Next Article in Journal
Changing Relationships between Water Content and Spectral Features in Moso Bamboo Leaves under Pantana phyllostachysae Chao Stress
Next Article in Special Issue
Underlying Mechanisms of Urban Green Areas’ Influence on Residents’ Health—A Case Study from Belgrade, Serbia
Previous Article in Journal
Difference in Response of Caragana intermedia Photosynthesis to Soil Water Content in Different Afforestation Years and Related Threshold Effects in Alpine Sandy Lands
Previous Article in Special Issue
Forests Attenuate Temperature and Air Pollution Discomfort in Montane Tourist Areas
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Impact of City Forests on Haze Reduction—Implementation of the National Forest City Policy in China

College of Economics and Management, Nanjing Forestry University, Nanjing 210037, China
*
Author to whom correspondence should be addressed.
Forests 2023, 14(4), 703; https://doi.org/10.3390/f14040703
Submission received: 25 February 2023 / Revised: 20 March 2023 / Accepted: 24 March 2023 / Published: 29 March 2023
(This article belongs to the Special Issue Forest, Trees, Human Health and Wellbeing)

Abstract

:
The construction and development of China’s forest cities revolves around the goal of building an environmentally friendly society and achieving harmonious development between human beings and nature, which is essentially the implementation of the concept of green development and helps promote the sustainable development of sustainable cities. Based on the panel data of 263 prefecture-level cities in China from 2001 to 2020, the National Forest City Policy (NFCP) is used as a quasi-natural experiment to assess the effect of NFCP on haze pollution management and their heterogeneity using a time-varying DID (difference-in-differences) model, and a mediating effect model is used to analyze the mechanism of the effect of NFCP on haze pollution. The research found that (1) the NFCP can reduce urban haze pollution, and this finding remained robust after placebo tests and the replacement of explanatory variables; (2) the NFCP is more conducive to reducing haze pollution in the Pearl River Delta urban agglomeration, Type II large cities, cities in the eastern region, and cities east of the Hu Huanyong line; (3) the NFCP will improve urban green space coverage, raise residents’ awareness of environmental protection, and promote the development of tertiary industries, thereby promoting urban haze reduction.

1. Introduction

Air pollution, especially haze pollution, is one of the environmental problems that need to be addressed in China [1]. The construction of forest cities can alleviate the air pollution problems caused by rapid industrialization and urbanization, offset the CO2, SO2 and dust generated by industrial systems to a certain extent, help suppress haze pollution and improve the well-being of city residents [2]. In order to strengthen the construction of urban forests and rural greening, China implemented the National Forest City Policy (NFCP) in 2004, and the cumulative success of the NFC has reached 219 cities, with more than 400 cities in the NFC, which is expected to basically achieve the equalization of ecological services in forest cities by 2035 [3].
The purpose of the NFCP is to “let the forest come into the city, let the city embrace the forest”, and really do a good job of “planting green for the people, planting green benefit the people”. In terms of objectives, the NFCP seeks to achieve greening of the earth, greening of the heart and greening of the whole population. In terms of method, the NFCP emphasizes the use of artificial methods to create near-natural forests; in terms of function, the NFCP can enhance urban residents’ awareness of green living and ecological civilization. The selection of national forest cities involves 36 indicators in 5 major systems, including the forest network, forest health, ecological welfare, ecological culture, and organizational management [4,5]. Only cities with all indicators meeting the required standards are eligible to be awarded the honorary title of national forest city [6]. With the maturity of the management system, the NFCP provides an important way for China’s cities to follow the development path of production development, rich living and good ecology, and has received more and more extensive attention [7]. However, the social, economic and cultural impacts of the NFCP on local communities, especially the environmental aspects, have not been effectively assessed and empirical evidence on these aspects is still lacking. Therefore, research to assess the haze-reduction effect of the NFCP becomes significant.
Currently, studies related to the NFCP are divided into three main categories. The first category is concerned with the differences in the objectives of the NFCP in different types of cities. Highland mountain cities focus on improving the quantity and quality of forest city construction through the NFCP [8], Hilly cities hope that the NFCP can alleviate problems such as urban insularity [9]. In the case of the plain cities, the quality of urban life has been improved through the NFCP [10], Resource-depressed cities aim to take the NFCP as an opportunity to promote urban green development [11]. The second category is concerned with the functions of urban forests, fully affirming that the construction of urban forests has an important value for the local improvement of ecological connotation, enhancement of environmental carrying capacity, enhancement of carbon sequestration, emission reduction and oxygen release capacity, the improvement of residents’ ecological well-being, and promotion of local green economic development [12,13]. The third category is a comprehensive evaluation of the construction results of the NFCP. The system construction involves indicators of forest status, forest benefits and forest development [14], and the object selection revolves around a single city [15] or a certain urban cluster [16].
The research on haze is mostly focused on the exploration of its hazards and management paths. Among them, the hazards brought by haze mainly include the following three aspects. One is that the chemical composition of haze directly damages the human respiratory and neurological life systems [17]. Second, haze reduces the quality of economic development. Haze pollution not only hinders local attraction of foreign direct investment and the gathering of highly qualified talents [18,19], but also has potential financial evolutionary risk effects that constrain high-quality economic development [20]. Third, haze brings many negative impacts on the production and operation of microeconomic agents such as enterprises. Haze pollution raises the labor costs of enterprises [21], reduces the confidence of business investors [22], causes a decrease in the productivity of enterprises [23], and so on. The analysis of haze management includes not only economic activities such as energy restructuring [24], industrial transformation and upgrading [25], and new infrastructure investment [26], but also economic phenomena such as population agglomeration [27], industrial agglomeration [28], and urbanization [29]; governmental decisions such as innovative city construction [30], smart city construction [31], and national-level integrated big data test areas [32] are also included in the study and become some of the hot spots.
The relationship between the NFCP and haze pollution has been less directly studied in the literature, mostly using natural experiments to study the combined effects of urban forest construction. In terms of the ecological environment, forest vegetation generally regulates the atmospheric environment by chemical dust removal and physical dust reduction, absorbing carbon dioxide and filtering air pollution particles, etc. [33,34]. Some scholars also believe that ozone and PM2.5 may be produced indirectly when trees in cities reach a certain density [35]. In terms of economic value, the construction of urban forests has a positive effect on the carbon storage of cities [36], green economic development [37], forest tourism and forest recreation development [38]. In terms of people’s well-being, urban greening can enhance the quality of cities and relieve residents’ stress [39,40]. Overall, in the vast majority of cases, China’s urban forest construction can purify the atmosphere, reduce urban PM2.5 concentration and increase the concentration of negative air ions, while in some cities (such as Lanzhou and Kunming, etc.), the more green the area, the more serious the air pollution [41,42].
In summary, existing studies have explored the construction initiatives, roles and evaluation of urban forests as well as the hazards of haze pollution and its management, laying a solid foundation for further research in this paper. However, few in-depth studies have been conducted on whether the NFCP affects haze pollution and the variability of the impact, and there is a lack of positive answers to the question of how the policy affects haze pollution. The main marginal contributions of this paper are: firstly, the haze-reduction effect of the NFCP is explored in depth, and the ideas of NFCP research are broadened; Second, the heterogeneity of the haze-reduction effect of the NFCP is analyzed from the perspectives of urban population size, geographic and economic location, and urban clusters, which have an important practical significance for optimizing the layout of forest cities in China. Thirdly, we analyze the influence mechanism of the NFCP on haze pollution from the urban forest, social environmental awareness and industrial structure points of view, and provide reference for promoting high-quality construction of forest cities in China to effectively deal with haze pollution.

