Fig. 1 Causal loop diagram of the drivers of residential building construction. Cycles labeled by “R” are positive (reinforcing) feedback loops; “B” indicates negative (balancing) feedback loops.


Metabolism is a dynamic phenomenon. The word itself is based on the Greek word μεταβολη, meaning “change,” and the biological process it originally describes is a set of biochemical reactions and transformations of nutrients and energy. Urban metabolism is often reduced to a snapshot of the material stocks and flows in a city over a given time period, but this simplification tends to mask the richness and complexity of metabolic phenomena. In this project, we focus on developing an understanding of the dynamics of urban metabolism as a way to illuminate the functional relationships between physical material flows and socioeconomic behaviors of the city. In so doing, urban metabolism becomes more truthful and useful in efforts to compare global cities and also more salient to urban sustainability policymaking.


The Dynamics of Urban Metabolism project is rooted in the use of System Dynamics (SD) modeling to represent and clarify the stocks, flows of resource throughput and the feedbacks, delays, and interrelationships that govern the behavior of urban metabolic systems. A highly modular and multi-scale modeling framework, SD can be used to communicate basic dynamic principles or test hypotheses of the potential responses of the system to certain detailed policy stimuli, and everything in between. A long-term goal of the project is the development of a generalized dynamic model of urban metabolism that can be adapted for any global city. In the meantime, ongoing research focuses on exploring the metabolism of urban sub-systems, such as residential housing (Davis 2012) or the water system (Noiva Welling 2011), and characterizing specific relationships between resource throughput and urban economic growth like income elasticity of consumption.

Singapore: an extraordinary case

Although this project first arose from efforts to characterize material flows associated with the destruction and rebuilding of New Orleans following Hurricane Katrina (Quinn 2008), the bulk of the research has been concerned with the island city-state of Singapore. Singapore is uniquely well-suited for both the detailed quantification of urban material flows and the study of deeper urban metabolic dynamics. As a resource-poor city, all resources must be imported, and as a city-state, all movement of goods and materials in and out of the city are logged as international trade in the UN COMTRADE database. In addition, the city has undergone relatively recent and extremely rapid urbanization. As late as the early 1960s, Singapore remained a relatively small and underdeveloped equatorial port city; just fifty years later, it is considered one of the world’s great modern cities. Combining data availability, clear urban boundaries, and rapid urbanization, Singapore is an ideal case to develop new theory of the dynamics of urban metabolism. It has yet to be determined if these same unique characteristics strengthen or hinder generalizability to other global cities.

Fig. 2 Income elasticities of urban metabolic consumption in Singapore, 1962-2012. Eight metabolic flows are plotted against real (corrected for inflation) GDP on a log-log scale. Growth curves are modeled using a power law, with the exponent representing the elasticity of consumption. Income elasticity is the percent change in consumption as a result of a percent change in Singapore’s GDP.