Ecological Resilience

According to the definition of urban resilience [29], ecological resilience(ER) refers to the ability of ecosystems to adapt, change, and enhance the status quo while maintaining or rapidly restoring desired functions in the face of disturbances. Thus, ecological resilience consists of three main features: resistance, adaptation, and resilience. Resilience represents the ability of ecosystems to withstand external disturbances and is expressed in terms of five indicators: total sulfur dioxide emissions (X11), total chemical oxygen demand (COD) emissions (X12), total smoke (dust) emissions (X13), direct economic losses from natural disasters (X14), and energy consumption per unit of GDP (X15). Resilience indicates the ability of the ecosystem to maintain stability, which mainly includes the urban sewage treatment rate (X21), air quality excellence rate (X22), the comprehensive utilization rate of industrial solid waste (X23), and harmless treatment rate of domestic garbage (X24), and the potential of the ecosystem to recover to the pre-destructive state after being impacted by the impacts, which mainly includes the number of water resources per capita (X31), the green coverage of built-up areas (X32), park green space per capita (X33), and the strength of environmental protection expenditures (X34). (X32), green space per capita (X33), and environmental protection expenditure (X34) (Table 1).

Among the data of X11-X14, X21 is from the China Statistics Yearbook on Environmental (https:/data.cnki. net/yearBook/single?id=N2023070120) issued by China Economic and Social Big Data Research Platform, X15 is from provincial statistical yearbooks, and X22 are from provincial Bulletin of Environmental Condition. X23-X34 data from the China Statistical Yearbook published by the China Statistics Bureau (http://www. stats.gov.cn/sj/ndsj/).

Green Technology Innovation Efficiency

Green technological innovation efficiency (GTIE) is an important indicator of regional technological innovation capacity, and this paper considers the manpower, capital, and innovation activities of different provinces in technological innovation as input indicators and the number of green patents granted as output indicators (Table 2). in which, the number of green patents authorized includes the number of green invention patents authorized and the number of green utility model patents authorized, and the data comes from CNRDS (https:/www.cnrds.com/Home/Login).

Industry Digitization Efficiency

This paper constructs an industrial digitization efficiency measurement system (Table 3). The relationship between industrial digitization infrastructure construction inputs and economic outputs, taking the manpower, equipment, enterprise, and Internet infrastructure construction involved in the process of industrial digitization as the input indicators [30, 31], and the software revenue and e-commerce sales as the output indicators, where the software revenue includes the software product revenue and the software business revenue. Data from the China Statistical Yearbook (http://www.stats.gov.cn/sj/ndsj/).

Coupled Coordination Model

The coupled coordination model describes the interactions between subsystems in a system, accurately reflects the degree of coupling and dynamic changes between related systems, and reflects the coordination effect between systems while capturing the integrated development of related systems [32]. The industrial digitalization development and green technology innovation of socio-economic systems are affected by regional natural resource endowments, and thus there is a certain degree of coupling between ecological resilience, industrial digitalization, and green technology innovation. The IDE, GTIE, and ER coupling coordination degree models are constructed by referring to the way of modification of the traditional coupling degree model by MENG [33]. The model is described as follows:

where D_t denotes the coupling coordination degree (CCD) in year t, categorized according to Table 4, C_t denotes the coupling degree in year t, T_t denotes the coordination degree in year t, and ω_iis the weight of system i according to CRITIC. X_it, X_jt denotes the level value of system i,j in year t, and n is the number of systems.

Dagum Gini Coefficient

In terms of analyzing the internal and external differences between different regions in the degree of coupling coordination, according to the geographical division of China, the internal and external differences of different regions are analyzed through the Dagum Gini coefficient, which is an improvement of the traditional Gini coefficient, making up for the shortcomings of other methods used to measure regional disparities due to the failure to address the overlap of the data under investigation and better identifying the sources of regional disparities. The model formula is as follows:

Where G denotes the overall Gini coefficient, which measures the total difference between regions. m denotes the number of regions. n_i,n_j denote the number of provinces within region i(j), (i,j=1,2, …,m), n denotes the total number of provinces, y_ih, y_ir indicates the level of dynamic changes in the CCD of each province, and ȳ denotes the average value of CCD in all provinces.

