Are you an EPFL student looking for a semester project?
Work with us on data science and visualisation projects, and deploy your project as an app on top of Graph Search.
Surface temperature is one of the critical factors used to study microclimate conditions through Land surface Temperature (LST), a widely used data source. This paper tests a classification approach using moderate spatial satellite resolution images to extract surface properties of Local Climate Zones (LCZs) in a semi-arid region against measured LSTs. However, LST is used to study and explore the spatial relationship between temperature and urban morphology. Therefore, it is considered an alternative and reliable data source to complement in-situ measurement, especially in developing countries. The used methodology is based on two statistical approaches: 1) auto-spatial correlation using Getis Ord Gi* statistic, and 2) ANOVA test. Results showed an inversion effect of surface urban heat island - SUHI, where a strong correlation between specific LCZs and the surface temperature was found. The urban classes represented by open-low-rise, compact-low-rise have shown a significant decrease in surface temperature for almost two decades and in both summer and winter seasons compared to non-urban ones. Therefore, the outcomes of this study may have critical implications for urban planners who seek to mitigate SUHI effects in arid and semi-arid urban areas. Also, the approach used in this study can be helpful to follow the urban environment's spatial and temporal development as an application framework of sustainable development.
Vincent Kaufmann, Luca Giovanni Pattaroni, Marc-Edouard Baptiste Grégoire Schultheiss
Andrea Baccarini, Imad El Haddad, Lubna Dada, Houssni Lamkaddam