Impacts of Built Environment on Urban Vitality in Cultural Districts: A Case Study of Haikou and Suzhou

Abstract

In the context of urban development, bridging the gap between urban regeneration and people’s demand for high-quality built environments is a current focus of research. Exploring the vitality of certain kinds of urban districts is imperative for comprehending human needs for specific built environments and fostering urban renaissance and advancement. This urgency arises from the prevailing lack of in-depth studies on district vitality, as current research primarily provides a general assessment of street vitality. Thus, this study aims to explore the correlations between indicators of urban vitality in cultural districts and built environments, using Haikou and Suzhou as case studies and employing multiple data sources (e.g., Baidu heat maps and nighttime light) and measurements. By applying a logit regression model, we find the following: (1) Traffic network integration has a positive impact on daytime vitality in Haikou and nighttime vitality in Suzhou, but it negatively affects nighttime vitality in Haikou. (2) In terms of nighttime vitality, both commercial density and greenery positively influence the overall blocks and various subgroups. (3) The proportion of cultural facilities in Suzhou has a detrimental effect on daytime vitality, especially for blocks with fewer permanent residents and lower land values. The present study, while limited to selected cultural blocks in Suzhou and Haikou, establishes the groundwork for a better comprehension of how spatial vitality can be enhanced at the street segment level, thereby contributing to the investigation of the varying impacts that built environment factors have on urban vitality in tourism cities at different stages of development. It uncovers the inherent latent characteristics found within cultural blocks across diverse regions and offers innovative perspectives and recommendations for optimizing the sustainable development of urban blocks.

1. Introduction

The role of blocks, or street segments, within urban districts is multifaceted in the context of urban development. As vital hubs of public life within cities [1], they not only drive economic growth but also foster social interactions and communication, thereby enhancing community cohesion and a sense of belonging [2]. However, the rapid development of blocks driven by target-oriented objectives often neglects the needs and experiences of human beings in public spaces [3], leading to insufficient attention to gradual improvement processes for built environments [4]. Consequently, this trend inhibits the revitalization of blocks and, furthermore, is not conducive to the sustainable development of urban districts. Meanwhile, with the thriving progress of urban economies, individuals have higher expectations for both urban and block quality [5]. Therefore, inquiry into the types of built environments that can attract a larger population and enhance urban vitality has emerged as a prominent focal point in current research on urban renaissance and development.

In recent years, an increasing number of scholars have acknowledged the significance of blocks as essential components of social space structure [6]. It is defined as a series of intangible activities that emerge under the influence of built environment components, such as transportation systems, functional characteristics, and spatial quality [7]. This perspective helps unveil the interplay between urban vitality and associated built environment features. By creating high-quality public spaces, it becomes possible to fulfill individual needs in terms of physical experience and psychological cognition, thereby further promoting block revitalization [8]. For instance, we can meet the crowd’s travel needs when updating transportation system layouts to address accessibility issues for residents in and around blocks; we can support the close integration of functional characteristics with people’s daily life usage and, at the same time, foster mixed functions along with diverse activities and multicultural integration; we can also focus on building enclosures, livable scales, and diversified forms, as well as combining the static with the dynamic in the redesign of spaces to create human-centered public places [9]. Vibrant block spaces facilitate various social interactions and enhance residents’ sense of security and belonging to urban environments [10], which actively contribute to improving resident wellbeing and promoting sustainable urban development [11].

A quantitative assessment of block vitality has garnered significant attention across disciplines, including urban planning, geographic science, and the social sciences [12]. As suggested by Mehta [13], lively streets refer to those streets with a high level of human participation in a series of fixed or continuous activities, especially social activities. In earlier studies, it was primarily characterized through on-site surveys of human activities [14], interactions, and life experiences. However, this approach has several limitations, such as a higher cost and a more restricted sample size. With the continuous advancements in information technology, smart sensing devices can now collect urban data from multiple sources with varying spatiotemporal resolutions and capture a substantial number of human activities. This has emerged as a valuable resource for measuring block vitality, offering novel avenues for urban research and governance. Many studies have utilized diverse sources of data on social perception features (e.g., mobile phones [15], social media [16], population heat maps [17], traffic trajectories [18], nighttime light [19], street views [20], and Wi-Fi hotspots [21]) to effectively quantify the vitality of blocks. Given the increasing abundance of urban data, it is feasible to investigate the influencing factors of block vitality in a more detailed and comprehensive way. However, current research often examines cities at the street level and utilizes a narrower approach focusing on individual samples when trying to gain insights into urban block revitalization. As a result, we may fail to account for the diversity and heterogeneity of social, economic, and cultural activities within urban spaces, leading to somewhat ambiguous and even biased conclusions. Hence, there is a need to divide street data into finer blocks (e.g., districts) when describing vitality and its influencing factors. Furthermore, integrating data from various sources is crucial for attaining a more comprehensive understanding of neighborhood vitality and its influencing factors.

The paper focuses on blocks within cultural districts as the primary research objects, with a particular emphasis on integrating multidimensional public space data. This is because cultural districts, as some of the most significant public spaces in a city, carry urban memories and serve as essential components of the urban heritage environment [22]. They reflect urban characteristics and embody core values [23]. The paper employs a quantitative approach to investigate the correlations between the indicators of block vitality and those of built environment quality, aiming to explore the influences of built environment components on block vitality within cultural districts and narrow the gaps in urban vitality research. The ultimate goal is to delve deeper into the specific influencing mechanisms and patterns of urban vitality, providing a scientific basis for further enhancing urban blocks’ quality and promoting overall urban renaissance. The conclusions drawn from this study may be utilized to promote the renovation and revitalization of similar blocks.

The remainder of this paper is organized as follows: Section 2 reviews existing explorations of urban vitality and the corresponding built environment components that serve as influencing factors. Section 3 illustrates our research methodology, including the research framework, study area, variables, data, and regression model. Section 4 identifies the relevant influencing factors and discusses their effects. The conclusions are presented in Section 5, followed by a discussion on the limitations in the current research and some suggestions for further studies.

2. Related Studies

2.1. Explorations of Urban Vitality Assisted by Multisource Data

Public spaces serve as vital material carriers for people to engage in social activities and diverse interactions. The research on the vitality of urban blocks empowered by multisource data can assist urban planners, relevant governments, and community practitioners in gaining deeper understandings of the importance of micro-level block revitalization and symbiosis and their specific impacts on urban regeneration. In other words, it can generate relevant policies and transformation plans. In traditional research, the evaluation of vitality involves exploring theories and defining concepts, and it has been carried out through field observations and questionnaire surveys. Vitality is widely recognized as a central aspect of urban design and regeneration. Jane Jacobs proposed that vitality is the most fundamental characteristic of cities, wherein human interactions within physical spaces constitute urban diversity [24]. Lynch underscored that vitality is the reflection of settlements’ capacity to support individual functions and ecological needs [25]. Gehl stated that public spaces of good quality can enhance vitality through a variety of optional outdoor activities [26], contrasting with Le Corbusier’s overly rigid and inconvenient spaces. In recent years, Bentley has advocated that vibrant spaces can provide people with multiple functions, purposes, and choices [27]. Building on such understandings of urban form, Lichey delved into both the “space” and “society” aspects, stating that meticulous organizational planning and process-oriented urban design can truly stimulate urban vitality [28]. Based on theoretical insights related to urban vitality, the elements contributing to vitality at the urban development level are greatly influenced by human activities, land use, and economic growth. Recent studies have utilized Twitter data to capture human activities in urban spaces, examining their spatiotemporal distribution and mobility patterns to assess the relationship between diversity and vitality [29]. Pedestrian flow has been taken as a key indicator of urban vitality [30]. Dynamic vitality has been assessed in terms of intensity and stability, while the driving factors behind it have been investigated from the perspective of land use [31]. Moreover, geographical weighted regression models have been used to analyze the heterogeneity of the driving factors for urban vitality across various socioeconomic groups [32]. In this paper, urban vitality primarily centers on the vitality of urban blocks, which can be measured by combining the intensity of human activities on streets at different time points with the density of lighting facilities in the neighborhood.

In the research on the vitality of cultural districts, heritage and utility value have been emphasized, such as cultural environment, spatial environment, business environment, residential environment, attraction, and other aspects [33]. It has been discovered that the provision of small-scale, dispersed cultural facilities is crucial for vernacular cultural pursuits [34]. To delve into vitality and its determinants, recent studies have turned to location-based services data to create a standard measure. For instance, a study utilized Baidu heat maps to quantify spatiotemporal variations in urban vitality on weekdays and weekends, informing targeted planning policies based on regional disparities [35]. Estimates for metro-led underground spaces have been obtained through regression analyses using smart card and location-based service data [36]. In addition, the nighttime light level has been employed as a metric for urban and regional nighttime activities. For example, one study examined, for the first time, the correlation between nighttime light levels and gross regional product [37]. DMSP/OLS nighttime light images were divided into five different types of nighttime light areas in another study to explore the relationship between dynamic spatial expansion patterns and spatiotemporal characteristics [38]. Research utilizing nighttime light data to analyze urban vitality found that cities with high vitality usually boast convenient transportation and rich historical or cultural heritage [39]. However, most studies have relied on a single data source to identify the patterns of association between urban vitality and other urban characteristics. As a result, they sometimes encounter challenges with the data caused by seasonality. It is therefore difficult to effectively capture street spatial and functional features at a more “micro”-level with a single measure, limiting the comprehensiveness of discussions on mechanisms. This calls for adopting multiple data sources and measures of urban vitality.

