Performance Assessment of M5 Tree Model and Genetic Programming in Tabriz Station Reference Evapotranspiration Modeling

Evapotranspiration has a main role in water budget assessment and management. In plant water requirements and evapotranspiration volume calculation, firstly reference evapotranspiration (ET0) have to be computed and then plant water requirements can be estimated using different methods. In this research, firstly the reference evapotranspiration factor was calculated by standard FAO-Penman-Monteith formula via climatic data of Tabriz station, East Azerbaijan province. Climatic parameters include mean, minimum and maximum of air temperature also mean, minimum and maximum of relative humidity, rainfall, wind speed and sunshine hours were considered as an input of genetic programming and M5 tree models to estimation of monthly reference evapotranspiration as an output. Results showed that both of two approaches present exact results (determination coefficient for M5 tree model equal to 0.99 and for GP equal to 0.96) in estimating monthly reference evapotranspiration in Tabriz region, but M5 model tree, provides understandable, applicable and simple linear relations to estimate reference evapotranspiration.