Interaction Between Urban Densification and Green Spaces Using Multi-Temporal Sentinel Imagery - A Case Study of Munich

Abstract

Green spaces have been increasingly essential for improving the quality of urban environments and enhancing city-dwellers' physical and mental health. However, cities are getting denser due to the rapidly expanding amount of people living in cities, leading to a challenge of urban green space (UGS) provision and proportion. Therefore, this thesis's main objective is to detect the temporal dynamics of urban density levels and their impacts on the quantities and proportion of UGS in Munich, Germany, for the past six years (2015 – 2020). Two multispectral Sentinel-2 remotely sensed images in August 2015 and 2020 were chosen to classify the impervious surfaces (IS) and UGS by applying the maximum likelihood classification method. Besides, four density levels were defined in the study – green spaces (GS), low density (LD), medium density (MD), and high density (HD). To discover the dynamics of the urban density degrees, the NDVI (Normalized Difference Vegetation Index) change detection approach was conducted based on six TimeScan-Sentinel-2 products developed by DLR (German Aerospace Center). The results have shown that the supply of UGS only changed in a small proportion in Munich for recent years under the increasing pressure of urban densification. Although the city was getting denser with the GS levels declined from 48% (2015) to 29% (2020) and the HD levels increased from 8% (2015) to as doubled as 16% (2020), the UGS still occupied the dominant portion in Munich, with 46.5% in 2015 and 47% in 2020. The conditions of green spaces have remained stable thanks to the efforts of the Munich urban planning office. This paper provided detailed findings and insights on the impacts of urban densification on UGS in Munich, and could assist policymakers and city planners in conducting more sustainable urban management for densely built-up cities.