2.2. Practices of Evaluating Built Environment and Spatial Quality Components as Influencing Factors of Blocks

At the neighborhood level, studies have focused on analyzing the built environment elements within blocks and their impacts on street vitality. For example, compactness, walkability, functional mix, appropriate building density, and other factors have been considered crucial for urban vitality [40]. Composite urban morphological analyses involving streets, plots, and buildings define evaluation elements, including street size, accessibility, and other block dimensions [6]. When discussing the relationship between morphology and vitality, existing studies have considered density and typology as core factors while controlling for accessibility, functional mix, and block size [41]. Emphasis has been placed on pedestrian facilities in evaluating how street infrastructure affects pedestrian services [42]. Moreover, an evaluation of commercial street users has shown that the spatial accessibility of streets influences their participation in social activities, and mixed-use functions and pedestrian experience further contribute to space vitality [43]. One study has suggested using the characteristics of urban “density, mix, and access” to evaluate the walkability of urban blocks [44]. In the specific context of Suzhou, the impact of the built environment on the spatial variation in shared bicycle usage during weekdays and weekends is investigated in terms of density, diversity, and destination accessibility [45]. Social-spatial dynamics are quantified through factors such as community elements, public facilities distance, and public facility density in Suzhou [46]. A review on the differences between urban and rural bed and breakfast (B&B) distributions in Haikou found that the distance from a highway is crucial for the urban distribution of B&Bs, while road network density, representing transportation accessibility, is a significant factor influencing the rural distribution of B&Bs. Additionally, housing density, hotel service availability, and neighborhood residential density have also been found to affect the spatial distribution of B&Bs [47]. There has been increased attention to the influential relationship between individuals and street spatial characteristics and accessibility.

In recent years, deep learning algorithms have been employed to measure built environment indicators. For example, one study proposed the automatic evaluation of streetscape walkability using semantic segmentation and statistical modeling [48]. The DeepLabv3+ model was used to segment images captured by Google Street View into various categories, including different kinds of infrastructures, vehicles, pedestrians, and vegetation. This segmentation aimed to analyze the impact of the community-built environment on the survival rate of local social organizations [49]. With the assistance of deep learning algorithms and street view images, the differential relationships between built environments and vitality were captured through hierarchical regression, starting with morphological, functional, and human-scale features [50]. Additionally, classification and regression evaluations have been used for regional streets. Another study conducted a questionnaire-based survey and applied logistic regression analyses on the six dimensions of urban street design in Singapore [51]. A comprehensive vitality index was developed using social media data, incorporating the six dimensions of the built environment (i.e., neighborhood attributes, urban form, function, landscape, location, and street configuration) into a spatial regression model [52]. Through multi-object tracking and scene classification, five street built-environment variables related to vitality were measured, and the significances were evaluated [53]. A 5D framework has been established, comprising multiple data sources (density, diversity, design, destination accessibility, and transit distance), to systematically explore spatial and temporal correlations between built environment and urban vitality [54]. To explore the relative impacts of built environment factors on street vitality and assess potential changes over different time periods, a spatial panel Durbin model has been constructed [55]. In conclusion, an increasing number of studies have begun to establish comprehensive measurement and evaluation frameworks, moving beyond qualitative subjective judgments to assess the overall or specific environment of a region more objectively.

2.3. Research Gaps and Focus of This Study

Previous research has mainly focused on theoretical foundations, vitality composition and evaluation, feature classification, and the relationships among some elements. However, there are several research gaps, as follows: First, the quantification of urban vitality primarily relies on a single data source, lacking multidimensional and multilevel measurements, potentially oversimplifying complex urban issues. Second, there is a scarcity in the assessments and quantification indicators for comparable thematic cultural blocks between urban and human scales, often limited to single-block studies or overly generalized assessments of noncomparable blocks, requiring a more detailed exploration of neighborhood attributes.

This study develops a multidimensional hierarchical framework to explore block vitality, providing detailed explanations and quantification formulas for various indicators. The aim is to explore which factors can effectively promote the vitality of surrounding communities to form super blocks to improve the diversity, consumption and inclusiveness of a region. Furthermore, by conducting heterogeneous analysis and examining the impacts of similar built environment factors on vitality in tourist cities in different development phases, this study aims at obtaining strategies to uncover the inherent potential of local cultural blocks, enhance district image, improve resident wellbeing, promote urban economic development, and achieve “win–win” solutions for industrial and technological upgrades, as well as the environmental optimization and sustainability of urban blocks in the future.

3. Methodology

3.1. Construction of Research Dimension

This paper recognizes the intensity of people activities on streets as a dependent variable representing block vitality. Subsequently, nighttime lighting (NTL) level and daily crowd density at 10 am, 2 pm, and 6 pm, according to location-based services (LBS) data, were also chosen as dependent variables. For independent variables, we selected indicators from three dimensions—transport system, function feature, and spatial quality—following a literature review, which is provided in Section 2. The rationale for selecting independent and dependent variables is elaborated in Section 3.3. Finally, we obtained patterns of the variables involved in this study through descriptive and correlation analyses. We employed logit regression models to analyze the heterogeneity of different regions and influencing factors, which was followed by robustness checks.

In our analysis, by considering the geographic location, average residential house price, and number of households within blocks as moderators (i.e., classification variables), as well as examining the coefficient correlations of the model variables, we aim to capture the specific mechanisms of generating vitality in cultural blocks from different regions and historical backgrounds under the same system of built environment indicators. We seek to explore factors that can effectively promote the vitality of neighborhoods and surrounding communities (Figure 1).

3.2. Study Area

This study selected cultural renovation blocks, which emerged in the process of urban redevelopment and transformation, as the thematic research objects. Because of their location and historical features, they attract cultural enthusiasts and tourists. Many historically rich blocks have regained vitality through revitalization and integration into young people’s daily lives. As an urban renaissance pilot city in China, Suzhou boasts diverse historical and cultural blocks, while Haikou represents an emerging tourism city that has rapidly developed over the past three decades. The protection of historical and cultural blocks is relatively intact in Haikou. Thus, both cities serve as unique but reasonable choices of research objects. Building on existing research, our study utilized Baidu map LBS data and NPP-VIIRS nighttime light data to develop dependent variables measuring block vitality across multiple levels.

Additionally, we gathered a wide range of independent variables related to the built environment of blocks. As illustrated in Figure 2, Figure 3 and Figure 4, we selected the following four representative cultural districts as our study areas: (1) Pingjiang Road Historical and Cultural District (PJL), (2) Shantang Street Historical and Cultural District in Suzhou (STJ), (3) Haikou Arcaded Street District (QL), and (4) Fucheng Cultural District in Haikou (FC). These four districts represent historically and culturally significant districts in Suzhou and Haikou, respectively. This enabled us to conduct an in-depth exploration of the specific impact mechanisms of blocks from different regions and historical backgrounds under a suit of similar built environment indicators.

Suzhou is renowned for its exquisite gardens, ancient water streets, and rich cultural heritage. The Pingjiang Road Historical and Cultural District, with its narrow alleys and traditional houses, integrates the style of Suzhou Garden and represents Chinese classical water town culture. Similarly, the Shantang Street Historical and Cultural District preserves the building styles from the Ming and Qing dynasties, with its unique location along the river adding to its charm. In Haikou, the Sotto Porticos Street Area is known for its distinctive arcade architecture in the Nanyang style, showcasing unique features. The Fucheng Cultural District, situated in the historic city center of Haikou, reflects the distinctive residential architecture of the northern part of Hainan Island. Whereas Suzhou represents the traditional culture of historical cities in China, Haikou showcases the rapid growth and dynamic development in urban centers. These two cities exhibit different attractions in the tourism industry, highlighting the diverse facets of tourism cities in different stages of development in China.

The rationale for selecting block/street segments (hereafter, we refer to them as blocks for simplicity) as analytical units is multifaceted. First, blocks are commonly utilized in urban morphology and design analyses, enabling the integration of classical theories and existing research perspectives for practical applications. Second, blocks serve as inclusive representations of urban life and culture, with their enduring impacts significantly contributing to the characteristics of different cities and facilitating flexible and sustainable urban planning practices. Last but not least, themed cultural blocks have the potential to stimulate tourism and further enhance vitality. Through strategic planning and the creation of unique cultural environments, these blocks can attract tourists and residents alike, fostering distinctive tourism experiences and fostering the growth and prosperity of larger districts and areas.

3.3. Variables and Data Collection

In the empirical investigation, the selected four districts were further divided into 311 street segments and surrounding blocks based on traffic networks. Specifically, Haikou Arcaded Street District comprises segments numbered 1–105, the Fucheng Cultural District includes segments numbered 106–151, the Shantang Street Historical and Cultural District encompasses segments numbered 152–200, and the Pingjiang Road Historical and Cultural District consists of segments numbered 201–311. The length of each segment ranges from 50 to 200 m. Subsequently, data collection and analysis were conducted for each of these 311 research street segments.

3.3.1. Dependent Variables for Cultural Blocks

In our research, the dependent variables included daily crowd distribution density (based on LBS data) and NTL level. Various types of LBS data, such as mobile phone usage, public transportation card swipes, taxi trips, and taxi trajectory data, have been used to describe social vitality [56], identify living space features [57], and analyze transportation modes [58]. These data enable a more accurate description of large-scale spatial behavior patterns at the individual level [57], providing a basis for measuring activity intensity. Baidu heat maps use the geographic location data of the LBS platform’s mobile users to visualize real-time differences in relative crowd density in cities using distinct colors. In this paper, Baidu heat maps were extracted for the 311 street segments within the four study areas for weekdays (30 January; 1, 5, and 7 February 2024) and holidays (28 January; 3, 11, and 17 February 2024) at 10 am, 2 pm, and 6 pm. Subsequently, they were imported into ArcGIS for the purpose of vectorization processing. A 20 m buffer zone was established around the street centerline to extract and summarize heat map data at the street level. Finally, grid points were generated to extract heat values. The average values of the grid points were assigned on the basis of the color coding and used as the representative heat value of the street during that specific period. Based on existing studies, land use data and nighttime light data were utilized to analyze patterns in nighttime crowd flows in Haikou [59]. The NTL data in our study were sourced from the global 500 m resolution NPP-VIIRS-like dataset [60,61]. This dataset corrects NTL data from DMSP-OLS and NPP-VIIRS sensors using a cross-sensor calibration method, ensuring the parameters are consistent with the NPP-VIIRS data. The new NTL data in 2023 reflect the temporal changes in the population and light brightness at various scales more accurately.

