Revealing the Spatial Effects of New-Type Urbanization on Urban Ecological Resilience: Evidence from 281 Prefecture-Level Cities in China

Abstract

Enhancing urban ecological resilience (UER) is essential for achieving sustainable urban development, as it fosters balanced urbanization while ensuring long-term ecosystem stability. New-type urbanization (NU) plays a pivotal role in sustaining urban sustainable development; however, the mechanisms through which NU affects UER remain insufficiently understood. This study seeks to bridge this knowledge gap by analyzing panel data from 281 prefecture-level cities in China spanning the period from 2000 to 2022. Composite indices for NU and UER are constructed using the entropy-weighted TOPSIS method. The relationship between NU and UER is empirically examined through fixed-effects models, mediation analysis, and a spatial Durbin model. The findings reveal a general upward trend in both NU and UER over time, albeit with some fluctuations. NU significantly enhances UER through direct effects, indirect pathways, and spatial spillovers. The magnitude and nature of this impact vary across geographic regions and resource endowments. Resource-based cities, in particular, demonstrate a stronger capacity to optimize land resource allocation, improve land use structures, and reduce environmental pollution—thus contributing more effectively to UER enhancement. Furthermore, while NU has a positive effect on UER across all regions, the impact is more pronounced in central and western cities, where major national development strategies—such as the Western Development Strategy and the Rise of Central China Plan—are actively implemented. Among them, cities in the central region with higher levels of urbanization experience more substantial benefits from NU compared to those in the western region, where urbanization is progressing more gradually. This study contributes to a deeper understanding of the spatial effects of NU on UER and offers valuable policy implications for enhancing ecological resilience through sustainable urbanization pathways.

1. Introduction

Since the implementation of the reform and opening-up policy, China’s urbanization rate has increased rapidly, which has made significant contributions to the socio-economic development [1,2,3]. However, urbanization is rooted in the ecological environment, and the growing demand for ecosystem services has driven its continuous expansion, posing inevitable challenges and pressures on urban ecosystems [4,5,6]. On one hand, natural disasters resulting from ecological laws have adversely affected urban ecological systems; on the other hand, the extensive and unsustainable development practices of humans have led to ecological imbalances, undermining the environment for resilient urban development [7,8,9]. This indicates a coupling and interactive relationship between the process of urbanization and ecological evolution. The introduction of the National New-type Urbanization Plan (2014–2020) marked a critical milestone in China’s transition to a new stage of urbanization. New-type urbanization (NU) has become both a practical consensus and a strategic guide for China’s socio-economic development in the new era [6,10]. Under the framework of sustainable development, exploring the mechanisms by which NU influences urban resilience and identifying the root causes of lagging resilience in certain cities can provide valuable policy insights for promoting the coordinated development of the two.

Resilience refers to a system’s ability to maintain structural or functional stability when it encounters external disturbances or changes [11,12,13]. The concept of resilience was introduced by Lade et al. [14] (2017) in the field of ecology as a measure of an ecosystem’s ability to absorb changes and disturbances while maintaining its overall function. Over time, the scope has broadened and the concept is now widely applied in the assessment of social services, economic development, and natural ecosystems [13,15,16,17]. Urban resilience refers to a city’s ability to quickly recover its original equilibrium state when subjected to internal and external shocks [7,8]. Cities are complex, multifaceted systems, and the ecological resilience, together with social, economic, and cultural elements, constitutes their overall urban resilience. There is a general consensus among researchers that the concept of urban ecological resilience (UER) refers to the speed or ability of urban ecosystems to return to a stable state after being subjected to internal and external perturbations.

With the continuous development of China’s society and economy, the characteristics and extension of urbanization are also evolving [18,19]. In contrast to traditional urbanization, which is characterized by crude and radical development, NU integrates high-quality economic development with improvements in the ecological environment, contributing to the optimal allocation of resources [20,21]. This may, in turn, impact the enhancement of UER. With the clarification of the concept of NU, the research focus has gradually shifted to the following challenges and key factors influencing the process of NU. (i) Placing human development rights at the core is a key factor and a strategic choice for addressing many of the complexities facing China’s NU. (ii) The construction of sustainable special towns is of great significance for promoting the coordinated development of urban and rural areas and accelerating the urbanization process. (iii) Promoting NU should focus on improving the planning of urban–rural spatial integration and development, strengthening the coordination, linkage, and coherence of urban–rural spatial planning.

Furthermore, economic development, industrial structure optimization, technological advancement, urbanization quality, and industrial agglomeration constitute the pivotal driving forces in the urbanization process [2,6,18,22,23]. Previous investigations have predominantly examined the coupling relationship between urbanization and ecological resilience. These studies have consistently demonstrated that urbanization processes inevitably compromise ecological capacity, adversely affecting the demographic, economic, social, and spatial characteristics of urban areas, while simultaneously triggering ecological responses and altering the ecological conditions. However, the relationship between NU and ecological resilience remains poorly understood. Previous studies of ecological resilience have mostly focused on the provincial level, with few investigations at city level. Cities are often clustered based on geographic proximity, economic dependencies, and similar resource endowments, which in turn drive the convergence of regional urban ecological conditions [24,25]. Therefore, more comprehensive studies of UER are needed. The indicator systems used to assess and measure ecological resilience are not sufficiently clear, often emphasizing natural factors such as adaptability and restorative capacity while neglecting the risks posed by human social activities. Existing studies have rarely explored the dynamic spatiotemporal evolution and regional differences in UER. In summary, the existing research has provided a foundation for exploring the relationship between the NU and UER, with the gaps in the previous research being the focus of this study.

The main contributions of this study to sustainable urban development were as follows. (1) A theoretical framework was proposed to deepen the understanding of the mechanisms through which NU affects UER. (2) Based on the concepts and mechanisms of NU and UER, a five-dimensional evaluation index system was constructed for NU (population–economy–space–society–land) and a three-dimensional evaluation index system was constructed for UER (pressure–state–response: PSR), to present the resilience of ecosystems in adapting to NU development. (3) The heterogeneity and spatial spillover effects of the impact of NU on UER were investigated by regressing samples based on nationally defined criteria.

