Conflict or Coordination? Ecosystem Services Supply and Demand in Chinese Urban Agglomerations

Abstract

Urban agglomerations (UAs), which play a significant role in socioeconomic development and urbanization, are confronted with mounting ecological stress and a profound global imbalance in ecosystem services (ES). Understanding the conflict and coordination of knowledge about ES supply and demand (ESSD) can facilitate more efficacious guidance on the ecological sustainable development of UAs. Nevertheless, the characteristics of the conflict and coordination relationship between ESSD in Chinese UAs remain unclear, and further investigation into the interactive coercive relationship between ESSD is warranted. Consequently, we employed spatial regression and coupled coordination models to elucidate the conflict and coordination relationship between ESSD, utilizing multi-source data on Chinese UAs from 2000 to 2020. We found that ES supply in the UAs decreased, while ES demand increased. Furthermore, the coupling coordination degree between ESSD demonstrated an increase trend. The overall coupling coordination degrees between ESSD in UAs were 0.260, 0.285, and 0.311 in 2000, 2010, and 2020, respectively. The central UAs were identified as stress areas, whereas the peripheral areas were classified as non-stress areas. This study offered valuable insights into the interactive relationship between ESSD in UAs and provided a basis for formulating differentiated policies for the sustainable development of ecosystems and human activities.

1. Introduction

In recent decades, there has been a notable intensification of anthropogenic activities and an accompanying increase in demand for ecosystem services (ES), which has resulted in a critical imbalance and degradation of ES, particularly in urban agglomerations (UAs) [1,2,3,4,5]. Currently, UAs in China play a pivotal role in new-type urbanization progress, as well as in ecological conservation and the establishment of ecological barriers [6]. However, the restricted availability of resources and the implementation of more stringent environmental regulations in UAs have increasingly become the primary constraints on sustainable development. The current focus of government departments and researchers is on identifying ways to overcome the ecological limitations that impede sustainable development in UAs. This includes the provision of comprehensive technical solutions to transform UAs from concentrated areas of eco-environmental problems into high-quality development areas [2,3,7,8,9].

ES refers to the material products and spiritual enjoyment that ecosystems provide for human well-being [10,11,12]. ES supply (ESS) stands for the various services provided by the ecosystem for humans, while ES demand (ESD) stands for the services that humans derive from ecosystems for survival and development [13,14,15]. The balance of the ES supply and demand (ESSD) is indicative of the degree of stress experienced by the ecosystem [1]. An ecosystem that is out of balance will result in degradation and, consequently, an adverse impact on human well-being. Conversely, a balanced ecosystem can facilitate regional sustainability [16,17,18,19]. The initial ESSD concentrated primarily on the formulation and establishment of a theoretical framework. As research has progressed, the quantitative methods used to measure ESSD have become increasingly sophisticated, encompassing a wider range of scales and indicators [13,20,21,22,23,24,25,26,27]. Recently, an increased number of studies on ESSD have focused on exploring this relationship by building models to identify a sustainable development path for humans [5,15,16,28,29]. A large number of scholars have employed a variety of spatial analysis methods to model, monitor, and evaluate the spatiotemporal patterns and matching conditions of ESSD at national, regional, and urban scales using a series of spatial analysis methods [30,31,32,33,34,35]. In addition, a series of socioeconomic indicators have been used in prior research to represent the ESD [4,33,35].

The ESS represents the source, whereas the ESD represents the outcome, which signifies the process of ES flowing from nature to human society [13,36]. The spatial mismatch of ESSD gives rise to the spatial flow and conveying of ES, which is influenced by a variety of natural and anthropogenic factors [37,38,39,40]. An excessive amount of ESD in a specific region will inevitably result in a regional imbalance of ES, which, in turn, will lead to ecosystem pressure and degradation [17,19,32,41]. Previous studies of ESSD have focused primarily on the imbalance, trade-offs, decoupling, and correlation between these factors [4,15,33,34,42]. Furthermore, it investigates the spatial relationship and driving factors of ESSD through the use of Geo-detectors, Bayesian networks, coupled coordination models, the geographically weighted regression (GWR) model, and redundancy analysis. The objective is to provide references for the optimization of the spatial pattern of ecosystem compensation and ESSD matching [43,44,45,46,47,48,49,50,51]. There is a paucity of studies that have addressed the conflict and coordination relationship between ESSD, with an even smaller number that has identified areas of ecological pressure in terms of ESSD in UAs. It is imperative to study the conflict and coordination of ESSD to assess and improve ecological management for ESSD balance [5,29,52]. The capacity of the ESS will affect the ESD, even if an ES flow exists. Furthermore, the ESD will also affect the ESS. It is therefore essential to determine the spatial coercive relationship between ESSD to promote regional sustainable development [33,34,39,53]. In light of the growing discrepancy between the limited natural environmental resources and the prospective requirements of socioeconomic development, particularly in Chinese UAs where the relationship between humans and land is strained, it is crucial to undertake quantitative research on the conflict and coordination between ESSD and to delineate the regions experiencing ecological pressure for the advancement of green and high-quality UAs development.