3.3.2. Independent Variables for Built Environment and Classification Variables

This study required the manual selection of objects and on-site surveys, supplemented by the use of handheld cameras to capture street view images aligned with the line of sight (Figure 4). The street view images were acquired by invoking Baidu Street View’s API query via an HTTP URL. A total of four street view perspectives were captured as follows: two parallel to the road (front and back) and two perpendicular to the direction of the road (left and right), with each line of sight having a viewing angle of 90 degree. In the areas in which there were insufficient street view images, we used a handheld camera to supplement the street view images. Quantification was conducted using data such as Baidu heat maps and points of interest (POIs). The study interpreted the research blocks according to the following three dimensions: (1) transport system, (2) function feature, and (3) spatial quality. This included 14 indicators, in total, as independent variables for assessment (Table 1). The measures included in each dimension are listed below.

Table 1. Independent variables and data sources.

Domain	Secondary	Indicators and Formula
Transport System	(1.1) Road grade	Road grade (x) = (branch, secondary, main, etc.) We classified the roads and calculated their widths based on the original data.
	(1.2) Minimum distance to green space	$D_{green}=\min d_{\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{e}\mathrm{n}}$	$d_{\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{e}\mathrm{n}}$: The distance from the street center to the block park green space
	(1.3) Minimum distance to public transit	$D_{pt}=\min d_{\mathrm{p}\mathrm{t}}$	$d_{\mathrm{p}\mathrm{t}}$: The distance from the street center to the neighborhood bus stop
	(1.4) Traffic network integration	$I_n=\frac{m\left[log2\right(m+2)/3-1]+1}{(m-1)(Md-1)}$	$Md$: The mean depth
			$m$: The number of nodes in the spatial connectivity graph
Function Feature	(2.1) Degree of business aggregation	H=$-{\sum }_{i=1}^S\left(P_i\mathit{ln}{P}_i\right)$	$S$: The total number of service facilities
			$P_i$: The proportion of the i-th service facility to the total number
	(2.2) Proportions of living and public services	$f_n\left(x\right)=\frac{1}{n\pi h^2}{\sum }_{n=1}^{n}K\left(1-\frac{{(x-x_i)}^2-{(y-y_i)}^2}{h^2}\right)$	$K\left(\cdot \right):$ Kernel function
	(2.3) Proportion of catering facilities		$h$: The bandwidth, or search radius
	(2.4) Proportion of cultural facilities		${(x-x_i)}^2-{\left(y-y_i\right)}^2$: The distance between two points
	(2.5) Commercial density	$D_{poi}=\frac{N_p}{L_r}$	$N_p$: The quantity of the four types of points of interest (POIs), including dining options, shopping outlets, financial institutions, and lifestyle services
			$L_r$: The length of the street
Spatial Quality	(3.1) Width-to-height ratio	$A_r=\frac{W}{H}$	$\mathrm{W}$: The width of the street
			$H$: The height of buildings along the street
	(3.2) Greenery	$I_{green}=\frac{S_{green}}{S_{\mathrm{a}\mathrm{l}\mathrm{l}}}$	$S_{\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{e}\mathrm{n}}$: The total sum of pixels of trees and vegetation
			${\mathrm{S}}_{\mathrm{a}\mathrm{l}\mathrm{l}}$: The total sum of pixels of all labels recognized through semantic recognition
	(3.3) Sky openness	$I_{sky}=\frac{S_{\mathrm{s}\mathrm{k}\mathrm{y}}}{S_{\mathrm{a}\mathrm{l}\mathrm{l}}}$	${\mathrm{S}}_{\mathrm{s}\mathrm{k}\mathrm{y}}$: The total sum of pixels of the sky
	(3.4) Building enclosure	$I_{en}=\frac{S_{\mathrm{b}\mathrm{u}\mathrm{i}\mathrm{l}\mathrm{d}\mathrm{i}\mathrm{n}\mathrm{g}}+S_{\mathrm{w}\mathrm{a}\mathrm{l}\mathrm{l}}+S_{\mathrm{f}\mathrm{e}\mathrm{n}\mathrm{c}\mathrm{e}}+S_{\mathrm{p}\mathrm{o}\mathrm{l}\mathrm{e}}}{S_{\mathrm{a}\mathrm{l}\mathrm{l}}}$	$S_{\mathrm{b}\mathrm{u}\mathrm{i}\mathrm{l}\mathrm{d}\mathrm{i}\mathrm{n}\mathrm{g}}$: The total sum of pixels of buildings
			$S_{\mathrm{w}\mathrm{a}\mathrm{l}\mathrm{l}}$: The total sum of pixels of walls
			$S_{\mathrm{f}\mathrm{e}\mathrm{n}\mathrm{c}\mathrm{e}}$: The total sum of pixels of fences
			$S_{\mathrm{p}\mathrm{o}\mathrm{l}\mathrm{e}}$: The total sum of pixels of pillars
	(3.5) Car and pedestrian friendliness	$I_{ac}=\frac{S_{\mathrm{r}\mathrm{o}\mathrm{a}\mathrm{d}}+S_{\mathrm{s}\mathrm{i}\mathrm{d}\mathrm{e}\mathrm{w}\mathrm{a}\mathrm{l}\mathrm{k}}}{S_{\mathrm{a}\mathrm{l}\mathrm{l}}}$	$S_{\mathrm{r}\mathrm{o}\mathrm{a}\mathrm{d}}$: The total sum of vehicular space
			${\mathrm{S}}_{\mathrm{s}\mathrm{i}\mathrm{d}\mathrm{e}\mathrm{w}\mathrm{a}\mathrm{l}\mathrm{k}}$: The total sum of pixels of sidewalks

Please refer to Table 2 and Figure 5 for the summary statistics of these indicators.

Table 1. Independent variables and data sources.

Domain	Secondary	Indicators and Formula
Transport System	(1.1) Road grade	Road grade (x) = (branch, secondary, main, etc.) We classified the roads and calculated their widths based on the original data.
	(1.2) Minimum distance to green space	$D_{green}=\min d_{\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{e}\mathrm{n}}$	$d_{\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{e}\mathrm{n}}$: The distance from the street center to the block park green space
	(1.3) Minimum distance to public transit	$D_{pt}=\min d_{\mathrm{p}\mathrm{t}}$	$d_{\mathrm{p}\mathrm{t}}$: The distance from the street center to the neighborhood bus stop
	(1.4) Traffic network integration	$I_n=\frac{m\left[log2\right(m+2)/3-1]+1}{(m-1)(Md-1)}$	$Md$: The mean depth
			$m$: The number of nodes in the spatial connectivity graph
Function Feature	(2.1) Degree of business aggregation	H=$-{\sum }_{i=1}^S\left(P_i\mathit{ln}{P}_i\right)$	$S$: The total number of service facilities
			$P_i$: The proportion of the i-th service facility to the total number
	(2.2) Proportions of living and public services	$f_n\left(x\right)=\frac{1}{n\pi h^2}{\sum }_{n=1}^{n}K\left(1-\frac{{(x-x_i)}^2-{(y-y_i)}^2}{h^2}\right)$	$K\left(\cdot \right):$ Kernel function
	(2.3) Proportion of catering facilities		$h$: The bandwidth, or search radius
	(2.4) Proportion of cultural facilities		${(x-x_i)}^2-{\left(y-y_i\right)}^2$: The distance between two points
	(2.5) Commercial density	$D_{poi}=\frac{N_p}{L_r}$	$N_p$: The quantity of the four types of points of interest (POIs), including dining options, shopping outlets, financial institutions, and lifestyle services
			$L_r$: The length of the street
Spatial Quality	(3.1) Width-to-height ratio	$A_r=\frac{W}{H}$	$\mathrm{W}$: The width of the street
			$H$: The height of buildings along the street
	(3.2) Greenery	$I_{green}=\frac{S_{green}}{S_{\mathrm{a}\mathrm{l}\mathrm{l}}}$	$S_{\mathrm{g}\mathrm{r}\mathrm{e}\mathrm{e}\mathrm{n}}$: The total sum of pixels of trees and vegetation
			${\mathrm{S}}_{\mathrm{a}\mathrm{l}\mathrm{l}}$: The total sum of pixels of all labels recognized through semantic recognition
	(3.3) Sky openness	$I_{sky}=\frac{S_{\mathrm{s}\mathrm{k}\mathrm{y}}}{S_{\mathrm{a}\mathrm{l}\mathrm{l}}}$	${\mathrm{S}}_{\mathrm{s}\mathrm{k}\mathrm{y}}$: The total sum of pixels of the sky
	(3.4) Building enclosure	$I_{en}=\frac{S_{\mathrm{b}\mathrm{u}\mathrm{i}\mathrm{l}\mathrm{d}\mathrm{i}\mathrm{n}\mathrm{g}}+S_{\mathrm{w}\mathrm{a}\mathrm{l}\mathrm{l}}+S_{\mathrm{f}\mathrm{e}\mathrm{n}\mathrm{c}\mathrm{e}}+S_{\mathrm{p}\mathrm{o}\mathrm{l}\mathrm{e}}}{S_{\mathrm{a}\mathrm{l}\mathrm{l}}}$	$S_{\mathrm{b}\mathrm{u}\mathrm{i}\mathrm{l}\mathrm{d}\mathrm{i}\mathrm{n}\mathrm{g}}$: The total sum of pixels of buildings
			$S_{\mathrm{w}\mathrm{a}\mathrm{l}\mathrm{l}}$: The total sum of pixels of walls
			$S_{\mathrm{f}\mathrm{e}\mathrm{n}\mathrm{c}\mathrm{e}}$: The total sum of pixels of fences
			$S_{\mathrm{p}\mathrm{o}\mathrm{l}\mathrm{e}}$: The total sum of pixels of pillars
	(3.5) Car and pedestrian friendliness	$I_{ac}=\frac{S_{\mathrm{r}\mathrm{o}\mathrm{a}\mathrm{d}}+S_{\mathrm{s}\mathrm{i}\mathrm{d}\mathrm{e}\mathrm{w}\mathrm{a}\mathrm{l}\mathrm{k}}}{S_{\mathrm{a}\mathrm{l}\mathrm{l}}}$	$S_{\mathrm{r}\mathrm{o}\mathrm{a}\mathrm{d}}$: The total sum of vehicular space
			${\mathrm{S}}_{\mathrm{s}\mathrm{i}\mathrm{d}\mathrm{e}\mathrm{w}\mathrm{a}\mathrm{l}\mathrm{k}}$: The total sum of pixels of sidewalks

Please refer to Table 2 and Figure 5 for the summary statistics of these indicators.

