How can the urban landscape affect urban vitality at the street block level? A case study of 15 metropolises in China

Abstract

Urban vitality, as a metric, measures the attractiveness and competitiveness of a city and is a driver of development. As the physical and social space of human activities, the urban landscape has close connections with urban vitality according to classical theories. However, limited quantitative criteria for the urban landscape and gaps between macro urban planning and micro design create difficulties when constructing a vibrant city. In this study, we quantitatively examined the relationship between the urban landscape and urban vitality at the street block level using geospatial open data to discover where, how, and to what extent we could improve urban vitality, taking 15 Chinese metropolises as a case study. Results indicate that, among the three aspects of the urban landscape considered, the city plan pattern has the highest effect on stimulating vitality, followed by the land use and the patterns of building form. Specifically, the three-dimensional form of buildings has a greater effect than a two-dimensional form. In addition, convenient transportation, a compact block form, diverse buildings, mixed land use, and high buildings are the main characteristics of vibrant blocks. The results also show that the effects of the urban landscape have spatial variations and obvious diurnal discrepancies. Furthermore, over 20 and 33% of the blocks in these cities are identified as low-vitality blocks during the day and night, respectively, and are then categorized into six different types. The identification of the common characteristics of these low-vitality blocks can be taken as references for designing a vibrant urbanity.

Keywords

Urban vitality urban landscape street blocks diurnal discrepancy China

Introduction

Since the postindustrial era, the quality of urban space and urban vitality have received substantial attention because of the decline of many large cities worldwide, as represented by the city of Salford in the United Kingdom. Good and vibrant cities tend to attract investments and high-end talent, promote rapid development, increase their competitiveness, as well as enhance the happiness of residents and achieve social sustainability (Brenner, 2014). In the absence of such character, a series of social problems, such as stagnant economic development, waste of land resources, and brain drain, can hinder the healthy and orderly development of cities (Hall and Pfeiffer, 2000; Woodworth and Wallace, 2017). The development of more vibrant, dynamic, and diverse cities is the focus of citizens, urban researchers, and managers; it is also the main goal and principle of urban planning and design.

Many urban theories, including but not limited to the networked city, compact city, new urbanism, and smart growth, suggest that many urban landscape elements, such as mixed land use and walkable streets, are basic features of a vibrant city (Craven and Wellman, 1973; Jacobs, 1961; Katz et al., 1994; Smith, 2002). The urban landscape, also termed as “townscape” is a combination of all elements that are visually perceived within the city, including both artificial and natural elements (Conzen, 1960, 2004; Zhang et al., 2019). As influenced by the geographer M.R.G. Conzen, the urban landscape, i.e. the physical form of urban areas, has been widely explored and elucidated using increasingly effective morphogenetic methods (Oliveira and Medeiros, 2016; Ye et al., 2017). The recognition of urban landscape units is the fundamental aspect of these studies and entails three components, namely the city plan, building form, and land utilization. The relationship of the three urban landscape components with one another and with the activities that occur in a particular place must be understood. The material urban space itself does not have vitality but has an impact on people’s behavior and psychology, thereby affecting urban vitality.

Scholars have explored the path of vitality improvement from the perspective of architectural and microscopic landscape design (Gehl, 1987; Marcus and Francis, 1998; Montgomery, 1998). In short, early research on urban vitality and its association with the urban landscape mainly focused on the qualitative analysis or hierarchical quantitative evaluation with data and methodological limitations (Huang et al., 2019). Urban designers, planners, critics, and even sociologists have unique opinions on how to create spatial vitality based on empirical criteria and theoretical research results, increasing the difficulty of forming a unified standard and bringing certain challenges to urban construction and landscape design. Since the late 1990s, surveys, observations, and quantitative methods have been used to analyze the relationship between urban landscape elements and urban vitality based on case studies (Filion and Hammond, 2003; Ravenscroft, 2000). However, conducting quantitative research and drawing universal conclusions in large study areas have been difficult during this period. In the current era, modern technologies enable the collection of large amounts of dynamic data with accurate spatial information, which provide the basis for accurately quantifying human movement and describing city details (Liu and Long, 2016; Song et al., 2013; Vanderhaegen and Canters, 2017; Zhang and Seto, 2011). Thus, quantitative methods are being used to explore how elements of urban landscapes affect urban vitality (Chen et al., 2017; Lunecke and Mora, 2018; Malizia and Motoyama, 2016).

However, limited quantitative research has been conducted and has mainly been focused on different aspects of the urban landscape as no uniform criteria to quantify it exist, especially at the street block (a street-enclosed area filled by buildings) level across different cities. Moreover, many essential elements are missing in the literature, such as building height, orientation, shape of the footprint, arrangements, block slope, and greening. Furthermore, urban activities and vibrant places vary between day and night, but few studies have considered such differences. Zhang et al. (2019) established a new quantification system based on Conzen’s theory and quantified the urban landscape using 28 indicators from three aspects, namely the city plan, building forms, and land use patterns. This system provides a foundation for research on the systematic quantification of the urban landscape, and we apply this system here to quantitatively analyze and empirically study the relationship between the urban landscape and urban vitality.

During the past decades, China has experienced rapid urbanization and continuous population growth without sufficient attention to urban vitality, thus leading to great damage to the urban ecological environment and numerous social problems (Chen et al., 2015; Lai et al., 2016; Zhang et al., 2020). However, the situation has begun to change in recent years (Bai et al., 2012; Yao et al., 2011). In 2017, the “Urban Renovation and Ecological Restoration (URER)” policy, which aimed to highlight the new requirements for urban development, was proposed and formally implemented by the Chinese government. Chinese cities are now facing new opportunities for urban transformation and sustainable development. However, most urban vitality studies focus on overall city features across multiple cities or in a certain single megacity (Huang et al., 2019; Jin et al., 2017; Tang et al., 2018; Wu et al., 2018a). In comparison, intraurban vitality has gained little attention even though it is essential for fine-scale urban planning (Yue et al., 2019). Many low-vibrancy areas remain inside cities, even in metropolises, because of rapid construction, poverty, incomplete infrastructure, and other reasons. Therefore, fine-scale research on intraurban vitality involving multiple cities is needed to draw universal conclusions about urban transformation.

