Research on Impact Mechanisms of Digital Economy on High-Quality Development of Forestry

Abstract

Forests, as a critical component of terrestrial ecosystems, have long been recognized for their vital roles as water, money, grain, and carbon repositories, collectively known as the “Four Repositories” concept. The significance of forests in the context of economic and ecological development has become increasingly prominent. As a novel economic paradigm and a new driver of growth, the digital economy ($DIG$) can significantly improve the management, operation, and preservation of forests, thus offering new prospects for advancing the high-quality development of forestry ($HQDF$) in China. This study constructs indicator systems for the levels of $HQDF$ and $DIG$ using panel data from 30 provinces in China over the period from 2012 to 2021. The empirical analysis utilizes a fixed effect model to assess the influence of $DIG$ in promoting $HQDF$ in China. Moreover, the mediation analysis model is employed to examine the underlying mechanisms. There are four main results as follows. First, the baseline regression results show that $DIG$ can directly drive $HQDF$ in China. Second, the mechanism analysis reveals that $DIG$ positively influences $HQDF$ through three mediating channels: strengthening environmental regulation, enhancing public environmental awareness, and increasing innovation in forestry. Third, the heterogeneity analysis indicates that the positive impact of $DIG$ on $HQDF$ is more pronounced in northern provinces, economically advanced regions, and provinces with stronger fiscal support for forestry. Fourth, further analysis reveals significant spatial spillover effects of $DIG$ on $HQDF$ in China. The study findings not only clarify the driving mechanisms of $DIG$ in $HQDF$, but also provide valuable policy insights for exploring practical pathways for $HQDF$ in China in the modern context.

1. Introduction

In recent times, the intensification of global warming and the rising frequency of extreme weather events have underscored the urgent need for sustainable development and climate resilience worldwide. Forests, serving as vital carbon sinks and biodiversity hotspots, are essential in combating climate change and maintaining ecological equilibrium. The United Nations’ Sustainable Development Goals (SDGs) emphasize the necessity of sustainable forest management and technological innovation to address environmental challenges. Against this global backdrop, China’s commitment to ecological civilization, as articulated in its national policies, aligns with international efforts to harmonize economic growth with environmental stewardship. China has consistently placed great emphasis on taking effective measures to combat climate change. The report of the 18th National Congress of the Communist Party of China (CPC) incorporated ecological civilization construction into the five-sphere integrated plan, while the report of the 20th CPC National Congress highlighted the need to “protect nature and the ecological environment as we protect our eyes”. It is evident that the CPC has consistently placed ecological civilization construction at a prominent position within its overall agenda. The emphasis on ecological civilization building is a distinctive feature of high-quality economic and social development in the new era.

Forests, as a vital component of terrestrial ecosystems, play multiple roles, including climate regulation, water conservation, carbon sequestration, and contributing to cultural landscapes and economic development. By the end of 2023, the total forestry output value driven by forest resources had reached CNY 9.2 trillion, making a significant contribution to the national economy. It also serves as a key battleground in the construction of ecological civilization. In this study, the high-quality development of forestry ($HQDF$) is defined as the dual improvement of forest quantity and quality, while simultaneously balancing the economic and ecological values of forests. Actively promoting $HQDF$ and fully harnessing the roles of forests in “conserving water, bringing in economic benefits, boosting grain production, and serving as a carbon sink” is an effective approach to establishing a new development pattern where economic and ecological benefits complement and reinforce each other with shared goals.

However, at present, China’s high-quality forestry development still faces numerous constraints and developmental bottlenecks, such as the need for enhanced regulation and improved innovation capabilities. As a global digital powerhouse, China has achieved remarkable progress in the digital sector, offering new pathways to tackle these challenges. The “14th Five-Year Plan” underscores the significance of “accelerating digital development and constructing a digital China”. The digital economy ($DIG$), based on AI, big data, and so on, has become deeply embedded across multiple domains of economic and social progress. $DIG$ can offer novel approaches and opportunities to address the challenges faced in achieving $HQDF$. Accelerating the digitization of forestry not only helps advance ecological civilization construction but also injects new momentum into $HQDF$ in China. A critical issue that requires urgent attention is how to better leverage $DIG$ to drive ecological civilization construction, fully integrate digitization with environmental sustainability, and turn $DIG$ into a “catalyst” for $HQDF$.

The existing literature on the impact of $DIG$ on forestry development primarily concentrates on several key areas: Firstly, in terms of strengthening forest resource management, Buonocore et al. [1] introduce a framework for developing a forest digital twin to enhance forest risk management, with the ultimate aim of directing capital toward sustainable forest management practices. Venanzi et al. [2] reviewed the latest applications of intelligent technologies in assessing the sustainability of forest management from 2019 to 2023, highlighting that precision forestry technologies are tools that support forest managers, rather than replace them. Secondly, in terms of enhancing forest benefits, Nitoslawski et al. [3] introduced the concept of “smart urban forestry” and emphasized the potential of digital infrastructure to improve forest benefits. $DIG$ can significantly improve the green total-factor productivity of forestry, and enhancing total factor productivity in forestry is instrumental in promoting the green, high-quality development of the forestry industry [4]. Scholz et al. [5] summarized how digital technologies are used to enhance forest supply chains. Thirdly, in terms of actualizing the value of ecological products, Kong et al. [6], taking Lishui City in Zhejiang Province as an example, found that $DIG$ has a dual-threshold effect on forest ecological product value conversion efficiency. When $DIG$ is greater than the first threshold but less than the second, it significantly impacts the efficiency of forest ecological product value conversion. Other scholars have explored the mechanisms and feasible paths for leveraging $DIG$ to empower the realization of forest ecological product value [7,8]. Additionally, some scholars have examined how cutting-edge digital technologies boost the subjective and objective well-being of forest farmers [9]. The above studies collectively indicate that $DIG$ has a “constructive” impact across multiple aspects of forestry development. However, it is undeniable that the $DIG$ also has a “destructive” side. According to some scholars who analyzed Finland’s revised Forest Information Act, there is a tension between the right to know and the right to privacy when it comes to opening up forest data and information. They also emphasized that the management of open data should not be neglected [10].

