The urban heat island (UHI) effect, characterized by higher temperatures in urban areas than in the surrounding rural environment, is being intensified in cities worldwide by growing urbanization and climate change. Measuring the UHI effect and its space-time patterns is needed for evidence-based decision-making and improved urban climate resilience. However, the limited spatial coverage of official and authoritative data often prevents in-depth analysis. Crowdsourced data may offer promising alternatives and can be integrated with official data sources to increase the spatial coverage of in-situ measurements, provided that the accuracy-related concerns are effectively addressed. Given the above considerations, in this thesis, we exploit crowdsourced air temperature data from the Netatmo amateur weather network to get insights into the UHI effect in the Metropolitan City of Milan (Northern Italy). We propose an automatic cleaning workflow (published open access on GitHub) to pre-process air temperature measurements and obtain analysis-ready time series. Specifically, we use official data from the Regional Agency for Environmental Protection (ARPA) as a reference to clean Netatmo air temperature observations. Geostatistical analyses are performed on the pre-processed data to investigate summertime temperature extremes and identify hot spots and cold spots in the study area. Finally, correlation analysis is carried out with multi-temporal Local Climate Zone (LCZ) maps. The research results indicate that crowdsourced data can effectively add value to the UHI assessment, provided detailed data cleaning is performed. The proposed methodology entirely relies on free and open-source software, thus enabling easy replication beyond the case study in this work.
Il fenomeno dell'isola di calore urbana, caratterizzato da temperature più elevate nelle città rispetto all'ambiente rurale circostante, si sta intensificando in tutto il mondo con la crescente urbanizzazione e i cambiamenti climatici. Per migliorare la resilienza delle città ed attuare strategie di mitigazione efficaci sono necessari strumenti e dati che permettano di misurare tale fenomeno. Tuttavia, la limitata disponibilità e copertura spaziale di dati ufficiali non permette di comprendere a fondo il fenomeno. I dati raccolti tramite \textit{crowdsourcing} possono offrire una soluzione al problema, a patto che i problemi legati all'accuratezza di tali dati vengano efficacemente indirizzati. Sulla base di queste considerazioni, la presente tesi si concentra sull'utilizzo di dati di temperatura dell'aria raccolti dalle centraline della rete amatoriale Netatmo per l'analisi dell'isola di calore urbana nella Città Metropolitana di Milano (Nord Italia). Viene proposta una procedura di pre-processamento automatico delle osservazioni (il cui codice è stato pubblicato su GitHub) per ottenere serie temporali di temperatura \textit{analysis-ready}. La procedura proposta utilizza i dati ufficiali dell'Agenzia Regionale per la Protezione dell'Ambiente (ARPA) come riferimento per pre-processare le osservazioni di temperatura di Netatmo. I dati pre-processati sono stati successivamente utilizzati per effettuare analisi geostatistiche, al fine di investigare i picchi di temperatura estivi e individuare gli \textit{hot spot} e \textit{cold spot} nell'area di studio. Infine, è stata analizzata la correlazione delle misure di temperatura con mappe multi-temporali delle zone climatiche locali (\textit{Local Climate Zone}, LCZ). I risultati ottenuti indicano che i dati raccolti tramite \textit{crowdsourcing} possono portare vantaggi significativi per analisi di dettaglio dell'isola di calore urbana, a patto di effettuare un'adeguata pulizia preliminare dei dati. La metodologia proposta si basa sull'utilizzo di strumenti \textit{free open-source}, che permettono di replicare le analisi ad altri casi di studio.
Exploring crowdsourced temperature data from Milan for urban heat island analysis
AHMED OMER AHMED MUKHTAR;Ahmed Mohamed Eltahir Yassin
2023/2024
Abstract
The urban heat island (UHI) effect, characterized by higher temperatures in urban areas than in the surrounding rural environment, is being intensified in cities worldwide by growing urbanization and climate change. Measuring the UHI effect and its space-time patterns is needed for evidence-based decision-making and improved urban climate resilience. However, the limited spatial coverage of official and authoritative data often prevents in-depth analysis. Crowdsourced data may offer promising alternatives and can be integrated with official data sources to increase the spatial coverage of in-situ measurements, provided that the accuracy-related concerns are effectively addressed. Given the above considerations, in this thesis, we exploit crowdsourced air temperature data from the Netatmo amateur weather network to get insights into the UHI effect in the Metropolitan City of Milan (Northern Italy). We propose an automatic cleaning workflow (published open access on GitHub) to pre-process air temperature measurements and obtain analysis-ready time series. Specifically, we use official data from the Regional Agency for Environmental Protection (ARPA) as a reference to clean Netatmo air temperature observations. Geostatistical analyses are performed on the pre-processed data to investigate summertime temperature extremes and identify hot spots and cold spots in the study area. Finally, correlation analysis is carried out with multi-temporal Local Climate Zone (LCZ) maps. The research results indicate that crowdsourced data can effectively add value to the UHI assessment, provided detailed data cleaning is performed. The proposed methodology entirely relies on free and open-source software, thus enabling easy replication beyond the case study in this work.File | Dimensione | Formato | |
---|---|---|---|
Exploring crowdsourced temperature data from Milan for urban heat island analysis.pdf
accessibile in internet per tutti
Descrizione: Exploring crowdsourced temperature data from Milan for urban heat island analysis
Dimensione
140.97 MB
Formato
Adobe PDF
|
140.97 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/223727