2. Theoretical Analysis and Research Hypotheses

The NFCP is to take forest vegetation as the main body, taking into account ecological construction and industrial development, and to improve the sustainable development ability of the city comprehensively through urban planning, combining policy guidance (pushing) and market incentives, reflecting the dual meaning of reducing haze pollutant emissions and improving the quality of urban green development [43]. On the one hand, the NFCP is guided by the concept of green development and covers numerous green development initiatives, which will have a direct positive impact on urban haze reduction [44]. On the other hand, the NFCP, as an external force to promote the green transformation of cities, provides a top-down impetus for urban green transformation and can also indirectly promote urban haze reduction [45]. The NFCP, as a kind of urban green infrastructure construction project, depends not only on the natural conditions in which they are located, but is also closely related to the economic status of the cities themselves [46]. In general, places with better economic development also have a greater ability to invest in urban environmental types of infrastructure. Accordingly, the following hypothesis is proposed:
H1: 
The NFCP can reduce haze pollution, and there is regional heterogeneity in the haze-reduction effect of NFCP.
The NFCP suppresses haze pollution by developing urban forests. As an important means of national forest city construction, urban forests are mostly configured with trees, irrigation and grass structures, with a large leaf area and complex spatial structures, which are not only conducive to absorbing carbon dioxide, nitrogen oxides, ammonia, sulfur dioxide and other substances, but also can effectively adsorb dust and suspended small particles through their own forest trees [47]. According to statistical data, urban PM2.5 mass concentrations are lowest within vegetated areas and highest in non-vegetated areas in urban areas [48]. The PM2.5 mass concentration inside the forest decreases faster than outside the forest, and the PM2.5 mass concentration on the roads protected by forests is smaller than the air PM2.5 mass concentration on the external roads [49,50]. This illustrates the ability of urban forest vegetation to combat pollution and reduce haze. Practice has shown that the large-scale construction of green infrastructure such as urban parks, nature reserves, and road forests is a low-cost and effective long-term way to reduce atmospheric particulate matter and clean the atmosphere [51]. Accordingly, the following hypothesis is proposed:
H2: 
The NFCP can reduces haze pollution by developing urban forests and improving green coverage in built-up areas.
The NFCP reduces haze pollution by raising social awareness of environmental protection. The State Forestry and Grassland Administration has included indicators related to cultivating public awareness of environmental protection, providing ecological culture sites, strengthening nature education for all, and promoting ecological culture in the Forest City Development Plan, so the NFCP will greatly increase society’s awareness of environmental protection. For urban residents, it is conducive to the practice of a “green” lifestyle and the concept of ecological civilization [52]. For the government, it will increase investment in environmental protection and implement stricter environmental regulations [53]. The increase in social awareness of environmental protection will force enterprises to make technological innovations and stimulate them to develop and apply green production models, emission equipment, and pollution control technologies to reduce the cost of environmental pollution and thus achieve the goal of haze reduction [54]. Accordingly, the following hypothesis is proposed:
H3: 
The NFCP reduces haze pollution by raising social awareness of environmental protection.
The NFCP promotes urban haze reduction through industrial structure adjustment. Cities can take the opportunity of national forest city construction and implement the new development concept to effectively promote the optimization and upgrading of industrial structures, thus producing environmental improvement effects. On the one hand, many resource-based cities and old industrial bases have adopted green development strategies, the development of the green economy, to promote the transformation of industry from underground to above ground, from resource consumption to environmentally friendly transformation, as part of the “three high” enterprises in the “pollution sanctuary” driven by the initiative to transfer industry out of the national forest city construction area [55]. On the other hand, the NFCP has cultivated a number of forest-based green industries through planning guidance, project support, science and technology drive, and policy support, forming a forest ecological system and green industrial system that is compatible with urban and rural development, such as forest tourism, forest recreation, flower cultivation, green forest products, etc. [56]. The correlation and spillover effects of such industries can provide strong support for the green development of the city as well as reduce the haze pollution in the city. In addition, the success of the NFCP directly expands the green spaces of the city, improves the soft environment for regional investment and attracts investment from foreign enterprises [57]. A high-quality habitat environment will also attract high-end talents and intellectual capital to gather, which will not only promote the high-quality development of the city economy, but also lead to the upgrading of the industrial structure, which in turn will affect the level of haze pollution in the city [58]. Accordingly, the following hypothesis is proposed:
H4: 
The NFCP reduces haze pollution by adjusting the urban industrial structure.

3. Materials and Methods

3.1. Data and Variables

The data on urban economic and social indicators used in this paper were obtained from the China Urban Statistical Yearbook (2002–2021), the China Regional Economic Statistical Yearbook (2002–2021), the Wind database, and the work reports of each city government. For a small number of prefecture-level cities with missing data in some years, the average growth rate method was used to fill them in. Since this paper studies the haze-reduction effect of national forest cities at the prefecture level, four municipalities directly under the central government, including Beijing, Tianjin, Shanghai, and Chongqing, were excluded. In addition, cities with serious data deficiencies such as Hong Kong, Macao, Taiwan, and Tibet Autonomous Region, as well as cities such as Chaohu and Laiwu whose administrative divisions changed during the study period, were excluded from the study sample. The list of national forest cities was obtained from official documents published by the State Forestry and Grassland Administration and the National Greening Committee. The final study sample was determined to be 263 cities at prefectural level and above in China, including 157 cities in the treatment group and 106 cities in the control group. The value variables involved were converted to the actual values using the consumer price index with 2001 as the base period, and foreign capital was used to convert to RMB at the exchange rate of the year and then deflated. The results of the descriptive statistics of the relevant variables are shown in Table 1.
Given the purpose of this paper, our empirical estimation required haze data for each city, the development of national forest cities, and other explanatory variables.
(1)
Annual average concentration of urban PM2.5. Haze is a collective term for the excess content of various suspended particles in the atmosphere, and PM2.5 is the main component of haze, which has the greatest impact on human health and the quality of the atmosphere; therefore, the key to managing haze is to solve the PM2.5 problem [59]. This paper used ArcGis 10.5 software to extract the raster data of global annual average PM2.5 concentration published by Columbia University to obtain the annual average PM2.5 concentrations of 263 cities in China, which were used to indicate the degree of urban haze pollution.
(2)
Core explanatory variable: whether the NFC was created successfully or not (DID = treat∗time). Where treat is a group dummy variable, if the NFCP is successful, then treat will take the value of 1 and be included in the experimental group, otherwise treat will take the value of 0 and be included in the control group. time is a time dummy variable, and the value of NFCP is 1 in the year of success and the following years, and the value of the rest of time variables is 0. Since Guiyang succeeded in creating China’s first national forest city in 2004, a total of 219 cities in China have succeeded in building national forest cities up to now. In view of the fact that the research subjects of this paper are prefecture-level cities, and some national forest cities are county-level cities or district and county-level cities, they are not taken into account, and the cities whose administrative levels have changed during the study period (such as Chaohu and Laiwu) and Hong Kong, Macao and Taiwan are also not included; 263 city samples were finally collated.
(3)
Three mediating variables. First, we used the greening coverage of urban built-up areas as a proxy variable for urban forestry. Second, we used the harmless disposal rate of domestic waste to measure the social awareness of environmental protection. Third, we used the share of tertiary industry value added in GDP to describe the degree of industrial structure shift.
(4)
In the empirical test of the impact of the NFCP on urban PM2.5 concentrations, a series of factors affecting urban PM2.5 concentrations were also controlled for, including: urban economic development (lnagdp), secondary industry development (indu), level of urbanization (urb), external openness (fdi), technological innovation (te), energy intensity (en), transportation intensity (tra), population density (lnpopden) [60,61].
Table 1 gives the descriptive statistical analysis of each variable, including the name, definition, observed value, mean, standard deviation, minimum and maximum values of these variables.

3.2. The Benchmark Regression Model

The NFCP, which began in 2004, was born out of the policy shock of haze pollution, so the NFCP can be seen as a quasi-natural experiment. The standard DID requires that all individuals in the treatment group start to receive the policy shock at exactly the same point in time. Time-varying DID is applicable to the case where the processing-period time points of individuals are not exactly the same. Considering that the NFCP is promoted in batches year by year, this paper uses a time-varying DID model to identify the impact of NFCP on urban haze [62]. The time-varying DID regression model is as follows.
l n P M 2.5 i t = α + β D I D i t + i = 1 N γ j X i t + C i t y f e + Y e a r f e + δ i t
where l n P M 2.5 i t denotes the annual average PM2.5 concentration in city i in year t. D I D i t is the interaction term of two dummy variables treat and time, i.e., D I D i t = t r e a t × t i m e , and its coefficient β measures the net effect of NFCP on urban PM2.5, X i t denotes control variables, specifically including urban economic development, secondary industry development, urbanization level, openness to the outside world, technological innovation, energy emission intensity, transportation intensity, population density and other variables; N is the number of control variables. C i t y f e , Y e a r f e , and δ i t denote city fixed effects, year fixed effects, and random disturbance terms, respectively.