By analyzing the above literature, it was found that although several scholars have examined the impact of $DIG$ on forestry development, few have defined the concept of $HQDF$ and thoroughly examined the mechanisms through which $DIG$ drives $HQDF$. To fill these gaps, this study seeks to define and measure $HQDF$ across provinces and investigate the impact of $DIG$ on $HQDF$ as well as its underlying mechanisms. The marginal contributions of this study are as follows: Firstly, by comprehensively considering forest quantity and quality, as well as the economic and ecological values of forests, an indicator system for $HQDF$ in China is constructed from four dimensions: forest resources, forest management, forest economic value, and forest ecological value. This system provides a scientific framework to measure the level of $HQDF$ across Chinese provinces, identify weaknesses in the country’s forestry development, and offer a basis for formulating relevant policies. Secondly, the study confirms that $DIG$ has a positive impact on $HQDF$ in China. It also clarifies the mechanisms through which $DIG$ influences $HQDF$, including the intensity of environmental regulation, public environmental awareness, and the level of innovation in the forestry sector. Thirdly, the study examines the spatial spillover effects of $DIG$ on $HQDF$ in neighboring provinces. The findings reveal that, while promoting $HQDF$ in the local region, $DIG$ also promotes $HQDF$ in adjacent regions.

2. Hypothesis Development

Guided by the aspiration to realize the “Beautiful China initiative” and after decades of persistent effort, China has undoubtedly become a large forestry country, with forest area and stock volume maintaining dual growth for thirty consecutive years. However, there still exists a gap between the current state and the development goals of becoming a global leader in forestry. For example, while the growth of forest area has slowed, and expansion cannot be unlimited, maintaining stable growth in stock volume faces certain development bottlenecks. Therefore, continuously and proactively advancing $HQDF$ is the essential path for China to become a forestry powerhouse. In this context, $HQDF$ necessitates transcending the constraints of one-sided development thinking. It focuses on enhancing both the quantity and quality of forests, while balancing the economic and ecological values of forests. The goal is to achieve concurrent progress in both quantity and quality, and to foster the integrated development of the economic and ecological values of forests. At the same time, $DIG$ represents a novel economic paradigm that emerged alongside the birth and evolution of the Internet. Especially since the Fourth Industrial Revolution, it has exhibited high levels of innovation, strong penetration, and broad coverage, playing a critical role in various sectors, including culture, politics, and ecological development [11,12,13]. In this context, to achieve the “Beautiful China initiative” and realize modernization with harmonious integration of humanity and nature, it is crucial to harness the strengths of $DIG$ to drive $HQDF$. This can be manifested in four aspects:

Firstly, digital technologies like GIS and satellite remote sensing can be employed to collect data on terrains, soil types, climate, and existing vegetation types and coverage in target regions [1]. By comparing these data with information stored in databases, the most suitable tree species for planting in a given area, which would rapidly increase both forest coverage and stock volume in the region, can be identified. Furthermore, the development of digital forest monitoring systems can track tree growth. An integrated monitoring system can accurately reflect forest growth patterns and environmental changes [14], enabling timely adjustments to management strategies and appropriate human interventions, thereby promoting the sustainable development of forest resources.

Secondly, technologies such as UAVs and sensors can be deployed to observe and identify pests and diseases, enabling the timely implementation of targeted control measures [15]. Additionally, digital platforms or applications can facilitate information sharing and collaboration among forest managers, scientists, and policymakers. These platforms can aggregate diverse data sources and offer real-time tracking of pest and disease outbreaks, along with recommendations for prevention and control. Furthermore, digital economic tools, exemplified by the creation of digital economy platforms and data markets, can promote the translation of scientific research into practical applications and the dissemination of technology [16], while attracting additional funding and resources for forest pest and disease management.

Thirdly, under the wave of $DIG$, live-streaming e-commerce, with its unique advantages, provides new marketing models and business opportunities for the marketing of diverse products. The live-streaming economy has effectively increased product sales and expanded their market reach, demonstrating a broad and profound economic impact [17]. The “long-tail effect” of e-commerce can effectively promote and sell forest products with low demand and limited market recognition, addressing issues of information asymmetry [18]. The growth of $DIG$ enables a more comprehensive industrial chain for forest products. Through the use of IoT technology and big data, real-time tracking and management of forest products can be achieved, reducing transportation losses and inventory costs, while enhancing logistics efficiency [19]. As forest farmers gain more benefits, they are incentivized to expand forest product production, ultimately fostering a virtuous cycle in the realization of forestry economic value.

Fourthly, as $DIG$ has developed, processing and disseminating information have become much more efficient. By disseminating green and low-carbon concepts through online platforms, various sectors can participate actively in forest conservation, ecological restoration, and sustainable utilization [20]. Moreover, digital technologies enable the accurate assessment and measurement of the value of forest ecosystems, directing more funding and resources toward projects that aim to realize these ecological values, such as the establishment and maintenance of forestry nature reserves and forest parks. Additionally, blockchain technology can be used to create carbon emission trading platforms, with smart contracts ensuring the transparency and credibility of carbon sequestration projects [21]. This approach encourages both businesses and individuals to support forest conservation and reforestation projects through the purchase of carbon credits, thereby increasing carbon sequestration and promoting the realization of forest ecological value.

After comprehensively analyzing the four aspects, we propose the first hypothesis:

H1:

$DIG$ can promote $HQDF$ in China.

