Assessing the Resilience of Urban Social–Ecological–Technological Systems in the Beijing–Tianjin–Hebei Urban Agglomeration

Abstract

The resilience of urban agglomerations (UAs) is recognized because of their ability to withstand, adapt to, and recover from natural disasters and social threats. However, limited information on the resilience of specific urban agglomerations may hinder their sustainable development. The emerging concept of Social–Ecological–Technological system (SETS) resilience presents a novel framework for understanding and evaluating the resilience of UAs. Taking the Beijing–Tianjin–Hebei urban agglomeration (BTHUA) as a case study, we constructed a comprehensive resilience assessment framework. By incorporating the coupling coordination degree (CCD) model, modified gravity model, standard deviation ellipse, and obstacle degree model, we systematically evaluated the BTHUA’s SETS resilience. The results show that from 2010 to 2022, both the SETS resilience and its CCD in the BTHUA improved significantly. All the cities reached the coordination stage, with CCD values exceeding 0.6. The key cities enhanced their influence on the surrounding cities, resulting in a more robust and interconnected intercity resilience network. However, the BTHUA still confronts challenges in resource endowment, technological innovation, and public services, which warrant a more integrated and systematic approach to enhance regional SETS resilience.

1. Introduction

Urban areas will be home to nearly 68% of the global population by 2050, highlighting the rapid expansion of urban landscapes [1]. Alongside this expansion, a confluence of complex natural and social threats has emerged [2,3]. Urban resilience, which emphasizes reducing urban vulnerability and achieving sustainable development, provides a novel paradigm for addressing these multifaceted challenges and has been increasingly applied across various fields of urban governance, including urban planning, economic recovery, disaster risk management, social transformation, and climate change studies [4,5,6,7,8]. Globally, both the 2030 Agenda for Sustainable Development [9] and the New Urban Agenda [10] have emphasized the importance of building resilient cities. In China, urban agglomerations (UAs), as highly developed spatial forms of integrated cities, are regarded as the primary form for advancing urbanization and serve as significant carriers of socioeconomic accumulation and industrialization [11,12]. Consequently, enhancing resilience through systematic measures and integrated capabilities has become a fundamental issue for UAs to maintain vitality and achieve sustainable development [13,14].

Since Holling first introduced the concept of resilience to the field of ecological studies as the capacity of ecosystems to withstand disturbances while maintaining their basic functional characteristics [15], scholars have interpreted its meaning in various ways based on different disciplines and perspectives. In the context of urban systems, resilience generally refers to the capacity to respond or reorganize in ways that preserve essential functions, identity, and structure, while also retaining the ability to adapt, learn, and transform in the face of anticipated or unexpected threats, shocks, and other disruptions [16,17]. This capacity determines the performance and long-term sustainability of urban systems when facing internal and external pressures [18,19]. Cities are dynamic and complex coupled systems of natural–social symbiosis, whose subsystems exhibit characteristics such as nonlinear dynamics, feedback mechanisms, and high interconnectivity [20,21]. Numerous studies have illuminated the complex interactive relationship between humans and ecosystems in cities from the perspective of social–ecological resilience [22,23,24], which has significantly inspired urban managers to transition from linear thinking to resilience-oriented approaches. Scholars have devised diverse methodologies for the quantitative assessment of urban resilience. Among these, indices serve as indispensable tools for quantifying resilience across various scenarios and temporal dimensions [25]. Many studies have developed indicator frameworks rooted in urban subsystems, such as economic, social, ecological, infrastructural, and institutional dimensions. This approach not only enhances urban resilience management but also optimizes resource allocation guidance across different subsystems to boost overall resilience [26,27].

As technologies have advanced to become more intricate, automated, decentralized, accessible, and intelligent, there is a growing recognition that technological systems should be viewed as a distinct but interrelated domain that aligns with the complexity of social and ecological structures and dynamics, rather than merely being a component of the social systems [28,29]. Technological resilience emphasizes both the robustness of physical and digital infrastructure and the integration of innovation and adaptive technologies to maintain functionality and performance under stress [30]. By integrating technological solutions with broader environmental policies and social initiatives, societies can more effectively adapt to and mitigate the impacts of environmental change, ultimately fostering a more resilient and sustainable future [31,32]. Nevertheless, technologically deterministic approaches, which emphasize technological solutions while neglecting broader social and ecological contexts, may exacerbate inequality, create a false sense of security, and ultimately heighten long-term vulnerabilities [33]. Thus, it is essential to embrace a Social–Ecological–Technological system (SETS) perspective to integrate social, ecological, and technological resilience, which can create opportunities for novel approaches to adaptation and transformation in complex environments while better managing risks and trade-offs [34,35,36].

The intricate mutual promotion and inhibition among subsystems is critical for the well-being of urban systems. Positive interactions among these subsystems can create synergistic effects, enhancing collective performance beyond the sum of their individual functionalities [37]. In the face of disturbances, coordinated responses among subsystems help maintain overall stability, whereas the pursuit of independent goals without coordination can lead to unsustainable outcomes and potentially system collapse [38]. The term “coupling” refers to the interaction and mutual influence between two or more subsystems [39], while “coordination” emphasizes the collaborative efforts of subsystems within complex systems to collectively achieve shared goals [40]. Scholars have dedicated substantial efforts to quantify the coupling coordination degree (CCD) of the resilience among urban subsystems [41,42,43]. However, current research efforts still lack a comprehensive evaluation framework that integrates technological resilience. In some cases, the complexity of urban system processes and patterns—both within individual cities and across interconnected urban regions—represents a critical challenge for today‘s urban planners, policymakers, and managers, especially in the context of UAs’ construction. This complexity arises because sustainable choices made in one location may not hold true if they impose social, economic, or environmental trade-offs elsewhere [44]. Although research on urban spatial resilience evaluation exists, there remains a deficiency in dynamically analyzing the spatiotemporal evolution of resilience across continuous time series and different scales [45]. This gap also involves ignoring the spatial heterogeneity of resilience and the spatial connections between cities within the UAs.

