Applied Research in Geographical Sciences

Applied Research in Geographical Sciences

Determination of the best time series model for forecasting annual rainfall of selected stations of Western Azerbaijan province

Authors
Somayeh Soltani gerdfaramarzi
Aref Saberi
Morteza Gheisouri
Abstract
Rainfall is one of the most important components of the water cycle and plays a very important role in the measurement of climate characteristic in any area. Limitations such as lack of sufficient information about the amount of rainfall in time and space scale and complexity of the relationship between meteorological elements related to rainfall, causes the calculation of these parameters using the conventional method not to be implemented. One method of evaluating and forecasting of rainfall in each region is time series models. In this research, to predict the average annual rainfall synoptic station at Mahabad, Uromiya and Mako in West Azarbayejan provience during 1984-2013, linear time series ARIMA was used. To investigate model static, Auto Correlation Function (ACF) and Partial Auto Correlation Function (PACF) was applied and with differencing method, the non-static data transformed to static data. In next step, stochastic models to estimate the annual rainfall average were used. With regard to the evaluation criterion such as T, P-VALUE < 0.05 and Bayesian Information Creterion (BIC), ARIMA (1,0,0), ARIMA (0,1,1) and ARIMA (0,1,1) models was determined as a suitable model for predicting annual rainfall in the three selected stations at Uromiya, Makoo and Mahabad. In the following, the annual rainfall for 3 (2013-2016) years is forecasted which based on rainfall data in that time, the adjusted model was acceptable.
Keywords
Prediction
autocorrelation
annual rainfall
Time Series
Arima
Aguilera A. M. Escabias M. Valderrama M. (2007). Forecasting binary longitudinal data by a functional PC-ARIMA model. Computational statistical & data analysis, (52): 3187-3197.

Borland P. Montana A. (1996). Forecasting of storm rain full by combined use of rider, rain gages and linear models, Atmospheric research, 42:199-216.

Box G.E. Jenkins G.M. Reinsel G.C. Ljung G.M. (2015). Time series analysis: forecasting and control. John Wiley & Sons.

Brockwell P.J. Davis R.A. (1996). Introduction to time series and forecasting. Springer Verlag, New York, Inc.

Chatfield C. (1996). The analysis of time series: an introduction. 5th edition. Chapman and Hall.UK.

Matalas N. C. (1967). Time series analysis. Water Resources Research, 3(3):817-829.
McKerchar A.I. Delleur J.W. (1974).Application of seasonal parametric linear stochastic models to monthly flow data. Water Resources Research, (10): 246-255.

Naill P.E. Momani M. (2009. Time Series Analysis Model for Rainfall Data in Jordan: Case Study for Using Time Series Analysis. American Journal of EnvironmentalSciences, 5(5): 599-604.

Ruhf R. Cutrim E.M. (2003). Cutrim. Time seriesanalysis of 20 years of hourly precipitation in southwest Michigan. Journal of Great Lakes Research, 29(2): 256-267.

Zou P. Jingsong, Y. Jianrong F. Guangming L. Dongshun L. (2010). Artificial neural network and time series models for predicting soil salt and water content. Agricultural water management, 97: 2009-2019.