Research on the impact of digital infrastructure construction on ecological resilience in the Yellow River Basin—empirical evidence from 75 cities in the Yellow River Basin

Well-developed digital infrastructure serves as a crucial vehicle and driving force for enhancing high-quality development and ecological environmental protection in the Yellow River Basin. Using a sample of 75 cities in the Yellow River Basin from 2014 to 2023, this paper investigates the impact mechanism and spillover effects of digital infrastructure construction on the basin’s ecological resilience. The findings reveal that digital infrastructure construction in the Yellow River Basin significantly improves the level of ecological resilience, and this promotive effect exhibits an increasing trend over time. Further mechanism analysis indicates that digital infrastructure construction drives ecological resilience enhancement through effects of resource allocation optimization and innovation efficiency upgrading, and it demonstrates spatial linkage spillover effects. Heterogeneity tests show that the promotive effect of digital infrastructure is more pronounced in upstream and downstream cities of the Basin, constrained by the traditional industrial structure and functional positioning of the middle reaches. Furthermore, the study finds that the enhancing effect of digital infrastructure construction on ecological resilience is even greater in cities abundant in renewable energy. This research provides empirical evidence for objectively assessing the impact of digital infrastructure construction on the ecological resilience of the Yellow River Basin, contributing to the accelerated advancement of ecological protection and high-quality development, thereby facilitating the sustainable development of the Yellow River Basin.

1 Introduction

The Yellow River Basin, serving as a crucial ecological barrier and economic zone in China, has its ecological security and sustainable development directly linked to the nation’s strategic landscape. In recent years, under the combined pressures of intensified climate change and human activities, the basin faces multiple challenges, including water scarcity, soil erosion, and habitat degradation. Enhancing ecological resilience has, therefore, become a core objective for achieving the modernization goal of “harmonious coexistence between humanity and nature.” The proposals for the 15th Five-Year Plan explicitly advocate to “moderately advance the construction of new infrastructure” and “accelerate the comprehensive green transformation of economic and social development,” while identifying technologies like digital twin as key supports for improving watershed governance efficacy. This strategic direction opens new avenues for ecological management in the Yellow River Basin. Digital infrastructure, forming the core foundation of the digital economy, with its characteristics of virtual iteration and efficient synergy, aligns well with the transformative needs of basin ecological governance that emphasizes “source control and front-end innovation.” Although existing research has explored the synergy between ecological protection and high-quality development in the Yellow River Basin, most studies lack an in-depth investigation into the interactive mechanisms between digital infrastructure and ecological resilience. Research gaps remain, particularly concerning spatial spillover effects and regionally differentiated driving pathways. How digital infrastructure can systematically enhance the resistance capacity, adaptive capacity, and restorative capacity of a basin’s ecosystem through optimizing resource allocation and strengthening cross-regional coordination remains a critical question demanding resolution. Therefore, this study focuses on digital infrastructure—encompassing communication systems, data centers, and intelligent platforms—and analyzes its driving mechanisms and spatial spillover effects on ecological resilience in the Yellow River Basin. It aims to elucidate how digital infrastructure enhances the adaptive, restorative, and transformative capacities of the basin’s ecosystem. This research not only responds to the strategic directives of the 15th Five-Year Plan proposals to “build a Digital China” and “strengthen ecological security barriers,” but also provides empirical insights for intelligent and coordinated ecological governance in river basins.

Based on existing research, the literature closely related to this paper primarily falls into the following two categories. The first category consists of studies focusing on the measurement methods and the coupling coordination relationship of ecological resilience. Firstly, regarding measurement methods, Yuan et al. (2022) established a framework of indicators influencing the resilience of resource-based cities. Using Changzhi, China, as a case study, they evaluated and investigated the distribution of ecological resilience. Jiang et al. (2025) constructed evaluation models for both economic resilience and ecological resilience. Their research found that ecological resilience and economic resilience mutually inhibit each other’s development and exhibit spatial heterogeneity. Li and Wang (2023) employed the entropy weight method to assess the ecological resilience of Chinese cities from 2005 to 2020. Their study revealed a gradual decrease in the fluctuation of ecological resilience levels but emphasized the need to strengthen collaboration between neighboring cities to narrow the ecological resilience gap. Secondly, concerning the coupling coordination relationship, Sheng et al. (2025) systematically elaborated on the coupling relationship between ecological resilience and high-quality economic development. They argued that strong ecological resilience not only provides fundamental ecological products but also injects sustained momentum into economic “quality and efficiency improvement” by compelling industrial green transformation and attracting green investments. Zhou and Xu (2025) were the first to link “ecological resilience” with “new quality productive forces” for a coupling analysis. Their research indicates that ecological resilience provides a stable environmental guarantee for new technologies and industries, while new quality productive forces, in turn, reinforce and enhance ecological resilience through green technological innovation. Existing literature has conducted multi-dimensional and multi-level research on ecological resilience in the Yellow River Basin, providing crucial theoretical support for ecological protection and high-quality development in the basin. However, there remains scope for further exploration regarding the influencing factors of ecological resilience specifically within the Yellow River Basin.

The second category of literature primarily involves extensive discussion on the economic effects of digital infrastructure construction. Firstly, regarding micro-level enterprise effects, Yan (2025) found that local digital infrastructure construction can drive innovation among regional enterprises, promoting increased innovation investment and output growth, helping companies expand their innovation boundaries and develop new technologies (Shen et al., 2023), and ultimately enabling Chinese exporting firms to move up the global value chain (GVC). Secondly, in terms of macro-level economic effects, Liu (2025) and Huang et al. (2025) studied the impact of digital infrastructure on urban economic resilience. Their research found that digital infrastructure can significantly enhance urban economic resilience, with technological innovation being an important transmission mechanism. The higher the level of technological innovation, the more digital infrastructure enhances urban economic resilience. Liu et al. (2025) demonstrated the impact of digital infrastructure development levels on “carbon reduction and efficiency enhancement.” Their study found that digital infrastructure has a significant carbon reduction effect, which increases gradually over time and manifests as a narrowing regional gap spatially. Finally, concerning regional coordinated development, Zhou and Chao (2023), based on a study of the Yellow River Basin, found that from 2013 to 2021, the coupling coordination degree of the ternary system of digital infrastructure, economic resilience development, and ecological environmental protection in the Yellow River Basin showed a stable upward trend. However, the spatial correlation of coupling coordination between the upper and lower reaches was weak, with spatial connection strength showing a multi-core agglomeration pattern. Zhang and Li (2023) found that network infrastructure weakens traditional geographical advantages. Compared to transportation infrastructure, network infrastructure has a greater positive impact on inclusive green growth.