G can be decomposed into the coefficient of variation within groups G_w, the coefficient of variation between groups G_b, and the coefficient of hypervariable density G_t, which satisfies the relationship G = G_w + G_b +G_t. Among them, G_w denotes the distributional difference within region i, and G_b denotes the distributional difference between regions i, j, and G_t reflecting the effect of the phenomenon of overlap of the regions on the overall Gini coefficient G. The specific coefficient decomposition equation is as follows. The specific coefficient decomposition equations are as follows:

Eq. (5) measures the contribution of differences in the dynamic level of CCD within region i to the total Gini coefficient, where, λi=ni/n, si=λiy―i/y―. G_ii indicates the intra-region Gini coefficient, as computed by Eq. (6).

Eqs. (7) and (8) measure the contribution of inter-area CCD differences and the intensity of inter-area differences to the total Gini coefficient, respectively. Where Gij denotes the inter-area Gini coefficient, which is calculated by Eq. (9), D_ij denotes the relative impact between areas i,j, which is calculated by Eq. (10), and (1-D_ij) denotes the hypervariance density.

Where d_ij, p_ij denote the difference in CCD between regions i, j. d_ij can be interpreted as the mathematical expectation of the sum of the sample values of y_ih-y_jr>0 in regions i, j. p_ij can be regarded as the mathematical expectation of the sum of sample values of y_jr-y_ih >0 in regions i, j.

where, F_i(j)(y)denotes the cumulative distribution function for region i(j).

Spatial Econometric Model

In order to comprehensively consider the coordinated development of industrial digitalization, green technology innovation, and ecological resilience, a spatial econometric regression model containing industrial digitalization, green technology innovation, and ecological resilience is constructed:

where y_it is the explanatory variable CCD, y_(I,t-1) is the first-order lag, d_i^’X_iδ represents the spatial lag term of the explanatory variable, and d_i^’ is the ith row of the spatial weight matrix w. u_i is the spatial effect, γ_i is the time effect, ε_it is the error term, β is the coefficient to be assessed, and τ,ρ,δ,λ indicate the parameters of the model determination test, which are passed by the LR, LM, and Hausman tests, A spatial Durbin model if λ = 0, the spatial lag model if λ = δ = 0, and then a spatial lag model if τ = ρ = δ = 0.

Simultaneously, in a bid to determine the applicability of the model, the condition of significant spatial correlation needs to be fulfilled when performing spatial measurements, which is ascertained by means of the global Moran’s index (Morans’ I):

where n is the number of districts, wij means the weight matrix obtained from Eq. (14), and xi refers to the coupling coordination level of the ith district. x― represent the average value of the coupling coordination level of the whole.

Fig. 2.

China's 30 Provinces Coupling Coordination Degree, 2011-2021.

Provincial Coupling Coordination Degree

The IDE-GTIE coupling coordination degree (Fig. 2a) measurement results show that all provinces and cities show a steady upward trend during the study period, among which the coupling coordination degree of Beijing is higher, from 0.5388 (barely coordinated) to 0.8623 (good coordinated) during the study period, and the other provinces and cities are from severe disorders, moderate disorders to on the verge of disorders, and quantitatively the results show that, in the process of industrial Chinese provinces, they are committed to moving from energy-driven to technology-driven. At the end of the study period, the coupling coordination of Shandong, Jiangsu, and Guangdong provinces decreased due to the impact of COVID-19.

The overall performance of the IDE-ER system coupling (Fig. 2b) in each province is good, and at the end of the study period, most of the provinces are in the barely coordinated as well as the elementary coordinated stage, but the coupling coordination degree is low in Heilongjiang, Jilin, Gansu, Ningxia, Xinjiang, Yunnan, and Guizhou. Of which, Gansu, Ningxia, and Xinjiang improved from 0.1376 and 0.1247 (severe disorder) to 0.3037.0.3371 (mild disorder) during the study period, showing a good upward trend to a certain extent, but Heilongjiang, Jilin, Yunnan, and Guizhou changed less during the study period, mainly due to the fact that the industrial foundation of these regions is poorer, and the traditional industries account for a relatively large, the level of industrial digitalization development is low, the radiation capacity for the ecological environment is weak, and the degree of interactive coupling influence is low.

