Volume 8, Issue 4 (2020)                   ECOPERSIA 2020, 8(4): 199-208 | Back to browse issues page

XML Print


Environmental Sciences Department, Shahid Rajaee Teacher Training University, Tehran, Iran , msakizadeh@gmail.com
Abstract:   (113 Views)
Aims: The main objective of the current study was to assess the efficiency of four-time series prediction methods to forecast the values of total dissolved solids (TDS) using a time series of over sixteen years.
Materials & Methods: The applied methods comprised of autoregressive integrated moving average (ARIMA) as the most traditional method, two neural network based techniques including multilayer perceptron (MLP) along with extreme learning machines (ELM) and a novel approach known as temporal hierarchies (TH) which was applied for the first time in water resources and water quality researches.
Findings: It was found that with respect to the forecasting accuracy, the MLP outperforms the ARIMA model for the training series where the MAPE (%) and MASE (mg/l) were reduced from 5.109 to 3.146 and 0.553 to 0.323, respectively. On the other hand, the forecasting accuracy of ELM was lower than that of MLP however the respective out-of-sample generalization ability of this model was higher with MAPE and MASE values of 6.526 and 0.683.
Conclusion: Meanwhile, it was concluded that temporal hierarchies gave the best results for the test part of time series. The main shortcoming of neural network based approaches was their reduced out-of-sample prediction due to overfitting. Based on the results, TH is a viable alternative for conventional time series forecasting techniques.
Full-Text [PDF 623 kb]   (31 Downloads)    
Article Type: Original Research | Subject: Pollution (Soil, Water and Air)
Received: 2019/07/28 | Accepted: 2019/12/24 | Published: 2020/09/22
* Corresponding Author Address: Environmental Sciences Department, Shahid Rajaee Teacher Training University, Tehran, Iran

