Study on Enhancement Effect of Climate-Resilient City Pilot Policy Construction on Urban Ecological Resilience

Abstract

Under the severe situation of increasing global climate change, it is urgent to improve the ability of cities to cope with climate change and achieve sustainable development. As a key institutional arrangement for China’s climate adaptation, the climate-resilient city initiative has been piloted in 67 cities across two batches since 2017, aiming to foster urban resilience through systematic governance. Based on complex adaptive system theory, this study constructs an urban ecological resilience evaluation framework under the “Pressure–State–Response” (PSR) model. Using panel data from 243 prefecture-level cities from 2010 to 2022 and a difference-in-differences model, it empirically examines the impact of climate-resilient city construction on ecological resilience, further exploring the moderating mechanism of government attention to environmental protection and spatial heterogeneity effects. Key findings include the following: (1) climate-resilient city construction significantly enhances urban ecological resilience, with pilot cities experiencing an average increase of approximately 0.74%; (2) government attention to environmental protection strengthens policy effectiveness, demonstrating a significant positive moderating effect; and (3) policy effects show notable regional variations, with more pronounced improvements in resource-based cities, western regions, and ecologically vulnerable areas.

1. Introduction

The global ecological landscape is undergoing profound and uncertain restructuring. Climate warming has triggered a shift in extreme weather events from isolated incidents to multiple, compound, and systemic crises. These evolving threats pose severe risks to the structural stability and functional continuity of urban ecosystems, presenting a critical test of national adaptive capacity [1]. In April 2025, Beijing issued its first orange-level gale warning in nearly a decade, leading to widespread disruptions in public transport and urban infrastructure—an event underscoring the intensifying challenges posed by climate change to urban resilience and operational continuity.

The United Nations’ 2030 Agenda for Sustainable Development emphasizes building inclusive, safe, resilient, and sustainable cities and human settlements. As a climate-sensitive and climate-impacted region, China has actively pursued strategies to strengthen urban adaptation capacity since the 18th National Congress of the Communist Party of China (CPC), specifically proposing the development of climate-resilient cities. In 2017, the National Development and Reform Commission (NDRC) and the Ministry of Housing and Urban-Rural Development (MOHURD) selected 28 regions as pilot areas for climate-resilient city construction. In 2023, this initiative was further institutionalized with the issuance of the “Notice on Deepening the Climate-Resilient City Construction Pilot Program”, jointly released by eight ministries. The policy aims to scale pilot experiences to all prefecture-level cities by 2035, highlighting the national commitment to adaptive urban governance and sustainable transformation [2].

Urban systems are complex, adaptive social–ecological, increasingly exposed to climate-induced stressors [3]. Although functionally robust, cities are inherently vulnerable to systemic shocks. Enhancing urban resilience has thus emerged as a key policy priority to ensure climate adaptation, public safety, and ecological sustainability [4]. Within this domain, ecological resilience represents a critical subdimension, reflecting an ecosystem’s ability to absorb disturbances, reorganize, and sustain core functions. Originally a biophysical concept [5], resilience has since been extended to urban contexts, becoming integral to national environmental security strategies [6]. However, rapid urbanization has led to ecosystem degradation, with green space fragmentation, air pollution, and biodiversity loss reducing cities’ adaptive capacities. Between 2010 and 2024, China’s average urban green coverage rose only 4.4%, while per capita CO₂ emissions reached 9 tons, substantially higher than the global average of 5.2 tons1, underscoring the ecological challenges of urban expansion [7,8].

Recent academic research on ecological resilience has increasingly focused on three key areas: measurement methods, influencing mechanisms, and spatiotemporal evolution. Regarding measurement methods, comprehensive indicator system approaches are predominant. Commonly used models include the Resistance–Adaptability–Resilience (RAR) model, the Driving forces–Pressure–State–Impact–Response (DPSIR) model, the Risk–Connectivity–Potential (RCP) model, the Pressure–State–Response (PSR) model, and the Scale–Density–Morphology (SDM) model [9,10,11,12]. These indices are typically calculated using methods like the entropy weight method, coefficient of variation method, and CRITIC method. Zhao et al. [13] constructed an evaluation index system for urban ecological resilience in 35 major Chinese cities based on the DPSIR model, quantified the urban ecological resilience index, and analyzed urban ecosystem sustainability by constructing ecological networks. Some scholars also employ the ecological footprint method to measure ecological resilience [14]. For instance, Zhao et al. [15] assessed ecological resilience levels at multiple spatial scales (basin, provincial, and municipal levels) within the Yellow River Basin using an energy-based ecological footprint model. We ultimately adopted the PSR model because it most accurately captures the core mechanism between direct environmental pressure and governance response. It allows us to set aside more macro-level and indirect driving forces, thereby concentrating our analysis on the specific policy interventions triggered by changes in environmental conditions.

Concerning influencing factors, Chu et al. [7] constructed a city ecological resilience index from the dimensions of state, pressure, and response. Using a multi-period difference-in-differences (DID) approach, they empirically analyzed the positive impact of smart city construction on urban ecological resilience, revealing the beneficial role of enhanced information technology levels in improving ecosystem coping capacity. Lu, F. et al., 2024 [16] constructed a PSR model to analyze the social network and influencing factors of urban ecological resilience among prefecture-level cities in the Yellow River Basin. In terms of spatiotemporal evolution, researchers frequently employ coupling coordination models and spatial analysis techniques to identify distribution patterns and dynamic trends. Wang et al. [17] measured the coupling coordination degree between urbanization and ecological resilience in the Pearl River Delta, discovering a concentric pattern centered around cities near the Pearl River Estuary, with coordination increasing towards the periphery. Li et al. [18] constructed a three-dimensional spatial vector model to assess the ecological resilience of urban agglomerations in the Yellow River Basin and explored their spatiotemporal evolution characteristics using GIS technology.