The Beijing–Tianjin–Hebei urban agglomeration (BTHUA), serving as China’s capital economic circle, innovation engine, and critical ecological region [46], features SETS that are both typical and highly complex. The region currently faces significant tensions among population aggregation, economic development, urban expansion, and ecological construction. Previous research on the urban resilience of the BTHUA has primarily focused on individual resilience subsystems or the coupling between socioeconomic and ecological subsystems. Here, we conducted a quantitative assessment of the resilience of the urban system in the BTHUA using a developed SETS framework, investigating the interactive coupling effects of SETS resilience and its spatiotemporal dynamics. Specifically, we aimed to (1) estimate the SETS resilience, (2) examine the spatiotemporal characteristics of the coupling coordinated development of SETS resilience and the interactions among cities, and (3) identify the key obstacle factors. Consequently, this study is expected to contribute to the refinement and enhancement of the theoretical framework for UAs’ resilience. It provides empirical evidence for understanding and measuring SETS resilience at the regional scale, offering novel perspectives on the sustainable development of resilient cities. By emphasizing the coupling coordination of subsystems and intercity interactions as crucial pathways to regional sustainability, the research not only deepens theoretical understanding but also serves as a practical reference for formulating policies to enhance UAs’ resilience.

2. Materials and Methods

2.1. Study Area

The BTHUA is located in Northern China, administratively encompassing Beijing, Tianjin, and 11 prefecture-level cities in Hebei Province, with a total area of 218,000 km² (Figure 1). The BTHUA accounted for 7.8% of China’s population and contributed 8.9% to the nation’s annual GDP between 2010 and 2022. However, disparities in socioeconomic development levels, variations in resource endowments, and the polarization effects caused by the concentration of population and resources in urban areas have led to regional development imbalances in the BTHUA [42]. The cooperative enhancement of urban resilience and the promotion of coordinated development in the BTHUA have become more crucial than ever for addressing challenges and achieving sustainable development.

2.2. Analytical Framework and Indicator System

The study constructed an analytical framework of SETS resilience, positioning TecR as equally significant as EcoR and SocR. The dimensions of these three subsystems were systematically conceptualized, and the underlying mechanisms were disentangled through an analysis of their spatiotemporal coupled interactions (Figure 2).

SocR, EcoR, and TecR are interconnected components of urban systems. Ecosystems serve as the spatial carriers and material foundations for urban development. EcoR maintains the structure and function of ecosystems, thereby supporting the continuous provision of ecosystem services [47], which are fundamental for SocR and urban well-being. Social development and human activities profoundly impact EcoR by altering ecological structures, encroaching on natural spaces, and exploiting resources [48]. High SocR usually means more efficient resource allocation mechanisms, strong social organization capacity, and adaptive policies, so as to provide a guarantee for technological innovation and improve TecR. In turn, an improvement in TecR could help strengthen infrastructure, stabilize economic development, and improve public services, particularly in the digital age [49]. Smart technologies such as web-based platforms and smartphone apps can support evidence-based decision-making and offer learning and knowledge-sharing opportunities to enhance adaptive capacity [50]. In addition, TecR contributes to EcoR by enabling better monitoring, management, and restoration of ecosystems [51]. Adequate infrastructure and high technological performance in pollution control and energy saving help reduce human pressure on the well-being of ecosystems [52]. Correspondingly, EcoR provides a natural foundation for technological innovation by providing a variety of raw materials, serving as sources of inspiration, and creating a suitable environment.

In this study, SocR was categorized into three dimensions: economic strength, public service, and social stability. Economic strength serves as the foundation for SocR, providing the necessary resources and financial support to withstand and recover from various external shocks. This ensures that communities can access essential goods and services even during crises. The per capita GDP and per capita disposable income of urban residents reflect the level of macroeconomic development, indicating that urban residents have an economic foundation to bear risk-related losses. A higher share of the tertiary industry signifies a more rational regional economic structure [53]. Effective public services, such as healthcare, education, basic insurance, and emergency response, help maintain social cohesion, promote social equity and inclusion, ensure that basic needs are met during and after disruptions, and enhance the capacity of communities to respond to crises [8]. These services are also essential for long-term resilience. Social stability involves strong social networks, effective governance, and a shared sense of identity, which help reduce vulnerability and foster collective action [54]. Human development serves as the foundation for social stability and drives overall social progress. A rational population growth rate and adequate financial investments in social security and employment can maintain basic urban functions, ensure a sufficient workforce, and support social development. The urban–rural gap is a critical factor affecting urban stability, as a narrower urban–rural income gap reduces social vulnerability and fosters more balanced urban–rural development [25].

EcoR was divided into three dimensions: resource endowment, landscape pattern, and ecological quality. The availability and distribution of natural resources form the material basis for ecosystem functioning and human well-being. Adequate resource endowments, particularly water, land, and forests, ensure that ecosystems can sustain their functions and support biodiversity. The landscape pattern encompasses the spatial arrangement and configuration of different land-cover types within an ecosystem, and the connectivity, fragmentation, and heterogeneity of landscapes significantly impact resilience. A high edge density index and impervious area ratio typically indicate a more fragmented ecosystem [55]. A diverse and well-connected landscape can enhance EcoR by facilitating species movement, ecosystem service flows, and the buffering of disturbances from land-use conflicts [56]. Enhancing ecological quality through conservation and restoration efforts is crucial for long-term resilience. High concentrations of pollutants, such as PM2.5, threaten ecological quality. Healthy vegetation growth forms the foundation for maintaining the structure and quality of ecosystems. High ecological quality ensures the structural and functional stability of ecosystems under stress, while promoting faster recovery from damage [57].