3.3. Parallel Trend Test Model

The time-varying DID model requires the same trend of change between the experimental and control groups before the policy occurs, i.e., it satisfies the parallel trend test hypothesis [63]. To verify this, it is necessary to test whether there is a significant systematic difference in the trends between the experimental and control group cities before and after the NFCP. In this paper, parallel trend tests are conducted using the event study method with 2003 as the base period. The test formula is as follows:
l n P M 2.5 i t = c 0 + k = 3 16 c k P i t k + i = 1 N γ j X i t + C i t y f e + Y e a r f e + σ i t
In Formula (2), P i t k is the core explanatory variable, k ≠ 0 and k is an integer, and other variables are set in accordance with Formula (1); when t c r e a t e _ y e a r i 16 , P i t 9 = 1 , otherwise P i t 9 = 0 ; when t c r e a t e _ y e a r i 3 , P i t 3 = 1 , otherwise P i t 3 = 0 ; when t c r e a t e _ y e a r i = k , P i t k = 1 , otherwise P i t k = 0 .

3.4. Mechanism Test Model

Based on the previous theoretical analysis, to test the mechanism of the role of the NFCP in influencing the degree of urban haze pollution, we selected three mediating variables, urban forest, social environmental awareness and industrial transfer [64], and constructed the following mediating effect model.
M e d i a t o r i t = α 2 + β 2 D I D i t + i = 1 N γ 2 X i t + ε 2 i + μ 2 t + σ 2 i t
l n P M 2.5 i t = α 3 + β 3 D I D i t + λ M e d i a t o r i t + i = 1 N γ 3 X i t + ε 3 i + μ 3 t + σ 3 i t
Among these, the mediating variable M e d i a t o r i t denotes the garden green area, social environmental awareness, and industrial structure of the city. To eliminate the effect of the magnitude, the index data are logarithmically processed. If the coefficient β 2 and the coefficient λ are both significant, the mediating effect holds. Further, if β 3 is also significant and has the same sign as  β 2 × λ , then M e d i a t o r i t has a partial mediating effect, and its contribution to the total effect is ( β 2 × λ ) / ( β 2 × λ + β 3 ) . In addition, to ensure the robustness of the empirical results, the bootstrap method is used to verify the existence of mediating effects.

4. Results

4.1. Benchmark Regression

In this paper, the DID model was used to identify the impact of NFCP on PM2.5, and the benchmark results are shown in Table 2. The coefficients of the core explanatory variable DID were found to be significantly negative regardless of whether fixed effects were included or not, indicating that the NFCP is conducive to reducing urban PM2.5 concentration. The coefficient of DID in column (4) was −0.023, which passed the 1% significance level test, indicating that the implementation of the NFCP can reduce the urban PM2.5 concentration by 2.30%, i.e., hypothesis H1 is valid.
The regression results in column (4) were used as a benchmark to observe the control variables. The estimated coefficient of lnagdp was negative and passed the significance test at the 5% level, reflecting that an increase in the level of economic development reduces PM2.5. The estimated coefficient of urb was positive and passed the significance test at the 10% level, reflecting that the increase in urbanization level brings about an increase in PM2.5, which is consistent with the fact that urbanization in China is in the medium-term acceleration stage. The estimated coefficient of te was negative and passed the test at the 1% level, indicating that cities with higher levels of technology will have lower PM2.5. The estimated coefficient of en was positive and passed the test at the 5% level, implying that the more energy is consumed, the higher its PM2.5 will be. The coefficient of Lnpopden was significantly negative, which means that population agglomeration is beneficial for reducing PM2.5 level. This indicates that the “clustering” effect of population in China’s cities can effectively reduce urban haze pollution.
The parallel test is shown in Figure 1, It can be seen that the difference between urban PM2.5 in the control and experimental groups before the policy implementation was not significantly different from the difference between the two in the base period (2003), satisfying the prerequisites for using the DID model. In addition, although the coefficient of lnPM2.5 was not significant in the years after the successful creation of national forest cities, the magnitude of the coefficient decreases with time, which indicates that the creation of NFC has a significant negative effect on PM2.5.

4.2. Robustness Tests

(1) Handling selection bias: PSM-DID. Since the NFCP is not a natural experiment in the strict sense, the haze-reduction effect of the NFCP separated by the double-difference method may still have the problem of self-selection, so this paper used the panel data transformation method and the period-by-period matching method in turn, to further test the robustness of the multi-time PSM-DID model. First, the level of urban economic development, secondary industry development, urbanization level, openness to the outside world, technological innovation, energy intensity, transportation intensity, and population density were set as matching variables. Then two new data sets were obtained according to cross-sectional PSM and period-by-period PSM, respectively, and the two data sets were tested for balance and analyzed for matching effects. In the cross-sectional PSM, the nearest neighbor matching method was used to generate the optimal control group satisfying the common support condition for all cities with successful NFC, and the non-common support part was excluded; while in the period-by-period matching method, the data obtained from year-by-year matching according to the characteristics of the city sample were combined vertically to generate the panel data required for regression. Finally, the haze-reduction effect of the NFCP was re-estimated using the time-varying DID method. The regression results of PSM-DID for both methods are shown in columns (1)–(2) of Table 2, The coefficient of DID was still significantly negative, which is consistent with the conclusion that “ NFCP can reduce urban PM2.5 concentration”.
(2) Replace the low haze measurement indicator. In order to avoid the bias of the results caused by the way the explanatory variables are measured, this paper used PM10 and industrial sulfur dioxide removal as proxies to re-run the robustness test, and the results are shown in columns (3)–(4) in Table 3. It was found that the NFCP has had a significant negative effect on urban PM10 and a significant positive effect on industrial sulfur dioxide removal, indicating that NFCP significantly reduces urban PM10 concentration and has a positive effect on urban industrial sulfur dioxide removal, which also indicates the robustness of the baseline regression results.
(3) Excluding other policy interference. During the study period, the construction of comprehensive demonstration cities for energy saving and emission reduction fiscal policies, which was carried out in three batches after 2011, and the Action Plan for the Prevention and Control of Atmospheric Pollution (the “Ten Atmospheric Articles”) issued by the State Council in 2013 were closely related to urban PM2.5 concentrations. To control the impact of these two policies on the estimation results as much as possible, the dummy variables of the year of implementation of the two policies, namely, the comprehensive demonstration cities of energy-saving and emission reduction fiscal policies and the “Ten Atmospheric Articles”, were added to the baseline regression model in turn. Columns (5)–(6) of Table 3 are the estimated results after excluding the above-mentioned policy interference. It was found that the core explanatory variable coefficient DID was still significantly negative after controlling for both types of policies, which indicates to some extent that the NFCP has a significant haze-reduction effect.
(4) Placebo test. Although the city characteristic variables have been effectively controlled in this paper, the assessment results of the haze-reduction effect of the NFCP may still be influenced by some unobserved city characteristic factors. Therefore, further placebo tests of the model were needed. A randomly selected number of cities equal to the number of successful NFC were selected from all sample cities as the experimental group for the placebo test [65]. The specific approach was as follows, using Stata software to randomly select 158 cities as the experimental group, and the policy time was randomly given, repeated 500 times, and the dummy variable l n P M 2.5 _ C r e a t e r a n d o n obtained from 500 groups was regressed, and the regression coefficients were distributed as follows. The distribution of regression coefficients is shown in Figure 2. It was found that the β r a n d o n values generated during the randomization process showed a normal distribution with 0 as the mean, and the vast majority of coefficients had p-values higher than 0.1, while the estimated coefficient of the actual policy was −0.023, which is significantly different from the placebo test results. This could also exclude to some extent that the baseline regression results above were caused by other unobservable factors.

4.3. Heterogeneity Analysis

The heterogeneity test in this paper focuses on answering the following questions: Does the haze-reduction effect of the NFCP differ significantly depending on the population size of the city? Does the effect of NFCP differ depending on the geographical location of cities? Is there any difference in the haze-reduction effect of NFCP in different forest city groups?

4.3.1. Urban Population Size Heterogeneity Analysis

The haze-reduction effects of NFCP in cities with different population sizes were investigated by dividing the sample cities into six groups according to their population sizes and conducting regressions on each of the six groups of sample cities based on China’s 2014 Notice on Adjusting the Criteria for City Size Classification. Since the number of small-scale cities was small and the regression results unreliable, only the regression results for cities of medium scale or above are reported, as shown in columns (1)–(5) of Table 4.
The coefficients of DID were all significantly negative, and the absolute values of the coefficients were larger for the sample group of large cities, indicating that the NFCP has a stronger inhibitory effect on haze pollution in large cities. The main reasons for this phenomenon are: firstly, that larger cities have economic agglomeration effects and relatively more efficient resource allocation and utilization, which, combined with stricter environmental regulations and a stronger willingness to improve the environment, make it easier to implement various environmental protection measures in large cities. Secondly, industries in large cities are dominated by service or high-end manufacturing industries, which are less polluting than those in heavy industrial cities. Thirdly, the role of selection of large cities is very significant; only those companies with high production efficiency, advanced technology, environmental protection to meet the requirements, stand firm in the locale, objectively playing a positive role in reducing the city’s haze pollution. Further studies have shown that a larger urban population size is not better, and that the haze-reduction effect of the NFCP is strongest in large Type II cities, and relatively weak in mega cities and above. This is related to the function of the city, Type II large cities are mainly tourist cities, the rapid development of wholesale trade, in Type I large cities, manufacturing is more concentrated, and mega cities and above are mainly provincial capitals, the crowding effect of a large-scale concentration of population may offset part of the NFCP to reduce the haze effect. From this conclusion, it is clear that the key to combating PM2.5 does not lie in the size of the urban population, but in the innovation of urban governance, industrial upgrading and technological advancement.

4.3.2. Heterogeneity Analysis of Urban Geo-Economic Location

According to the geographic location of cities, the cities were divided into four groups of east, central, west and northeast for regression, so as to explore the geographic location heterogeneity of the haze-reduction effect of the NFCP, and the results are shown in columns (1)–(4) of Table 5.
The regression results of the subgroups show that the regression coefficients of DID were all significantly negative, and the NFCP had the greatest inhibiting effect on haze pollution in the eastern region, while the inhibiting effect on haze pollution in the northeastern region was relatively weak. The reason for this is that, as a kind of green infrastructure, the construction of an urban forest, in addition to relying on certain natural conditions, requires a large amount of capital investment for its construction and maintenance. Due to its waterfront location and long-term capital accumulation, the eastern region has traditionally been the center of economic development in China. Under the guidance of the new development concept, it is more capable and willing to improve the ecological environment of the city. As a heavy industrial base in China, the northeast region is subject to industrial transformation and places more emphasis on economic revitalization, which may cause a lack of investment in its green infrastructure and affect the construction efforts of urban forestry. The difference in the effect of the NFCP on haze reduction in central and western regions may be due to the fact that the Central Rising Strategy has promoted the upgrading and transformation of the industrial structure in the central region, thus reducing the degree of haze pollution in the central region, while the western region has taken over most of the polluting industries transferred from the east and central regions.
The Hu Huanyong line is a geographical dividing line between population development level and economic and social patterns in China, and there are obvious differences in population density, urbanization level and economic development level between its east and west sides. In this regard, this paper investigates the impact of NFCP on PM2.5 in cities on the east and west sides of the Hu Huanyong line. From the estimated results in columns (5)–(6) of Table 5, it is clear that the haze-reduction effect of the NFCP on PM2.5 in the cities east of the Hu Huanyong line was significantly higher than that in cities west of it. The reason for this is that the haze pollution is heavier in the eastern part of Hu Huanyong line and relatively less in the western part, while the difference in the level of economic development directly leads to the difference in the investment in the construction of NFCP on both sides, which is eventually reflected in the difference in the effect of haze reduction.

4.3.3. Heterogeneity of Forest City Clusters

China’s “13th Five-Year Plan” forestry development plan clearly proposed the building of six national forest city clusters in Beijing-Tianjin-Hebei, the Pearl River Delta, the Yangtze River Delta, Chang-Zhu-Tan, Central Plains, and Guanzhong-Tianshui by 2020, and currently, the Pearl River Delta national forest city cluster has passed the national assessment and acceptance and is officially completed. In this context, this paper analyzed the haze-reduction effects of the NFCP in different forest city clusters, as shown in columns (1)–(6) of Table 6.
The NFCP can significantly reduce the haze-pollution level in the Pearl River Delta, Yangtze River Delta, and Central Plains urban agglomerations, while the haze-reduction effect on the Beijing-Tianjin-Hebei, Chang-Zhu-Tan, and Guanzhong-Tianshui urban agglomerations is not obvious, and the haze control effect of the NFCP on the Pearl River Delta region is significantly better than other urban agglomerations. The possible explanation lies in the different effects of the NFCP on the effect of haze control within urban agglomerations, which is closely related to the leading industries and the state of urban forest construction in urban agglomerations. The modern service and manufacturing industries in the Pearl River Delta city cluster are very well developed, and it has a relatively low proportion of heavy industry, a high degree of urbanization and forest coverage, and relatively good urban greening work, so compared with other city clusters, the effect of the NFCP on haze-pollution suppression is most significant [66]. The Yangtze River Delta urban agglomeration is dominated by light industries such as textiles, electronics and foodstuffs, and the concentration of innovative resources has contributed to the improvement of the regional ecological environment by optimizing the industrial structure [67]. The Central Plains city cluster has taken over part of the industrial transfer from Beijing, Tianjin, Hebei, Shandong and other regions, but by building a “three screens and four corridors” ecological network, creating a “green ecological development demonstration area”, and compiling urban air quality deadlines to meet the plan, the regional ecological environment has been effectively improved [68]. The Beijing-Tianjin-Hebei city cluster is China’s influential heavy chemical industry base; the energy consumption structure is coal-based, coupled with more coal burning in the autumn and winter seasons, compared with other cities, The haze-reduction effect of the NFCP is not significant [69]. Despite the accelerated pace of regional integration development of the Chang-Zhu-Tan urban agglomeration, the degree of its economic development determines the low effect of industrial transfer, and the forest coverage rate is lower than the average level of Hunan Province, so there is much room for progress in the construction of national forest cities [70]. The Guanzhong-Tianshui Economic Zone is dominated by non-ferrous metal processing, building materials, coal and other industries, and lacks a coordinated policy for regional environmental protection and development, and the haze problem is more prominent [71].

4.4. Mechanism Test

Table 7 shows the test results of the mediating mechanism. The regression coefficient of NFCP was significantly positive when urban forest, environmental awareness and industrial structure were the explanatory variables, and the coefficient was significantly negative after including each mediating variable and the NFCP into the model regression at the same time. The confidence interval of the bootstrap test for each mediating variable did not include 0, which indicates that the NFCP can improve urban greenery coverage, enhance social awareness of environmental protection, promote industrial structure transformation, and thus reduce haze pollution. Hypothesis H2, Hypothesis H3, and Hypothesis H4 are all verified.
From the perspective of urban forest, the construction of NFC can promote the construction of urban forest, which is mainly reflected in the increase in forest area, recreational parks and urban green coverage. Urban forest can not only purify the air pollution, but also this effect will be more obvious as the trees grow or the forest area increases. In terms of environmental awareness, the construction of national forest cities helps strengthen the concept of ecological civilization among residents. Firstly, the construction of national forest cities requires cities to be equipped with a certain number of popular science places, such as science museums in parks or nature reserves that are open to the public for free; secondly, the construction of NFC requires ecology-related activities to be held at least five times a year, and voluntary tree planting activities to be organized for city residents at regular intervals. Finally, with the development of forest cities, the increase in green space per capita and the improvement of living conditions, the requirements for the environment will naturally become higher and higher. In terms of industrial structure, China does propose in the Forest City Development Plan to establish an ecological barrier and develop and expand forest-related green industries, such as forest farming and flower cultivation. On the one hand, stopping indiscriminate logging and continuing to expand ecological spaces; on the other hand, developing greening service industries such as forest tourism, recreation and oxygen bars according to the existing forest resources.