As well as contributing directly to $HQDF$, $DIG$ can also indirectly contribute to this development through environmental regulations, public environmental awareness, and forestry innovation. Below is a detailed analysis:

It is possible to enhance the effectiveness of environmental regulations and their enforcement as $DIG$ develops. Through digital monitoring and management systems, such as GIS, GNSS, and UAVs, forestry management agencies can more efficiently oversee and manage forest resources, ensuring their sustainable use [22]. Digital technologies provide more accurate data and information, facilitating the real-time monitoring and management of forest resources. For instance, blockchain technology can be employed to ensure the legality of timber sources, effectively preventing illegal logging and the misuse of forest resources. By utilizing this technology, forest resources are monitored in real time, precisely recorded, and strict penalties can be imposed on those who damage them.

Behavior is determined by awareness, and cultivating and enhancing public environmental awareness is crucial to society’s green transformation [23]. Online information platforms facilitate the dissemination of environmental knowledge. The widespread use of online education and social media enables the public to better understand forestry’s importance and its relationship with ecological balance. This heightened awareness helps stimulate public enthusiasm for forest conservation, thereby promoting green consumption and sustainable development. Further, digital technologies facilitate public participation in environmental protection. As a typical application of the digital economy, Ant Forest digitizes users’ low-carbon behavior and enhances their environmental awareness through gamification and socialization in the process of obtaining “green energy”. When the energy accumulates to a certain extent, they can apply for tree planting or natural conservation areas, contributing to $HQDF$.

Through big data and data analytics technologies, $DIG$ provides rich data resources for innovation [24]. Digital tools, as innovative technological foundations, reduce the costs and barriers associated with innovation. $DIG$ transcends geographical boundaries, creating an open and collaborative innovation ecosystem. In this ecosystem, different organizations and individuals have the opportunity to collaborate on innovation projects through digital platforms, jointly addressing complex challenges and driving technological progress. The cultivation of new high-quality productive forces has emerged as an essential requirement and a central focus for driving high-quality economic and social development, with continuous innovation being its most prominent characteristic. Furthermore, innovation can rapidly expand into global markets, as digital platforms and online marketplaces provide innovators with opportunities to reach global audiences, thereby fostering international cooperation and technological exchange. Chen et al. [4] empirically tested that green technological innovation is one of the key mechanisms through which $DIG$ enhances the green total factor productivity of forestry. Patents, serving as carriers for innovation activities, are an effective incentive system for innovation. Forestry patents can be turned into practical applications to reduce pest and disease incidence, enhance the productivity of forestry operations, and restore vegetation in desertification areas. Consequently, forestry will be developed in a high-quality manner.

Synthesizing the aforementioned analysis and building on the validation of Hypothesis 1, we proceed to investigate the channels through which $DIG$ influences $HQDF$. Afterwards, we propose the second hypothesis, which can be divided into the following three aspects:

H2:

$DIG$ promotes  $HQDF$ through three channels: enhancing the intensity of environmental regulations, raising public environmental awareness, and increasing the intensity of forestry innovation.

(1) $\mathrm{DIG}$ promotes $HQDF$ in China by enhancing the intensity of environmental regulations.

(2) $DIG$ promotes $HQDF$ in China by raising public environmental awareness.

(3) $DIG$ promotes $HQDF$ in China by increasing the intensity of forestry innovation.

3. Empirical Design

3.1. Variables Selection

3.1.1. Dependent Variable

To assess the level of high-quality development of forestry ($HQDF$),

Table 1. Indicator system for the high-quality development of forestry.

Primary Index	Secondary Index	Index Calculation Method (Unit)
Forest resources	Forest coverage rate	%
	Forest volume per unit area	Forest volume/forest area (cubic meters/hectare)
Forest management	Forest disease control rate	%
	Forest pest control rate	%
Forest economic value	Output value of forestry per unit area	Total output value of forestry industry/forest area (yuan/hectare)
	Production of major economic forest products	Main economic forest product output (tons), logarithm taken
Forest ecological value	Proportion of land area occupied by natural reserves in the forestry system	%
	Number of forest parks	(units), logarithm taken
	Carbon sink per unit area of forest	Forest carbon sink/forest area (tons/hectare)

shows the indicator system developed for assessing forest quality and quantity while balancing their economic and ecological values. The two secondary indicators—forest coverage rate and forest volume per unit area—serve as proxies for each province’s forest resource abundance. Forest pests and diseases, which disrupt the ecological balance and hinder sustainable forest development, are critical factors to consider. Therefore, the rates of forest pest and disease control are used as secondary indicators of forest management effectiveness. In addition, forests possess significant economic value. The output per hectare of forest area represents the forest’s economic contribution, and the production of major economic forest products further reflects the forests’ practical economic impact. As forests are the main component of terrestrial ecosystems, their ecological value is an important criterion of their quality. In this context, three secondary indicators are selected: forest parks, nature reserves, and carbon sequestration functions. Given that forest surveys are typically conducted every five years, the geometric growth rate method was used to interpolate forest volume data. Following previous studies, the forest carbon storage was estimated using the forest volume expansion method, where the carbon sequestration for a given year is the difference between the carbon storage in that year and the previous year [25]. The specific calculation methods for the indicators are detailed in Appendix A. To synthesize the level of $HQDF$, this study uses the entropy weight method. In contrast to subjective weighting methods, the entropy weight method minimizes human bias [26].