References
1. Deng W, Wang G, Zhang X. A novel hybrid water quality time series prediction method based on cloud model and fuzzy forecasting. Chemom Intell Lab Syst. 2015;149(Part A):39-49. [Link] [DOI:10.1016/j.chemolab.2015.09.017]
2. Bahar Gogani M, Douzbakhshan M, Shayesteh K, Ildoromi AR. New formulation of fuzzy comprehensive evaluation model in‎ groundwater resources carrying capacity analysis. ECOPERSIA. 2018;6(2):79-89. [Link]
3. Vafakhah M, Janizadeh S, Khosrobeigi Bozchaloei S. Application of several data-driven techniques for rainfall-runoff modeling. ECOPERSIA. 2014;2(1):455-69. [Link]
4. Liong SY, Phoon KK, Pasha MF, Doan CD. Efficient implementation of inverse approach for forecasting hydrological time series using micro GA. J Hydroinform. 2005;7(3):151-63. [Link] [DOI:10.2166/hydro.2005.0013]
5. Bazrafshan O, Salajegheh A, Bazrafshan J, Mahdavi M, Fatehi Maraj A. Hydrological drought forecasting using ARIMA models (case study: Karkheh Basin). ECOPERSIA. 2015;3(3):1099-117. [Link]
6. Wang WC, Chau KW, Xu DM, Chen XY. Improving forecasting accuracy of annual runoff time series using ARIMA based on EEMD decomposition. Water Resour Manag. 2015;29(8):2655-75. [Link] [DOI:10.1007/s11269-015-0962-6]
7. Arya FK, Zhang L. Copula-based Markov process for forecasting and analyzing risk of water quality time series. J Hydrol Eng. 2017;22(6):04017005. [Link] [DOI:10.1061/(ASCE)HE.1943-5584.0001494]
8. Elkiran G, Nourani V, Abba SI. Multi-step ahead modelling of river water quality parameters using ensemble artificial intelligence-based approach. J Hydrol. 2019;577:123962. [Link] [DOI:10.1016/j.jhydrol.2019.123962]
9. Fabro AY, Ávila JG, Alberich MV, Sansores SA, Camargo-Valero MA. Spatial distribution of nitrate health risk associated with groundwater use as drinking water in Merida, Mexico. Appl Geogr. 2015;65:49-57. [Link] [DOI:10.1016/j.apgeog.2015.10.004]
10. Li L, Jiang P, Xu H, Lin G, Guo D, Wu H. Water quality prediction based on recurrent neural network and improved evidence theory: A case study of Qiantang River, China. Environ Sci Pollut Res. 2019;26(19):19879-96. [Link] [DOI:10.1007/s11356-019-05116-y]
11. Dottridge J, Jaber NA. Groundwater resources and quality in northeastern Jordan: Safe yield and sustainability. Appl Geogr. 1999;19(4):313-23. [Link] [DOI:10.1016/S0143-6228(99)00012-0]
12. Park SC, Yun ST, Chae GT, Yoo IS, Shin KS, Heo CH, et al. Regional hydrochemical study on salinization of coastal aquifers, western coastal area of South Korea. J Hydrol. 2005;313(3-4):182-94. [Link] [DOI:10.1016/j.jhydrol.2005.03.001]
13. Taneja K, Ahmad Sh, Ahmad K, Attri SD. Time series analysis of aerosol optical depth over New Delhi using Box-Jenkins ARIMA modeling approach. Atmos Pollut Res. 2016;7(4):585-96. [Link] [DOI:10.1016/j.apr.2016.02.004]
14. Shirmohammadi B, Vafakhah M, Moosavi V, Moghaddamnia A. Application of several data-driven techniques for predicting groundwater level. Water Resour Manag. 2013;27(2):419-32. [Link] [DOI:10.1007/s11269-012-0194-y]
15. Mirzavand M, Ghazavi R. A stochastic modelling technique for groundwater level forecasting in an arid environment using time series methods. Water Resour Manag. 2015;29(4):1315-28. [Link] [DOI:10.1007/s11269-014-0875-9]
16. Faruk DÖ. A hybrid neural network and ARIMA model for water quality time series prediction. Eng Appl Artif Intell. 2010;23(4):586-94. [Link] [DOI:10.1016/j.engappai.2009.09.015]
17. Kumar S, Tiwari MK, Chatterjee Ch, Mishra A. Reservoir inflow forecasting using ensemble models based on neural networks, wavelet analysis and bootstrap method. Water Resour Manag. 2015;29(13):4863-83. [Link] [DOI:10.1007/s11269-015-1095-7]
18. Kourentzes N, Barrow DK, Crone SF. Neural network ensemble operators for time series forecasting. Expert Syst Appl. 2014;41(9):4235-44. [Link] [DOI:10.1016/j.eswa.2013.12.011]
19. Haykin SS, Haykin SS, Haykin SS, Haykin SS. Neural networks and learning machines. 3rd Edition. Upper Saddle River: Prentice Hall; 2009. [Link]
20. Athanasopoulos G, Hyndman RJ, Kourentzes N, Petropoulos F. Forecasting with temporal hierarchies. Eur J Oper Res. 2017;262(1):60-74. [Link] [DOI:10.1016/j.ejor.2017.02.046]
21. Stevens A, Nocita M, Tóth G, Montanarella L, Van Wesemael B. Prediction of soil organic carbon at the European scale by visible and near infrared reflectance spectroscopy. PLoS One. 2013;8(6):e66409. [Link] [DOI:10.1371/journal.pone.0066409]
22. Hyndman RJ, Koehler AB. Another look at measures of forecast accuracy. Int J Forecast. 2006;22(4):679-88. [Link] [DOI:10.1016/j.ijforecast.2006.03.001]
23. An T, Huang Y, Li G, He Z, Chen J, Zhang Ch. Pollution profiles and health risk assessment of VOCs emitted during e-waste dismantling processes associated with different dismantling methods. Environ Int. 2014;73:186-94. [Link] [DOI:10.1016/j.envint.2014.07.019]
24. Hounslow A. Water quality data: Analysis and interpretation. Boca Raton: CRC Press; 2018. [Link] [DOI:10.1201/9780203734117]
25. Zarasvandi A, Mirzaee S. Geochemistry of the Karkheh River sediments, Khuzestan Province, Iran: Evidences for natural contamination. Res J Appl Sci. 2009;4(1):35-40. [Link]
26. Chitsazan M, Faryabi M, Zarrasvandi AR. Evaluation of river-aquifer interaction in the north part of Dezful-Andimeshk district, SW of Iran. Arab J Geosci. 2015;8(9):7177-89. [Link] [DOI:10.1007/s12517-014-1686-2]
27. Nasrabadi T, Nabi Bidhendi GR, Karbassi AR, Hoveidi H, Nasrabadi I, Pezeshk H, et al. Influence of Sungun copper mine on groundwater quality, NW Iran. Environ Geol. 2009;58(4):693-700. [Link] [DOI:10.1007/s00254-008-1543-2]
28. Chatfield Ch. The analysis of time series: an introduction. Boca Raton: CRC Press; 2016. [Link]
29. Laio F, Di Baldassarre G, Montanari A. Model selection techniques for the frequency analysis of hydrological extremes. Water Resour Res. 2009;45(7):W07416. [Link] [DOI:10.1029/2007WR006666]
30. Kim S, Kim H. A new metric of absolute percentage error for intermittent demand forecasts. Int J Forecast. 2016;32(3):669-79. [Link] [DOI:10.1016/j.ijforecast.2015.12.003]
31. Makridakis S, Wheelwright SC, Hyndman RJ. Forecasting methods and applications. Hoboken: John Wiley & Sons; 2008. [Link]
32. Srivastava N, Hinton G, Krizhevsky A, Sutskever I, Salakhutdinov R. Dropout: A simple way to prevent neural networks from overfitting. J Mach Learn Res. 2014;15(1):1929-58. [Link]