Based on the urban landscape quantification system, this paper explores the extent to which the urban landscape affects daytime vitality (DV) and nighttime vitality (NV) at the street block level in 15 metropolises of China. Its aim is to provide a reference for formulating more reasonable and targeted URER strategies to build vibrant cities. This paper has three main parts: (1) the methods for measuring influencing factors and their expected impacts, (2) the analysis of the influences of landscape elements and diurnal discrepancies at the multicity scale, and (3) low-vitality block analysis and targeted advice at the intracity scale.

Methodology and data resources

Methodology

Measurement of urban vitality

Early Western scholars in the field of urban planning conducted abundant theoretical studies on the connotations, effect factors, and characteristics of urban vitality (Gehl, 1987; Jacobs, 1961; Lynch and Joint Center for Urban Studies, 1960; Montgomery, 1998; Mumford, 1961). Urban vitality is abstract and complex, and no uniform definition and standard exist in different disciplines because of its rich meanings (Long and Huang, 2019; Ye et al., 2018). Meanwhile, human activities are an indispensable component of city vibrancy. The vitality of cities generally refers to the ability of attracting lively businesses and human activities, which largely depends on the spatio-temporal gathering of human flows. Therefore, many scholars support the idea that urban vitality can be defined by the abundance of urban activities or the intensity of residential activities (Jin et al., 2017; Wu et al., 2018a; Yue et al., 2017), which is adopted in this study. However, urban activities rise and decline with time, represented as cities being active at certain times and inactive at others, or becoming consistently vibrant with different types of activities. Thus, suitable proxies must be selected for measuring urban vitality at different periods.

Currently, tracking records from hand-held global positioning systems (Yue et al., 2017), traffic smart card data (Sulis et al., 2018), social media check-in data (Jin et al., 2017; Wu et al., 2018b), nighttime light data (Levin and Duke, 2012), and small catering data (Long and Huang, 2019; Ye et al., 2018) have been widely used to measure urban vitality. As expected, these sources have strengths and weaknesses. Small catering businesses can be deemed as an indicator of the urban attractiveness of places, but they cannot reflect all dimensions of urban vitality. Urban areas with flourishing small catering businesses tend to be more vibrant, because maintaining this type of business is difficult without pedestrian flows and intensive urban activities. In other words, places that satisfy the locational needs of small catering businesses tend to be densely populated and promote leisure activities, such as walking, resting, and so on (Dong et al., 2019). Meanwhile, unlike large dining restaurants and department stores, small catering businesses are too small to alter their neighboring urban landscapes and thus more flexible and reflect the existing urban vitality. Therefore, the distribution of small catering businesses can be taken as a current reflection of urban vitality (Ye et al., 2018).

Urban NV can be quite different from DV. Light is regarded as the most obvious characteristic of a city at night compared to rural areas, and diverse nightlife activities represent NV. In fact, social activities usually occur in bright areas at night, while dark places are the most unpopulated or poorest areas inside the city (Bennett and Smith, 2017; Hadfield, 2015; Rudlin and Falk, 1999). Furthermore, residents’ feeling of safety and comfort relates to the presence or absence of urban nighttime lights, which is also an important aspect of inducing urban activities. In some cases, some stores or parks are open the entire night without any visitors, but they meet the potential demand and generate a special atmosphere to differentiate from the dark, lonely, and unfamiliar places (Hadfield, 2015; Van Liempt et al., 2015). The improvements and high resolution of the recently released nighttime light data set, i.e. the Visible Infrared Imaging Radiometer Suite (VIIRS) imagery, eliminate three critical problems of saturation, blooming, and on-board calibration; it has a resolution of approximately 500 meters (the Luojia 1-01 nighttime light imagery has a higher resolution of 130 meters, which, however, began to be available after June 2018.) (Bennett and Smith, 2017; Li et al., 2019). Moreover, nighttime light imagery can serve as a reliable proxy for indicating the intensity of urban activities at fine-scale and their variations over time (Levin and Duke, 2012; Levin and Zhang, 2017; Liu et al., 2016). Substantial correlations between these nighttime light data and some socio-economic activities have been reported, including economic production, population, and energy consumption (Wang et al., 2018; Zhang and Seto, 2011). Therefore, small catering businesses and nighttime light data were used in the current study to measure the city’s DV and NV, respectively. Specifically, kernel density analysis was adopted to measure the distribution of small catering businesses within a radius of 500 meters, and a weight was set according to the number of comments to acquire the average values in different blocks. Furthermore, the average of the VIIRS nighttime light data was used as the NV value.

Quantification of factors influencing urban vitality

The factors affecting urban vitality mainly include population density, building density, functional density, mixed land use, open space, and architectural ages, along with the pattern, accessibility, walkability, and connectivity of streets (Katz et al., 1994; Montgomery, 1998; Sung and Lee, 2015; Zeng et al., 2018), most of which belong to urban landscape components. We quantified impact factors considering two aspects: urban landscape components and other factors. As mentioned previously, the urban landscape has three aspects, namely the city plan, building forms, and land use patterns. The detailed quantification indicators according to Zhang et al. (2019) are listed in Table 1. To capture the considerable effects of the environment on livability and attractiveness, NGR, reflecting the externality of green space, was added to the quantification system as a supplement to SA_GRE, an indicator describing each block’s ability to provide open spaces (including squares, parks, and so on). Population and functional densities were used as control variables and were regarded as significant factors affecting vitality in most previous studies (Long and Huang, 2019; Yue et al., 2017).

