1. Introduction
The United Nations Development Goal 11 underscores the necessity for inclusive, secure, and sustainable urban environments [1]. Within this context, urban green space (UGS) plays an important role [2]. UGS is defined as the vegetated land within urban planning boundaries, including parks, outdoor sports facilities, play areas, and greenways [3], which offers both environmental and social benefits [4]. It establishes numerous interconnections with human health and well-being [5]. Additionally, ecological restoration and adequate urban green space provision facilitate healthy urban development and mitigate income inequality among urban residents [6,7]. Therefore, the adequate provision of UGS stands as a crucial factor in the establishment of sustainable cities and healthy, green human settlements.
Equity in UGS constitutes a vital area of research for evaluating city sustainability and residents’ quality of life [8]. Many research teams have employed the Gini coefficient to assess inequalities in green space exposure in urban areas [9,10]. This coefficient quantifies inequalities in green space exposure among residents within a region based on the Lorentz curve [11]. Some studies indicate that relatively equitable UGS provision is more prevalent in cities with favorable climatic and natural conditions [12,13]. Conversely, research in developed countries suggests that UGS equity correlates with socio-economic factors, including urban economic development, population size, residents’ income, and education levels [14,15]. Economic policies exert varying effects on green space equality in Chinese cities, influenced by factors such as urban land area, per capita gross regional product (GRP), housing prices, and housing space per capita [6,16]. Cities with larger populations, higher GRPs, and greater development tend to experience more equitable UGS provision [17].
It has been demonstrated that the rapid expansion of urban land leads to a reduction in green space provision [18], resulting in ecological challenges and exacerbating the consequences of extreme weather events [19]. In 2015, China witnessed a 57.23% decrease in urban green space compared to 1990 [20]. Urbanization tends to diminish green space provision and exacerbate social disparities among urban residents [21]. Wu et al. [22] observed a more rapid reduction in green space exposure inequalities in developing regions globally compared to developed regions. Conversely, Jiao et al. [23] found that rapid regional urbanization leads to UGS inequality, which hampers long-term urban development. Nonetheless, the influence of rapid urbanization on green space equity remains an underexplored area (research on the relationship between rapid urbanization and green space equity). China has experienced significant urbanization up to 2020. Ecological degradation and the strain on ecosystem services resulting from China’s rapid economic growth are critical issues in its pursuit of sustainability [24,25]. The environmental effect of urban development is multifaceted [26], and the supply of urban green space in China typically follows a pattern of transitioning from more to less and then from less to more. While economically developed cities exhibit higher rates of urban greening, per capita green space supply tends to inversely correlate with socio-economic development [27]. China’s urbanization holds global significance, with land finance serving as a crucial driver of this process [28,29]. However, research on the relationship between land finance and UGS supply equity has been limited. Some studies have demonstrated a positive causal link between land finance and urban sprawl, highlighting its role in the local capital accumulation system [30]. Land finance is associated with various forms of urban expansion [31], which substantially hampers the development of the green economy [32] and poses a latent risk to sustainable development [33]. According to data from the Ministry of Finance, China’s land transfer revenue in 2020 amounted to CNY 8414.2 billion, reflecting a year-on-year increase of 15.9%.
In light of these considerations, we propose the utilization of population-weighted green space exposure and the Gini coefficient to evaluate UGS provision and inequality in large cities. This approach involves collecting data on natural, economic, and social factors, particularly those linked to urbanization, that impact UGS distribution in 140 major mainland Chinese cities. The primary objectives of this study are as follows: (1) to evaluate green space exposure levels and inequalities in Chinese cities; (2) to identify the principal urban factors contributing to green space inequality, with a focus on urbanization; (3) to elucidate the implications of China’s distinctive land finance phenomenon on green space inequalities; and (4) to formulate strategies for addressing deficiencies and inequalities in urban green space exposure.
2. Materials and Methods
2.1. Large Cities and Income Groupings
In this research, we employed the 2020 Statistical Yearbook of Urban Construction in China by the National Bureau of Statistics of China as our primary data source and selected 140 cities (including municipalities and prefectural-level cities) within mainland China with built-up areas exceeding 100 km2 as the criterion for inclusion in the sample. According to previous studies [8,34], the whole of China is divided into seven regions, namely, Northeast China, North China, East China, South China, Central China, Northwest China, and Southwest China. The regional divisions are primarily based on two considerations. First, geographic divisions have holistic characteristics. Natural factors, such as mountains and rivers, together with human elements, like provinces, cities, and villages, jointly shape the spatial integrity of the region. Secondly, there are significant differences between geographic regions. China is vast, and there are notable regional differences in locations, natural conditions, climate, population density, industrial structures, and economic levels.
This approach to city selection was deliberately chosen to prevent the inadvertent exclusion of cities with extensive built-up areas but relatively small populations [35], thus enhancing the rigor and precision of our sample selection. To categorize these 140 cities, we referenced the Statistical Bulletin of the People’s Republic of China on National Economic and Social Development in 2020 [36], and according to the five-level income grouping of the national residents, all surveyed households are arranged in order of per capita disposable income from low to high and divided into five equal parts. The lowest 20% of income households are in the low-income group, and similarly, the city is divided into five groups: low income, lower-middle income, middle income, upper-middle income, and high income (Figure 1).
2.2. Green Space, Green Space Exposure Levels, and Inequalities Assessment
To ensure the precision of the green space exposure levels and inequality assessment results and maintain spatial comparability, we rigorously followed a three-step approach to establish the assessment framework, as illustrated in Figure 2. These steps included (1) green space mapping, (2) the evaluation of green space exposure levels, and (3) the computation of the green space Gini coefficient.
2.2.1. Green Space
The green space provision assessment system incorporated the normalized difference vegetation index (NDVI) and green space coverage area (GA). The NDVI serves as a gauge of vegetation coverage within large urban areas, offering insights into vegetation growth conditions. Meanwhile, the GA includes the areas of various green spaces used for gardening and greening, including park green spaces, production green spaces, protective green spaces, ancillary green spaces, and other green spaces, and is used to reflect the utilization of green spaces in large urban built-up areas. The NDVI data were sourced from the Google Earth Engine (GEE), derived from 2020 Landsat 5/7/8 remote sensing data, was rigorously processed to eliminate cloud and shadow interference, and possessed a spatial resolution of 30 m [37]. Additionally, we extracted information pertaining to GAs, including tree cover, shrubland, grassland, wetland, moss, and lichen, within the selected large cities’ built-up areas, utilizing the 1 m spatial resolution Chinese land cover dataset [38]. Notably, SinoLC-1, based on a deep learning framework, was developed using data from Global Land Cover (GLC) products, Open Street Map (OSM), and Google Earth imagery, achieving an overall accuracy of 73.61%. In this study, we employed the 2020 10 m resolution built-up area dataset generated through GEE and Sentinel data calculations [39]. Additionally, we utilized the 2020 China 100 m spatial resolution population data from the Worldpop website to compute various UGS variables.
2.2.2. Green Space Exposure Level Assessment
The percentage of GA (%GA) in built-up areas is a key assessment criterion for green cities, as recognized by the World Health Organization (WHO) [40]. It is computed by dividing the urban green space area by the urban built-up area. The data for this indicator were sourced from the China Urban Construction Statistics Yearbook.
