Reference Evapotranspiration Prediction from Limited Climatic Variables Using Support Vector Machines and Gaussian Processes

Climatic variables collected from weather stations evenly distributed in all regions of Turkey were used to study the potential Gaussian Process Regression (GPR) and Support Vector (SVR) predicting reference evapotranspiration (ET0). The as input features for GPR SVR models solar radiation, mean temperature, wind speed, relative humidity, month year. corresponding ET0 values calculated using Food Agriculture Organization recommended equation FAO 56 PM climatic measurements same stations. Results show that regression with high accuracies are possible models. most effective variable prediction was found be radiation. Relative humidity had lowest impact on model accuracies.