Table 1.

Influencing factors and expected impacts.

Category of indicators			Indicators	Code	Expected impact
Urban landscape quantification indicators	City plan pattern	Street system	Public transportation convenience degree	PTCD	+
			Road intersection quantity	RIQ	+
			Road intersection separation distance	RISD	−
			Near-road building density	NRBD	+
			Near-road building expandability	NRBE	+
		Block pattern	Area	A	−
			Fractal dimension	FD	−
			Spatial compact ratio	SCR	+
			Block relief degree	BRD	−
			Adjacent block number per unit length	ULABN	+
			Mean of block elevation	MBE	−
		Building arrangement	Eccentricity degree of building distribution	BDE	−
			Dispersion degree of building distribution	BDD	+
			Max of building area	Max_BA	+
	Building forms pattern	Building 2D form	Mean of building area	MBA	+
			Building area-weighted orientation index	AWBOI	+
			Building area-weighted fractal dimension	AWBFD	−
			Building area-weighted spatial compact ratio	AWBSCR	+
			Building fractal dimension variation coefficient	BFDVC	+
			Building spatial compact ratio variation coefficient	BSCRVC	+
		Building 3D form	Mean of building height	MBH	+
			Building height variation coefficient	BHVC	+
			Proportion of the tower building	PTB	+
	Land use pattern	Land use function	Near-block greening rate	NGR	+
			Residential service ability	SA_RES	+
			Public service ability	SA_PUB	+
			Commercial service ability	SA_COM	+
			Industrial service ability	SA_IND	+
			Transportation service ability	SA_TRA	+
			Green service ability	SA_GRE	+
			Land use diversity	LUD	+
		Land use intensity	Building expandability	BE	+
		Land use intensity	Building density	BD	+
Control factors			Population density	PD	+
Control factors			Functional density	POID	+

Theoretically, extremely high or low indicator values may have an effect opposite to that expected, but these extreme phenomena are unlikely to exist in real cities for a long time. See Supplemental Table B for the explanations of each indicator and the Supplemental Materials for the analysis of the expected impact of the urban landscape on urban vitality.

Analytical methods

Urban vitality, as a representation of human activity intensity, usually presents an agglomeration effect and spatial dependency, and conventional statistical analysis, which ignores such phenomena, may not be enough to explain urban vitality at the local urban scale. This study used the global linear regression and geographically weighted regression (GWR) approaches to examine the impact of the urban landscape on urban vitality. First, the ordinary least squares (OLS) method was used to estimate the overall statistical correlations between the dependent and explanatory variables and identify the major determinants of urban vitality from the global view. Then, Moran’s I was used to examine the existence of spatial autocorrelation of urban vitality before employing the GWR model. Finally, GWR was utilized to investigate spatial non-stationarity and reveal the local variations of the regression parameters for the explanatory variables. Pseudo t-tests were also performed to test the statistical significance of the regression parameters, which were obtained by dividing the parameter estimates by the standard errors. The detailed equations and descriptions for the above methods can be found in the Supplemental Materials.

Study area and data sources

The urban built-up area of 15 metropolises in China was selected as our study area, including Shanghai, Beijing, Guangzhou, Shenzhen, Tianjin, Chongqing, Nanjing, Wuhan, Hangzhou, Chengdu, Suzhou, Shenyang, Qingdao, Changsha, and Xi’an (Supplemental Fig. A). These cities have representativeness, covering the major geographical, climatic, economic, political, and cultural districts of China. They are the most dynamic cities in China and are considered the template for city development. In addition, the data for these large cities are relatively complete, reliable, and easy to collect. The data include two types: (1) remote sensing image products and (2) geospatial “big data” (Supplemental Table A). The built-up areas were divided into blocks by streets and water bodies according to the research by Liu and Long (2016), and the blocks with incomplete data were excluded from further analysis. Different levels of roads (e.g. expressways, arterial roads, branch roads, and living streets) were derived from OpenStreetMap (OSM).

Results and discussion

Characteristics of urban vitality

According to the different degrees of frequency distribution skewness, natural logarithmic transformation was performed for DV, PD, and POID, whereas square root transformation was used for NV. Taking the overall perspective across cities, DV and NV show obvious differences (Figure 1). For example, among these cities, Tianjin is relatively inactive at daytime but extremely active at nighttime, whereas Qingdao displays the exact opposite tendency. Shanghai, Beijing, and Shenyang are active, whereas Chongqing, Nanjing, and Changsha exhibit poor performance in terms of urban vitality. From the spatial perspective, Figure 2 illustrates that urban vitality is attenuated from the urban center to the urban fringe, with higher speed at nighttime than at daytime; this phenomenon can be manifested as smaller standard deviation ellipses and steeper lines at night in most of the cities. In addition, the DV and NV centers do not completely overlap, and the high-vitality centers are more concentrated at night and more distributed and extensive during the day. Overall, the vitality of Chongqing and Changsha and the NV of Guangzhou and Nanjing require the most improvement.

Figure 1.

Median and mean of Ln(DV) and Sqrt(NV) in different cities.

Figure 2.

Distribution of urban vitality (the standard deviational ellipses, calculated according to the blocks’ distribution with Ln(DV) and Sqrt(NV) as weights, represent 1× standard deviation, 2× standard deviation, and 3× standard deviation from the city center to the periphery).