Prior research has provided a comprehensive examination of ES in UAs, indicating that excessive demand for ES represents a primary driver of ES degradation [1,5,17,54]. However, there is still a paucity of understanding regarding the conflict and coordination between ESSD in Chinese UAs. To gain a comprehensive understanding of the interactive coercive relationship between ESSD in Chinese UAs, it is essential to develop and implement regulatory measures that promote sustainable ecosystem development in Chinese UAs. Previous research on ESSD has primarily focused on the analysis of spatiotemporal features, mismatches, trade-offs, the service flow of ESSD, influencing factors, green infrastructure, biodiversity, prediction and simulation, ecosystem management, ecological compensation, and zoning aspects [43,44,45,46,47,48,49,50,51,55]. Previous research on ESSD has been conducted at multiple scales using a variety of methods from multiple perspectives [56,57,58]. While previous research has adequately addressed the relationship between ESSD, the conflict and coordination between ESSD remains unclear, particularly in UAs where the human–land relationship is intensifying [4,33,41]. Furthermore, the identification of ecological pressure areas in UAs based on ESSD is conducive to an understanding of the ecologically fragile, sensitive, and degraded areas. Human–land conflicts in UAs are particularly evident; however, previous studies have focused primarily on ESSD in single UAs, while few studies have considered UAs in China as research areas to understand the conflict and coordination relationship between ESSD and the identification of ecological stress areas [1,17,55]. The contributions of this study included the following: (1) the selection of Chinese UAs as the study areas, which compensated for the deficiencies of previous studies on Chinese UAs; (2) the introduction of coupled coordination models and spatial regression models, which provided a comprehensive explanation of the interactive stress effects of ESS and ESD in Chinese UAs; and (3) the identification of stress areas of ESD in UAs, which offered spatial insights for zonal ecological planning in UAs.

In 2020, China’s overall urbanization rate reached 63%, while the urbanization rate of UAs exceeded 80%. The 19 UAs account for the majority of the country’s population, economy, and urban land. UAs are regarded as a superior organizational regional spatial form in the context of industrialization and urbanization [6,59]. A strategic pattern of urbanization has emerged in Chinese UAs [6,59]. However, the rapid development of UAs has resulted in significant challenges related to resource and environmental concerns. This has been accompanied by a notable expansion of impervious surfaces and a marked decline in ecological land. The current focus of discussion among government departments and researchers is the balancing of the relationship between environmental protection and socioeconomic development, the balancing of ESSD, the alleviation of ecological pressure and conflicts in UAs, and the transformation of UAs into model areas for high-quality socioeconomic development and high-level ecological protection.

To achieve this objective, the spatiotemporal patterns of ESSD in Chinese UAs were initially quantified in the present study. Subsequently, the coldspots and hotspots of ESSD in Chinese UAs were analyzed using hotspot analysis. Thereafter, the coupling coordination and spatial regression models were employed to elucidate the conflict and coordination relationship between ESSD. Finally, a differential local bivariate spatial autocorrelation model was employed to identify the ecological stress areas in the Chinese UAs. The research objectives were as follows: (1) to evaluate the spatiotemporal patterns of ESSD in Chinese UAs; (2) to analyze the conflict and coordination relationship between ESSD in Chinese UAs; and (3) to identify the ecological stress areas in Chinese UAs.