3.1.2. Core Independent Variable

To assess the level of digital economy ($DIG$) development, an indicator system for $DIG$ is constructed on the basis of the existing literature [27,28], which is referenced in the 2022 China Digital Economy Development Research Report proposed by the China Center for Information Industry Development. $DIG$ is divided into four dimensions in this study: digital infrastructure, digital industrialization, industrial digitalization, and digital governance. Data valuation is not included in the indicator system since the data factor market in China is not yet fully mature and the related mechanisms are still underdeveloped, making it difficult to select suitable indicators for quantification [29]. Developing $DIG$ relies on digital infrastructure, which is represented by its construction and utilization. As the name suggests, digital industrialization denotes the process of generating tradable products and services using digital technologies, which is calculated by measuring the size of the digital industry and the number of employees within it. The process of integrating digital technologies into the industrial sector is called industrial digitalization, and is further subdivided into agricultural digitalization, industrial digitalization, and service sector digitalization based on the three major industry categories. Besides the indicators above, this study evaluates digital governance from two perspectives: government service capability and government attention to digitalization. Government digital attention is measured according to Jin et al. [30], who collected provincial government work reports from 2012 to 2021, applying text analysis to calculate the frequency of digital economy-related terms per thousand words. To synthesize an indicator system for $DIG$, the study also uses the entropy weight method. The indicator system is illustrated in

Table 2. Indicator system for the digital economy.

Primary Index	Secondary Index	Three-Level Index
Digital infrastructure	Infrastructure construction	Optical cable line density
		Per capita mobile telephone exchange capacity
	Infrastructure utilization	Number of internet broadband users per 100 people
		Number of internet domain names per 100 people
		Mobile phone penetration rate
		Per capita mobile SMS volume
Digital industrialization	Industry scale	Per capita total telecommunications service volume
		Per capita software business revenue
		Per capita information technology service revenue
	Industry employee	Proportion of employment in the information transmission, software, and information technology services industry
Industrial digitalization	Agricultural digitalization	Degree of agricultural electrification
		Rural broadband access penetration rate
		The average number of computers owned per 100 rural households at the end of the year
		Agricultural meteorological observation stations
	Industrial digitalization	Proportion of revenue from high-tech industries
		Average expenditure on technology introduction by industrial enterprises above designated size
		Average expenditure on the assimilation and absorption of technology by industrial enterprises above designated size
		Average expenditure on the purchase of domestic technology by industrial enterprises above designated size
		Average expenditure on technological transformation by industrial enterprises above designated size
	Service digitalization	Per capita express delivery service revenue
		Proportion of enterprises engaged in e-commerce transactions
		Proportion of e-commerce sales in GDP
		The breadth of digital financial inclusion
		The depth of digital financial usage
		The digitalization level of inclusive finance
Digital governance	Government service capability	Government online services capability index
		Number of Weibo accounts of government agencies per 100 people
	Government attention	Government digital attention
		Proportion of fiscal expenditure on science and technology

.

3.1.3. Control Variables

Control variables are selected at both the forestry and provincial levels: (1) the number of forestry system practitioners at the end of the year ($PRAC$), with the logarithmic transformation applied; (2) the cumulative forestry investment completed since the beginning of the year ($INVEST$), also transformed logarithmically; (3) per capita regional GDP index (previous year = 100) ($GDPP$); (4) industrial structure ($INDUS$), measured as the ratio of the value added in the tertiary sector to that in the secondary sector; (5) population density ($POPU$).

3.1.4. Mechanism Variables

(1)

Environmental Regulation Intensity ($REGU$): Local governments can reduce illegal logging and other deforestation activities by strengthening environmental law enforcement, thereby promoting $HQDF$. Drawing from the work of Yuan and Zhang [31], environmental administrative penalty cases handled at the provincial level are logarithmized to gauge the intensity of local environmental regulation.

(2)

Public Environmental Awareness ($AWARE$): The role of the public in environmental protection should not be overlooked. The public can express their concerns about environmental protection through channels such as letters and visits, urging the government to take strict actions against illegal activities that harm the environment. Therefore, this study represents public environmental awareness using the logarithm of the number of environmental complaint letters [32].

(3)

Forestry Innovation Efforts ($INNOV$): Innovation is a driving force behind high-quality development across various industries, and patents are frequently employed as a measure of technological innovation [33]. Hence, the logarithm of forestry-related patents is used in this study to measure innovation efforts in forestry.

3.2. Model Settings

3.2.1. Baseline Regression Model

The baseline regression model in this study is specified as follows:

$${HQDF}_{it}={\beta }_0+{\beta }_1{DIG}_{it}+{\beta }_2{Z}_{it}+{\lambda }_i+{\epsilon }_{it}$$

(1)

In the model, $i$ represents the provincial regions; t represents the years; ${HQDF}_{it}$ denotes the level of forestry high-quality development, while ${DIG}_{it}$ indicates the level of digital economy development. $Z_{it}$ is a vector of control variables selected from both the forestry sector and provincial differences. ${\beta }_0$ is the intercept coefficient of the regression, ${\beta }_1$ represents the effect of $DIG$ on $HQDF$ in China, and ${\beta }_2$ is the vector of coefficients for the control variables. ${\lambda }_i$ represent the province-fixed effects, while ${\epsilon }_{it}$ is the random error term.

3.2.2. Mediation Effect Model

Referring to Hayes [34] and Baron and Kenny [35], the mediation effect model for mechanism analysis is as follows:

$$M_{it}={\alpha }_0+{\alpha }_1{DIG}_{it}+{\alpha }_2{Z}_{it}+{\lambda }_i+{\epsilon }_{it}$$

(2)

$${HQDF}_{it}={\delta }_0+{\delta }_1{DIG}_{it}+{\delta }_2{M}_{it}+{\delta }_3{Z}_{it}+{\lambda }_i+{\epsilon }_{it}$$

(3)

In the equation, $M_{it}$ represents the mechanism variables to be examined in this study, which include relevant indicators of environmental regulation intensity, public environmental awareness, and forestry innovation efforts. The remaining variables align with those in Equation (1).