In comparison, the coupling coordination of GTIE-ER in each province (Fig. 2c) performed poorly but trended positively, with most of the provinces being in the stage of severe disorder, moderate disorder at the beginning of the study period, and mild disorder at the end of the study period, which indicates that the interaction coupling of GTIE-ER is developing in a benign direction. Among the 30 provinces, GTIE-ER has a high coupling coordination degree in regions with better economic development, such as Beijing, Shanghai, and Shandong, and at both, it also has a high coupling coordination degree in regions with a lower level of economic development, like Gansu, Qinghai, and Xinjiang. It shows that the interactive coupling of green technology innovation and ecological resilience has a "U" trend in the economic perspective, i.e., in the lower stage of economic development, the region based on the local natural conditions of the resources of the traditional industrial economic activities, and thus most of the innovation activities are related to the ecological resources. In the mid-stage of economic development, the industry's dependence on natural resources is weakened, and the green technology innovation activities around the industry also reduce the interactive coupling relationship with the ecology. When economic development reaches a certain level, some of the benefits of industrial development empower the ecological environment, and the innovation activities also strengthen the interactive coupling relationship with the ecology.

According to the measurement results (Fig. 2d), the coupling coordination degree of IDE-GTIE-ER in most of the provinces showed a certain upward tendency, with lower changes in Beijing, Shanghai, Heilongjiang, Shandong, and Guangdong, while the coupling coordination degree of Shandong and Guangdong slightly decreased at the end of the study period due to the COVID-19. In Beijing, the IDE and GTIE both remain high in the study period, while the ER is low, and all three are relatively stable in the study period, so the improvement of the level of coupling coordination degree in Beijing requires the improvement of ecological resilience. Shanghai stays in the stage of mild disorder during the study period. As a more economically developed region in China, it has good economic conditions in industrial digitization and technological innovation, but Shanghai's industry and manufacturing industry account for a lower proportion of the economic system, and the degree of ecological dependence is lower, so the degree of interaction between industrial digitization and green technological innovation and ecological resilience is weaker. Heilongjiang has lower industrial digitalization efficiency and green technology innovation efficiency during the research period, i.e., the utilization rate of industrial digitalization infrastructure and green technology innovation resource inputs in Heilongjiang is relatively low, which is fundamentally due to the poorer foundation of industrial development and fewer modern industries. Shandong and Guangdong show an upward trend in IDE during the study period, and ER maintains the average level in China (0.4024), but the innovation efficiency of GTIE is lower.

In general, the average level of coupled coordination of IDE, GTIE, and ER systems is shown as IDE-ER > IDE-GTIE > IDE- GTIE -ER > GTIE -ER. i.e., in the process of system evolution, the IDE is most closely related to the ER, and China, which mainly focuses on the real economy, has a high degree of relevance to the ecological resources during the process of industrial development, and on the basis of this, an industrial digitization transformation will likewise have a high connection with the ecological environment. In the context of strengthening the construction of ecological civilization, the industrial digitization transformation needs green technology innovation to provide motive power, while the industrial digitization provides direction and platform for green technology innovation, as a result of which IDE has a high interaction effect with GTIE and ER in the process of development, and the synergistic development level of GTIE and ER is the lowest. On the one hand, technological innovation is enterprise-oriented, and even if it is green technological innovation, most of it is green technological research for the improvement of traditional processes, with less technological research and development for the ecological field. On the other hand, technological research and development and popularization have a lagging effect, and green technology has a weaker direct impact on the ecological environment in the premise that it fails to give full play to the benefits of scale. On the one hand, technological innovation is enterprise-oriented, and even if it is green technological innovation, most of it is green technological research for the improvement of traditional processes, with less technological research and development for the ecological field. On the other hand, technological research and development and popularization have a lagging effect, and green technology has a weaker direct impact on the ecological environment in the premise that it fails to give full play to the benefits of scale. The average level of co-evolution of the IDE- GTIE -ER system is higher than that of GTIE -ER, indicating that the industrial system contributes to the improvement in the level of technological and ecosystem co-evolution.

