This PhD thesis develops a physically based framework to quantify Land Surface Temperature (LST) in urban environments and to investigate the Surface Urban Heat Island Intensity (SUHII) at the intra-urban scale. The analysis focuses on spatial thermal contrasts within cities, comparing different urban and peri-urban typologies, including historical centers, residential and industrial areas, and surrounding rural environments. The proposed methodology is based on a coupled representation of surface water and energy exchanges at the soil–atmosphere interface, referred to as the Surface Water Energy Balance (SWEB) framework. The model explicitly links soil moisture dynamics to latent and sensible heat fluxes, enabling a physically consistent simulation of LST variability under different land-cover and climatic conditions. The SWEB model was calibrated and validated in two pilot cities with contrasting climatic and urban characteristics: Milano and Amsterdam. In Milano, simulations were performed at daily scale on a 500 m × 500 m grid for the period 2010–2022, including climate projections up to 2100 under multiple scenarios. Short-, mid-, and long-term horizons were analyzed. In Amsterdam, the focus was on the historical city center within the Singelgracht canal system, using a finer spatial resolution (100 m × 100 m) and hourly simulations for 2020–2024. In both case studies, the cooling role of urban blue–green infrastructure was assessed. Meteorological forcing was derived from automatic weather stations, while soil and land-cover properties were obtained by integrating satellite data with local urban datasets. Vegetation in Milano was represented through a Vegetated Fraction (VF), whereas built-up, green, and water fractions were used for Amsterdam. Satellite-derived LST was employed for model calibration, using MODIS data for Milano and Landsat 8 imagery for Amsterdam, complemented by flux-tower observations in the Amsterdam city center. In Milano, mean summer LST reaches 35.37 °C, with differences of about 3.7 °C between paved and green surfaces and peaks exceeding 4.5 °C. A 10% increase in vegetation cover corresponds to an average LST reduction of 0.23 °C. Heatwave summers, such as 2015 and 2022, exhibit LST values approximately 4 °C above the multi-year mean. In Amsterdam, daytime LST averages 23.99 °C, with built-up areas warmer than green and water surfaces, resulting in a mean SUHII of +1.72 °C relative to green areas and +4.21 °C relative to water bodies. Future projections indicate substantial changes in urban thermal conditions. In Milano, summer LST is projected to decrease by up to 2.66 °C under SSP1–2.6, while SUHII remains high (+1.55 °C). Under SSP5–8.5, LST increases by 3.98 °C and SUHII weakens (+0.51 °C) due to reduced soil moisture and limited evapotranspiration. Consistently, results from Amsterdam highlight the strong cooling effect of urban water bodies, with LST values about 4 °C lower than surrounding built-up areas. Overall, this thesis provides a transferable and physically based framework for assessing urban LST and SUHII, supporting urban planning and climate adaptation strategies under present and future climate conditions
Questa tesi di dottorato sviluppa un quadro modellistico fisicamente basato per quantificare le Land Surface Temperature (LST) in ambiente urbano e analizzare l’Intensità dell’Isola di Calore Urbana Superficiale (Surface Urban Heat Island Intensity, SUHII) alla scala intra-urbana. L’analisi si concentra sui contrasti termici spaziali all’interno delle città, confrontando diverse tipologie urbane e peri-urbane, inclusi centri storici, aree residenziali e industriali, e ambienti rurali circostanti. L’approccio proposto si basa su una rappresentazione accoppiata dei bilanci di acqua ed energia alla superficie suolo–atmosfera, denominata Surface Water Energy Balance (SWEB). Il modello collega la dinamica dell’umidità del suolo ai flussi di calore latente e sensibile, consentendo una simulazione fisicamente coerente della variabilità delle LST in funzione della copertura del suolo e delle condizioni climatiche. Il modello SWEB è stato calibrato e validato in due città pilota con differenti caratteristiche climatiche e urbanistiche: Milano e Amsterdam. A Milano, le simulazioni sono state condotte a scala giornaliera su una griglia di 500 m × 500 m per il periodo 2010–2022, includendo proiezioni climatiche fino al 2100 secondo diversi scenari. Ad Amsterdam, l’analisi si è concentrata sul centro storico delimitato dalla Singelgracht, utilizzando una risoluzione spaziale di 100 m × 100 m e una scala temporale oraria per il periodo 2020–2024. In entrambi i casi studio è stato analizzato il ruolo mitigativo delle infrastrutture urbane blu–verdi. Le forzanti meteorologiche derivano da stazioni automatiche, mentre le proprietà del suolo e la copertura del suolo sono state ottenute integrando dati satellitari e dataset di pianificazione urbana. Le LST satellitari sono state utilizzate per la calibrazione del modello, impiegando dati MODIS per Milano e immagini Landsat 8 per Amsterdam, integrate in quest’ultimo caso con osservazioni da flux tower. A Milano, le LST medie estive raggiungono 35.37 °C, con differenze di circa 3.7 °C tra superfici pavimentate e aree verdi; un aumento del 10% della copertura vegetata comporta una riduzione media delle LST di 0.23 °C. Le estati con frequenti ondate di calore mostrano valori medi superiori di circa 4 °C rispetto alla media pluriennale. Ad Amsterdam, le LST diurne medie sono pari a 23.99 °C, con temperature più elevate nelle aree edificate rispetto a quelle verdi e ai corpi idrici, determinando una SUHII media di +1.72 °C e +4.21 °C rispetto alle superfici verdi e d’acqua. Le proiezioni climatiche indicano una significativa modifica delle condizioni termiche urbane nel XXI secolo. A Milano, le LST estive sono previste diminuire fino a 2.66 °C nello scenario SSP1–2.6, mentre nello scenario SSP5–8.5 aumentano di 3.98 °C con una riduzione della SUHII, dovuta alla diminuzione dell’umidità del suolo e dell’efficienza evaporativa. Coerentemente, il caso di Amsterdam evidenzia il ruolo mitigativo dei corpi idrici urbani, caratterizzati da LST inferiori di circa 4 °C rispetto alle superfici edificate. Nel complesso, questa tesi fornisce un quadro quantitativo e fisicamente basato per la valutazione delle LST urbane e delle SUHII, offrendo uno strumento trasferibile a supporto della pianificazione urbana e delle strategie di adattamento climatico.
