Spatial identification of "production living ecological" spaces in urban-rural regional system by integrating multiple source data

 

The precise identification of the "Production-Living-Ecological" spaces (PLES) plays a crucial role in  optimizing urban functional zones, constructing livable cities, and promoting balanced urban-rural development.  However, present studies on the functional identification of PLES exhibit a deficiency in comprehensive understanding and application of quantitative methods that integrate and interact with spatial elements. It is urgent to integrate multi-source geographical big data, considering the functional characteristics of different urban-rural regional systems, to establish a coherent and effective scheme for identifying spatial functions. To address this need, this study established three indices—Spatial Function Strength index (SFS), Spatial Function Coverage  index (SFC), and Spatial Function Interaction index (SFI) —from Point of Interest (POI), land cover, and mobile communication record, respectively. Utilizing road networks as the basic spatial unit for analysis, a decision tree was constructed for interpretation. Furthermore, landscape pattern indices were employed to analyze the spatialfunction characteristics at multiple scales including landscape, class and patch scale. The findings revealed significant functional disparities across various urban-rural systems. As increasing urbanization intensifies, there is an observed increase in spatial type diversity whereas the aggregation index of similar space decrease, along with the increase of shape complexity and patch density. The analysis identifies 13 distinct PLES patterns, notably,   ecological spaces predominantly occupy rural areas, while living spaces are primarily urban. The morphology and  distribution of production spaces vary with the dominant industries in different urban-rural systems. Fusion spaces  generally mirror the pattern of adjacent spaces, whereas interaction spaces are chiefly found in the transition zones  between urban and rural areas. Additionally, landscape pattern indices at the patch scale provide additional  evidence supporting systematic principles governing PLES from a more refined perspective. This study also  highlights those specific areas characterized by high spatial diversity but low agglomeration, providing new  scientific guidance for urban spatial planning and management.