evolution approach / Modélisation journalière des matières en suspension par une approche d’évolution différentielle floue , 2004 .

[25]  R. Brereton,et al.  Support vector machines for classification and regression. , 2010, The Analyst.

[26]  Özgür Kişi,et al.  Multi-layer perceptrons with Levenberg-Marquardt training algorithm for suspended sediment concentration prediction and estimation / Prévision et estimation de la concentration en matières en suspension avec des perceptrons multi-couches et l’algorithme d’apprentissage de Levenberg-Marquardt , 2004 .