Drought Forecasting Using Tree Multi-Scale Model

Document Type : Original Article

Authors

1 Assistant Professor, Dept. of Civil Eng., University of Ayatollah Boroujerdi

2 MSc Student, Department of Civil Engineering, University of Ayatollah Boroujerdi

Abstract

Drought is one of the environmental phenomena which causes severe damage to human societies. To survey the climatic changes conditions, especially drought, drought indices were as a strong management tool. In addition to the drought index, drought forecasting using appropriate models may be appropriate in control and drought management. In this research, information concerning precipitation related to two basins located in Hamadan and Lorestan Province was used to calculate the SPI and EDI indices, which are among the most-widely used indices for monitoring drought. The calculated indices were employed as the input for the WM5 models that were used in predicting drought. In the best-case scenario, the coefficients of determination for the EDI index in the M5 model and in WM5 model were 0.95 and 0.99, respectively. Moreover, the coefficients of determination for the SPI index in the M5 model and in WM5 model, in the best-case scenario, were 0.90 and 0.95, respectively. This suggests indicated the use of the WM5 was superior to and enjoyed great accuracy compared to the M5 model. The reason for this superiority is that both short- and long-term changes in time series are monitored in WM5. Moreover, the WM5 is more capable than the M5 in monitoring the maximum values of time series.

Keywords


[1] Palmer, W. C. (1965). Meteorological drought. Washington, DC: US Department of Commerce, Weather Bureau.
[2] McKee, T.B., Doesken, N.J., & Kleist, J. (1993). “The relationship of drought frequency and duration to time scales”, In Proceedings of the 8th Conference on Applied Climatology, 17(22), 179-183.
[3] Jain, V.K., Pandey, R.P., Jain, M.K., & Byun, H.R. (2015). “Comparison of drought indices for appraisal of drought characteristics in the Ken River Basin”, Weather and Climate Extremes, 8, 1-11.
[4] Bazrafshan, O., Salajegheh, A., Bazrafshan, J., Mahdavi, M., & Fatehi Marj, A. (2015). “Hydrological drought forecasting using ARIMA models (Case Study: Karkheh Basin)”, Ecopersia, 3(3), 1099-1117. 
[5] کماسی، م.، اعلمی، م.، نورانی، و. (1391). "پیش‌بینی خشکسالی با نمایه‌ی SPIبه روش مدل‌سازی ANFIS برمبنای خوشه‌بندی C-Mean فازی"، نشریه‌ی آب و فاضلاب، شماره‌ 4، ش.ص. 90- 102.
[6] Maca, P., & Pech, P. (2016). “Forecasting SPEI and SPI Drought Indices Using the Integrated Artificial Neu-ral Networks”, Computational Intelligence and Neuroscience, 2016(2016), 1-17.
[7] Savice, D.A., Walters G.A., & Davidson, J. (1999). “A Genetic programming approach to rainfall- runoff modeling”, Water Resources Management, 13, 219-231.
[8] Hassanzadeh, Y., Abdi, A., Talatahari, S., & Singh, V.P. (2011). “Meta-Heuristic Algorithms for hydrologic frequency analysis”, Water Resources Management, 25(7), 1855-1879.
[9] Cannas, B., Fanni, A., See, L., & Sias, G. (2006). “Data preprocessing for river flow forecasting using neural networks: wavelet transforms and data partitioning”, Physics and Chemistry of the Earth, 31(18), 1164-1171.
[10] Djerbouai, S., & Souag– Gamane, D. (2016). “Drought Forecasting Using Neural Networks, Wavelet Neural Networks, and Stochastic Models: Case of the Algerois Basin in North Algeria”, Water Resources Management, 30(7), 2445-2464.
[11] Özger, M., Mishra, A.K., &Singh, V.P. (2012). “Long lead time drought forecasting using a wavelet and fuzzy logic combination model: A case study in texas”, Journal of Hydrometeorology, 13(1), 284-297.
[12] Londhe, S. N., & Dixit, P.R. (2011). “Forecasting stream flow using model trees”, International Journal of Earth Sciences and Engineering, 4(6), 282-285. 
[13] Alipour, A., Yarahmadi, J., & Mahdavi, M. (2014). “Comparative Study of M5 Model Tree and Artificial Neural Network in Estimating Reference Evapotranspiration Using MODIS Products”, Journal of Climatology, 2014.
[14] ستاری، م.، رضازاده، ت.، جودی، ع.، نهرین، ف. (1392). "پیش‌بینی مقادیر بارش ماهانه با استفاده از شبکه‌های عصبی مصنوعی و مدل درختی M5 (مطالعه‌ی موردی: ایستگاه اهر)"، پژوهش‌های جغرافیای طبیعی، دوره 46، شماره 2، ش.ص. 247-260.
[15] Pal, M., & Deswal, S. (2009). “M5 model tree based modeling of reference evapotranspiration”, Hydrolog-ical Processes, 23(10), 1437-1443.
[16] Quinlan, J.R. (1986). “Introduction of decision trees”, Machine learning, 1(1), 81-106.
[17] امامی‌فر، س.، رحیمی‌خوب، ع.، نوروزی، ع.ا. (1393). "ارزیابی مدل درختی M5 و شبکه‌ی عصبی مصنوعی برای برآورد متوسط روزانه‌ی دمای هوا براساس داده‌های دمای سطح زمین سنجنده‌ی مودیس"، تحقیقات آب و خاک ایران، دوره 45، شماره 4، ش.ص. 423-433.
[18] Nourani, V., Komasi, M., & Alami, M.T. (2011). “Hybrid wavelet- genetic programming approach to opti-maize ANN modeling of rainfall – runoff Process”, Journal of Hydrologic Engineering, 17(6), 724-741.
CAPTCHA Image