3.3. Data Sources and Descriptive Statistics

This study uses a sample of 30 provinces (excluding Tibet) in mainland China from 2012 to 2021. The sample data primarily come from sources such as the China Forestry and Grassland Statistical Yearbook, China Population & Employment Statistical Yearbook, China Statistical Yearbook for Regional Economy, China Statistical Yearbook on High Technology Industry, China Industrial Statistical Yearbook, China Statistical Yearbook on Science And Technology, China Environment Yearbook, China Energy Statistical Yearbook, China Agricultural Yearbook, China Statistical Yearbook of the Tertiary Industry, the Survey and Evaluation Report on Provincial and Key City Integrated Government Service Capabilities, the China Internet Development Statistics Report, Peking University’s Digital Inclusive Finance Index, the China Forestry Intellectual Property Network, the Ministry of Ecology and Environment of China, the National Bureau of Statistics, provincial statistical yearbooks, and government work reports from each province. Missing data were imputed using a three-year moving average method. Descriptive statistics are illustrated in

Table 3. Descriptive statistics.

Variable	Observations	Mean	SD	Min	Median	Max
$HQDF$	300	0.306	0.083	0.105	0.306	0.641
$DIG$	300	0.168	0.098	0.056	0.146	0.733
$PRAC$	300	5.344	1.135	1.629	5.468	8.059
$INVEST$	300	0.756	0.441	0.049	0.708	2.473
$GDPP$	300	106.948	2.499	96.400	107.100	113.800
$INDUS$	300	1.283	0.711	0.549	1.120	5.297
$POPU$	300	47.514	70.793	0.791	29.290	392.587
$REGU$	300	7.896	1.124	4.431	7.841	10.713
$AWARE$	300	7.921	1.077	4.357	8.107	10.810
$INNOV$	300	7.150	1.116	3.664	7.240	9.695

. There is a noticeable variation between the maximum and minimum values of $HQDF$, where the maximum exceeds the minimum by more than sixfold. Similarly, the maximum $DIG$ was 0.733, while the minimum was 0.056, indicating that the maximum is over thirteen times the minimum. These results indicate significant disparities between the provinces in the levels of $HQDF$ and $DIG$.

4. Empirical Results

4.1. Baseline Regression Results

The baseline regression results are presented in

Table 4. Results of the baseline regression.

Variable	$\mathit{H}\mathit{Q}\mathit{D}\mathit{F}$
	(1)	(2)	(3)
$DIG$	0.426 ***	0.400 ***	0.191 ***
	(0.028)	(0.032)	(0.056)
$PRAC$		−0.005	−0.005
		(0.010)	(0.010)
$INVEST$		0.026 ***	0.026 ***
		(0.007)	(0.007)
$GDPP$			0.001
			(0.001)
$INDUS$			0.050 ***
			(0.010)
$POPU$			0.002
			(0.001)
Constant	0.234 ***	0.245 ***	0.029
	(0.005)	(0.059)	(0.117)
Province FE	Y	Y	Y
N	300	300	300
R2	0.469	0.495	0.543

Note: *** indicate significant coefficients at the 1% levels, with standard errors in parentheses.

. Columns (1) to (3) show the results of progressively adding control variables at the forestry and provincial levels, based on separate regressions of $HQDF$ and $DIG$. As shown in column (1), $DIG$ significantly impacts $HQDF$, without any control variables. In column (2), forestry-level control variables are incorporated into the model in (1), and the positive effect remains significant, suggesting that the inclusion of forestry practitioners and investment does not alter the positive relationship between $DIG$ and $HQDF$. Column (3) further incorporates provincial-level control variables to account for the potential confounding factors. After controlling for various confounding factors, $DIG$ continues to significantly drive $HQDF$ in China at the 1% level. Thus, Hypothesis 1 is supported.

4.2. Robustness Test and Endogeneity Treatment

4.2.1. Robustness Test

The results of robustness test are shown in

Table 5. Results of the robustness test.

Variable	$\mathit{H}\mathit{Q}\mathit{D}\mathit{F}$
	(1)	(2)	(3)	(4)
	Factor Analysis	One-Period Lag	Shorten the Time Window	Tobit Model
$DIG$	0.047 ***	0.169 ***	0.421 ***	0.191 ***
	(0.011)	(0.056)	(0.066)	(0.052)
Controls	YES	YES	YES	YES
Constant	0.035	0.001	0.092	−0.245
	(0.116)	(0.145)	(0.145)	(0.189)
Province FE	Y	Y	Y	N
sigma_u	/	/	/	0.000
				(0.001)
sigma_e	/	/	/	0.024 ***
				(0.001)
N	298	270	240	300
R2	0.551	0.502	0.587	/

Note: *** indicate significant coefficients at the 1% levels, with standard errors in parentheses.

. (1) Modification of Core Explanatory Variable Synthesis Method: The KMO test is conducted first, yielding a KMO statistic of 0.875, indicating that factor analysis is appropriate for synthesizing $DIG$. The regression shows that the coefficient of $DIG$ is significantly positive after replacing the synthesis method. (2) Lagging the Core Explanatory Variable by One Period: Because of the possibility of lag effects caused by $DIG$, $DIG$ is lag-aged by one period before regression analysis. The coefficient remains significantly positive at the 1% level. (3) Shortening the Time Window: The sample period is shortened to 2012–2019 in order to alleviate the effects of the COVID-19 pandemic. According to the regression results, the previous conclusions are valid. (4) Changing the Model Estimation Method: Since the dependent variable has a range of (0, 1), the model estimation method is changed to the Tobit model. The result remains robust.

4.2.2. Endogeneity Treatment

To address potential endogeneity issues, this study follows the approach of Nunn and Qian [36] and employs a Shift-Share Instrumental Variable (IV) to replace $DIG$. The share component is derived from Wang and Shao [37], using the number of post offices per million people in 1984. Historically, regions with more post offices tended to have better-developed digital economies. At the same time, the number of national internet broadband access users in the previous year, reflecting temporal variation, is used as the shift component. The interaction term of the aforementioned two variables is employed as an instrumental variable. The results are illustrated in

Table 6. Results of the endogeneity treatment.