5. Discussion

This paper used a time-varying DID model to find out whether the NFCP could promote urban haze reduction, and whether there is regional heterogeneity in this haze-reduction effect. Further, using a mediating effect model, we found that the NFCP can promote urban haze reduction by increasing urban forest coverage, enhancing residents’ awareness of environmental protection, and promoting the development of tertiary industries.
For the haze-reduction effect of the NFCP, we think the government management should actively play the guiding role in national forest city construction, and implement zoning classification and grading management for the cities according to the natural conditions, leading functions and economic location. The haze-reduction effect of the NFCP is heterogeneous, and each city needs to build a unique national forest city construction system by taking into account its own geographical features, location conditions and existing economic status, and formulate a forest city development plan that meets the actual local conditions. Relevant regulatory regulations are also essential, and the management department strengthens the supervision, inspection and guidance services for the NFCP through hard regulation and soft constraints, adjusts the NFC construction measures in a timely manner, and may adopt withdrawal mechanisms when necessary.
The development of an urban forest has a haze-reduction effect. As an important green infrastructure, urban forests should be incorporated into the planning and construction of cities along with urban buildings, brownfields, squares and other facilities. In the choice of urban greening methods, we can choose the urban greening mode of mainly trees, evergreen and native species, and actively explore three-dimensional greening methods in the form of roofs, walls and bridges to organically combine the forest and the urban space.
The improvement of social awareness of environmental protection can also promote urban haze reduction. With the development of the Internet, various new ways to use publicity have emerged. We can use new media such as Facebook and WeChat to carry out environmental protection publicity, making them important places to carry out environmental protection publicity. Residents themselves should develop a sense of ownership and realize that they are the ultimate beneficiaries of a protected urban environment.
For forest city construction, we should not only let it have the effect of haze reduction, but also bring about sustainable development of the urban economy. City managers should take the construction of a national forest city as an opportunity to continuously optimize the forestry industry structure in the region. On the one hand, local governments need to build an interconnected ecological forest network system according to the natural geographical conditions, create forest towns with urban cultural characteristics, forest parks, forest homes, forest oxygen bars and other bases, and on the other hand, they should develop ecological industries such as planting, breeding, tourism, vacation, leisure and recreation based on the forest resources in the region to promote the green growth of the urban economy.
In future research, we will focus on micro studies of the impact of national forest cities on the perceptions and behaviors of urban residents. For example, questions such as which aspects or types of forest city construction will affect the physical and psychological health of urban residents; and what kind of forest city (quantity, quality, and configuration of types and spaces) can accommodate the interests of different social groups in an urban society with changing demographic characteristics.

6. Conclusions

This paper used the NFCP as a quasi-natural experiment to estimate the impact and mechanism of action of the NFCP on urban PM2.5, based on the panel data of 263 prefecture-level cities in China from 2001 to 2020, using a time-varying DID model and a mediating effects model, and found that:
(1)
Overall nationwide, the construction of NFC is beneficial in reducing urban PM2.5 concentrations, with an average decrease of 2.25% in PM2.5 concentrations in cities with a successful NFC; a finding that remains robust after placebo tests and replacement of explanatory variables, respectively.
(2)
The haze-reduction effect of the NFCP is heterogeneous, and the haze-reduction effect of the NFCP is more significant for the Pearl River Delta urban agglomeration, Type II large cities, cities in the eastern region and cities east of the Hu Huanyong line.
(3)
The NFCP promotes urban haze reduction by increasing the area of urban green spaces, raising residents’ awareness of environmental protection, and promoting the shift of industrial structure to tertiary industries.

Author Contributions

C.H.: data curation, original draft, visualization, formal analysis, writing—reviewing and editing. J.C.: supervision, methodology, software, validation. J.D.: conceptualization, writing, investigation, funding acquisition. All authors have read and agreed to the published version of the manuscript.

Funding

This research is funded National Key Fund for Philosophy and Social Sciences, grant number 22AKS017.

