Prediction of evaporation in arid and semi-arid regions: a comparative study using different machine learning models

• Published in: Engineering applications of computational fluid mechanics 2020-01, Vol.14 (1), p.70-89
• Main Authors: Yaseen, Zaher Mundher, Al-Juboori, Anas Mahmood, Beyaztas, Ufuk, Al-Ansari, Nadhir, Chau, Kwok-Wing, Qi, Chongchong, Ali, Mumtaz, Salih, Sinan Q., Shahid, Shamsuddin
• Format: Article
• Language: English
• Subjects: arid and semi-arid regions; Arid regions; best input combination; Comparative studies; Developing countries; Evaporation; Gene expression; Geoteknik; Humidity; Hydrology; LDCs; Machine learning; Neural networks; Performance prediction; predictive model; Rainfall; Regression analysis; Relative humidity; Semi arid areas; Soil Mechanics; Support vector machines; Water resources; Weather stations; Wind speed
• ISSN: 1994-2060; 1997-003X; 1997-003X
• DOI: 10.1080/19942060.2019.1680576

Evaporation, one of the fundamental components of the hydrology cycle, is differently influenced by various meteorological variables in different climatic regions. The accurate prediction of evaporation is essential for multiple water resources engineering applications, particularly in developing countries like Iraq where the meteorological stations are not sustained and operated appropriately for in situ estimations. This is where advanced methodologies such as machine learning (ML) models can make valuable contributions. In this research, evaporation is predicted at two different meteorological stations located in arid and semi-arid regions of Iraq. Four different ML models for the prediction of evaporation - the classification and regression tree (CART), the cascade correlation neural network (CCNNs), gene expression programming (GEP), and the support vector machine (SVM) - were developed and constructed using various input combinations of meteorological variables. The results reveal that the best predictions are achieved by incorporating sunshine hours, wind speed, relative humidity, rainfall, and the minimum, mean, and maximum temperatures. The SVM was found to show the best performance with wind speed, rainfall, and relative humidity as inputs at Station I (R 2  = .92), and with all variables as inputs at Station II (R 2  = .97). All the ML models performed well in predicting evaporation at the investigated locations.