Regional Coupling Coordination Degree

In order to promote coordinated regional development, 30 provinces were divided into four regions according to China's regional planning [40], and the coupling coordination degree of each region was measured (Fig. 3).

Fig. 3.

China's coupling coordination degree by region, 2011-2021.

In the research period, the coupling coordination degree of IDE-GTIE-ER in the eastern region (Fig. 3a) fluctuates between the verge of disorder and barely coordination. According to the results of the research on the coupling coordination degree among IDE, GTIE, and ER, the overall stability of IDE-GTIE and IDE-ER is good and the coupling coordination degree is good, while the GTIE-ER coupling degree is low and volatile, with the change interval Between 0.3560-0.5735, which shows that the lower coupling degree in the eastern region is mainly due to the efficiency of technological innovation, mainly due to the fact that although the eastern region has good technological innovation resources, the proportion of green technological innovation patents in the authorized patents is relatively low, which leads to the lower efficiency of green technological innovation.

The coupling coordination degree of the IDE-GTIE-ER system in the middle region shows an upward trend, with an average annual growth rate of 1.31%. According to Fig. 3b, among the coupling coordination degrees of the subsystems in the middle region, the GTIE-ER coupling degree is low, but unlike the eastern region, which faces pressure from factor cost and international competition, the middle region mainly faces the quantitative crude economy, especially in Shanxi, Henan, and Anhui agricultural provinces, where labor-intensive industries occupy a large proportion and the problem of insufficient green innovation endowment is more prominent.

Based on Fig. 3c, the coupling coordination degree of the IDE-GTIE-ER system in the western region shows an upward trend with an average annual growth rate of 1.7%, in which the coupling coordination degree of the IDE- ER subsystem is lower than that of other subsystems. Compared with other regions in China, the western region has poorer conditions for overall economic development, and the industrial structure is mainly dominated by the energy industry and mineral resource extraction and processing industry. Due to environmental constraints, industrial development in the western region is strongly correlated with regional resources, and the level of industrial digitalization is low.

Based on Fig. 3d, the coupling coordination degree of the IDE-GTIE-ER system in northeast China is on the rise overall, but it shows strong instability during the study period. In the early stage of the study, the coupling coordination degree of the system shows fluctuating and rising, in which the IDE-GTIE subsystem shows a significant rise, mainly due to the transformation of northeast China's old industrial zones, and due to the low degree of marketization, in the course of the economic restructuring process, the Need for industrial digitization and green technology innovation together, in the late stage of the study, the coupling coordination degree of IDE-GTIE decreases, and with the proposal of green ecological civilization construction, the industrial development takes ecological factors into account, and the coupling coordination degree of IDE-ER gradually increasing.

From Fig. 3, the coupling coordination degree level of the IDE-GTIE-ER system in each region is slightly higher in the northeast than in other regions, by analyzing the subsystems, while the overall system development level is lower in the East, mainly due to the higher efficiency of industrial digitization in the east and the relatively low green technology innovation and ecological resilience. Subsystems' coupling coordination degree of IDE-GTIE and IDE-ER sub-systems is higher in the northeast and east than in the middle and west, respectively, while gtie -er is higher in the west and northeast than in the east and middle, as shown by results of the subsystems' synergistic development level research. East and northeast have good industrial foundations, with the continuous promotion and optimization of regional coordination development strategy, the sustainability of the economic system has been enhanced, while the digital transformation, green technology innovation, and ecological transformation based on the industrial foundation for the system interaction has been increasing, while the western industrial development foundation is poorer, and the digital construction of the industry is difficult to a certain extent, but the western energy industry has a better heritage, and the green technological innovation has improved energy utilization efficiency and reduced resource depletion, thus reducing environmental pressure to a certain extent. To a certain extent, green technology innovation has improved energy utilization efficiency and reduced resource loss, thus reducing environmental pressure. Middle Region, as an important transportation hub as well as a production base for agricultural products, and an important industrial base for energy and raw materials in China, has a weak urban industrial infrastructure, a relative lack of conditions for green technological innovation on an industrial basis, and a low coupling coordination degree of the system.