Data Availability Statement

Data will be available if necessary.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Zhang, Y.; Chen, J. An empirical study of the efficiency of haze pollution governance in Chinese cities based on streaming data. Sci. Total. Environ. 2020, 739, 139571. [Google Scholar] [CrossRef]
  2. Bonilla-Duarte, S.; González, C.C.; Rodríguez, L.C.; Jáuregui-Haza, U.J.; García-García, A. Contribution of Urban Forests to the Ecosystem Service of Air Quality in the City of Santo Domingo, Dominican Republic. Forests 2021, 12, 1249. [Google Scholar] [CrossRef]
  3. Zhang, Y.; Zhang, T.; Zeng, Y.; Cheng, B.; Li, H. Designating National Forest Cities in China: Does the policy improve the urban living environment? For. Policy Econ. 2021, 125, 102400. [Google Scholar] [CrossRef]
  4. Li, X.; Zhao, C. Can national forest city construction mitigate air pollution in China? Evidence from a quasi-natural experiment. Environ. Geochem. Health 2022. [Google Scholar] [CrossRef]
  5. Liu, H.M. Research on the key work of national forest city construction. For. Econ. 2018, 40, 77–79. (In Chinese) [Google Scholar]
  6. Zhang, J.G.; Ge, Y. Comparative study on Spatial distribution and influencing factors of National Forest City and National Garden City. J. Nanjing For. Univ. 2022, 46, 41–49. (In Chinese) [Google Scholar]
  7. Zhang, Q.; Yang, L.; Xu, S. The Relationships of Supporting Services and Regulating Services in National Forest City. Forests 2022, 13, 1368. [Google Scholar] [CrossRef]
  8. Peng, J.S.; Zhu, K.; Xu, H.; Huang, X.Y.; Dai, J.P.; Huang, R.J. On the Development of mountainous Forest cities in Yunnan Plateau. For. Resour. Manag. 2021, 1, 23–29. (In Chinese) [Google Scholar]
  9. Li, X.; Guo, J.; Qi, S. Forestland landscape change induced spatiotemporal dynamics of subtropical urban forest ecosystem services value in forested region of China: A case of Hangzhou city. Environ. Res. 2020, 193, 110618. [Google Scholar] [CrossRef]
  10. Chen, C.M.; Chen, G.C.; Liang, H.S. Planning and Exploration of high-quality forest city construction in Guangdong-Hong Kong-Macao Greater Bay Area: A case study of Foshan City. For. Sci. Technol. Inf. 2021, 53, 19–23. (In Chinese) [Google Scholar]
  11. Zhang, Z. Research on the dilemma and countermeasures of green development in cities with declining forest resources—A case study of Yichun City. Econ. Res. Ref. 2016, 63, 68–73. (In Chinese) [Google Scholar]
  12. Şevik, H.; Cetin, M.; Belkayalı, N. Effects of Forests on Amounts of CO2: Case Study of Kastamonu and Ilgaz Mountain National Parks. Pol. J. Environ. Stud. 2015, 24, 253–256. [Google Scholar] [CrossRef] [PubMed]
  13. Chen, W.Y.; Li, X. Urban forests’ recreation and habitat potentials in China: A nationwide synthesis. Urban For. Urban Green. 2021, 66, 127376. [Google Scholar] [CrossRef]
  14. Pregitzer, C.C.; Charlop-Powers, S.; Bradford, M.A. Natural Area Forests in US Cities: Opportunities and Challenges. J. For. 2021, 119, 141–151. [Google Scholar] [CrossRef]
  15. Wang, Y.Z.; Tong, G.R.; Zhang, S.H.; Wang, W.F.; Cheng, Z.P.; Chen, H.F. Study on the effect and development of Forest City Construction in Fuzhou City—A case study of establishing National Forest City. J. Anhui Agric. Sci. 2020, 48, 104–107. (In Chinese) [Google Scholar]
  16. Xu, C.; Dong, L.; Yu, C.; Zhang, Y.; Cheng, B. Can forest city construction affect urban air quality? The evidence from the Beijing-Tianjin-Hebei urban agglomeration of China. J. Clean. Prod. 2020, 264, 121607. [Google Scholar] [CrossRef]
  17. Guo, J.; Xiong, Y.; Shi, C.; Liu, C.; Li, H.; Qian, H.; Sun, Z.; Qin, C. Characteristics of airborne bacterial communities in indoor and outdoor environments during continuous haze events in Beijing: Implications for health care. Environ. Int. 2020, 139, 105721. [Google Scholar] [CrossRef]
  18. Guan, Y.; Zhai, Z.; Wang, Y.; Wu, D.; Yu, L.; Lei, Z. Foreign direct investment, environmental regulation, and haze pollution: Empirical evidence from China. Environ. Sci. Pollut. Res. 2022, 29, 27571–27584. [Google Scholar] [CrossRef]
  19. Jiang, X.; Fu, W.; Li, G. Can the improvement of living environment stimulate urban Innovation?—Analysis of high-quality innovative talents and foreign direct investment spillover effect mechanism. J. Clean. Prod. 2020, 255, 120212. [Google Scholar] [CrossRef]
  20. Gan, T.; Yang, H.; Liang, W. How do urban haze pollution and economic development affect each other? Empirical evidence from 287 Chinese cities during 2000–2016. Sustain. Cities Soc. 2021, 65, 102642. [Google Scholar] [CrossRef]
  21. Wei, L.-Y.; Liu, Z. Air pollution and innovation performance of Chinese cities: Human capital and labour cost perspective. Environ. Sci. Pollut. Res. 2022, 29, 67997–68015. [Google Scholar] [CrossRef] [PubMed]
  22. Li, B.; Guo, P.; Zeng, Y. The Impact of Haze on the Availability of Company Debt Financing: Evidence for Sustainability of Chinese Listed Companies. Sustainability 2019, 11, 806. [Google Scholar] [CrossRef] [Green Version]
  23. Zhang, F.; Li, Y.; Li, Y.; Xu, Y.; Chen, J. Nexus among air pollution, enterprise development and regional industrial structure upgrading: A China’s country panel analysis based on satellite retrieved data. J. Clean. Prod. 2022, 335, 130328. [Google Scholar] [CrossRef]
  24. Liu, X.; Zhao, T.; Chang, C.-T.; Fu, C.J. China’s renewable energy strategy and industrial adjustment policy. Renew. Energy 2021, 170, 1382–1395. [Google Scholar] [CrossRef]
  25. Ma, T.; Cao, X. The effect of the industrial structure and haze pollution: Spatial evidence for China. Environ. Sci. Pollut. Res. 2022, 29, 23578–23594. [Google Scholar] [CrossRef]
  26. Zhang, M.; Liu, X.; Ding, Y. Assessing the influence of urban transportation infrastructure construction on haze pollution in China: A case study of Beijing-Tianjin-Hebei region. Environ. Impact Assess. Rev. 2021, 87, 106547. [Google Scholar] [CrossRef]
  27. Li, X.; Zhou, M.; Zhang, W.; Yu, K.; Meng, X. Study on the Mechanism of Haze Pollution Affected by Urban Population Ag-glomeration. Atmosphere 2022, 13, 278. [Google Scholar] [CrossRef]
  28. Ye, Y.; Ye, S.; Yu, H. Can Industrial Collaborative Agglomeration Reduce Haze Pollution? City-Level Empirical Evidence from China. Int. J. Environ. Res. Public Health 2021, 18, 1566. [Google Scholar] [CrossRef]
  29. Yang, Y.; Tang, D.; Yang, X. Investigating the spatio-temporal variations of the impact of urbanization on haze pollution using multiple indicators. Stoch. Environ. Res. Risk Assess. 2021, 35, 703–717. [Google Scholar] [CrossRef]
  30. Fan, F.; Cao, D.; Ma, N. Is Improvement of Innovation Efficiency Conducive to Haze Governance? Empirical Evidence from 283 Chinese Cities. Int. J. Environ. Res. Public Health 2020, 17, 6095. [Google Scholar] [CrossRef]
  31. Feng, Y.; Hu, S. The Effect of Smart City Policy on Urban HazePollution in China: Empirical Evidencefrom a Quasi-Natural Experiment. Pol. J. Environ. Stud. 2022, 31, 2083–2092. [Google Scholar] [CrossRef]
  32. Che, S.; Wang, J. Digital economy development and haze pollution: Evidence from China. Environ. Sci. Pollut. Res. 2022, 29, 73210–73226. [Google Scholar] [CrossRef] [PubMed]
  33. Baumgardner, D.; Varela, S.; Escobedo, F.J.; Chacalo, A.; Ochoa, C. The role of a peri-urban forest on air quality improvement in the Mexico City megalopolis. Environ. Pollut. 2012, 163, 174–183. [Google Scholar] [CrossRef] [PubMed]
  34. Ding, J.; Lun, X.; Ma, W.; Zhao, L.; Cao, Y.; Sun, F.; Li, R. Deposition of water-soluble inorganic ions in PM2.5 in a typical forestry system in Beijing, China. For. Ecosyst. 2018, 5, 36. [Google Scholar] [CrossRef]