2. Materials and Methods

2.1. Study Area

UAs have become the primary areas of urbanization and socioeconomic development in China, exhibiting an advanced spatial form of urban development. In the context of China’s 14th Five-Year Plan, 19 UAs were proposed as a means of facilitating China’s major economic development forward and establishing these areas as growth poles within the country (Figure 1). In the context of rapid urbanization, continuous socioeconomic development, and ecological degradation, urban agglomerations (UAs) have become a subject of considerable interest. The high concentration of population, capital, and construction land has resulted in a significant expansion of the ESD of UAs. In some areas, there is a significant discrepancy between ESSD, indicating that demand greatly exceeds supply. Ecosystems provide a range of benefits to human society, including the provision of essential resources and the maintenance of well-being. It is therefore important to identify the conflict and coordination relationship between ESSD in the UAs to further promote high-quality development in UAs.

2.2. Data Sources

The data on land use and the normalized difference vegetation index (NDVI) for the years 2000, 2010, and 2020 were obtained from the Data Center for Resources and Environmental Sciences, Chinese Academy of Sciences (RESDC) (http://www.resdc.cn, accessed on 3 August 2024) [60,61]. It has been widely adopted by existing research institutes as a reliable source of data. The economic density (ED) data were derived from the RESDC. Furthermore, population density (PD) data with a 100 m resolution were obtained from Open Spatial Demographic Data and Research (https://www.worldpop.org, accessed on 3 August 2024).

2.3. Method

Figure 2 provides a visual representation of the research flow chart presented in this study. First, the ESS and ESD were calculated using multi-source data, then the spatiotemporal distribution patterns of ES balance was analyzed. Moreover, the hotspot analysis model was conducted to ascertain the principal concentration areas of alterations in ESSD. Second, the interactive coercive relationship between ESS and ESD was revealed through the application of the coupled coordination and spatial regression model. Ultimately, the bivariate spatial autocorrelation model was employed to identify the stress areas within the UAs.

2.3.1. ESS, ESD, and ES Balance

Ecosystem services supply. The theoretical framework for ES measurement proposed by Costanza et al. (1997) [10] has had a considerable impact on global ES assessment [11]. Xie et al. (2008) [62] undertook a revision of the ES measurement based on Chinese expert knowledge drawn from Costanza et al. (1997) [10]. Chen et al. (2022) further revised the ES value equivalent based on the biomass of cultivated land, thereby incorporating consideration of ES [54]. By the enhanced ES measurement framework put forth by Chen et al. (2022) [54], the present study quantified the different types of total ESS in Chinese UAs in 2000, 2010, and 2020. The specific equations were as follows:

$$VCI=({\displaystyle \frac{NDVI-NDV{I}_{\min }}{NDV{I}_{\max }-NDV{I}_{\min }}})\times 100\%,$$

(1)

$$ES{S}_{c-corrected}={\displaystyle \frac{VC{I}_i}{\overline{VC{I}_f}}}\times {\displaystyle \sum _{j=1}^m{\displaystyle \sum _{i=1}^n\left(LU{A}_i\times V{C}_{ij}\right)}},$$

(2)

where ESS_c-corrected is the evaluation result of the ES value after correction; VC_ij represents the equivalent factor of the i-th land use type in the j-th unit; LUA_i signifies the area of the i-th land use type; n and m are the number of ecosystems and ES categories, respectively; VCI denotes the vegetation condition index, which is calculated using NDVI; and VCI_i and $\overline{VC{I}_f}$ are, respectively, the average annual VCI of the cultivated land in the i-th unit and China. In line with previous studies, this study employed the land-averaged ES value of each county as a representative value of ESS [4,32,33].

Ecosystem services demand. In previous studies, ESD was defined as the goods or services that humans consume [13]. ESD can be considered the extent of human demand for a particular ES [33]. This may be exemplified by land development intensity and the degree of disturbance, to a certain extent [4,33]. Socioeconomic indicators are employed to quantify the extent of human demand for ES. According to prior research, these indicators employ the proportion of developed land (PDL), PD, and ED to indicate the ESD [4,32,33]. These indicators were employed in the formulation of a comprehensive multi-index methodology. The equation used was as follows:

$$ES{D}_i=PD{L}_i\times \ln P{D}_i\times \ln E{D}_i,$$

(3)

where ESD_i is the ESD index of the i-th unit. Similarly, PLD_i, PD_i, and ED_i correspond to the PDL, PD, and ED of the i-th unit, respectively.

ES Balance. To further measure the relative development of ESSD, a method was devised to calculate the relative balance development coefficient (E) of ESSD was constructed. The equation used was as follows:

$$E_i=SES{S}_i-SES{D}_i,$$

(4)

where SESS_i and SESD_i represent the standardized ESS and ESD of the i-th unit, respectively. If E_i is around 0, it indicates that the ESS and ESD are in balance.

