土地城市化过程的波纹扩散规律--基于中国234个地级市的实证分析

A Wave-shaped Diffusion Law for Physical Spatial Processes of Land Urbanization:Evidence from 234 Prefecture-Level Cities in China

研究目的:实证分析和定量模拟土地城市化过程的波纹扩散规律,为理解土地城市化过程,认识城市发展阶段、空间结构以及开展城市空间扩张模拟提供新的理论和方法。研究方法:基于中国234个地级市1990—2019年不透水面数据,采用圈层梯度分析法研究增量城市土地密度分布特征,验证土地城市化过程的波纹扩散规律,并选择高斯函数进行定量模拟。研究结果:(1)不同规模城市其土地城市化过程均表现出明显的波纹扩散特征,高斯模型拟合精度R2均值达到0.88;(2)人口规模越大的城市其土地城市化速度越快,空间分布也更为分散;(3)基于土地城市化过程的高斯拟合模型能较好地推导城市土地分布的宏观梯度递减模式,推导结果与观测结果相关性达到0.91(R2均值)。研究结论:不同等级规模城市其土地城市化过程均表现出明显的波纹扩散规律,且可用高斯模型进行定量化表达和分析,这一规律是对既有城市空间理论的有益补充。

The purpose of this paper is to empirically analyze and quantitively stimulate a wave-shaped diffusion law for physical spatial processes of land urbanization,providing new theory and method for understanding the process of urbanization,and the stage and spatial structure of urban development as well as for conducting simulation of urban expansion.The research methods are as follows.The spatiotemporal characteristics of the physical spatial processes of urban land expansion are explored based on annual new urban land data of 234 prefecture-level cities in China from 1990 to 2019,using a concentric-ring based gradient analysis.The results show that the physical spatial processes of urban land expansion occur in cities with different sizes,presenting a wave-shaped diffusion pattern,and can be quantitatively modelled by a Gaussian-based model.The fitting accuracy(R2)reaches an average of 0.88.The parameters of the Gaussian model can help characterize urban land expansion processes.Analysis of model parameters show that cities with larger urban population usually have larger urban extent,faster urban expansion,and more scattered urban development.The consistency(represented by R2)between the observed and deducted macroscale urban land distribution reaches 0.91,which verifies the feasibility of the Gaussian-based model.In conclusion,the wave-shaped diffusion law of urban land expansion processes and its quantitative model are valuable supplements to the existing urban theory.