Finding the secret formula: How can we quantitatively understand cities' growth, based on their street network structure?
Paper in proceedings, 2017
Cities are complex, perhaps one of the most complex kinds of structure created by humans. Some cities have been planned top-down to a large extent while others have grown organically on their own. The outcome of these planning and growth processes are the various morphological building and street patterns seen in cities. There are strong reasons to believe that the street system in a city has a crucial effect on land use and building development (Hillier, 1996). It is therefore essential to understand the complex street networks in our cities. This is not only to categorize cities but also to be able to switch-over our transport system towards a higher degree of sustainability. Despite extensive research in the fields of urban planning and urban history, there are still very few consistent ways of quantitatively describing and classifying cities. The perspectives used in research so far have mainly categorized them based on visual judgements of their morphology (Kostof & Tobias, 1991). It is problematic that classifications are mainly based on subjective judgement and lack a quantitative measure. To understand cities and their characteristics, it is necessary to find methods and measures that can describe these characteristics and also be suitable for comparisons between cities. By using space syntax methods, it is possible to find measures and properties of the cities' street networks, which sometimes seems to exhibit patterns with scale free properties (Jiang, 2007). Thanks to recent progress in the field of complexity studies, it is now also possible to test if and to which degree a network has scale free properties. One commonly used approach is to test whether the distribution of a certain property in a collection of elements fit a power law distribution(Clauset et al., 2009). The results of this study show that the degree distribution for a city's street network seems to fit a power law for cities grown organically. On the other hand, cities that are planned to a large extent do not have that good fit. This result seems sound, since a distribution that fits a power law is a signature of a multiplicative growth processes. Another interesting finding following this result, is that this way of quantitatively classifying cities seems to correlate well with earlier attempts of qualitative morphological classification.