2.3.2. Hotspots of ESSD

Hotspot analysis is an econometric model employed to investigate the attributes of spatial agglomerations. At present, hotspot analysis is a widely employed method for the monitoring of ecological and socioeconomic changes [63,64,65]. The output of a hotspot analysis comprises two values: the Z-score and the p-value. If a positive Z-score is observed in a region and the p-value is in the significant interval, this indicates that the region in question represents a significant concentration of high values. If a negative Z-score occurs in a region and the p-value is in the significant interval, this signifies that the region has a major concentration of low values. This study introduced the use of hotspot analysis to explore the principal concentration areas of changes in ESSD.

2.3.3. Coupling Coordination Model

Coupling coordination analysis is a method derived from the physical sciences and is employed in a range of disciplines to quantify interactions between multiple systems [4,15]. The current state of the ES in UAs is shaped by the combined influence of the supply and demand systems of the ES. Accordingly, the interactive coercive relationship between the two systems can be quantified using a coupled coordination model [66,67]. Specifically, coupling analysis can reflect the intensity of the interaction between the ESS and ESD; however, it is unable to elucidate the level of development in coupling coordination [66,67]. Accordingly, this method was employed to examine the coupling coordination between the ESS and ESD. The equations used were as follows:

$$C={\left\{SES{S}_i\cdot SES{D}_i/\left({\left(SES{S}_i+SES{D}_i)/2\right)}^2\right)\right\}}^{1/2},$$

(5)

$$D_i={\sqrt{C_i\times T}}_i,$$

(6)

$$T_i=\alpha \times SES{S}_i+\beta \times SES{D}_i,$$

(7)

where C_i represents the coupling degree of the ESSD of the i-th unit, D_i denotes the coupling coordination degree of the ESSD of the i-th unit, and T_i signifies the overall coordination index of the ESSD. It was assumed that ESSD systems were of equal importance (α = β = 0.5).

2.3.4. Spatial Regression Models

An integrated regression approach was subsequently employed to investigate the spatial correlation between ESSD during the 2000–2010 and 2010–2020 periods. The ordinary least squares model was initially employed, followed by four spatial regression models: the spatial lag model (SLM), the spatial error model (SEM), the SEM with lag dependence (SEMLD), and the GWR model. First, the ordinary least squares model was employed to ascertain the association between ESSD, without accounting for spatial dependence and heterogeneity. Subsequently, the SLM, SEM, and SEMLD models were employed in GeoDa 095i software to ascertain the association between ESSD, with the consideration of the spatial lag or spatial error terms taken into account [2,63,68]. The GWR model was used, which was constructed with GWR software (version 4.0) to reveal the spatially non-stationary response of ESS change to ESD change [69]. The changes in ESSD during the 2000–2010 and 2010–2020 periods, as summarized for 1768 counties within Chinese UAs, were the dependent and independent variables incorporated in these models.

2.3.5. Identifying Ecological Stress Areas

The present study employed bivariate autocorrelation to identify stress regions in ESD. The discrepancy between the ESSD was primarily examined through the clustered regions that were identified by bivariate autocorrelation. First, bivariate spatial autocorrelation was performed for the ESSD, and then four types of clustered regions were identified. The regions exhibiting clustering types of low-high and high-low were identified and extracted and were subsequently classified as stress or non-stress regions, respectively [70]. In particular, the low-high clustering results indicate that the clustered region has a decrease in ESS and an increase in ESD above the mean, indicating that the region was a stress region. Conversely, the high-low clustering demonstrated that the rise in ESS and decline in ESD within this region were above average, indicating that this region was a non-stress region.

3. Results

3.1. ESS, ESD, and ES Balance in Chinese UAs

Significant discrepancies were observed in the ES among the various categories of Chinese UAs (Figure 3). In terms of the internal structure of ES in Chinese UAs, the hydrological regulation function constituted the predominant ecosystem function type. During the period between 2000 and 2020, all types of ES experienced a decline, except the hydrological regulation function, which demonstrated an increase. Significant spatial heterogeneity was observed in ESS, ESD, and ES balance across the Chinese UAs (Figure 4). A clear spatial correlation was observed between the low-value areas of ESS and the high-value areas of ESD. In particular, the Ha-Chang UAs, Dianzhong UAs, mid-Yangtze River UAs, and Hu-Bao-E-Yu UAs exhibited elevated ESS. The ESD was observed to be higher in UAs situated in the lower reaches of the Yangtze River and the northern China Plain. The calculation of the ES balance index revealed that the ESS exceeded the ESD in the Ha-Chang, Dianzhong, mid-Yangtze River, northern Tianshan Mountains, and Hu-Bao-E-Yu UAs. Conversely, the ESD exceeded the supply in the Cheng-Yu UAs and eastern coastal region. In the time series, the ESD in the North China Plain and Yangtze River Delta exhibited a rapid increase over the 20 years, whereas the ESS demonstrated minimal discernible change.

