Capability Evaluation of Hybrid Wavelet-Principal Component Analysis-Random Forest Approach in Simulating the River Flow

Document Type : Research Article


1 Water engineering department, Faculty of civil and surveying engineering, Graduate University of Advanced technology, Kerman, Iran

2 Water Engineering Department, Civil and surveying engineering Faculty, Graduate University of Advanced Technology


Simulating the flow for managing the water allocation in drought and wet periods is of great importance. According to the researches conducted during several decades in this regard, computational intelligence methods combined with wavelets are known to be effective. In this paper, Wavelet-Principal Component Analysis-Random Forest (WPCARF) hybrid approach is proposed to model the daily flow of the Polroud river. In the proposed model, first, hydrometric data is preprocessed by wavelet transform and applied to the PCA along with meteorological data. Afterward, their output vectors were entered into the random forest network. The results have shown that the PCA algorithm can improve the performance accuracy and speed of the model, despite reducing the input vectors and simplifying them. Also, it can integrate a model with increased simulation time and input vectors uncertainty having a lower impact on model capability leading to a more uniform decreasing trend. Furthermore, preprocessing the data accompanied by PCA could enhance the agreement index by 5 and 8 percent during one and three days of the simulation and increase the model ability for a more accurate simulation of river flow. On the other hand, results for the best-proposed hybrid model during the one-day-ahead simulation time were R=0.911 and RMSE=7.095 m3/s, while these values were R=0.817 and RMSE=8.681 m3/s in the best hybrid model for three-day-ahead simulation time. This indicates the adequate capacity of the proposed hybrid model for long-term simulation times.