TecR was conceptualized into three dimensions: urban infrastructure, technological innovation, and technical performance. Robust and well-designed urban infrastructure, including physical constructs like buildings, roads, and bridges, as well as essential services such as water supply, electricity, and communication networks ensure that a city or organization can withstand various pressures without significant degradation of services [51]. Urban areas and communities with inadequate infrastructure are more likely to be affected by extreme weather events associated with climate change, as well as resource insufficiency brought about by rapid urbanization [35]. The number of internet subscribers and the urban road area reflect the connectivity for material and information flows within and beyond the city. Adequate water supply infrastructure ensures sufficient water availability to meet residents’ needs [41]. Technological innovation can continuously optimize system architecture, provide effective solutions and methods for improving system efficiency and adaptability, and enable more robust responses to uncertain global changes in the future [58]. Cities with diverse, flexible, and adaptable knowledge bases and a high level of TecR were better able to avoid technological crises, recover more quickly from crisis events, and be competitive in the long run [59]. Considering data availability, the number of granted patents, the proportion of science and technology financial expenditures, and R&D investment intensity were selected as the technological innovation indicators. Technical performance ensures that existing technologies and systems operate at optimal levels. It can be enhanced through continuous monitoring, maintenance, and system upgrades, which not only strengthen the overall stability and reliability of the technological system but also enhance urban resilience to disruptions [36]. Lower CO₂ emissions, water consumption, and sulfur dioxide emissions per CNY 10,000 of GDP indicate that a city’s technological system is more effective in reducing endogenous pollution and addressing challenges like climate change and resource scarcity.

Overall, based on the definition of urban resilience and the established analytical framework, we selected 9 dimensions and 27 indicators from the social, ecological, and technological aspects of the urban system to form the evaluation index system (

Table 1. The evaluation system of resilience for the BTHUA.

Subsystem	Dimension	Indicator	Unit	Property *	Weight †
Social Resilience	Economic strength	Per capita GDP	yuan	Positive	0.12
		Per capita disposal income of urban resident	yuan	Positive	0.10
		Proportion of the secondary and tertiary industry added value in GDP	%	Positive	0.08
	Public service	Number of doctors per 10,000 people	person	Positive	0.10
		Number of students in colleges and universities	person	Positive	0.21
		Number of employees participating in basic endowment insurance per 10,000 people	person	Positive	0.11
	Social stability	Index of urban–rural income gap	%	Negative	0.08
		Annual growth rate of permanent resident population	%	Positive	0.05
		Proportion of financial expenditures on social security and employment	%	Positive	0.14
Ecological Resilience	Resource endowment	Water resource per capita	m3	Positive	0.13
	Landscape pattern	Land area per 10,000 people	km2	Positive	0.23
		Green space per capita	m2	Positive	0.10
		Edge density index	-	Negative	0.13
		Aggregation Index	-	Positive	0.10
		Ratio of impervious area	%	Negative	0.09
	Ecological quality	PM 2.5	μg/m3	Negative	0.08
		Gross primary productivity	gC/m2	Positive	0.10
		Fraction Vegetation Coverage	%	Positive	0.05
Technological Resilience	Infrastructure	Number of subscribers of internet per 10,000 household	number	Positive	0.11
		Urban road area per capita	m2	Positive	0.07
		The density of water supply pipes in built-up areas	km/km2	Positive	0.11
	Technological innovation	Number of patents granted per 10,000 people	piece	Positive	0.19
		Proportion of financial expenditures on science and technology	%	Positive	0.18
		R&D investment intensity	%	Positive	0.15
	Technical performance	CO2 emissions per CNY 10,000 of GDP	ton	Negative	0.07
		Water consumption per CNY 10,000 of GDP	ton	Negative	0.06
		Volume of sulfur dioxide emission per 10,000 GDP	ton	Negative	0.05

* “Positive” represents indicators that make positive contributions to resilience and “Negative” represents indicators that have negative impacts on resilience; ^† the weights of the indicators were determined by the Entropy–CRITIC weight method.

). The selection of indicators was based on related theories and existing research results, enabling a comprehensive, scientific, systematic, and objective reflection of the resilience of each subsystem. All data had high authority, availability, and mature measurement methods, ensuring the feasibility of the evaluation work.

2.3. Data Source

The raw data of the indicators of the BTHUA between 2010 and 2022 used in this study were primarily sourced from the China City Statistical Yearbook, China Statistical Yearbook, Beijing Statistical Yearbook, Tianjin Statistical Yearbook, Hebei Economic Yearbook, and Statistical Bulletin of National Economic and Social Development. The edge density index and Aggregation Index were calculated based on the land-use data (30 m) from China’s CLUD [54] using the R version 4.4.1 and R package Landscapemetrics [55]. The annual average value of PM2.5 of each city was calculated based on Shen et al. [56]. The GPP data were derived from the Resource and Environmental Science Data Platform. Among these indicators, the value-based indicators such as the per capita GDP and per capita disposable income of urban residents were a constant price. The indicators expressed as per capita, per 10,000 people, or percentage ratios were calculated based on the permanent population. To address missing data in specific years, we applied linear interpolation techniques.

2.4. Methods

2.4.1. Entropy–CRITIC Weight Method

The entropy weight method and CRITIC method are both commonly used to assign weights to indicators. By combing these two methods, we can effectively capture the diversity, dispersion, and correlation among different features and enhance the accuracy of the weight assigning process [60].

