The aim of this study was to investigate the use of artificial neural networks (ANNs) as a simulation tool to predict thermal
changes in urban areas. Istanbul Airport, which is a recently constructed mega project in Istanbul, was selected as the study
area. In the study, the texture transfer method which allows creating synthetic images is used as a new image processing
technique. This method allows land-use changes to be represented by texture pixels transfers that represent similar land-use
types before the construction phase of a project. According to the results, while the surface temperature predictions for the
warmed and cooled points were obtained with 79% and 54% accuracy respectively for one-to-one overlapping, 97% accuracy
rates were reached within 300 m. This study revealed that the effect on the urban climate from surface changes that might
occur in cities can be predicted by employing an ANN algorithm.