Global and local influences of the urban landscape on urban vitality

OLS regression analysis and global relationships

Correlation and linear regression analyses were used to test the effect of urban landscape indictors on urban vitality, with 14 dummy and 2 control variables. The adjusted R²-values of the DV and NV models were 0.443 and 0.463, respectively. As can be seen, the models passed the F-test at the 99% significance level, and the regression standardized residuals show normal distributions (Supplemental Fig. B).

The research results show that most landscape elements have significant influences, which are consistent with the previous assumptions (Table 2). This means that street accessibility, space utilization, block pattern, greening, mixed land use, and building forms and arrangement are all influential factors of urban vitality, which is in line with expectations and previous theoretical research.

Table 2.

Regression analysis results for urban vitality.

Dependent variables		Ln(DV)	Sqrt(NV)	VIF	Dependent variables		Ln(DV)	Sqrt(NV)	VIF
Independent variables		Standardized coefficients		VIF	Independent variables		Standardized coefficients		VIF
Urban landscape variables	Ln(PTCD)	0.144***	0.187***	3.524	Dummy variables	Beijing	0.052***	0.046***	3.976
	Ln(RIQ)	0.119***	0.252***	1.501		Shanghai	0.046***	0.188***	4.843
	Ln(NRBE)	−0.025***	−0.050***	1.375		Guangzhou	0.022***	−0.030***	2.789
	SCR	0.032***	0.109***	1.115		Shenzhen	−0.044***	0.053***	2.989
	BRD	0.016***	−0.019***	1.506		Chengdu	0.050***	0.204***	2.381
	Ln(BDD)	0.045***	0.035***	1.894		Hangzhou	0.069***	0.063***	2.224
	Ln(Max_BA)	0.071***	0.113***	1.444		Wuhan	0.033***	0.040***	2.202
	AWBOI	0.034***	−0.020***	1.143		Nanjing	0.054***	0.026***	2.415
	AWBFD	0.000	−0.015***	1.149		Tianjin	0.047***	0.233***	2.392
	Ln(MBH)	0.028***	0.018***	1.701		Suzhou	0.103***	0.195***	2.712
	Ln(BHVC)	0.060***	0.039***	2.079		Xi’an	0.050***	0.177***	2.175
	Ln(PTB)	0.064***	0.054***	1.609		Changsha	0.025***	0.068***	1.700
	Ln(NGR)	0.026***	0.007**	1.134		Shenyang	0.030***	0.204***	1.869
	Ln(SA_PUB)	0.030***	−0.087***	1.629		Qingdao	0.066***	0.033***	1.946
	Ln(SA_TRA)	0.072***	0.119***	1.878
	Ln(SA_GRE)	0.073***	0.014***	1.160	Control variables	Ln(PD)	0.033***	0.050***	1.956
	LUD	0.028***	0.010**	2.236	Control variables	Ln(POID)	0.438***	0.242***	4.442

DV: daytime vitality; NV: nighttime vitality; VIF: variance inflation factor.

***p < 0.01, **p < 0.05.

More specifically, (1) among all the landscape elements, the city plan pattern, which affects urban vitality mainly through street accessibility and building arrangement, has a higher influence than the patterns of land use and building forms. For example, street accessibility indicators (PTCD and RIQ) have the highest impact among all the landscape quantification indicators; in other words, convenient public transport and high road network density are beneficial for stimulating urban vitality. (2) Mixed land use makes blocks more functional, thereby increasing the attractiveness of different human activities and the opportunities for pedestrians to communicate and interact with each other. Meanwhile, assembled crowds drive the development of various industries, such as services and commerce, which in turn, help maintain the block’s vitality. In particular, the urban green space has many functions (except service functions), including ecology, landscape, and mental and physical health promotion, and offers positive externalities to the adjacent blocks’ vitality, as shown by the quantitative results. Therefore, vibrant blocks are concentrated in the city center, such as commercial or mixed-use blocks with large shopping malls, green-oriented blocks with parks or plazas, traffic-oriented blocks with train stations or airports, and public service-oriented blocks with museums or hospitals. The arrangement of buildings in these blocks is not traditionally enclosed but is centered on a few main buildings with an open space. (3) Building form has a vital effect on urban vibrancy, whereas 2D building forms and placement have lower effects than the 3D building forms. Specifically, building height diversity and tower buildings are significant influencing factors.

Notably, NRBE and NRBD have a negative effect on urban vitality in this study, although they usually represent the quantity of ground commercial stores and have been positively correlated with the vitality of commercial streets in some studies (Xu et al., 2018). However, high NRBE and NRBD mean low openness of street walls, which attracts more commercial activities but far fewer social activities (Xu et al., 2018). Likewise, insufficient near-road shops and solid walls lacking two-way visual permeability work against commercial activities (Gehl, 1987; Jacobs, 1993). Generally, near-road buildings are dominated by residential, commercial, and mixed-use buildings, and street-facing buildings are for mixed or commercial use. However, only one or two sides of street-facing blocks have commercial functions in China, indicating that only a few near-road spaces have been efficiently used. Furthermore, the materials, ages, and detailed design of buildings, particularly near-road buildings, are intimately associated with commercial and social activities, but they have not been considered in the current study because these features are difficult to quantify in large areas.

Urban landscape indicators are dependent on one another as they are different aspects of the urban landscape that interrelate and interact with each other. Therefore, some indicators that have significant positive or negative correlations with urban vitality were excluded from the regression model to avoid collinearity and increase reliability (Supplemental Table C and Fig. C). Among them, RISD and A have negative effects when they are above a certain value but positive effects when they are small, which is consistent with other research. BD and BE have strong correlations with urban vitality, which conforms to the concept of the compact city; that is, urban densification can transform the morphology and function of cities, indicating some important aspects of a vibrant and sustainable city. Meanwhile, ULABN is highly positively correlated with POID, which has the strongest influence on vitality, and negatively correlated with A. In other words, small-scaled blocks, which are closely connected with their neighbors, have more opportunities to use the services from neighboring blocks and are more likely to stimulate vitality.