Standardization:

Positive indicators:

$$x_{ij}^{\prime }={\displaystyle \frac{x_{ij}-x_{min}}{x_{max}-x_{min}}}$$

(1)

Negative indicators:

$$x_{ij}^{\prime }={\displaystyle \frac{x_{max}-x_{ij}}{x_{max}-x_{min}}}$$

(2)

where $x_{ij}^{\prime }$ is the standardized value of $x_{ij}$, and $x_{max}$ and $x_{min}$ are the maximum and minimum values of the jth indicator, respectively.

Calculate the weight using the Entropy–CRITIC method:

$$p_{ij}={\displaystyle \frac{x_{ij}^{\prime }}{{\sum }_{i=1}^m{x}_{ij}^{\prime }}}$$

(3)

$$E_j=-{\displaystyle \frac{1}{ln m}}{\sum }_{i=1}^m p_{ij}ln p_{ij}$$

(4)

$$W_1={\displaystyle \frac{1-E_j}{{\sum }_{j=1}^n(1-E_j)}}$$

(5)

$$W_2={\displaystyle \frac{{\sigma }_j{\sum }_{i=1}^m \left(1-r_{ij}\right)}{{\sum }_{j=1}^n \left[{\sigma }_j{\sum }_{i=1}^m \left(1-r_{ij}\right)\right]}}$$

(6)

$$W_j=\alpha W_1+(1-\alpha )W_2$$

(7)

where $W_1$ is the weight obtained by the entropy weight method; $p_{ij}$ is the probability of the occurrence of a particular indicator value; $E_j$ is the entropy value of the jth indicator; $W_2$ is the weight obtained by the CRITIC method; ${\sigma }_j$ is the standard deviation; $r_{ij}$ denotes the linear correlation coefficient between i and j; ${\sigma }_j{\sum }_{i=1}^m \left(1-r_{ij}\right)$ is the information given by the jth indicator; $W_j$ is the weight of obtained by the Entropy–CRITIC weight method; and $\alpha$ = 0.5 [42].

2.4.2. Coupling Coordination Degree Model

The CCD model can explain the interdependencies, associations, and relationships between the different systems [61]. This study adopted the CCD model to investigate the spatiotemporal characteristics of the coupling coordination relationships between the resilience level of social, ecological, and technological subsystems.

$$C=3\times {\left[{\displaystyle \frac{SocR\times EcoR\times TecR}{{\left(SocR+EcoR+TecR\right)}^3}}\right]}^{{\displaystyle \frac{1}{3}}}$$

(8)

$$T={\gamma }_{1}SocR+{\gamma }_{2}EcoR+{\gamma }_{3}TecR$$

(9)

$$\mathit{CCD}=\sqrt{C\times T}$$

(10)

where C and CCD are the coupling degree and coordination degree of $SocR$, $EcoR$, and $TecR$, respectively. T is the comprehensive coordination value of $SocR$, $EcoR$, and $TecR$; ${\gamma }_1$, ${\gamma }_2$, and ${\gamma }_3$ are the weight coefficients. In view of the equal importance of these three subsystems, ${\gamma }_1$ = ${\gamma }_2$ = ${\gamma }_3$ = 1/3. To better conduct the comparisons, the resilience level was classified into three stages and eight types of CCD as shown in

Table 2. Classification of the CCD level.

Level	CCD Value	Stage
Extreme imbalance	0.000~0.200	Disorder stage
High imbalance	0.201~0.300
Moderate imbalance	0.301~0.400	Transition stage
Low imbalance	0.401~0.500
Primary coordination	0.501~0.600
Moderate coordination	0.601~0.700	Coordination stage
Good coordination	0.701~0.800
Excellent coordination	0.801~1.000

.

2.4.3. Standard Deviation Ellipse Model

The standard deviation ellipse model is frequently employed in research to analyze spatial distribution patterns and directional trends of geographic data [62,63]. It has been adopted to reflect the spatial evolution pattern of the CCD [64].

$$\mathrm{t}\mathrm{a}\mathrm{n}\theta ={\displaystyle \frac{\left({\sum }_{i=1}^m {\bar{X}}_i^2-{\sum }_{i=1}^m {\bar{Y}}_i^2\right)+\sqrt{\left[{\left({\sum }_{i=1}^m {\bar{X}}_i^2-{\sum }_{i=1}^m {\bar{Y}}_i^2\right)}^2\right]+4{\left({\sum }_{i=1}^m X_i-Y_i\right)}^2}}{2{\sum }_{i=1}^m X_i-Y_i}}$$

(11)

$${\sigma }_x={\displaystyle \frac{1}{n}}{\sum }_{i=1}^m {\left(\overline{X_i}\mathrm{c}\mathrm{o}\mathrm{s}\theta -{\bar{Y}}_i\mathrm{s}\mathrm{i}\mathrm{n}\theta \right)}^2+4{\left({\sum }_{i=1}^m X_i-Y_i\right)}^2$$

(12)

$${\sigma }_y={\displaystyle \frac{1}{n}}{\sum }_{i=1}^m {\left(\overline{X_i}\mathrm{s}\mathrm{i}\mathrm{n}\theta -{\bar{Y}}_i\mathrm{c}\mathrm{o}\mathrm{s}\theta \right)}^2$$

(13)

where ($\bar{X}$, $\bar{Y}$) is the weighted average center; ${\bar{X}}_i$ and ${\bar{Y}}_i$, respectively, represent the coordinate deviation from the location of each research object to the average center; tanθ represents the rotation angle; and ${\mathsf{\sigma }}_x$ and ${\mathsf{\sigma }}_y$ are standard deviations along the long axis and short axis, respectively.

