Estimation of Maximum, Mean and Minimum Air Temperature in Tabriz City Using Artificial Intelligent Methods

Estimating air temperature is one of the important issues in agricultural planning and in water resources management which can be accomplished by using different methods such as empirical, semi-empirical and intelligent methods. In the present study, Adaptive Neuro Fuzzy Inference System, Artificial Neural Networks and Genetic Programming were used to estimate air temperature in the synoptic station of Tabriz City, northwest of Iran. Considering the statistical indices, all three models were able to estimate accurately minimum, mean and maximum air temperature. In spite of slight differences in the prediction accuracy and errors by the models, Adaptive Neuro Fuzzy Inference System, Artificial Neural Networks and Genetic Programming were in the order of priority. Also explicit solutions that show the relation between input and output variables are presented based on Genetic Programming. This adds to the superiority of Genetic Programming over the other two models.