A review of the aforementioned literature reveals that while existing studies provide substantial insights into digital infrastructure construction and ecological resilience separately, most are based on independent or segmented dimensional perspectives examining the economic effects of digital infrastructure. There is a lack of discussion on the ecological benefits of digital infrastructure, particularly regarding its driving mechanisms and network spatial effects on ecological resilience. Therefore, critical questions remain: Can digital infrastructure construction enhance ecological resilience in the Yellow River Basin? What is the intrinsic mechanism through which it operates? Furthermore, does the impact of digital infrastructure exhibit dynamic persistence and spillover effects? Addressing these questions will help accurately assess the macro-level driving effects of digital infrastructure construction, deepen the understanding of its role in shaping the ecological resilience of the Yellow River Basin, and provide empirical evidence to inform relevant policy-making.

Compared to the existing literature, the potential marginal contributions of this paper are as follows: First, this study enriches and expands the literature on the determinants of ecological resilience in the Yellow River Basin. While existing research has focused more on measuring and analyzing ecological resilience, it has overlooked the mechanistic investigation of the driving role of digital infrastructure construction. This paper, from the perspective of digital infrastructure construction, explores its impact mechanism on the ecological resilience of the Yellow River Basin, thereby providing a valuable addition to the literature on high-quality development and ecological environmental protection in the region. Second, based on the attributes and functional characteristics of digital infrastructure construction, this paper constructs an indicator system across three dimensions: communication infrastructure, data infrastructure, and intelligent infrastructure. It derives a composite digital infrastructure construction index through calculation, distinguishing it from existing studies that often examine the effects of digital infrastructure from a single dimension. Third, this paper provides an in-depth analysis of the driving effects of digital infrastructure construction through factor resource allocation optimization and innovation efficiency upgrading. It further examines the spatial spillover effects of digital infrastructure construction, thereby offering empirical evidence for effectively enhancing the ecological resilience of the Yellow River Basin by integrating data from empirical analysis.

2 Theoretical analysis and research hypotheses

Digital infrastructure can be defined as a new-generation infrastructure system with innovation as its core driver, ubiquitous information networks as its physical foundation, and data as its key factor of production. It is designed to empower the intelligent transformation of the entire society and the intelligent upgrading of industrial chains. Currently, the construction of digital infrastructure has emerged as an internal driving force and crucial support for advancing high-quality development and ecological environmental protection in the Yellow River Basin. While altering the conditions for economic activities, it significantly optimizes the allocation of factor resources, accelerates the improvement of innovation efficiency, and exerts spatial linkage effects by amplifying the multiplier effects of digital infrastructure construction, thereby promoting the enhancement of ecological resilience in the Yellow River Basin (Figure 1).

Figure 1

Figure 1. Theoretical logic diagram.

2.1 Resource allocation optimization effect

New Economic Geography theory posits that reductions in transportation costs lead to agglomeration economies, externalities, and economies of scale. Digital infrastructure construction enhances inter-regional accessibility and lowers the transportation costs for factors and products. The resulting agglomeration and scale effects contribute to optimizing the spatial allocation of factor resources (Tsekouras et al., 2016) and improving the efficiency of resource allocation across regions (Redding and Turner, 2015; Abula and Aihemaiti, 2025). On the one hand, digital infrastructure construction in the Yellow River Basin promotes the enhancement of the Basin’s ecological resilience by optimizing the structure of resource allocation throughout the watershed and improving the efficiency of resource allocation. On the other hand, it optimizes resource allocation (Vial, 2021; Acemoglu and Restrepo, 2019), thereby avoiding resource waste, reducing waste pollutant discharge, and consequently mitigating environmental pollution. By performing an intensive-orientation function for ecological and environmental protection, it more efficiently unleashes the potential of factor utilization, thus elevating the ecological resilience level of the Yellow River Basin. Based on this, the following hypothesis is proposed:

Hypothesis I: Digital infrastructure construction can drive the improvement of ecological resilience in the Yellow River Basin by optimizing basin-wide resource allocation and enhancing resource allocation efficiency.

2.2 Innovation-driven effect

The continuous improvement of digital infrastructure, such as big data and artificial intelligence, can expand process innovation across production stages and reduce resource occupancy in internal low-efficiency production links (Czernich et al., 2011), ultimately leading to a leap in regional innovation efficiency. On the one hand, through the innovation efficiency upgrading effect, digital infrastructure construction in the Yellow River Basin can increase innovation output within the region (Xu et al., 2025), realizing a “low-input, high-output” economy conducive to low-carbon development and enhancing ecological resilience (Li et al., 2025). On the other hand, upgraded innovation efficiency can transform the approach to ecological protection in the Yellow River Basin through technological innovation, shifting from “end-of-pipe treatment” to “source prevention and control, along with front-end innovation.” This transformation compels a shift in economic growth drivers, promotes a change in the economic development model, and thereby boosts the level of ecological resilience. Based on this, the following hypothesis is proposed:

Hypothesis II: Digital infrastructure construction can drive the improvement of ecological resilience in the Yellow River Basin by accelerating the integration of innovative knowledge and enhancing the level of technological innovation in ecological governance.

