Geography Reference
In-Depth Information
purposes. For example, we might begin by
dividing land use into 'rural' versus 'urban' on the
basis of the results of a standard maximum
likelihood classification of a satellite image; a
narrower definition of urban land use might be as
urban land that is 'built-up', based on
supplementary information from the census or
other socio-economic sources. Such sources might
also be used to discriminate between residential
and other (commercial, industrial) built-up land
uses, and/or information might be incorporated
about the numbers of properties in the area. The
resultant classified digital images might then be
used to create a range of graphical and numerical
indices of urban morphology.
One such graphical tool that has been widely
used in applied urban analysis is the 'density
gradient', which measures the rate of change in
the proportion of land that is filled with 'urban'
land uses with increased distance from the historic
centre of the settlement. This is measured as (N(r)/
(pr
2
)), that is, the number of occupied pixels in
each distance band
r
divided by the annular area
of that distance band. Longley and Mesev (1997)
use this kind of approach to identify the density
gradients of four different facets to urban form of
Norwich, UK, illustrated in Figure 44.6 —
'urban', 'built-up', 'residential' and 'numbers of
properties'.
As this figure shows, the differences between
these apparently similar categories are more than
semantic and can heavily condition whether and
to what extent we might consider that the profile
is characterised by discontinuities and 'density
craters'. However, the optimistic message
contained in this figure is that, once the differences
between different conceptions of 'urban-ness'
have been clearly grasped, it is possible to develop
a range of customised indicators of urban
morphology. These GIS-based data models are
informing our thinking about the ways in which
urban settlements fill space, as well as providing
detailed information as to the morphology of
particular settlement structures.
CONCLUSION
The shape, form and dimension of cities are today
changing faster than ever before, as processes of retail
decentralisation, demographic change and ongoing
industrial restructuring exact changes in urban
morphology of unprecedented scale.Yet at the same
time our ability to measure, model and simulate
change is developing too, thanks to the GIS and
digital data revolutions, which now allow us to
invoke a full range of modelling assumptions and
simplifications at different scales. This is important
because in all the best classical theories, scale plays an
important role and empirical modelling requires us
to generalise between scales. The resulting 'real-
world' indicators of urban morphology will rarely
exhibit the clear geometry and orderliness of high
school geography models: however, just as GIS makes
it possible to depict real-world spatial distributions
with greater precision than ever before, so
developments in spatial analysis are allowing
traditional models of urban structure to be recast in
new and exciting ways (see Box 44.2).
The early models of urban structure suggested
that cities were clearly organised, simply ordered,
and thus predictable. It followed that they could
be planned in such a way that the quality of life of
their residents could be directly improved by
manipulating their physical form. During the last
twenty years, many geographers have been
overwhelmed by the complexity that characterises
Box 44.2
Modelling in a GIS environment
SUMMARY
•
Models are simplifications of reality that can be used
to create scenarios.
•
The ways in which we decide to simplify reality (e.g.
what
is
urban land?) conditions the way in which we
measure geographical phenomena. This in turn
conditions the way we analyse them and (depending
on the particular application) the likely outcome of
analysis
•
Problems of the spatial scale of data and the level of
geographical aggregation pose particular problems
for geographical modelling (see Openshaw 1984)
•
GIS is
a
useful applied technology for creating digital
models that can be used to explore the world around
us (see Longley
et al
. 1998).
