Evaluation of Urban Resilience and Its Influencing Factors: A Case Study of the Yichang–Jingzhou–Jingmen–Enshi Urban Agglomeration in China

Abstract

With the increasing frequency of various uncertainties and disturbances faced by urban systems, urban resilience is one of the vital components of the sustainability of modern cities. An indicator system is constructed to measure the resilience levels of the Yichang–Jingzhou–Jingmen–Enshi (YJJE) urban agglomeration during 2010–2023 based on four domains—economy, ecology, society, and infrastructure. This paper analyzes the spatiotemporal differentiation of resilience in YJJE in conjunction with the entropy weight method, Getis–Ord Gi* model, and robustness testing. Then, the factor contribution model is used to discern key driving elements of urban resilience. Finally, the CA-Markov model is implemented to predict urban resilience in 2030. The results reveal that the values of resilience in YJJE increase at a rate of 3.25%/a and continue to rise, with the differences among cities narrowing over the examined period. Furthermore, the urban resilience exhibits a significant spatially heterogeneity distribution, with Xiling, Wujiagang, Xiaoting, Yidu, Zhijiang, Dianjun, Dangyang, Yuan’an, Yiling, and Duodao being the high-value agglomerations of urban resilience, and Hefeng, Jianli, Shishou, and Wufeng being the low-value agglomerations of urban resilience. The marked heterogeneity of resilience in the YJJE urban agglomeration reflects the disparity in economic progress across the study area. The total amount of urban social retail, financial expenditure per capita, GDP per capita, park green space area, urban disposable income per capita, and number of buses per 10,000 people surface as the key influencing factors in relation to urban resilience. Finally, the levels of resilience among cities within YJJE will reach the medium level or higher than medium level in 2030. Xiling, Wujiagang, Xiaoting, Zhijiang, Dianjun, Dangyang, and Yuan’an will remain significant hot spots of urban resilience, while Jianli will remain a significant cold spot. In a nutshell, this paper can provide scientific references and policy recommendations for policymakers, urban planners, and researchers on the aspects of urban resilience and sustainable city.

1. Introduction

As the most complex social ecological system, cities are extremely susceptible to disturbances and impacts from the climate and ecological change, globalization, natural disasters, social security accidents, economic crises, public health incidents, and so on [1,2,3]. Humanity has entered the Anthropocene dominated by human activities [4], and urbanization has become one of the most transformative trends in human society in the 21st century [5]. However, while urbanization and industrialization have brought well-being to human beings, they have also created numerous problems that constrain the sustainable development of cities [6,7]. Resilience is regarded as a vital component of sustainability [8]. Urban resilience is important to the sustainability of cities as it improve cities’ ability to offset, resist, and reduce external uncertain shocks and disturbances, and to bolster urban social ecological system sustainability [9,10,11,12,13,14,15,16].

Currently, several countries and organizations have proposed projects and international initiatives on urban resilience [7,14]. The United Nations New Urban Agenda and 2030 Agenda, and the U.S. Rockefeller Foundation Project “Global 100 Resilient Cities”, emphasized that resilience is one of the core goals of urban social ecological system sustainable development [6,17,18]. In 2024, the Chinese government pointed out the significance of resilience in promoting the high-quality development of China in the Resolution of the Central Committee of the Communist Party of China on Further Deepening Reform Comprehensively to Advance Chinese Modernization [7,8,9,10,11,12,19,20].

The ecologist Holling first introduced the term resilience to the field of ecology in 1973 when he published a seminal article about the resilience and stability of ecological systems [21]. Since then, resilience research has expanded from the field of ecology to the different fields of engineering, economics, society, psychology, ethnology, etc. [22,23,24,25]. In the 21st century, the term resilience has emerged in research on sustainable cities [17,26]. Scholars explored the definition of urban resilience from the various perspectives of bioscience, ecology, economics, management, geography and engineering [7,20,27].

