Exploring the Relationship between Spatiotemporal Distribution of Urban Vibrancy and Neighborhood Attributes by Coupling Multi-Source Data: A Case of Nanshan District in Shenzhen

The dynamic pulse of urban life has captivated scholars, policymakers, and city planners. Understanding the intricate mechanisms that shape urban vibrancy is pivotal for optimizing urban vibrancy patterns and fostering sustainable livable cities. However, previous studies have predominantly emphasized the relationship between objective neighborhood attributes and urban vibrancy. Systematically incorporating a multidimensional set of neighborhood factors, including human vision, functionality, morphology, location, and street configuration, and utilizing spatial regression models in resonance with advanced algorithms facilitates a more profound comprehension of the spatiotemporal differentiation of urban vibrancy within urban spaces and the driving or inhibiting factors involved. This work focuses on the Nanshan District of Shenzhen City as the study area, and has three objectives: (i) one-week passenger flow data that characterized the spatiotemporal distribution of urban vibrancy were provided; (ii) broadly collected Street View Images (SVI) were incorporated as a visional environmental factor, together with functional and morphological factors, into understanding the influencing mechanism of urban vibrancy; and finally, (iii) machine learning tools were employed to apply the Random Forest Regression (RFR) algorithm in exploring the independent driving role of the characteristic factors behind the temporal distribution of urban vibrancy, and the Geographically Weighted Regression (GWR) model was used to probe the influence of the characteristic factors on the spatial distribution of this dynamic concept. The results are as follows: 1) Nanshan District emerges a mosaic of multi-vibrancy centers, where urban functions and business forms emerge as the dominant causative factors that shape the urban vibrancy and human perceptions, exhibiting unignorable effects. Evidently, employment and commuting activities have a ripple effect on economic urban vibrancy throughout the center of the administrative district, making weekday vibrancy surpass that of weekend; 2) Variance of the spatiotemporal distribution of urban vibrancy is improved or inhibited by the characteristic factors, among which, the entertainment function plays a major role as the driving force; additionally, enclosure, Point Of Interest (POI) diversity, road density, and building density have promoting effects on the overall urban vibrancy. The effects of greenness, openness, business function, traffic function, residence function, and living function variate depends on space changes as well as the inhibiting effects of casual function; 3) Neighborhood with compact buildings, composite functions, and green surroundings motivate the surge of urban vibrancy; therefore, the creation of spaces with mixed functionality and compact structure, routes with diverse street view interface elements and open sky views, places with multi-level and highly accessible blue-green landscapes, and cultivation of new urban functions through urban renewal are encouraged during urban construction. This study reveals the importance of human vision in the relationship with urban vibrancy, provides insights into the dynamic evolution of urban vibrancy, and makes recommendations for stock planning for a sustainable and lively city, creating opportunities for people to enjoy various daily life activities by providing convenience, services, and novel cultural experiences, especially in the areas of shopping, catering, infrastructure, and transportation facilities. This study helps urban planners deepen their understanding of urban vibrancy, thereby promoting the implementation of people-oriented urban planning and management.