GWR analysis and spatial variations

Considering the potential spatial non-stationarity, we supplemented the GWR model to analyze the spatial variations of the effects of urban landscape variables. We used the same set of variables to construct the GWR model. The comparison between the results of OLS regression and the GWR model is shown in Table 3. The results of Moran’s I statistics illustrate the significant strong positive spatial autocorrelations for most dependent and independent variables and justify the use of the GWR model (Supplemental Table D). The results show that the GWR model leads to significant improvement compared with OLS regression. First, the significant decrease of the Akaike information criteria shows the much better performance of the GWR model in exploring the relationships between the urban landscape and urban vitality. Second, the significant increases of R² and adjusted R² and the decrease of the residual squares indicate that the GWR model has a better goodness-of-fit. Finally, the much lower Moran’s I value of residuals in the GWR model suggests a higher spatial independence.

Table 3.

Comparison between OLS regression and GWR.

Dependent variables	Ln(DV)		Sqrt(NV)
Models	OLS	GWR	OLS	GWR
Residual squares	30,617	21,669	29,482	12,998
AICs	123,895	109,401	121,817	82,338
R ²	0.44	0.58	0.46	0.75
Adjusted R²	0.44	0.55	0.46	0.73
Moran’s I of residuals	0.14**	0.06**	0.34**	0.12**

AIC: Akaike information criteria; DV: daytime vitality; GWR: geographically weighted regression; NV: nighttime vitality; OLS: ordinary least squares.

**Indicates significance at 0.05 level.

Compared with the global OLS regression results, the parameter estimates of GWR show significant spatial variations, and all landscape variables have both positive and negative parameters with different proportions of significant values, indicating that the unified parameters in the global regression model might be problematic (Supplemental Figs. D-G). The positive or negative influences of urban landscape variables do not hold true for all blocks in the study area. Based on the proportions of significant positive and negative values, results indicate that the characteristics of the GWR parameters are generally consistent with those of the OLS regression. As shown in Supplemental Table E, the major explanatory variables in the OLS regression (e.g. PTCD, RIQ, Ln(Max_BA), and Ln(POID)) have very high proportions of significant values, and their positive or negative effects in most of the significant street blocks are also consistent with the OLS results. Meanwhile, the variables with little influence or weak significance in the OLS regression (e.g. Ln(NGR) and AWBFD) have few significant street blocks in the entire region and show apparent divisions of positive and negative results. Furthermore, GWR reveals the spatial variations of urban landscape effects on urban vitality for different areas in the cities, which are highly related to the socio-economic and environmental contexts of these areas (Fang et al., 2019).

Diurnal variation

The results also indicate that the effects of urban landscape variables on DV and NV have obvious discrepancies. However, the diurnal variation issue has been seldom discussed in previous studies. Residents’ nightlife is a crucial part of urban activities and can boost the local economy and shape a unique culture (Bianchini, 1995; Hollands and Chatterton, 2003). People have more flexible time after work, and such free time drives many social and economic activities, which is perhaps why transportation accessibility (PTCD, RIQ) and service facilities both (SA_TRA) have high impacts at night. Restricted by service time, the vitality of public service facilities and open parks decreases at night. In contrast, people are more likely to gather in specific beautifully lit places, such as iconic buildings, commercial buildings, and plazas. Bars, dance clubs, karaoke clubs, and saunas are essential components of Chinese nightlife. In addition, some factories are also at the center of economic vitality at night. Another important issue related to nightlife is safety (Schwanen et al., 2012). These reasons may explain why SA_GRE and NGR have low effects, whereas Max_BA has a high effect and why BRD and SA_PUB have opposite effects during the day and at night. Overall, transportation accessibility, greening rate, and iconic buildings are the most important landscape factors influencing DV, whereas transportation accessibility, iconic buildings, and block forms are vital to NV.

Development strategies for low-vitality blocks

The assessment of existing urban landscapes in terms of performance, which aims to identify the conditional characters of low-vitality blocks rather than directly applicable values, can serve as helpful references for designing a “vibrant urbanity” (Pont and Haupt, 2010). URER is a vital task for building livable, vibrant, and characteristically modern cities in China, which aims to improve urban green space systems, strengthen intensive mixed land use and service capabilities, complete urban public service facilities, increase public spaces, and optimize urban form and building design management. In this study, we divided urban vitality into five levels using natural breaks, and the blocks with low vitality (including the lowest and lower vitality blocks) tend to be deemed as requiring restoration or regeneration. Low-vitality blocks account for 20.35 and 33.61% by day and at night, respectively (Supplemental Table F), whereas only 12.73% of the blocks have low vitality all day. Except for some 2D building forms’ characteristics, low-vitality blocks have particularly different landscape characteristics compared with other blocks, which are divided into six categories using K-means clustering (Figure 3, Supplemental Fig. H), including isolated blocks with complex shapes (Type I), and blocks with monotonous buildings (Type II), insufficient open space (Type III), fewer land functions (Type IV), inconvenient transportation (Type V), and inefficient land use (Type VI). From the perspective of spatial distribution, most of these blocks are distributed at the urban fringe, and the blocks near urban boundaries are dominated by Types I, III, and V blocks. The number and size of the other types are small, and the distributions are scattered (Figure 4). The characteristics of these six types of low-vitality blocks might provide some potential objectives and solutions for URER policy implementation (Figure 3, Supplemental Tables G–H).

Figure 3.

Mean value of landscape indicators for the low-vitality blocks.

Figure 4.

Distribution of low-vitality blocks in Beijing.