2.4.4. Modified Gravity Model

Cities can be considered as nodes in the regional resilience network, and they establish their connections through their interactions. The gravity model, which is frequently used for quantifying intercity linkages, has been adapted for studies in many fields such as sociology, economics, and ecological studies [65,66]. To explore the dynamic evolution of CCD spatial interactions among various cities, this study used the modified gravity model and regarded the CCD as a city’s quality [64,67].

$$R_{AB}=k_{AB}{\displaystyle \frac{M_A{M}_B}{d_{AB}^2}}$$

(14)

$$k_{AB}={\displaystyle \frac{M_A}{M_A+M_B}}$$

(15)

where $M_A$ and $M_B$ represent the CCD values of city A and city B, respectively; $d$_AB is the geographical distance between city A and city B; $R_{AB}$ is the CCD’s spatial connectivity strength; and k_AB is the empirical constant defined by the ratio of city A’s CCD to the sum of the CCDs of city A and city B. The linkage strength was then classified into five categories (I to V), ranging from weak to strong. This study followed prior research to retain spatial connections whose strength exceeded the multiyear average value of spatial connection strength [64].

2.4.5. Obstacle Degree Model

To better formulate countermeasures and suggestions for promoting resilience for different cities and the BTHUA, this study used the obstacle degree model to identify the obstacle factors constraining the well-being of resilience and quantify the degree of obstacles [68,69]. The formula is as follows:

$$F_j=\gamma \times W_{\mathrm{j}}$$

(16)

$$T_{ij}=1-x_{ij}^{\mathrm{\text{'}}}$$

(17)

$$I_{ij}={\displaystyle \frac{F_j\times T_{ij}}{{\sum }_{i=1}^m F_j\times T_{ij}}}$$

(18)

where $F_j$ is the factor contribution of the jth indicator; $\mathsf{\gamma }$ is the weight coefficient of each subsystem; $T_{ij}$ and $I_{ij}$ denote the the deviation and obstacle degree indicator j for city i, respectively; and the $x_{ij}^{\prime }$ is the standardized indicator values calculated by Equations (1) and (2).

3. Results

3.1. The SETS Resilience of the BTHUA

The SocR values showed an uneven upward trend from 2010 to 2022, accompanied by the enhancement of economic strength, social security, and social stability. Its average values reached 0.281, 0.325, 0.400, and 0.465 in 2010, 2014, 2018, and 2022, respectively, (Figure 3a). SocR experienced its fastest growth from 2014 to 2018 at an annual rate of 5.28%, followed by slower growth rates of 3.87% from 2018 to 2022 and 3.68% from 2010 to 2014. The maximum value ascended from 0.537 in 2010 to 0.789 in 2022, while the minimum value climbed from 0.179 in 2010 to 0.343 in 2022. The disparity between the maximum and minimum values of regional social indicators expanded from 0.358 in 2010 to 0.430 in 2018, before undergoing a slight contraction after 2019. Cities in the region showed pronounced disparities in their SocR values and growth rates, highlighting the ongoing challenge of uneven social development. Beijing and Tianjin sustained high-level SocR values since 2010, retaining their status as the top two cities in terms of SocR among the 13 cities from 2010 to 2022, with average annual growth rates of 3.25% and 4.23%, respectively. Cities that had lower initial SocR values in 2010 generally witnessed higher growth rates. Notably, Chengde, Zhangjiakou, and Cangzhou achieved the highest average annual growth rates, at 6.52%, 5.57%, and 5.41%, respectively, demonstrating their potential in ScoR growth.

The EcoR values for 2010, 2014, 2018, and 2022 were 0.377, 0.376, 0.411, and 0.439, respectively, with an average annual growth rate of 1.27% (Figure 3b). Between 2010 and 2014, the region experienced a modest overall decline. Among the thirteen cities, seven saw a decrease in their EcoR during this period, with the most significant average annual decline of 1.64% occurring in Hengshui. From 2014 to 2018, the EcoR of most cities experienced the most substantial improvement. Specifically, the EcoR of Langfang, Beijing, Cangzhou, and Tianjin achieved average annual growth rates of 5.48%, 5.14%, 4.03%, and 3.04%, respectively. Subsequently, from 2018 to 2022, a moderate growth trend persisted. Regarding spatial distribution, Chengde and Zhangjiakou, situated in the northern mountainous regions, possess relatively large areas of natural ecological land and abundant natural resources, consistently maintaining relatively high levels of EcoR. This underscored their crucial roles in bolstering regional EcoR. Nevertheless, cities in the plain areas, characterized by scarce land and water resources and relatively fragmented landscape patterns, exhibited higher average annual growth rates of EcoR as resource conservation and ecological restoration efforts advanced.

The TecR values in the BTHUA increased progressively from 0.336 in 2010 to 0.363 in 2014, 0.408 in 2018, and 0.444 in 2022 (Figure 3c). The average annual growth rate of TecR demonstrated a similar trend to that of SocR, with the average annual growth rates being 2.95% from 2014 to 2018, 2.13% from 2018 to 2022, and 1.99% from 2010 to 2014. In terms of spatial distribution, as the core of scientific and technological innovation in the BTHUA and even across the country, Beijing always maintained the highest level of TecR, with the value increasing from 0.537 in 2010 to 0.629 in 2022. Tianjin maintained the second-highest level most of the time. Among the 11 cities in Hebei Province, cities with a relatively high TecR include Langfang, Cangzhou, and Baoding. Their geographical proximity to Beijing and Tianjin, coupled with relatively well-developed infrastructure and advanced technological systems, enabled them to sustain higher TecR levels within Hebei Province. The top three cities in terms of the average annual growth rate of TecR were Xingtai (4.88%), Baoding (3.94%), and Shijiazhuang (3.61%), all of which were in Hebei Province.