Regional Variance Analysis

Due to the estimation results in Fig. 4, the overall difference between IDE, GTIE, and ER in each region Fig. 4d tends to decrease during the study period, but it rises in 2019 and keeps decreasing thereafter, and according to the decomposition of the differences, the main reason for the overall increase in the differences in 2019 is due to the increase in the differences between the groups.

By comparing the overall differences in the coupling degree coordination of the subsystems (Fig. 4a,b,c), the overall differences in the IDE-GTIE subsystems were maintained in the range of 0.11-0.15 during the study period, and the results of the difference decomposition showed that the contributions of the inter-group differences, intra-group differences, and hyperdensity differences to the overall differences were more evenly distributed, basically maintained in the range of 0.3- 0.7. The IDE-ER subsystems showed a decreasing trend during the study period, mainly because the intergroup differences decreased continuously during the study period. The study period showed a decreasing trend, mainly due to the fact that the between-group differences were decreasing during the study period. The overall differences of the GTIE-ER subsystem basically remain stable during the study period, and unlike the IDE-GTIE subsystem, the overall differences of the GTIE-ER subsystem are mainly affected by the changes in the differences between groups. Through the spatial distribution, the intergroup differences of IDE, GTIE, and ER synergistic development are mainly realized as the differences between the eastern region and the northeastern and western regions, and according to the coefficients of differences of each subsystem, the overall differences of the synergistic development of the subsystems are mainly originated from the intergroup differences.

Fig. 4.

Decomposition results of regional differences in the coupling coordination degree of the system.

It can be seen from the comparison that the overall differences in synergistic development of IDE, GTIE, and ER mainly originated from intergroup differences, while in the process of synergistic development, in the trend of change by the IDE-ER intergroup differences in the role of the greater influence, while IDE-GTIE and GTIE-ER to a certain extent to maintain the IDE, GTIE, and ER synergistic development of the stability of the IDE, GTIE, and ER, in particular, by the intergroup differences between the GTIE-ER. The effect of the difference was more influential.

Mechanism Analysis

Regression Results

The results of the spatial double fixed-effects spatial lag model measurements for each region in China are shown in Table 5.

Eastern region regression results show that the influence coefficient of a single system shows negative influence at the significance level, while the influence of an interacting system is positive, and the influence relationship is in the order of GTIE×ER (3.2164) > ER×IDE (1.311) > IDE×GTIE (0.0806). It can be seen that the interaction effects of GTIE, IDE, and ER are much larger than the interaction effects of GTIE and IDE. This is mainly due to the relatively better development of high-tech industries in the eastern region, which possesses certain resources and technologies for green technology innovation and industrial digital transformation. The investment in ecological resilience construction is relatively weak, while the industrial digital transformation efficiency is better than the green technology innovation efficiency. Therefore, green technology innovation efficiency is superior to green technology innovation efficiency. At the same time, industrial digital transformation efficiency is better than green technology innovation efficiency. Therefore, the joint effect of green technology innovation efficiency and ecological resilience is stronger than other types of system interactions.

Mid-region regression results show that the effects of IDE and GTIE on the overall level of synergistic development are 0.7342 and 1.5657, respectively, and are significant. Meanwhile, the interaction test results show that only the interaction between GTIE and ER is significant. The linear effect coefficients of GTIE and ER are greater than the one-way effect coefficients when significance is valid. That is, the influence relationship is shown as GTIE+ER>GTIE×ER>GTIE,ER, presenting a two-factor nonlinear enhancement trend [34]. Results from the regression show that improving the efficiency of green technology innovation and strengthening the application of green technology in the construction of ecological resilience can help to enhance the level of regional synergistic development. On the one hand, compared with the eastern region, the central region's industrial digital foundation is relatively weak, but its natural resource endowment is not weaker than that of the eastern region, and green technological innovation has a certain application space, while green technological innovation has a certain knowledge spillover effect [35]. On the other hand, the central region, as a region more closely linked to the east, will also be affected by the eastern region. Therefore, strengthening the construction of green technology innovation infrastructure in the central region and improving the efficiency of green technology innovation can help to enhance the level of regional synergistic development.