Scholars have constructed models for quantitatively measuring and monitoring urban resilience by using multiple indicators, and then they have analyzed the spatial-temporal differences and the major or primary factors of urban resilience among different cities or urban agglomerations [7,20,28]. For example, domestic and international scholars constructed a comprehensive urban resilience assessment framework and an indicator system covering the dimensions of society, economy, housing and infrastructure, ecological environment and climate, community, culture, institution and governance, physical and technological, social network and individual behavior, and they used subjective or objective weighting methods to measure the urban resilience of communities, urban networks, cities, megacities, economic circles, economic belt, urban agglomerations, provinces, countries, and the world [8,9,10,11,28,29,30,31]. Next, the academic community analyzed the spatiotemporal evolution of urban resilience and frequently applied the methods of Getis–Ord Gi analysis, exploratory spatial data analysis and the Theil index to assess the difference in urban resilience among different regions and periods [32,33,34,35]. Finally, scholars adopted mathematical models, geographic detectors, and statistical methods to explore the driving factors behind urban resilience [36,37]. They found that natural factors, economic factors and social factors affect the spatial and temporal differences of urban resilience [36,38], and they put forward specific suggestions to improve the urban resilience [20,39]. In China, recent research studies on urban resilience in different geographic contexts were conducted by scholars, such as in the eastern, western, central, southern, northern regions of China [7,9,10,15,20,26,33,36]. Due to the accessibility of data, city data at the prefecture-level or above prefecture-level were usually used by domestic scholars in urban resilience research studies, and very limited data at the levels of the community, social network, or individual exist in the current research [8,9,10,11].

To sum up, the above scholars have provided the theoretical support and methodological guidance for this study. However, urban resilience is still a complicated and interdisciplinary matter requiring more in-depth research. Firstly, the current research lacks robustness analysis for different standardized methods while ignoring the uncertainty of urban resilience assessment caused by using a single standardization method. Secondly, there is very limited research on urban resilience prediction for the future. Thirdly, current research studies are mostly conducted at the spatial scales of prefecture-level or above prefecture-level. Research studies at the county-level or multi spatial scales are still lacking. Fourthly, more recent data are required in current studies.

The YJJE urban agglomeration is an important component of the middle reaches of the Yangtze River (MRYR), which is a major urban agglomeration in China. The YJJE urban agglomeration plays a pivotal role in the development of Hubei Province, China, with robust economic dynamics and notable urban expansions. Yichang in the YJJE urban agglomeration is one of the regional centric cities of Central China. The cities in the YJJE urban agglomeration are promoting the construction of livable, resilient, smart cities.

To the best of our knowledge, the objectives of this paper are: (1) to construct an urban resilience evaluation indicator system from the four dimensions of economy, ecology, society, and infrastructure, and scientifically measure urban resilience based on robustness analysis; (2) to investigate the spatiotemporal evolution characteristics of urban resilience in the YJJE urban agglomeration during 2010–2023, considering two spatial scales of urban resilience: prefecture-level and county-level; (3) to examine the factors that influence the urban resilience of the study area during the study period; and (4) to predict the urban resilience of the YJJE urban agglomeration in 2030 and provide reference experience for exploring urban resilience in the future.

In Section 2, the study area, indicator system, data sources, and methods are introduced. In Section 3, the spatiotemporal differentiation characteristics of urban resilience are measured and analyzed, the drivers of urban resilience are identified, and the urban resilience in 2030 is predicted. The discussion and conclusions of this paper are presented in Section 4 and Section 5, respectively.

2. Materials and Methods

2.1. Study Area

The YJJE urban agglomeration in Hubei Province, China, was chosen as the study area in this paper. It connects Central China and Southwest China and encompasses 4 prefecture-level cities (Yichang, Jingzhou, Jingmen, and Enshi) and 34 county-level cities (Figure 1). Due to vigorous economic vitality and significant urban expansion, the YJJE urban agglomeration plays an increasingly important role in radiating to the surrounding regions, and it occupies a vital position in the development of Hubei Province. According to the criteria of the Chinese Government [40], Yichang is classified as a metropolis. Jingzhou and Jingmen are medium-sized cities, while Enshi is a small city. Moreover, Yichang, Jingzhou, Jingmen, and Enshi encompass 13 county-level cities, 8 county-level cities, 5 county-level cities, and 8 county-level cities, respectively (Figure 1). According to the data on the official website of the Hubei Provincial Statistics Bureau (https://tjj.hubei.gov.cn/tjsj/sjkscx/tjnj/gsztj/, accessed on 20 July 2024), the GDP per capita of Yichang, Jingzhou, Jingmen, and Enshi is CNY 1,468,000, CNY 614,000, CNY 891,000, and CNY 436,000, respectively. According to data on the website of the Hubei Meteorological Service (http://hb.cma.gov.cn/, accessed on 22 July 2024), the annual mean temperature of Yichang, Jingzhou, Jingmen, and Enshi is 17.1 °C, 16.5 °C, 16.7 °C, 16.2 °C, while the annual precipitation in these cities is 1161.0 mm, 1083.9 mm, 961.8 mm, 1600.0 mm, respectively. The climate type of the YJJE urban agglomeration is a subtropical monsoon climate. Mountains and hills are located in the western part of the YJJE urban agglomeration, while the Jianghan Plain is located in the eastern part of it (Figure 1).