Table 2. Descriptive statistics of the built environment variables.

Domain	Independent Variables	Haikou		Suzhou
		Mean	SD	Mean	SD
Transport System	(1.1) Road grade	0.223	0.261	0.213	0.191
	(1.2) Traffic network integration	0.534	0.239	0.553	0.230
	(1.3) Minimum distance to green space	0.387	0.254	0.519	0.249
	(1.4) Minimum distance to public transit	0.339	0.237	0.264	0.223
Function Feature	(2.1) Degree of business aggregation	0.485	0.179	0.620	0.174
	(2.2) Proportions of living and public services	0.189	0.153	0.275	0.194
	(2.3) Proportion of catering facilities	0.430	0.242	0.419	0.244
	(2.4) Proportion of cultural facilities	0.042	0.095	0.281	0.199
	(2.5) Commercial density	0.452	0.254	0.214	0.171
Spatial Quality	(3.1) WH ratio	0.190	0.228	0.339	0.250
	(3.2) Greenery	0.250	0.251	0.210	0.183
	(3.3) Sky openness	0.333	0.205	0.375	0.191
	(3.4) Building enclosure	0.515	0.265	0.479	0.206
	(3.5) Car and pedestrian friendliness	0.256	0.222	0.202	0.237

SD stands for standard deviation; WH stands for width to height. The sample size is 311.

Table 2. Descriptive statistics of the built environment variables.

Domain	Independent Variables	Haikou		Suzhou
		Mean	SD	Mean	SD
Transport System	(1.1) Road grade	0.223	0.261	0.213	0.191
	(1.2) Traffic network integration	0.534	0.239	0.553	0.230
	(1.3) Minimum distance to green space	0.387	0.254	0.519	0.249
	(1.4) Minimum distance to public transit	0.339	0.237	0.264	0.223
Function Feature	(2.1) Degree of business aggregation	0.485	0.179	0.620	0.174
	(2.2) Proportions of living and public services	0.189	0.153	0.275	0.194
	(2.3) Proportion of catering facilities	0.430	0.242	0.419	0.244
	(2.4) Proportion of cultural facilities	0.042	0.095	0.281	0.199
	(2.5) Commercial density	0.452	0.254	0.214	0.171
Spatial Quality	(3.1) WH ratio	0.190	0.228	0.339	0.250
	(3.2) Greenery	0.250	0.251	0.210	0.183
	(3.3) Sky openness	0.333	0.205	0.375	0.191
	(3.4) Building enclosure	0.515	0.265	0.479	0.206
	(3.5) Car and pedestrian friendliness	0.256	0.222	0.202	0.237

SD stands for standard deviation; WH stands for width to height. The sample size is 311.

Figure 5. Descriptive analysis of indicators. The cross (×) refers to the mean value for each variable in each district.

Figure 5. Descriptive analysis of indicators. The cross (×) refers to the mean value for each variable in each district.

The first dimension, transport system, reflects various characteristics related to the within-block transportation and that are essential for enabling pedestrian activities, exerting a crucial impact on vitality. The evaluation indicators include road grade, minimum distance to green space, minimum distance to public or mass transit, and traffic network integration. Road grade was determined on the basis of vehicle speed and street width, categorized as main roads, secondary roads, etc. The minimum distance to green space measures the shortest distance from the street center point to parks and green areas, addressing residents’ basic needs and attracting more foot traffic, as green spaces serve as primary recreational areas. The minimum distance to public transit measures the shortest distance from the street center point to bus stops, reflecting the accessibility for individuals using public transportation to reach block destinations. Traffic network integration analyzes the degree of agglomeration or dispersion among elements in a spatial system by combining space syntax [62], indicating a spaces’ ability to attract transportation arrivals as a destination, thereby reflecting its centrality in the entire system. Higher traffic network integration implies higher accessibility, stronger centrality, and easier aggregation of foot traffic.

The “functional feature” dimension primarily reflects the functional composition and diversity of business formats within blocks, focusing on the proportions of living and public services (e.g., pharmacies, telecommunication offices, and public toilets), catering facilities (e.g., restaurants and tea houses), and cultural facilities (e.g., art galleries, libraries, and museums). These functional establishments not only meet people’s basic living needs but also stimulate consumption, significantly influencing vitality. The degree of business aggregation reflects the mix of functional formats that can be obtained by applying the equation corresponding to the Shannon–Wiener complexity index [63]. A higher concentration of functional establishments is advantageous in shortening travel distances to different destinations. Finally, the commercial density indicates the convenience of service functions.

The “spatial quality” dimension centers on studying the streets composing the blocks, primarily addressing people’s needs for the comfort and convenience of public spaces. Indicator constructions involved establishing a buffer zone for each street, where the intersection of the street buffer zone and buildings constitutes street-side buildings. The width-to-height (WH) ratio was obtained by dividing the street width by the average height of street-side buildings. Greenery measures the level of street vegetation; sky openness refers to the percentage of sky area visible to human eyes, significantly impacting the comfort of street environments; building enclosure measures the degree of enclosure formed by buildings, walls, and other structures, whereby taller surrounding objects can create a more confined spatial sensation, impeding a pleasant pedestrian experience; car and pedestrian friendliness assesses the proportion of road areas for pedestrians and motorized and nonmotorized vehicles. These indicators were obtained through semantic segmentation recognition of researcher-captured street view images using the model trained on the Ade20k dataset in Deeplab v3 via PyCharm 2014.1.2 [64], calculating relevant pixel areas and their ratios to the total pixel area.

In addition, the study divided blocks into multiple neighborhood units with a radius of 200 m, obtaining the average residential housing price and number of households from the GeoQ app, developed by Beijing GISUNI Information Technology Co., Ltd. (Beijing, China). These were used as classification criteria for the subsequent logit regression analysis.

3.4. Ordered Logistic Regression Model

The study employed an ordered logistic regression model, which is a statistical method suitable for analyzing ordinal (from low to high) dependent variables. In our research, the dependent variables were the daily crowd density and nighttime activity level, with $k$ categories. The model can thus be expressed by Equations (1)–(3), as follows:

$$\mathrm{l}\mathrm{o}\mathrm{g}\mathrm{i}\mathrm{t}\left(p_1\right)=\mathrm{l}\mathrm{o}\mathrm{g}\frac{p_1}{1-p_1}={\alpha }_1+\beta x$$

(1)

$$\mathrm{l}\mathrm{o}\mathrm{g}\mathrm{i}\mathrm{t}\left(p_1+p_2\right)=\log \frac{p_1+p_2}{1-p_1-p_2}={\alpha }_2+\beta x$$

(2)

$$\mathrm{l}\mathrm{o}\mathrm{g}\mathrm{i}\mathrm{t}\left(p_1+p_2+\dots +p_k\right)=\mathrm{l}\mathrm{o}\mathrm{g}\frac{p_1+p_2+\dots +p_k}{1-p_1-p_2-\dots -p_k}={\alpha }_k+\beta x$$

(3)

and $p_1+p_2+\dots +p_k=1$. When $k=2$, $p_2=1-p_1$, and the model is reduced to only Equation (1), and it is simply called the logit regression model. In our regression, we let 1 indicate values below the median of the standardized regional data, and 2 indicate values above the median of the standardized regional data. Then, $p_1$ and $p_2$ denote the proportion of values below and above the median.

The logit regression model was used to explore the impacts of the built environment indicators across four historical and cultural blocks in Suzhou and Haikou on their respective neighborhood vitalities. The study sought to uncover the influencing mechanisms using the average crowd density at different time points from the LBS and NTL levels as dependent variables to assess street vitality at multiple levels. Socioeconomic factors (e.g., number of households and other relevant economic variables) were integrated as classification variables. These variables were introduced to explain any variation in the block vitality that was not related to the built environment indicators. In subsequent sections, we check the robustness of the regression model.

4. Main Results and Discussions

4.1. Descriptive Statistics and Correlation Analysis Results

Because of the inclusion of multiple data sources for the variables in this study, such as road vectors, POIs, and street view images, with different processing methods and incomparable magnitudes and units, we first standardize the raw data obtained from them. The calculation formula is as follows:

$$x_{\mathrm{s}\mathrm{t}\mathrm{a}\mathrm{n}\mathrm{d}\mathrm{a}\mathrm{r}\mathrm{d}\mathrm{i}\mathrm{z}\mathrm{e}}=\frac{x-x_{\mathrm{m}\mathrm{i}\mathrm{n}}}{x_{\mathrm{m}\mathrm{a}\mathrm{x}}-x_{\mathrm{m}\mathrm{i}\mathrm{n}}}$$

(4)

The indicators across different districts are summarized in Figure 5. First, the performances of the dependent variables are portrayed on the basis of the LBS and NTL data influenced by built environment indicators in different cities and blocks. Second, the varied performances of each independent variable across four districts in two cities are demonstrated, categorized on the basis of the transport systems, function feature, and spatial quality dimensions. Last, the analysis presents the potential role of classification variables.

Table 2 presents the summary statistics of the built environment variables for the researched blocks in Haikou and Suzhou. As a government-designated pilot city, Suzhou has a larger urban area, higher economic development level, and higher population density compared to Haikou. The mean results indicate that the most significant differences between the indicators in the blocks of the two cities are the proportion of cultural facilities, shop density, and street aspect ratio. The proportion of cultural facilities in Haikou’s blocks is more concentrated at low values compared to Suzhou, with a smaller standard deviation. This reflects that Suzhou has better local cultural publicity and display, while the cultural facilities in Haikou streets are generally insufficient.

