PDF 20.134kB - TOBIAS-lib - Universität Tübingen

78 S. Siedentop and S. Fina
A further shortcoming of the empirical sprawl debate is that many studies do not link the
addressed land use dimensions and their related indicators to specific sprawl impacts. This
failure leads to the misconception that the (alleged) impacts of sprawl can be measured with
uniform sprawl indicators. It also neglects the fact that different sprawl dimensions correspond
with different impacts. For example, ecological impacts such as meso-climatic or
hydrological changes are an outcome of landcover changes whereas economic impacts like
efficiency losses of public transport systems are caused by declining densities and the spatial
dispersion of urban land uses (Nuissl, Haase, Lanzendorf, and Wittmer 2009).
Furthermore, many measurement concepts are linked to a certain scale of observation or
spatial geography (e.g., metropolitan areas). The reason for this lies in the use of data sources
(e.g., land use data aggregated to the level of metropolitan areas), or in the scale for which the
analysis is carried out. The problem here is that indicators used in these concepts could be
restricted in their use for other scales. This applies to measures of centrality as the degree to
which urban development is close to a city centre (central business district) and other metrics
whose implementation requires the definition of a fixed spatial reference point.
3. An impact-oriented empirical concept of urban sprawl
3.1. General approach
Our measurement concept is intended to overcome some of the empirical limitations
described above. Firstly, we work with different indicators to indicate different impacts
caused by sprawl. Secondly, our approach can deal with the static and process nature of
urban sprawl. Therefore, we suggest operational indicators that characterise the conditions
of land use and use others to address land use changes over time. Thirdly, our indicators can
be used for all spatial units (administrative or non-administrative units such as river basins or
air pollution sheds) and various geographical scales above the neighbourhood level (city,
metropolitan, national).
Our conceptual framework takes into account that different dimensions of sprawl
correspond with environmental, social and economic impacts of urban land use change
(see Figure 1 and Table 2). At first, sprawl-type developments contribute to declining urban
densities (density dimension). As noted above, declining densities are an outcome of low
density development at the urban fringe and density losses within the urbanised area as an
effect of household dynamics and rising affluence. ‘Low density sprawl’ imposes pressure
on the economic efficiency of technical infrastructures and increases transportation demand
(Burchell et al. 1998; Schiller and Siedentop 2005). Low density living is further propelling
automobile dependence (Newman and Kenworthy 2006). Other studies present evidence
that people living in sprawling areas are likely to walk less and more likely suffer from
obesity and hereby linked chronic diseases than people who live in less sprawling areas
(Ewing, Pendall, and Chen 2002).
A second dimension of sprawl refers to the change of land use pattern (pattern dimension),
operationalised with geometric measures. According to this dimension, sprawl
describes the transition of a compact urban form to a dispersed urban land use pattern. A
typical feature of this sprawl dimension is an irregular, discontinuous urban form with a
highly fragmented mosaic of different land uses. ‘Pattern sprawl’ can typically be found in
suburban and exurban parts of metropolitan areas. Researchers claim that ‘pattern sprawl’ is
responsible for efficiency losses of urban services such as road infrastructure or sewer
systems (Burchell et al. 1998; Doubek and Zanetti 1999). There is also evidence that
spatially dispersed urban functions contribute to larger travel distances (Cervero 1996;