2.2. Construction of the Indicator System

The geographical-temporal difference in resilience in the YJJE urban agglomeration during 2010–2023 will be investigated in this paper based on the indicator system, which consists of the 4 domains of economy, ecology, society, and infrastructure.

Economy resilience is the capability of an economic system to deal with external disturbances and risk shocks, and to reduce losses [20,41]. Referring to relevant recent studies and due to the accessibility of data, we select six indicators, industrial structure, GDP, savings deposit, governmental financial expenditure, social retail amount, and fixed asset investment, based on the structure, foundation, and activity of the economy.

Ecology resilience emphasizes the sustainable capability of the urban ecosystem in the city development process [20,42]. Referring to relevant recent studies and due to the accessibility of data, we select six indicators, cover of green area, air quality level, treatment capacity of waste, and so on, based on the environmental quality and purification capacity of the urban ecosystem.

Society resilience reflects the ability of the social system to cope with external pressures, disturbances, uncertainties and crises [43]. Referring to relevant recent studies and due to the accessibility of data, we select six indicators, disposable income, education level, medical care level, grain yield level and social management level, based on social security, public healthcare security, social and public management.

Infrastructure resilience emphasizes the capability and material base of the urban infrastructure to guarantee the healthy operation and development of a city [7,20]. Referring to relevant recent studies and due to the accessibility of data, we select six indicators, public transport, power supply, pipe supply, water supply and gas supply, based on the supply and distribution of the municipal establishment and infrastructure.

The indicator system consists of 24 indicators (

Table 1. Comprehensive evaluation indicator system for urban resilience in the YJJE urban agglomeration.

Domains	Indicators	Unit	Variable
Economy	GDP per capita	CNY 10,000	a1
resilience	The proportion of tertiary industry in GDP	%	a2
	Savings deposit per capita	CNY 10,000	a3
	Financial expenditure per capita	CNY 10,000	a4
	Total amount of urban social retail	CNY 10,000	a5
	Total fixed asset investment	CNY 10,000	a6
Ecology	Greening coverage rate of built-up area	%	b1
resilience	Proportion of days with air quality index (AQI) < 100 in a year	%	b2
	Park green space area	ha	b3
	Treatment rate of living waste in city	%	b4
	Comprehensive utilization rate of general industrial solid waste	%	b5
	Domestic sewage treatment rate	%	b6
Society	Urban disposable income per capita	CNY 10,000	c1
resilience	Number of hospital beds per 10,000 people	per 10,000 people	c2
	The investments on education	CNY 10,000	c3
	Grain yield per capita	kg	c4
	Number of medical technical personnel per 10,000 people	per 10,000 people	c5
	Public management and social organization personnel per 10,000 people	per 10,000 people	c6
Infrastructure	Number of buses per 10,000 people	per 10,000 people	d1
resilience	Per capita power supply	kw·h/person	d2
	Road area per capita	m2/person	d3
	Density of urban drainage pipes	km/km2	d4
	Per capita water supply	m3/person	d5
	Gas penetration rate	%	d6

), and the properties of these indicators are positive.

2.3. Method

2.3.1. Standardization Method

In this paper, three standard methods of transformation were used to standardize each indicator for the urban resilience evaluation. The range standardization method is shown in Formula (1). The formulas for the z-transformation standardization method and interquartile range standardization method can be found in reference [44]. Then, three standardized values of each indicator were used in the next steps, such as the specific calculations of the entropy weight method and robustness analysis. With a general focus on a robust method, one of the three standardization methods was selected after robustness analysis for the calculation of the comprehensive urban resilience index.

The range standardization method:

$$X_i={\displaystyle \frac{x_i-x_{i\min }}{x_{i\max }-x_{i\min }}}$$

(1)

where X_i is the standardized index value of the indicator i, while x_i, x_imax, and x_imin are the initial value, the maximum value and minimum value of the indicator i, respectively [44].

