Reference Evapotranspiration Modeling Using Heuristic Computing Model in Distinct Climate Stations

Tahsin, Ala (2020) Reference Evapotranspiration Modeling Using Heuristic Computing Model in Distinct Climate Stations. Eurasian Journal of Science & Engineering, 6 (1). pp. 89-103. ISSN 24145629

Text (Research Article) eajse.v6i1p89.pdf - Published Version Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (1MB)

Abstract

Reference evapotranspiration (ET0) plays important roles in environmental, hydrological and agricultural studies and its accurate prediction is significant in water resources management and water productivity increase. This study focused on evaluating the ability of support vector regression (SVR) model for modeling ET0 in arid and semiarid climate stations of Iraq. For comparison, multiple linear regression (MLR) and calibrated Hargreaves and Samani (HS) empirical models were also applied. Daily meteorological data from Basra and Erbil stations including minimum, maximum and mean temperatures, relative humidity, wind speed, precipitation, solar radiation and surface pressure were collected for two consecutive years (2017 – 2018) and used as inputs to the models. FAO 56 PenmanMonteith was used as the benchmark ET0. Root mean square error (RMSE) and Nash Sutcliffe efficiency criterion (NSE) were the performance evaluation criteria employed. The results revealed that, all the applied models led to reliable results, but SVR model provided the best performance with NSEs of 0.9949, 0.9871 and RMSEs of 0.0009, 0.0016 in the validation phase for Basra and Erbil stations, respectively. The general results implied that SVR model could be employed successfully for estimation of ET0 in arid and semiarid climate stations of Iraq.

Actions (login required)

View Item