A research team from the Upper Euphrates Basin Developing Centre succeeds in modelling daily transpiration evaporation in dry areas (Anbar Governorate) using artificial intelligence applications

2024-06-20

A research team from the Upper Euphrates Basin Developing Centre - University of Anbar has demonstrated the modelling of evapotranspiration (ET) evaporation processes in the city of Ramadi using one of the applications of artificial intelligence, artificial neural networks. Dr. Atheer Salim Al-Mawla, the first researcher in the team, stated that the research, tagged ((Predictive Modelling of Daily Evapotranspiration in Arid Regions Using Artificial Neural Networks)), has obtained acceptance for publication in a solid scientific journal indexed within the Scopus (Q2) database. "The process of determining evaporation with high accuracy in arid lands is one of the most important processes in hydrological studies, and it is of great importance in the effective management of water resources and hydrological modelling, as well as in the management of irrigation operations, because these areas are permanently linked to the issue of water scarcity." Predicting evaporation and transpiration is a vital step towards water resources management.

The study aims to develop an artificial neural network model to predict evaporation in arid and semi-arid areas such as the western region (Anbar Governorate). RBFNN and GRNN models were used with six data input in this study. The input data are maximum temperature, minimum temperature, AVA, temperature, humidity, wind speed and solar radiation, ANN modeling was achieved using MATLAB with hyperbolic sigmoid transfer function for both input and output layers. Several statistical indicators were used to examine the accuracy of the model's prediction The results showed that the current model is a powerful model and has the ability to predict evaporation and transpiration with high accuracy The superiority of the GRNN model is very clear compared to the RBFNN model where the coefficient of determination for the GRNN model was more than 96% compared to 94% for the RBFNN model and the mean square error for the GRNN model was 0.4 compared to 0.52 for the RBFNN model. In addition to Dr. Atheer Salim, the research team included Dr. Bashir Khalil and Dr. Ahmed Saud.

This research is the third to be published this year by the research team after the publication of two other research papers in cooperation with researchers from the centre (Dr. Haitham Abdul Mohsen Afen and Prof. Dr. Ammar Hatem Kamel) in partnership with researchers from other international universities.

#university_of_anbar
#Upper_Euphrates_Basin_Developing_Center