Exploring the influence of building morphology on surface temperatures: A multi-city analysis in Europe

Rapid urbanization is transforming cities, along with altering the urban forms. This results in diverse building structures and materials that often intensify the thermal environment, making cities hotter than their surroundings. Despite the growing risks, the role of urban forms in influencing land surface temperature (LST) for informing climatic neutral urban planning and design remains insufficiently understood. Previous research has primarily focused on landscape metrics, spectral indices, and surface attributes affecting LST and to understand urban heat patterns. Comprehensive 2D metrics describing urban forms are scarce particularly in European cities, despite better data availability. This study investigates the impacts of urban forms on LST in Paris, Rotterdam, Milan, and Vienna across different climate conditions. An urban morphological approach is adopted to analyse building-level urban forms with a set of twenty-four urban morphometrics (UMMs) calculated using momepy. A Random Forest Regression (RFR) model is employed at a grid level of 70 by 70 m (spatial resolution of ECOSTRESS data) to explore the relationship between UMMs and LST. The RFR model explains over 80 % of LST variability, with mean height, orientation, alignment, building adjacency, and interbuilding distance as key influential factors. LST correlates positively with mean building height in most cities. Orientation significantly contributes to LST variation, but its impact depends on the local context. Building adjacency and alignment have non-linear cooling effects across all cities. This study provides a new perspective by encompassing a wide range of UMMs to study LST patterns, crucial for effective urban heat mitigation strategies.