3.2. Hotspots of ESS, ESD, and ES Balance in Chinese UAs

The Moran’s I of ESS and ESD change were 0.115 and 0.195, respectively, during 2000–2010, and 0.347 and 0.132, respectively, during the period 2010–2020. Both were found to be statistically significant at the 0.0001, indicating that their change values exhibited significant spatial clustering characteristics. Subsequently, a hotspot analysis was employed to investigate the spatial alterations of ESS and ESD, with the findings presented in Figure 5. In particular, the hotspots of ESS during 2000–2010 were concentrated in the central part of the mid-Yangtze River UAs and the eastern part of the Cheng-Yu UAs. In contrast, the hotspots of ESD were primarily located in major metropolitan regions (such as Beijing, Shenzhen, Shanghai, Guangzhou, and Chengdu). The areas identified as coldspots for ESS exhibited notable changes during the period under review, with the greatest concentration of such areas observed in Beijing, southern Guangdong, Shanghai, northern Tianshan Mountain UAs, and the northwestern part of the Hu-Bao-E-Yu UAs, while the coldspot areas for ESD did not display a similar degree of concentration. The period between 2010 and 2020 saw the majority of changes to the ESS hotspot areas concentrated in the Hu-Bao-E-Yu UAs and its surrounding regions, as well as in the northern part of the Dianzhong UAs and the western section of the Ha-Chang UAs. The distribution of hotspot areas of ESD changes exhibited similarities to that observed during the 2000–2010 period. However, in addition to the regions previously identified, the mid-Yangtze River UAs and the northern Tianshan Mountain UAs also demonstrated the presence of hotspot areas. The coldspot areas of ESS changes during 2010–2020 were concentrated mainly in the mid-Yangtze River UAs, Yangtze River Delta, and eastern part of the Ha-Chang UAs, while the coldspot areas of ESD changes were still not significant. Conversely, there was a discernible spatial correlation between the hotspot areas of the ESS changes and the coldspot areas of the ESD changes. In contrast to ESS, ESD exhibited a notable growth, indicating an increase in ESD within the UAs.

3.3. Spatial Coupling Coordination Degree between ESSD in Chinese UAs

The overall coupling coordination degree indices for the ESSD in the UAs were 0.260, 0.285, and 0.311 in 2000, 2010, and 2020, respectively. Figure 6 illustrates the coupling coordination degree between the ESSD. The areas of discrepancy could be identified, and recommendations for future policy could be provided. Overall, there was no discernible temporal variation in the coupling coordination degree between ESSD during the study period. A total of 96.156% of the county units exhibited an increase in coupling coordination degree during the period between 2000 and 2010, while this value was 86.263% during the subsequent period between 2010 and 2020. This suggests that the coupling coordination degree between ESSD in UAs has increased, which is a positive phenomenon. Furthermore, the spatial distribution revealed high coupling coordination in the North China Plain, Yangtze River Delta, and Pearl River Delta UAs. The regions exhibiting low coupling coordination degrees were primarily situated in the west-central and northeastern UAs.

3.4. Spatiotemporal Impact of ESD on ESS

The spatial dependence diagnostics based on the ordinary least squares model estimation are shown in

Table 1. Diagnostic items of the ordinary least squares method.

Diagnostic Item	2000–2010	2010–2020
Moran’s I (error)	4.454 ***	38.341 ***
LM (lag)	14.838 ***	1367.799 ***
Robust LM (lag)	1.705	73.374 ***
LM (error)	16.982 ***	1305.810 ***
Robust LM (error)	3.849 **	11.386 ***
LM (SARMA)	18.687 ***	1379.185 ***
Breusch–Pagan test	6005.326 ***	66.496 ***
Koenker–Bassett test	312.391 ***	20.402 ***
Log-likelihood	5179.990	2586.080
Akaike information criterion	−10,356.000	−5168.150
Schwartz’s Bayesian information criterion	−10,345.000	−5157.200

Notes: *** p ≤ 0.001, ** p ≤ 0.01, * p ≤ 0.05.