2.3.2. Entropy Weight Method

After Section 2.3.1, the entropy weight method was used to calculate the weight of each indicator. The specific formulas are as follows:

$$P_{ij}={\displaystyle \frac{X_{ij}}{{\displaystyle \sum _{j=1}^n{X}_{ij}}}} (n=1, 2, \dots , 34)$$

(2)

$$e_i=-{\displaystyle \frac{1}{ln{n}_j}}{\displaystyle {\sum }_{j=1}^n{P}_{ij}ln{P}_{ij}}$$

(3)

$$d_i=1-e_i$$

(4)

$$w_i=d_i/{\displaystyle {\sum }_{i=1}^m{d}_i} (m=1, 2, \dots , 24)$$

(5)

where P_ij is the normalized index value of the indicator i in the city j, X_ij is the standardized index value of the indicator i in the city j in Section 2.3.1. e_i and d_i are the information entropy and difference coefficient of the indicator i, respectively. n is the total number of cities, and m is the total number of indicators. Finally, w_i is the weight of the indicator i.

2.3.3. Robustness Analysis

The construction of composite indicators involves the selection of various methods in the development process. However, this may result in various issues of uncertainty due to the selection of the methods of imputation, normalization, standardization, weighting, and aggregation [45]. In the calculation of the comprehensive urban resilience index, the different selections of data standardization methods can lead to changes in the rankings among different cities, which further increases the uncertainty of the composite indicator [45]. Robustness checking can help to enhance the transparency and credibility of the urban resilience indices [45,46]. In this paper, three standardization methods were used to calculate the urban resilience index of 4 prefecture-level cities and 34 county-level cities in 2010, 2015, 2020, and 2023, and then to obtain a ranking of 12 urban resilience index values at the prefecture-level or county-level. Furthermore, three box plots of the urban resilience index rankings in the YJJE urban agglomeration were constructed based on the variation range, which were further used to measure the robustness of the calculation results concerning urban resilience. The smaller the variation range in the ranking of urban resilience, the stronger the robustness [45,46]. Thus, the one standardization method with the strongest robustness was selected from the three standardization methods to calculate the urban resilience index.

2.3.4. Calculation of Urban Resilience Index

The constructed evaluation model formula of urban resilience is as follows:

$$URI={\sum }_{i=1}^m{w}_i{X}_i^{\prime } (m=1, 2, \dots , 24)$$

(6)

where URI is the urban resilience value of the city in the YJJE urban agglomeration and m is the total number of variables. w_i is the weight of the indicator i, while $X_i^{\prime }$ is the standardized index value of the indicator i, which is calculated using the one of the three standardization methods in Section 2.3.1 after the robustness analysis. The larger the URI value, the higher the urban resilience. In addition, the levels of urban resilience are shown in

Table 2. The levels of urban resilience.

Level	Low	Relatively Low	Medium	Relatively High	High
Urban resilience value	[0.0, 0.2)	[0.2, 0.4)	[0.4, 0.6)	[0.6, 0.8)	[0.8, 1.0]

.

2.3.5. Getis–Ord Gi* Model

The Getis–Ord Gi* model is usually used in relation to local spatial autocorrelation statistics, and it is an effective method for discerning the hot spot (high-value clusters), cold spot (low-value clusters), and not significant region in the spatial distribution of resilience [7,47]. The confidence levels of the hot spot or cold spot encompass 99%, 95%, and 90%. The specific formulas are:

$$G_i^{*}={\displaystyle \frac{{\displaystyle \sum _{j=1}^n{w}_{(i,j)}{x}_j}-\overline{x}{\displaystyle \sum _{j=1}^n{w}_{(i,j)}}}{S\sqrt{\frac{[n{\displaystyle \sum _{j=1}^n\underset{(i,j)}{\overset{2}{w}}-{({\displaystyle \sum _{j=1}^n{w}_{(i,j)}})}^2}]}{n-1}}}}$$

(7)

$$\overline{x}={\displaystyle \frac{{\displaystyle \sum _{j=1}^n{x}_j}}{n}}$$

(8)

$$s=\sqrt{{\displaystyle \frac{{\displaystyle \sum _{j=1}^n\underset{j}{\overset{2}{x}}}}{n}}-{(\overline{x})}^2}$$

(9)

where $G_i^{*}$ is the value of local $G_i^{*}$ index, x_j is the urban resilience value of city j in the YJJE urban agglomeration, w_(i,j) is the spatial weight between city i and city j, and n is the total number of cities [48]. In this paper, the specific calculations of the Getis–Ord Gi* model are completed in the ArcGIS software platform.

