Application of wavelet-artificial intelligence hybrid models for water quality prediction: a case study in Aji-Chay River, Iran

• Published in: Stochastic environmental research and risk assessment Vol. 30; no. 7; pp. 1797 - 1819
• Main Authors: Barzegar, Rahim, Adamowski, Jan, Moghaddam, Asghar Asghari
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.10.2016; Springer Nature B.V
• Subjects: Adaptive systems; Aquatic Pollution; Artificial intelligence; Artificial neural networks; Chemistry and Earth Sciences; Computational Intelligence; Computer Science; Earth and Environmental Science; Earth Sciences; Environment; Freshwater; Learning theory; Math. Appl. in Environmental Science; Mathematical models; Neural networks; Original Paper; Parameters; Physics; Probability Theory and Stochastic Processes; Rivers; Salinity; Statistical analysis; Statistics for Engineering; Waste Water Technology; Water Management; Water Pollution Control; Water quality; Wavelet; Wavelet transforms; China; ISW, Iran; Water quality prediction; Discrete wavelet transform; Iran; Aji-Chay River; Artificial intelligence models
• ISSN: 1436-3240; 1436-3259
• DOI: 10.1007/s00477-016-1213-y

The accuracy of Artificial Neural Network (ANN), Adaptive Neuro-Fuzzy Inference System (ANFIS), wavelet-ANN and wavelet-ANFIS in predicting monthly water salinity levels of northwest Iran’s Aji-Chay River was assessed. The models were calibrated, validated and tested using different subsets of monthly records (October 1983 to September 2011) of individual solute (Ca 2+ , Mg 2+ , Na + , SO 4 2− and Cl − ) concentrations (input parameters, meq L −1 ), and electrical conductivity-based salinity levels (output parameter, µS cm −1 ), collected by the East Azarbaijan regional water authority. Based on the statistical criteria of coefficient of determination (R 2 ), normalized root mean square error (NRMSE), Nash–Sutcliffe efficiency coefficient (NSC) and threshold statistics (TS) the ANFIS model was found to outperform the ANN model. To develop coupled wavelet-AI models, the original observed data series was decomposed into sub-time series using Daubechies, Symlet or Haar mother wavelets of different lengths (order), each implemented at three levels. To predict salinity input parameter series were used as input variables in different wavelet order/level-AI model combinations. Hybrid wavelet-ANFIS (R 2  = 0.9967, NRMSE = 2.9 × 10 −5 and NSC = 0.9951) and wavelet-ANN (R 2  = 0.996, NRMSE = 3.77 × 10 −5 and NSC = 0.9946) models implementing the db4 mother wavelet decomposition outperformed the ANFIS (R 2  = 0.9954, NRMSE = 3.77 × 10 −5 and NSC = 0.9914) and ANN (R 2  = 0.9936, NRMSE = 3.99 × 10 −5 and NSC = 0.9903) models.