Table 5.

Spatio-temporal double-fixed spatial lag model Regression Results.

	Variant	Model1	Model2	Model3	Model4	Model5	Model6
Eastern region	IDE	-0.1307***			-0.1778***	-0.653***
	GTIE		-0.0656		-0.1832***		-1.5014***
	ER			0.0524		-0.5494***	-0.8218***
	IDE×GTIE				0.0806*
	ER×IDE					1.311***
	GTIE×ER						3.2164***
	R2	0.5343	0.4733	0.4050	0.3819	0.5999	0.6146
	Log-likelihood	238.11	223.02	222.09	248.34	240.33	229.68
Middle Region	IDE	0.7342***			0.6437***	0.7531***
	GTIE		1.5657***		0.8131***		1.67***
	ER			0.1631		-0.3013***	0.4091*
	IDE×GTIE				0.0016
	ER×IDE					-0.0061
	GTIE×ER						0.0241*
	R2	0.7480	0.6956	0.4643	0.8832	0.7016	0.6461
	Log-likelihood	249.85	190.51	179.6	292.15	194.12	251.29
Western Region	IDE	0.3629***			-0.2045***	0.2926
	GTIE		0.2086***		-0.0783		0.5158**
	ER			-0.7926***		-0.3701	-0.9196***
	IDE×GTIE				3.3007***
	ER×IDE					0.0766
	GTIE×ER						-0.4365
	R2	0.5261	0.2596	0.3787	0.4237	0.5524	0.4873
	Log-likelihood	222.37	214.13	214.23	267.27	223.22	220.84
Northeast Region	IDE	0.2524**			-0.3981***	-0.8194***
	GTIE		-0.0327		0.0005		2.6834***
	ER			-0.0152		-0.5584***	1.1527***
	IDE*GTIE				3.0902***
	ER*IDE					3.8919***
	GTIE*ER						-6.2575***
	R2	0.8069	0.6930	0.6895	0.7227	0.7308	0.8750
	Log-likelihood	106.98	103.77	103.71	125.65	109.97	124.1
Spatial fixed		YES	YES	YES	YES	YES	YES
Time fixed		YES	YES	YES	YES	YES	YES

[i] Note: ***Significant at the 1% level, **Significant at the 5% level, *Significant at the 10% level, hereafter.

Fig. 5.

Synergistic development theory analysis for the synergistic evolution of industrial digitalization, green technology innovation and ecological resilience.

Fig. 6.

Driving mechanism for synergistic development of industrial digitization, green technology innovation, and ecological resilience.

Western region's measurement results show that there is a significant positive effect of one-way regression results of IDE and GTIE. According to the results of the interaction test, only the interaction effect of GTIE × IDE shows obvious significance, while in the relationship of the influence coefficient, GTIE × IDE > GTIE,IDE, shows a two-factor enhancement trend. From the measurement results, the western region is more affected by the interaction effect of GTIE and IDE. Unlike the eastern region, the average level of GTIE and IDE in the western region is lower than that in the eastern region, while the level of ER is not much different, so the marginal contribution of the enhancement of GTIE and IDE to regional synergistic development is much larger than that of the eastern region and the contribution of ER to the western region.

The northeast region's regression results show that only IDE has a positive effect on the overall synergy development level at the 1% significance level in the one-way regression. According to the interaction test, IDE×GTIE and ER×IDE both show positive influence effects on the overall synergistic development and at the same time show a two-factor enhancement trend in the quantitative relationship. GTIE×ER has a significant negative effect, but GTIE+ER shows a positive effect at the 1% level of significance, which suggests that the IDE plays a major role in the synergistic development of the system in the Northeast region, whereas the linear regression results of GTIE and ER indicate that the IDE plays a major role in the systematic synergistic development. GTIE, ER linear combination can also promote system development, but the influence role is lower than the mutual interaction with IDE, that is, ER * IDE>IDE*GTIE>GTIE+ER confirmed the results of the one-way test.