This paper investigates the universality and robustness of scaling laws for
urban systems, according to the work by Bettencourt, Lobo and West among
others, using England and Wales as a case study. Initial results employing the
demarcations for cities from the European Statistical Commission digress from
the expected patterns. We therefore develop a method for producing multiple
city definitions based on both morphological and functional characteristics,
determined by population density and commuting to work journeys. For each of
these realisations of cities, we construct urban attributes by aggregating high
resolution census data. The approach produces a set of more than twenty
thousand possible definitions of urban systems for England and Wales. We use
these as a laboratory to explore the behaviour of the scaling exponent for each
configuration. The analysis of a large set of urban indicators for the full
range of system realisations shows that the scaling exponent is notably
sensitive to boundary change, particularly for indicators that have a nonlinear
relationship with population size. These findings highlight the crucial role of
system description when attempting to identify patterns of behaviour across
cities, and the need for consistency in defining boundaries if a theory of
cities is to be devised.
%0 Journal Article
%1 Arcaute2014Constructing
%A Arcaute, E.
%A Hatna, E.
%A Ferguson, P.
%A Youn, H.
%A Johansson, A.
%A Batty, M.
%D 2014
%J Journal of The Royal Society Interface
%K cities, universality scaling
%N 102
%P 20140745
%R 10.1098/rsif.2014.0745
%T Constructing cities, deconstructing scaling laws
%U http://dx.doi.org/10.1098/rsif.2014.0745
%V 12
%X This paper investigates the universality and robustness of scaling laws for
urban systems, according to the work by Bettencourt, Lobo and West among
others, using England and Wales as a case study. Initial results employing the
demarcations for cities from the European Statistical Commission digress from
the expected patterns. We therefore develop a method for producing multiple
city definitions based on both morphological and functional characteristics,
determined by population density and commuting to work journeys. For each of
these realisations of cities, we construct urban attributes by aggregating high
resolution census data. The approach produces a set of more than twenty
thousand possible definitions of urban systems for England and Wales. We use
these as a laboratory to explore the behaviour of the scaling exponent for each
configuration. The analysis of a large set of urban indicators for the full
range of system realisations shows that the scaling exponent is notably
sensitive to boundary change, particularly for indicators that have a nonlinear
relationship with population size. These findings highlight the crucial role of
system description when attempting to identify patterns of behaviour across
cities, and the need for consistency in defining boundaries if a theory of
cities is to be devised.
@article{Arcaute2014Constructing,
abstract = {{This paper investigates the universality and robustness of scaling laws for
urban systems, according to the work by Bettencourt, Lobo and West among
others, using England and Wales as a case study. Initial results employing the
demarcations for cities from the European Statistical Commission digress from
the expected patterns. We therefore develop a method for producing multiple
city definitions based on both morphological and functional characteristics,
determined by population density and commuting to work journeys. For each of
these realisations of cities, we construct urban attributes by aggregating high
resolution census data. The approach produces a set of more than twenty
thousand possible definitions of urban systems for England and Wales. We use
these as a laboratory to explore the behaviour of the scaling exponent for each
configuration. The analysis of a large set of urban indicators for the full
range of system realisations shows that the scaling exponent is notably
sensitive to boundary change, particularly for indicators that have a nonlinear
relationship with population size. These findings highlight the crucial role of
system description when attempting to identify patterns of behaviour across
cities, and the need for consistency in defining boundaries if a theory of
cities is to be devised.}},
added-at = {2019-06-10T14:53:09.000+0200},
archiveprefix = {arXiv},
author = {Arcaute, E. and Hatna, E. and Ferguson, P. and Youn, H. and Johansson, A. and Batty, M.},
biburl = {https://www.bibsonomy.org/bibtex/2eb17632c0249ab25b769188152aaf8bb/nonancourt},
citeulike-article-id = {11870794},
citeulike-linkout-0 = {http://dx.doi.org/10.1098/rsif.2014.0745},
citeulike-linkout-1 = {http://arxiv.org/abs/1301.1674},
citeulike-linkout-2 = {http://arxiv.org/pdf/1301.1674},
day = 19,
doi = {10.1098/rsif.2014.0745},
eprint = {1301.1674},
interhash = {b6906f3df35c50caaceb2666c5da8fb7},
intrahash = {eb17632c0249ab25b769188152aaf8bb},
issn = {1742-5662},
journal = {Journal of The Royal Society Interface},
keywords = {cities, universality scaling},
month = nov,
number = 102,
pages = 20140745,
posted-at = {2013-09-19 10:59:54},
priority = {2},
timestamp = {2019-08-01T15:36:58.000+0200},
title = {{Constructing cities, deconstructing scaling laws}},
url = {http://dx.doi.org/10.1098/rsif.2014.0745},
volume = 12,
year = 2014
}