Jordan Journal of Earth and Environmental Sciences Developing Reference Crop Evapotranspiration Time Series Simulation Model Using Class a Pan: A Case Study for the Jordan

The greatest environmental challenge that Jordan faces today is the scarcity of water resources. Evapotranspiration (ET) affects water resources and it is considered an important process in aridic hydrologic systems. The estimated long-term average of ET in Jordan is over 90 % of the total precipitation; nevertheless, there have been no attempts to model reference crop evapotranspiration using a time series approach in Jordan. In this study, a seasonal time series Autoregressive and Moving Average (ARIMA) mathematical model is described. It is used for forecasting monthly reference crop evapotranspiration (ETo) without using weather data based on past historical records (1973-2002) of measured pan evaporation at Central Jordan Valley: an arid to semi-arid region. The developed ARIMA (1, 0, 0) (0, 1, 1)12 model provides reasonable and acceptable forecasts, comparing its performance with a computed reference evapotranspiration from measured pan evaporation parameter. The forecasting performance capability of three tentative ARIMA models was assessed using Root Mean Squared Forecasting Error, Mean Absolute Forecasting Percentage Error, and Maximum Absolute Forecasting Percentage Error. The developed model allows local farmers and water resource managers to predict up to 60 months with a percentage error less than 11 % of the mean absolute forecasting. The potential to make such predictions is crucial in optimizing the needed resources for effective management of water resources. Furthermore, the developed model offers a simple, accurate, and an easy short and long-term forecasting in the valley. This would develop a robust strategy for irrigatio