In recent years, urban environments have been subject to increasing pressure generated by a raised demand for space, resulting in the degradation of crucial environmental services. Consequently, there is a pressing need to explore the spatiotemporal dynamics of Land Use/Land Cover (LULC) changes. The field of geomatics, with its interdisciplinary nature encompassing remote sensing, geographic information systems (GIS), and spatial analysis, plays a pivotal role in addressing the pressing challenges of monitoring and assessing land use and land cover dynamics. This research introduces an innovative geospatial methodology that employs advanced remote sensing techniques and GIS to identify, measure, and visualize the patterns of urban transformations. Focusing on a case study of 15 municipalities in southern Italy, the authors utilized Landsat multispectral satellite imagery obtained from the Google Earth Engine platform, covering sixteen years. The images were processed and classified using an unsupervised K-means machine learning algorithm, which segmented the imagery into clusters based on pixel similarity, delineating built-up changes from 2005 to 2021. The accuracy of this classification was rigorously measured by comparing the model predictions with the ESA Land Cover dataset, employing Precision, Recall, and Fl-Score as the evaluation metrics. Specifically, for 2021, the Fl-Score for built-up areas was recorded at 0.64, and for vegetation cover, a robust 0.91. This methodology successfully measured a land consumption of approximately 167 hectares between 2005 and 2021, offering a robust decision-support framework for urban planning and environmental conservation. This approach demonstrates significant potential in aiding policymakers and urban planners in making informed decisions to mitigate the adverse effects of urban growth on the environment.

A Novel Geospatial Methodology for Measuring and Mapping Spatiotemporal Built-Up Dynamics Based on Google Earth Engine and Unsupervised K-Means Clustering of Multispectral Satellite Imagery

Vitale, Alessandro
;
Salvo, Carolina;Lamonaca, Francesco
2024-01-01

Abstract

In recent years, urban environments have been subject to increasing pressure generated by a raised demand for space, resulting in the degradation of crucial environmental services. Consequently, there is a pressing need to explore the spatiotemporal dynamics of Land Use/Land Cover (LULC) changes. The field of geomatics, with its interdisciplinary nature encompassing remote sensing, geographic information systems (GIS), and spatial analysis, plays a pivotal role in addressing the pressing challenges of monitoring and assessing land use and land cover dynamics. This research introduces an innovative geospatial methodology that employs advanced remote sensing techniques and GIS to identify, measure, and visualize the patterns of urban transformations. Focusing on a case study of 15 municipalities in southern Italy, the authors utilized Landsat multispectral satellite imagery obtained from the Google Earth Engine platform, covering sixteen years. The images were processed and classified using an unsupervised K-means machine learning algorithm, which segmented the imagery into clusters based on pixel similarity, delineating built-up changes from 2005 to 2021. The accuracy of this classification was rigorously measured by comparing the model predictions with the ESA Land Cover dataset, employing Precision, Recall, and Fl-Score as the evaluation metrics. Specifically, for 2021, the Fl-Score for built-up areas was recorded at 0.64, and for vegetation cover, a robust 0.91. This methodology successfully measured a land consumption of approximately 167 hectares between 2005 and 2021, offering a robust decision-support framework for urban planning and environmental conservation. This approach demonstrates significant potential in aiding policymakers and urban planners in making informed decisions to mitigate the adverse effects of urban growth on the environment.
2024
979-8-3503-8501-4
Built-up Dynamics
GeoAI
Google Earth Engine
Multispectral Satellite Imagery
Remote Sensing
File in questo prodotto:
Non ci sono file associati a questo prodotto.

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/20.500.11770/373418
 Attenzione

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

Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? ND
social impact