Variable	$\mathit{H}\mathit{Q}\mathit{D}\mathit{F}$
$DIG$	0.850 ***
	(0.197)
Controls	Y
Constant	0.288
	(0.278)
Province FE	Y
Anderson canon. corr. LM statictic	32.438
	[0.000]
Cragg–Donald Wald F statictic	32.006
Stock–Yogo 10% critical value	16.38
N	300
Centered R2	0.875

Note: *** indicate significant coefficients at the 1% levels, with standard errors in parentheses.

, which reject the weak instrument test and the underidentification test. The results show that the conclusions of this study are not affected by potential endogeneity issues.

4.3. Mechanism Test

To verify the aforementioned study hypotheses and explore the mechanisms, this study aims to analyze its mediating effects from three perspectives: strengthening environmental regulation, enhancing public environmental awareness, and boosting innovation in forestry.

4.3.1. Environmental Regulation Intensity

To ensure that forestry is developed in a high-quality manner, it is essential to implement environmental regulations and corresponding penalties for violations such as land encroachment and the illegal logging of natural forests. The first step is to monitor and document these violations. The application of digital monitoring systems and big data platforms can enhance the efficiency of government oversight of forests. Accordingly, this study first examines how $DIG$ affects $HQDF$ from a regulatory perspective. Based on the coefficient shown in column (1) of

Table 7. Results of the mechanism test.

Variable	(1)	(2)	(3)	(4)	(5)	(6)
	$\mathit{R}\mathit{E}\mathit{G}\mathit{U}$	$\mathit{H}\mathit{Q}\mathit{D}\mathit{F}$	$\mathit{A}\mathit{W}\mathit{A}\mathit{R}\mathit{E}$	$\mathit{H}\mathit{Q}\mathit{D}\mathit{F}$	$\mathit{I}\mathit{N}\mathit{N}\mathit{O}\mathit{V}$	$\mathit{H}\mathit{Q}\mathit{D}\mathit{F}$
$DIG$	2.971 **	0.170 ***	5.039 ***	0.128 **	4.521 ***	0.154 ***
	(1.258)	(0.056)	(1.101)	(0.056)	(0.855)	(0.059)
$REGU$		0.007 **
		(0.003)
$AWARE$				0.012 ***
				(0.003)
$INNOV$						0.008 **
						(0.004)
Controls	Y	Y	Y	Y	Y	Y
Constant	10.803 ***	−0.046	6.411 ***	−0.050	13.429 ***	−0.079
	(2.626)	(0.119)	(2.300)	(0.115)	(1.785)	(0.128)
Province FE	Y	Y	Y	Y	Y	Y
N	300	300	300	300	300	300
R2	0.192	0.554	0.262	0.570	0.665	0.550

Note: **, *** respectively indicate significant coefficients at the 5%, and 1% levels, with standard errors in parentheses.

, $DIG$ significantly increases environmental regulation intensity at the 5% level, indicating that it is effective at strengthening environmental regulation. In column (2), the coefficient is positive at the 5% level, suggesting that forestry development is enhanced by higher levels of environmental regulation. Furthermore, after incorporating the environmental regulation intensity indicator, the coefficient of $DIG$ remains significantly positive. Therefore, Hypothesis 2 (1) is confirmed.

4.3.2. Public Environmental Awareness

Awareness of the environment among the public plays a vital role in influencing government policy. In the digital economy, digital media platforms serve as an important channel for exposing environmental information, enabling the public to engage in public opinion oversight [38]. This, in turn, prompts the government to place greater emphasis on $HQDF$, steering $DIG$ toward supporting such development. Based on the coefficient shown in column (3) of Table 7, $DIG$ significantly increases public environmental awareness at the 1% level, indicating that it is effective at raising public environmental awareness. In column (4), the coefficient is positive at the 1% level, suggesting that forestry development is enhanced by higher levels of public environmental awareness. Furthermore, after incorporating the public environmental awareness indicator, the coefficient of the $DIG$ remains significantly positive. Therefore, Hypothesis 2 (2) is confirmed.

4.3.3. Forestry Innovation Efforts

$DIG$ is reshaping the way society operates and profoundly altering the patterns and pathways of innovation. Through it, knowledge and information are more efficiently shared, costs associated with acquiring knowledge are reduced, and entry barriers to innovation are lowered [27]. The massive growth of data and advanced technologies in the digital economy presents forestry with new opportunities for innovation. Innovation is a critical driver of sustainable development and a key support for the advancement of ecological civilization [39]. Based on the coefficient shown in column (5) of Table 7, $DIG$ significantly increases forestry innovation efforts at the 1% level, indicating that it is effective at enhancing forestry innovation. In column (6), the coefficient is positive at the 1% level, suggesting that $HQDF$ is enhanced by higher levels of forestry innovation. Furthermore, after incorporating the forestry innovation efforts indicator, the coefficient of $DIG$ remains significantly positive. Therefore, Hypothesis 2 (3) is confirmed.

4.4. Heterogeneity Analysis

4.4.1. Geographic Location

Compared to northern provinces, southern provinces predominantly have collective forests, which are generally managed at a higher level, and their forests have reached a better level of development. In contrast, forests in northern provinces are more scattered, and their management is relatively underdeveloped. As $DIG$ has advanced, the level of forest management in northern provinces has significantly improved, promoting high-quality forestry practices. As a result, $DIG$ has a smaller effect on $HQDF$ in southern provinces, as illustrated in columns (1) and (2) of

Table 8. Results of the heterogeneity analysis.