However, the commercial density of the streets in Haikou is higher than that in Suzhou, indicating that there are more types of shops rather than cultural facilities on the crowded streets in Haikou. This indicates a greater business development intensity and better living convenience on a block scale in the smaller city of Haikou. In terms of the street WH ratio, Suzhou has a higher average after standardization, possibly due to protection policies in Suzhou’s ancient city area that limit building heights in cultural blocks. This may make the streets in Suzhou appear more spacious but could potentially undermine street vitality. On the other hand, the lower WH ratio of the streets in Haikou reflects a potentially more depressing environment. Although this could promote the vigor of the blocks, it can also cause discomfort and safety issues in the visual field. The spatial distribution differences of the indicators in these two cities could lead to variations in neighborhood vitality.

At this stage, we simultaneously conducted a multicollinearity analysis and examined the correlation matrix to assess the reliability of the model’s control variables. Typically, multicollinearity is a concern if the tolerance (TOL) is less than 0.1 or the variance inflation factor (VIF) exceeds 10. In this study, any variable with a VIF value greater than 5 would be excluded from the model. The VIF examinations (Figure 6) indicate that all independent variables included have a value below 5; meanwhile, the TOL of all variables is above 0.28. Therefore, the influence of multicollinearity could be negligible for the independent variables in our research data. Moreover, with the variables not following a normal distribution, the Spearman’s correlation matrix (Figure 7) reveals that the correlation coefficients between any two independent variables are less than 0.8. They represent various dimensions of objective performance within the cultural block, thus serving effectively as covariates.

A higher street building enclosure often corresponds to a greater minimum distance to parks and green spaces. Moreover, a larger proportion of living and public services, coupled with convenient access to transportation facilities, typically correlates with a higher proportion of cultural facilities. This offers residents a more diverse array of cultural activities and entertainment options. Simultaneously, better car and pedestrian friendliness, or poorer accessibility to green spaces, could influence the proportion of cultural facilities, potentially resulting in a negative impact on residents’ quality of life in the neighborhood. Despite a positive correlation, the diversity and complexity of urban planning for block renaissance should be thoroughly considered.

4.2. Baseline Logistic Regression for Block Vitality

This paper employed a logit regression model to investigate the impact of three types of factors (i.e., transport system, function feature, and spatial quality) on the vitality of cultural blocks. We categorized block vitality into the following two types: daytime block vitality (assessed with heat maps depicting pedestrian activity at 10 am, 2 pm, and 6 pm) and nighttime block vitality (evaluated using NTL data). The objective was to comprehensively explore which factors effectively contribute to the vibrancy of streets and surrounding blocks, thereby enhancing diversity, consumerism, and inclusivity in the area.

The first columns in

Table 3. Logit regression of daytime block vitality at 10 am.

Domain	Independent Variables	Overall	Haikou	Suzhou	High Population	Low Population	High Land Value	Low Land Value
Transport System	Road grade	0.27 (0.27)	−3.88 (−1.55)	2.02 (1.07)	0.19 (0.14)	3.96 (1.59)	−6.11 *** (−2.91)	1.13 (0.53)
	Traffic network integration	−1.29 ** (−2.15)	2.81 *** (2.67)	−0.37 (−0.24)	−2.33 *** (−2.64)	1.45 (1.32)	−3.06 ** (−2.20)	1.15 (1.12)
	Minimum distance to green space	1.34 ** (2.38)	4.31 *** (4.14)	−0.52 (−0.46)	0.24 (0.27)	2.60 *** (2.72)	0.35 (0.32)	0.60 (0.56)
	Minimum distance to public transport	−0.31 (−0.52)	−1.78 * (−1.87)	0.49 (0.40)	0.63 (0.76)	−2.79 ** (−2.42)	−2.26 ** (−2.41)	0.26 (0.22)
Function Feature	Degree of business aggregation	2.97 *** (3.63)	2.05 (1.44)	1.10 (0.79)	3.79 *** (3.28)	2.44 * (1.76)	1.71 (1.39)	3.12 * (1.72)
	Proportion: living + public services	0.26 (0.35)	−1.88 (−1.34)	0.99 (0.75)	−1.02 (−0.83)	1.24 (1.01)	1.85 (1.58)	−1.98 (−1.44)
	Proportion: catering facilities	2.71 *** (4.24)	−0.65 (−0.66)	4.66 *** (3.10)	3.01 *** (2.97)	2.05 * (1.93)	3.94 *** (3.32)	1.90 * (1.71)
	Proportion: cultural facilities	−0.96 (−1.20)	−2.80 (−1.10)	−2.06 (−1.29)	0.87 (0.66)	−3.02 ** (−2.11)	0.95 (0.66)	−5.66 *** (−3.44)
	Commercial density	1.55 ** (2.55)	−1.08 (−1.19)	8.64 *** (3.26)	1.53 * (1.80)	2.91 ** (2.48)	1.86 * (1.71)	−1.74 (−1.52)
Spatial Quality	WH ratio	−0.23 (−0.24)	4.45 (1.51)	−4.79** (−2.51)	0.01 (0.00)	−3.78 (−1.58)	3.36** (2.06)	−0.57 (−0.24)
	Greenery	−0.74 (−1.00)	−0.69 (−0.62)	−0.14 (−0.08)	−1.49 (−1.11)	−0.87 (−0.70)	−0.56 (−0.45)	−1.46 (−1.11)
	Sky openness	−1.67 ** (−2.19)	−2.25 * (−1.83)	−1.26 (−0.72)	−2.55 * (−1.83)	−0.22 (−0.20)	−1.90 (−1.33)	−1.66 (−1.38)
	Building enclosure	0.56 (0.80)	0.04 (0.04)	0.58 (0.36)	0.11 (0.08)	−0.15 (−0.15)	1.66 (1.48)	−0.98 (−0.81)
	Car and pedestrian friendliness	1.27 * (1.91)	−0.40 (−0.35)	2.05 (1.62)	2.24 ** (2.50)	−1.30 (−1.01)	2.05 * (1.94)	−0.73 (−0.59)
Constant		−2.78 *** (−2.74)	−1.26 (−0.92)	−2.52 (−1.09)	−2.12 (−1.11)	−3.35 ** (−2.26)	−2.44 (−1.64)	0.92 (0.44)

t-Statistics are in parentheses. *** p < 0.01, ** p < 0.05, and * p < 0.1. WH stands for width to height. The sample size is 311. Results are obtained via Stata 18.0.

,

Table 4. Logit regression of daytime block vitality at 2 pm.

Domain	Independent Variables	Overall	Haikou	Suzhou	High Population	Low Population	High Land Value	Low Land Value
Transport System	Road grade	−0.79 (−0.70)	−0.50 (−0.21)	1.99 (0.97)	−0.46 (−0.31)	2.81 (1.11)	−7.90 *** (−3.36)	1.95 (0.89)
	Traffic network integration	−0.21 (−0.34)	0.91 (0.94)	2.31 (1.43)	−1.30 (−1.45)	1.93 * (1.65)	−1.03 (−0.77)	1.17 (1.11)
	Minimum distance to green space	1.88 *** (3.19)	3.28 *** (3.37)	0.56 (0.46)	1.24 (1.35)	3.34 *** (3.17)	1.24 (1.10)	0.72 (0.67)
	Minimum distance to public transport	−0.29 (−0.48)	−1.87 * (−1.86)	2.06 * (1.74)	−0.40 (−0.47)	−1.03 (−0.93)	−1.92 ** (−2.06)	0.57 (0.46)
Function Feature	Degree of business aggregation	0.03 (0.04)	−1.07 (−0.71)	−1.89 (−1.44)	−0.63 (−0.61)	0.87 (0.61)	−1.23 (−1.04)	−1.29 (−0.75)
	Proportion: living + public services	−0.48 (−0.59)	2.34 * (1.67)	−3.33 ** (−2.39)	−1.01 (−0.79)	0.04 (0.03)	1.17 (0.95)	−2.15 (−1.32)
	Proportion: catering facilities	2.09 *** (3.21)	−1.12 (−1.14)	2.57 (1.60)	2.33 ** (2.35)	2.34 ** (2.03)	3.92 *** (3.20)	1.17 (1.00)
	Proportion: cultural facilities	−0.20 (−0.25)	1.59 (0.81)	−4.43 ** (−2.48)	−0.18 (−0.14)	−0.27 (−0.19)	0.45 (0.35)	−1.06 (−0.73)
	Commercial density	3.33 *** (5.00)	2.80 *** (2.95)	7.83 *** (2.86)	2.61 *** (3.01)	5.38 *** (4.17)	0.89 (0.86)	4.32 *** (3.01)
Spatial Quality	WH ratio	2.97 *** (2.71)	2.04 (0.77)	−0.19 (−0.10)	4.50 *** (2.81)	−1.39 (−0.57)	7.93 *** (3.88)	0.43 (0.18)
	Greenery	0.35 (0.45)	1.46 (1.33)	2.12 (1.16)	0.27 (0.20)	0.05 (0.03)	−0.21 (−0.17)	1.57 (1.06)
	Sky openness	−1.65 ** (−2.06)	−2.92 ** (−2.16)	−0.80 (−0.45)	−2.23 (−1.60)	−0.60 (−0.49)	−3.76 ** (−2.53)	−1.13 (−0.87)
	Building enclosure	0.74 (1.00)	2.10 * (1.95)	0.75 (0.46)	0.67 (0.49)	0.30 (0.27)	1.55 (1.35)	−0.50 (−0.37)
	Car and pedestrian friendliness	0.83 (1.18)	0.11 (0.10)	1.11 (0.95)	1.78 * (1.92)	−1.46 (−1.09)	1.01 (0.93)	0.33 (0.26)
Constant		−3.15 *** (−2.93)	−2.72 * (−1.85)	−2.49 (−1.08)	−1.93 (−1.01)	−5.67 *** (−3.31)	−2.03 (−1.34)	−1.22 (−0.53)

t-Statistics are in parentheses. *** p < 0.01, ** p < 0.05, and * p < 0.1. WH stands for width to height. The sample size is 311. Results are obtained via Stata 18.0.

,

Table 5. Logit regression of daytime block vitality at 6 pm.

