Machine Learning–Driven Assessment of Urban Heat Island Intensity in Athens: Integrating Remote Sensing, Landscape Metrics, and Green Infrastructure for Climate-Resilient Urban Planning

Abstract

Environmental and health risks are increasing in large Mediterranean cities like Athens, where high urban density and limited vegetation intensify the Urban Heat Island (UHI) effect. This study analyzes the spatial patterns of surface urban heat in Athens, Greece, using satellite-derived Land Surface Temperature (LST) as a proxy for UHI conditions. LST datasets were derived from Landsat 8 thermal imagery and Sentinel-2 forecasts within Google Earth Engine (GEE) to achieve a high-resolution 10 m grid. Environmental variables, including vegetation (NDVI), building intensity (NDBI), and topography (DEM and slope), were analyzed alongside landscape metrics (PLAND, LPI, FRAC, ED, and LSI) to quantify the influence of both urban composition and spatial configuration.
The research implemented a dual-track predictive framework: neighborhood-level machine learning models—Random Forest (RF), Support Vector Regression (SVR), Decision Tree (DT), and XGBoost—and a pixel-level Convolutional Neural Network (CNN) for spatial prediction. Results demonstrate that building intensity (NDBI) is the dominant driver of higher surface temperatures, while vegetation (NDVI) and topography contribute to localized cooling. Among the neighborhood-level models, Random Forest demonstrated the most consistent performance, achieving a maximum coefficient of determination (R2) of 0.5153 in August and a Root Mean Squared Error (RMSE) of approximately 1.75°C. In contrast, although the CNN model had low statistical accuracy (R2 < 0.20), due to data configurations and sensor noise, it managed to capture thermal points that were invisible in tables.
Finally, the study highlights the great importance of integrating remote sensing, GIS-based landscape measurements, and machine learning for the changes that the climate crisis will bring. It thus appears that urban planning, at least in Mediterranean cities, should go a step beyond simple green space and, to more effectively address heat, should take into account spatial geometry and topographic factors. However, the study was limited by the small sample size of 53 administrative neighborhoods, thus limiting the statistical power of high-variance models such as XGBoost and CNN.

Popular Abstract

Every summer, Athenians and their visitors increasingly experience heat waves, which last longer and are often uncomfortable or even cause health problems. Especially in the city center, where it is densely built and has little vegetation, there are areas that are too hot, and residents would like the situation to change because climate change creates additional problems and anxiety.
This study analyzes the distribution of urban heat in the center of Attica and the environmental factors that affect the temperature from place to place. Using satellite images, geographic information systems (GIS) and artificial intelligence techniques, the study examined 53 neighborhoods throughout the city during the summer of 2024.
The analysis showed that the hottest neighborhoods are those with more construction (buildings, paved surfaces) and less greenery. Trees and greenery in general contributed to the reduction in surface temperatures, while heat was trapped by the rest of the city's materials. It was also found that the layout of green space also plays a major role. That is, along with the amount of greenery, its location and spatial pattern influenced the temperature pressure in the area.
Various machine learning models were used to identify the most important factors that affect urban temperatures. Of these models, Random Forest produced the most reliable results. Building density was identified as the strongest factor contributing to the increase in temperatures, while vegetation and topographic features, such as altitude, helped retain heat in some areas.
Simply increasing total green space will not improve the quality of urban life. City planners should plan for future megalopolis growth by reviewing how to connect the lots with green space throughout the city. When considering how much more green space a city will need for future growth, planners should investigate how to best utilize the available green infrastructure, such as trees, parks, and other green areas, to make cities cooler, thereby decreasing the potential for excessive heat produced by future heat waves due to climate change.
City governments in the Mediterranean region are experiencing many of the challenges of climate change as their cities continue to grow and develop. There is an unprecedented amount of data available to city officials to identify and analyze their cities' heat islands. Examples of these tools can include satellite data, GIS, and machine learning algorithms that can help city officials better understand the causes of urban heat islands. For instance, by using urban heat island data, Athens city officials are better able to provide their residents with a safer, more livable community.

• author: Fyntanidis, Athanasios
• supervisor: Pengxiang Zhao; Rachid Oucheikh
• organization: Department of Earth and Environmental Sciences (MGeo)
• course: GISM01 20261
• year: 2026
• type: H2 - Master's Degree (Two Years)
• subject: Earth and Environmental Sciences
• keywords: Keywords: Urban Heat Island (UHI), Land Surface Temperature (LST), Athens, Remote Sensing, Google Earth Engine (GEE), Landscape Metrics, Machine Learning, Support Vector Regression (SVR), Random Forest (RF), Convolutional Neural Network (CNN), Urban Planning, Climate Resilience.
• publication/series: Master Thesis in Geographic Information Science
• report number: 214
• language: English
• id: 9238013
• date added to LUP: 2026-06-16 09:01:53
• date last changed: 2026-06-16 09:01:53