Another crucial indicator for gauging green space exposure levels in large cities is population-weighted green space exposure [41]. To calculate this indicator, we extracted data to the center of the mass of a grid created for urban built-up areas. Subsequently, we utilized the population distribution within the grid to depict green space exposure levels [42]. During the data processing, all individuals were considered to be located at the center of the mass of the grid. The calculation for this indicator is as follows:
(1)
where denotes the green space exposure of the selected large city in 2020, denotes the sum of population data within the grid, and denotes the greenery level within a specific buffer zone around the center of the mass of the grid.In particular, we developed two indicators, namely, GENDVI and GEGA, to appraise the degree of green space exposure in each city. These indicators were quantitatively assessed using the summation of the population, the average NDVI, and the average GA within the buffer zone of the urban grid. Given the context of low green space exposure levels and high population density in mainland China, we designated the area of each grid as 1 km2 [41].
2.2.3. Assessment of Green Space Exposure Inequalities
Chen et al. [43] utilized the Gini coefficient to quantify inequalities in green space exposure within densely populated urban regions. The methodology, depicted in Figure 3, involves calculating the green space Gini coefficient based on the ratio of two areas: the area between the equality line and the Lorenz curve, and the area beneath the equality line. This calculation is conducted on a plot where the x-axis represents the cumulative distribution of the population, ordered by increasing green space exposure, and the y-axis reflects the cumulative proportion of green space exposure in a city. The green space Gini coefficient is computed using the following formula:
(2)
where n denotes the total population within the built-up area of the selected city, denotes the cumulative proportion of the population in the grid, and denotes the cumulative proportion of green space exposure possessed by the corresponding population in the grid. The green space Gini coefficient varies between 0 (indicating perfect equality) and 1 (indicating absolute inequality). A lower Gini coefficient signifies a more equitable distribution of green space exposure among city residents. We categorized the green space Gini coefficients into five brackets: low inequality (0–0.4), moderate inequality (0.4–0.55), intermediate inequality (0.55–0.7), high inequality (0.7–0.85), and very high inequality (0.85–1).Using this formula, we computed the green space Gini coefficients for all the selected large cities. Subsequently, two green space assessment indicators, namely, GiniNDVI and GiniGA, were derived from the mean values of NDVI and GA, respectively, within each population grid.
2.3. Correlation of Urban Factors with Green Space Exposure Levels and Inequalities
2.3.1. Urban Factors
To investigate the relationship between green space provision and the natural and socio-economic characteristics of large cities, we collected a total of 14 factors, as detailed in Table 1. These factors were grouped into three categories: natural factors (variables 1–3), economic factors (variables 4–11), and social factors (variables 12–14). Previous studies have established links between most of these factors and UGS exposure levels or inequalities [13,15,44,45]. However, to mitigate the potential effect of the coronavirus disease 2019 (COVID-19) pandemic on data accuracy [46,47], we opted to utilize 2019 economy-related data rather than 2020 data.
2.3.2. Correlation Analysis
In this study, the UGS variables include the five indicators detailed in Section 2.2, namely, green space exposure levels (%GA, GENDVI, and GEGA) and green space exposure inequality (GiniNDVI and GiniGA). To investigate the associations between urban green space provision and pertinent factors, we employed Spearman correlation analyses and a geographic detector in spatial stratified heterogeneity (SSH) analyses, following the approach outlined by Song et al. [49].
Spearman Correlation
Spearman correlation analysis, a bivariate technique, has been employed by various research teams to examine the relationships between UGS variables and urban factors [13,15,50]. One advantage of selecting Spearman is its focus on the relative rank of values, making it particularly suitable when assessing the importance of the rank order of values rather than the exact differences in the original value series [51]. This approach proved effective in systematically investigating the associations between the natural, economic, and social factors of the selected cities and the UGS variables, yielding reasonably accurate correlation coefficients.
Geographical Detector
A segment-based geographical detector model offers a valuable approach for analyzing the correlations and variations between UGS variables and urban factors across cities belonging to different income brackets and geographical regions. The geographic detector approach, notable for not assuming linearity in the relationship between explanatory and response variables [41]), has found widespread application in diverse contexts of SSH, including land cover, carbon emissions, and urban green space [52].
In this study, we employed two geographic detectors: a factor detector model and an interaction detector model. The factor detector employs the Q value (0 ≤ Q ≤ 1) as a metric to depict the relative significance of explanatory variables in elucidating the spatial heterogeneity of response variables [41]. In essence, it helps assess the relative importance of UGS variables in explaining urban factors, with a higher Q value denoting a stronger connection between the UGS indicators and the urban factors. Subsequently, an interaction detector was utilized to identify any synergistic effects (joint effects) between urban factors and UGS variables, thereby uncovering interactions between the explanatory and response variables. In this study, the interaction detector was applied to compare the importance (Q-value) of combined variables against the importance of each independent variable [52], yielding the final test results.
3. Result
3.1. Green Space Provision
3.1.1. Green Space Exposure in the Chinese Mainland
In this study, %GA, GENDVI, and GEGA were utilized to assess green space exposure levels in selected large cities within mainland China. Figure 4 and Figure 5 depict the green space exposure levels in these cities, represented by colored circles ranging from red to green and varying in size. Box plots illustrate the distribution of green space exposure levels across the seven geographic regions and income brackets. The Supplementary Materials includes codes for the seven geographic regions and income divisions, along with %GA, GENDVI, and GEGA statistics for each geographic region and income class in the Supplementary Data Table S2 and Table S3, respectively. Among these three UGS indicators, %GA and GENDVI (standard deviation < 0.05) exhibit relatively stable patterns, while GEGA (standard deviation = 0.197) shows more significant fluctuations. The differences observed between %GA, sourced from the China Urban Statistical Yearbook, and the variables GENDVI and GEGA can be attributed to variations in the statistical methodologies and data sources.
The research data reveal disparities in urban green space exposure levels across different geographic regions and income divisions. When arranging the green space exposure data in descending order, the top three geographic regions for %GA are East China (0.398), South China (0.397), and Northeast China (0.390). Regarding GENDVI, the top three regions are Southwest China (0.281), Central China (0.251), and South China (0.248), while for GEGA, the top three regions are Northeast China (0.289), Southwest China (0.257), and Northwest China (0.254). These findings indicate that large cities in South, Southwest, and Northeast China exhibit a distinct advantage in green space exposure, while North China shows comparatively lower urban green space exposure levels. Additionally, North China (0.068) and East China (0.084) exhibit significantly lower urban exposure levels in the GEGA indicator compared to the other five geographic regions.
When considering income brackets, it can be observed that %GA and GENDVI tend to be lower in cities with lower income. Nonetheless, the differences in %GA and GENDVI among the three income divisions are minimal and not notably distinct. In contrast, the GEGA indicators reveal a substantial variation. Cities in the upper-middle-income bracket (0.075) exhibit less than half the GEGA of cities in the middle- and lower-middle-income brackets (0.184 and 0.174). Additionally, the standard deviation of GEGA data for cities in the upper-middle-income division (0.080) is significantly smaller than that of cities in the middle- and lower-middle-income divisions (0.213 and 0.232). This observation affirms that the trend of cities in the upper-income division having markedly smaller GEGA than those in other income brackets is not isolated but rather a widespread phenomenon.
3.1.2. Inequality in Green Space Provision
In this study, the green space Gini coefficients (GiniNDVI and GiniGA) were employed to assess inequalities in green space exposure among the selected large cities within mainland China. As depicted in Figure 6, the colored circles, ranging from red to green and varying in size, represent the green space Gini coefficients, with smaller values indicating a more equitable distribution of green space. The box plots illustrate the distribution of green space Gini coefficients across seven geographic regions and income divisions. The Supplementary Materials provides the GiniNDVI and GiniGA statistics for each geographic region and income division, which can be found in Tables S2 and S3. In terms of results, most of the cities assessed show green space inequality (Gini > 0.7), with 90 cities for the GiniNDVI indicator and 84 cities for the GiniGA indicator.