Domain	Independent Variables	Overall	Haikou	Suzhou	High Population	Low Population	High Land Value	Low Land Value
Transport System	Road grade	−1.89 (−1.50)	−1.45 (−0.59)	−4.34 (−1.12)	−2.72 (−1.39)	0.92 (0.36)	−10.14 *** (−3.76)	−0.76 (−0.36)
	Traffic network integration	−0.18 (−0.27)	1.85 * (1.85)	−1.81 (−0.77)	0.31 (0.33)	−0.24 (−0.19)	−1.02 (−0.71)	1.05 (1.04)
	Minimum distance to green space	2.23 *** (3.51)	3.35 *** (3.43)	4.86 ** (2.35)	2.20 ** (2.21)	2.65 ** (2.49)	2.62 ** (2.05)	0.12 (0.12)
	Minimum distance to public transport	−0.23 (−0.35)	−1.34 (−1.34)	2.20 (1.27)	−0.67 (−0.72)	0.08 (0.07)	−2.09 * (−1.93)	0.63 (0.53)
Function Feature	Degree of business aggregation	1.69 * (1.95)	0.66 (0.44)	0.08 (0.05)	0.17 (0.15)	6.07 *** (3.16)	2.42 * (1.77)	−0.51 (−0.31)
	Proportion: living + public services	−1.77 * (−1.90)	1.35 (0.95)	−10.04 *** (−3.79)	−3.20 ** (−2.16)	−0.49 (−0.32)	−2.07 (−1.38)	−1.77 (−1.11)
	Proportion: catering facilities	3.09 *** (4.25)	−0.77 (−0.78)	3.51 (1.38)	4.00 *** (3.55)	2.13 * (1.85)	5.85 *** (4.03)	1.27 (1.16)
	Proportion: cultural facilities	−0.16 (−0.19)	1.17 (0.59)	−8.08 *** (−2.77)	0.14 (0.11)	−1.17 (−0.82)	0.74 (0.51)	−1.54 (−1.08)
	Commercial density	4.19 *** (5.75)	2.74 *** (2.89)	22.24 *** (3.39)	3.37 *** (3.53)	7.04 *** (4.40)	3.12 *** (2.67)	3.69 *** (2.79)
Spatial Quality	WH ratio	3.66 *** (2.99)	3.11 (1.12)	4.54 (1.39)	5.47 *** (2.95)	0.22 (0.09)	8.35 *** (3.67)	2.93 (1.22)
	Greenery	−0.16 (−0.18)	−0.70 (−0.63)	6.66 ** (2.52)	0.01 (0.00)	0.90 (0.64)	0.34 (0.23)	−0.20 (−0.15)
	Sky openness	−2.22 ** (−2.55)	−2.70 ** (−2.06)	0.90 (0.33)	−2.95 * (−1.87)	−0.59 (−0.46)	−5.83 *** (−3.00)	−1.11 (−0.88)
	Building enclosure	0.90 (1.11)	0.72 (0.68)	4.65 ** (2.09)	0.75 (0.50)	1.36 (1.11)	2.04 (1.47)	−0.55 (−0.42)
	Car and pedestrian friendliness	0.78 (1.04)	−0.14 (−0.12)	0.32 (0.21)	2.17 ** (2.10)	−2.07 (−1.48)	1.09 (0.87)	−0.16 (−0.13)
Constant		−4.43 *** (−3.66)	−2.98 ** (−2.01)	−7.61 ** (−2.34)	−3.90 * (−1.82)	−7.77 *** (−3.75)	−4.77 ** (−2.53)	−0.70 (−0.32)

t-Statistics are in parentheses. *** p < 0.01, ** p < 0.05, and * p < 0.1. WH stands for width to height. The sample size is 311. Results are obtained via Stata 18.0.

and

Table 6. Logit regression of nighttime block vitality.

Domain	Independent Variables	Overall	Haikou	Suzhou	High Population	Low Population	High Land Value	Low Land Value
Transport System	Road grade	−3.09 *** (−2.75)	6.94 ** (2.42)	−14.00 *** (−3.17)	−3.05 ** (−2.01)	−2.33 (−0.89)	4.68 * (1.67)	−10.01 *** (−3.34)
	Traffic network integration	0.59 (0.98)	−3.13 *** (−2.96)	3.24 ** (2.03)	1.56 * (1.69)	0.04 (0.04)	−1.32 (−0.89)	0.08 (0.09)
	Minimum distance to green space	−0.50 (−0.87)	−1.36 (−1.36)	−0.33 (−0.27)	0.82 (0.87)	−1.31 (−1.31)	−1.10 (−0.85)	2.41 ** (2.11)
	Minimum distance to public transport	−0.36 (−0.61)	0.28 (0.29)	−1.36 (−1.20)	−1.19 (−1.40)	−0.08 (−0.07)	−1.37 (−1.47)	0.93 (0.82)
Function Feature	Degree of business aggregation	3.59 *** (4.25)	4.99 *** (3.34)	4.35 *** (3.00)	2.51 ** (2.28)	8.86 *** (3.99)	5.08 *** (3.55)	3.34 * (1.94)
	Proportion: living + public services	1.43 * (1.89)	2.34 (1.59)	3.51 ** (2.50)	1.29 (0.91)	3.08 ** (2.23)	2.67 ** (2.17)	1.63 (1.12)
	Proportion: catering facilities	−0.98 (−1.58)	0.13 (0.13)	1.00 (0.74)	−0.34 (−0.35)	−1.16 (−1.03)	0.44 (0.41)	−1.78 * (−1.68)
	Proportion: cultural facilities	−0.20 (−0.25)	3.10 (1.27)	0.56 (0.45)	−0.97 (−0.76)	1.61 (1.26)	2.01 (1.42)	−0.16 (−0.11)
	Commercial density	2.06 *** (3.24)	2.80 *** (2.73)	2.77 (1.33)	2.00 ** (2.18)	3.17 ** (2.48)	2.76 ** (2.20)	3.19 *** (2.68)
Spatial Quality	WH ratio	3.11 *** (2.84)	−8.16 ** (−2.44)	8.64 *** (3.20)	2.58 * (1.73)	3.25 (1.33)	−1.15 (−0.62)	10.49 *** (3.48)
	Greenery	3.08 *** (3.82)	2.69 ** (2.30)	5.25 *** (2.91)	4.60 *** (3.07)	2.39 * (1.77)	3.90 *** (3.04)	2.40 * (1.72)
	Sky openness	−1.66 ** (−2.17)	−0.93 (−0.74)	1.32 (0.72)	0.16 (0.12)	−4.14 *** (−3.07)	0.04 (0.03)	−2.36 * (−1.78)
	Building enclosure	1.46 ** (2.14)	0.83 (0.81)	7.66 *** (3.88)	3.50 ** (2.44)	0.92 (0.88)	3.63 *** (2.91)	−1.35 (−1.19)
	Car and pedestrian friendliness	0.31 (0.45)	0.13 (0.11)	0.48 (0.37)	0.28 (0.29)	0.29 (0.20)	−1.60 (−1.35)	0.83 (0.67)
Constant		−3.33 *** (−3.36)	−2.78 ** (−1.98)	−11.19 *** (−4.17)	−5.20 ** (−2.57)	−6.19 *** (−3.37)	−4.98 *** (−3.16)	−3.68 * (−1.84)

t-Statistics are in parentheses. *** p < 0.01, ** p < 0.05, and * p < 0.1. WH stands for width to height. The sample size is 311. Results are obtained via Stata 18.0.

present the baseline logistic regression model estimates for block vitality. According to the results, in terms of functional feature, a higher degree of business aggregation and commercial density in blocks leads to increased vitality. This conclusion is quite comprehensible due to the diverse range of businesses within a block, which attracts more consumers to engage in shopping, dining, or entertainment activities, thereby increasing foot traffic and opportunities for social interactions. Simultaneously, a higher commercial density also tends to generate more economic activities and employment opportunities in the area, thus promoting economic development. From the perspective of spatial quality, increasing the WH ratio of streets within a reasonable range positively impacts block vitality. This phenomenon can be attributed to the presence of characteristics such as openness, brightness, good ventilation, easy orientation, and ample parking facilities. These features are more likely to promote good neighborliness and enrich social network resources. As a result, they attract numerous consumers for shopping and entertainment purposes compared to isolated group activity patterns formed in closed, narrow, and/or disorderly layouts. Meanwhile, it is observed that increased sky openness adversely affects the all-day vibrancy of blocks. In other words, higher levels of sky openness correspond to lower levels of block activity level. This is possibly due to the fact that higher sky openness often indicates a lack of commercial activities and cultural entertainment facilities within blocks. When there are not enough places along its streets offering people chances to participate or consume, the entire block tends to appear deserted.

Further explorations reveal a positive correlation between the minimum distance to green space and the daytime vitality of blocks, which holds significant importance. The presence of green spaces farther away from streets creates a natural transitional zone that effectively satisfies people’s needs for natural landscapes, healthy lifestyles, and social interactions. This ultimately enhances residents’ quality of life and promotes sustainable development. Additionally, a higher proportion of catering facilities is correlated with greater block vitality during daytime hours, highlighting the growing demand for food and social venues among individuals. As urbanization progresses and living standards improve, people’s expectations for dining experiences and social interactions are steadily rising. Consequently, an abundant array of high-quality restaurants, cafes, and other dining establishments within a block will attract more foot traffic and contribute to heightened levels of daytime vibrancy.

Regarding the nighttime vitality in blocks, road grade has a detrimental impact. As a road moves up the hierarchy, the nighttime vitality of relevant blocks decreases. This is because higher-grade roads attract more traffic, which can result in noises and air pollution during nighttime hours, causing inconvenience and discomfort for nearby residents. Consequently, there might be restrictions on establishing residential or commercial facilities along roads of high grades, leading to a decline in nighttime liveliness around that area. Furthermore, unlike daytime vitality, the proportion of living and public services plays a pivotal role in enhancing nighttime vitality. During nighttime hours, people’s needs and behavioral patterns shift toward seeking entertainment venues, leisure activities, and social gatherings. Therefore, having abundant public service facilities within a block effectively enhances its nighttime vitality. Lastly, two elements related to spatial quality, greenery and building enclosure, positively influence nocturnal liveliness. The higher the streets’ greenery and building enclosure, the greater their promotion of nighttime vitality within relevant areas. This inspires governments to establish more public green spaces while preserving natural landscapes and encourages businesses to open up on to streets to create a vibrant and interactive urban atmosphere.

