Urban Spatiotemporal Energy Flux

Urban energy systems are often studied in a very similar way in the sense that the characteristics of the underlying physical infrastructure are weighted as the main determinants of energy use predictions, while the behavior of the human population in relation to this systemthe so-called ``energy consumers''in time and urban spaces is effectively neglected. The spatial and temporal variations in infrastructure-population interactivity greatly complicate urban energy systems; the unremitting growth in population and advances in technology mean that the dynamic interrelationship between the population and urban environment will continue to grow exponentially, resulting in increasing uncertainties, unreliable predictions and poor management decisions given the inadequacy of existing approaches. In this dissertation, I explore the interdependencies of spatiotemporal fluctuations of human mobility as an indicator for human activities and energy use in urban areas in three main studies. First, I show that the fluctuations of intra-urban human mobility and energy use have an underlying structure across both time and space, and that human mobility can indeed be used as a predictor for energy use in both dimensions. Second, I examine how one of the dominant drivers of this structure, namely individuals' location-based activities, influence patterns in energy supply and demand across building types (i.e. residential and commercial buildings) and show how variations in the human mobility networks of two distinct urban populations (the so-called returners and explorers) can explain fluctuations in energy use. Third, I introduce an integrated approach for predicting urban energy use across time and space by incorporating these interdependencies. Generating predictive models that capture the spatiotemporal variations in these determinants in urban settings, as suggested in this research, will contribute to our understanding of how variations in urban population activities for particular times and locations influence can be applied to estimate energy use patterns in surrounding areas.