In summary, existing research provides a foundation for exploring the impact of climate-resilient cities on ecological resilience. However, significant gaps remain: (1) studies often focus on regional urban agglomerations, lacking systematic analysis at the national scale; and (2) research content primarily concentrates on measuring ecological resilience and its spatiotemporal evolution, paying insufficient attention to core questions such as how climate-resilient city policies affect ecological resilience and what their underlying mechanisms are. As a key strategy for enhancing urban resilience, what are the actual effects of implementing climate-resilient city policies? Can they genuinely strengthen urban ecological resilience? These questions require in-depth investigation. Particularly, current research commonly adopts a linear “policy–outcome” analytical framework, overlooking the crucial moderating role that governmental behavioral attributes and attention allocation during policy implementation may have on policy effectiveness.

Therefore, this study addresses these gaps by examining the effects of China’s climate-resilient city pilot policy on urban ecological resilience using panel data from 243 cities between 2010 and 2022. Employing a difference-in-differences (DID) approach, the policy is treated as a quasi-natural experiment. A Pressure–State–Response (PSR) framework is adopted to construct a comprehensive resilience index. Importantly, this study introduces “government environmental attention” as a moderating variable, recognizing that the success of policy implementation hinges not only on institutional design but also on the prioritization and allocation of administrative attention at the local level.

The marginal contributions of this study are (1) constructing a “policy instrument–government attention–ecological resilience” theoretical framework, introducing government environmental attention as a core moderating variable into climate adaptation policy research, thereby offering a new theoretical lens to explain the spatiotemporal heterogeneity of policy effects; and (2) establishing a national-scale “Pressure–State–Response” full-chain ecological resilience evaluation index system covering 243 prefecture-level cities, systematically revealing the spatially heterogeneous impact patterns of climate-resilient city policies on ecological resilience, thereby laying a methodological foundation for designing differentiated governance strategies.

2. Policy Background and Theoretical Analysis

2.1. Policy Background

China’s climate-resilient city construction represents a systematic institutional response to escalating climate risks, integrating ecological adaptation into national urban governance. The evolution of this policy framework is illustrated in Figure 1. Since the 1990s, following the release of the China Agenda 21, the concept of “climate change adaptation” has gradually entered national policy discourse. The 2013 National Climate Change Adaptation Strategy2 marked a pivotal step, embedding climate risk assessment into urban planning and calling for improved drainage systems and ecological corridors to enhance urban resilience.

From 2015 to 2016, the Ministry of Housing and Urban-Rural Development (MOHURD) initiated 30 sponge city pilots, targeting green infrastructure to mitigate pluvial flood risks. Concurrently, the National Development and Reform Commission (NDRC) and MOHURD released the Work Plan for Climate-Resilient City Pilots, adopting a “pilot-first, tailored implementation” principle and launching applications for 30 cities. In 2017, 28 pilot cities were formally approved, spanning diverse climatic and economic zones. The inclusion of “resilient cities” in the 14th Five-Year Plan further institutionalized the agenda, promoting the integration of climate adaptation with modernization of urban governance. The 2022 National Climate Change Adaptation Strategy 2035, issued by the Ministry of Ecology and Environment (MEE) and 16 other ministries, further emphasized the role of cities as critical nodes in national climate adaptation efforts [19]. In 2023, eight departments including the Ministry of Ecology and Environment jointly issued the “Notice on Deepening the Climate-Resilient City Construction Pilot Program”, and in 2024, they announced the list of 39 pilot projects for deepening climate-resilient city construction across the country.

2.2. Theoretical Framework

The complex adaptive systems (CAS) theory emphasizes that open systems possess self-organizing, self-adaptive, and emergent capabilities, enabling them to maintain relatively stable internal structures and functions despite external disturbances [20]. Urban systems, characterized by high heterogeneity and nonlinearity, are increasingly vulnerable to the compounding impacts of climate change. The main thread running through the theoretical logic of complex adaptive systems is adaptation. Adaptation is not only a core keyword of the CAS theory but also a fundamental feature of city climate resilience governance. The term adaptation has thus become the convergence point between the CAS theory and city climate-resilience governance. As a form of institutional adaptation, the construction of climate-resilient cities aims to enhance the systemic resilience of urban ecosystems through coordinated policy instruments and technological interventions. These measures work across the “Pressure–State–Response” (PSR) dimensions to improve the capacity of urban systems to sustain dynamic equilibrium under increasing stress. This approach marks a shift from traditional reactive disaster response models towards a more proactive paradigm of complex system governance and dynamic resilience [21]. The main thread running through the theoretical logic of complex adaptive systems is adaptation. Adaptation is not only a core keyword of the CAS theory but also a fundamental feature of climate-resilience governance. The term adaptation has thus become the convergence point between the CAS theory and climate-resilience governance.

However, the effectiveness of institutional interventions is highly dependent on the attention allocation of governing actors. Based on Ocasio’s [22] attention-based view (ABV), government attention is a finite and scarce governance resource that determines the extent to which problems are “seen”, resources are “allocated”, and policies are “implemented” [23]. In this study, local government attention to environmental affairs is conceptualized as a moderating mechanism for policy implementation, which plays a decisive role in the effectiveness and sustainability of policies [24,25]. That is, higher levels of attention are more likely to manifest as stronger policy commitments, optimized resource allocation, and stricter monitoring and feedback during the advancement of climate-resilient city policies, thereby amplifying the marginal effect of policy instruments on ecological resilience [26].

To systematically assess the structure of ecological resilience, this study adopts the PSR model as its core analytical framework. Initially developed by the United Nations Environment Programme (UNEP) and the Organization for Economic Co-operation and Development (OECD), the PSR model has been widely employed in environmental management research to examine the dynamic interplay among environmental pressures, system states and institutional responses [27].

Specifically, in pressure dimension, extreme weather events caused by climate change and chronic risks form a compound impact, forcing cities to reconstruct their risk perception frameworks. In the state dimension, the vulnerability of the urban ecological base and the sensitivity of population and economic agglomeration form a cumulative effect. Dynamic restoration necessitates interventions through both natural and artificial means [28].

In the response dimension, policy implementation fosters a multi-tiered policy framework combining institutional constraints, market incentives, and social mobilization. This is achieved by establishing climate adaptation planning guidelines, innovating climate bond financing mechanisms, and promoting community participatory governance, systematically enhancing the city’s resilience capacity to cope with extreme climate events [29]. Based on this, this study constructs an urban ecological resilience evaluation indicator system using the PSR model. Among these, pressure resilience refers to the adverse impacts of human socio-economic activities on the ecological environment; state resilience refers to the current status of the urban environment and ecosystems and response resilience refers to the measures taken by humans in response to environmental issues [30]. The specific research framework is shown in Figure 2.

2.2.1. Enhancing Effect of Climate-Resilient City Construction on Ecological Resilience

One of the primary objectives of constructing climate-resilient cities is to enhance the service functions of urban ecosystems and improve urban ecological resilience. The state has continuously explored and promoted various urban development models, including sponge cities, climate-resilient cities, low-carbon cities, and waste-free cities. Existing scholarship has empirically validated the positive impact of these urban pilot policies on urban resilience. Zhang et al. [31] demonstrated that the sponge city construction pilot policy significantly improved urban water ecological environment quality. The climate-resilient city pilot policy effectively enhances urban resilience levels through industrial structure upgrading and increased public environmental awareness. Furthermore, its synergistic interaction with similar urban pilot policies yields an even more pronounced improvement in ecological resilience [32].

Specifically, constructing an urban ecological security barrier is a key component of climate-resilient city development. Implementing urban ecological restoration projects enriches urban biodiversity, mitigates the urban heat island effect, and maintains the stability of ecological chains, thereby enhancing ecological resilience [33]. Simultaneously, climate-resilient cities advocate for the construction of resilient infrastructure. For instance, developing infrastructure focused on the urban water cycle effectively ensures stable urban water supply, strengthening urban ecological resilience from the water resource dimension [34].

Based on this analysis, this study proposes the following:

Hypothesis 1 (H1).

Climate-resilient city construction has a significant promoting effect on urban ecological resilience.

2.2.2. Heterogeneous Effects of Climate-Resilient City Construction on Ecological Resilience

China is a vast country, and there are significant differences between regions in terms of climate conditions, environmental pressures, and levels of economic development [35]. Consequently, the impact of the climate-resilient city construction pilot policy on ecological resilience is likely to exhibit heterogeneity. Variations in the financial capital, human resources, and material inputs allocated by governments to climate-resilient city construction influence policy implementation effectiveness [36]. Comparatively, large cities possess distinct advantages in terms of capital investment, governance capacity, and infrastructure layout [37].

At the same time, differences in local government environmental attention can also lead to regional heterogeneity in environmental policy enforcement, resulting in spatially heterogeneous effects of climate-resilient city construction on ecological resilience. Additionally, the initial levels of ecological resilience vary across cities due to differences in ecological resource endowments [38]. For example, resource-based cities, whose primary industries are the extraction and export of natural resources such as minerals and fossil fuels, tend to have slower economic growth and lower environmental governance effectiveness. As a result, most of these cities are still in the early stages of developing resilience [39], and the implementation of pilot policies may have a significant impact on the ecological resilience of such cities.

Therefore, this study proposes the following:

Hypothesis 2 (H2).

The impact of climate-resilient city construction on urban ecological resilience exhibits spatial heterogeneity.

2.2.3. Moderating Effect of Government Environmental Attention

Government environmental attention is closely linked to local government decision-making and governance effectiveness in environmental protection [40]. “Attention” is originally a psychological concept, which Jones [41] introduced into the field of policy research, proposing that bounded rationality and attention shifting are important causes of policy stability or sudden change. For a long time, ecological and environmental protection has been regarded as a task with weak incentives, and local governments have long paid little attention to ecological and environmental protection [42]. However, in recent years, central macro policies, supervision policies, and cadre evaluation systems have guided local governments to allocate more attention to ecological protection, aiming to balance economic development with ecological environmental conservation [6,43].