Type I blocks have large sizes and few networks. In line with the previous work of Pont and Haupt (2010), large islands have relatively little exposure and weak connections with their neighbors, which hamper the stimulation of interaction and mixture. As mentioned by Jacobs (1961), high coverage and few networks, combined with high build intensity, will concentrate movements in the streets, which would then conflict with the accessibility and walkability of the space. Thus, a compromise is required. Furthermore, complex and uncompact forms are not conducive to the full usage of spaces inside blocks.

For Type II blocks, particularly the low-vitality blocks by day, the standardization of a building’s form and single building type lead to the absence of diversity. Jacobs (1961) highlighted the necessity of diversity and a mix of functions when densities become relatively high. In addition, the direction of buildings is unsuitable for daylight access, which is important for human well-being and energy saving. Hence, the architecture principle of “applicable, economical, green and beautiful” must be implemented for this type of block to enrich their building types and promote the harmony and beauty of architectural texture, color, form, and environment.

For Type III blocks, the low greening rate and the lack of parks, squares, and cultural tourist attractions are the main causes of low vitality. Open spaces for residents’ fitness, leisure, and public activities can be viewed as necessary elements contributing to an attractive form of urbanity. Consequently, the key issue is to improve the ecological function of the blocks and rationally arrange park squares according to population size.

The common landscape features of Type IV blocks are low mixed land use, poor service capabilities, and inadequate supporting facilities. As discussed above, land use mixture and good access to urban amenities contribute a lot to attractiveness of spaces. Thus, these blocks should diversify their land use functions and promote the construction of various public facilities (e.g. culture, sports, medical care, education, and so on) and residential supporting facilities (e.g. parking lots, convenience stores, logistics, and so on).

For Type V blocks, poor accessibility, which is represented by low network density, long distances between intersections, and inaccessible public transportation (such as subways and buses), restricts the development of urban vitality. Potential activity intensity is not only dependent on the number of people within walkable distance but also related to transportation modes and the accessibility to these modes in this area (Pont and Haupt, 2010).

The building density of Type VI blocks is extremely low, and the buildings are extremely short. As described in many studies (e.g. Pont and Haupt, 2010), density thresholds are critical to the different performances of the urban landscape, although a very high density may lead to complications and negative consequences according to Jacobs (1961). Thus, population density, development intensity, and land use efficiency should be improved to optimize the building patterns and humanize street spaces for these types of blocks.

Limitations and prospects

The concepts of a 24-hour city and a night economy have been widely recognized by urban planners and local governments since the 1990s (Heath, 1997). Thus, conducting research on urban systems during daytime and after night falls is significant. Anthropogenic activities at night, which are mainly represented by the electrical illumination of population centers, light up the Earth’s surface. In this study, VIIRS images serve as a reliable proxy for economic activities at night because of their superior spatial resolution and improved radiometric saturation compared with the images provided by the Defense Meteorological Satellite Program’s Operational Linescan System. However, urban vitality is a very broad and complicated concept, with socio-economic and cultural dimensions, which cannot be completely and precisely represented by small catering businesses and nighttime light data. For example, several large street blocks located in the fringe of cities were identified as low vitality, which might be urban parks or suburban malls. It is inappropriate to define these kinds of blocks as low vitality, because they are frequently used by urban residents without intensive small catering businesses and night lights. Thus, it might be better to apply the proposed method in the central-city areas or adopt real-time density and mobility data to indicate concentration of activities. Furthermore, the materials, ages, and detailed design of buildings, particularly near-road buildings, are intimately associated with commercial and social activities, but they were not considered in this study due to data unavailability and difficulties in quantifying these features in large areas. The development strategies for low-vitality blocks can vary if the performance of the urban landscape is assessed from different perspectives, and solutions should be critically judged based on all conditions that contribute to urbanity. Furthermore, conducting further research on the scope and impact mechanism of green space externality, which has been largely ignored in the current research, is valuable. We analyzed 15 mega-cities in China to examine whether our research results are applicable to different types of cities. In future studies, the proposed analytical framework can be applied to other small and mid-sized cities in China or to cities in other countries for further examination. The average size of street blocks in this research is relatively large (about 0.11 square kilometers), which may be caused by the sparse OSM road networks (Liu and Long, 2016). Fine street maps can be used to alleviate this deficiency and generate more realistic street blocks in Chinese cities.

Conclusions

Taking 15 metropolises in China as a case study, this paper empirically examines the major determinants of urban vitality, reveals the spatial variations and diurnal discrepancies of urban landscape effects, and provides some policy implications for the improvement of low-vitality blocks. To our knowledge, despite some limitations, this is one of the first studies to quantitatively analyze this relationship for multiple cities at the street block level. Meanwhile, after quantifying urban vitality from the daytime and nighttime perspectives, results reveal that the influences of the urban landscape and the distribution of urban vitality have spatial variations and obvious diurnal discrepancies. Additionally, the location and characteristics of low-vitality blocks in each city are identified and categorized, and some targeted suggestions are provided to promote urban quality and vitality in conjunction with the URER policy. The findings of this work may serve as evidence and provide guidance to authorities in tailoring strategies toward improved place making and urban revitalization.

Supplemental Material

sj-pdf-1-epb-10.1177_2399808320924425 - Supplemental material for How can the urban landscape affect urban vitality at the street block level? A case study of 15 metropolises in China

Supplemental material, sj-pdf-1-epb-10.1177_2399808320924425 for How can the urban landscape affect urban vitality at the street block level? A case study of 15 metropolises in China by Anqi Zhang, Weifeng Li, Jiayu Wu, Jian Lin, Jianqun Chu and Chang Xia in Environment and Planning B: Urban Analytics and City Science

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Anqi Zhang is a PhD candidate in the Department of Urban Planning and Design at the University of Hong Kong. She received her Master’s degree in Architectural Design and Theory from Peking University and Bachelor’s degree in Land Resources Management from Wuhan University. Her research interests include urban morphology, spatial analytics, and application of geographical open data in urban and environmental studies.

