City-Level Decoupling-Full Report

living are less than similar standards of living across all income categories in rural areas, but urban material and energy use will vary with the context, based on such factors as urban form, especially density; the efficiency of the built structures and related 'green building' standards and regulations; advanced industrialised versus newly industrialised; high or low per capita income; the effect of the urbanisation rate on the likelihood of infrastructure backlogs and related infrastructure investment levels (e.g. India/China vs North America/Europe); income and/or spatial inequalities (e.g. more egalitarian European cities tend to be more dependent on public transport than North and South American cities that tend to be highly dependent on the private car); effectiveness of urban planning (limited in African cities versus high in European, US and Chinese cities); and geographical and topographical differences that will directly affect the requirements for infrastructures such as storm water drainage, energy for heating/cooling, what densities are possible (e.g. steep inclines are generally not useful for high rise buildings). Urbanisation per se does not seem to result in increased material and energy use per capita. Based on direct flows only, material and energy use per capita in cities is usually lower than rural areas for the same level of income. Rising household income and therefore levels of resource consumption per capita increase material and energy use, so when urbanisation is correlated with rising GDP per capita, then upwardly mobile households will tend to concentrate in cities. At the same time, the capacity for innovations in cities could encourage the decoupling of rising income (up to a point) from rates of resource use. 4.3 Comparative material flow analysis of cities The Building Technology Programme at Massachusetts Institute of Technology (MIT) has been applying urban MFA to a variety of architectural and urban planning challenges (see http://bt.mit.edu/). One of their recent

Whereas the DMI per capita for Paris was 8.8 tons in 2003, it was 11 tons for Paris and its suburbs and 12.3 t for Paris plus the region. DMC (corrected) per capita was 5 t, 4.6 t and 7.1 t for each region respectively. Thus more than half the DMI from all three regions are returned to local natural systems. Also, the recycling rate of 0.7-0.9 t per capita is a fifth of DMC (corrected) which sets an interesting benchmark for comparative studies. This kind of detailed quantification of urban material flows makes it possible to define what decoupling could mean for a particular city in quite specific terms. To retrofit or design new urban infrastructures so as to achieve the goals of decoupling for resource efficiency or resource productivity, the focus will need to be on the 'converter', 'demand' and 'reconverter' technologies (discussed in later chapters). However, wider spatial and economic factors play roles that are unrelated to these specific technologies. The most important will be the stage of development. During the early stages of development, relative decoupling with a focus on materials will be the strategic focus, but later on as NAS becomes less important the focus will shift to rising energy demands as income levels rise. The unique configuration of cities can give rise to very different levels of DMC/cap even with similarities at national level evidence. 62 For example, DMC/cap was 20.8 t per year for Lisbon, 18 for Singapore, 7.6 for Geneva, 5 for Paris, 3.6 for London, and 3.3 for Cape Town (these figures are for direct flows and ignore indirect flows). The energy requirements for mobility are determined primarily by population density. Where high densities are correlated with good public transit systems and deterrents to private car use, energy requirements for mobility can be lowered dramatically. Similarly, the operational energy requirements of buildings (which far exceed the energy embodied in the constructed building) could be reduced by as much as 80% by changing the way they are designed and operated. Although there may be some rebound effects (if there are no recessionary conditions), in general the energy requirements for urban

38