2.3 Spatial linkage and spillover effects

Digital infrastructure construction enhances inter-city collaboration, enriches regional economic connections and information exchange, and generates spillover effects to cities along its network by strengthening the driving influence on peripheral neighboring areas (Jin et al., 2025). This fosters interconnected and coordinated development among the upper, middle, and lower reaches of the Yellow River Basin (Zhang and Zhang, 2020). On the one hand, digital infrastructure construction integrates the economic activities of the Yellow River Basin into a cohesive whole. Through positive spatial spillover effects, it enables rapidly growing areas to stimulate economic development in slower-growing regions (Miao and Zhang, 2021; Wang et al., 2024), thereby enhancing the overall ecological resilience of the Basin. On the other hand, from the perspective of spatial linkage and spillovers, digital infrastructure construction continuously strengthens coordinated ecological governance across the entire Basin (Manny et al., 2022; Ma et al., 2023). By leveraging mechanisms such as ecological compensation and green finance, it activates the wealth appreciation function of ecological protection, consequently boosting the ecological resilience of the Yellow River Basin. Based on this, the following hypothesis is proposed:

Hypothesis III: Digital infrastructure construction can drive the improvement of ecological resilience in the Yellow River Basin by promoting inter-regional coordinated governance and exerting spatial linkage and spillover effects.

3 Methods

3.1 Econometric model specification

To thoroughly investigate the driving mechanisms of digital infrastructure construction on ecological resilience in the Yellow River Basin, and drawing on existing research while considering data availability, this paper establishes the baseline regression model shown in Equation (1):

$\begin{array}{ll}E{R}_{\mathit{it}}={\beta }_0+{\beta }_{1}D{I}_{\mathit{it}}+\lambda X_{\mathit{it}}+{\mu }_i+{\gamma }_t+{\epsilon }_{\mathit{it}}& (1)\end{array}$

Where: The subscript $i$ denotes individual cities. The subscript $t$ denotes years. $E{R}_{\mathit{it}}$ represents the ecological resilience index of the Yellow River Basin for region $i$ in period $t$ . $D{I}_{\mathit{it}}$ indicates the level of digital infrastructure development. $X_{\mathit{it}}$ is a vector of control variables that affect ecological resilience. ${\mu }_i$ represents city fixed effects, which control for all time-invariant heterogeneity across cities. ${\gamma }_t$ represents time fixed effects, which absorb common shocks that affect all cities during the sample period. ${\epsilon }_{\mathit{it}}$ is the random error term, assumed to be independently and identically distributed.

3.2 Variable selection

3.2.1 Explained variable: ecological resilience (ER)

Ecological resilience refers to the “protective barrier” of an ecosystem against risk shocks, manifested as its resistance, adaptation, and restoration capacities. This paper constructs an evaluation indicator system for ecological resilience from three dimensions: resistance, adaptability, and restorative capacity within the Yellow River Basin. Resistance primarily includes sulfur dioxide (SO₂) emissions, water consumption per 10,000 yuan of GDP, and energy consumption per 10,000 yuan of GDP. It reflects the robustness of the ecosystem when subjected to shocks. Adaptability mainly comprises forest coverage rate, green coverage rate in built-up areas, per capita park green space area, and per capita water resources. It characterizes the adjustment capacity of the ecosystem under shock. Restorative Capacity includes the urban domestic sewage treatment rate, soil erosion control area, comprehensive utilization rate of industrial solid waste, and investment in industrial pollution control. It represents the recovery capacity and potential of the ecosystem after being impacted. The specific indicator system is shown in Table 1.

Table 1

Table 1. Evaluation indicator system for ecological resilience in the Yellow River Basin.

3.2.2 Core explanatory variable: level of digital infrastructure construction (DI)

Digital infrastructure is an infrastructure system led by innovation-driven development, grounded in information networks, and utilizes data factors as key production inputs, aiming to promote the intelligent upgrading of related industrial chains (Henfridsson and Bygstad, 2013). It primarily includes communication infrastructure, data infrastructure, and intelligent infrastructure. First, communication infrastructure encompasses communication network infrastructure represented by mobile communications and optical fiber networks. Communication enhances information exchange efficiency, resolves information asymmetry, and enables the co-construction and sharing of resources and information among cities in the Yellow River Basin. Therefore, this paper employs the number of internet access ports, the number of year-end mobile phone subscribers in the city, and the number of mobile base stations to measure the construction of communication infrastructure in the Yellow River Basin. Second, data infrastructure refers to infrastructure represented by data centers and intelligent computing centers. Data centers facilitate the extraction of value from massive datasets, ensure data fulfills its factor function, and empower the digital transformation of traditional infrastructure. Consequently, this paper specifically uses the registered capital level of big data enterprises, the number of data trading platforms, and the proportion of manufacturing data information input to measure the construction of data infrastructure in the Yellow River Basin. Finally, intelligent infrastructure refers to infrastructure represented by artificial intelligence, among others. As a capital-extending technology, AI enhances production efficiency and benefits by altering factor combination methods, creating a powerful engine for economic growth. Thus, this paper utilizes the number of urban AI patent inventions, industrial robot installation density, and the number of industrial internet patents to measure the construction of intelligent infrastructure in the Yellow River Basin. The construction of the digital infrastructure construction indicator system is detailed in Table 2.

Table 2

Table 2. Indicator system for digital infrastructure construction.

3.2.3 Control variables

Drawing on relevant literature and an analysis of factors influencing ecological resilience in the Yellow River Basin, this paper selects the following control variables:

1. Institutional Quality: Measured by the ratio of a city’s fiscal budget revenue to its GDP.

2. Foreign Investment Level: Measured by the ratio of the actual utilized foreign direct investment (FDI) in a city to its GDP.

3. Urban Investment Intensity: Measured by the share of a city’s total fixed-asset investment in its GDP.

4. Industrial Structure Upgrading: Measured by the ratio of the output value of a city’s tertiary industry to that of its secondary industry.

5. Financial Development Level: Measured by the ratio of the total balance of deposits and loans in financial institutions at year-end to the city’s GDP.

6. Social Consumption Level: Measured by the share of a city’s total retail sales of consumer goods in its GDP.

7. Foreign Trade Development: Measured by the ratio of a city’s total import and export value to its GDP.

3.2.4 Mediating variables

(1) Optimization of Resource Allocation Efficiency (ora): This paper employs the degree of factor market development from the Fan Gang Marketization Index as the measurement for this mechanism variable. The primary rationale is that a higher degree of factor market development indicates more sufficient market-based allocation of factors in the region, smoother factor mobility, and effectively reflects the level of rational allocation of factor resources.