2.3.6. Factor Contribution Model

To investigate the influence of each indicator on the urban resilience of the YJJE urban agglomeration, we used the factor contribution model [48] to identify the contribution value of each indicator to the urban resilience, and the specific calculation is as follows:

$$O_j={\displaystyle \frac{I_j\times W_j}{{\displaystyle \sum _{j=1}^n{I}_j\times W_j}}}$$

(10)

where O_j is the contribution value of the driving factor j on the urban resilience; I_j is gap between the actual value and the optimal target value of the driving factor j, which is usually expressed as the difference between 1 and the standardization value of indicator j; W_j is the weight of the driving factor j; and n is the total number of driving factors.

Factors with O_j of more than 10% are regarded as key obstacle factors [48].

2.3.7. CA-Markov Model

The Markov chain can control the temporal variation, while the Cellular Automata (CA) conversion rules can control the spatial variation [49]. Thus, the CA-Markov model combines the CA model with the Markov chain, giving itself the ability to predict and analyze both spatial and temporal changes. The model can integrate the continuum spatial distribution elements and the possible spatial distribution transitions into the Markov chain, so that the Markov chain provides a spatial prediction of the system state at the t + 1 time node based on the system state change rules during the t − 1 to t time node [50].

In this paper, the CA-Markov model is used to predict the urban resilience of the YJJE urban agglomeration in 2030. The specific processes encompass the (1) transition probability matrix operation, (2) transition suitability image creation, and (3) spatial distribution simulation. The specific calculations were completed with using the IDRISI software platform. In the processes, we chose the urban resilience map in 2010 as the earlier image, while the resilience map in 2020 was chosen as the later image in the Markovian transition estimator. Then, the prefix for the output conditional probability images was named 1020, and the number of time periods between the earlier and later images or between the later and predicted images was 10. The proportional error was 0.15. Next, the transition probability matrix file was generated, and then the CA/Markov change prediction was finished based on it. More details can be found in reference [51].

2.4. Data Sources

This study used the 4 prefecture-level cities and 34 county-level cities in the YJJE urban agglomeration as the case study. The data used in this study were derived from the Statistical Yearbook of Yichang, Jingzhou, Jingmen, and Enshi, China Urban Statistical Yearbook, China Statistical Yearbook (County-Level), Environmental Quality Annual Report or Environmental Statistics Bulletin of Yichang, Jingzhou, Jingmen, and Enshi, Government Gazette on the Urban Construction Status of Yichang, Jingzhou, Jingmen, and Enshi.

3. Results

3.1. Results of the Robustness Analysis

In this paper, three box plots (Figure 2, Figure 3 and Figure 4) of the urban resilience index rankings in the YJJE urban agglomeration were constructed based on the interquartile range standardization method, the range standardization method, and the z-transformation standardization method, respectively. In these box plots, the highest ranking order and lowest ranking order showed the variation range in the urban resilience index rankings of each city in the YJJE urban agglomeration. The stable ranking order was the order with the most occurrence frequency among the urban resilience index rankings based on the different standardization methods. From the comparison of Figure 2, Figure 3 and Figure 4, the variation range of the majority of cities was the smallest in the urban resilience index rankings based on the range standardization method. The smaller the variation range in the ranking of urban resilience, the stronger the robustness [45,46]. Thus, the range standardization method was selected from the three standardization methods in this paper to calculate the urban resilience index.

3.2. Spatial-Temporal Differentiation Characteristics of Urban Resilience

The results of the spatial-temporal differentiation characteristics of urban resilience showed that the average value of YJJE was 0.471 in 2023 (

Table 3. The mean value of urban resilience in the YJJE urban agglomeration and its prefecture-level cities.

Name	2010	2015	2020	2023
Yichang	0.379	0.427	0.461	0.496
Jingzhou	0.294	0.350	0.401	0.450
Jingmen	0.356	0.422	0.456	0.504
Enshi	0.276	0.350	0.395	0.430
YJJE urban agglomeration	0.331	0.390	0.430	0.471

