Local Climate Zones (LCZs) are a standardized classification system for describing urban and rural landscape based on surface cover and morphology, dividing the land surface into 17 classes based on built-up and land cover characteristics. This approach provides a unified framework for comparing cities globally and analyzing the impact of urban morphology on its microclimate. LCZs are now widely used in urban climate modeling and Earth Observation data are exploited for their identification, by applying machine/deep learning models on satellite imagery, where the collection of training and validation sites from local knowledge plays an important role. With the goal of mapping LCZs for the city of Bologna, the analyses were carried out using a PRISMA (L2D) image and a Harmonized Sentinel-2 (L2A) image. Urban Canopy Parameters (UCPs) were included as additional features for enhancing the classification performance. Specifically, The Building height layer was obtained from the municipal cadastral map of Bologna (scale 1:2000) by resampling it to 10 and 30 meters (for Sentinel-2 and PRISMA respectively), while the impervious surface fraction and tree canopy height layers were derived from Copernicus datasets provided by the European Space Agency. Training and validation sites were defined in GIS environment by using the Sentinel-2 acquisition, the building height layer, and high resolution imagery as reference. Finally, Random Forest and Support Vector Machine were used to classify the images in Phyton environment. The performance of the classifiers was ultimately assessed using a consistent set of training and validation sites for both PRISMA and Sentinel-2 imagery. Principal components derived from PRISMA significantly improved classification results, highlighting the importance of applying feature reduction techniques to hyperspectral data. Future developments will focus on the characterization of LCZs through time series analysis of thermal infrared data, leveraging both satellite observations and ground-based measurements, assessing the vulnerability of the municipality of Bologna to Urban Heat Island effect.

Lo Grasso, G., Ventura, M., Mandanici, E., Bitelli, G. (2025). Preliminary Results for Local Climate Zones mapping of Bologna. AIT.

Preliminary Results for Local Climate Zones mapping of Bologna

G. Lo Grasso;M. Ventura;E. Mandanici;G. Bitelli
2025

Abstract

Local Climate Zones (LCZs) are a standardized classification system for describing urban and rural landscape based on surface cover and morphology, dividing the land surface into 17 classes based on built-up and land cover characteristics. This approach provides a unified framework for comparing cities globally and analyzing the impact of urban morphology on its microclimate. LCZs are now widely used in urban climate modeling and Earth Observation data are exploited for their identification, by applying machine/deep learning models on satellite imagery, where the collection of training and validation sites from local knowledge plays an important role. With the goal of mapping LCZs for the city of Bologna, the analyses were carried out using a PRISMA (L2D) image and a Harmonized Sentinel-2 (L2A) image. Urban Canopy Parameters (UCPs) were included as additional features for enhancing the classification performance. Specifically, The Building height layer was obtained from the municipal cadastral map of Bologna (scale 1:2000) by resampling it to 10 and 30 meters (for Sentinel-2 and PRISMA respectively), while the impervious surface fraction and tree canopy height layers were derived from Copernicus datasets provided by the European Space Agency. Training and validation sites were defined in GIS environment by using the Sentinel-2 acquisition, the building height layer, and high resolution imagery as reference. Finally, Random Forest and Support Vector Machine were used to classify the images in Phyton environment. The performance of the classifiers was ultimately assessed using a consistent set of training and validation sites for both PRISMA and Sentinel-2 imagery. Principal components derived from PRISMA significantly improved classification results, highlighting the importance of applying feature reduction techniques to hyperspectral data. Future developments will focus on the characterization of LCZs through time series analysis of thermal infrared data, leveraging both satellite observations and ground-based measurements, assessing the vulnerability of the municipality of Bologna to Urban Heat Island effect.
2025
Smart Earth Observation for a Sustainable Future - Abstract book
96
96
Lo Grasso, G., Ventura, M., Mandanici, E., Bitelli, G. (2025). Preliminary Results for Local Climate Zones mapping of Bologna. AIT.
Lo Grasso, G.; Ventura, M.; Mandanici, E.; Bitelli, G.
File in questo prodotto:
Eventuali allegati, non sono esposti

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11585/1031000
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact