Geoscience Reference
In-Depth Information
multiplication of works, often with an operational aim. As a matter of fact, this user-
friendly tool enables the user to test the impact of a new planning, transportation
infrastructure or a new urbanized area, and perform prospective. Thus, it is a tool
that helps the user planify (see Engelen and Uljee for the case of the Netherlands
[ENG 97], Lajoie
et al.
, [LAJ 07] for a prospective approach applied to Réunion, for
example).
Stanilov and Batty [STA 10] proposed a theoretical framing for this family of
models of land use dynamics by CA. They distinguished two scales involved in
change:
- the local scale at which the attraction and repulsion forces operate between the
different land-uses. Indeed, land uses such as residential occupations, recreational
areas and commercial activities tend to attract each other. Conversely, industrial
activities are repulsive for residential land use;
-the regional scale corresponding to the global physical framework,
particularly in terms of transport infrastructure and organization of the network of
activity centers. It determines the accessibilities to the key elements of the urban
space.
The authors propose an analogy between the genetic code of individuals and the
generic rules that govern the transformations of land use. This “genetic code” of the
city would determine the potential urban development in the long-term. Therefore,
these rules operate at all times, regardless of the economic climate. The latter has an
influence, but does not come first in the explanation of change. Mechanisms such as
the distance to the Central Business District (CBD) and train stations have enduring
influence and play the role of the “genetic code”. The economic conditions affect
only afterward, in a secondary manner, creating the context in which the generic
rules operate.
Stanilov and Batty apply their CA model, using the platform Metronomica, to
the evolution of land use in West London over a period of 130 years. The years
1875, 1895 and 1915 are used for the calibration and then forecasts are made and
compared with the states observed for the years 1935, 1960, 1985 and 2005. The
authors stress upon the overall good quality of model predictions for these dates.
They conclude that a small number of variables are thus sufficient to give account,
in general terms, of the spatial organization of land use transformations on the long
term. The model has notably reproduced the transitions from a nucleated type of
growth (with separate urban cores), to a diffuse type, and finally to an infill type
over a century. The ability of the model to reproduce a realistic change over such a
long period corroborated, according to the authors, the hypothesis that the rules
defined at the local scale, which govern the changes of land use, are consistent
through time and independent of the socio-economic context.