). From 2010 to 2023, the YJJE urban agglomeration’s urban resilience increased by 0.140, with a rate of 3.25%/a. From the perspective of prefecture-level cities, the value of Yichang increased by 0.117 during 2010–2023, with a rate of 2.37%/a. At the same time, the urban resilience of Jingzhou increased by 0.156, with a rate of 4.08%/a; the urban resilience of Jingmen increased by 0.148, with a rate of 3.20%/a; and the urban resilience of Enshi increased by 0.154, with a rate of 4.29%/a (Table 3). The urban resilience values of all the prefecture-level cities displayed an increasing trend between 2010 and 2023, and this indicated that the capability of YJJE in terms of dealing with external uncertainties, shocks, disturbances, and risks is improving. Meanwhile, the differences in resilience among the cities within the study area were narrowing over the examined period. Among all 34 county-level cities, Xiling had the highest value of urban resilience in 2010, which was 0.560, while Hefeng had the lowest value, which was 0.171. Furthermore, 10 county-level cities reached the medium level of urban resilience in 2010, while 18 county-level cities reached the relatively low level and 6 county-level cities remained at the low level (Figure 5). In 2015, Xiling reached the relatively high level of urban resilience, while 14 county-level cities reached the medium level and the remaining 19 county-level cities reached the relatively low level (Figure 5). In 2020, Xiling also reached the relatively high level of urban resilience, while Wufeng had the lowest value (0.30) of urban resilience among all 34 county-level cities. Meanwhile, 18 county-level cities reached the medium level of urban resilience in 2020, while the remaining 15 county-level cities reached the relatively low level (Figure 5). In 2023, Xiling had the highest value of urban resilience among the county-level cities in the YJJE urban agglomeration, which was 0.660, while Wufeng had the lowest value, which was 0.342. Furthermore, 6 county-level cities reached the relatively high level of urban resilience in 2023, while 21 county-level cities reached the medium level and the remaining 7 county-level cities reached the relatively low level (Figure 5). From 2010 to 2023, the urban resilience values of all the county-level cities showed an upward trend, and this indicated that the resilience of the urban system is improving. Meanwhile, the urban resilience values of the regional centers have consistently remained higher than those of other regions, such as Xiling, Wujiagang, Yiling, Jingzhou, Dongbao, and Enshi. Additionally, the urban resilience of the county-level cities near mountainous areas has consistently remained lower than the average value of the YJJE urban agglomeration or the prefecture-level cities, such as Wufeng, Hefeng, Badong, and Xuan’en.

The marked heterogeneity of urban resilience was observed by the Getis–Ord Gi* model across the YJJE urban agglomeration from 2010 to 2023. Xiling, Wujiagang, Xiaoting, Yidu, Zhijiang, Dianjun, Dangyang, Yuan’an, Yiling, and Duodao showed high-value agglomerations of urban resilience, while Hefeng, Jianli, Shishou, and Wufeng showed low-value agglomerations of urban resilience (Figure 6). From 2010 to 2020, Xiling, Wujiagang, Xiaoting, Dianjun, and Zhijiang stood out as the significant hot spots (99% confidence), while Yuan’an and Dangyang exhibited 95% confidence as a hot spot. During the study periods, Yiling became a hot spot (90% confidence) only once, which was in the year 2010. Similarly, Duodao became a hot spot (90% confidence) only once, which was in the year 2023. As a hot spot, the significance levels of Yidu varied in the period 2010–2023. And the confidence level of Yidu was 95% in 2010 and 2015, while it was 90% in 2020 and 2023. In 2023, Wujiagang and Dangyang stood out as the significant high-value clusters with the confidence level of 99%, while Xiling, Xiaoting, Dianjun, Zhijiang, and Yuan’an exhibited high-value clusters with the confidence level of 95%. As a cold spot, the significance levels of Jianli varied in the period 2010–2023. And the confidence level of Jianli was 95% in 2010, 2015 and 2020, while it was 90% in 2023. Similarly, the significance level of Hefeng, which was a low-value cluster, varied during the study periods. The confidence level of Hefeng was 95% in 2010, while it was 90% in 2015, 2020 and 2023. As a low-value cluster, the significance level of Shishou also varied in the study periods. And the confidence level of Shishou was 95% in 2010 and 2015, while it was 90% in 2020. Wufeng stood out as a cold spot (90% confidence) only once, which was in the year 2010.

3.3. Driving Factors of Urban Resilience

An analysis was conducted based on the factor contribution model (Formula (10)) to discern the major influencing factors of resilience in the YJJE urban agglomeration from 2010 to 2023 (Figure 7). The results showed that the contribution values of the six indicators (Figure 7) to urban resilience were more than 10%, and this indicated that these six indicators were the key influencing elements of resilience in the YJJE urban agglomeration between 2010 and 2023.