According to the research data, disparities in the Gini coefficient of UGS exist across different geographic regions and income divisions. When considering Gini coefficients from small to large, the three geographic regions with the lowest GiniNDVI values are South China (0.672), Central China (0.709), and Southwest China (0.718). Similarly, the three geographic regions with the lowest GiniGA values are Southwest China (0.685), Central China (0.708), and South China (0.715). This indicates that these three geographical regions exhibit a more equitable distribution of green space exposure based on the green space Gini coefficient assessment within mainland China. However, it is noteworthy that the mean values of GiniNDVI and GiniGA for the vast majority of geographical regions (six-sevenths) are above 0.7, indicating a higher level of inequality. Only two regions have green space Gini coefficients in the medium inequality range, with the average value of GiniNDVI in South China at 0.672 and the average value of GiniGA in Southwest China at 0.685. This underscores the prevalence of inequalities in UGS provision in large cities across China. For income brackets, there is no significant difference in the GiniNDVI indicators observed among large cities, with values differing by less than 0.015. However, cities in the upper-middle-income bracket tend to have higher GiniGA (0.792), where GEGA (0.075) in this income bracket is less than half of the other income brackets (see Table S3 in Supplementary Data).
3.2. Linkages Between Urban Factors and Green Space Exposure Levels and Inequalities
In this study, SSH analyses were conducted at multiple levels, including mainland China, seven geographic regions, and income brackets, with the aim of elucidating the connections and correlations between UGS variables and urban factors.
3.2.1. Correlation Analyses in Mainland China
As shown in Figure 7, Spearman correlation was performed at the national level in this analysis. A heatmap was employed to visualize the correlation between UGS variables (%GA, GENDVI, GEGA, GiniNDVI, and GiniGA) and urban factors, and only significant coefficients at p < 0.05 are displayed. In the heatmap, blue signifies a positive correlation, red indicates a negative correlation, and gray represents no significant correlation. Detailed statistical data for all UGS variables and urban factors can be found in Table S1 in the Supplementary Materials.
Table 2 presents the results of two sets of SSH analyses, namely, the factor detector and interaction detector, conducted for the selected cities on a nationwide scale. Factors deemed significant (p < 0.05) by the factor detector were considered robust explanations for the correlations between UGS variables and urban factors (natural, economic, and social), and these significant factors are listed in Table 2. In the interaction detector analysis, it was observed that all response variables (UGSs) of the selected large cities exhibited nonlinear-enhanced or bi-enhanced interaction effects with their corresponding explanatory variables (14 urban factors). The nonlinear-enhanced interaction effect indicates that the explanatory power of the interaction between two factors is greater than the sum of the explanatory power of the two individual factors. The nonlinear enhanced interaction effect means that the interaction between two factors contributes more to the explanatory power of the two independent factors, while the bi-enhanced interaction effect suggests that the interaction of the two factors enhances the explanatory power of either of the independent factors (further details can be found in the Supplementary Data). At the national level, the Q values present a difference. Typically, temperature (Temp) and precipitation (PPTN) have the largest associations with green space exposure and inequality, and these two factors also play a predominant role in the interactions. Anyway, Temp and PPTN are particularly evident in the case of the GENDVI indicator, where the Q values for Temp and PPTN reach 0.578 and 0.470, respectively.
Combined with these two nationwide analyses mentioned above, it can be concluded that adequate and equitable urban green spaces are more probably to exist under the following conditions: suitable climate (Temp and PPTN), rapid economic development (GRPPC), fast urbanization (VBA, LF, and LT), and better residential experience (ANRPH and AFHPC).
3.2.2. Correlation Analyses Across Geographical Regions
In this research, SSH correlation analyses were conducted for each of the seven geographic regions in China. However, due to the limited number of samples in some geographic regions (with fewer than 10 samples in Northwest China), the results of the analyses in certain regions may not have achieved optimal statistical power. The average Q values (p < 0.05) for the UGS variables in the seven geographic regions are summarized in Table S4 in the Supplementary Materials. More detailed results of the factor detector data across geographic regions can be found in Table S5.
In Northeast and Southwest China, no urban factors were found to be significantly correlated (p < 0.05) with the UGS variables. In East China, where the largest urban sample size is located, climate and economic development become significant factors affecting urban green space exposure. In Northwest China, the results demonstrate that CPM, GRPPC1, LF5, and LT1 are strongly correlated with each other in terms of green space exposure levels (p < 0.001 and Q > 0.990), and VBA1 was found to be strongly correlated with green space inequality (p < 0.001 and Q > 0.990). Table S6 in the Supplementary Materials summarizes the top three nonlinear-enhanced interaction effects of urban factors. It was observed that in most geographical regions, climate and land finance play a predominant role in the interactions.
3.2.3. Correlation Analyses Across Income Brackets
Tables S7 and S8 in the Supplementary Materials show the summary and complete results of the factor detector models across income brackets. It becomes evident that Temp and PPTN are significantly and consistently linked to urban green space provision across all income divisions. In the middle- and lower-middle-income divisions, the housing conditions of the inhabitants (ANRPH and AFHPC) also exhibit significant explanatory power for green space exposure. In the middle-income bracket, which includes the largest number of cities, most of the urban factors (eleven-fourteenths) are significantly correlated (p < 0.05) with UGS patterns. Table S9 in the Supplementary Materials presents the results of the interaction detector for different income divisions. The UGS patterns in the upper-middle-income bracket are still mainly nonlinearly enhanced by the interactions of climatic and other factors, while the influence of economic and social factors becomes progressively more pronounced in the lower-income divisions.
4. Discussion
Urban green space provision serves as a pivotal indicator in the pursuit of the United Nations Sustainable Development Goals (SDGs), particularly SDG 11.7, and plays a crucial role in advancing the construction of ecological civilizations in cities. In this comprehensive study, leveraging high-resolution remote sensing data spanning the entire nation, we employed five carefully selected UGS variables in conjunction with an array of 14 urban factors and rigorously evaluated green space exposure levels and inequalities in large cities. These selected cities (built-up areas > 100 km2) can often be taken as exemplars of urban development in China, offering a unique lens through which to appraise green space provision and its interplay with urban factors.
4.1. The Correlation Between UGS Indicators and Urban Factors in Mainland China
As an essential public natural resource, UGS plays a pivotal role in enhancing the ecological functions of cities and has a profound effect on the well-being and social welfare of urban populations [53]. The findings reveal a prevalent deficiency in UGS variables across the majority of large cities, and the mean values of variables are not satisfactory (GENDVI = 0.232, GEGA = 0.152, GiniNDVI = 0.733, and GiniGA = 0.748). These basic findings are consistent with research showing that the available UGS primarily caters to a select few, resulting in substantial inequalities in green space exposure [41]. An examination of the correlations between UGS variables and the 14 urban factors for large cities reveals several noteworthy insights. Favorable natural conditions, characterized by adequate precipitation and favorable temperatures, tend to facilitate the provision of better green spaces in urban areas. Simultaneously, improved UGS provision contributes to better natural conditions, as evidenced by lower PM2.5 indicators. This result ties well with previous studies on urban green space equity [12,13]).