. The findings demonstrated that the spatial regression model exhibited superior performance compared to the ordinary least squares model. It can thus be concluded that the spatial regression models (SLM, SEM, and SEMLD) were more appropriate for elucidating the spatial correlation between ESS and ESD alterations. A comparison of the log-likelihood, Akaike information criterion, and Schwartz’s Bayesian information criterion of the four models revealed that the spatial regression models significantly enhanced the interpretability of the regression relationship to a large extent. The SEMLD model exhibited the greatest log-likelihood value and the smallest Akaike information criterion and Schwartz’s Bayesian information criterion values, thereby demonstrating superior performance (

Table 2. Regression results of the SLM, SEM, and SEMLD.

Explanatory Variables	2000–2010			2010–2020
	SLM	SEM	SEMLD	SLM	SEM	SEMLD
ESD changes	−0.335 *** (0.012)	−0.347 *** (0.012)	−0.339 *** (0.012)	−0.192 *** (0.020)	−0.190 *** (0.021)	−0.136 *** (0.018)
Constant	0.003 *** (0.001)	0.002 *** (0.001)	0.003 *** (0.001)	0.001 (0.001)	−0.015 ** (0.005)	0.006 *** (0.001)
Spatial lag term	0.217 *** (0.046)		0.182 *** (0.055)	0.768 *** (0.026)		1.012 *** (0.021)
Spatial error term		0.318 *** (0.059)	−0.139 *** (0.068)		0.790 *** (0.026)	−669 *** (0.084)
Log-likelihood	5191.640	5194.325	5195.019	2870.790	2865.616	2951.006
Akaike information criterion	−10,377.300	−10,384.600	−10,384.000	−5735.590	−5727.230	−5896.010
Schwartz’s Bayesian information criterion	−10,360.900	−10,373.700	−10,367.600	−5719.150	−5716.280	−5879.580
R2	0.361	0.365	0.362	0.362	0.361	0.401

Notes: *** p ≤ 0.001, ** p ≤ 0.01, * p ≤ 0.05.

). All associations between ESS and ESD changes were found to be statistically significant, with a notable negative association observed between the years 2000 and 2020. The SEMLD model was used to facilitate interpretation, which revealed that for every 1% increase in ESD change, there was a 0.339% decrease in ESS change from 2000 to 2010 and a 0.136% increase in ESS change from 2000 to 2020. The estimation of the spatial lag term revealed that a 1% increase in ESS in the surrounding area during the period 2000–2010 was associated with a 0.182% increase in ESS in the local region, while this value was 1.012% during the subsequent period (2010–2020). This indicated that the magnitude of the ESS was not solely contingent on the local region, but also subject to the influence of external factors from surrounding regions.

The global association between ESSD changes was demonstrated through the utilization of spatial regression models. Subsequently, the GWR model was employed to calculate the locally varying coefficients, to elucidate the spatial heterogeneity, and to reveal the spatial nonstationary features. To provide a more detailed explanation of the GWR results, the estimated coefficient maps for the two periods are presented in Figure 7. The relationship between ESSD changes exhibited spatial variability across different periods. In terms of the time series, the regression coefficients for the two time periods were found to be significantly different. A total of 97.739% of counties exhibited negative regression coefficients during the 2000–2010 period, compared to 91.068% during the 2010–2020 period. This suggests that the negative impact of ESD changes on ESS changes was decreasing, while the positive effect was increasing. This also indicated that the coupling between the ESSD was further enhanced. The spatial pattern of the regression coefficients demonstrated that the changes in each UA was statistically significant. During the period 2000–2010, the ESD had a positive effect on the ESS in the peripheral areas of some UAs in the northwest and northeast, whereas in the majority of other areas, the ESD had a negative effect on the ESS. During the period 2010–2020, the impact of ESD on ESS was positive in the core of some developed UAs and negative in the majority of other areas.