Foremost, the total amount of urban social retail (a5), financial expenditure per capita (a4), and GDP per capita (a1) surfaced as the key influencing factors in relation to urban resilience, emphasizing the critical importance of the economic resilience domain and the necessity of economic support. The marked heterogeneity of resilience in the YJJE urban agglomeration reflected the disparity in economic progress across the study area. Notably, the park green space area (b3) emerged as a major driving factor, indicating the strong influence of urban ecology and environment on resilience. The urban disposable income per capita (c1) was discerned as a key influencing factor, presenting the vital role of the society resilience domain and the necessity of social security. The number of buses per 10,000 people (d1) stood out as a driving factor, showing the crucial influence of infrastructure with regard to urban resilience. In terms of the remaining 18 factors, even though these factors showed relatively lower contribution values in the study area, their presence emphasized the importance of them.

3.4. Modeling Changes in Urban Resilience in the Future

In this paper, the CA-Markov model was used to predict the urban resilience of the YJJE urban agglomeration in 2030. The results showed that the average value of urban resilience in the YJJE urban agglomeration was 0.563 in 2030 (Figure 8). From the perspective of prefecture-level cities, Yichang, Jingzhou, Jingmen, and Enshi all reached the medium level of urban resilience in 2030. Yichang had the highest value of urban resilience, which was 0.586, while Enshi had the lowest value, which was 0.530 (Figure 8). From the perspective of county-level cities, Xiling had the highest value of urban resilience among all 34 county-level cities in 2023, which was 0.809, while Wufeng had the lowest value, which was 0.430. Furthermore, Xiling reached the high level of urban resilience in 2023, while 11 county-level cities reached the relatively high level of urban resilience and the remaining 22 county-level cities reached the medium level (Figure 8).

The Getis–Ord Gi* model was used in this paper to identify high-value clusters (hot spot) and low-value clusters (cold spot) of urban resilience in 2030. Xiling, Wujiagang, Xiaoting, Zhijiang, Dianjun, Dangyang, and Yuan’an showed high-value agglomerations of urban resilience, while Jianli showed low-value agglomerations of urban resilience (Figure 8). In 2030, Xiling was identified as a hot spot with 90% confidence, while Wujiagang, Xiaoting, Zhijiang, Dianjun, Dangyang, and Yuan’an were identified as a hot spot with 95% confidence. Meanwhile, Jianli was identified as a cold spot with 90% confidence (Figure 8).

4. Discussion

Urbanization has become one of the most important development trends in the world today [52]. Improving urban resilience can increase a city’s ability to cope with external uncertainties, shocks, disturbances, and risks [53]. Scholars have explored spatiotemporal evolution and the driving factors of urban resilience in various regions of China [7,20,52,53,54]. Moreover, the Yangtze River Economic Belt (YREB) and the urban agglomerations in the MRYR have recently become a focus of urban resilience research in China [55,56,57,58,59,60,61]. Scholars found that the resilience of cities appeared to be on an upward trend in the MRYR from 2009 to 2020 [20,56,59], and the resilience values of 110 prefecture-level cities in the YREB were improved during 2010–2019 [53]. Similarly, this study revealed that the urban resilience values of the YJJE urban agglomeration, an important component of the MRYR, showed an upward trend from 2010 to 2023. Scholars found that the urban resilience of Yichang reached the medium level in 2019 [53]. Similarly, this paper revealed that Yichang reached medium level of resilience in 2020. Scholars found that the MRYR increased at a rate of 3.86%/a in the field of urban resilience between 2005 and 2018 [20]. Similarly, this study revealed that the YJJE urban agglomeration increased at a rate of 3.25%/a in the aspect of urban resilience from 2010 to 2023.

Scholars have argued that the MRYR exhibited heterogeneity in the field of urban resilience, and the value of Yichang was higher than that of Jingzhou, Jingmen, and Enshi in 2019 [53,55]. Similarly, this paper revealed that the YJJE urban agglomeration showed a spatially heterogeneous distribution in the aspect of resilience, and the value of Yichang was higher than that of Jingzhou, Jingmen, and Enshi in 2020. Furthermore, Xiling, Wujiagang, Xiaoting, Yidu, Zhijiang, Dianjun, Dangyang, Yuan’an, Yiling, and Duodao were the high-value agglomerations of urban resilience, and Hefeng, Jianli, Shishou, and Wufeng were the low-value agglomerations of urban resilience during 2010–2023. Moreover, scholars have argued that the economic resilience of the MRYR exerted a strong influence or driving force on the urban resilience of the MRYR during 2005–2019 [20,55]. Similarly, this study revealed that economic resilience was the major force behind urban resilience. Scholars found that the income of urban areas and local fiscal revenue were the key factors affecting the values of resilience in the MRYR during 2005–2020 [59]. Similarly, this study revealed that the financial expenditure per capita and urban disposable income per capita were the critical driving factors behind urban resilience in the YJJE urban agglomeration between 2010 and 2023. In other regions of the Yangtze River, such as the Yangtze River Delta (YRD), scholars found that differences in economic development were responsible for the heterogeneity in urban resilience [9]. Similarly, this study revealed that the marked heterogeneity of resilience reflected the disparity in economic progress across the study area. In the Headwater Region of the Yangtze River, the per capita disposable income was a critical influencing factor in relation to the resilience of the pastoral community [48]. Similarly, this study found that the urban disposable income per capita was a key driving factor behind resilience. In the period of 2011–2019, the regional differences in urban resilience within the YRD were narrowing [61]. Similarly, this study found that the differences in resilience among the cities within the study area were narrowing over the period of 2010–2023.

