Detecting, Modeling and Predicting Vertical Urban Growth: An exploratory review
2019-09-17T13:44:32Z (GMT) by
Cities across the globe have been growing both horizontally and vertically to accommodate growing populations within limited land. Despite extensive research on urban sprawl, contemporary research on vertical urban development remains limited. This paper presents an exploratory review of the literature on vertical urban growth (VUG) based on academic papers published over the past three decades. Three thematic areas were identified, including approaches defining and extracting VUG, the identification of factors driving VUG, as well as models simulating and predicting VUG. Our results show that a diverse range of methods exist in defining and measuring VUG, ranging from the use of building heights or building floors to urban intensity, floor-to-area ratio, and urban functional complexity. While some research used building information data over time to define VUG, the most commonly used data derives from remote sense imagery, including Synthetic Aperture Radar imagery and Lidar data. Recent research uses Google Earth and OpenStreetMap as source data to extract building heights or number of building floors. On the other hand, only limited work is available in understanding and modeling the driving factors that lead to vertical urban growth. Cellular automata (CA) modeling, which has been commonly used to simulate horizontal urban expansion over the past four decades, are beginning to be used to simulate VUG and generate future growth scenarios. We argue that future research in this field should address the challenges in modeling vertical urban growth, particularly reliable data to parameterization, the configuration and validation of the 3D urban models, and the simulation of VUG under different scenarios and driving factors. We also call for more research to simulate the processes of urban demolition and shrinkage in the vertical dimensions, as well as the underground expansion in our urban space.