Main Subjects

[1] D. P. Solomatine, A. Ostfeld, Data-driven modelling: some past experiences and new approaches, Journal of hydro informatics, 10(1) (2008) 3-22.
[2] N. Khairuddin, A.Z. Aris, A. Elshafie, T. Sheikhy Narany, M.Y. Ishak, N.M. Isa, Efficient forecasting model technique for river stream flow in tropical environment. Urban Water Journal, 16(3) (2019) 183-192.
[3] H. Tongal, M. J. Booij, M, Simulation and forecasting of stream flows using machine learning models coupled with base flow separation, Journal of hydrology, 564 (2018) 266-282.
[4] D. Hussain, A.A. Khan, A, Machine learning techniques for monthly river flow forecasting of Hunza River, Pakistan, Earth Science Informatics, (2020) 1-11.
[5] V. Nourani, A. H. Baghanam, J. Adamowski, O. Kisi, Applications of hybrid wavelet–artificial intelligence models in hydrology: a review, Journal of Hydrology, 514 (2014) 358-377.
[6] N. Nourani, A. Davanlou Tajbakhsh, A. Molajou, H. Gokcekus, Hybrid wavelet-M5 model tree for rainfall-runoff modeling, Journal of Hydrologic Engineering, 24(5) (2019) 04019012.
[7] Y. Sun, J. Niu, B. Sivakumar, A comparative study of models for short-term streamflow forecasting with emphasis on wavelet-based approach. Stochastic Environmental Research and Risk Assessment, 33(10) (2019) 1875-1891.
[8] K. Roushangara, R. Ghasempourb, Monthly precipitation prediction improving using the integrated model based on kernel-wavelet and complementary ensemble empirical mode decomposition, CEEJ (XML). In Persian.
[9] A. D. Mehr, An improved gene expression programming model for streamflow forecasting in intermittent streams, Journal of hydrology, 563 (2018) 669-678.
[10] M. Rezaie-Balf, S. Fani Nowbandegani, S. Z. Samadi, H. Fallah, S. Alaghmand, An ensemble decomposition-based artificial intelligence approach for daily streamflow prediction. Water, 11(4) (2019) 709.
[11] F. J. Chang, P. A. Chen, Y. R. Lu, E. Huang, K. Y. Chang, Real-time multi-step-ahead water level forecasting by recurrent neural networks for urban flood control, Journal of Hydrology, 517 (2014) 836-846.
[12] S. Kabir, S. Patidar, G. Pender, Investigating capabilities of machine learning techniques in forecasting stream flow, In Proceedings of the Institution of Civil Engineers-Water Management, 173 (2) (2020) 69- 86.
[13] R. Noori, A. Farokhnia, S. Morid, H. Riahi Madvar, (2009). Effect of input variables preprocessing in artificial neural network on monthly flow prediction by PCA and wavelet transformation, Journal of Water and Wastewater, (1) (2009) 13-22. In Persian.
[14] M.R. Najafi, H. Moradkhani, T. C. Piechota, Ensemble streamflow prediction: climate signal weighting methods vs. climate forecast system reanalysis, Journal of Hydrology, 442 (2012) 105-116.
[15] C. Prieto, N. Le Vine, D. Kavetski, E. García, R. Medina, Flow prediction in ungauged catchments using probabilistic Random Forests regionalization and new statistical adequacy tests, Water Resources Research, 55(5) (2019) 4364-4392.
[16] H.A. Afan, M.F. Allawi, A. El-Shafie, Z.M. Yaseen, A.N. Ahmed, M.A. Malek, A. Sefelnasr, Input attributes optimization using the feasibility of genetic nature inspired algorithm: Application of river flow forecasting, Scientific Reports, 10(1) (2020) 1-15.
[17] R. Noori, A.R. Karbassi, A. Moghaddamnia, D. Han, M.H. Zokaei-Ashtiani, A. Farokhnia, M.G. Gousheh, Assessment of input variables determination on the SVM model performance using PCA, Gamma test, and forward selection techniques for monthly stream flow prediction, Journal of Hydrology, 401(3-4) (2011) 177-189.
[18] Y. Hassanzadeh, K. A. Abdi, N. M. Shafiei, S. Khoshtinat, Daily streamflow forecasting of Nooranchay river using the yybrid model of Artificial Neural Networks-Principal Component Analysis, Journal of Soil and Water Science, 25 (3) (2015) 53- 63. In Persian.
[19] M. Ehteram, H.A Afan, M. Dianatikhah, A. N. Ahmed, C. Ming Fai, M. S. Hossain, A. Elshafie, Assessing the predictability of an improved ANFIS model for monthly streamflow using lagged climate indices as predictors, Water, 11 (6) (2019) 1130.
[20] L. Diop, A. Bodian, K. Djaman, Z. M. Yaseen, R.C. Deo, A. El-Shafie, L. C. Brown, The influence of climatic inputs on stream-flow pattern forecasting: case study of Upper Senegal River, Environmental Earth Sciences, 77(5) (2018) 182.
[21] S.J. Hadi, M. Tombul, (2018). Monthly streamflow forecasting using continuous wavelet and multi-gene genetic programming combination, Journal of Hydrology, 561 (2018) 674-687.
[22] R. M. Adnan, Z. Liang, S. Heddam, M. Zounemat-Kermani, O. Kisi, B. Li, Least square support vector machine and multivariate adaptive regression splines for streamflow prediction in mountainous basin using hydro-meteorological data as inputs, Journal of Hydrology, 586 (2020) 124371.
[23] M. Motamednia, A. Nohegar, A. Malekian, M. Saberi Anari, K. Karimi Zarchi, Impacts of combining meteorological and hydrometric data on the accuracy of streamflow modeling, Environmental Resources Research, 7(2) (2019) 147-164.
[24] Y-F. Sang, A review on the applications of wavelet transform in hydrology time series analysis, Atmospheric Research, 122 (2012) 8-15.
[25] S. Mallat, A wavelet tour of signal processing: the sparse way 3rd edn, New York: Academic, (2008).
[26] L. Breiman, Random forests. Machine learning, 45(1) (2001) 5-32.
[27] H. Abdi, L. J. Williams, Principal component analysis, Wiley interdisciplinary reviews: computational statistics, 2(4) (2010) 433-459.
[28] F. Anctil, M.H. Ramos, Verification Metrics for Hydrological Ensemble Forecasts, Handbook of Hydro Meteorological Ensemble Forecasting, (2019) 893-922
[29] C. J. Willmott, on the validation of models, Physical geography, 2(2) (1981) 184-194.
[30] B. G. Tabachnick, L. S. Fidell, Experimental designs using ANOVA, Belmont, CA: Thomson/Brooks/Cole, (2007).