Evapotranspiration prediction using M5T method and Ritchie equation for St. Johns, FL, USA

Abstract

Evapotranspiration (ET) estimation takes an important role in hydraulic designs and agricultural yield. Even it is nonnegligible for hydraulic designers and irrigation engineers it is not clear enough to estimate or calculate ET because of direct and indirect parameters effects. In this study Solar Radiation (SR), Air Temperature (AT), Relative Humidity (RH) and Wind Speed (U) meteorological parameters are used to create a M5T model. 1158 daily RH, U, AT and SR records are used to create model and 385 daily values are used to test it. Data set is taken from St. Johns, Florida, USA weather station. The test set is also applied to the Ritchie empirical formula. M5T model and Ritchie formula Results are compared with daily ET records using determination coefficient. Determination coefficient is found 0.966 for M5T model and 0.913 for Ritchie formula. According to the determination coefficient, Mean Square Error (MSE) and Mean Absolute Error (MAE) statistics, it is understood that M5T method can be used for daily ET estimation effectively.