Additionally, the urban factors representing eight distinct economic types are all positively correlated with urban green space exposure levels and negatively correlated with green space exposure inequalities. This implies that rapid urban development and higher per capita GRP tend to lead to more robust UGS provision, in line with previous studies [6,16]. Furthermore, the analysis of China’s unique land finance policy affirms this trend, suggesting that cities relying more on land finance tend to exhibit better green space exposure levels and equitabilities. It must be pointed out that our findings appear to be different from previous studies that have argued that land finance is not conducive to sustainable development [32,33]). These results underscore the potential for ecological preservation and enhanced UGS supply to complement and enhance economic development, representing a win-win situation. Cities where land grant premiums constitute a larger proportion of total annual budget revenue tend to enjoy more abundant and equitable green space exposure (Figure 7). We noticed that China adopts the strategy of “encouraging those having got prosperous first to help others catch up”, and the economic development level of each large city is also different. Thus, the results might be explained by the “Environmental Kuznets Curve”, which proposes that as a region’s economy advances, environmental pollution tends to follow a trajectory from low to high, eventually reversing course as the economy reaches a certain level of development, leading to a decline in environmental pollution [54]. Cities in the early stages of urban development, often characterized by higher land finance figures, exhibit a more robust preservation of UGS supply. Possibly, this phenomenon can be attributed to the relatively short duration of their development, which has resulted in a greater retention of original green spaces. Also, policies related to green and sustainable development can facilitate more effective and equitable patterns of urban green space [55]. Recent government initiatives, such as the Ministry of Construction’s directive mandating at least 30% public green space in new residential areas, have led to an increased supply of UGS, especially in regions with larger land grant area proportions. For instance, it has been proven that greening policies are the drivers of the improvement in UGS patterns in Shanghai [56].
Our results reveal that cities with larger growth in built-up areas and lower levels of education have a greater availability of UGS. And better housing conditions for urban dwellers signify more reasonable population densities within a city, therefore amplifying UGS exposure for residents and enhancing the equitability of green space exposure among the population. Unlike the perception that rapid urbanization hinders green space equity [18], we found that urban sprawl and improved residential conditions provide equitable UGS provision, echoing the findings of some other studies [16,17]. For example, less-educated populations and households with a lower living area are not exposed to green space equitably in Guangzhou [6]. Also, cities with higher levels of education predominantly belong to earlier development stages, where green space provisions are possibly not as comprehensive as in cities that have experienced more rapid recent development.
4.2. Marked Disparities in UGS Across Geographical Regions and Income Brackets
The study’s findings underscore the presence of geographic heterogeneity in the spatial distribution of urban green space supply across China. In most instances, these urban areas grapple with limited overall green space coverage. Usually, lower green space exposure (GE) tends to coincide with higher green space inequality (Gini), one prime example of which is Southwest China (with the highest GE and second lowest Gini among all geographical regions). More specifically, cities situated in the north and northeast regions of the country are inclined to exhibit lower levels of green space exposure and higher green space Gini coefficients. These regions are primarily characterized by arid and semiarid climates, featuring low average temperatures, extended winters, significant seasonal temperature variations, and limited and uneven annual precipitation [57]. These environmental conditions pose challenges to the growth of green vegetation and necessitate greater human and material resources to maintain even modest levels of green space provision. For cities in these ecologically fragile regions, implementing strategies to enhance urban water use efficiency through low-carbon and energy-saving methods is crucial [58]. Such measures may include the development of water-saving green space irrigation systems, the adoption of reclaimed water irrigation techniques, and the selection of drought-resistant vegetation for planting, among others.
Cities in Southwest China stand out, with notable advantages across UGS variables (the highest GE and second lowest Gini) compared to other geographic regions, indicating that they boast the most robust green space provision in China. Nevertheless, the green space exposure levels in East China (GEGA = 0.084) do not align with the region’s favorable natural conditions. This discrepancy can be attributed to East China’s status as an early epicenter of economic development in China, characterized by extensive urban development during a period when the emphasis on sustainable green development was less pronounced [59]. Consequently, disparities in UGS provision between the urban core and fringe areas in East China have emerged, contributing to lower-than-anticipated green space exposure [55]. Additionally, we found that the results related to GEGA and GiniGA from the high-resolution China land cover dataset indicate that cities in higher-income divisions exhibit lower levels of green space supply and equality than their counterparts in lower-income divisions. Such a circumstance can be primarily attributed to the earlier period of extensive and unreasonable urban expansion in China [5]. Newly developed cities may have an advantage over developed cities in terms of green space exposure and equity [55]. In light of these findings, municipalities in the middle- and upper-middle-income categories, which are financially more robust but exhibit slightly lower green space provision, should prioritize the preservation of seminatural vegetation [60] and undertake efforts to revitalize sites and introduce greenery to less green areas.
4.3. The Relative Roles Played by Urban Factors Are Diverse at the Levels of the Chinese Mainland, Geographical Regions, and Income Brackets
The results of the factor detector and interaction detector models reveal varying roles played by urban factors at these three levels. At the national level, it is evident that temperature and precipitation significantly influence green space provision (Table 2), which has already been proven by previous empirical findings [12,13]. Fundamental climatic conditions, such as suitable temperatures and adequate precipitation, are foundational prerequisites for green vegetation growth and contribute to a favorable green living environment for urban residents. Regional economic development is intrinsically linked with UGS variables, as higher levels of regional economic development are conducive to maintaining green space provision. For socio-economic aspects, we identified no particularly notable factors that have an especially significant influence on the UGS patterns, but most of them play a role. The expansion and economic growth of cities (changes in built-up area and land finance) have a strong explanatory power on the provision and equity levels of green space, which ties in well with previous studies on the connection between UGS and socio-economic factors [16,17].
At the geographic–regional level, as revealed in Tables S5 and S6 in the Supplementary Materials, the correlation between green space provision and natural factors becomes less dominant for certain cities within a given region in contrast to the mainland China analysis results. Instead, the explanatory power and interaction of economic and social factors in cities within a region become more pronounced (see Table S7 in the Supplementary Data), similar to the studies that focus more on socio-economic aspects when they select more detailed study areas, such as Guangzhou and Wuhan [6,23]. Regarding income brackets, as depicted in Tables S8 and S9 in the Supplementary Materials, natural factors continue to be the primary explanatory factors, while social factors have also turned out to be very important. As demonstrated by a case study of Europe [61], favorable housing conditions represent a reasonable population density, and large cities with less population pressure tend to have adequate and equitable UGSs [13].
4.4. Limitations
This study employed high-resolution remote sensing data to evaluate UGS provisions in mainland China’s major cities and identified two key drivers to establish correlations between UGSs and urban factors, particularly economic and social variables. Nonetheless, it is crucial to acknowledge certain limitations in this investigation.
First, the study did not disaggregate the quality of UGS. It is noteworthy that the quality of green space is a crucial factor that should not be overlooked [62]. Second, the model’s explanation of the assessment of accessible green space under dynamic population movement within a city is insufficient. Failure to account for green space accessibility hampers a comprehensive understanding of the study [63]. Third, the selection of a single year (2020) for UGS assessment hinders the exploration of changes in urban green space provisions over time in the selected cities, resulting in the absence of comparative studies of changes before and after the development. Next, when creating buffers (grids) for the city, the study exclusively utilized 1 km2 grids for UGS assessment. A more comprehensive comparative analysis could have been conducted by considering both larger and smaller buffers (grids) [43]. Finally, although the study mentions education levels as a factor, it does not sufficiently explore the role of ecological education in enhancing public awareness and engagement with UGS [64,65]. Ecological education could be an important element in fostering a deeper understanding of the value of UGS. While the study underscores the importance of public awareness and engagement, it misses an opportunity to advocate for integrating ecological education into urban planning. Programs that teach residents about the benefits of UGS could enhance both their use and preservation.
