Gene expression programming and artificial neural network to estimate atmospheric temperature in Tabuk, Saudi Arabia

Abstract

Climate change is not a myth. There is enough evidence to showcase the impact of climate change. Town planners and authorities are looking for potential models to predict the climatic factors in advance. Being an agricultural area in Saudi Arabia, Tabuk region gets greater interest in developing such a model to predict the atmospheric temperature.Therefore, this paper presents two diferent studies based on artifcial neural networks (ANNs) and gene expression programming (GEP) to predict the atmospheric temperature in Tabuk. Atmospheric pressure, rainfall, relative humidity and wind speed are used as the input variables in the developed models. Multilayer perceptron neural network model (ANN model), which is high in precession in producing results, is selected for this study. The GEP model that is based on evolutionary algorithms also produces highly accurate results in nonlinear models. However, the results show that the GEP model outperforms the ANN model in predicting atmospheric temperature in Tabuk region. The developed GEP-based model can be used by the town and country planers and agricultural personals