2025-05-15
来源：CITIES Volume: 162 DOI: 10.1016/j.cities.2025.105958
作者：张正峰、郭珊等
张正峰｜中国人民大学公共管理学院教授
郭珊｜中国人民大学公共管理学院副教授
摘要：
Understanding the spatiotemporal dynamics of industrial land is crucial for sustainable land use and industrial transformation. However, data and methodological limitations hinder large-scale, long-term studies of industrial land from a high-granularity, multi-dimensional perspective. This study develops a methodological framework using multi-source data to identify industrial land parcels, enabling a systematic analysis of both external attributes (including quantity scale and spatial pattern) and internal attributes (including functional structure and utilization intensity). The validity of the framework is demonstrated through a case study of the Southern Jiangsu Urban Agglomeration in China from 1990 to 2020. The overall accuracy of industrial land identification reaches 94.70 %, with detailed representations of parcel morphology. Based on this data, the study reveals the nonlinearity of industrial land evolution, as well as the interrelationships and asynchrony among various attributes. Spatial patterns at the parcel level are uncovered, including four types of spatial evolution. Finally, the study identifies potential dimensions and areas for industrial land upgrading and provides specific recommendations for addressing compatibility between different attributes in planning and correcting land misallocation in land supply. The study contributes a new analytical method for advancing industrial land theory and offers support for refined management practices.
关键词：
Industrial land; Land parcel identification; Spatiotemporal dynamics; Multidimensional framework; Multi-source big data