5. Conclusions
This study conducted an assessment of green space exposure levels and inequalities in 140 major cities in mainland China, utilizing high-resolution remote sensing data. The analysis included various urban factors, including natural, economic, and social aspects. The findings reveal that large cities in mainland China exhibit both low and unequal levels of UGS provision. Spatial heterogeneity was observed among cities, with those in the southern part of China generally possessing better UGS provision and equality, while cities in North China and Northwest China tended to have poorer levels. Moreover, cities in the process of recent development exhibited an advantage in terms of green space provision compared to already developed cities. A positive relationship between the economic development of cities and the adequacy and equity of UGS supply for residents was found, with fast-growing cities fairly reliant on land finance tending to have more adequate and equitable UGS provision. At the national level, temperature and precipitation are the most significant predictors of UGSs. Nonetheless, in regional and income analyses, the influence of socio-economic factors on UGS patterns is also quite significant. The study suggests that rapid urban development does not necessarily lead to a shortage of UGS provision. Effective and sustainable urban green development policies, tailored to local conditions and implemented correctly, can promote the coexistence of people and ecosystems. We recommend that local municipalities in China implement adaptive greening interventions adjusted to different local circumstances to enhance green space exposure levels and reduce green space inequalities.
Conceptualization, B.H. and J.G.; methodology, B.H. and Z.F.; software, M.Z.; validation, B.H., J.G. and Z.F.; formal analysis, M.Z.; investigation, J.G. and Z.F.; resources, B.H.; data curation, M.Z.; writing—original draft preparation, J.G. and Z.F.; writing—review and editing, B.H. and Z.F.; visualization, M.Z.; supervision, B.H.; project administration, B.H.; funding acquisition, B.H. All authors have read and agreed to the published version of the manuscript.
The raw data supporting the conclusions of this article will be made available by the authors on request.
The authors declare no conflicts of interest.
Footnotes
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Enlarge this image.Figure 2. Conceptual framework for assessing the correlation between green space provision and urban factors. Distribution of large cities in mainland China with an urban built-up area exceeding 100 square kilometers (a); Urban areas divided into grids for spatial analysis based on SinoLC-1, WorldPop, and NDVI data (b); Methodology for calculating green space exposure levels (%GA and GE) and green space exposure inequality (Gini Index) (c); Description of using Spearman correlation analysis and geographic detectors to identify the driving factors of green space exposure (d); Classification of driving factors into natural, economic, and social variables, among other urban factors (e).
Enlarge this image.Figure 3. Illustrative diagram of the Gini index of green space exposure inequality. As illustrated in Figure 3, the x-axis represents the cumulative share of residents arranged by their greenspaces in ascending order. The y-axis represents the cumulative share of greenspaces. The diagonal indicates a utopian condition of greenspace distribution. The Gini coefficient is defined as the ratio of the area bounded by the equality line and Lorenz curve to the area below the equality line.
Enlarge this image.Figure 4. Spatial distribution of %GA across large cities on the Chinese mainland (a); box plots of %GA aggregated by geographical regions (b); and income brackets (c).
Enlarge this image.Figure 5. Spatial distribution of GE calculated by GENDVI (a) and GEGA (b) across large cities on the Chinese mainland; box plots of GENDVI and GEGA aggregated by geographical regions (c); and income brackets (d).
Enlarge this image.Figure 6. Spatial distribution of the Gini coefficient calculated by GiniNDVI (a) and GiniGA (b) across large cities on the Chinese mainland; box plots of GiniNDVI and GiniGA aggregated by geographical regions (c); and income brackets (d).
Enlarge this image.Figure 7. Bivariate correlations between green space evaluation indicators (%GA, GENDVI, GEGA, GiniNDVI, and GiniGA) and urban factors.
Descriptions of urban factors.
| No. | Variables (Unit) | Code | Measure | Source |
|---|---|---|---|---|
| 1 | Mean temperature (°C) | Temp | Average annual temperature in built-up areas of cities in 2020 | National Tibetan Plateau Data Center |
| 2 | Annual Precipitation (mm) | PPTN | Annual precipitation in the built-up area of each city in 2020 | |
| 3 | Concentrations of PM2.5 (ug/m3) | CPM | Annual average concentration of PM2.5 in built-up areas of cities in 2020 | ChinaHighPM2.5 dataset [ |
| 4 | Variation in GRP per Capita from 2018 to 2019 ( | GRPPC1 | Ratio of GRP per capita from 2019 to 2018 in each city | China City Statistical Yearbook, 2017, 2020, 2021 |
| 5 | Variation in GRP per Capita from 2015 to 2019 ( | GRPPC5 | Ratio of GRP per capita from 2015 to 2019 in each city | |
| 6 | 1-year change in built-up area ( | VBA1 | Ratio of built-up area from 2019 to 2020 in each city | China Urban Construction Statistical Yearbook, 2017, 2020, 2021 |
| 7 | 5-year change in built-up area ( | VBA5 | Ratio of built-up area from 2016 to 2020 in each city | |
| 8 | 1 year of land finance ( | LF1 | Ratio of land grant premiums to general budget revenue from local finances in 2019 | China Land Market Network ( |
| 9 | 5 years of land finance ( | LF5 | Ratio of the sum of land grant premiums (2015–2019) to the sum of general budget revenue from local finances (2015–2019) | |
| 10 | 1 year of land transfer ( | LT1 | Ratio of land transfer area in 2019 to built-up area in 2020 | China Land Market Network ( |
| 11 | 5 years of land transfer ( | LT5 | Ratio of the sum of land transfer area (2015–2019) to built-up area in 2020 | |
| 12 | Average number of rooms per household ( | ANRPH | Average number of rooms per household for residents by city in 2020 | Seventh national census |
| 13 | Average floor area of housing per capita ( | AFHPC | Average floor area of housing per inhabitant by city in 2020 | |
| 14 | Average years of education (year) | AYE | Years of schooling per capita by city in 2020 |
Results of SSH analyses of selected cities in Chinese mainland.