Modello idrologico-energetico per la valutazione delle Land Surface Temperature e delle Isole di Calore Urbano: i casi Milano ed Amsterdam / Morgese, Sonia. - (2026 May 14).
Modello idrologico-energetico per la valutazione delle Land Surface Temperature e delle Isole di Calore Urbano: i casi Milano ed Amsterdam.
MORGESE, SONIA
2026-05-14
Abstract
This PhD thesis develops a physically based framework to quantify Land Surface Temperature (LST) in urban environments and to investigate the Surface Urban Heat Island Intensity (SUHII) at the intra-urban scale. The analysis focuses on spatial thermal contrasts within cities, comparing different urban and peri-urban typologies, including historical centers, residential and industrial areas, and surrounding rural environments. The proposed methodology is based on a coupled representation of surface water and energy exchanges at the soil–atmosphere interface, referred to as the Surface Water Energy Balance (SWEB) framework. The model explicitly links soil moisture dynamics to latent and sensible heat fluxes, enabling a physically consistent simulation of LST variability under different land-cover and climatic conditions. The SWEB model was calibrated and validated in two pilot cities with contrasting climatic and urban characteristics: Milano and Amsterdam. In Milano, simulations were performed at daily scale on a 500 m × 500 m grid for the period 2010–2022, including climate projections up to 2100 under multiple scenarios. Short-, mid-, and long-term horizons were analyzed. In Amsterdam, the focus was on the historical city center within the Singelgracht canal system, using a finer spatial resolution (100 m × 100 m) and hourly simulations for 2020–2024. In both case studies, the cooling role of urban blue–green infrastructure was assessed. Meteorological forcing was derived from automatic weather stations, while soil and land-cover properties were obtained by integrating satellite data with local urban datasets. Vegetation in Milano was represented through a Vegetated Fraction (VF), whereas built-up, green, and water fractions were used for Amsterdam. Satellite-derived LST was employed for model calibration, using MODIS data for Milano and Landsat 8 imagery for Amsterdam, complemented by flux-tower observations in the Amsterdam city center. In Milano, mean summer LST reaches 35.37 °C, with differences of about 3.7 °C between paved and green surfaces and peaks exceeding 4.5 °C. A 10% increase in vegetation cover corresponds to an average LST reduction of 0.23 °C. Heatwave summers, such as 2015 and 2022, exhibit LST values approximately 4 °C above the multi-year mean. In Amsterdam, daytime LST averages 23.99 °C, with built-up areas warmer than green and water surfaces, resulting in a mean SUHII of +1.72 °C relative to green areas and +4.21 °C relative to water bodies. Future projections indicate substantial changes in urban thermal conditions. In Milano, summer LST is projected to decrease by up to 2.66 °C under SSP1–2.6, while SUHII remains high (+1.55 °C). Under SSP5–8.5, LST increases by 3.98 °C and SUHII weakens (+0.51 °C) due to reduced soil moisture and limited evapotranspiration. Consistently, results from Amsterdam highlight the strong cooling effect of urban water bodies, with LST values about 4 °C lower than surrounding built-up areas. Overall, this thesis provides a transferable and physically based framework for assessing urban LST and SUHII, supporting urban planning and climate adaptation strategies under present and future climate conditions| File | Dimensione | Formato | |
|---|---|---|---|
|
Morgese_phd_thesis_revised.pdf
accesso aperto
Descrizione: Tesi def
Tipologia:
Tesi di dottorato
Dimensione
152.56 MB
Formato
Adobe PDF
|
152.56 MB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.