Variable	$\mathit{H}\mathit{Q}\mathit{D}\mathit{F}$
	(1)	(2)	(3)	(4)	(5)	(6)
	Northern Provinces	Southern Provinces	Developed Provinces	Underdeveloped Provinces	Strong Fiscal_Support	Weak Fiscal_Support
$DIG$	0.264 ***	0.031	0.258 ***	0.055	0.290 ***	0.173
	(0.057)	(0.097)	(0.065)	(0.116)	(0.072)	(0.108)
Controls	Y	Y	Y	Y	Y	Y
Constant	0.263 **	0.112	0.381 ***	−0.763 ***	0.284 *	−0.218
	(0.123)	(0.181)	(0.143)	(0.189)	(0.169)	(0.175)
Province FE	Y	Y	Y	Y	Y	Y
N	130	170	160	140	150	150
R2	0.509	0.670	0.602	0.652	0.525	0.632

Note: *, **, *** respectively indicate significant coefficients at the 10%, 5%, and 1% levels, with standard errors in parentheses.

.

4.4.2. Economic Development Level

Based on the classification of regions according to per capita GDP rankings by the National Bureau of Statistics, the provinces are categorized into underdeveloped and developed provinces. The underdeveloped provinces include Xinjiang, Gansu, Qinghai, Ningxia, Sichuan, Guizhou, Yunnan, Guangxi, Hainan, Henan, Hebei, Liaoning, Jilin, and Heilongjiang, totaling 14 provinces. The remaining 16 provinces are considered more economically developed. Utilizing digital technologies to enhance $HQDF$ yields relatively lower economic benefits than applying them to other sectors. As a result, underdeveloped provinces are more likely to prioritize using $DIG$ to drive the development of higher-profit industries, treating “economic growth” as a primary incentive. In contrast, developed provinces tend to focus on broader development goals [40]. In these provinces, “environmental governance” becomes an additional guiding principle beyond “economic growth”. As a result, officials in developed provinces tend to direct technological advancements toward forestry, aiming to improve local ecological conditions, as illustrated in columns (3) and (4) of Table 8.

4.4.3. Forestry Fiscal Support Intensity

$DIG$ requires substantial financial investments [41]. Compared to other sectors, forestry has a longer investment cycle and is a highly policy-driven industry [42]. Because of this, society has a limited willingness to invest in forestry, making government financial support vital. In regions with strong fiscal support for forestry, $DIG$ can be more effectively applied to forestry, thereby promoting its high-quality development. Forestry fiscal support intensity ($\mathit{\text{Fiscal\_Support}}$) is calculated as the ratio of national investment in the forestry sector to the total forest area. The greater the national fiscal investment per unit of forest area, the stronger the fiscal support for forestry. Forestry fiscal support intensity is categorized into two groups based on the median value, as illustrated in columns (5) and (6) of Table 8.

4.5. Further Analysis

In the previous section, we explored whether $DIG$ is driving $HQDF$ locally. Nonetheless, focusing solely on the direct impact is not sufficient to capture the full effect of $DIG$, as its high penetration may lead to significant spillover effects in neighboring areas. Thus, we employed a spatial econometric model for further analysis. This approach helps to provide a more targeted theoretical foundation for the formulation of relevant policies by helping to understand the regional impacts of $DIG$.

The spatial weight matrix in this study is based on the inverse of geographic distance between two cities. The formula for calculating the Global Moran’s I is illustrated in Equation (4).

$$Mora{n}^{’}s I=\left[\sum _{i=1}^n\sum _{j=1}^n{\omega }_{ij}\left(f_i-\overline{f}\right)\left(f_j-\overline{f}\right)\right] /\left(s^2\sum _{i=1}^n\sum _{j=1}^n{\omega }_{ij}\right)$$

(4)

In this model, $s^2=\sum _{i=1}^n{\left(f_i-\overline{f}\right)}^2∕n$, $\overline{x}=\sum _{i=1}^n{f}_i∕n$, $n$ is the number of regions. $f_i$ and $f_j$ are the observed values of $HQDF$ in regions $i$ and $j$, respectively. The Global Moran’s I for $HQDF$, as presented in

Table 9. Results of Global Moran’s I.

Year	$\mathit{M}\mathit{o}\mathit{r}\mathit{a}\mathit{n}\mathbf{’}\mathit{s}\mathbf{ }\mathit{I}$	Year	$\mathit{M}\mathit{o}\mathit{r}\mathit{a}\mathit{n}\mathbf{’}\mathit{s}\mathbf{ }\mathit{I}$
2012	0.189 ***	2017	0.169 **
	(0.095)		(0.092)
2013	0.175 **	2018	0.254 ***
	(0.095)		(0.094)
2014	0.119 *	2019	0.288 ***
	(0.095)		(0.094)
2015	0.224 ***	2020	0.258 ***
	(0.094)		(0.094)
2016	0.223 ***	2021	0.265 ***
	(0.092)		(0.094)

Note: *, **, *** respectively indicate significant Moran’s I at the 10%, 5%, and 1% levels, with standard deviation in parentheses.

, significantly rejects the null hypothesis of spatial independence, indicating the presence of spatial autocorrelation. In addition, the results of the Global Moran’s I calculations are consistently positive, suggesting that $HQDF$ in a given region has a positive spillover effect on neighboring regions.

The Spatial Lag Model (SLM) accounts for the spatial dependence of the dependent variable. Specifically, it posits that the dependent variable in a given region is influenced not only by the explanatory variables within the same region but also by the dependent variables of neighboring regions [43]. In this study, the Spatial Lag Model is utilized to examine how the digital economy affects $HQDF$ in neighboring provinces, as shown in Equation (5):

$${HQDF}_{i,t}={\gamma }_1{DIG}_{i,t}+{\gamma }_2{Z}_{i,t}+\rho W\times {Forestry}_{i,t}+{\epsilon }_{it}$$

(5)

In this model, $W$ is the spatial weight matrix. The regression results in

Table 10. Estimation results of SLM.