When governments direct high levels of attention to environmental protection, “political momentum” can effectively drive resource concentration and policy implementation [44]. This manifests in actions such as formulating and enforcing stricter environmental regulations, adjusting local industrial and fiscal subsidy policies, and optimizing resource allocation [45,46]. Simultaneously, the higher-level government’s focus on the ecological environment has compelled relevant personnel to actively take measures to ensure that performance targets are met and even exceeded [47].

Based on this analysis, this study proposes the following:

Hypothesis 3 (H3).

Government environmental attention has a positive moderating effect on the relationship between climate-resilient city construction and urban ecological resilience.

3. Research Design and Data Sources

3.1. Indicator System Construction

To investigate the impact of climate-resilient city in response to stress or emergencies, this study constructs an urban ecological resilience index as the core explained variable.

This index is a key indicator used to assess a city’s comprehensive ability to limit pollution, maintain ecological conditions, and enhance environmental patent capabilities when faced with ecological and environmental pressures or emergencies [7,48]. Building on existing research [31], environmental performance evaluation methodologies, and urban socio-economic characteristics, the index (see

Table 1. Construction of resilience index system for urban ecological environment.

Tier-1 Indicator	Tier-2 Indicator	Tier-3 Indicator	Unit	Attribute
Urban Ecological Resilience Index (UERI)	Urban Ecological Pressure Resilience Index (UEPI)	Per Capita Industrial Wastewater Discharge	ton per people	−
		Per Capita Industrial Sulfur Dioxide (SO2) Emissions	ton per people	−
		Per Capita Industrial Soot Emissions	ton per people	−
		Per Capita Carbon Emissions	ton per people	−
		Annual Average Concentration of Inhalable Particles (PM10)	μg/m3	−
	Urban Ecological State Resilience Index (UESI)	Per Capita Water Resource Availability	m3/people	+
		Green Coverage Rate of Built-up Areas	%	+
		Per Capita Park Green Space Area (Municipal Districts)	hectares per 10,000 people	+
		Per Capita Built-up Area (Municipal Districts)	km2 per 10,000 people	+
	Urban Ecological Response Resilience Index (UERI)	Industrial Sulfur Dioxide (SO2) Removal Rate	%	+
		Industrial Soot Removal Rate	%	+
		Municipal Solid Waste Harmless Treatment Rate	%	+
		Centralized Sewage Treatment Rate	%	+
		Industrial Solid Waste Comprehensive Utilization Rate	%	+

) is decomposed into three dimensions—pressure resilience, state resilience, and response resilience [7]. Given that various indicators may have a positive or negative impact on the overall resilience index of a city, this study draws on the method proposed by Zhou et al. [49] to normalize all indicator values and uses the entropy weight method for weighting, enabling comprehensive calculation of each city’s ecological resilience index.

The state resilience index reflects the current quality of the urban ecological environment, where its magnitude directly indicates ecosystem health and the city’s foundational capacity to sustain sound ecological conditions. Consequently, this study designates it as a secondary indicator of urban ecological resilience. To accurately characterize state resilience, it is further broken down into per capita water resource availability, built-up area green coverage rate, per capita park green space area in municipal districts, and per capita built-up area [31,50].

The pressure resilience index evaluates ecological and environmental stresses faced by cities [7,31], with its value revealing a city’s resistance capacity to maintain ecological stability under pressure. Higher values denote superior ability to mitigate environmental stress impacts. Selected tertiary indicators include per capita industrial wastewater discharge, per capita industrial sulfur dioxide emissions, per capita industrial soot emissions, per capita carbon emissions, and annual average concentration of inhalable particulate matter [31,50].

The response resilience index captures the speed and efficacy of urban actions addressing ecological issues, directly influencing problem-solving efficiency and serving as a critical measure of environmental governance proficiency. It encompasses pollution control, environmental restoration, and emergency coordination capabilities, quantified through tertiary indicators: industrial sulfur dioxide removal rate, industrial soot removal rate, municipal solid waste harmless treatment rate, centralized sewage treatment rate, and industrial solid waste comprehensive utilization rate [7,50].

3.2. Entropy Weight Method

This study employs the entropy weight method to normalize data and assign weights to indicators. The specific computational procedure is as follows:

Step 1:

Indicator Data Normalization

Given the directional attributes (positive/negative) of indicators, differentiated standardization methods are applied:

$$\text{Positive} \text{indicator}: Y_{it}={\displaystyle \frac{X_{it}-min{X}_{it}}{max{X}_{it}-min{X}_{it}}}$$

(1)

$$\text{Negative} \text{indicator}: Y_{it}={\displaystyle \frac{{maxX}_{it}-X_{it}}{max{X}_{it}-min{X}_{it}}}$$

(2)

where X_it indicates the original value of indicator i in year t; maxX_it and minX_it indicate the maximum and minimum values of indicator i; Y_it indicates the normalized value of indicator i in year t.

Step 2:

Dimensionless Treatment

Proportional transformation eliminates dimensional effects by calculating relative proportions:

$${\omega }_{it}={\displaystyle \frac{Y_{it}}{\sum _{i=1}^m{Y}_{it}}}$$

(3)

where m indicates the total number of years

Step 3:

Information Entropy Calculation

Information entropy is computed based on information theory principles:

$$e_i=-{\displaystyle \frac{1}{lnm}}\sum _{t=1}^m{\omega }_{it} \ast ln{\omega }_{it}$$

(4)

Step 4:

Weight Assignment Mechanism

Indicator weights are determined by entropy values:

$$W_i={\displaystyle \raisebox{1ex}{$d_i$}\!\left/ \!\raisebox{-1ex}{$\sum _{i=1}^n{d}_i$}\right.}$$

(5)

3.3. Model Setting

In this study, the explanatory variable is urban ecological resilience (UERI). The explanatory variable is a double difference variable (did), indicating whether climate-resilient city pilot projects have been launched, and the control variables are four indicators: economic development level (ECO), industrial structure (STR), urbanization level (URB), and openness to foreign investment (FDI) [7,43,51].

Economic development level (ECO) is measured by regional GDP per capita. Economic development may encourage governments and businesses to invest more funds in environmental protection and governance, promote green technology and environmental innovation, and enhance urban ecological resilience. However, it may also lead to resource over exploitation and environmental pollution, potentially undermining ecological resilience.

Industrial structure (STR) is quantified as the ratio of value-added from secondary industry to tertiary industry. A higher proportion of secondary industry typically implies greater resource consumption and environmental degradation due to its high resource intensity. However, the tertiary industry typically has lower pollution intensity and resource consumption. Therefore, as the industrial structure undergoes optimization and upgrading, particularly during the transition to low-pollution, low-energy-consuming service industries, the ecological resilience of cities will be enhanced.

Urbanization level (URB) is represented by the urbanization rate. Increasing urbanization often correlates with population growth, potentially elevating resource consumption and pollutant emissions. This process may also reduce green space coverage and biodiversity, posing significant challenges to ecological resilience.

Openness to foreign investment (FDI) is measured by foreign direct investment. Inflows of foreign capital may introduce advanced production technologies and efficient management practices, positively contributing to green development and ecological conservation, thereby strengthening urban ecological resilience.

Based on this analytical framework, we construct the following difference-in-differences model to evaluate the policy effect of climate-resilient city development:

$${UERI}_{it}={\alpha }_i+{\alpha }_1{did}_{it}+\sum {\mu }_j{Control}_{jit}+{\epsilon }_{it}$$

(6)

where UERI_it indicates the urban ecological resilience of city i in year t. did_it indicates the core explanatory variable (1 if city i implemented the pilot policy in year t; 0 if otherwise). Control_jit indicates vector of control variables (j denotes variable index). ε_it indicates the error term. $\alpha$_i indicates the intercept term.

3.4. Moderating Effect Model

Government environmental attention exerts a positive moderating effect on the relationship between climate-resilient city development and urban ecological resilience. Enhanced governmental attention can strengthen climate-resilient city initiatives and amplify ecological resilience through policy optimization, resource allocation, and heighten public awareness. However, excessive focus on climate-resilient city construction may divert attention away from other important areas, thereby affecting the overall development of society. This diversion of attention may partially undermine the sustainability of urban ecological-resilience enhancement.

Drawing on methodological approaches from He et al. [50] and Jiang et al. [52], this study constructs a moderating effect model with the natural logarithm of government environmental attention (att) as the moderating variable. Government environmental attention is indicated by the total frequency of environmental protection terms appearing in government reports. The frequency data was obtained from municipal government work reports using Python 3.8.10 crawler technology. The model specification is as follows:

$${UERI}_{it}={\gamma }_0+\theta {did}_{it}+\rho {lnatt}_{it}+\sigma {did}_{it}\times ln{att}_{it}+\sum {\mu }_j{Control}_{jit}+{\vartheta }_{it}+{\epsilon }_{it}$$

(7)

3.5. Data Sources and Descriptive Statistics

This study conducts an empirical analysis of the impact of the first batch of climate-resilient city pilots (2017) on urban ecological resilience. Constrained by policy implementation cycles and data availability, panel data from 243 prefecture-level cities in China spanning 2010–2022 were selected. The analysis focuses exclusively on the first batch of pilot cities, since the second batch initiated in 2024 had not manifested observable policy effects in data prior to 2022. In the sample selection process, cities with data missing rates exceeding 20% were excluded (certain cities in the Xinjiang Uygur Autonomous Region were excluded due to incomplete statistical information3). Ultimately, 25 climate-resilient cities with complete data were selected from the first batch of pilot cities.

The data are sourced from the China City Statistical Yearbook, China Urban Construction Statistical Yearbook, China Energy Statistical Yearbook, China Environmental Statistical Yearbook, and provincial and municipal statistical yearbooks from 2010 to 2022. For missing data in certain years, simple interpolation and extrapolation methods were used to fill in the gaps to ensure the continuity and comparability of the data.

The specific descriptive statistical results are shown in

Table 2. Variable definition and descriptive statistics.

Variable Type	Variable Name	Definition	Mean	Std. Dev.	Min	Max
Explained Var.	Urban Ecological Resilience	Calculated using entropy weight method	0.7220	0.0298	0.5362	0.8256
Core Exp. Var.	DID	Dummy variable (1 = climate-resilient city pilot; 0 = otherwise)	0.0495	0.2169	0.0000	1.0000
Moderating Var.	Government Environmental Attention	Natural logarithm of total frequency of environmental keywords in gov. reports	8.1436	2.2616	−2.3025	13.0819
Control Variables	Economic Development Level	Regional GDP per capita (10,000 yuan/person)	5.8153	3.5186	0.5862	25.6908
	Industrial Structure	Ratio: Secondary industry VA / Tertiary industry VA (%)	1.0700	0.5428	0.1156	7.8494
	Urbanization Level	Urbanization rate (%)	0.5765	0.1525	0.3390	1.1779
	Openness to Foreign	Foreign direct investment (million USD)	1117	3201	0.0000	1,000,242.7000
	Investment		0.1160	0.9860

. The mean value of the core explanatory variable, urban ecological resilience, is 0.7220, indicating a relatively high average level of urban ecological resilience. The standard deviation is 0.0298, with data distribution relatively concentrated. The range of values is between 0.5362 and 0.8256, suggesting that there are certain differences in urban ecological resilience across different regions, potentially indicating spatial heterogeneity; The mean value of the moderating variable, government environmental attention, is 5684.5550, with significant regional variations. These disparities imply differential policy implementation effects of climate-resilient city initiatives across localities.

4. Results

4.1. Base Regression

This study employs a difference-in-differences (DID) model to analyze the impact of climate-resilient city pilot policy on urban ecological resilience. The baseline regression results are presented in

Table 3. Benchmark regression results.

Variable	Model 1	Model 2
did	0.0257 *** (8.9982)	0.0074 *** (2.9216)
GDP per capita		0.0035 *** (8.2276)
Sec/Ter Industry VA		−0.0479 *** (−15.9183)
Urbanization rate		0.0940 *** (5.7656)
FDI		0.0000 * (3.5690)
Constant	0.7207 *** (1.5 × 103)	0.6912 *** (69.2575)
Observations	2870	2870
R2	−0.0646	0.2114

Robust standard errors in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

. As shown by the results, regardless of whether control variables are included in the analysis, the coefficient for climate-resilient city development in enhancing urban ecological resilience is significantly positive (at the 1% significance level), with a coefficient of 0.0074. This indicates that the experimental group experienced an overall increase of approximately 0.74% in urban ecological resilience levels compared with the control group before and after policy implementation. The regression results confirm that establishing climate-resilient cities can significantly enhance urban ecological resilience by effectively improving a city’s ability to withstand and recover from extreme climate events, thereby ensuring that cities can maintain functional stability, ecological health, and sustainable development in the face of climate change challenges.

Control variable analysis reveals that both regional per capita GDP and urbanization rate exhibit significantly positive effects on urban ecological resilience at the 1% significance level, indicating that improvements in economic development levels enhance urban ecological resilience by promoting capital inflows into environmental protection departments. Meanwhile, increases in urbanization rates can reduce the impact of external disturbances on urban ecosystems through scientifically sound urban planning and refined management, thereby enhancing urban ecological resilience. At the 1% significance level, the added value of the secondary industry and the added value of the tertiary industry have a significant negative impact on urban ecological resilience, indicating that the reduction in the secondary industry and the increase in the tertiary industry can reduce environmental pollution through industrial transformation and enhance urban ecological resilience. At the 10% significance level, the amount of foreign direct investment has a significant but relatively small impact on urban ecological resilience.

4.2. Parallel Trend Test

To validate whether the baseline regression satisfies the parallel trends assumption, we estimate a dynamic difference-in-differences model following Beck et al. [53]. Using 2017 as the baseline, the implementation year of climate-resilient city pilots, we examine policy effects from 2010 to 2022. To avoid multicollinearity, data from the previous period (2016) was excluded from the regression analysis.

The results are shown in Figure 3. During the period from 2010 to 2017, the estimated coefficients were not significant and remained at relatively low levels. This indicates that prior to the implementation of the climate-resilient city policy, there were no significant differences between the treatment group and the control group, satisfying the assumption of parallel trends. That is, if the policy had not been implemented, the ecological resilience levels of cities in the treatment group and control group would have followed similar trends. The coefficients become significantly positive starting in 2020, confirming that climate-resilient city development exerts a substantial positive impact on urban ecological resilience. The delayed emergence of policy effects suggests implementation lags, likely attributable to the time required for policy formulation, execution, and tangible ecosystem responses.

4.3. PSM-DID

When conducting research using the difference-in-differences (DID) method, inherent differences between cities prevent random selection of pilot cities, making it challenging to ensure that the treatment group and control group have completely consistent individual characteristics prior to policy implementation. Given that the DID method may introduce selection bias due to this non-randomness, we employ propensity score matching (PSM) to match cities across groups, followed by regression analysis using the difference method on the matched samples [54].

The specific procedure involves predicting each city’s probability of being selected as a pilot using Logit regression based on urban characteristics, Control_i, with the model defined as follows:

$$p_i=p\left({Control}_i\right)=prob[{treat}_i=1|{Control}_i]$$

(8)

where p_i is the conditional probability of city i being selected as a pilot city; Control_i is the set of city characteristic variables; and treat_i is the pilot policy dummy variable (1 if pilot city; 0 if otherwise).

The analysis results are presented in

Table 4. PSM-DID regression results.

Variable	PSM
DID	0.0129 *** (3.8444)
GDP per capita	0.0014 (1.1136)
Sec/Ter Industry VA	−0.0422 *** (−5.3159)
Urbanization rate	0.0706 (1.4834)
FDI	0.0000 (0.7167)
Constant	0.7086 *** (25.2941)
Observations	731
R2	0.216

Robust standard errors in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

. It demonstrates that after propensity score matching, the DID estimated coefficients, signs, and significance levels are consistent with the baseline regression results. This indicates that the regression results obtained using the PSM-DID method further validate and support the conclusions of the baseline regression.

4.4. Placebo Test

To address potential standard error bias [55], we conduct 500 random samplings across all 243 prefecture-level cities and perform benchmark regression analysis. As shown in Figure 4, the kernel density distribution of estimated coefficients clusters around zero, exhibiting normal distribution characteristics, consistent with the expectations of placebo test. These results confirm that our findings pass the placebo test, indicating that the observed effects of climate-resilient city policies are not attributable to unobserved confounding factors.

4.5. Heterogeneity Analysis

The preliminary descriptive statistics revealed differences in ecological resilience across cities. Given this, the study conducted a detailed heterogeneity analysis to explore the impact mechanism of establishing climate-resilient cities on urban ecological resilience. Results are presented in

Table 5. Heterogeneity analysis results.

Variable	Model 1 National	Model 2 Eastern China	Model 3 Central and Western China	Model 4 Northern China	Model 5 Southern China
DID	0.0074 *** (2.9216)	−0.0001 (−0.0268)	0.0083 *** (2.6025)	0.0200 *** (4.8938)	−0.0014 (−0.4391)
GDP per capita	0.0035 *** (8.2276)	0.0035 *** (7.2474)	0.0035 *** (4.6843)	0.0037 *** (4.8461)	0.0037 *** (6.6000)
Sec/Ter Industry VA	−0.0479 *** (−15.9183)	0.0480 *** (−12.5434)	−0.0481 *** (−10.7408)	0.0483 *** (−10.5218)	−0.0461 *** (−11.4792)
Urbanization rate	0.0940 *** (5.7656)	0.0671 *** (3.8041)	0.1253 *** (4.0539)	0.1000 *** (4.3611)	0.0894 *** (3.6864)
FDI	0.0000 * (3.5690)	−0.0000 ** (−2.1110)	0.0000 *** (5.0668)	0.0000 (1.2728)	0.0000 *** (3.7275)
Constant	0.6912 *** (69.2575)	0.7129 *** (59.0808)	0.6737 *** (40.2640)	0.6843 *** (45.2470)	0.6915 *** (50.2050)
Observations	2870	1318	1318	1059	1806
R2	0.2114	0.2452	0.2452	0.2179	0.2171

Notes: Robust standard errors in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

.

According to Model 1, the coefficient of the impact of implementing climate-resilient city policies nationwide on urban ecological resilience is significantly positive at the 1% level, confirming its overall effectiveness. Further analysis in Models 2 and 3 reveals a distinct regional disparity. While the policy impact remains statistically insignificant in Eastern China, it shows significantly positive coefficients at the 1% level in the central and western regions. Based on the results of Models 4 and 5, the coefficient for the northern region is significantly positive, while that for the southern region is not significant. Collectively, these findings establish that climate-resilient cities pilot policies may have a more significant impact on cities with originally poor environmental conditions.

4.6. Moderating Effect Test

Based on the moderation effect model constructed in the preceding section, the results are shown in

Table 6. Adjustment effect results.

Variable	Base Regression	Adjustment Effect Model
DID	0.0074 *** (2.9216)	−0.0268 * (−1.9453)
GDP per capita	0.0035 *** (8.2276)	0.0031 *** (10.5659)
Sec/Ter Industry VA	−0.0479 *** (−15.9183)	−0.0093 *** (−6.6002)
Urbanization rate	0.0940 *** (5.7656)	0.1255 *** (15.0594)
FDI	0.0000 * (3.5690)	−0.0000 (−1.0054)
lnatt		0.0006 (0.3392)
lnatt × DID		0.0028 * (1.7446)
Constant	0.6912 *** (69.2575)	0.6367 *** (39.8085)
Observations	2870	2750
R2	0.2114	0.2609

Notes: Robust standard errors in parentheses. *** p < 0.01, ** p < 0.05, * p < 0.1.

. The interaction term between the logarithm of environmental attention and the difference-in-differences variable is significantly positive, indicating that the government’s environmental attention can exert a positive promotional effect on the implementation of policies to establish climate-resilient cities. Referring to the baseline regression results, establishing climate-resilient cities can significantly promote urban ecological resilience. Based on the research results of Rao et al. [56], government environmental attention plays a key positive moderating role in the complex transmission mechanism of establishing climate-resilient cities to enhance urban ecological resilience.

When local governments demonstrate greater concern for environmental protection, it can lead to a greater allocation of public resources toward environmental protection. This accelerates the implementation of climate-resilient city policies while optimizing resource allocation also effectively enhances urban ecological resilience. This demonstrates the positive link between policy execution efficiency and local governments’ environmental awareness.

5. Discussion

5.1. Implication of Climate-Resilient City Pilot Policy on Urban Ecological Resilience

5.1.1. The Impact of Climate-Resilient City Development on Urban Ecological Resilience

Through this study, it was found that building climate-resilient cities can significantly enhance urban ecological resilience. By comprehensively integrating urban development planning with climate change adaptation strategies, it is possible to achieve the comprehensive optimization and strengthening of the urban ecological environment. Thereby enhancing the city’s ability to withstand extreme weather events such as floods, droughts, and heatwaves, as well as its adaptability to long-term climate change challenges such as rising temperatures and changes in precipitation patterns [57,58]. Jiujiang City in Jiangxi Province was long plagued by urban waterlogging. In the past, heavy rainfall often resulted in water accumulation exceeding 50 cm in depth. Due to the advancement of climate-resilient city pilot policy, they upgraded their drainage infrastructure. The city has successfully prevented waterlogging since the beginning of last flood season. Currently, over 80% of the urban area has significantly enhanced its flood resilience, meeting drainage standards capable of handling once-in-30-year rainstorms. Enhanced resilience and recovery capabilities not only enable cities to swiftly restore functions and minimize losses during disasters but also maintain the balance and health of urban ecosystems in the face of long-term climate change. This significantly mitigates the impact of extreme weather events and long-term climate change on the economy [59].