Weifeng Li is currently an Associate Professor in the Department of Urban Planning and Design at the University of Hong Kong. His research interests are in the areas of environmental sustainability, land use-transportation planning interactions, and the environmental and energy impacts, as well as the use of GIS and spatial analysis in urban planning. He holds both his bachelor’s and master’s degrees in Geography from Peking University, and did the PhD study at Massachusetts Institute of Technology.

Jiayu Wu is an Assistant Professor at College of Agriculture and Biotechnology, Zhejiang University in China, expertise in the field of urban landscape, protected area, and planning policy.

Jian Lin is a Professor at College of Urban and Environmental Sciences, Peking University in China, expertise in the field of land use policy and planning practice.

Jianqun Chu is an Associate Professor at College of Urban and Environmental Sciences, Peking University in China, expertise in the field of land use policy and planning practice.

Chang Xia is a PhD candidate in the Department of Urban Planning and Design at the University of Hong Kong. He received his Master’s and Bachelor’s degree in Land Resource Management from Wuhan University. His research interests include urban and land use modeling, urban morphology, and application of geographical open data in urban and environmental studies.

ORCID iD

Chang Xia

Supplemental material

Supplemental material for this article is available online.

References

Bai

Chen

Shi

(2012) Landscape urbanization and economic growth in China: Positive feedbacks and sustainability dilemmas. Environmental Science & Technology 46(1): 132–139.

Bennett

Smith

(2017) Advances in using multitemporal night-time lights satellite imagery to detect, estimate, and monitor socioeconomic dynamics. Remote Sensing of Environment 192: 176–197.

Bianchini

(1995) Night cultures, night economies. Planning Practice & Research 10(2): 121–126.

Brenner

(2014) Implosions/Explosions: Towards a Study of Planetary Urbanization. Berlin: Jovis.

Chen

Yan

Gao

, et al. (2015) Quantifying circular urban expansion patterns of compact Chinese cities: The case of Yangtze River Delta, China. Environment and Planning B: Planning and Design 42(2): 279–299.

Chen

Liu

(2017) Analyzing parcel-level relationships between urban land expansion and activity changes by integrating Landsat and nighttime light data. Remote Sensing 9(2): 164.

Conzen

MRG

(1960) Alnwick, Northumberland: A Study in Town-Plan Analysis. London: George Philip.

Conzen

MRG

(2004) Thinking About Urban Form: Papers on Urban Morphology, 1932–1998. Oxford and New York: Peter Lang.

Craven

Wellman

(1973) The network city. Sociological Inquiry 43(3–4): 57–88.

10.

Dong

Ratti

Zheng

(2019) Predicting neighborhoods’ socioeconomic attributes using restaurant data. Proceedings of the National Academy of Sciences of the United States of America 116(31): 15447–15452.

11.

Fang

(2019) Does hotel location tell a true story? Evidence from geographically weighted regression analysis of hotels in Hong Kong. Tourism Management 72: 78–91.

12.

Filion

Hammond

(2003) Neighbourhood land use and performance: The evolution of neighbourhood morphology over the 20th century. Environment and Planning B: Planning and Design 30(2): 271–296.

13.

Gehl

(1987) Life Between Buildings: Using Public Space. New York: Van Nostrand Reinhold.

14.

Hadfield

(2015) The night-time city. Four modes of exclusion: Reflections on the urban studies special collection. Urban Studies 52(3): 606–616.

15.

Hall

Pfeiffer

(2000) Urban Future 21: A Global Agenda for Twenty-First Century Cities. London: E & FN Spon.

16.

Heath

(1997) The twenty‐four hour city concept – A review of initiatives in British cities. Journal of Urban Design 2(2): 193–204.

17.

Hollands

Chatterton

(2003) Producing nightlife in the new urban entertainment economy: Corporatization, branding and market segmentation. International Journal of Urban and Regional Research 27(2): 361–385.

18.

Huang

Zhou

, et al. (2019) Evaluating and characterizing urban vibrancy using spatial big data: Shanghai as a case study. Environment and Planning B: Urban Analytics and City Science. Epub ahead of print 13 February 2019. DOI: 10.1177/2399808319828730.

19.

Jacobs

(1993) Great streets. ACCESS Magazine 1(3): 23–27.

20.

Jacobs

(1961) The Death and Life of Great American Cities. New York: Vintage Books, pp.143–222.

21.

Jin

Long

Sun

, et al. (2017) Evaluating cities’ vitality and identifying ghost cities in China with emerging geographical data. Cities 63: 98–109.

22.

Katz

Scully

Bressi

(1994) The New Urbanism: Toward an Architecture of Community. Vol. 10. New York: McGraw-Hill.

23.

Lai

Huang

Yang

, et al. (2016) Carbon emissions from land-use change and management in China between 1990 and 2010. Science Advances 2(11): e1601063.

24.

Levin

Duke

(2012) High spatial resolution night-time light images for demographic and socio-economic studies. Remote Sensing of Environment 119: 1–10.

25.

Levin

Zhang

(2017) A global analysis of factors controlling VIIRS nighttime light levels from densely populated areas. Remote Sensing of Environment 190: 366–382.

26.

, et al. (2019) A preliminary investigation of Luojia-1 night-time light imagery. Remote Sensing Letters 10(6): 526–535.

27.

Liu

Long

(2016) Automated identification and characterization of parcels with OpenStreetMap and points of interest. Environment and Planning B: Planning and Design 43(2): 341–360.