Variable	$\mathit{H}\mathit{Q}\mathit{D}\mathit{F}$
$DIG$	0.132 **
	(0.056)
$W\times HQDF$	0.170 **
	(0.069)
Controls	Y
Province FE	Y
Direct effect	0.135 **
	(0.058)
Indirect effect	0.035 *
	(0.020)
$\mathsf{\rho }$	0.224 ***
	(0.082)
sigma2_e	0.001 ***
	(0.000)
Log-likeihood	698.476
N	300

Note: *, **, *** respectively indicate significant coefficients at the 10%, 5%, and 1% levels, with standard errors in parentheses.

show that both the direct and indirect effect coefficients are significantly positive, suggesting that as $DIG$ develops in a region, it not only promotes the local $HQDF$ but also generates technological spillovers that positively affect $HQDF$ in neighboring regions. Spatial spillovers are caused by the digital economy’s inherent permeability. Technology, in digital form, can be disseminated and flow across space and time, overcoming geographical and temporal limitations, enabling rapid technology transfer to adjacent areas [44]. Given the similar natural conditions of neighboring regions, technology can be more effectively applied to forests in these areas, ultimately enhancing $HQDF$.

5. Conclusions and Implications

This study utilizes panel data collected from 30 provinces of China from 2012 to 2021 to construct an indicator system for $HQDF$ by combining nine indicators from four dimensions: forest resources, forest management, economic value of forestry, and ecological value of forestry. Similarly, an indicator system for $DIG$ is developed, with 29 indicators across four dimensions: digital infrastructure, digital industrialization, industrial digitalization, and digital governance. The main conclusions are as follows: (1) $DIG$ significantly drives $HQDF$. This conclusion holds after robustness tests. (2) The mechanism analysis reveals that $DIG$ promotes $HQDF$ through three channels: enhancing environmental regulation intensity, raising public environmental awareness, and increasing forestry innovation efforts. (3) The heterogeneity analysis reveals that $HQDF$ is more strongly driven by $DIG$ in northern, economically developed, and fiscally supported provinces. (4) The spatial econometric results indicate that in addition to promoting local $HQDF$, $DIG$ also has a positive spatial spillover effect on neighboring regions as well.

Based on the aforementioned conclusions, this study proposes significant implications for the development of global forestry management and digital economy:

Firstly, invest more in forestry digital infrastructure to support the efficient management of forest resources and big data analysis. For example, developing countries in Southeast Asia, Africa, and South America also face similar challenges such as deforestation and unsustainable practices. For real-time forest monitoring and carbon trading, they can learn from China’s approach, which combines GIS, drones, and blockchain technology. These technologies, although targeted to specific environments, provide scalable solutions to enhance transparency and efficiency in forest governance.

Secondly, improve enforcement measures to effectively strengthen environmental regulations, expand the scope of public environmental awareness campaigns, and increase financial and policy support for forestry innovation. The government should leverage digital technologies to monitor, record, and strictly penalize behaviors such as illegal logging and deforestation in real-time. For each case of forest destruction, immediate action should be taken, with penalties imposed to deter further violations. Enhancing public environmental awareness through educational initiatives, such as environmental awareness campaigns and volunteer activities, so that the government can engage a broader societal audience in promoting and applying $DIG$ in $HQDF$. Public engagement through digital platforms, as demonstrated in China, aligns with global trends in citizen science and participatory environmental governance. Nations like Brazil and Indonesia, where community involvement is critical to combating illegal logging, could replicate such approaches to amplify conservation efforts. Further, the government should focus on training forestry workers in digital skills to prepare them for the changes in technology ahead. Through policy guidance and financial support, the government can incentivize innovation in forestry enterprises to improve overall competitiveness. Collaborative efforts between forestry and technology companies should be promoted to establish innovation incubators, providing opportunities for science and technology to be transformed and applied, and fostering an open, transparent, and efficient forestry innovation system for sustainable and high-quality growth.

Thirdly, provide targeted subsidies and carry out capacity building activities for underdeveloped areas to enhance the digital literacy of forestry workers. Establish “Digital Forestry Transition Funds” to subsidize the procurement of essential technologies (e.g., drones, IoT sensors) in underdeveloped areas. Launch mobile digital literacy campaigns in rural areas, utilizing apps and offline modules to educate forest-dependent communities on leveraging digital tools for sustainable practices. Develop province-specific training programs for forestry workers, and partner with tech firms and universities to deliver workshops and certifications. Economic imperatives may conflict with ecological preservation efforts in less-developed regions. To address these trade-offs, policymakers should prioritize ecological compensation mechanisms tailored to regional contexts. For example, in provinces with fragile ecosystems, governments could implement payment for ecosystem service (PES) schemes, where downstream beneficiaries compensate upstream communities for forest conservation. Digital platforms could facilitate the transparent tracking of conservation outcomes and fund distribution.

Fourthly, promote the sharing of experiences and collaborate across regions for high-quality development. Emphasis should be placed on cross-regional cooperation and coordination, particularly in neighboring areas where successful case studies can be replicated. By establishing cross-regional cooperation mechanisms, sharing advanced technologies and experiences, and setting common sustainable development goals, the overall enhancement and protection of forest ecosystems can be achieved. This collaborative effort will not only foster high-quality forestry development locally but will also generate positive spatial spillover effects in surrounding areas, injecting momentum into broader ecological conservation and economic development.

However, it is crucial to recognize that due to data limitations, this study is unable to conduct a comprehensive international comparison. Future studies should adopt a global perspective to achieve a more holistic understanding of the role of $DIG$ in promoting $HQDF$. Moreover, this study did not explore the extended impacts, obstacles, or potential hazards that $DIG$ may pose to the sustainable advancement of the forestry sector, including issues like unequal access to digital resources or excessive dependence on technological solutions. Future research should explore these long-term effects and challenges in more detail, especially through longitudinal studies to thoroughly examine these risks.