| UGS Variables | Factor Detector | Interaction Detector | ||
|---|---|---|---|---|
| Urban Factors | Q Value | |||
| %GA | PPTN | 0.093 * | Temp∩VBA1 | 0.308 |
| VBA1 | 0.083 * | VBA1∩LT5 | 0.256 | |
| AFHPC | 0.077 * | PPTN∩VBA1 | 0.256 | |
| Temp | 0.074 * | VBA1∩AFHPC | 0.254 | |
| CPM∩VBA1 | 0.248 | |||
| VBA1∩VBA5 | 0.240 | |||
| GENDVI | Temp | 0.578 *** | Temp∩CPM | 0.684 |
| PPTN | 0.470 *** | Temp∩AYE | 0.655 | |
| AFHPC | 0.265 *** | PPTN∩CPM | 0.580 | |
| GRPPC1 | 0.206 *** | PPTN∩AYE | 0.563 | |
| ANRPH | 0.198 *** | PPTN∩LT1 | 0.549 | |
| VBA5 | 0.191 *** | CPM∩AFHPC | 0.456 | |
| GRPPC5 | 0.158 *** | |||
| LF5 | 0.151 *** | |||
| LT5 | 0.125 *** | |||
| LF1 | 0.121 ** | |||
| VBA1 | 0.104 ** | |||
| AYE | 0.068 * | |||
| GEGA | CPM | 0.113 * | Temp∩AFHPC | 0.237 |
| PPTN∩ANRPH | 0.205 | |||
| CPM∩GRPPC1 | 0.195 | |||
| CPM∩AFHPC | 0.187 | |||
| CPM∩LF5 | 0.173 | |||
| GRPPC5∩AFHPC | 0.171 | |||
| GiniNDVI | Temp | 0.252 *** | PPTN∩ANRPH | 0.401 |
| PPTN | 0.170 *** | Temp∩AFHPC | 0.385 | |
| CPM | 0.106 ** | Temp∩GRPPC5 | 0.379 | |
| GRPPC1 | 0.103 ** | PPTN∩GRPPC5 | 0.357 | |
| GRPPC5 | 0.101 ** | Temp∩ANRPH | 0.349 | |
| VBA5 | 0.098 ** | Temp∩VBA1 | 0.340 | |
| ANRPH | 0.097 ** | |||
| LF1 | 0.089 ** | |||
| LF5 | 0.083 * | |||
| GiniGA | PPTN | 0.135 *** | PPTN∩LT5 | 0.312 |
| Temp | 0.097 ** | PPTN∩VBA1 | 0.288 | |
| VBA1 | 0.095 ** | PPTN∩LT1 | 0.269 | |
| ANRPH | 0.095 ** | VBA5∩ANRPH | 0.265 | |
| VBA5 | 0.089 * | PPTN∩LF5 | 0.241 | |
| LT5 | 0.068 * | PPTN∩VBA5 | 0.233 | |
p < 0.05 (*). p < 0.01 (**). p < 0.001 (***).
Supplementary Materials
The following supporting information can be downloaded at:
References
1. UN. Transforming Our World: The 2030 Agenda for Sustainable Development; United Nations: New York, NY, USA, 2015.
2. Leo, D.N.; Escobedo, J.F.; Dubbeling, M. The role of urban green infrastructure in mitigating land surface temperature in Bobo-Dioulasso, Burkina Faso. Environ. Dev. Sustain.; 2016; 18, pp. 373-392. [DOI: https://dx.doi.org/10.1007/s10668-015-9653-y]
3. Lo, A.Y.H.; Jim, C.Y. Citizen attitude and expectation towards greenspace provision in compact urban milieu. Land Use Policy; 2012; 29, pp. 577-586. [DOI: https://dx.doi.org/10.1016/j.landusepol.2011.09.011]
4. Sathyakumar, V.; Ramsankaran, R.A.A.J.; Bardhan, R. Linking remotely sensed Urban Green Space (UGS) distribution patterns and Socio-Economic Status (SES)-A multi-scale probabilistic analysis based in Mumbai, India. Gisci. Remote Sens.; 2019; 56, pp. 645-669. [DOI: https://dx.doi.org/10.1080/15481603.2018.1549819]
5. Tonne, C.; Adair, L.; Adlakha, D.; Anguelovski, I.; Belesova, K.; Berger, M.; Brelsford, C.; Dadvand, P.; Dimitrova, A.; Giles-Corti, B. et al. Defining pathways to healthy sustainable urban development. Environ. Int.; 2021; 146, 106236. [DOI: https://dx.doi.org/10.1016/j.envint.2020.106236]
6. Yang, W.; Yang, R.; Zhou, S. The spatial heterogeneity of urban green space inequity from a perspective of the vulnerable: A case study of Guangzhou, China. Cities; 2022; 130, 103855. [DOI: https://dx.doi.org/10.1016/j.cities.2022.103855]
7. Ma, Z.; Tian, X.; Zhang, P. Could ecological restoration reduce income inequality? An analysis of 290 Chinese prefecture-level cities. Ambio; 2023; 52, pp. 802-812. [DOI: https://dx.doi.org/10.1007/s13280-022-01815-y]
8. Huang, Y.; Lin, T.; Xue, X.; Zhang, G.; Liu, Y.; Zeng, Z.; Zhang, J.; Sui, J. Spatial patterns and inequity of urban green space supply in China. Ecol. Indic.; 2021; 132, 108275. [DOI: https://dx.doi.org/10.1016/j.ecolind.2021.108275]
9. Wen, C.; Albert, C.; Von Haaren, C. Equality in access to urban green spaces: A case study in Hannover, Germany, with a focus on the elderly population. Urban For. Urban Green.; 2020; 55, 126820. [DOI: https://dx.doi.org/10.1016/j.ufug.2020.126820]
10. Chen, Y.; Ge, Y.; Yang, G.; Wu, Z.; Du, Y.; Mao, F.; Liu, S.; Xu, R.; Qu, Z.; Xu, B. et al. Inequalities of urban green space area and ecosystem services along urban center-edge gradients. Landsc. Urban Plan.; 2022; 217, 104266. [DOI: https://dx.doi.org/10.1016/j.landurbplan.2021.104266]
11. Althoff, T.; Sosic, R.; Hicks, J.L.; King, A.C.; Delp, S.L.; Leskovec, J. Large-scale physical activity data reveal worldwide activity inequality. Nature; 2017; 547, 336. [DOI: https://dx.doi.org/10.1038/nature23018]
12. Chen, W.Y.; Wang, D.T. Urban forest development in China: Natural endowment or socioeconomic product?. Cities; 2013; 35, pp. 62-68. [DOI: https://dx.doi.org/10.1016/j.cities.2013.06.011]
13. Han, Y.; He, J.; Liu, D.; Zhao, H.; Huang, J. Inequality in urban green provision: A comparative study of large cities throughout the world. Sustain. Cities Soc.; 2023; 89, 104229. [DOI: https://dx.doi.org/10.1016/j.scs.2022.104229]
14. Chen, J.; Kinoshita, T.; Li, H.; Luo, S.; Su, D.; Yang, X.; Hu, Y. Toward green equity: An extensive study on urban form and green space equity for shrinking cities. Sustain. Cities Soc.; 2023; 90, 104395. [DOI: https://dx.doi.org/10.1016/j.scs.2023.104395]
15. Nesbitt, L.; Meitner, M.J.; Girling, C.; Sheppard, S.R.J.; Lu, Y. Who has access to urban vegetation? A spatial analysis of distributional green equity in 10 US cities. Landsc. Urban Plan.; 2019; 181, pp. 51-79. [DOI: https://dx.doi.org/10.1016/j.landurbplan.2018.08.007]
16. Wu, L.; Kim, S.K. Exploring the equality of accessing urban green spaces: A comparative study of 341 Chinese cities. Ecol. Indic.; 2021; 121, 107080. [DOI: https://dx.doi.org/10.1016/j.ecolind.2020.107080]
17. Lu, Y.; Chen, R.; Chen, B.; Wu, J. Inclusive green environment for all? An investigation of spatial access equity of urban green space and associated socioeconomic drivers in China. Landsc. Urban Plan.; 2024; 241, 104926. [DOI: https://dx.doi.org/10.1016/j.landurbplan.2023.104926]
18. Kong, F.; Yin, H.; Nakagoshi, N. Using GIS and landscape metrics in the hedonic price modeling of the amenity value of urban green space: A case study in Jinan City, China. Landsc. Urban Plan.; 2007; 79, pp. 240-252. [DOI: https://dx.doi.org/10.1016/j.landurbplan.2006.02.013]
19. Zhang, W.; Villarini, G.; Vecchi, G.A.; Smith, J.A. Urbanization exacerbated the rainfall and flooding caused by hurricane Harvey in Houston. Nature; 2018; 563, 384. [DOI: https://dx.doi.org/10.1038/s41586-018-0676-z]
20. Huang, Y.; Lin, T.; Zhang, G.; Jones, L.; Xue, X.; Ye, H.; Liu, Y. Spatiotemporal patterns and inequity of urban green space accessibility and its relationship with urban spatial expansion in China during rapid urbanization period. Sci. Total Environ.; 2022; 809, 151123. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2021.151123]
