Rethinking progress in urban analysis and modeling: models, metaphors, and meaning - Urban Remote Sensing: Monitoring, Synthesis and Modeling in the Urban Environment - page 358

Modeling cities as an organism has also been greatly facili-
tated by the biologically inspired computing paradigm, ranging
from neural computing and genetic algorithms to evolutionary
programming. Due to sharing common data structure, cellular
automata (CA) and agent-based (AB) models have increasingly
stepped out of the artificial world (Besussi and Cecchini, 1996),
and been integrated with GIS (Takeyama and Couclelis, 1997;
Liu, 2008) as an integral part of geosimulation tools (Benenson
and Torrens, 2004). Viewed from a larger context, CA and AB
models are increasingly becoming part of the emerging generative
social science (Epstein, 2006) aiming to simulate diverse social
processes across space.
In many ways similar to the dominance of Lowry models in
the social physics tradition, the social biology tradition has been
dominantly by models developed from cellular automata (CA)
and agent-based models (ABM) - what Portugali (1999) called
FACS (free agent in a cellular space) model. As demonstrated in
some chapters in this topic, there have been major methodological
accomplishments during the past 15 years. The driving metaphor
behind this new modeling paradigm using cellular automata and
agent-based modeling is more biologically than mechanically
motivated, by conceiving cities as growing cells 3 .
Different from the Lowry and Forester types of urban models,
CA and AB models conceive cities as complex systems involving
a large but finite number of intelligent and adaptive agents.
The behaviors of these agents are contingent on the availability
of information and subject to modification of their rules of
action based upon new flows of information. This continual
and dynamic change of agents behavior makes prediction and
measurement using old rules of science impossible. CA and AB-
based urban models have been characterized by new concepts
and theories based on non-linear dynamics. After about a decade
of theoretical reflection (Couclelis, 1985, 1988, 1997), CA and AB
models have been applied to simulate complex real world issues
(Clarke and Gaydos, 1998; Sui and Zeng, 2001; Waddell, 2002;
Yang and Lo, 2003; Brown et al ., 2004; Torrens, 2006; Liu and
Seto, 2008).
Just like the physical/mechanical metaphors, conceiving cities
as organism also has unintended consequences. For example,
the idea of life cycles is embedded in all organism metaphors.
According to Roberts (1991), the uncritical use of a life cyclical
approach allowed the idea of city to slip from being an image to
being a cause in accounts of urban decline, and has served a pre-
dominantly conservative urban policy. The use of the biological
metaphor in general and the life cycle metaphor in particular,
is profoundly ideological as it espouses inevitability, fatalism,
determinism, and inexorability (Furbey, 1999). According to
the policies motivated by biological metaphors, any government
intervention has been criticized as tinkering with some natural
order of things, which has led to policies of planning shrinkage,
and managing decline and dispersion. A deterministic, biologi-
cally motivated urban triage often directly attacks empowering
views of polity and place.
Nonetheless, CA and ABM-based simulations are emerging
as part of the generative social science, which is changing the new
frontier of science (Casti, 1997; Wolfram, 2002). For instance,
Batty (2009) argued that we can develop a digital breeder to
simulate urban growth according to different scenarios. Spiller
(2009) even suggested that:

new architectural flora and fauna. We might be able to make
truly sustainable and green materials whose biodegradability
is simply a natural side effect of these technologies (p. 131).
Remote sensing technologies are going to play increasingly
important roles in the urban modeling following the social
biology tradition. Instead of simply being tools to study
cities from the sky (Campanella, 2001), remote sensing now
is regarded as ''the X-ray crystallography'' for urban ''DNA''
(Wilson, 2009)

.

26.3.4 Cities as arenas - the spatial
event tradition

The spatial event tradition is relatively new, perhaps the least
developed and least noticed among the four traditions, but
perhaps has the greatest potential and momentum for new
growth. This new urban modeling tradition conceptualizes cities
as spatial events (Batty, 2002). Conceptually, this tradition is
related to Pepper's contextualism world hypothesis. Consistent
with Hagerstrand's time geography, the spatial event tradition is
closely allied with the development of people-based GIS (Miller,
2007) and the concept of a real-time city (Foth, 2008).
Attempting to move away from any pre-conceived notion of
how cities work, this tradition aims to understand how individual
events occur spontaneously within a city. According to Batty
(2002), this tradition grew out of a need for a more temporal
emphasis in our theories and models. For quite some time, we
have placed too much emphasis on finding equilibrium instead
of on understanding the dynamics of urban change despite
Wilson's (1981) early attempt to do so from the perspective of
catastrophe and bifurcation. By conceiving cities as arenas for
spontaneous events, we can better understand rhythms of urban
life that take place in both time and space. Nowadays we have
the technologies to analyze events by their location, duration,
intensity, and sometimes volatility. More than in any other
tradition, humans are becoming the censors (Goodchild, 2007) as
a part of the participatory sensing (http://research.cens.ucla.edu),
and urban modeling has become closer to story-telling and
narrative development (Guhathakurta, 2002).
The metaphor of ''cities as arena'' has meanings not only
in a physical sense, but also increasingly in a virtual sense. The
growth of web 2.0 and related user-generated content in the
form of spatially and temporally tagged information may greatly
facilitate the study of cities as spatial events (Hudson-Smith et al .,
2008, 2009). For example, Girardin, Vaccari, and Gerber (2009)
used geotagged photos tourists voluntarily posted on flickr.com
as the primary data source to reconstruct tourist flows. Similarly,
Dykes et al . (2008) used volunteered geographic information to
describe the diversity of places.
The emerging participatory sensing has enabled citizens
to record and transit data about the sourrounding environ-
ment using means like their cell phones. High school students
can develop a mashup in Google Earth using realistic three-
dimensional imageries for showing potential impacts of global
climate change in their hometown. The US Centers for Disease
Control and Prevention (CDC) has been tracking the diffusion
of H1N1 flu across the United States by closely monitoring
tweets on Twitter. Urban planners have also started using Second
Life, simulating potential planning scenarios and policy impacts

we can use the natural imperatives of plants and maybe ani-
mal cells as a means to 'rewire' them to create huge rafts of