3.5. Ecological Stress Areas

The stress areas of the ESSD in the UAs were derived using local bivariate spatial autocorrelation (Figure 8). During the period between 2000 and 2010, 12.154% of the counties within the UAs were identified as areas of stress, while 24.929% were identified as non-stress areas. The areas identified as experiencing stress were predominantly situated in the eastern coastal UAs and in major cities such as Beijing and Shanghai. In contrast, the non-stress areas were primarily located in the ecological conservation zones on the periphery of the core cities. The findings for the period between 2010 and 2020 indicated an increase in the number of stress areas within the UAs, with 15.998% of counties identified as such and 14.867% classified as non-stress areas. The identified stress areas were predominantly situated within the Yangtze River Delta, the Guangdong-Hong Kong-Macao Greater Bay Area, and within the boundaries of major urban centers, including Beijing, Tianjin, Wuhan, and Changsha. The distribution of non-stressed areas was similar to that observed during the 2000–2010 period, although its proportion was significantly reduced.

4. Discussion

4.1. Interpretation of Findings

The degradation of ecosystems in UAs has resulted in a gradual reduction in the capacity of these systems to provide essential services, a phenomenon that has attracted considerable attention within the academic community [2]. The findings revealed that hydrological regulation constituted the primary ecosystem function in UAs. During the study period, all ecosystem functions exhibited a decline to varying degrees, except hydrological regulation. The increasing demand for ES in UAs, which are the main areas of population, economy, and construction land in China as a result of rapid urbanization, still requires the proper functioning of ES provisioning systems. The functioning of ecosystems is a determining factor in human well-being. However, since the 21st century, evolving socioeconomic policies and land management practices have posed a threat to the proper functioning of ecosystems [71,72,73]. Prior research has indicated that land use/land cover change represents a primary driver of ESSD alteration, with population growth representing a secondary factor [29]. The exponential growth of the global population has resulted in an increased demand for ESD. Similarly, the rapid growth of the global economy is often preceded by ecological damage. While the area of forested land plays a significant role in ESS [43,48], an expansion of construction land also disrupts landscape connectivity, which, in turn, leads to a reduction in ESS and, consequently, ecological degradation [74]. It can be concluded that ESSD are not independent systems; rather, they interact and constrain each other [33,51]. This may also explain the observed increase in ESD while ESS decreases in UAs. Furthermore, the central regions of UAs frequently exhibit imbalances in the ES, with the ESD exceeding the ESS to a considerable degree. In the majority of UAs, the core–edge model of socioeconomic development was observed to prevail over the edge–core model of ES. It is inherent to the nature of large cities that an ESD gap will be created [29]. Furthermore, they frequently lack effective ecological conservation land, which contributes to additional imbalances in the ESSD. Similarly, a Matthew effect was identified in the changes observed in ESD. Areas exhibiting a high level of ESD continued to demonstrate an upward trend in this regard. This may have been due to the siphoning effect of large cities, with the increasing ESD being triggered by the concentration of factors [29]. However, the dynamics of ESS changes exhibited disparate patterns. The Loess Plateau region demonstrated a further augmentation in ESS capacity due to the prolonged implementation of an ecological restoration project. Conversely, other regions tended towards a reduction in ESS capacity.

4.2. Conflict and Coordination Relationship between ESS and ESD

In comparison to other regions, UAs are distinctive in terms of their physical geography and socioeconomic conditions. UAs represent the primary areas of concentration for human activity, economic investment, and land development and are frequently the epicenters of the most severe conflicts between ESS and ESD [29]. In comparison with other studies [46,49], it was determined that the coupling coordination relationship between ESS and ESD in UAs remained at a low level [4], indicating that ESS and ESD in UAs are characterized by a conflicting relationship. Nevertheless, the rising coupling coordination relationship between ESS and ESD throughout the study period suggests a sustained enhancement in the interrelationship between ESS and ESD in UAs. The achievement of high-quality urbanization is contingent upon the provision of robust support for healthy ecosystems [46]. The formulation of effective urban development policies can facilitate the restoration of ecosystems through the allocation of technical and financial resources [66]. Some studies have also identified a potential U-shaped correlation between urbanization and ES, suggesting that this relationship could facilitate further coordination of ESD and ESS in UAs [68,74].

Furthermore, the stress areas of ESD that require particular attention were identified through the use of local differential bivariate autocorrelation. The aforementioned stressed areas were predominantly situated in regions exhibiting the highest socioeconomic levels in China. To maintain human well-being, it is necessary to increase the level of ESS in areas of high population density. Nevertheless, the persistent disparity between socioeconomic advancement and the provision of essential services continues to impede the attainment of superior developmental standards in these regions. Consequently, the equilibrium between ES in areas of stress, while upholding the trajectory of augmented ESS collaboration, persists as a significant challenge for decision-makers.