4.3. Heterogeneity Analysis by Stratifying Cultural Blocks

Considering that inherent attributes of blocks significantly influence the vitality to a certain extent, we examined the following three moderators: geographic location, resident population, and land value. Stratified analyses were conducted to explore variations in the influencing factors of block vitality based on different attributes. This facilitates targeted analysis and enables the development of corresponding strategies tailored to specific circumstances in order to promote sustained healthy growth in such areas. The results are presented in the last six columns of Table 3, Table 4, Table 5 and Table 6.

4.3.1. Geographic Location

The results demonstrate that a high degree of business aggregation positively impacts the daytime vitality of both cities’ blocks, with Suzhou experiencing a larger effect. In Suzhou, the presence of numerous commercial establishments contributes to a thriving city center, attracting a multitude of tourists and consumers seeking shopping, dining, and entertainment options. It not only stimulates local economic growth but also enhances the city’s overall image and reputation. The traffic network integration has a positive impact on the daytime vitality of Haikou’s blocks. However, in Suzhou, there is a lack of significant influence in block vitality during the day due to traffic network integration. This disparity can be attributed to variations in developmental trajectories and underlying conditions between these two cities, as follows: Haikou, as an emerging city, has made rapid improvements to its local road network, public transportation, as well as rapid transit systems through vigorous integration and optimization efforts in recent years; whereas Suzhou, being historically rich and economically developed, has long prioritized infrastructure development leading to a highly matured and efficient transportation system. In terms of function feature, the proportion of cultural facilities in Suzhou district has a detrimental impact on enhancing daytime vitality. The higher the proportion of cultural facilities, the lower the daytime vitality of the district. This contradicts our understanding and may be attributed to the growing demand for leisure and entertainment as cities develop and economies prosper, resulting in an increased emphasis on cultural facility construction by local governments. However, the excessive pursuit of cultural facility construction often leads to the neglect of other essential supporting services, thereby limiting people’s mobility and time spent within this district. Concentrating on spatial quality, sky openness negatively affects the daytime vitality of blocks in Haikou. With an increase in the degree of sky openness, the daytime vitality level in Haikou’s blocks decreases. This can be attributed to the fact that excessive sky openness results in a lack of shading and physical barriers between buildings. Under Haikou’s hot and sunny climate, the absence of shaded areas discourages pedestrians from staying or walking outdoor, thereby suppressing the daytime vitality level of blocks. However, no significant impact of sky openness on daytime vitality was observed in Suzhou, indicating regional differences in its influence.

Traffic network integration has a detrimental impact on the nighttime vitality of Haikou, thereby hindering its potential as a tourist destination in the dark. This can be attributed to the significant traffic congestion experienced in well-connected blocks during evening hours, which impedes pedestrian flow and restricts residents from engaging in leisure activities such as nighttime strolls, shopping, and enjoying local cuisine. In contrast, Suzhou’s blocks exemplify how harmonious alignment between urban traffic networks and ancient fabric injects new vitality into its nighttime economy. Furthermore, it is observed that both degree of business aggregation and greenery contribute to nighttime vitality in both Suzhou’s and Haikou’s blocks. Consequently, this suggests that governments can optimize business structures to promote healthy competition among merchants while simultaneously incorporating plant landscapes into public spaces to create pleasant environmental atmospheres. However, the presence of excessive public service facilities (e.g., pharmacies, telecommunication offices, and government institutions) in Suzhou’s blocks tends to dampen the nighttime vitality of the area. This is primarily due to their substantial occupation of land resources and limited nighttime visitation or closure. As a result, the availability of commercial establishments and entertainment venues within a block diminishes accordingly, thereby restricting options for evening activities. Nevertheless, this phenomenon does not apply to Haikou, indicating that Haikou has effectively achieved a more harmonious equilibrium between public service facilities and commercial entertainment venues through meticulous planning and management. Thus, it has successfully cultivated a captivating and dynamic nightlife ambiance.

4.3.2. Resident Population

The classification of blocks can also be based on the number of resident households in the surrounding area. In terms of functional feature, our findings indicate that both blocks with high and low resident populations experience significant positive impacts on daytime vitality from a higher proportion of catering facilities and increased commercial density. Nevertheless, the influence of catering facilities on daytime vitality is more pronounced in blocks with a larger population. This might be attributed to their greater demand for diverse restaurants, cafes, etc. On the other hand, commercial density has a more prominent effect on daytime vitality in blocks with fewer permanent households. High commercial density fills residents’ daily necessities and plays a crucial role in activating social life while improving quality of life. In addition, car and pedestrian friendliness has positive impacts on the daytime liveliness of blocks with more resident households nearby. These areas tend to attract more businesses. Businesses highly value customer flow and convenience, preferring locations that can meet consumer demands while offering ample parking facilities and pedestrian accessibility. This further stimulates economic, cultural, and social development within blocks. However, this relationship is not as apparent in blocks with fewer residents, necessitating flexible adjustments by governments to strike a balance based on actual circumstances. Additionally, it is interesting to note that in blocks with a smaller resident population there appears to be a positive relationship between the minimum distance to green space and daytime vitality. This phenomenon could potentially be attributed to the environmental benefits associated with green spaces. Typically designed as serene and tranquil areas, these green spaces offer a peaceful retreat from noise and stress, thereby creating a more serene and comfortable atmosphere within the surrounding blocks.

For enhancing nighttime vitality, the degree of business aggregation, commercial density, and greenery all exhibit a significant positive impact, irrespective of the number of resident households within the blocks. Further, examining experimental data reveals that, compared to blocks with more resident households, the degree of business aggregation and commercial density exert a more pronounced influence on nighttime vitality in blocks with fewer residents. This suggests that prioritizing the development of suitable specialized economic models and improving the quality of the business environment is crucial in areas with lower resident populations. Only by striking a balance between supplying high-quality goods and services and creating a comfortable and convenient consumer environment can genuine nighttime vitality be stimulated, attracting more external mobile consumers and thereby promoting sustained and steady economic development in the region. Additionally, the positive impact of greenery on the nighttime vitality of densely populated blocks surpasses that on less populated ones. This suggests that areas with a higher permanent population exhibit relatively elevated levels of nighttime activities, making it crucial for these crowded regions to have adequate green spaces to alleviate stress, provide recreational opportunities, and enhance the environment. Increased greenery can significantly contribute to the vitality and overall quality of life in these blocks. Moreover, interestingly, we find a detrimental impact of road grade on the nighttime vitality in densely populated blocks. As the road grade increases, there is a gradual decline in the nighttime vitality within these blocks. This signifies that higher levels of road grade can lead to amplified traffic flow, sound pollution, and environmental contamination. These adverse effects not only disrupt residents’ daily lives but also impede the cultivation of a pleasant nocturnal ambiance. Consequently, urban managers face novel challenges in promoting block vitality while ensuring seamless traffic management and safety measures are upheld. Lastly, sky openness exerts a certain negative influence on streets with fewer permanent households during nighttime hours. Issues such as interference, privacy deprivation, and noise and light pollution necessitate attention from relevant authorities through appropriate interventions.

4.3.3. Land Value

The study also categorizes blocks into two distinct groups based on land value, as follows: those with high (i.e., above-median) residential housing prices and those with low (i.e., below-median) residential house values. This was to further examine the disparities in the influencing factors on their daytime and nighttime vitality. Regarding daytime vitality, the role of commercial density remains unaffected by the residential housing price classification, exerting a profound impact on both high-value and low-value blocks. However, in terms of the magnitude of influence, commercial density plays a more positive role in enhancing daytime vitality for low-value blocks. Even though highly attractive commercial facilities may be lacking, effectively increasing the number and variety of small retail stores, restaurants, or service industry businesses would significantly enhance daytime activities in these areas. By improving convenience and meeting residents’ basic needs in low-value blocks, a subsequent enhancement in land value can be achieved. Therefore, it is crucial to fully consider the characteristics of different blocks during urban development and implement corresponding measures to enhance regional diversity, consumption capacity, and inclusiveness for comprehensive sustainable development goals. The daytime block vitality is adversely impacted by road grade, particularly for high-value land, thereby diminishing the attractiveness. To enhance the daytime vitality and agglomeration effect of these blocks, it is imperative to carefully balance the relationship between the road grade and the surrounding commercial facilities. Moreover, the proximity to public transit confers advantages in terms of aggregating vitality for high-value land within blocks, underscoring the necessity for governments to strategically plan the layout of mass transit stops. By integrating scientific planning and management measures, a harmonious equilibrium can be achieved between optimizing urban spatial resource allocation and sustainable development objectives. In terms of spatial quality, several noteworthy findings emerge. The WH ratio positively impacts the daytime vitality of high-value blocks. Nevertheless, excessive sky openness can potentially diminish their vitality during daylight hours. This is due to the risk of overexposing the internal environment, compromising privacy, increasing optical radiative pollution, and impeding social connections within blocks, as well as influencing commercial development and cultural atmosphere. Conversely, there are few significant factors that affect the daytime vitality of low-value blocks. Further exploration into characteristics such as building types, sizes, functions, population structure, consumption habits, and cultural backgrounds is warranted to comprehensively comprehend operational mechanisms and future development trends for blocks with low land value.

