This Ph.D. thesis presents research that provides new insights into urban studies using novel multi-source big data and techniques with the scope to improve understanding of urban development. City is a complex system which provides residence, employment, transportation and other utility functions for human well-being. To facilitate the development of sustainable urban pathways, such as reduction of traffic-related air pollution, the focus of my thesis was on systematic analysis of the relationship between different components. Novel geospatial big data and deep learning techniques open new opportunities to systematically understand human mobility and achieve real-time monitoring and dispatching road resource in a closed loop system. Based on multi-source novel geospatial data and analytic approaches, the objectives of this thesis addressed the follow questions:


1.How can high spatial resolution functional zones be mapped with satellite remote sensing imagery?


2.How can population density at fine scale be mapped using satellite night-time light data?


3.What are the spatio-temporal patterns of traffic congestion and what is the most influential urban form factor, which is associated with traffic congestion?


4.What are the spatio-temporal patterns of traffic-related air pollutant emissions? What is the relationship between traffic-related air pollutant emissions and functional zones?


5.How does building density impact the land surface temperature? Is there any difference regarding land surface temperature among cities across different climatic zones?