In most cities of the BTHUA, significant pairwise positive correlations exist among SocR, EcoR, and TecR, demonstrating the synergistic effects among the three subsystems (Figure 4). Generally, the SocR-TecR correlation was the strongest, followed by that between EcoR-TecR, while the correlation between SocR-EcoR was relatively weak. The degree of synergy among the three systems varied across cities. For instance, Beijing, Tianjin, and Shijiazhuang exhibited strong and significant linkages (correlation coefficients greater than 0.9, p < 0.001) among SocR, EcoR, and TecR. By contrast, cities like Cangzhou and Chengde showed weak or insignificant synergistic effects. This discrepancy can be attributed to factors such as local industrial structures, development priorities, or policy orientations, which hinder the integration and mutual reinforcement of SocR, EcoR, and TecR.

3.2. The Spatiotemporal Changes of Coupling Coordination Development in the BTHUA

With the synergistic growth of SocR, EcoR, and TecR, the CCD of SocR, EcoR, and TecR demonstrated consistent growth over time in the BTHUA. The average CCD values across all cities were 0.482 in 2010, 0.547 in 2014, 0.616 in 2018, and 0.653 in 2022, with an annual average growth rate of 2.71%. From the perspective of the kernel density curve dynamics for the 13 cities in the BTHUA, the curves had been gradually shifting rightward (Figure 5), indicating a sustained upward trend in the CCD of SocR, EcoR, and TecR during 2010–2022. In terms of curve morphology, the left tail shortened, and the main peak transformed from a “broad and flat” shape to a “sharp and narrow” one. The CCD of most cities became more concentrated at the level of moderate coordination, with a reduction in the number of cities exhibiting a low CCD. As for the polarization characteristics, the kernel density curve gradually evolved from a “multi-peak” shape to a “double-peak” shape, signifying an overall decrease in the intercity disparities of the CCD within the BTHUA.

In 2010, the top three cities in terms of the CCD were Beijing (0.660), Hengshui (0.598), and Tianjin (0.547), while Cangzhou had the lowest CCD value of 0.329—resulting in a maximum–minimum difference of 0.231. The distribution of CCD classifications showed 3 cities in moderate imbalance, 6 in low imbalance, 3 in primary coordination, and 1 in moderate coordination, with 12 out of 13 cities in a transitional stage. As shown in Figure 6, the number of cities classified as moderate imbalance and low imbalance gradually decreased. By 2014, all cities had surpassed the moderate imbalance level. In 2018, Beijing became the first city to achieve good coordination. By 2022, Beijing maintained its leadership with a CCD value of 0.764. All cities achieved a CCD of over 0.6, and the maximum–minimum gap narrowed to 0.158. The number of cities with CCD values classified as moderate coordination and good coordination was 12 and 1, respectively, with all cities entering the coordination stage.

In addition, the overall dynamic spatial evolution of the CCD was analyzed using standard deviation ellipses for the years 2010, 2014, 2018, and 2022 (Figure 7 and

Table 3. The results of the standard deviation ellipse of the CCD in 2010, 2014, 2018, and 2022.

Year	CenterX	CenterY	XStdDist	YStdDist	Rotation (°)	Flatness (%)
2010	2,527,964.198	4,597,275.394	126.340	255.172	22.243	50.488
2014	2,525,369.788	4,590,548.178	128.740	259.588	21.791	50.406
2018	2,526,436.333	4,596,557.055	129.802	259.881	20.856	50.053
2022	2,525,910.888	4,596,941.887	130.568	259.767	20.864	49.737

). The center position of the CCD remained concentrated near the geographical center of the BTHUA. The center shifted slightly southwestward from 2010 to 2014 and northeastward from 2014 to 2022, indicating faster CCD growth in southwestern cities in the first period and more rapid increases in northeastern cities thereafter. The rotation angles of the CCD ellipses changed from 22.24° in 2010 to 20.86°in 2022, reflecting accelerated CCD development in the northeast. The flatness decreased continuously from 50.49% in 2010 to 49.74% in 2022, suggesting a reduction in intercity CCD disparities and a trend toward a more balanced spatial distribution of the CCD across the BTHUA.

3.3. Spatial Connection of CCD

From a holistic perspective, cities in the BTHUA demonstrated significant network characteristics in terms of the CCD of SocR, EcoR, and TecR, with no isolated cities. The spatial network of the CCD has become increasingly dense and complex, as evidenced by the growing number of intercity connections. Specifically, in 2010, 2014, 2018, and 2022 (Figure 8), the total number of linkages from city A to city B was 20, 29, 36, and 39, respectively, while that from city B to city A was 23, 26, 34, and 40, respectively. In addition, the spatial interactions between cities intensified over time, with the number of strong connections (above level IV) from city A to city B increasing from one in 2010 to eight in 2022.

At the sub-regional level, with the advancement of integrated development from 2010 to 2022, BTHUA cities gradually formed several CCD connection clusters, including the Beijing–Tianjin–Langfang cluster, Shijiazhuang–Hengshui cluster, and Xingtai–Handan cluster. Beijing consistently maintained strong linkages with neighboring cities, exerting far greater influence on surrounding areas than it received, which demonstrates Beijing’s leading role in the construction of the regional resilience network. Langfang, geographically adjacent to Beijing and Tianjin, was significantly influenced by these two cities, yet its own impact was relatively limited. As Langfang’s CCD increased over time, its influence on Beijing and Tianjin grew, while connections with adjacent cities like Baoding and Cangzhou also strengthened. With the advancement of regional integration, cities within Hebei Province became increasingly interconnected. Shijiazhuang, as the capital of Hebei Province, significantly influenced surrounding cities, especially Hengshui. Moreover, Xingtai and Handan in the southern BTHUA developed relatively close intercity linkages, forming a cohesive sub-regional cluster.