4.3. Policy Implications

The restoration of ecosystems and the achievement of coordinated balance in the UAs represent long-term challenges. It is incumbent upon policymakers to consider ways of improving the supply and demand patterns in UAs, as well as reversing the degradation of ecosystems in these areas. It is also incumbent upon researchers to consider further the identification of patterns of ES imbalance in UAs and of priority areas for restoration. In light of the aforementioned findings, the following potential policies are put forth for consideration.

Building an integrated framework for ecosystem management and economic development. The establishment of ecological protection zones in key ecological areas should not only strictly limit development activities but protect biodiversity and the integrity of ecological functions. It is also necessary to make full use of the spatial spillover effect of the ecological reserve, radiate the ecological functions of the reserve to the surrounding areas through ecological corridors and green infrastructure, promote the overall improvement of the ecological environment, and realize the benign interaction between ecological protection and economic development. In addition, ES is weighed and prioritized to ensure that the overall benefits are maximized [54].

A comprehensive eco-environmental pressure assessment is conducted for UAs with high socioeconomic levels, rapid urbanization, and significant eco-environmental pressure [68]. Specific ecological restoration plans have been developed to alleviate ecological pressure and restore ES. These plans encompass measures such as vegetation restoration, wetland protection, and river management. It is recommended that urbanization be promoted in a manner that is of the highest quality and that ecosystem restoration be integrated into urban planning and development strategies. To mitigate the adverse effects of urbanization on the ecological environment by facilitating the advancement of novel urbanization models. Concurrently, the technical and financial advantages of urbanization can be leveraged to facilitate the implementation of ecological restoration projects and technological innovation. Addressing the stress areas of ESD is a key consideration for policymakers. Toward that end, orderly ecological restoration work in stress areas may alleviate this imbalance.

Ultimately, the optimal management of the relationship between the restoration of ecosystems and urbanization in UAs may facilitate a mutually beneficial outcome [66]. Historically, urbanization has frequently resulted in detrimental impacts on the surrounding ecological environment. At this juncture, China has reached a pivotal point in its urbanization trajectory, presenting an opportunity to advance the technological sophistication of ecosystem restoration in UAs. This could be achieved by integrating ecosystem restoration into the urbanization process. The implementation of high-quality urbanization policies and programs can facilitate the provision of technical and financial support for a diverse range of projects about ecological restoration, thereby enhancing human well-being through the advancement of ecosystem restoration.

4.4. Limitations and Future Directions

This study reveals the conflict and coordination relationship between ESSD in China based on the coupled coordination and spatial regression model. It provides new evidence and perspectives for understanding ecological pressure in UAs in China. Further research could build on the present study in the following ways. First, while this study has considered the spatial dependence effect of ESSD, it is evident that the flow between the elements of UAs also has a significant impact on the conflict and coordination of ESSD in UAs. It is therefore recommended that future studies adopt a local coupling and a telecoupling theoretical framework [75]. Secondly, this study solely examined the interrelationship between ESSD, without taking into account the impact of policy regulations. In particular, further verification is required concerning the influence of UAs as policy guides on ESSD.

5. Conclusions

The spatiotemporal patterns of ESS, ESD, and ES balance in Chinese UAs were quantified, and coupled coordination and spatial regression models were introduced to analyze the interactive coercive relationships between ESSD. The findings indicated a decline in ESS within UAs, while ESD demonstrated an upward trend. The changes in ESSD in UAs had significant spatial dependence, and their Moran’s I was all larger than 0.3. The core UAs were the principal locations where ESD increased, while the peripheral UAs were the primary regions where ESS increased.

Furthermore, the degree of coupling coordination degree between the ESSD increased. The overall coupling coordination degrees between ESSD in UAs were 0.260, 0.285, and 0.311, respectively, in 2000, 2010, and 2020. During the period between 2010 and 2020, an increase in ESD was observed to contribute to an increase in ESS. These results indicate that the coupling coordination degree between the ESSD in the UAs is improving.

The differential local bivariate spatial autocorrelation model was employed to identify the stress areas of ESD, and the core UAs remained the primary stress areas. This study may serve as a scientific reference for achieving effective and differentiated ecosystem management tools in Chinese UAs and offers an example for future research on ESSD in UAs in other developing countries.