In terms of the nighttime vitality of blocks, the degree of business aggregation, commercial density, and the greenery consistently play a positive role. Irrespective of land value, these factors synergistically contribute to promoting nighttime development of blocks. By optimizing the industrial structure, increasing the merchant presence, and prioritizing urban ecological construction measures, we can further unleash the potential of the nighttime economy within blocks and create a more prosperous, secure, and livable urban environment. However, the impact of road grades on different blocks varies. Higher-grade roads contribute to the development of nighttime vitality in blocks with higher land values, attracting more commercial and entertainment facilities, increasing nighttime activities. These blocks typically possess well-designed, clean, and efficient transportation networks that enhance travel experiences for both residents and tourists alike. Conversely, higher-grade roads may hinder the growth of nighttime vitality in blocks with low land values. Therefore, it is important to promote a balanced expansion of nighttime vitality throughout the city by appropriately adjusting road grades, optimizing public transport systems, and fostering diversified economic development to ensure an enhanced quality of life and experience for all residents and visitors. Additionally, in order to enhance the nighttime vitality of areas with low land values, it is crucial to accentuate their proximity to green spaces. When these low-value lands are situated near parks and verdant areas, the suppression of block vitality during nighttime may be attributed to a dearth of nearby recreational amenities and commercial services, as well as inadequate development. Hence, when contemplating strategies for augmenting the nighttime vitality of low-value blocks, integration with the surrounding environment becomes imperative by providing top-notch supporting facilities, improving a culturally enriched ambiance, and enhancing dynamic social interactions.

4.4. Robustness Check

To further validate the main regression results, this study included three classification variables as additional independent variables (i.e., control variables) in the logit regression. This common practice in empirical research aims to examine the sensitivity of the regression coefficients of the main built environment factors to additional controls. Consistent coefficients are typically interpreted as evidence of structural validity [65]. As indicated in

Table 7. Logit regression with moderators as controls.

Domain	Independent Variables	10 am		2 pm		6 pm		Nighttime
		Coef.	t	Coef.	t	Coef.	t	Coef.	t
Transport System	Road grade	−1.26	−1.03	−1.96	−1.44	−3.04 **	−2.00	−1.91	−1.44
	Traffic network integration	−0.31	−0.47	0.93	−1.34	0.64	−0.89	0.15	−0.22
	Min. distance to green space	−0.23	−0.35	0.44	−0.64	0.84	−1.16	0.95	−1.36
	Min. distance to public transport	−0.30	−0.47	−0.12	−0.18	−0.01	−0.02	−0.65	−1.03
Function Feature	Degree of business aggregation	1.82 **	−2.10	−1.57 *	−1.80	0.34	−0.37	4.69 ***	−4.87
	Proportion: living + public services	−0.18	−0.21	−1.17	−1.34	−2.35 **	−2.44	1.99 **	−2.32
	Proportion: catering facilities	2.11 ***	−3.00	1.30 *	−1.79	2.19 ***	−2.78	−0.10	−0.15
	Proportion: cultural facilities	−3.01 ***	−2.66	−2.30 **	−2.09	−2.40 **	−2.07	1.71 *	−1.68
	Commercial density	1.13	−1.62	3.31 ***	−4.16	4.43 ***	−5.19	2.20 ***	−2.84
Spatial Quality	WH ratio	1.06	−0.89	4.25 ***	−3.12	4.85 ***	−3.23	1.99	−1.58
	Greenery	−0.62	−0.73	0.55	−0.62	0.55	−0.59	2.24 ***	−2.66
	Sky openness	−2.28 ***	−2.69	−2.37 ***	−2.67	−2.75 ***	−2.93	−2.09 **	−2.50
	Building enclosure	0.48	−0.63	0.68	−0.84	1.20	−1.40	1.12	−1.53
	Car and pedestrian friendliness	1.706 **	−2.43	1.08	−1.46	0.92	−1.18	0.30	−0.40
Moderators	Regional affiliation	−1.01 **	−1.99	−1.33 **	−2.48	−1.40 **	−2.49	0.93 *	−1.9
	Resident population	0.13	−0.42	0.54	−1.63	−0.13	−0.39	1.05 ***	−3.36
	Land value	−1.78 ***	−4.93	−1.86 ***	−4.85	−1.42 ***	−3.66	0.97 ***	−2.64
Constant		0.35	−0.28	−0.01	−0.01	−1.46	−1.06	−6.09 ***	−4.73

The t stands for t-statistics. *** p < 0.01, ** p < 0.05, and * p < 0.1. WH stands for width to height. The sample size is 311. Results are obtained via Stata 18.0.

, the trends and correlations obtained are generally consistent with the main results, strengthening the conclusions of this paper.

5. Conclusions

5.1. New Insights and Limitations

This study aimed to explore the correlations between indicators of block vitality and built environment in order to provide a more comprehensive understanding of vitality promotion. Additionally, we sought to investigate the heterogeneous impacts of built environment factors on vitality in tourism cities in different stages of development, aiming to uncover the inherent potential characteristics of cultural blocks in diverse regions and offer innovative perspectives and recommendations for optimizing urban environments and promoting sustainable development. The present study employed NTL and LBS data from four districts to construct dependent variables and further analyzed built environments using the block indicators of transport system, function feature, and spatial quality. A logit regression model was utilized in this study, and through the segmentation of various streets based on geographic location, resident population, and land value, our objective was to examine disparities in block vitality and its associated factors.

The main findings suggest that the traffic network integration has a positive impact on daytime vitality of Haikou and nighttime vitality of Suzhou, while it adversely affects the nighttime vitality of Haikou’s cultural blocks. This implies the necessity of exploring the relationship between built environment indicators and their effects in different time periods and regions. In terms of nighttime vitality, both commercial density and greenery consistently exert a positive influence in general and in various subgroups. It is worth noting that an increase in the proportion of cultural facilities decreases daytime vitality in Suzhou, particularly among subgroups with fewer permanent residents and lower land values. Moreover, sky openness negatively impacts daytime vitality in Haikou, especially within blocks characterized by higher residential populations and land values. It also has a detrimental effect on nighttime vitality with fewer permanent residents and lower land values.

With technological advancements, there are a growing number of interactions between virtual and physical spaces, reshaping urban consumption spaces, including cultural blocks. To meet consumer demands and adhere to the aesthetic expectations of internet celebrities, many urban areas have witnessed the transformation of traditional physical spaces, consequently altering the relationships among locations. Building upon previous research on the correlation between existing built environment characteristics and vitality, this paper utilizes more comprehensive data sources and measurement techniques to analyze the factors influencing vitality. This paper yields the following insights and conclusions: (1) The Chinese cultural blocks in this study showcase the features of pedestrian-friendly daily-life scenes, and they contain numerous informal spaces. Informality is deemed a vital planning concept [66]. However, because of the greater emphasis on material space renewal and lower awareness of comprehensive regeneration goals historically, coupled with the challenges related to objectively measuring data for such districts, incomplete information has not led to a sufficient policy-making basis. This paper, thus, takes a more comprehensive approach to assess daytime (LBS) and nighttime (NTL) vitality data and analyze the mechanisms influencing the transport and function indicators related to informal spaces within blocks, alongside spatial quality indicators reflecting these informal spaces. It helps us better understand how to cultivate a warm and dynamic built environment that caters to diverse human behaviors, physical sensation, and psychological cognition, and it seeks ways to rectify biases due to past data deficiencies and shape a superior built environment through refining concepts and rules. (2) Space constitutes a multidimensional system shaped by the amalgamation of diverse factors, and the analysis of urban space is essentially a form of consumption analysis [67]. In contemporary urban life, it is increasingly common to encounter various combinations of public and private spaces, leading to the emergence of diverse hybrid spaces [68]. Consequently, intricate spatial systems, functions, and qualities have different impact mechanisms depending on regional cultures, economic conditions, and residential population demographics. Thus, there is a need to redefine specific influencing factors through multidimensional explorations and categorizations. While establishing the primary logit regression model, this study introduces a new case study by delineating two distinct cities, Suzhou and Haikou, measures land on the basis of average residential housing prices, and uses the number of households to proxy resident population conditions. We incorporate economic, demographic, and spatial intricacies to elucidate the potential diverse impacts and drivers of built environment variables across various categories, thereby deriving more nuanced conclusions by group. (3) There are limited practices of grouping and classifying multisource data on built environment variables in a single investigation. While our stratified analyses align with Guo et al.’s findings [31], we also identify scenarios in which the integration of traffic network can adversely affect nighttime vitality (e.g., in Haikou). This necessitates exploring the impacts of built environment in various time periods and regions. Of note, greater distances from streets and green parks correlate with a higher degree of daytime vitality. This may explain the negative coefficient of vegetation’s impact on vitality, as found by Jiang et al. [50], suggesting that high-quality green spaces may be quieter during the day. Nevertheless, our study demonstrates their positive influence on nighttime vitality. Perhaps studies with even more refined groupings can provide greater clarity on this relationship. (4) In historical streets within old city centers, for which obtaining comprehensive information is challenging via street view images, data gaps were filled using uniformly standardized field-survey photos. The study identifies distinct characteristics of vitality in cultural blocks associated with built environment factors across different groups. In terms of urban renaissance management, these objective methods offer empirical supports for enhancing spatial activity within blocks while also emphasizing the necessity for ongoing exploration of the long-term sustainable operation path of urban regeneration under digital empowerment.

Admittedly, there are some limitations in this paper. First, in our measurement of block vitality using multiple data sources across different time periods, we faced challenges in establishing a uniform length of sampling periods. Future research should explore integrated data sources that offer a more cohesive representation of vitality across all time periods. Second, the sample size is not large compared to other studies. In fact, our focus on the selected cultural blocks in Suzhou and Haikou may limit the generalizability of our results. Expanding the sample size of cities and blocks will enhance the robustness of the results and yield broader and more universally applicable conclusions, which should be a direction for future research. Additionally, in exploring the multidimensional impacts of built environment factors on vitality, it would be advantageous to further enhance the explanatory dimensions. Subsequent studies could draw inspiration from Hu et al. and develop clear and intuitive district renovation plans based on environmental and social benefits. Dimensions such as the landscape environment, pedestrian activities and inclusiveness, plant diversity, and all-age design could be considered [69].

5.2. Concluding Remarks

In summary, the internal spaces of cultural blocks are closely intertwined with people’s social activities and daily interactions. Although the social informality issue is reflected in urban physical spaces and persists over the long term [70], accepting the authenticity, ambiguity, and variability in informal spaces and breaking away from rigid spatial planning paradigms can effectively promote the vitality of streets and surrounding communities. This would establish vibrant blocks infused with the essence of local cultural venues, thereby enhancing regional diversity, consumption, and inclusivity.