3.4. Obstacle Factor Identification

Through the obstacle degree model, the key barriers affecting the resilience of the BTHUA were identified at the dimension level, with their impact demonstrating significant changes from 2010 to 2022 (Figure 9). Resource endowment (E1) and technological innovation (T2) have consistently ranked among the primary obstacle factors. Specifically, the obstacle degree of E1 increased steadily from 17.31% in 2010 to 22.29% in 2022, making it the most primary constraint by 2022. The third obstacle factor, public service (S2), exhibited a marginal decline in its obstacle degree. From 2010 to 2016, economic strength (S1) was the fourth largest obstacle factor, while from 2017 onward, the obstacle degree of infrastructure (T1) exceeded that of S1, except in 2021 when S1’s obstacle degree rebounded slightly. In 2010, the obstacle degree of landscape pattern (E2) was slightly lower than that of social stability (S3). However, E2 demonstrated a pronounced upward trend, becoming the fourth largest obstacle factor in both 2020 and 2022. Correspondingly, S3’s obstacle degree showed a minor decline. Additionally, ecological quality (E3) and technical performance (T3) remained the least influential factors throughout the period, with both showing a consistent downward trend in their obstacle degrees.

To facilitate a more detailed analysis of disparities, Figure 10 illustrates the obstacle factors for each city in 2010 and 2022. While the major obstacle factors in most cities generally aligned with the regional pattern, distinct urban characteristics emerged, indicating variances in specific challenges and their impacts. For example, while T2 remained a crucial obstacle factor for many cities, its hindrance on Beijing’s resilience development gradually diminished, ceasing to be a major obstacle by 2022. In contrast, whereas S3 showed a downward trend in the obstacle degree across most cities, Beijing witnessed an increase in S3’s hindrance, which became the third largest obstacle factor in 2022, with an obstacle degree of 14.39%. Notably, despite E1’s high obstacle degree at the regional level, it was not a major obstacle factor for Chengde during the period, with its obstacle degree consistently below 10%. Although S1 was not a major regional obstacle factor, it retained significant influence on cities like Chengde, Zhangjiakou, Qinhuangdao, Handan, and Xingtai.

4. Discussion

4.1. The Resilience of the BTHUA

This study reveals that urban resilience in the BTHUA has trended upward since 2010, a pattern consistent with recent resilience research on UAs in China [43,45]. Such consistency is deeply rooted in China’s sustained efforts to advance urban resilience practices. The robust urban resilience has enhanced governance capabilities at the UA scale, thereby more effectively facilitating the BTHUA in achieving efficient governance and realizing ecologically friendly development.

From the perspective of subsystems, ScoR showed the highest growth rate, and this finding is consistent with the conclusions of other studies conducted in the BTHUA [25,42]. This rapid increase was primarily driven by economic expansion, improvements in social services, and the integrated development of urban and rural areas, all of which can be activated and enhanced through the implementation of targeted policies and coordinated collaborative efforts. TecR has been enhanced by continuous innovation and new technology applications, which help to update urban management concepts and strategies to address the misalignment between social development and ecosystem protection. Previous studies have demonstrated the promoting effect of sound infrastructure on urban resilience but overlooked the contributions of innovation capacity and technical efficiency, which may have led to underestimations of urban resilience levels. Also, TecR still requires deeper support from the social system, as its improvement has lagged behind that of SocR, which can be attributed to social factors such as policy orientation, institutional barriers, social equity issues, and public participation [70]. The enhancement of EcoR can be attributed to regional ecological restoration projects and resource conservation practices, yet its growth rate remains the lowest. This is primarily because the complex dynamics and threshold effects of ecosystems have made enhancing the key variables for maintaining ecosystem services a time-consuming process. Moreover, increasing urbanization pressure, compounded land, and water resource constraints have further constrained EcoR enhancement.

Urban resilience relies on multifaceted interactions and feedback among the components of urban subsystems, which manifest in diverse forms across urban areas. Studies on the BTHUA have shown that over the past decade, its economic–social and eco-environmental subsystems have achieved coupled and coordinated development, while disparities in the CCD among cities remain prominent [71,72]. Driven by the collective enhancement and mutual reinforcement of ScoR, EcoR and TecR, the CCD of all cities within the region has been improved. The region has shifted towards a sustainable development model that emphasizes the integration of social, ecological, and technological subsystems, rather than focusing solely on the welfare of the social subsystem. However, technological innovation and environmental improvement are not inherently positive. Without proper consideration of its wider inequalities and vulnerabilities, they can reinforce existing patterns of exclusion and even create new ones for disadvantaged groups [49].

The continuous growth in both the quantity and intensity of the spatial connections among cities in the BTHUA has demonstrated that intercity communication and cooperation within the region have become increasingly close. In particular, cities in the BTHUA have become more intricately connected through infrastructure, ecosystems, and socioeconomic systems, enabling the sharing of development strategies, resources, and technologies to enhance resilience. With key cities exerting spillover effects and regional integration advancing, an increasingly stable regional network and strong sub-regional linkages have emerged, making urban agglomerations more resilient. Maintaining more extensive connections is more conducive to reducing the vulnerability of the UA and enhancing its capacity to cope with risks [3,14].

Notably, the Beijing–Tianjin–Hebei coordinated development strategy, launched in 2014 to optimize the urban agglomeration’s development model, supports underdeveloped cities and promotes equitable resource allocation, and it has significantly enhanced regional resilience and propelled integrated development across the region. From 2014 to 2018, SocR and TecR achieved significant growth with the highest average annual increase, while EcoR transitioned from a period of fluctuation to a gradually upward trend. Since 2014, the intercity connection intensity has been continuously strengthened, driven by expanded industrial collaboration, a reinforced infrastructure network, and established coordinated pollution control mechanisms. Concurrently, the spatial distribution of the CCD has become more balanced. From 2018 to 2022, the rapid advancements of the preceding years likely approached a point of diminishing returns, where further improvements appeared to be increasingly constrained by resource limitations and systemic inefficiencies. Additionally, benefiting from the sound foundations of SocR, TecR, and EcoR in the BTHUA, the overall resilience still maintained slow and slight growth during the COVID-19 pandemic, despite the huge shocks brought by its outbreak [8,43].