(2) Innovation Efficiency Upgrading (inn_eff): This paper uses regional innovation activity to measure innovation efficiency upgrading, specifically employing the number of invention patents per 10,000 people in a city as the metric for innovation activity. The main reason is that innovation activity captures changes in regional innovation inputs and outputs, new product development, and other related areas, reflecting the quality, capability, and performance of regional innovation activities.

3.3 Research methods

To avoid the influence of subjective factors, this study uses the entropy weight method to determine the weight of each indicator and calculates the Digital Infrastructure Development Index and the Ecological Resilience Index accordingly. Since the secondary indicators of ecological resilience have different directions and dimensions, and the digital infrastructure indicators also have different dimensions, the range method is first used to standardize the indicators of digital infrastructure and ecological resilience (Equations 2, 3) before measuring the indicator system to eliminate the impact of indicator direction and dimension, making the data comparable. The specific processing method is as follows:

$\begin{array}{ll}\mathit{Yij}=\frac{\mathit{Xij}-min\left(\mathit{Xij}\right)}{max\left(\mathit{Xij}\right)-min\left(\mathit{Xij}\right)},X_{\mathit{ij}}\text{is}\mathrm{a}\text{positive indicator}& (2)\end{array}$

$\begin{array}{ll}\mathit{Yij}=\frac{max\left(\mathit{Xij}\right)-\mathit{Xij}}{max\left(\mathit{Xij}\right)-min\left(\mathit{Xij}\right)},X_{\mathit{ij}}\text{is}\mathrm{a}\text{negative indicator}& (3)\end{array}$

Here, i represents the province, j represents the measurement indicator, and X_ij and Y_ij represent the initial data of digital infrastructure and ecological resilience indicators and the data after standardization, respectively.

Then, calculate the proportion $P_{\mathit{ij}}$ and information entropy $e_j$ of the final standardized indicators in the total of each observation unit using the following equations (Equations 4, 5):

$\begin{array}{ll}{p}_{\mathit{ij}}=\frac{X_{\mathit{ij}}^{\text{'}}}{{\sum }_{i=1}^m{X}_{\mathit{ij}}^{\text{'}}}& (4)\end{array}$

$\begin{array}{ll}{e}_j=-\frac{1}{ln\left(n\right)}{\sum }_{i=1}^m{p}_{\mathit{ij}}ln\left(p_{\mathit{ij}}\right)& (5)\end{array}$

Again, calculate the coefficient of variation $d_j$ and the proportion $w_j$ for each indicator using the following equations (Equations 6, 7):

$\begin{array}{ll}{d}_j=1-e_j& (6)\end{array}$

$\begin{array}{ll}{w}_j=\frac{d_j}{{\sum }_{j=1}^n{d}_j}& (7)\end{array}$

Finally, calculate the index values $x_i$ for different provinces in each year of the Yellow River Basin using Equation (8):

$\begin{array}{ll}{x}_i={\sum }_{j=1}^n{w}_j\times X_{\mathit{ij}}^{\text{'}}& (8)\end{array}$

3.4 Data sources and description

The sample data for this study comprises 75 prefecture-level cities along the Yellow River Basin, covering the period from 2014 to 2023. Data for the control variables were primarily sourced from the China Statistical Yearbook, the China City Statistical Yearbook, the CSMAR database, the EPS database, and the Chinese Research Data Services Platform (CNRDS). A small number of missing values were supplemented using the linear interpolation method. The descriptive statistics for the relevant variables are presented in Table 3.

Table 3

Table 3. Descriptive statistics of variables.

4 Empirical results and analysis

4.1 Baseline regression

To examine the impact of the Digital Infrastructure Construction Index (DI) on Ecological Resilience (ER) in the Yellow River Basin, the baseline regression results based on the specification of econometric model (1) are presented in Table 4. Columns (1) and (2) show the results of the pooled OLS panel regression. Column (1) presents the results without including control variables and without controlling for two-way fixed effects. Column (2) adds the control variables to the specification in column (1). Columns (3) and (4) present the results of the panel regression with two-way fixed effects. Column (3) does not include control variables but controls for both time and region fixed effects. Column (4) introduces the control variables into the baseline model based on column (3). It can be observed that the regression coefficient for Digital Infrastructure Construction (DI) is 0.1341, which is statistically significant at the 1% level. The conclusions from the baseline regression indicate that digital infrastructure construction significantly enhances ecological resilience in the Yellow River Basin. This further underscores the importance of prioritizing digital infrastructure construction in the process of ecological environmental protection within the Yellow River Basin, and comprehensively strengthening the interconnectivity of digital infrastructure across the basin.

Table 4

Table 4. Benchmark regression results.

4.2 Endogeneity tests

The impact of digital infrastructure construction on the ecological resilience of the Yellow River Basin may potentially suffer from endogeneity issues. These issues generally arise from four sources: omitted variables, measurement errors, reverse causality, and sample selection problems. To address omitted variables, this study incorporates additional relevant control variables. For sample selection bias, the issue is mitigated by altering the sample composition. Regarding measurement error, alternative measurement methodologies are employed. Considering that the impact of digital infrastructure construction on the ecological resilience of the Yellow River Basin might be confounded by reverse causality, this study further utilizes an instrumental variable (IV) approach to address endogeneity. To ensure the robustness of the model, the Two-Stage Least Squares (2SLS) method is adopted for endogeneity testing. Regarding the selection of instrumental variables, the interaction term between the total postal service volume of each prefecture-level city in 1984 and the city’s infrastructure supporting fees from the previous year is used as the first instrumental variable (iv1). In terms of relevance, postal infrastructure was essential for dial-up internet access in the early days; therefore, cities with historically higher postal service volumes are more likely to be areas with better digital infrastructure development, satisfying the relevance condition. Regarding exogeneity, since the number of post offices in 1984 is a historical fixed value, it is unlikely to directly affect the current ecological resilience of the Yellow River Basin, thus satisfying the exogeneity assumption. However, as the city-level postal service volume in 1984 is a fixed value, to incorporate a time trend, this study constructs an interaction term between each city’s 1984 postal service volume and the one-period lagged municipal infrastructure supporting fees, using this as the instrumental variable (iv1) for the 2SLS regression. Furthermore, to ensure the robustness of the empirical results, this study also employs the number of telephones in each prefecture-level city in 1984 interacted with the one-period lagged city infrastructure supporting fees as a second instrumental variable (iv2) and reapplies the 2SLS method for empirical testing. The test results are shown in Table 5. Columns (1) to (4) present the results of the two-stage regression. Based on columns (1) and (3), for the null hypothesis of “underidentification,” the Kleibergen-Paap rk LM statistic p-values are significant at the 5% level, rejecting the null hypothesis. According to the regression results in columns (2) and (4), in the test for weak identification, the Kleibergen-Paap rk Wald F statistic exceeds the Stock-Yogo weak identification test critical value at the 10% level, indicating that the selection of instrumental variables is reasonably valid. Furthermore, results from columns (1) and (3) show that the endogenous variable is significantly positively correlated with the instrumental variables, satisfying the relevance condition. Simultaneously, the regression results in columns (2) and (4) remain significantly positive, indicating that digital infrastructure construction has a significant promoting effect on the ecological resilience of the Yellow River Basin, thereby corroborating the conclusions of the baseline model.

Table 5

Table 5. Endogeneity test regression results.

4.3 Robustness tests

To further enhance the reliability of the baseline regression results, this study conducts verifications from four aspects: changing the measurement method, controlling for omitted variables, replacing the data sample, and examining the lagged effects of digital infrastructure construction.

4.3.1 Changing the measurement method

The study further employs the “vertical and horizontal scatter degree” method to recalculate the level of digital infrastructure construction in the Yellow River Basin. Compared to methods such as the Analytic Hierarchy Process, Entropy Method, and Principal Component Analysis, the “vertical and horizontal scatter degree” method not only captures the characteristics of three-dimensional time-series data but also processes underlying data through bottom-up, layer-by-layer refinement, making the calculation results more comprehensive and scientific. Therefore, this study uses this comprehensive evaluation method to recalculate the level of digital infrastructure construction in the Yellow River Basin, obtaining a new Digital Infrastructure Construction Index. Table 6 presents the regression results using the “vertical and horizontal scatter degree” method to measure digital infrastructure construction. According to the regression results in column (1), the coefficient for digital infrastructure construction remains significantly positive at the 1% significance level, indicating that digital infrastructure construction has a significant promoting effect on the ecological resilience of the Yellow River Basin. Thus, the core finding of this study remains robust.

Table 6

Table 6. Robustness test regression results.

4.3.2 Controlling for omitted variables

Additional control variables are included to mitigate the impact of other potential channels affecting the ecological resilience of the Yellow River Basin. First, the technological level is controlled. Scientific research and development are not only an important driver of economic growth but also a favorable means of promoting ecological environmental protection, thereby enhancing the coupling coordination between high-quality development and ecological environmental protection in the Yellow River Basin. Therefore, the technological level (tech) is added to the regression model as a control variable to address potential omitted variable bias, where the technological level is measured by the share of science and technology expenditure in each prefecture-level city’s GDP. Second, the educational level is controlled. Improving the educational level can cultivate high-end production factors, and the optimal reallocation of factor resources contributes to promoting ecological environmental protection in the Yellow River Basin. Therefore, based on the baseline regression model, the educational level at the city level (edu) is included in the control variables to mitigate omitted variable bias and enhance the reliability of the regression results. Column (2) of Table 6 reports the relevant regression results. After controlling for the technological level (tech), educational level (edu), and other control variables, the regression coefficient for digital infrastructure construction in the Yellow River Basin remains significantly positive. This indicates that even after controlling for other potential omitted variables, digital infrastructure construction can still significantly enhance the ecological resilience of the Yellow River Basin, consistent with the conclusions of the baseline regression.

4.3.3 Replacing the data sample

Among the cities along the Yellow River Basin, provincial capitals are national economic centers with more advanced digital infrastructure construction, possessing absolute advantages in development, and their construction levels rank at the forefront nationally. Therefore, to verify that the promoting effect of digital infrastructure construction on the ecological resilience of the Yellow River Basin is more inclusive, the sample of provincial capitals is excluded. The baseline regression model is then reapplied to analyze digital infrastructure construction in prefecture-level cities of other provinces. According to the regression results in column (3) of Table 6, after replacing the data sample, digital infrastructure construction still has a promoting effect on the ecological resilience of the Yellow River Basin. Thus, the conclusions of this study remain robust.

4.3.4 Examining the lagged effects of digital infrastructure construction

The impact of digital infrastructure construction on ecological resilience often exhibits sustainability and persistence, and this process may involve certain time lags. Therefore, to more accurately identify the persistent impact of digital infrastructure construction on the ecological resilience of the Yellow River Basin—specifically, to determine whether digital infrastructure construction in a given year affects the ecological resilience of the Yellow River Basin in the next one to two years—this study re-estimates the regressions using the one-period and two-period lagged Digital Infrastructure Construction Index as the core explanatory variable. Columns (4) and (5) of Table 6 show the impacts of digital infrastructure construction on the ecological resilience of the Yellow River Basin with lags of one and two years, respectively. The results indicate that regardless of a one-period or two-period lag, the regression coefficients for the overall level of digital infrastructure construction (DI) remain significantly positive. Moreover, the positive impact is slightly higher with a one-year lag compared to a two-year lag. This result suggests that the impact of digital infrastructure construction on the ecological resilience of the Yellow River Basin has a long-term driving effect.

5 Further analysis

Based on the preceding analysis, this study further investigates the mechanisms, spatial spillover effects, and heterogeneous impacts of digital infrastructure construction on the ecological resilience of the Yellow River Basin, aiming to provide an in-depth analysis of its operational channels and differential influences from diverse perspectives.

5.1 Mechanism tests

5.1.1 Resource allocation optimization effect

Theoretical analysis indicates that digital infrastructure construction can reduce transaction costs, promote market integration, and enhance resource accessibility and allocation efficiency, thereby improving the ecological resilience of the Yellow River Basin. Column (1) of Table 7 presents the impact of digital infrastructure construction on the degree of factor market development. It can be observed that the regression coefficient for digital infrastructure construction is 3.2740, significant at the 1% level. This indicates that digital infrastructure construction can effectively enhance the development of factor markets, and also demonstrates that digital infrastructure construction can drive the improvement of ecological resilience in the Yellow River Basin by optimizing the allocation of factor resources, thus validating Hypothesis I.

Table 7

Table 7. Regression results of mechanism test.

5.1.2 Innovation efficiency upgrading effect

Based on the preceding theoretical analysis, digital infrastructure construction can accelerate the integration of innovative knowledge, enhance regional innovation efficiency, and achieve innovation-driven improvement of the ecological resilience of the Yellow River Basin. Column (2) of Table 7 shows the impact of digital infrastructure on regional innovation activity. According to the regression results, the coefficient of digital infrastructure construction on regional innovation activity is significantly positive at the 1% statistical level, indicating that digital infrastructure construction significantly boosts regional innovation vitality. This also proves that digital infrastructure construction can drive the ecological resilience of the Yellow River Basin by improving the level of regional innovation efficiency, thus validating Hypothesis II.

5.2 Spatial effect test

As previously discussed, digital infrastructure construction enhances inter-city collaborative cooperation. In particular, the extensive coverage of digital infrastructure continuously strengthens the exchange of information, resources, and personnel between regions. Consequently, digital infrastructure construction exhibits significant positive spatial spillover effects, thereby intensifying the spatial linkage effects among cities in the Yellow River Basin and better promoting the enhancement of the Basin’s ecological resilience. To verify the spatial linkage and spillover effects, this study introduces an economic-geographic spatial weight matrix based on spatial econometric methods to explore the potential impact of the spatial spillover effects of infrastructure construction on the ecological resilience of the Yellow River Basin. The specific regression results are shown in Table 8. The decomposed direct effect and indirect effect regression coefficients are both significantly positive. This indicates that digital infrastructure construction can not only promote the improvement of ecological resilience within the local region but also enhance the ecological resilience of neighboring areas in the Yellow River Basin through spatial linkage and spillover effects. Thus, Hypothesis 3 of this study is validated.

Table 8

Table 8. Spatial linkage spillover effect regression results.

5.3 Heterogeneity analysis

5.3.1 Heterogeneity across upper, middle, and lower reaches

In reality, the resource endowments and levels of economic development vary significantly across the upper, middle, and lower reaches of the Yellow River Basin, and the ecological and environmental challenges faced by each segment also differ. Consequently, both the level of digital infrastructure construction and the ecological resilience of the Yellow River Basin exhibit distinct heterogeneous characteristics in their spatial distribution along the river. Therefore, to delve deeper into the impact of digital infrastructure construction on ecological resilience in different segments of the Yellow River Basin, and with reference to the natural river basin boundaries defined in the Comprehensive Plan for the Yellow River Basin (2012–2030) by the Yellow River Conservancy Commission of the Ministry of Water Resources, this study divides the Basin into the upper, middle, and lower reaches using Hekou Town in Inner Mongolia and Taohuayu in Henan Province as nodal points. Separate regressions are conducted for these sub-samples. The sample includes 24 cities in the upper reaches, 30 cities in the middle reaches, and 21 cities in the lower reaches. The regression results presented in Table 9 show that, compared to the middle reaches, digital infrastructure construction has a greater promoting effect on the ecological resilience of the Yellow River Basin in the upper and lower reaches. The reasons for this finding may be as follows: In the upper reaches, leveraging superior natural resources and national policy support, the marginal promoting effect of digital infrastructure construction is more pronounced. In the lower reaches, the economy is developed, and the industrial base is solid, providing the talent and market conditions necessary for adopting digital technologies, which allows for a more complete realization of the dividends from digital infrastructure construction. In contrast, the middle reaches have a higher proportion of traditional heavy industries, making the digital transformation more challenging; thus, the impact of digital infrastructure is less pronounced.

Table 9

Table 9. Heterogeneity test regression results (I).

5.3.2 Heterogeneity based on renewable energy abundance

Cities along the Yellow River Basin possess abundant renewable energy resources (Zheng et al., 2023). The most significant ecological constraint facing the Yellow River Basin is water resources. The water-saving effect achieved through renewable energy utilization can conserve substantial amounts of industrial water use. This is crucial for ensuring the continuous flow of the Yellow River, restoring downstream wetlands, and maintaining the basic ecological functions of the river course, directly enhancing the ecological resilience of the basin’s ecosystem. Therefore, to verify whether the level of renewable energy abundance influences the promoting effect of digital infrastructure construction on ecological resilience in the Yellow River Basin, and based on the Renewable Energy Data Handbook 2015 from the National Renewable Energy Center, cities along the Yellow River Basin are categorized into renewable energy-abundant cities and renewable energy-scarce cities according to their renewable energy reserves. The sample includes 19 renewable energy-abundant cities and 56 renewable energy-scarce cities. The specific regression results are shown in Table 10. The results indicate that in renewable energy-abundant cities, the promoting effect of digital infrastructure construction on the ecological resilience of the Yellow River Basin is more pronounced. This suggests, to some extent, that accelerating digital infrastructure construction can help facilitate industrial transformation and upgrading in renewable energy-abundant cities within the Yellow River Basin, thereby driving the improvement of the Basin’s ecological resilience. It further demonstrates the necessity to expedite digital infrastructure construction in the Yellow River Basin to leverage its “dividend effect” on enhancing ecological resilience.

Table 10

Table 10. Heterogeneity test regression results (II).

5.3.3 Heterogeneity analysis based on time windows

In 2016, the state issued the *National Informatization Development Strategy Outline, which defined the scope of information infrastructure construction for the new era, primarily including digital infrastructure such as data centers, cloud computing, and the Internet of Things. Evidently, based on new-generation information technology, digital infrastructure has become a vital support for the digital transformation of the economy and society and for high-quality development, while also providing technical support for the digitalization, networking, and intellectualization of traditional infrastructure like transportation and energy. Consequently, to further investigate the temporal heterogeneous impact of digital infrastructure construction on the ecological resilience of the Yellow River Basin, this section divides the sample period into two time windows using 2016 as the cutoff point. The regression results are presented in Table 10. It can be observed that from 2017 to 2023, the regression coefficient for digital infrastructure construction is significantly higher than in the earlier period. This indicates that the impact of digital infrastructure construction on the ecological resilience of the Yellow River Basin exhibits temporal heterogeneity. Simultaneously, it proves that the promoting effect of digital infrastructure construction on the ecological resilience of the Yellow River Basin generally shows an increasing trend over time.

6 Conclusions and policy implications

6.1 Conclusion

This study, from the perspective of digital infrastructure construction in the Yellow River Basin, investigates its impact on the Basin’s ecological resilience and the underlying driving mechanisms. Utilizing panel data from 75 prefecture-level cities along the Yellow River Basin from 2013 to 2022, and employing panel fixed-effects models and mechanism testing models for empirical analysis, the reliability of the conclusions is enhanced through a series of robustness tests. Furthermore, the study examines the heterogeneous effects of digital infrastructure construction on ecological resilience in the Yellow River Basin. The main findings are as follows:

First, during the sample period, digital infrastructure construction in the Yellow River Basin significantly promoted the enhancement of its ecological resilience. This conclusion holds after a series of robustness checks, including the instrumental variable approach, replacing measurement methods, controlling for omitted variables, changing the data sample, and examining the lagged effects of digital infrastructure construction.

Second, mechanism analysis reveals that digital infrastructure construction drives the improvement of ecological resilience in the Yellow River Basin through resource allocation optimization effects and innovation efficiency upgrading effects. Additionally, digital infrastructure construction exhibits spatial spillover effects, meaning it can not only enhance ecological resilience in the local region but also promote ecological resilience in neighboring areas of the Yellow River Basin.

Third, further heterogeneity analysis indicates that: Regarding heterogeneity across the upper, middle, and lower reaches, digital infrastructure construction has a more significant impact on ecological resilience in the upper and lower reaches. In contrast, the middle reaches, characterized by a higher proportion of traditional industrial structure and bearing multiple tasks such as ecological governance and economic transformation with constrained resource allocation, experience a relatively lower promoting effect from digital infrastructure. In terms of renewable energy heterogeneity, digital infrastructure construction is more conducive to enhancing ecological resilience in cities relatively abundant in renewable energy within the Basin. Regarding time window heterogeneity, the enhancing effect of digital infrastructure construction on the ecological resilience of the Yellow River Basin generally shows an increasing trend over time.

6.2 Policy implications

Based on the findings, this study proposes the following policy recommendations: First, implement differentiated digital infrastructure layouts. The upper reaches should strengthen the construction of ecological monitoring networks and data centers to solidify the ecological security barrier of the Yellow River Basin. The lower reaches require intelligent upgrades to flood control and water dispatch systems, focusing on preventing and controlling soil erosion in the Basin. The middle reaches should prioritize the digital and green transformation of their traditional industries as a starting point, creating space for digital technology empowerment. Second, prioritize supporting the digital transformation of renewable energy-abundant cities. Cities rich in renewable resources should be designated as pilot demonstration zones for the coordinated development of digital infrastructure and ecological governance, receiving preferential treatment in project layout, funding investment, and technical support to fully unlock their potential for enhancing ecological resilience. Third, establish a long-term advancement mechanism. Continuously increase investment in digital infrastructure, dynamically track the effectiveness of technology applications, and break down administrative barriers through cross-provincial collaboration mechanisms. This will ensure the sustained release of the increasing marginal effects of digital infrastructure on ecological resilience, facilitating the synergistic advancement of ecological protection and high-quality development in the Basin.

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Author contributions

WZ: Data curation, Investigation, Validation, Writing – original draft, Writing – review & editing.

Funding

The author(s) declare that financial support was received for the research and/or publication of this article. This study was supported by the Major Humanities and Social Sciences Research Projects in Zhejiang higher education institutions (2024QN045) and the Quzhou Municipal Social Science Planning Project (25QSKG41LX).

Conflict of interest

The author declares that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author declares that no Gen AI was used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Abula, K., and Aihemaiti, Y. (2025). Digitalization and culture–tourism integration in China: the moderated mediation effects of employment quality, infrastructure, and new-quality productivity. Sustainability 17:8792. doi: 10.3390/su17198792

Crossref Full Text | Google Scholar

Acemoglu, D., and Restrepo, P. (2019). Automation and new tasks: how technology displaces and reinstates labor. J. Econ. Perspect. 33, 3–30. doi: 10.1257/jep.33.2.3

Crossref Full Text | Google Scholar

Czernich, N., Falck, O., and Kretschmer, T. (2011). Broadband infrastructure and economic growth. Econ. J. 121, 505–532. doi: 10.1111/j.1468-0297.2011.02420.x

Crossref Full Text | Google Scholar

Henfridsson, O., and Bygstad, B. (2013). The generative mechanisms of digital infrastructure evolution. MIS Q. 37, 907–931. doi: 10.25300/MISQ/2013/37.3.11

Crossref Full Text | Google Scholar

Huang, M. H., Zhang, W. G., and Lan, X. J. (2025). Research on the impact of digital infrastructure construction on urban economic resilience. Stat. Decis. 41, 101–106 (in Chinese). doi: 10.13546/j.cnki.tjyjc.2025.08.017

Crossref Full Text | Google Scholar

Jiang, D., Zhu, W., and Zhang, Z. (2025). The spatiotemporal coupling and synergistic evolution of economic resilience and ecological resilience in Africa. Sustainability 17:863. doi: 10.3390/su17030863

Crossref Full Text | Google Scholar

Jin, S. R., Qi, X. H., and Tang, S. Y. (2025). Can digital infrastructure construction improve the ESG performance of agricultural enterprises? Quasi-natural experiment based on the "broadband China" strategy. J Agrotech Econ 2025, 127–144. doi: 10.13246/j.cnki.jae.20250623.002

Crossref Full Text | Google Scholar

Li, Q., Ge, J., Zhang, X., Wu, X., Fan, H., and Yang, L. (2025). Assessment of the interaction between digital infrastructure and ecological resilience: insights from Yangtze River delta urban agglomeration in China. J. Clean. Prod. 486:144364. doi: 10.1016/j.jclepro.2024.144364

Crossref Full Text | Google Scholar

Li, G., and Wang, L. (2023). Study of regional variations and convergence in ecological resilience of Chinese cities. Ecol. Indic. 154:110667. doi: 10.1016/j.ecolind.2023.110667

Crossref Full Text | Google Scholar

Liu, N. (2025). Digital infrastructure, technological innovation and urban economic resilience. Statist. Decis. 41, 101–106. doi: 10.13546/j.cnki.tjyjc.2025.12.017

Crossref Full Text | Google Scholar

Liu, M. F., Li, X. Y., and Zhou, C. J. (2025). China's city-level digital infrastructure and its "carbon reduction and efficiency enhancement" effect. Econ. Geogr. 45, 20–30. doi: 10.15957/j.cnki.jjdl.2025.08.003

Crossref Full Text | Google Scholar

Ma, Y., Zheng, M., Xu, F., Qian, Y., Liu, M., Zheng, X., et al. (2023). Modeling and exploring the coordination relationship between green infrastructure and land use eco-efficiency: an urban agglomeration perspective. Environ. Sci. Pollut. Res. 30, 92537–92554. doi: 10.1007/s11356-023-28841-x,

PubMed Abstract | Crossref Full Text | Google Scholar

Manny, L., Angst, M., Rieckermann, J., and Fischer, M. (2022). Socio-technical networks of infrastructure management: network concepts and motifs for studying digitalization, decentralization, and integrated management. J. Environ. Manag. 318:115596. doi: 10.1016/j.jenvman.2022.115596,

PubMed Abstract | Crossref Full Text | Google Scholar

Miao, C. H., and Zhang, B. F. (2021). Zoning and classification regulation strategies of high-quality development in the Yellow River Basin. Econ. Geogr. 41, 143–153. doi: 10.15957/j.cnki.jjdl.2021.10.016

Crossref Full Text | Google Scholar

Redding, S. J., and Turner, M. A. (2015). Transportation costs and the spatial organization of economic activity. Handb. Regl. Urban Econ. 5, 1339–1398. doi: 10.1016/B978-0-444-59531-7.00020-3

Crossref Full Text | Google Scholar

Shen, K. R., Lin, J. W., and Fu, Y. H. (2023). Network infrastructure construction, information availability, and firm innovation boundary. China Industr. Econ., 57–75. doi: 10.19581/j.cnki.ciejournal.2023.01.014

Crossref Full Text | Google Scholar

Sheng, Y. C., Chen, X., Zhang, Y. F., Wang, R. X., Li, Q., et al. (2025). The mechanism of ecological resilience promoting high-quality economic development in the Yellow River Basin. Environ. Sci., 1–18. doi: 10.13227/j.hjkx.202503345

Crossref Full Text | Google Scholar

Tsekouras, K., Chatzistamoulou, N., Kounetas, K., and Broadstock, D. (2016). Spillovers, path dependence and the productive performance of European transportation sectors in the presence of technology heterogeneity. Technol. Forecast. Soc. Change 102, 261–274. doi: 10.1016/j.techfore.2015.09.008

Crossref Full Text | Google Scholar

Vial, G. (2021). Understanding digital transformation: a review and a research agenda. Manag. Digital Transf. 2021, 13–66. doi: 10.1016/j.jsis.2022.101720

Crossref Full Text | Google Scholar

Wang, H., Yan, Z. Y., Guo, G. Y., Zhang, X. F., and Yin, J. Y. (2023). Digital infrastructure policy and corporate digital transformation: "empowerment" or "disablement"? J. Quant. Tech. Econ. 40, 5–23. doi: 10.13653/j.cnki.jqte.20230314.002

Crossref Full Text | Google Scholar

Wang, Y., Zhang, X. F., and Lei, S. Z. (2024). Empirical test on the coupling coordination degree of digital economy, industrial development and ecological environment in the Yellow River Basin. Statist. Decis. 40, 108–113. doi: 10.13546/j.cnki.tjyjc.2024.04.019

Crossref Full Text | Google Scholar

Xu, L., Li, X., Deng, X., and Zhang, K. (2025). Digital infrastructure and total factor carbon productivity: evidence from China's broadband China pilot. Environ. Dev. Sustain. 2025, 1–29. doi: 10.1007/s10668-025-07013-6,

PubMed Abstract | Crossref Full Text | Google Scholar

Yan, X. N. (2025). The effect of digital infrastructure on corporate innovation: from the perspective of new quality productivity. Bus. Manag. J. 47, 40–58. doi: 10.19616/j.cnki.bmj.2025.08.003

Crossref Full Text | Google Scholar

Yuan, Y., Bai, Z., Zhang, J., and Xu, C. (2022). Increasing urban ecological resilience based on ecological security pattern: a case study in a resource-based city. Ecol. Eng. 175:106486. doi: 10.1016/j.ecoleng.2021.106486

Crossref Full Text | Google Scholar

Zhang, T., and Li, J. C. (2023). Network infrastructure, inclusive green growth and regional disparity: causal inference based on double machine learning. J. Quant. Tech. Econ. 40, 113–135. doi: 10.13653/j.cnki.jqte.20230310.005

Crossref Full Text | Google Scholar

Zhang, K. Y., and Zhang, Y. (2020). The evolution of regional economic differences in the Yellow River Basin at different spatial scales. Econ. Geogr. 40, 1–11. doi: 10.15957/j.cnki.jjdl.2020.07.001

Crossref Full Text | Google Scholar

Zheng, J., Chen, Z., Zhang, T., Huang, X., and Wang, X. (2023). Regional sustainable and renewable energy development in China: a comprehensive assessment and influencing factors. Energy Rep. 9, 76–80. doi: 10.1016/j.egyr.2023.09.119

Crossref Full Text | Google Scholar

Zhou, W. H., and Chao, X. J. (2023). Analysis on the coupling coordination development of digital infrastructure, economic development resilience and ecological environment protection in the Yellow River Basin: based on the ternary system coupling coordination model. J. Arid Land Resourc. Environ. 37, 1–9. doi: 10.13448/j.cnki.jalre.2023.203

Crossref Full Text | Google Scholar

Zhou, W. H., and Xu, C. Y. (2025). Measurement of the coupling coordination level and interactive response between new quality productivity and ecological resilience in the Yellow River Basin. Arid Zone Res. 2025, 1–12. doi: 10.3724/SP.J.1148.2025.05123

Crossref Full Text | Google Scholar