Research Project

CAREER: Understanding the Impact of Space and Time on Household Urban Energy and Resource Consumption

Principal Investigator

Derrible, Sybil

Start Date

2016-08-15

End Date

2021-07-31

Funding Source

National Science Foundation

Abstract

Ultimately, the amount of energy and resources that Americans consume depends heavily on where as well as how they live. Thus residents of northern states tend to consume more energy at home than residents of southern states, because heating requires more energy than cooling. Similarly, people living in suburbs tend to consume more energy for transportation than people living in city centers because longer commutes consume more energy. Similar observations can be made for all types of energy and resources, from electricity and transportation to natural gas, water, and even food. This Faculty Early Career Development (CAREER) grant will help to discover the fundamental principles that govern how much "where" and "how" people live matters for energy and resource consumption. The contributions from this research will directly assist in the development of effective policies for more sustainable communities that consume less energy and resources. Moreover, a better understanding of energy flows in cities will provide planners and engineers with information that will enable them to design smarter and more resilient infrastructure systems that are decentralized and distributed. In addition, this grant will have a significant impact for the broader public as it includes the development of a smartphone application to enable anyone to calculate their daily carbon footprint and track their performance over time. This research places itself at the nexus of urban metabolism and complexity theory. Urban metabolism is the study of flows of material and energy in cities. The main hypothesis of this research is that urban metabolism follows distinct mathematical laws at the community scale that can be captured using elements of complexity theory. Various mathematical laws (e.g., power law, lognormal distribution, uniform distribution) will be applied using agent-based modeling techniques to generate a theoretical space that will include every possible community profile in terms of energy and resource consumption. These laws will then be tested for individual communities using freely available data from municipal open data portals. Additionally, this research will utilize elements of machine learning to classify communities based on their energy and resource consumption patterns, and network science to better understand the inter- and co-dependence between the usage of electricity, water, natural gas, and transportation infrastructure. The research will therefore offer a significant contribution to help design a society that is smarter and more resilient while being more sustainable by requiring less energy and fewer resources to thrive.