21. Wuestemann, H.; Kalisch, D.; Kolbe, J. Access to urban green space and environmental inequalities in Germany. Landsc. Urban Plan.; 2017; 164, pp. 124-131. [DOI: https://dx.doi.org/10.1016/j.landurbplan.2017.04.002]
22. Wu, S.; Chen, B.; Webster, C.; Xu, B.; Gong, P. Improved human greenspace exposure equality during 21st century urbanization. Nat. Commun.; 2023; 14, 6460. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37833296]
23. Jiao, H.; Li, C.; Yu, Y.; Peng, Z. Urban Public Green Space Equity against the Context of High-Speed Urbanization in Wuhan, Central China. Sustainability; 2020; 12, 9394. [DOI: https://dx.doi.org/10.3390/su12229394]
24. Wood, S.L.R.; Jones, S.K.; Johnson, J.A.; Brauman, K.A.; Chaplin-Kramer, R.; Fremier, A.; Girvetz, E.; Gordon, L.J.; Kappel, C.V.; Mandle, L. et al. Distilling the role of ecosystem services in the Sustainable Development Goals. Ecosyst. Serv.; 2018; 29, pp. 70-82. [DOI: https://dx.doi.org/10.1016/j.ecoser.2017.10.010]
25. Weng, C.; Bai, Y.; Chen, B.; Hu, Y.; Liu, S.; Wu, L. Virtual built-up land transfers and induced ecosystem service potential loss embodied in China’s inter-regional trade. Land Degrad. Dev.; 2024; 35, pp. 597-607. [DOI: https://dx.doi.org/10.1002/ldr.4938]
26. Wang, J.; Zhang, Y.; Zhang, X.; Song, M.; Ye, J. The spatio-temporal trends of urban green space and its interactions with urban growth: Evidence from the Yangtze River Delta region, China. Land Use Policy; 2023; 128, 106598. [DOI: https://dx.doi.org/10.1016/j.landusepol.2023.106598]
27. Li, F.; Wang, X.; Liu, H.; Li, X.; Zhang, X.; Sun, Y.; Wang, Y. Does economic development improve urban greening? Evidence from 289 cities in China using spatial regression models. Environ. Monit. Assess.; 2018; 190, 541. [DOI: https://dx.doi.org/10.1007/s10661-018-6871-4]
28. Wu, Q.; Li, Y.; Yan, S. The incentives of China’s urban land finance. Land Use Policy; 2015; 42, pp. 432-442.
29. Ye, L.; Wu, A.M. Urbanization, land development, and land financing: Evidence from Chinese cities. J. Urban Aff.; 2014; 36, pp. 354-368. [DOI: https://dx.doi.org/10.1111/juaf.12105]
30. Tong, D.; Chu, J.; MacLachlan, I.; Qiu, J.; Shi, T. Modelling the Impacts of land finance on urban expansion: Evidence from Chinese cities. Appl. Geogr.; 2023; 153, 102896. [DOI: https://dx.doi.org/10.1016/j.apgeog.2023.102896]
31. Liu, Y.; Fan, P.; Yue, W.; Song, Y. Impacts of land finance on urban sprawl in China: The case of Chongqing. Land Use Policy; 2018; 72, pp. 420-432. [DOI: https://dx.doi.org/10.1016/j.landusepol.2018.01.004]
32. Hou, S.; Song, L.; Wang, J.; Ali, S. How Land Finance Affects Green Economic Growth in Chinese Cities. Land; 2021; 10, 819. [DOI: https://dx.doi.org/10.3390/land10080819]
33. Zheng, H.; Wang, X.; Cao, S. The land finance model jeopardizes China’s sustainable development. Habitat Int.; 2014; 44, pp. 130-136. [DOI: https://dx.doi.org/10.1016/j.habitatint.2014.05.008]
34. Xie, H.; Du, L.; Liu, S.; Chen, L.; Gao, S.; Liu, S.; Pan, H.; Tong, X.; Li, S.; Chen, J. et al. Dynamic Monitoring of Agricultural Fires in China from 2010 to 2014 Using MODIS and GlobeLand30 Data. ISPRS Int. J. Geo-Inf.; 2016; 5, 172. [DOI: https://dx.doi.org/10.3390/ijgi5100172]
35. Sun, L.; Chen, J.; Li, Q.; Huang, D. Dramatic uneven urbanization of large cities throughout the world in recent decades. Nature Communications; 2020; 11. [DOI: https://dx.doi.org/10.1038/s41467-020-19158-1]
36. National Bureau of Statistics. Statistical Bulletin of the People’s Republic of China on National Economic and Social Development in 2020; National Bureau of Statistics: Beijing, China, 2021.
37. Yang, J.; Dong, J.; Xiao, X.; Dai, J.; Wu, C.; Xia, J.; Zhao, G.; Zhao, M.; Li, Z.; Zhang, Y. et al. Divergent shifts in peak photosynthesis timing of temperate and alpine grasslands in China. Remote Sens. Environ.; 2019; 233, 111395. [DOI: https://dx.doi.org/10.1016/j.rse.2019.111395]
38. Li, Z.; He, W.; Cheng, M.; Hu, J.; Yang, G.; Zhang, H. SinoLC-1: The first 1 m resolution national-scale land-cover map of China created with a deep learning framework and open-access data. Earth Syst. Sci. Data; 2023; 15, pp. 4749-4780. [DOI: https://dx.doi.org/10.5194/essd-15-4749-2023]
39. Sun, J.; Sun, Z.; Guo, H.; Wang, J.; Jiang, H.; Gao, J. A dataset of built-up areas of Chinese cities in 2020. China Sci. Data; 2022; 7, pp. 1-15. [DOI: https://dx.doi.org/10.11922/11-6035.csd.2021.0087.zh]
40. Pereira Barboza, E.; Cirach, M.; Khomenko, S.; Iungman, T.; Mueller, N.; Barrera-Gomez, J.; Rojas-Rueda, D.; Kondo, M.; Nieuwenhuijsen, M. Green space and mortality in European cities: A health impact assessment study. Lancet Planet. Health; 2021; 5, pp. E718-E730. [DOI: https://dx.doi.org/10.1016/S2542-5196(21)00229-1]
41. Song, Y.; Chen, B.; Ho, H.C.; Kwan, M.-P.; Liu, D.; Wang, F.; Wang, J.; Cai, J.; Li, X.; Xu, Y. et al. Observed inequality in urban greenspace exposure in China. Environ. Int.; 2021; 156, 106778. [DOI: https://dx.doi.org/10.1016/j.envint.2021.106778]
42. Song, Y.; Huang, B.; Cai, J.; Chen, B. Dynamic assessments of population exposure to urban greenspace using multi-source big data. Sci. Total Environ.; 2018; 634, pp. 1315-1325. [DOI: https://dx.doi.org/10.1016/j.scitotenv.2018.04.061]
43. Chen, B.; Wu, S.; Song, Y.; Webster, C.; Xu, B.; Gong, P. Contrasting inequality in human exposure to greenspace between cities of Global North and Global South. Nat. Commun.; 2022; 13, 4636. [DOI: https://dx.doi.org/10.1038/s41467-022-32258-4] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35941122]
44. Wang, R.; Feng, Z.; Pearce, J.; Yao, Y.; Li, X.; Liu, Y. The distribution of greenspace quantity and quality and their association with neighbourhood socioeconomic conditions in Guangzhou, China: A new approach using deep learning method and street view images. Sustain. Cities Soc.; 2021; 66, 102664. [DOI: https://dx.doi.org/10.1016/j.scs.2020.102664]
45. Laszkiewicz, E.; Kronenberg, J.; Mohamed, A.A.; Roitsch, D.; De Vreese, R. Who does not use urban green spaces and why? Insights from a comparative study of thirty-three European countries. Landsc. Urban Plan.; 2023; 239, 104866. [DOI: https://dx.doi.org/10.1016/j.landurbplan.2023.104866]
46. Ayittey, F.K.; Ayittey, M.K.; Chiwero, N.B.; Kamasah, J.S.; Dzuvor, C. Economic impacts of Wuhan 2019-nCoV on China and the world. J. Med. Virol.; 2020; 92, pp. 473-475. [DOI: https://dx.doi.org/10.1002/jmv.25706] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32048740]
47. Fernandes, N. Economic Effects of Coronavirus Outbreak (COVID-19) on the World Economy. 2020; Available online: https://ssrn.com/abstract=3557504 (accessed on 20 May 2022).
