打印

作者

Longxu Yan;De Wang;Shangwu Zhang;Carlo Ratti

来源

CITIES,Vol.113,Issue1,Article 103156

语言

英文

关键字

Urban centers;Big data;Non-local function;Localization;Shanghai

作者单位

College of Architecture and Urban Planning, Tongji University, China;Singapore-MIT Alliance for Research and Technology, Singapore;Senseable City Lab, Department of Urban Studies and Planning, Massachusetts Institute of Technology, USA;College of Architecture and Urban Planning, Tongji University, China;Singapore-MIT Alliance for Research and Technology, Singapore;Senseable City Lab, Department of Urban Studies and Planning, Massachusetts Institute of Technology, USA

摘要

Along with the rising space-of-flow, non-local functions are identified as a comparable force to local urban-hinterland relations that determine urban development. However, little evidence and explanation have been presented for how urban centers (UCs) are made of local and non-local functions. By examining the case of Shanghai, this paper endeavors to fill the gap with multiple big/open datasets. We first identify UCs using a new method as well as employment/activity densities inferred from cell phone data. Then, local and non-local forces are measured using functional and interaction metrics derived from theoretical deduction based on classic economic-space theories. Our result demonstrates that non-local functions have become a comparable force as local functions. Spatial analyses show that non-local functions generally agglomerate in about 30 UCs (27%) in the inner-ring Puxi area, presumably indicating a locational preference for local services and demand; local functions are distributed more like the central places diagram in Isard (1956). Additionally, specialization analysis of UCs shows clear evidence that non-local functions cause both centralization and decentralization in UCs, which can be well explained by the deductive theoretical patterns. This paper contributes to the literature of urban spatial structure with new evidence as well as a generic theoretical explanation and transferable methodologies.