Meanwhile, as one of China’s largest UAs, the BTHUA still faces some difficulties in enhancing its resilience. This study has identified several key barriers to the current development of the BTHUA, such as resource constraints, insufficient technological innovation, and inadequate social services. These findings will provide a more comprehensive reference for formulating targeted and integrated policy recommendations.

4.2. Policy Implications

To enhance urban resilience and promote coordinated and sustainable development in the BTHUA, the following recommendations are proposed.

First, it is crucial to develop comprehensive resilience enhancement strategies that are grounded in the interdependence among SocR, EcoR, and TecR. Given that inadequate infrastructure and technological innovation are major obstacles for most cities in the BTHUA, TecR should be fully considered in the formulation of urban resilience enhancement strategies. When allocating capital and resources, prioritizing and continuously optimizing interactions among the three resilience subsystems is essential to maximize their collective contributions to overall resilience. To ensure effective implementation, a robust evaluation and monitoring mechanism should be established to regularly track and assess the progress and impact of resilience enhancement initiatives, thereby providing a basis for continuous improvement and adaptation.

Second, it is important to enhance the radiation-driven role of key cities to optimize regional development patterns. Although urban development in the BTHUA has become more balanced, and several sub-regional high-performance clusters have emerged, significant gaps remain between cities. Given that key cities like Beijing exhibit high resilience and strong connectivity with surrounding cities, leveraging their advantages in human capital, economic development, infrastructure, and high-tech industries can enhance the learning and adaptive capabilities of neighboring cities. It is essential to continuously improve cooperation mechanisms to promote the joint construction and sharing of public services, thereby narrowing the regional development gap. Urban ecological spaces and infrastructure should be interconnected to form regional infrastructure and ecological corridor networks, facilitating resource and population flows, reducing redundancy, and strengthening intercity cooperation, thus enhancing the capacity for a coordinated response to regional challenges such as natural disasters or economic fluctuations.

Third, implementing city-specific policies tailored to each urban context is vital for enhancing UA resilience. Considering the differences in coordinated development levels among cities and the specific challenges they face, formulating and implementing differentiated strategies that adapt to local conditions and address specific needs is critical. For example, cities such as Chengde and Zhangjiakou, where the CCD was moderate but EcoR was high, should accelerate the development of eco-friendly industries to promote socioeconomic development while preserving their ecological strengths. Meanwhile, cities such as Langfang and Xingtai, which had relatively high TecR but constraints in economic strength and public services, should leverage their technological advantages to accelerate industrial and social transformation. By promoting intensive and efficient resource use, these cities can overcome limitations and bolster their resilience. Overall, each city in the BTHUA should actively harness its unique assets, mitigate weaknesses, foster heterogeneous advantages, and achieve mutual benefits.

4.3. Limitations of This Study

There are some limitations that should be further studied and clarified in the future. First, given the complex and deep interactions among SocR, TecR, and EcoR, as well as their dependence on policy orientation, government capacity, and the flows of capital, resources, population, and technology between regions [27,73], future research could employ methodologies like a dynamic Bayesian network [74], complex network analysis [75], and spatial panel regression models [76] to better quantify resilience and elucidate the coupling relationships. Second, the indicator system may not be perfect due to the limitation in data availability, necessitating more refined and long-term data to expand and enrich it, thereby better revealing evolutionary patterns. In particular, TecR indicators such as the patent self-sufficiency rate, ecological monitoring coverage, and technology transfer efficiency can more effectively reflect the coupled development between technological and socio-ecological systems. These metrics are vital for avoiding misjudgments of technological resilience by capturing the tangible interactions between technological innovation and its practical impacts on social and ecological contexts [77,78].

5. Conclusions

As China’s capital economic circle, innovation engine, and critical ecological zone, the BTHUA represents a typical Social–Ecological–Technological urban system confronted with the dual challenges of internal urban development disparities and external shocks. This study employed the SETS framework to analyze the spatiotemporal coupling characteristics of BTHUA’s resilience subsystems and intercity relationships and identified the key obstacle factors. By highlighting the independence and interconnectedness of TecR with SocR and EcoR as well as exploring resilience characteristics at both the urban and agglomeration spatial scales, this research contributes by providing practical recommendations for enhancing the overall resilience of the urban agglomeration.

The results indicate that from 2010 to 2022, when the SocR, EcoR, and TecR in the BTHUA exhibited unsteady upward trends, their CCD steadily grew over time. The intercity gaps in the CCD within the BTHUA narrowed, and the spatial distribution of resilience increasingly tended towards balance, while Beijing consistently maintained a leading position, and disparities among cities still persisted. By 2022, all the cities had reached the coordination stage. The spatial network of the CCD became increasingly dense and complex, with intercity connections growing continuously and no isolated cities left. In particular, key cities like Beijing progressively strengthened their linkages with surrounding cities, demonstrating a strong radiation effect. This led to the emergence of several urban clusters, including the Beijing–Tianjin–Langfang, Shijiazhuang–Hengshui, and Xingtai–Handan clusters. Moreover, the obstacle factors and their impact on urban resilience in the BTHUA underwent dynamic evolution. At the regional level, while the constraint exerted by economic strength on regional resilience has gradually weakened, resource endowment, public services, and technological innovation persist as the primary constraints, with their relative significance having shifted over time. At the local level, although each city generally aligns with the regional trends, notable disparities in the nature and intensity of these obstacles still exist. Hence, we propose a strategy to enhance the coordinated and coupled development of SETS resilience in the BTHUA, simultaneously leveraging the role of key cities to strengthen the regional resilience network and developing targeted approaches to promote sustainable regional development.