28.

Liu

Delahunty

Zhao

, et al. (2016) These lit areas are undeveloped: Delimiting china’s urban extents from thresholded nighttime light imagery. International Journal of Applied Earth Observation and Geoinformation 50: 39–50.

29.

Long

Huang

(2019) Does block size matter? The impact of urban design on economic vitality for Chinese cities. Environment and Planning B: Urban Analytics and City Science 46(3): 406–422.

30.

Lunecke

MGH

Mora

(2018) The layered city: Pedestrian networks in downtown Santiago and their impact on urban vitality. Journal of Urban Design 23(3): 336–353.

31.

Lynch

and Joint Center for Urban Studies (1960) The Image of the City. Cambridge: MIT Press.

32.

Malizia

Motoyama

(2016) The economic development–vibrant center connection: Tracking high-growth firms in the DC region. The Professional Geographer 68(3): 349–355.

33.

Marcus

Francis

(1998) People Places: Design Guidelines for Urban Open Space. 2nd ed. New York: Wiley.

34.

Montgomery

(1998) Making a city: Urbanity, vitality and urban design. Journal of Urban Design 3(1): 93–116.

35.

Mumford

(1961) The City in History: Its Origins, Its Transformations, and Its Prospects. Vol. 67. Boston, MA: Houghton Mifflin Harcourt.

36.

Oliveira

Medeiros

(2016) Morpho: Combining morphological measures. Environment and Planning B: Planning and Design 43(5): 805–825.

37.

Pont

Haupt

(2010) Spacematrix: Space, Density and Urban Form. Rotterdam: NAI.

38.

Ravenscroft

(2000) The vitality and viability of town centres. Urban Studies 37(13): 2533–2549.

39.

Rudlin

and Falk

(1999) Building the 21st Century Home: The Sustainable Urban Neighbourhood. Oxford: Architectural Press.

40.

Schwanen

Van Aalst

Brands

, et al. (2012) Rhythms of the night: Spatiotemporal inequalities in the nighttime economy. Environment and Planning A: Economy and Space 44(9): 2064–2085.

41.

Smith

(2002) New globalism, new urbanism: Gentrification as global urban strategy. Antipode 34(3): 427–450.

42.

Song

Popkin

Gordon-Larsen

(2013) A national-level analysis of neighborhood form metrics. Landscape and Urban Planning 116: 73–85.

43.

Sulis

Manley

Zhong

, et al. (2018) Using mobility data as proxy for measuring urban vitality. Journal of Spatial Information Science 16: 137–162.

44.

Sung

Lee

(2015) Residential built environment and walking activity: Empirical evidence of Jane Jacobs’ urban vitality. Transportation Research Part D: Transport and Environment 41: 318–329.

45.

Tang

Lin

, et al. (2018) Exploring the influence of urban form on urban vibrancy in Shenzhen based on mobile phone data. Sustainability 10(12): 4565.

46.

Vanderhaegen

Canters

(2017) Mapping urban form and function at city block level using spatial metrics. Landscape and Urban Planning 167: 399–409.

47.

Van Liempt

Van Aalst

Schwanen

(2015) Introduction: Geographies of the urban night. Urban Studies 52(3): 407–421.

48.

Wang

Zhou

, et al. (2018) Mapping population density in China between 1990 and 2010 using remote sensing. Remote Sensing of Environment 210: 269–281.

49.

Woodworth

Wallace

(2017) Seeing ghosts: Parsing china’s “ghost city” controversy. Urban Geography 38(8): 1270–1281.

50.

Ren

, et al. (2018a) Check-in behaviour and spatio-temporal vibrancy: An exploratory analysis in Shenzhen, China. Cities 77: 104–116.

51.

Song

, et al. (2018b) Urban form breeds neighborhood vibrancy: A case study using a GPS-based activity survey in suburban Beijing. Cities 74: 100–108.

52.

Guan

, et al. (2018) The cause and evolution of urban street vitality under the time dimension: Nine cases of streets in Nanjing city, China. Sustainability 10(8): 2797.

53.

Yao

Wang

, et al. (2011) Urbanization in China needs comprehensive scientific thinking: Exploration of the urbanization mode adapted to the special situation of China. Geographical Research 30(11): 1947–1955.

54.

Liu

(2018) How block density and typology affect urban vitality: An exploratory analysis in Shenzhen, China. Urban Geography 39(4): 631–652.

55.

Yeh

Zhuang

, et al. (2017) “Form syntax” as a contribution to geodesign: A morphological tool for urbanity-making in urban design. Urban Design International 22(1): 73–90.

56.

Yue

Chen

Zhang

, et al. (2019) Spatial explicit assessment of urban vitality using multi-source data: A case of Shanghai, China. Sustainability 11(3): 638.

57.

Yue

Zhuang

Yeh

, et al. (2017) Measurements of POI-based mixed use and their relationships with neighbourhood vibrancy. International Journal of Geographical Information Science 31(4): 658–675.

58.

Zeng

Song

, et al. (2020) Spatially explicit assessment on urban vitality: Case studies in Chicago and Wuhan. Sustainable Cities and Society 40: 296–306.

59.

Zhang

Xia

Chu

, et al. (2019) Portraying urban landscape: A quantitative analysis system applied in fifteen metropolises in China. Sustainable Cities and Society 46: 101396.

60.

Zhang

Seto

(2011) Mapping urbanization dynamics at regional and global scales using multi-temporal DMSP/OLS nighttime light data. Remote Sensing of Environment 115(9): 2320–2329.

61.

Zhang

Chen

(2020) Exploring the relationships between urban form metrics and the vegetation biomass loss under urban expansion in China. Environment and Planning B: Urban Analytics and City Science 47(3): 363–380.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

14.28 MB