Forecasting of Pomegranate (Punica granatum L.) evapotranspiration by using Seasonal ARIMA Model

This paper deals with the stochastic modeling of weekly reference crop evapotranspiration (ET) by using Seasonal ARIMA model for the period of 1984–2009 with a total of 1352 readings for Pune district in Maharashtra. An ARIMA model of 1^st order was selected based on observing autocorrelation function (ACF) and partial autocorrelation function (PACF) of the weekly time series of ET_r. The model parameters were obtained by using maximum likelihood method with the help of three tests (i.e., standard error, ACF and PACF of residuals and Akaike Information Criteria) for adequacy test of the selected model. The ARIMA model that passed the adequacy test was selected for forecasting. One year ahead forecast of ET_r values were obtained with the help of these selected models and compared with the values of ET_r obtained from the climatological data of 2010 by root mean square error (RMSE). The Seasonal ARIMA (1, 1, 1) (0, 1, 1) with lower RMSE is finally selected for forecasting of weekly pomegranate evapotranspiration values in Ld⁻¹ t⁻¹. The seasonal values i.e., 1026, 2673, 7337, 10145 and 12878 Lyr⁻¹ tr⁻¹ of forecasted pomegranate evapotranspiration (ET_r) for 1^st to 5^th year pomegranate trees were to be applied. The obtained results showed that the forecasted ET_P values by Seasonal ARIMA model are very effective and reliable.