48. Wei, J.L.; Li, Z. ChinaHighPM2.5: Big Data Seamless 1 km Ground-Level PM2.5 Dataset for China. Zenodo; 2019; [DOI: https://dx.doi.org/10.5281/zenodo.6398971]
49. Song, Y.; Wang, J.; Ge, Y.; Xu, C. An optimal parameters-based geographical detector model enhances geographic characteristics of explanatory variables for spatial heterogeneity analysis: Cases with different types of spatial data. Gisci. Remote Sens.; 2020; 57, pp. 593-610. [DOI: https://dx.doi.org/10.1080/15481603.2020.1760434]
50. Artmann, M.; Chen, X.; Ioja, C.; Hof, A.; Onose, D.; Ponizy, L.; Lamovsek, A.Z.; Breuste, J. The role of urban green spaces in care facilities for elderly people across European cities. Urban For. Urban Green.; 2017; 27, pp. 203-213. [DOI: https://dx.doi.org/10.1016/j.ufug.2017.08.007]
51. Liu, Y. A Short Note on Spearman Correlation: Impact of Tied Observations. Available online: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2933193 (accessed on 17 July 2020).
52. Wang, J.-F.; Zhang, T.-L.; Fu, B.-J. A measure of spatial stratified heterogeneity. Ecol. Indic.; 2016; 67, pp. 250-256. [DOI: https://dx.doi.org/10.1016/j.ecolind.2016.02.052]
53. Lu, Y.; Chen, L.; Liu, X.; Yang, Y.; Sullivan, W.C.; Xu, W.; Webster, C.; Jiang, B. Green spaces mitigate racial disparity of health: A higher ratio of green spaces indicates a lower racial disparity in SARS-CoV-2 infection rates in the USA. Environ. Int.; 2021; 152, 106465. [DOI: https://dx.doi.org/10.1016/j.envint.2021.106465]
54. Shahbaz, M.; Ozturk, I.; Afza, T.; Ali, A. Revisiting the environmental Kuznets curve in a global economy. Renew. Sustain. Energy Rev.; 2013; 25, pp. 494-502. [DOI: https://dx.doi.org/10.1016/j.rser.2013.05.021]
55. Feng, F.; Yang, X.; Jia, B.; Li, X.; Li, X.; Xu, C.; Wang, K. Variability of urban fractional vegetation cover and its driving factors in 328 cities in China. Sci. China-Earth Sci.; 2024; 67, pp. 466-482. [DOI: https://dx.doi.org/10.1007/s11430-022-1219-2]
56. Wu, Z.; Chen, R.; Meadows, M.E.; Sengupta, D.; Xu, D. Changing urban green spaces in Shanghai: Trends, drivers and policy implications. Land Use Policy; 2019; 87, 104080. [DOI: https://dx.doi.org/10.1016/j.landusepol.2019.104080]
57. Shi, L.; Halik, U.; Abliz, A.; Mamat, Z.; Welp, M. Urban Green Space Accessibility and Distribution Equity in an Arid Oasis City: Urumqi, China. Forests; 2020; 11, 690. [DOI: https://dx.doi.org/10.3390/f11060690]
58. Topi, C.; Esposto, E.; Govigli, V.M. The economics of green transition strategies for cities: Can low carbon, energy efficient development approaches be adapted to demand side urban water efficiency?. Environ. Sci. Policy; 2016; 58, pp. 74-82. [DOI: https://dx.doi.org/10.1016/j.envsci.2016.01.001]
59. He, Q.; Zeng, C.; Xie, P.; Tan, S.; Wu, J. Comparison of urban growth patterns and changes between three urban agglomerations in China and three metropolises in the USA from 1995 to 2015. Sustain. Cities Soc.; 2019; 50, 101649. [DOI: https://dx.doi.org/10.1016/j.scs.2019.101649]
60. Zlender, V.; Thompson, C.W. Accessibility and use of peri-urban green space for inner-city dwellers: A comparative study. Landsc. Urban Plan.; 2017; 165, pp. 193-205. [DOI: https://dx.doi.org/10.1016/j.landurbplan.2016.06.011]
61. Xu, C.; Du, X.; Jian, H.; Dong, Y.; Qin, W.; Mu, H.; Yan, Z.; Zhu, J.; Fan, X. Analyzing large-scale Data Cubes with user-defined algorithms: A cloud-native approach. Int. J. Appl. Earth Obs. Geoinf.; 2022; 109, 102784. [DOI: https://dx.doi.org/10.1016/j.jag.2022.102784]
62. Knobel, P.; Maneja, R.; Bartoll, X.; Alonso, L.; Bauwelinck, M.; Valentin, A.; Zijlema, W.; Borrell, C.; Nieuwenhuijsen, M.; Dadvand, P. Quality of urban green spaces influences residents’ use of these spaces, physical activity, and overweight/obesity. Environ. Pollut.; 2021; 271, 116393. [DOI: https://dx.doi.org/10.1016/j.envpol.2020.116393]
63. Ekkel, E.D.; de Vries, S. Nearby green space and human health: Evaluating accessibility metrics. Landsc. Urban Plan.; 2017; 157, pp. 214-220. [DOI: https://dx.doi.org/10.1016/j.landurbplan.2016.06.008]
64. Korcz, N.; Kamińska, A.; Ciesielski, M. Is the Level of Quality of Life Related to the Frequency of Visits to Natural Areas?. Forests; 2024; 15, 2257. [DOI: https://dx.doi.org/10.3390/f15122257]
65. Ciesielski, M.; Mohytych, V.; Korcz, N. The role and significance of Cultural Ecosystem Services (CES) based on a survey of forest users in the Warsaw metropolitan area. Sylwan; 2024; 168, pp. 533-548.
66. Wang, J.; Xu, C. Geodetector: Principle and prospective. J. Geogr.; 2017; 72, pp. 116-134.
© 2025 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.
Details
1 China Land Surveying and Planning Institute, Beijing 100035, China;
2 School of Land Science and Technology, China University of Geosciences, Beijing 100083, China;