Spatial Modeling of the Impact of Climate Change on Thermal Comfort Using Geospatial Techniques and Artificial Neural Networks: A Case Study of Northwest Jordan

The extensive use of high-resolution digital elevation data, along with continuous improvements in computing power and geographic information system (GIS) tools. Has driven the development of spatial data processing, management, and spatial interpolation methods. This study aims to construct a high-quality spatial distribution map of thermal comfort in densely populated areas of northwestern Jordan using climate data collected from six meteorological stations between 1991 and 2024, based on the indoor temperature index (IAT). To analyze the spatial variability of climate elements, a digital elevation model (DEM) with a spatial resolution of 30 m was used and resampled to a 0.5-km grid. Spatial interpolation employed inverse distance weighting (IDW), with each grid cell using data from the three nearest meteorological stations. The results showed that areas with higher temperatures inside the villas were clearly concentrated in the summer, especially in the lowlands near the Jordan Valley. Indicating that these areas are more susceptible to thermal stress. The model results also show that it performs well in predicting thermal comfort, with a coefficient of determination (R2) between 0.95 and 0.98 and mean squared error (MSE) between 0.35 and 0.50. which reflects the ability of these models to represent the relationship between climate variables and predict thermal comfort levels with a high degree of accuracy. The results indicate significant spatiotemporal differences in thermal comfort within the study area, with longer durations of heat stress in summer. This highlights the importance of combining geospatial methods with numerical simulations in studying the impacts of climate change and supporting urban planning and climate adaptation strategies.