Multi-scale observations and urban canopy modelling of heat exposure in a Mediterranean coastal city

Urban areas in the Mediterranean basin are increasingly exposed to thermal stress as a result of climate change and ongoing urbanization, creating an urgent need for urban climate information that supports heat-risk assessment and adaptation strategies at city scale. This study presents an integrated multiscale assessment of the urban microclimate in Bari (southern Italy), a mid-sized Mediterranean coastal city, with the aim of disentangling the relative contributions of sea–land breeze dynamics and urban morphological characteristics to intra-urban thermal variability. The analysis combines three complementary approaches. First, in situ observations of air temperature and relative humidity were collected during summer 2023 using a dense network of eight canyon-level sensors distributed across neighborhoods characterized by different distances from the coastline, building density, vegetation cover, and land use. Second, satellite-derived land surface temperature (LST) from ECOSTRESS was employed to provide a spatially continuous view of surface thermal patterns at different times of the day. Third, the recently developed offline MLUCM BEP+BEM urban canopy model (Pappaccogli et al., 2025) was applied and evaluated against observations as a science-based tool for representing intra-urban thermal variability under realistic mesoscale forcing. Observations reveal a pronounced coastal–inland gradient in both air temperature and humidity, particularly during daytime, driven by the onset and persistence of sea-breeze circulations. Coastal locations experience moderated warming and higher humidity, whereas inland districts exhibit stronger heating and daytime drying, amplifying thermal stress. Satellite LST confirms these patterns, highlighting persistent hotspots in dense urban fabrics and large impervious areas, while also capturing the diurnal evolution of surface thermal contrasts. Model results demonstrate that MLUCM BEP+BEM improves the representation of intra-urban variability compared to reanalysis data alone, particularly in reproducing canopy-level temperature differences across neighborhoods. While mesoscale forcing largely controls the background climate signal, microscale processes associated with urban geometry, surface properties, vegetation, and anthropogenic heat contribute substantially to spatial variability and are effectively captured by the model. The relative importance of these contributions varies with distance from the coastline and the choice of boundary forcing. Overall, this work highlights the necessity of integrating observations, remote sensing, and urban canopy modeling to accurately characterize thermal environments in Mediterranean coastal cities. The proposed framework is transferable to other coastal contexts and provides a robust basis for assessing urban heat exposure and for the development of urban climate services that support climate-sensitive planning and the evaluation of mitigation and adaptation strategies under current and future climate conditions. 

This work is supported by ICSC – Centro Nazionale di Ricerca in High Performance Computing, Big Data and Quantum Computing, funded by European Union – NextGenerationEU (CUP F83C22000740001).

Reference
Pappaccogli, G., Zonato, A., Martilli, A., Buccolieri, R., and Lionello, P.: MLUCM BEP + BEM: an offline one-dimensional multi-layer urban canopy model based on the BEP + BEM scheme, Geosci. Model Dev., 18, 7129–7145, https://doi.org/10.5194/gmd-18-7129-2025, 2025.