Logistic Models of Fractal Dimension Growth for Spatio-Temporal Dynamics of Urban Morphology

Yanguang Chen

DOI: https://doi.org/10.48550/arXiv.1606.03538

2016-06-11

Physics and Society

Abstract:Urban form and growth can be described with fractal dimension, which is a measurement of space filling of urban evolution. Based on empirical analyses, a discovery is made that the time series of fractal dimension of urban form can be treated as a sigmoid function of time. Among various sigmoid functions, the logistic function is the most probable selection. However, how to use the model of fractal dimension growth to explain and predict urban growth is a pending problem remaining to be solved. This paper is devoted to modeling fractal dimension evolution of different types of cities. A interesting discovery is as follows: for the cities in developed countries such as UK, USA and Israel, the comparable fractal dimension values of a city's morphology in different years can be fitted to the logistic function; while for the cities in developing countries such as China, the fractal dimension data of urban form can be fitted to a quadratic logistic function. A generalized logistic function is thus proposed to model fractal dimension growth of urban form. The logistic model can also be employed to characterize the change of the dimensions of multifractals. The generalized logistic models can be used to predict the missing values of fractal dimension, to estimate the growth rate of the dimension, and thus to reveal the spatio-temporal process and pattern of a city's growth. Especially, the models lay a foundation for researching the correlation between urban form and urbanization and for developing the theory of spatial replacement dynamics.