Scholars have proposed policy recommendations for the government to build resilient cities and improve sustainable urban development. For example, enhancing the levels of economic integrations, connections, collaborations, complementary capabilities, and coordination between different cities is a valid pathway for the reinforcement of urban resilience [9,48,61]. Strengthening the levels of openness, championing the protection of the environment, amplifying the support for the economy, and reinforcing the synergies of regional development are also the key suggestions and insights of scholars [45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61]. In this paper, we suggest that reinforcing the openness level, amplifying the governmental support for finance, strengthening the economic coordination or integration, improving the social security and supply, and advancing regional environment conservation are the aspects or areas for future research on urban resilience within the YJJE urban agglomeration.

In this paper, we used multiple methods to provide a robust and scientifically rigorous approach to ensure the reliability of measuring and analyzing urban resilience. However, there are uncertainties in this study. The study focuses solely on the YJJE urban agglomeration at the prefecture-level and county-level, which might lack research on other spatial resolutions within the YJJE and limit the generalizability of the findings to other regions or countries. Although the data sources are derived from various government statistical materials, there might still be inaccuracies in these sources. The choice of indicators for measuring urban resilience might not capture all the relevant aspects. For example, short-term economic fluctuation is not accounted for in the domain of economy resilience; microclimate, local biodiversity, specific ecological interactions, and extreme climate events are not accounted in the domain of ecology resilience; local governance structure, community engagement, social networks, individual behaviors, and cultural differences are not accounted for in the domain of society resilience; and the quality and maintenance of infrastructure are not accounted for in the domain of infrastructure resilience. Due to the accessibility of data, we need to make in-depth efforts to obtain more recent data on various aspects and improve the indicator system based on the obtained data in the future. The chosen method might still introduce biases or inaccuracies, such as the selected standardization method. The prediction for 2030 using the CA-Markov model is based on historical data and current trends, which might not account for unexpected future developments, deviations or disruptions. The policy recommendations we propose might not fully consider the practical challenges of implementing these recommendations in different political and economic contexts.

5. Conclusions

This study rigorously measured, assessed and predicted the spatial-temporal differentiation of urban resilience and discerned the key driving elements in the YJJE urban agglomeration from 2010 to 2023. The conclusions are as follows:

(1)

The urban resilience of the YJJE urban agglomeration increased at a rate of 3.25%/a and continues to rise, with the differences among cities narrowing over the period of 2010–2023. Meanwhile, the urban resilience values of the regional centers have consistently remained higher than those of county-level cities near mountainous areas. A marked heterogeneity was discerned, with Xiling, Wujiagang, Xiaoting, Yidu, Zhijiang, Dianjun, Dangyang, Yuan’an, Yiling, and Duodao being the hot spots of urban resilience, and Hefeng, Jianli, Shishou, and Wufeng being the cold spots of urban resilience. In 2023, all the prefecture-level cities and 27 county-level cities within the YJJE reached the medium level or higher than medium level of urban resilience.

(2)

The total amount of urban social retail, park green space area, financial expenditure per capita, urban disposable income per capita, GDP per capita, and number of buses per 10,000 people stood out as the key influencing elements in relation to urban resilience.

(3)

The urban resilience among cities within the YJJE will reach the medium level or higher than medium level in 2030. Xiling, Wujiagang, Xiaoting, Zhijiang, Dianjun, Dangyang, and Yuan’an are high-value agglomerations of urban resilience, while Jianli is a low-value agglomeration of urban resilience.

(4)

Policymakers could focus on several crucial aspects, which include reinforcing the openness level, amplifying the governmental support for finance, strengthening the economic coordination or integration, improving the social security and supply, and advancing regional environment conservation.