Prediction of surface temperature changes caused by the construction of Istanbul's mega airport using ANN in remote sensing

Over the past 25 years, Istanbul has experienced an imbalance in its urban climate due to uncontrolled urban development and loss of green spaces, making it vulnerable to abrupt weather changes caused by global climate change. The issue of urban climate change is multi-factorial, involving physical, social and economic dimensions, and should be addressed through effective planning strategies. It is crucial to accurately predict the future and ensure the success of the strategies implemented to address this challenge.

This research focuses on investigating the predictability of thermal changes in urban environments through the use of Artificial Neural Networks (ANN) in remote sensing. Istanbul Airport, a significant urban mega-project constructed in Istanbul, has been selected as the study site. The research uses the texture transfer method, a novel image processing technique that facilitates the generation of synthetic images. This approach allows the detection of thermal changes caused by land use changes in the project zones by applying texture transfers to pixels representing similar land uses according to the pre-construction project. The results show that the prediction accuracy can be up to 95% within a 300m neighbourhood. The results demonstrate that artificial neural network algorithms can effectively predict the impact of thermal changes on urban climate.

Keywords: Urban climate change, remote sensing, GIS, ANN, thermal change detection