Parts | Days | Selection | Search | Updates | Downloads | Help
SOE: Fachverband Physik sozio-ökonomischer Systeme
SOE 15: Focus Session: Cities as complex systems
SOE 15.3: Talk
Wednesday, March 22, 2017, 15:30–15:45, GÖR 226
Zipf's Law for Australia: An outlier? — Somwrita Sarkar1 and •Peter Alexander Robinson2 — 1Faculty of Architecture, Design and Planning, The University of Sydney, Australia — 2School of Physics, The University of Sydney, Australia
Various studies of city size distributions across the planet have shown what is now known as Zipf's law, which says that the distribution of city sizes fits a power law: the number of cities with populations greater than N is roughly proportional to 1/N. We explore this relationship and its historical evolution for city size distributions in Australia (from 1922 to 2015), and find that Zipf's Law does not hold for Australia. Rather, a different organization emerges with (a) the Zipf exponent close to 0.7, suggesting a 'flatter' distribution with the biggest cities having larger sizes, and (b) a 'missing middle' with no middle-sized cities (of approximately 500,000 to 1 million). Curiously, when the five largest cities (Sydney, Melbourne, Brisbane, Perth and Adelaide) are removed and the analysis re-performed for the smaller cities, the exponent is close to unity, maintaining itself as stable from 1922 to 2015. These empirical findings will be presented and some preliminary thoughts on this outlier behavior will be discussed. Two hypotheses will be considered: (a) the possibility that smaller (regional) and larger cities in Australia are following different growth processes, and (b) the possibility that each state in the country could be behaving as a 'country'. Relationships of these hypotheses with Australia's colonial history and relatively young urban settlement will be discussed. Implications for urban and regional development will also be discussed.