  35. Zhou, Y.; Liu, H.; Zhou, J.; Xia, M. Simulation of the Impact of Urban Forest Scale on PM2.5 and PM10 based on System Dynamics. Sustainability 2019, 11, 5998. [Google Scholar] [CrossRef] [Green Version]
  36. Carretero, E.M.; Moreno, G.; Duplancic, A.; Abud, A.; Vento, B.; Jauregui, J.A. Urban forest of Mendoza (Argentina): The role of Morus alba (Moraceae) in carbon storage. Carbon Manag. 2017, 8, 237–244. [Google Scholar] [CrossRef]
  37. Liao, L.; Zhao, C.; Li, X.; Qin, J. Towards low carbon development: The role of forest city constructions in China. Ecol. Indic. 2021, 131, 108199. [Google Scholar] [CrossRef]
  38. Tapak, L.; Abbasi, H.; Mirhashemi, H. Assessment of factors affecting tourism satisfaction using K-nearest neighborhood and random forest models. BMC Res. Notes 2019, 12, 2019. [Google Scholar] [CrossRef] [Green Version]
  39. Olsson, O. Changed availability of urban fringe forests in Sweden in 2000–2010. Scand. J. For. Res. 2013, 28, 386–394. [Google Scholar] [CrossRef]
  40. Ostoić, S.K.; van den Bosch, C.C.K.; Vuletić, D.; Stevanov, M.; Živojinović, I.; Mutabdžija-Bećirović, S.; Lazarević, J.; Stojanova, B.; Blagojević, D.; Stojanovska, M.; et al. Citizens’ perception of and satisfaction with urban forests and green space: Results from selected Southeast European cities. Urban For. Urban Green. 2017, 23, 93–103. [Google Scholar] [CrossRef]
  41. Shi, T.; Liu, M.; Hu, Y.; Li, C.; Zhang, C.; Ren, B. Spatiotemporal Pattern of Fine Particulate Matter and Impact of Urban So-cioeconomic Factors in China. Int. J. Environ. Res. Public Health 2019, 16, 1099. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  42. Zhao, X.; Zhou, W.; Han, L.; Locke, D. Spatiotemporal variation in PM(2.5) concentrations and their relationship with so-cioeconomic factors in China’s major cities. Environ. Int. 2019, 133, 105145. [Google Scholar] [CrossRef] [PubMed]
  43. He, W.; Cheng, Y.; Lin, Y.; Zhang, H. Microeconomic effects of designating National Forest Cities: Evidence from China’s publicly traded manufacturing companies. For. Policy Econ. 2022, 136, 102669. [Google Scholar] [CrossRef]
  44. Xie, H.; Tan, X.; Yang, C.; Li, C. Does Urban Forest Control Smog Pollution? Evidence from National Forest City Project in China. Sustainability 2022, 14, 12897. [Google Scholar] [CrossRef]
  45. Wang, F.; Wang, H.; Liu, C.; Xiong, L.; Qian, Z. The Effect of Green Urbanization on Forestry Green Total Factor Productivity in China: Analysis from a Carbon Neutral Perspective. Land 2022, 11, 1900. [Google Scholar] [CrossRef]
  46. Lee, D.; Choe, H. Estimating the Impacts of Urban Expansion on Landscape Ecology: Forestland Perspective in the Greater Seoul Metropolitan Area. J. Urban Plan. Dev. 2011, 137, 425–437. [Google Scholar] [CrossRef]
  47. Deng, C.; Huang, Z.; Zhang, X.; Zhao, H.; Jiang, S.; Ren, Y. Correlation between Vegetation Structure and Species Diversity in Traditional Villages in Karst Topographic Regions of the Zunyi City, China. Plants 2022, 11, 3161. [Google Scholar] [CrossRef]
  48. Xiao, Y.; Li, J.; Kuang, Y.; Tong, F.; Xi, D.; Chen, B.; Shi, X.; Pei, N.; Huang, J.; Pan, Y. Air TSP and air TSP in and outside the forest of Doctorashan in Guangzhou during rainy season. Acta Ecol. Sin. 2013, 33, 6209–6217. (In Chinese) [Google Scholar]
  49. Ren, Q.W.; Wang, C.; Qie, G.F. Study of characteristic changes of air microorganism concentration of urban green space. Chin. Urban For. 2006, 4, 34–37. (In Chinese) [Google Scholar]
  50. Chen, B.; Liu, H.L.; Zhao, D.B.; Chen, P.F.; Lu, S.W.; Li, S.N. Delayed payment of green tree species in the Xishan Mountains of Beijing in autumn PM2. 5 Ability and its relationship with leaf surface AFM characteristics. Chin. J. Appl. Ecol. 2016, 3, 777–784. (In Chinese) [Google Scholar]
  51. Liu, X.; Wang, X.; Dong, T. Impacts of Urban Shrinkage on Haze Pollution-Evidence from China. Math. Probl. Eng. 2022, 2022, 3952442. [Google Scholar] [CrossRef]
  52. Olivero-Lora, S.; Meléndez-Ackerman, E.; Santiago, L.; Santiago-Bartolomei, R.; García-Montiel, D. Attitudes toward Resi-dential Trees and Awareness of Tree Services and Disservices in a Tropical City. Sustainability 2020, 12, 117. [Google Scholar] [CrossRef] [Green Version]
  53. Kim, J.-H.; Kwon, O.-S.; Ra, J.-H. Urban Type Classification and Characteristic Analysis through Time-Series Environmental Changes for Land Use Management for 31 Satellite Cities around Seoul, South Korea. Land 2021, 10, 799. [Google Scholar] [CrossRef]
  54. Park, J.; Kim, J. Economic impacts of a linear urban park on local businesses: The case of Gyeongui Line Forest Park in Seoul. Landsc. Urban Plan. 2019, 181, 139–147. [Google Scholar] [CrossRef]
  55. Wang, Y.; Chen, H.; Long, R.; Sun, Q.; Jiang, S.; Liu, B. Has the Sustainable Development Planning Policy Promoted the Green Transformation in China’s Resource-based Cities? Resour. Conserv. Recycl. 2022, 180, 106181. [Google Scholar] [CrossRef]
  56. Li, L.; Li, F.; Tao, C.; Cheng, B. The impact of spatial agglomeration on export of forest products manufacturing in China: Evidence from enterprises’ data. J. Sustain. For. 2019, 38, 743–754. [Google Scholar] [CrossRef]
  57. Shkaruba, A.; Kireyeu, V.; Likhacheva, O. Rural-urban peripheries under socioeconomic transitions: Changing planning contexts, lasting legacies, and growing pressure. Landsc. Urban Plan 2017, 165, 244–255. [Google Scholar] [CrossRef]
  58. Zhou, H.; Liu, Y.; He, M. The Spatial Interaction Effect of Green Spaces on Urban Economic Growth: Empirical Evidence from China. Int. J. Environ. Res. Public Health 2022, 19, 10360. [Google Scholar] [CrossRef]
  59. Liu, X. Dynamic evolution, spatial spillover effect of technological innovation and haze pollution in China. Energy Environ. 2018, 29, 968–988. [Google Scholar] [CrossRef]
  60. Han, X.; Li, H.; Liu, Q.; Liu, F.; Arif, A. Analysis of influential factors on air quality from global and local perspectives in China. Environ. Pollut. 2019, 248, 965–979. [Google Scholar] [CrossRef]
  61. Zhou, L.; Dai, Y. The Influencing Factors of Haze Tolerance in China. Int. J. Environ. Res. Public Health 2019, 16, 287. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  62. Sun, W.; Li, R.; Cai, R.; Ji, Z.; Cheng, M. The impact of solar energy investment in multilateral development banks on techno-logical innovation: Evidence from a multi-period DID method. Front. Energy Res. 2023, 10, 1085012. [Google Scholar] [CrossRef]
  63. Ma, J.; Hu, Q.; Shen, W.; Wei, X. Does the Low-Carbon City Pilot Policy Promote Green Technology Innovation? Based on Green Patent Data of Chinese A-Share Listed Companies. Int. J. Environ. Res. Public Health 2021, 18, 3695. [Google Scholar] [CrossRef]
  64. Rijnhart, J.J.M.; Valente, M.J.; Smyth, H.L.; MacKinnon, D.P. Statistical Mediation Analysis for Models with a Binary Mediator and a Binary Outcome: The Differences Between Causal and Traditional Mediation Analysis. Prev. Sci. 2021. [Google Scholar] [CrossRef] [PubMed]
  65. Liu, Q.; Lu, Y. Firm investment and exporting: Evidence from China’s value-added tax reform. J. Int. Econ. 2015, 97, 392–403. [Google Scholar] [CrossRef] [Green Version]
  66. Fang, Z.; Zhang, Y.; Wang, Z.; Zhang, L. Spatial agglomeration of exhibition enterprises on a regional scale in China. Chin. Geogr. Sci. 2017, 27, 497–506. [Google Scholar] [CrossRef]
  67. Xu, H.; Jiao, M. City size, industrial structure and urbanization quality—A case study of the Yangtze River Delta urban ag-glomeration in China. Land Use Policy 2021, 111, 105735. [Google Scholar] [CrossRef]
  68. Wang, F.; Li, R.; Yu, C.; Xiong, L.; Chang, Y. Temporal-Spatial Evolution and Driving Factors of the Green Total Factor Productivity of China’s Central Plains Urban Agglomeration. Front. Environ. Sci. 2021, 9, 686725. [Google Scholar] [CrossRef]