2. Theoretical Analysis and Research Hypotheses

2.1. Characteristics of UER

The term UER refers to the ability of a city to maintain its original state after experiencing either internal or external impacts [8,26]. This study summarized the UER process using the pressure-state-response (PSR) model, which encompasses three major subsystems: pressure resilience, state resilience, and response resilience, based on the process of action [27,28]. Pressure refers to the strength of the city’s response to negative impacts. These pressures on UER primarily stem from waste and pollutants generated by production and daily residential living activities [29]. A smaller pressure impact on a city translates to a better developmental environment and stronger pressure resilience. State refers to the stability of natural resource conditions, ecological elements, and the urban ecological structure [6]. Under the dual influence of pressure and response, the more reasonable the configuration of ecological elements, the stronger the state’s resilience. Response refers to the ability of an ecosystem to adapt to changes following a pressure shock [13]. Sufficient financial support and waste management capacity are essential for an effective response.

The role of UER can be expressed as follows: when a city is impacted, pressure resilience is the first subsystem to perceive the risk and the pressure is transferred to the other subsystems. Under pressure, state resilience is altered, and information about these changes is conveyed to the response resilience subsystem. The response resilience subsystem then adjusts the urban ecosystem to adapt to the new environment under the combined effect of pressure and changes in the state. The outcome of this adjustment feeds back into both the state resilience and pressure resilience subsystems, leading to the formation of a new urban ecosystem equilibrium. Together, these three subsystems enable the urban ecosystem to reach a new equilibrium.

2.2. Influence Mechanisms of the NU on UER

In this study, NU was divided into five dimensions of population, economy, society, space, and land [10,20,24]. The overall NU process inevitably alters the resource endowment and environmental conditions in urban areas, thereby affecting UER. This study analyzed the mechanism through which NU influences UER from the perspectives of direct, indirect, heterogeneous, and spatial spillover effects. A theoretical framework was constructed to analyze how NU enhances UER, as shown in Figure 1.

2.2.1. The Direct Effects of NU on UER

New-type urbanization (NU) directly enhances UER through five key pathways. First, population urbanization enables the effective integration of over 100 million agricultural migrants and permanent residents into cities, improving overall education and labor quality. This transformation of labor into human capital boosts social production capacity and enhances the adaptive capabilities of cities. Second, social urbanization fosters the adoption of ecological civilization and people-centered development principles, promoting green, low-carbon, and efficient resource utilization, which strengthens urban ecological resilience. Third, economic urbanization provides critical financial support for environmental protection, including funding for green infrastructure and centralized pollution control, offering economic security for sustainability initiatives and the development of green technologies. Fourth, spatial urbanization contributes to more rational urban planning and spatial organization, based on ecological capacity and development potential. This reduces disordered expansion and overexploitation, ensuring cities grow within their environmental limits. Fifth, land urbanization strengthens urban–rural integration by enhancing economic, cultural, and informational exchanges, helping to dissolve the urban–rural divide. This promotes coordinated development and creates synergies that support ecological sustainability. Overall, NU fosters a win–win relationship between economic development and environmental protection by optimizing spatial layouts, upgrading industrial structures, and reinforcing ecological governance across all five dimensions. These efforts, grounded in ecological civilization and people-centered values, not only advance urban–rural integration but also significantly boost UER.

Thus, Hypothesis 1 is proposed:

Hypothesis 1. 

The advancement of NU can directly enhance urban ecological resilience.

2.2.2. Indirect Effects of NU on UER

New-type urbanization (NU) indirectly enhances UER by promoting the optimal allocation of social resources—primarily through industrial structure optimization and improved innovation capacity [13,29]. From the perspective of industrial structure optimization (STRUCT), NU facilitates skill agglomeration and lifestyle transformation. The concentration of skilled labor supports the upgrading of traditional industries, while evolving consumption patterns drive demand for emerging industries and productive services [10]. These dynamics foster the development of new industries, contributing to a more resilient urban economy. Additionally, NU promotes efficient land use through scientific planning, curbing disorderly expansion and reducing resource waste. This not only strengthens STRUCT but also enhances UER.

Hypothesis 2 is therefore proposed:

Hypothesis 2. 

NU can improve UER through industrial structure optimization.

Simultaneously, NU enhances urban innovation capacity (INNO) by attracting talent, capital, and technology, integrating innovation factors into urban development. Improvements in digitalization, informatization, and transport infrastructure accelerate the diffusion and spillover of technology [21]. Technological innovation reduces dependence on traditional inputs like land, energy, and labor, improving resource efficiency and economic productivity. This, in turn, enriches urban resource endowment, mitigates environmental pressures, and boosts UER.

Hypothesis 3 is therefore proposed:

Hypothesis 3. 

NU can enhance UER through the optimization of innovation resource allocation.

2.2.3. Heterogeneity Analysis of NU Affecting UER

(1) Heterogeneity of resource endowment. Resource-based cities in China have long been constrained by overdependence on resource extraction and processing, which has resulted in environmental degradation, ecological risks, industrial decline, and employment pressures [1]. NU, however, offers a pathway to mitigate these challenges by restructuring urban development patterns. First, NU promotes industrial diversification and the growth of knowledge- and service-based sectors, thereby reducing excessive reliance on resource-intensive industries and creating new employment opportunities. Second, NU emphasizes green and low-carbon development, which facilitates cleaner production, promotes technological upgrading, and enhances the efficiency of resource utilization. Third, by strengthening infrastructure, public services, and urban–rural integration, NU improves the allocation of labor, capital, and land resources, thus contributing to the optimization of urban resource endowment. Finally, NU’s focus on ecological protection and sustainable land use not only helps restore degraded ecosystems but also enhances the long-term resilience of resource-based cities. Collectively, these mechanisms illustrate how NU can serve as a transformative driver for addressing the long-standing structural and ecological issues in resource-based cities. Therefore, NU offers new development strategies for resource-based cities by enhancing resource endowment and optimizing resource utilization, thereby strengthening UER. In contrast, non-resource-based cities benefit from a high aggregation of production factors, a low proportion of secondary industries, and a favorable ecological environment. These advantages mean that NU may not significantly enhance the ecological resilience of these cities.

Therefore, Hypothesis 4 was proposed.

Hypothesis 4. 

Comparing with non-resource-based cities, resource-based cities can more significantly improve UER through NU.

(2) Heterogeneity of urban zones. Due to significant differences in geographic location, resource factors, industrial structure, and policy orientation across the regions of China, there are considerable disparities in the willingness and capacity for development. As a result, the regional development imbalance in the process of economic growth has gradually increased [2,20]. The development imbalance between the eastern, central, and western regions of China poses a challenge that may hinder NU in promoting the balanced development of UER. The eastern region is more economically developed, has a higher concentration of financial institutions and enterprises, and is more digitized than the other regions. Cities in the east benefit from better infrastructure, including advanced Internet and communication networks, as well as a high degree of agglomeration of production factors. These advantages mean that the enhancement effect of NU on UER is weak in eastern cities. Conversely, the western regions, while less developed, contain large areas of agricultural land and a larger number of cities than the other regions, which limits the effect of NU on improving UER. For cities in central China, where economic and ecological development is more balanced, the enhancement effect of NU on UER is expected to be stronger than in cities in the east and west.

Therefore, Hypothesis 5 was proposed.

Hypothesis 5. 

The enhancement of UER by NU will be the strongest in the central part of the country, followed by the western part, with the weakest effect in the eastern part of the country.

2.2.4. Spatial Spillover Effects of NU Affecting UER

As the level of urbanization continues to increase, the ability of individual cities to develop independently is increasingly limited. As an advanced phenomenon of regional spatial morphology, city clusters are a hallmark of the rapid development of the national economy and the growing level of modernization, with urban development exhibiting spatial agglomeration and spatial dependency characteristics [21,22]. First, the promotion of NU will generate spatial spillover effects, such as technological and knowledge spillovers. Advanced technologies and production methods will be transferred to neighboring areas [1]. This will strengthen inter-city technological cooperation, promote regional industrial agglomeration, optimize the industrial structure, and facilitate the improvement of ecological resilience in neighboring cities [30]. Second, NU will help to integrate dispersed technological resources. This will promote cross-city dissemination and the exchange of technology, encourage research and development cooperation among various technological innovation entities, and lead to the formation of a green technological innovation network with complementary advantages. This promotes the enhancement of ecological resilience in neighboring cities.

Therefore, Hypothesis 6 was proposed.

Hypothesis 6. 

The promotion of NU has a positive spatial spillover effect on the enhancement of ecological resilience in neighboring cities.

3. Methodology

3.1. Evaluations of the NU and UER

3.1.1. Evaluations of the New-Type Urbanization

Based on the existing research [10,20,21,22], we decomposed the NU process into five dimensions: population, economy, society, space, and land urbanization. We then constructed an evaluation system for the NU index, where a higher index value indicates a higher level and quality of urbanization. To eliminate the influence of data with different scales on the evaluation, the indexes were normalized using the extreme deviation standardization method. The weights of the indexes were then determined using the entropy weight TOPSIS comprehensive evaluation method. Finally, the comprehensive development level of urbanization was calculated using the weighted sum method.

First, we apply the min–max normalization method to standardize the indicator data. Then, the entropy weight method is employed to determine the weight of each indicator. Finally, the TOPSIS method is used to calculate the comprehensive index of NU, as shown in the following formula.

$$C_i={\displaystyle \frac{D_i^{-}}{D_i^{+}+D_i^{-}}};\left(i=1,2,\dots ,m\right)$$

(1)

$$D_i^{+}=\sqrt{{\displaystyle \sum }_{i=1}^n{\left(v_{ij}-v_j^{+}\right)}^2}$$

(2)

$$D_i^{-}=\sqrt{{\displaystyle \sum }_{i=1}^n{\left(v_{ij}-v_j^{-}\right)}^2}$$

(3)

$$V^{+}=\underset{m\ge i\ge 1}{\mathit{max}}{v}_{ij}$$

(4)

$$V^{-}=\underset{m\ge i\ge 1}{\mathit{min}}{v}_{ij}$$

(5)

In Equations (1)–(5), $C_i$ denotes the NU value of the ith region, where 0 ≤$C_i$≤ 1. $V^{+}$ and $V^{-}$ represent the positive and negative ideal solutions, respectively. $D_i^{+}$ and $D_i^{-}$ denote the Euclidean distances from the ith region to the positive and negative ideal solutions, respectively.

3.1.2. Evaluations of the Urban Ecological Resilience

Based on the characteristics of UER and the PSR model, the UER process was divided into three dimensions: pressure resilience, state resilience, and response resilience. This approach combined the actual development of the study area, supplements and screened the indicators according to the basic principles of indicator selection, and ultimately constructed an evaluation index system. In addition to the three dimensions of pr-essure resilience, state resilience, and response resilience, there are also eight indicators (

Table 1. Evaluation index system for UER.

Target Layer	Intermediate Layer	Indicator Layer	Unit	Attribute	Weight
UER	Pressure resilience	Carbon dioxide emissions per capita	kg/person	−	0.0802
		Water resources per capita	m3/person	+	0.0800
	State resilience	Energy consumption per 10,000 Yuan of GDP	tons of standard coal/ten thousand Yuan	−	0.1466
		Construction land per capita	m2/person	+	0.0553
		Park green space per capita	m2/person	+	0.2963
		Green coverage rate of built-up area	%	+	0.1641
	Response resilience	Urban sewage treatment rate	%	+	0.1739
		Domestic waste treatment rate	%	+	0.0033

A plus sign (+) indicates a positive effect, while a minus sign (−) indicates a negative effect.

). The entropy weight TOPSIS comprehensive evaluation method was then used to determine the weights of each indicator, and the comprehensive UER index was calculated using the weighted sum method. The calculation formula is the same as Equations (1)–(5).

3.2. Model Settings

3.2.1. Benchmark Regression and Mediation Effect Model Selection

To test the enhancement of UER by NU, we first constructed a benchmark regression model for the direct effect mechanism as follows:

$${\mathrm{UER}}_{\mathrm{it}}={\mathrm{a}}_0+{\mathrm{a}}_1{\mathrm{NU}}_{\mathrm{it}}+{\mathrm{aX}}_{\mathrm{it}}+{\mu }_{\mathrm{i}}+{\delta }_{\mathrm{t}}+{\epsilon }_{\mathrm{it}}$$

(6)

In Equation (6), ${\mathrm{UER}}_{\mathrm{it}}$ is the UER level; ${\mathrm{NU}}_{\mathrm{it}}$ is the NU level; ${\mathrm{X}}_{\mathrm{it}}$ is a series of control variables; ${\mathrm{a}}_0$ is a constant term; ${\mathrm{a}}_1$ is the coefficient of the core explanatory variables; $\mathrm{a}$ is the parameter to be estimated; ${\mu }_{\mathrm{i}},$ ${\delta }_{\mathrm{t}}$, and ${\epsilon }_{\mathrm{it}}$ denote the individual effect, the time effect, and the random perturbation term, respectively.

To test this hypothesis, after the coefficient passes the significance and robustness tests, the following mediation effect model is constructed for testing:

$${\mathrm{UER}}_{\mathrm{it}}={\gamma }_0+{\gamma }_1{\mathrm{NU}}_{\mathrm{it}}+{\gamma }_2{\mathrm{M}}_{\mathrm{it}}+{\gamma \mathrm{X}}_{\mathrm{it}}+{\mu }_{\mathrm{i}}+{\delta }_{\mathrm{t}}+{\epsilon }_{\mathrm{it}}$$

(7)

In Equation (7): ${\mathrm{M}}_{\mathrm{it}}$ represents the mediating variables, specifically including industrial structure optimization (STRUCT) and innovation resource allocation optimization (INNO); ${\gamma }_0$ is a constant term; ${\gamma }_1$ is the coefficient of the core explanatory variables; ${\gamma }_2$ is the coefficient of the mediating variables; and $\gamma$ is the parameter to be estimated.

3.2.2. Spatial Durbin Model

To test Hypothesis 6, after performing the Lagrange multiplier (LR), likelihood ratio (LR), and Wald tests, we selected the spatial Durbin model to examine the spatial spillover effects of NU on UER. The spatial Durbin model includes both the spatial lag terms of the dependent and independent variables, making it an important model for examining the spatial relationships of geographic elements. The formula is as follows:

$${\mathrm{UER}}_{\mathrm{it}}=\alpha +{\rho \mathrm{WUER}}_{\mathrm{it}}+{\beta }_{\mathrm{k}}{\mathrm{NU}}_{\mathrm{it}}+{\theta }_{\mathrm{k}}{\mathrm{X}}_{\mathrm{it}}+{\mu }_{\mathrm{i}}+{\delta }_{\mathrm{t}}+{\epsilon }_{\mathrm{it}}$$

(8)

In Equation (8), ${\mathrm{UER}}_{\mathrm{it}}$ is the UER level; ${\mathrm{NU}}_{\mathrm{it}}$ is the NU level; ${\mathrm{X}}_{\mathrm{it}}$ is a series of control variables; $\alpha$ is a constant term; ${\beta }_{\mathrm{k}}$ denotes the elasticity coefficient of NU on UER; ${\theta }_{\mathrm{k}}$ denotes the elasticity coefficient of the control variables on UER; and ${\mu }_{\mathrm{i}}$, ${\delta }_{\mathrm{t}}$, and ${\epsilon }_{\mathrm{it}}$ denote the individual effect, the time effect, and the random perturbation term, respectively.

3.3. Variable Selection

3.3.1. Dependent Variable

The dependent variable in this study was UER, which was divided into pressure resilience, state resilience, and response resilience subsystems based on its characteristics and in line with the measurement methods applied in the existing literature. First, carbon dioxide emissions per capita (kg/person) and water resources per capita (m³/person) were selected as the basis for pressure resilience. Second, energy consumption per 10,000 Yuan of GDP (tons of standard coal/ten thousand Yuan), construction land per capita (m²/person), park green space per capita (m²/person), and the green coverage rate of built-up areas (%) were selected as the basis for state resilience. Finally, the urban sewage treatment rate (%) and domestic waste treatment rate (%) were used as the measurement basis for response resilience. Table 1 gives the details of the indicators.

3.3.2. Independent Variable

The independent variable in this study was NU. Based on previous studies, the NU indicator system was constructed from five dimensions: population, economy, society, space, and land. Specifically, the urbanization rate (%) and the proportion of urban employment (%) were used as indicators of population urbanization; the level of economic development (Yuan), the level of industrial and service development (%), and the level of residents’ income (Yuan) were used as indicators of economic urbanization; the proportion of urban built-up land area (%) and the level of residential living space (m²/person) were used as indicators of land urbanization; the level of residents’ education (person) was used as an indicator of social urbanization; and the density of regional highways (km/km²) and urban population density (km²/person) were used as indicators of spatial urbanization.

Table 2. Evaluation index system of NU.

Target Layer	Intermediate Layer	Indicator Layer	Unit	Attribute	Weights
NU	Population urbanization	Urbanization rate	%	+	0.1004
		Proportion of urban employment	%	+	0.1011
	Economic urbanization	Level of economic scale	Yuan	+	0.1006
		Level of industrial and service development	%	+	0.1011
		Level of residents’ income	Yuan	+	0.0989
	Land urbanization	Proportion of urban built-up land area	%	+	0.1008
		Level of residential living space	m2/person	+	0.1006
	Social urbanization	Level of residents’ education	person	+	0.0977
	Spatial urbanization	Density of regional highways	km/km2	+	0.0996
		Urban population density	km2/person	+	0.0988

A plus sign (+) indicates a positive effect.

presents the specifics of the indicators.

3.3.3. Intervening Variables

Optimization of Industrial Structure (STRUCT). The reason for choosing STRUCT as the mediating variable in this study was twofold. First, in the context of global economic integration and rapid industrial and social changes, countries and regions are facing unprecedented development opportunities and challenges. Additionally, an advanced industrial structure is also a key indicator of the quality of economic growth. To achieve sustained and healthy economic and social development, the optimization of industrial structure has become a crucial measure for assessing and promoting the quality of economic growth. This indicator can accurately reflect the process of improving industrial production efficiency from low to high levels, and can also reflect the flow of resources from low-productivity industries to high-productivity industries. In this study, energy consumption per unit of industrial added value is used to measure industrial structure optimization. Because this is an inverse indicator, it must be standardized before measurement.

Optimization of Innovation Resource Allocation (INNO). In the knowledge-based economy, innovation has become the primary driving force behind economic growth. This study selected INNO as an indicator because optimizing the allocation of innovation resources ensures that limited resources (such as capital, skill, and technology) are efficiently and accurately invested in the most promising and impactful innovation projects. This accelerates technological iteration, drives industrial upgrading, and transitions the economy from being factor-driven to innovation-driven. At the same time, in the context of globalization, scientific and technological competition among countries has become increasingly fierce. Optimizing the allocation of innovation resources and upgrading national innovation capacity is a crucial strategy to address the new challenges of global competition and cooperation in science and technology. By concentrating national resources, focusing on key core technologies, and building an autonomous and controllable innovation system, national security and economic development can be effectively safeguarded. A city’s innovation resources cannot be measured solely by the number of patent applications. Therefore, the intensity of R&D funding was selected as an indicator to measure the optimization of innovation resource allocation.

3.3.4. Control Variables

With reference to existing studies, relevant variables were selected to control for the possible impact of NU on UER: population concentration (POP), measured by urban nighttime lighting data (lm/10,000 m²); scientific research support (SCI), measured by R&D funding (10,000 Yuan); overall wage level (WAG), measured by the average wage of employees (10,000 Yuan); fixed asset investment scale (ASI), measured in terms of land-averaged fixed asset investment (10,000 Yuan/km²); and the status of national economic development (NE), measured by the number of industrial enterprises above a certain scale (units).

3.4. Data Sources

The data sources included: (1) nighttime lighting data, obtained from the global 500 m resolution Suomi National Polar-orbiting Partnership (S-NPP) Visible Infrared Imaging Radiometer Suite (VIIRS) nighttime lighting dataset (https://www.geodata.cn/data/datadetails.html?dataguid=8213124601985&docId=90, accessed on 6 May 2025); (2) energy consumption data, sourced from the China Energy Statistics Yearbook; and (3) other social statistics, sourced from the China Urban Statistical Yearbook, China Rural Statistical Yearbook, and China Environmental Statistical Yearbook.

Table 3. Descriptive statistics of variables.

Type of Variable	Name of Variable	Average	Std	Max	Min
Dependent variable	UER	0.3432	0.0681	0.5876	0.1387
Independent variable	NU	0.4647	0.0962	0.7824	0.1651
Intervening variable	STRUCT	3.1354	2.3461	14.56	0.03
	INNO	2.2388	0.2128	2.7497	1.5633
Control variable	POP	5.9600	5.9369	41.5596	0.0833
	SCI	45.9647	128.6647	2944.3945	0.176
	WAG	4.1537	2.9407	22.1678	0.0010
	ASI	1370.8826	2403.0950	44,862	1.5475
	NE	1163.7436	1656.5957	18,792	19

presents the descriptive statistics of the data used in this study. Some of the missing data were input using the sliding average interpolation method. Among this, the control variables exhibit significant standard deviations, reflecting lower data stability, whereas the remaining variables demonstrate consistent stability.

4. Empirical Results and Analysis

4.1. Spatiotemporal Patterns of the NU and UER, 2000–2022

4.1.1. Urban Ecological Resilience

Figure 2 shows the evolutionary characteristics of the spatial and temporal patterns of UER capacity in 281 prefecture-level cities in China from 2000 to 2022. The UER composite index of each city increased annually, with only slight decreases observed between 2000–2005 and 2020–2022. In 2000, the UER composite index of most cities was below 0.3432, indicating that the overall UER was relatively low. By 2005, the number of cities with an UER composite index below 0.3432 had increased, and nearly all cities in the western region had an index below this threshold. From 2010 onward, the UER composite index of cities nationwide increased significantly each year. By 2020, the UER capacity of cities across the country had stabilized and was greater than 0.3432, with some cities exceeding 0.4794. However, a marginal decline in the overall UER level was observed across Chinese cities by 2022. Nevertheless, the composite UER index for China continues to exhibit an upward trajectory overall.

4.1.2. New-Type Urbanization

Figure 3 depicts the spatiotemporal evolution of the composite index of NU for 281 prefecture-level cities in China from 2000 to 2022, indicating a steady and significant annual increase in NU levels nationwide. The development trajectory reveals a distinct spatial pattern: urbanization initially advanced along the eastern coastal areas, subsequently gained momentum in central China, and later expanded into the western regions. In 2000, NU indices across all regions were uniformly low, reflecting the generally underdeveloped stage of urbanization. By 2005, however, clear regional disparities began to emerge, with eastern coastal cities demonstrating rapid and sustained growth in their NU indices, while central, western, and inland eastern cities experienced comparatively moderate improvements. Over the following years, all regions exhibited consistent year-on-year growth, ultimately achieving relatively high and convergent levels of NU nationwide by 2022. This evolutionary process not only mirrors China’s broader trajectory of economic development and spatial restructuring but also shows a statistically significant and regionally differentiated correlation with UER patterns, highlighting the interactive dynamics between urbanization and sustainable development.

4.2. Benchmark Regression

Before conducting the benchmark regressions, a Hausman test was performed to determine whether to use a fixed effects or random effects model when analyzing panel data. The results are presented in

Table 4. Hausman test results.

Variables	(1) UER	(2) UER
NU	0.194 ***	0.195 ***
	(11.64)	(10.51)
POP	−0.000	−0.001 **
	(−1.26)	(−2.49)
SCI	−0.000 ***	−0.000 ***
	(−8.99)	(−9.00)
WAG	0.012 ***	0.013 ***
	(25.93)	(23.53)
ASI	−0.000	−0.000
	(−0.64)	(−0.61)
NE	−0.000	−0.000 **
	(−1.58)	(−2.14)
_cons	0.208 ***	0.209 ***
	(33.79)	(31.32)
N	6463	6463
p-values		0.000
R2		0.5646

Standard errors are in parentheses. ** p < 0.05, *** p < 0.01.

.

The final test results yielded a p-value of 0.0002, i.e., less than 0.01. The null hypothesis was therefore rejected, indicating that the fixed effects model was a better fit for regression than the random effects model. Therefore, this study used a fixed effects model to regress Equation (3), and

Table 5. Benchmark regression results.

Variables	(1) UER	(2) UER	(3) UER	(4) UER
NU	0.113 ***	0.123 ***	0.123 ***	0.123 ***
	(0.014)	(0.018)	(0.030)	(0.047)
POP		0.000	0.000	0.000
		(0.000)	(0.000)	(0.001)
SCI		−0.000 ***	−0.000 ***	−0.000
		(0.000)	(0.000)	(0.000)
WAG		−0.001 *	−0.001	−0.001
		(0.001)	(0.001)	(0.001)
ASI		−0.000 ***	−0.000 ***	−0.000
		(0.000)	(0.000)	(0.000)
NE		0.000 ***	0.000 ***	0.000 *
		(0.000)	(0.000)	(0.000)
_cons	0.235 ***	0.288 ***	0.288 ***	0.288 ***
	(0.008)	(0.007)	(0.014)	(0.020)
N	6463	6463	6463	6463
R2	0.807	0.808	0.808	0.808
YearFE	YES	YES	YES	YES
IDFE	YES	YES	YES	YES
F	84.745	20.278	11.966	3.516

Standard errors are in parentheses. * p < 0.1, *** p < 0.01.

presents the benchmark regression results. Column (1) in the table shows the regression results without control variables, while column (2) displays the results after including control variables. The contrast between columns (1) and (2) highlights the impact of the control variables. To eliminate the effects of heteroskedasticity and correlations between samples within the same cluster, column (3) adjusts for robust standard errors in the ordinary least squares (OLS) estimation, and column (4) adjusts for cluster-robust standard errors to provide more accurate regression results. In columns (1) and (2), the regression coefficients of the independent variables are positive and pass the 1% significance test, suggesting that NU enhances overall UER, thus supporting Hypothesis 1.

Among the control variables, the regression coefficients for POP and NE were positive, indicating that excessively dense populations and rapid economic development lead to increased energy consumption and carbon emissions, which will harm the environment and reduce UER. The coefficients for SCI, WAG, and ASI were negative, indicating that excessive scientific research funding, increased residents’ incomes, and increased fixed asset investment do not significantly impact the city’s ecological resilience. The limited influence of financial support, the development of the fixed asset economy, and human capital in the study area on the optimal allocation of resources and environmental elements is a plausible explanation for this observation.

4.3. Endogenous Issue

The statistical properties of OLS estimators rely on the critical assumption that the error term is not correlated with the explanatory variables. If, for any reason, there is a correlation between the error term and the explanatory variables, the validity of our previous estimation results may be questioned. The possible reasons for such correlations include endogenous associations. Although the study controlled for variables affecting UER as much as possible, endogeneity due to omitted variables may still arise. Specifically, some control variables may exhibit an inverse relationship with NU. To address this, all control variables were lagged by one period to reduce potential endogeneity issues. The test results are presented in column (1) of

Table 6. Endogeneity test results.

	Lag the Control Variable by One Period	Instrumental Variable Method
Variables	(1) UER	(2) NU	(3) UER	(4) NU	(5) UER
NU	0.0769 ***		0.2061 ***		0.1523 **
	(5.33)		(6.39)		(2.52)
IPR		0.0048 ***		0.0040 ***
		(32.34)		(20.70)
CPPR		0.0001**		0.0001 ***
		(2.35)		(2.60)
Constant	0.2463 ***	0.2613 ***	0.2988 ***	0.2540 ***	0.3079 ***
	(30.40)	(37.74)	(16.53)	(24.68)	(11.12)
Control Variables	YES	NO	NO	YES	YES
YearFE	YES	YES	YES	YES	YES
IDFE	YES	YES	YES	YES	YES
F			(19.93)		(19.93)
Cragg-Donald			1049.23		358.602
Observations	6182	281	281	281	281
R-squared			0.153		0.156

Standard errors are in parentheses. ** p < 0.05, *** p < 0.01.

. The effect of NU in enhancing UER was still apparent, with the significance and signs of the regression coefficients being consistent with the benchmark regression results.

Second, cities with a higher UER tend to have more advanced digitalization and informatization technologies, which in turn provide NU with a “first-mover advantage”, creating an endogeneity issue in the causality judgment of the study. To mitigate this endogeneity, we employed the instrumental variable (IV) method. We selected the Internet penetration rate (IPR) and the cell phone penetration rate (CPPR) in 2018 as instrumental variables for NU. These values were measured by the number of Internet-connected households and the number of cell phone subscribers per 100 people, respectively. We then conducted an instrumental variables-two-stage least squares (IV-2SLS) regression, with the results presented in columns (2) to (5) of Table 6. Specifically, columns (2) and (3) represent the IV-2SLS without control variables, with column (2) showing the first stage of the regression and column (3) showing the second stage. Columns (4) and (5) represent the IV-2SLS with control variables, with column (4) showing the first stage and column (5) showing the second stage. The results indicated that the hypothesis that NU enhances UER still holds and passes the significance test at the 5% level, regardless of whether control variables are included, after addressing the endogeneity issue.

Additionally, in the weak identification test of the instrumental variables, the Cragg–Donald F statistic exceeded the critical value at the 10% level in the Stock–Yogo weak identification test, and the test result rejected the hypothesis of a weak correlation of instrumental variables, indicating that the selection of instrumental variables was valid. In conclusion, the previous results remained robust after addressing the endogeneity issue.

4.4. Robustness Test

To further validate the robustness of these findings, additional tests were conducted by shortening the sample period, adjusting the observation window, excluding policy interference, and excluding the effect of outliers.

Shortening the sample period: Considering the impact of the COVID-19 epidemic on land use activities, the sample period was shortened to 2000–2019 for the regression analysis. The regression results, shown in column (1) of

Table 7. Robustness test results.

Variables	Shortening the Sample Period (1) UER	Changing the Observation Window (2) UER	Excluding Policy Interference (3) UER	Excluding the Effect of Outliers (4) UER
NU	0.324 ***		0.123 ***	0.144 ***
	(0.032)		(0.019)	(0.017)
L2.NU		0.209 ***
		(0.023)
_cons	0.170 ***	0.205 ***	0.287 ***	0.282 ***
	(0.015)	(0.010)	(0.007)	(0.007)
N	5620	5901	6233	5816
R2	0.823	0.7812	0.810	0.789
YearFE	YES	YES	YES	YES
IDFE	YES	YES	YES	YES
F	40.610	0.7798	20.269	24.713

Standard errors are in parentheses. *** p < 0.01.

, indicated that the regression coefficient for NU remained significantly positive after reducing the sample period, thereby confirming the robustness of the benchmark regression results.

Changing the observation window: In correlation studies, it is common practice to lag the explanatory variables by one or two periods in regression analyses to address the endogeneity problem and enhance the robustness of the model, leading to more accurate analytical results. In this study, we lagged the independent variable, NU, by two years to exclude potential confounding factors, thereby ensuring more reliable and accurate conclusions. As shown in column (2) of Table 7, the regression coefficient for NU remained significantly positive after changing the observation window, confirming the robustness of the benchmark regression results.

Excluding policy interference: To eliminate the impact of localized policies on the comprehensive green transformation of cities’ ecological resilience, we conducted regressions on the remaining sample after excluding ten cities, including Wuxi and Fuzhou. As shown in column (3) of Table 7, the benchmark regression results remained robust even after excluding policy interference.

Excluding the effect of outliers: To mitigate the impact of outliers on the regression results, we truncated the top and bottom 5% of the explanatory variables. The test results are shown in column (4) of Table 7. The benchmark regression results remained robust even after excluding sample selection bias.

4.5. Mechanism Test

To verify Hypotheses 2 and 3, whether NU influences UER through STRUCT and INNO was investigated using the mediation effect model. Columns (1) and (3) of

Table 8. Mechanism test results.

Variables	(1) INNO	(2) UER	(3) Sturct	(4) UER
NU	1.790 ***	0.175 ***	10.442 ***	0.121 ***
	(0.018)	(0.023)	(0.403)	(0.015)
INNO		0.035 ***
		(0.010)
STRUCT				0.001
				(0.000)
_cons	1.407 ***	0.339 ***	7.988 ***	0.285 ***
	(0.008)	(0.015)	(0.188)	(0.007)
N	6463	6463	6463	6463
R2	0.967	0.807	0.867	0.807
YearFE	YES	YES	YES	YES
IDFE	YES	YES	YES	YES
F	9638.661	38.379	670.116	33.461

Standard errors are in parentheses. *** p < 0.01.

show the results obtained for estimating the impact of NU on STRUCT and INNO, respectively. The results showed that NU positively affected both STRUCT and INNO. From this, we inferred that the promotion of NU facilitates the optimization of innovation resource allocation by increasing the intensity of R&D investment. Additionally, NU contributes to STRUCT, thereby reducing energy consumption per unit of industrial added value.

Second, columns (2) and (4) show the effects of NU on UER after adding the mediating variables. The results indicated that both STRUCT and INNO positively impacted UER. Additionally, compared to the benchmark regression results, the regression coefficient for NU decreased but remained significant. This suggests that STRUCT and INNO play a role in mediating the effect. Specifically, the mediating effect of optimizing INNO was 0.063 (1.790 × 0.035). This was higher than the mediating effect of STRUCT, which was 0.010 (10.442 × 0.001).

To strengthen the credibility of the mechanism test results, we also used the Sobel and bootstrap tests to validate the findings. The test results, shown in

Table 9. Results of the Soble and Bootstrap tests.

Variables	STRUCT		INNO
Sobel P > |Z|	0.000		0.000
	_bs_1	_bs_2	_bs_1	_bs_2
[95%conf.interval]	[0.15, 0.176]	[0.251, 0.288]	[0.239, 0.310]	[0.119, 0.197]

Note: _bs_1 represents the indirect effects, while _bs_2 represents the direct effects.

, indicated that both tests produced statistically significant results, confirming that NU can influence UER through STRUCT and INNO.

As discussed in the theoretical analysis above, NU helps optimize the allocation of limited social resources, which in turn promotes STRUCT and INNO. Struct and INNO ultimately further enhance UER. In summary, STRUCT and INNO are two key mechanisms through which NU enhances the ecological resilience of cities, thereby verifying Hypotheses 2 and 3.

4.6. Heterogeneity Analysis

4.6.1. Heterogeneity of Resource Endowment

The samples were divided into resource-based and non-resource-based cities for a group regression, based on their local government classification. The regression results in columns (3) and (4) of

Table 10. Heterogeneity analysis results.

Variables	Eastern Cities	Central Cities	Western Cities	Resource-Based Cities	Non-Resource-Based Cities
NU	0.156 ***	0.338 ***	0.195 **	0.250 ***	0.074
	(0.048)	(0.039)	(0.076)	(0.060)	(0.047)
_cons	0.279 ***	0.247 ***	0.296 ***	0.224 ***	0.313 ***
	(0.019)	(0.019)	(0.016)	(0.026)	(0.023)
N	2300	2277	1874	2588	3853
R2	0.811	0.833	0.804	0.810	0.808
YearFE	YES	YES	YES	YES	YES
IDFE	YES	YES	YES	YES	YES
F	10.650	15.596	5.284	17.551	2.446

Standard errors are in parentheses. ** p < 0.05, *** p < 0.01. Note: According to the “three major zones” division established by the National Bureau of Statistics, China’s central region consists of eight provinces: Anhui, Henan, Shanxi, Hubei, Hunan, Jiangxi, Jilin, and Heilongjiang; the eastern region includes 11 provinces: Beijing, Tianjin, Hebei, Liaoning, Shandong, Jiangsu, Shanghai, Zhejiang, Fujian, Guangdong, and Hainan; and the western region comprises 12 provinces: Chongqing, Yunnan, Sichuan, Guizhou, Tibet, Guangxi, Xinjiang, Qinghai, Ningxia, Gansu, Shaanxi, and Inner Mongolia.

show that NU significantly improved UER in resource-based cities, but the effect was not significant for non-resource-based cities. Thus, Hypothesis 4 was verified. Compared to non-resource-based cities, resource-based cities face issues such as a single industrial structure, an excessive proportion of industrial land, and large areas of idle industrial land. Therefore, resource-based cities have more potential to optimize their land resource allocation, improve land use structure, and control environmental pollution, making it easier for NU to enhance UER.

4.6.2. Heterogeneity of Urban Zones

The regression results in columns (1), (2), and (3) of Table 10 show that NU enhanced the ecological resilience of central cities more than it did for western and eastern cities, with the effect in western cities being slightly higher than in eastern cities. Hypothesis 5 was therefore supported. Compared to eastern cities, central and western cities, which are currently undergoing the implementation of the western development strategy and the rise of central China, respectively, are better positioned to demonstrate the impact of NU on UER. Moreover, the effect of urbanization on ecological resilience was more pronounced in central cities, which have reached a high level of urbanization, than in western cities where urbanization is progressing more slowly.

4.7. Spatial Spillover Effects

4.7.1. Selection of the Spatial Econometric Model

The NU process in cities exhibits spatial correlation effects due to the “strong linkages” of socio-economic factors. Therefore, spatial spillovers must be considered when assessing its impact on UER. Moran’s I values of the NU and UER indexes were significantly positive in most years, indicating a spatially positive correlation between NU and UER in each city. This supports the appropriateness of using a spatial econometric model in this study.

To determine the appropriate spatial econometric regression model, we conducted LM, LR, and Wald tests, with the results shown in

Table 11. Model selection test results.

Spatial Panel Model Test		Value	p-Value
LM test	Moran’s I	12.956	0.000
	LM-lag	283.040 ***	0.000
	Robust-LM-lag	54.856	0.000
	LM-error	2954.655 ***	0.000
	Robust-LM-error	2726.471	0.000
LR test	LR-SDM/SEM	1338.58 ***	0.000
	LR-SDM/SAR	4168.45 ***	0.000
Wald test	Wald-SDM/SEM	14.63 ***	0.000
	Wald-SDM/SAR	17.99 ***	0.000

*** p < 0.01.

. Because the results of the LM, LR, and Wald tests were all significant, and the Hausman test results determined that a fixed-effects model was preferable to a random-effects model, we ultimately selected the two-way fixed-effects spatial Durbin model for this analysis.

4.7.2. Spatial Durbin Model Analysis

Table 12. Spatial Durbin model regression results.

Variables	Individual Fixed Effect	Time Fixed Effect	Two-Way Fixed-Effect
Main
NU	0.169 ***	0.189 ***	0.214 ***
	(0.019)	(0.024)	(0.042)
Spatial
rho	0.711 ***	−0.061	0.060
	(0.024)	(0.211)	(0.060)
Variance
sigma2_e	0.001 ***	0.002 ***	0.001 ***
	(0.000)	(0.000)	(0.000)
Direct
NU	0.171 ***	0.189 ***	0.214 ***
	(0.019)	(0.024)	(0.042)
Indirect
NU	0.400 ***	−0.008	0.006
	(0.031)	(0.032)	(0.007)
Total
NU	0.571 ***	0.181 ***	0.221 ***
	(0.043)	(0.044)	(0.044)
N	6463	6463	6463
R2	0.366	0.361	0.384
Year FE	YES	YES	YES
IDFE	YES	YES	YES

Standard errors are in parentheses. *** p < 0.01.

presents the results of the benchmark regression examining the impact of NU on UER. To further emphasize the explanatory power of the two-way fixed-effects spatial Durbin model, we compared these results with those from the individual fixed-effects and time fixed-effects models. The spatial Durbin model with two-way fixed effects in column (3) had a higher R² value, a larger absolute value of the log-likelihood, and a better fit compared to the single fixed-effects models in columns (1) and (2). Therefore, we analyzed the regression results from the two-way fixed-effects spatial Durbin model.

Column (3) of Table 12 shows that the regression coefficient for NU was significantly positive, indicating that the promotion of NU positively affected the UER of the cities involved, thereby further validating Hypothesis 1. Column (3) also presents the decomposition results of the spatial effects of NU on UER. The coefficients for the direct and indirect effects of NU on UER were 0.214 and 0.006, respectively, indicating that the promotion of NU in a city has a significant positive impact on its ecological resilience and a weak positive impact on the ecological resilience of neighboring cities. Hypothesis 6 was therefore verified.

5. Conclusions and Policy Recommendations

5.1. Conclusions

Based on panel data from 281 prefecture-level cities in China between 2000 and 2022, this study constructed an evaluation index system for NU and UER and employed the entropy weight TOPSIS comprehensive evaluation method to calculate the NU and UER comprehensive indices for each city over the period of 2000 to 2022. The mechanisms, heterogeneity, and spatial spillover effects of NU on UER were empirically investigated. The key findings were as follows. First, NU significantly improved overall UER, a conclusion that remained robust even after addressing endogeneity issues and conducting a series of robustness tests. Second, NU enhanced UER by optimizing the industrial structure and innovatively reallocating resources. Third, the impact of NU on UER varied across urban zones and resource endowments, with the effect being more pronounced in central and resource-oriented cities. Fourth, NU exhibited a spatial spillover effect on UER, with the direct effect outweighing the indirect effect. The findings highlight that NU significantly improved UER in resource-based and central cities, but the effect was not significant for non-resource-based, western and eastern cities. Notably, the study also uncovered that the promotion of NU in a city has a significant positive impact on its ecological resilience and a weak positive impact on the ecological resilience of neighboring cities.

These findings emphasize the need for differentiated policy approaches and targeted interventions. In terms of benefits, this research contributes to understanding how urban development and UER can be better integrated, offering new perspectives for urban ecological planning.

5.2. Policy Recommendations

The findings have several implications for enhancing UER while advancing the high-quality construction of NU. First, the development concept of NU should be reformulated to accelerate its progress, which can be achieved by fostering the comprehensive and coordinated development of population, economy, land, society, and space. The construction of NU must be closely aligned with the improvement of UER. To effectively construct NU there is a need for efficient and intensive land use, which can be achieved through the optimal allocation of land resources based on the actual demands for different land use types during social and economic transformation. At the same time, the principles of an ecological civilization and people-oriented development must be embedded throughout the entire construction of NU process. By integrating the enhancement of economic efficiency, the reduction in emissions, carbon sequestration, and the improvement of the ecological environment, the construction of NU should be advanced efficiently while simultaneously establishing a long-term mechanism to enhance ecological resilience.

Second, in recent years, as China’s economic development has entered a new phase, the country has transitioned from a period of rapid growth to a stage of high-quality development. The economy currently faces a complex situation marked by the convergence of three key developmental phases: a slowdown in growth, structural adjustment challenges, and the process of digesting the effects of previous stimulus policies. These overlapping factors have made traditional development unsustainable. Issues such as resource wastage and diminished resilience are becoming increasingly prominent, hindering high-quality economic development. In promoting the construction of NU, regions should more strongly support the optimization of innovative resource allocation and industrial structure to ensure the continuous improvement of UER. The government should actively provide policy and financial support for STRUCT. Local governments must consider the current situation and goals of urban development and introduce policies that guide the transformation and upgrading of traditional industries with high energy consumption and emissions. At the same time, regions should actively foster low-carbon, high-efficiency industries through tax reductions and subsidies, while considering both consumer demand and the city’s factor endowment.

Finally, cities should consider the spatial spillover effects from neighboring cities while taking into account their zones and resource endowments. They should formulate policies tailored to local conditions, based on a clear understanding of the stage of development and current status of the city’s NU. This will facilitate the efficient use of land resources and promote low-carbon transformation. Eastern and resource-oriented cities should accelerate the development of NU, making it a key strategy for enhancing the ecological resilience of these cities. For cities in central and western China, improving factor allocation capacity is crucial. This can be achieved by introducing skills development, technology, and advanced management practices from large cities, thereby better leveraging the impact of NU on UER.

However, this paper also has some limitations. On one hand, the academic community has not yet reached a consensus on the evaluation criteria and index system for NU. Due to data limitations, multi-dimensional processes, such as urban–rural integration, have not been included. On the other hand, the relationship between the two and the mechanisms underlying their interaction at a more detailed level have yet to be thoroughly explored. In the future, the indicator system can be further refined, and the scope of research can be expanded to provide more detailed guidance for better promoting the construction of NU and improving the overall ecological resilience of cities.