However, this promotional effect exhibits significant spatial heterogeneity, with the policy yielding more pronounced results in central and western regions as well as resource-based cities, such as Xi’an, Shaanxi Province and Taiyuan, Shanxi Province. By contrast, economically advanced eastern cities like Shenzhen and Hangzhou, which already benefit from stronger environmental infrastructure and higher baseline resilience, show relatively limited marginal gains from the policy [60]. This divergence likely stems from Eastern China’s pre-existing environmental advantages, which dilute the observable impact of new policy interventions. Conversely, implementation in less-developed central and western regions substantially enhances ecological resilience and environmental conditions. This provides an important basis for further developing differentiated, precise, and targeted environmental protection policies.

5.1.2. The Moderating Role of Government Environmental Protection Attention

Government attention, as a scarce governance resource, guides the direction of policy formulation and implementation [22]. When the government focuses more attention on environmental protection, it triggers a series of positive chain reactions. At the policy implementation level, higher environmental attention prompts the government to re-examine the logic of administrative resource allocation based on the theory of resource optimization in public management [61]. Governments allocate more human, material, and financial resources to environmental protection departments to ensure the precise and comprehensive implementation of environmental policies.

Guided by ecosystem resilience theory, this concentrated resource investment coupled with efficient policy enforcement progressively strengthens urban environmental systems. The resulting ecological improvements simultaneously boost cities’ resistance to climate disruptions while amplifying their recovery and adaptive capacities [62,63]. Consequently, urban ecosystems develop reinforced structural and functional stability when confronting extreme climate events, ultimately achieving sustainable development and genuine climate adaptation.

5.2. Future Directions for Climate-Resilient City Development

Strengthen integrated, multi-level governance mechanisms. Local governments should establish cross-departmental coordination frameworks to enhance policy coherence across sectors [64]. Implementation plans should be phased and adapted to regional ecological characteristics and developmental conditions. Moreover, integrating climate-resilient city construction with existing initiatives such as sponge cities, low-carbon cities, and waste-free cities will foster synergy and minimize policy fragmentation [65].

Promote spatially differentiated and risk-based adaptation strategies. Policymakers should tailor climate-resilient city initiatives to the specific risk profiles and environmental vulnerabilities of different regions. Eastern seaboards should prioritize coastal defense infrastructure [66]. Inland regions require drought-resilient systems to address aridification. Tailored to Southern China’s distinct climatic challenges, an integrated adaptation system synergistically combats compound humidity–heat–rainfall threats. As for Northern China, it must focus on extreme cold-weather resilience and air quality enhancement. Additional support should be directed toward ecologically fragile areas in Central, Western, and Southern China through targeted financial transfers and resource allocation to ensure balanced development and resilience co-benefits [67].

Embed climate adaptation considerations into regulatory, fiscal, and planning frameworks. Local governments should incorporate climate resilience indicators into urban planning, land-use approvals, and public investment decisions. Economic instruments, including fiscal subsidies, tax incentives and green finance tools, should be leveraged to incentivize private sector engagement in green buildings, renewable energy, and energy-efficient technologies [68]. Finally, enhancing public awareness and multi-stakeholder participation will foster a broader societal foundation for climate adaptation [69].

5.3. Limitations and Prospects

This study primarily focuses on the direct impact and moderating role of climate-resilient city pilot policies on urban ecological resilience but lacks in-depth exploration of the dynamic assessment and feedback mechanisms of policy implementation effects. Furthermore, due to data availability constraints, this study is based solely on the first batch of pilot cities. Future research could expand the dataset to include more recent batches of pilot cities, which allow for a longitudinal comparison and a more robust assessment of the sustainability and temporal dynamics of policy impacts. Additionally, it would also be useful to compare China’s programs with similar efforts in other countries. Such international comparisons could help identify strategies that work across different contexts and improve global approaches to urban climate adaptation.

6. Conclusions

Amid escalating global climate risks and increasing ecological vulnerability in urban areas, the promotion of climate-resilient cities has become a pivotal institutional strategy to enhance urban sustainability and adaptive capacity. Drawing on panel data from 243 prefecture-level cities in China from 2010 to 2022, this study employed a difference-in-differences (DID) model to empirically assess the impact of climate-resilient city construction on urban ecological resilience, while further exploring the moderating role of government environmental attention and spatial heterogeneity. Empirical analysis reveals that the ecological resilience level of pilot cities increased by an average of approximately 0.74% following the policy’s implementation. The policy effects became statistically significant only starting in 2020, indicating a certain time lag between policy execution and the emergence of ecological benefits. Furthermore, the policy impact exhibited significant regional heterogeneity, with more substantial improvements observed in areas possessing a weaker ecological foundation. Further analysis demonstrates that local governments’ attention to environmental protection played a crucial positive moderating role in this process, amplifying the positive influence of the pilot policy on ecological resilience.