  69. Pan, Y.; Tian, S.; Li, X.; Sun, Y.; Li, Y.; Wentworth, G.R.; Wang, Y. Trace elements in particulate matter from metropolitan regions of Northern China: Sources, concentrations and size distributions. Sci. Total. Environ. 2015, 537, 9–22. [Google Scholar] [CrossRef]
  70. Wang, L.; Ye, W.; Chen, L. Research on Green Innovation of the Great Changsha-Zhuzhou-Xiangtan City Group Based on Network. Land 2021, 10, 1198. [Google Scholar] [CrossRef]
  71. Dong, L.; Shang, J.; Ali, R.; Rehman, R.U. The Coupling Coordinated Relationship Between New-type Urbanization, Eco-Environment and its Driving Mechanism: A Case of Guanzhong, China. Front. Environ. Sci. 2021, 9, 638891. [Google Scholar] [CrossRef]
Figure 1. Parallel trend test. Note: the successive broken lines reflect the marginal effects created by national forest cities, and the dotted lines above and below are confidence intervals below the 95% level.
Figure 1. Parallel trend test. Note: the successive broken lines reflect the marginal effects created by national forest cities, and the dotted lines above and below are confidence intervals below the 95% level.
Forests 14 00703 g001
Figure 2. Distribution of estimated coefficients after random processing (500 samples). Note: the X-axis is the estimated coefficients generated randomly 500 times, the empty circle is the p-value of the estimated coefficients, and the left vertical line is the estimated coefficient of the actual policy.
Figure 2. Distribution of estimated coefficients after random processing (500 samples). Note: the X-axis is the estimated coefficients generated randomly 500 times, the empty circle is the p-value of the estimated coefficients, and the left vertical line is the estimated coefficient of the actual policy.
Forests 14 00703 g002
Table 1. Descriptive statistics of the variables.
Table 1. Descriptive statistics of the variables.
VariablesDefinitionNMeanS.DMinMax
Urban haze levelsAnnual average urban PM2.5 concentration (μg/m3)52603.7340.3601.1504.687
Development of NFCPCity is an NFC in that year = 1; otherwise = 052600.1880.3910.0001.000
Urban economic developmentExpressed in logarithmic GDP per capita (CNY)526010.1500.9014.59513.056
Development of secondary industryAdded value of secondary industry/GDP (%)52600.4710.1120.1070.910
Urbanization levelNon-agricultural employment population/total population (%)52600.3800.2050.0741.000
External opennessTotal imports and exports/GDP (%)52600.3740.5690.0003.720
Technological innovationThe number of patent applications authorized per 104 people (pieces/104 persons)52609.73026.8180.002409.448
Energy intensityIndustrial power consumption/added value of secondary industry (KWh/104 CNY)52600.3170.5960.01314.125
Transportation intensityPer capita freight volume (ton/person)526032.16938.6050.089483.247
Population densityPopulation per square kilometer (person/km2)52605.7840.8542.3037.903
Table 2. Benchmark regression results.
Table 2. Benchmark regression results.
Variables L n P M 2.5
(1)(2)(3)(4)
DID−0.206 ***−0.028 ***−0.113 ***−0.023 ***
(0.009)(0.007)(0.0123)(0.007)
lnagdp −0.063 ***−0.034 **
(0.001)(0.013)
indu 1.027 ***0.013
(0.065)(0.044)
urb 0.0840.050 *
(0.053)(0.029)
fdi −0.074 ***0.020 **
(0.013)(0.008)
te −0.0025 ***−0.001***
(0.001)(0.000)
en −0.024 ***0.009 **
(0.008)(0.004)
tra −0.001 *0.000
(0.000)(0.000)
lnpopden 0.107−0.091 **
(0.080)(0.041)
_cons3.772 ***3.638 ***3.322 ***4.422 ***
(0.002)(0.005)(0.432)(0.251)
Time FENOYESNOYES
Urban FENOYESNOYES
N5260526052605260
R20.1960.7500.4090.756
Note: The values of ***, ** and * are significant at 1, 5 and 10%, respectively. The values of t are shown in brackets and are the same in the following tables.
Table 3. Robustness tests.
Table 3. Robustness tests.
Variables(1)(2)(3)(4)(5)(6)
Cross-Section PSMYear-by-Year PSMPM10Industrial Fume RemovalEnergy Saving and Emission ReductionAtmosphere 10
DID−0.020 ***
(−2.935)
−0.016 **
(−2.346)
−0.030 *
(0.016)
0.107 **
(0.051)
−0.022 ***
(0.005)
−0.020 ***
(0.008)
fiscal_city 0.018
(0.0167)
air −0.140 ***
(0.0305)
Control variablesYESYESYESYESYESYES
_cons6.617 ***
(0.479)
11.19 ***
(1.561)
4.435 ***
(0.252)
4.422 ***
(0.251)
Time FEYESYESYESYESYESYES
Urban FEYESYESYESYESYESYES
N521749905260526052605260
R20.75670.76760.3290.1760.7560.757
Table 4. Heterogeneity analysis of urban population size.
Table 4. Heterogeneity analysis of urban population size.
City Type(1)(2)(3)(4)(5)
Medium CityLarge CityType II Large CitiesType I Large CitiesMega Cities and above
Population0.5 M< and <1 M≥1 M1 M< and ≤3 M3 M< and ≤5 M>5 M
DID−0.106 *
(0.0551)
−0.108 ***
(0.00629)
0.139 ***
(0.011)
−0.086 ***
(0.011)
0.055 ***
(0.010)
Control variablesYESYESYESYESYES
_cons3.216 ***
(0.452)
3.313 ***
(0.218)
3.493 ***
(0.322)
1.698 ***
(0.495)
3.580 ***
(0.413)
Time FEYESYESYESYESYES
Urban FEYESYESYESYESYES
N1755013164515511817
R20.2950.3970.2910.4390.492
Table 5. Heterogeneity analysis of urban geo-economic location.
Table 5. Heterogeneity analysis of urban geo-economic location.
Location(1)(2)(3)(4)(5)(6)
Eastern RegionCentral RegionWestern RegionNortheast RegionCities East of the Hu LineCities West of the Hu Line
DID−0.104 ***
(0.011)
−0.097 ***
(0.010)
−0.071 ***
(0.013)
−0.051 **
(0.023)
−0.113 ***
(0.013)
−0.080 **
(0.0334)
Control variablesYESYESYESYESYESYES
_cons4.094 ***
(0.280)
1.467 ***
(0.366)
3.268 ***
(0.326)
−1.351
(1.083)
3.607 ***
(0.468)
2.785 ***
(0.644)
Time FEYESYESYESYESYESYES
Urban FEYESYESYESYESYESYES
N1660158013406804880380
R20.4470.5610.3940.2150.4300.281
Table 6. Heterogeneity analysis of urban clusters.
Table 6. Heterogeneity analysis of urban clusters.
City Cluster(1)(2)(3)(4)(5)(6)
Pearl River DeltaBeijing-Tianjin-HebeiYangtze River DeltaChang-Zhu-TanCentral PlainsGuanzhong-Tianshui
DID−0.247 ***
(0.033)
−0.044
(0.028)
−0.086 ***
(0.016)
−0.029
(0.044)
−0.038 **
(0.016)
−0.049
(0.035)
Control variablesYESYESYESYESYESYES
_cons1.658
(1.031)
−0.462
(1.140)
2.237 ***
(0.721)
−20.76 ***
(6.184)
1.644 ***
(0.566)
1.247
(1.487)
Time FEYESYESYESYESYESYES
Urban FEYESYESYESYESYESYES
N18024050060520140
R20.5940.6690.5720.8260.5540.489
Table 7. Results of intermediate effect test.
Table 7. Results of intermediate effect test.
Variables(1)(2)(3)(4)(5)(6)
Urban ForestPM2.5Environmental AwarenessPM2.5Industry StructurePM2.5
DID0.044 ***
(0.016)
−0.023 ***
(0.007)
0.192 ***
(0.039)
−0.023 ***
(0.007)
0.066 ***
(0.009)
−0.023 ***
(0.007)
Urban forest −0.014
(0.010)
Environmental Awareness −0.003
(0.003)
Industry structure −0.039
(0.036)
Control variablesYESYESYESYESYESYES
_cons−4.310 ***
(0.992)
4.379 ***
(0.253)
−4.346 ***
(1.588)
4.494 ***
(0.270)
−2.065 ***
(0.417)
4.344 ***
(0.248)
Bootstrap test confidence interval−0.1429269 < _bs_2 < −0.1056222−0.1406999 < _bs_2 < −0.1034399−0.1255899 < _bs_2 < −0.0853589
Time FEYESYESYESYESYESYES
Urban FEYESYESYESYESYESYES
N525952595063506352605260
R20.2810.7560.1160.7630.7340.756
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Hu, C.; Chen, J.; Dong, J. Impact of City Forests on Haze Reduction—Implementation of the National Forest City Policy in China. Forests 2023, 14, 703. https://doi.org/10.3390/f14040703

AMA Style

Hu C, Chen J, Dong J. Impact of City Forests on Haze Reduction—Implementation of the National Forest City Policy in China. Forests. 2023; 14(4):703. https://doi.org/10.3390/f14040703

Chicago/Turabian Style

Hu, Chao, Jian Chen, and Jiayun Dong. 2023. "Impact of City Forests on Haze Reduction—Implementation of the National Forest City Policy in China" Forests 14, no. 4: 703. https://doi.org/10.3390/f14040703

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop