ESTIMATING REFERENCE EVAPOTRANSPIRATION FOR MIDDLE EUPHRATES AREA USING ARTIFICIAL NEURAL NETWORKS (ANNs)

This study has been established for Middle Euphrates region which includes Hilla , Kerbala , Najaf , Muthanna and Diwaniya governorates , Which one of the very important areas of Iraq in terms of geographical location played this space, as well as being rich in water resources , This was due to the Euphrates River which made it the land of sedimentary soils suitable for agriculture so it is the main artery of the country's economy, From this standpoint emerged importance of this study, As we know that Iraq is classified as hot areas because its orbit near the equator, the amounts of water to evaporation taking place in this region took to rise gradually, especially in recent years, Therefore, it is necessary to have a quality control it, The primary purpose of research is to generate an equation practical and easy application for calculating reference evapotranspiration depending on the FAO Penman-Montieth using technique Artificial Neural Networks (ANNs), which computational models based on the structure and functions of biological networks and its considered nonlinear statistical data modeling tools where the complex relationships are modeled or patterns are formed , Artificial neural systems are the most beneficial researcher's subjects that deal with many sciences. This study depending on the climatic data were recorded from 1995 to 2015 years and was obtained coefficient of determination (R2 ) and root mean square error (RMSE) with respect FAO Penman-Montieth , the values of ANNs models (R2 and RMSE) are (0.969, 0.43) , (0.9713, 0.40) , (0.9707, 0.37) , (0.9681, 0.48) and (0.9742,0.42) for Hilla , Kerbala , Najaf , Muthanna and Diwanyia respectably.