Remote sensing modeling of urban density dynamics across 36 major cities in China: Fresh insights from hierarchical urbanized space

Abstract

Rapid urbanization is considered to be of high relevance for global/local environment change, but its multi-dimensional change features over time have rarely been reported, especially for the process, states and consequences of urban density growth at hierarchical scales of urbanized area. To fill this gap, here, we firstly integrated 27-years’ worth of consistent nighttime light (NTL) time series with land cover dataset, to reconstruct 4 basic, 13 archetypes of temporal trajectories of urban density dynamics with an objected-oriented analysis, and measured their spatio-temporal patterns over hierarchical urbanized space that spanned from landscape patches to 36 major cities in China. Our results showed that, from the micro patch dynamics, the logistic pattern is striking among 4 models over most of time periods, accounting for ~45.63% of the total patch amounts and ~43.73% of the total patch area, respectively, while the exponential pattern contributes the least. Moreover, the structural transition of thematic features revealed their spatial gradient delay from established patches to newly developed patches, from infilling patches to outlying patches. In terms of the space of cities, seven typical clusters of urbanization evolution were identified, however, under different policies, reform practice and developmental history. Additional three alternative states of urban density were also organized for estimated future pathways. Therefore, according to our new insights into inter- and intra-city analysis, both the differentiated strategies for horizontal cities and trade-offs of spatial configuration for urban interior areas were informed to formulate effective science-policy interface.