Prediction and evaluation of average monthly temperature using time series models

Temperature is one of the most important climate parameters that used in many studies. This parameter is very important in assessment of climate changes and agricultural science, so that the temperature rise is one of the important environmental challenges for humans. Therefore its evaluation and prediction in long term can be effective in correct management of water and soil resources and preparation of plant water requirement. In this research were used Time series models, ARAR model and ITSM software to Prediction and evaluation of monthly temperature in Fasa Station (47 years from 1967 to 2014). In this research average monthly temperature in Fasa Station predicted for 6 next years (from 2015 to 2020). Result showed that based on the autocorrelation and partial autocorrelation diagrams, the AR Burg (26, 1) model with AICC index equivalent 2609.91 was the best fitted to the considered data set. According to results coefficient of model (Z(t-1)) is insignificant at 21, 22 and 23 lags therefore this coefficient set zero. The P-value of the turning points test statistic for different lags is greater than the significance level of 5%, which indicates that the residuals are uncorrelated, therefore can be said predicted data are thrust.