Study on the impact of forest fire prevention policy on the health of forest resources

Jingyi Zhang; Chang Chen; Yang Zhang; Chengliang Wu

Dziennik i szczegóły wydania

Applied Mathematics and Nonlinear Sciences

Informacje o czasopiśmie

Format: Czasopismo
eISSN: 2444-8656
Pierwsze wydanie: 01 Jan 2016
Częstotliwość wydawania: 2 razy w roku
Języki: Angielski

Forest fires threaten not only the forest ecosystem but also the safety of human health and their property. The Chinese Government has issued corresponding policies to strengthen the emphasis on forest fire prevention. Therefore, this paper uses panel data from 31 provinces, municipalities and autonomous regions (except Hong Kong, Macao and Taiwan) in China from 2011 to 2018 to establish a multiple regression model to study the relationship between the forest fire prevention policy and the health of forest resources, and it draws relevant conclusions. We put forward relevant suggestions accordingly to promote the improvement of the health of forest resources.

Keywords

forest fire prevention
forest health
pest control operation

Artykuł

Introduction

The Forest Law of the People’s Republic of China stipulates that the state shall strengthen the protection of forest resources and play the functions of forest to store water and soil, regulate climate, improve the environment, maintain biodiversity and provide forest products. About forest resources, people’s governments at various levels shall take measures mainly for natural restoration and combining natural restoration with artificial restoration to protect and restore forest ecosystems scientifically. Newly planted seedling woodlands and other places that should be closed off for cultivation shall be organised by the local people’s governments to close off mountains for cultivation. The healthy development of forest resources is the result of implementing the scientific conclusion put forward by General Secretary Xi Jinping that ‘clear waters and green mountains are golden mountains and silver mountains’, and it is the direction of the joint efforts of the state, governments at all levels and forestry departments.

The policies related to forest fire prevention involve many aspects, but the effect is not satisfactory, and the number of fires is high every year. Many scholars have done a lot of research on the causes of forest fires and fire prevention measures, but most of the research is only in at theoretical level and has not been applied to the actual work. As for the impact of forest fires, the occurrence of fires will have an impact on the quantity and quality of forest resources, but there is a lack of research about their degree of impact specifically.

Therefore, this article focuses on the forest fire prevention policy in forest protection policies issued by the state. Forest fire prevention is an important step and measure to protect forest resources, and it is an important part of the construction of an ecological protection system. First, we sort out the national policies, notices and research related to forest fire prevention. Second, we collect relevant data and establish an empirical analysis model. Finally, we conclude and make recommendations based on the results of the empirical analysis model.

Literature review

Forest fire prevention is of great significance in the protection of resources and the safety of people’s lives. After the founding of the People’s Republic of China, the State Council and the CPC Central Committee issued a series of documents on forest fire prevention. Especially since the Greater Khingan Range fire on May 6, 1987, the CPC Central Committee, the State Council and people’s governments at all levels paid more attention to forest fire prevention work.

The newly revised ‘Forest Fire Prevention Regulations’ implemented on January 1, 2009, clearly stated that the National Forest Fire Prevention Command Agency was responsible for organising, coordinating and guiding the nation’s forest fire prevention work. The competent forestry department of the State Council was responsible for the supervision and management of national forest fire prevention, undertaking the daily work of the national forest fire prevention command organisation and formulating national forest fire prevention plans and national emergency plans for major forest fires. Relevant departments of the State Council and local people’s governments at various levels were responsible for building forest fire prevention infrastructure, storing fire prevention materials and improving the forest fire prevention command system. After the 19th National Congress of the Communist Party of China, Chinese Government departments carried out a new round of institutional reforms, transferring the forest fire prevention and fire control command function that originally belonged to the forestry department to the emergency management department [1]. The Emergency Management Department has a Fire Rescue Bureau and a Forest Fire Bureau, which are responsible for preparing national emergency plans and national forest fire prevention plans, guiding emergency response work and promoting the construction of emergency plan systems and plan drills. The Emergency Management Department manages the emergency rescue team, the public security firefighting force after the transformation and the armed police forestry force. These teams are mainly responsible for fire material storage, fire prevention and fire rescue.

Second, it is reflected in the promulgation of relevant fire policies and laws and regulations. In 2009, the State Forestry Administration and the National Development and Reform Commission jointly issued the ‘National Forest Fire Prevention Mid- and Long-Term Development Plan’, which was the first national-level forest fire prevention plan since the founding of New China. It put forward forest fire prevention requirements from four aspects and further established and improved the three major systems of forest fire prevention, firefighting and protection, and gradually established a long-term forest fire prevention mechanism. The newly revised ‘Forest Fire Prevention Regulations’ implemented in 2009 mentioned that the competent forestry department of the State Council was responsible for preparing national forest fire prevention plans and national emergency plans, particularly for major forest fires. Relevant departments of the State Council and local people’s governments at various levels were responsible for building forest fire prevention infrastructure, storing fire prevention materials and improving the forest fire prevention command system. The third revised version of the ‘Forest Law of the People’s Republic of China’, which was implemented on July 1, 2020, proposed that governments at all levels can do a good job in forest fire prevention by establishing forest protection organisations and hiring forest guards.

Forest ecosystem health is a relatively new term, which was derived from the concept of ecosystem health, but there is no clear conclusion on its specific concept [2]. However, most of the studies believe that forest health is an artificial state [3], which requires the forest system not only to achieve self-regulation and recovery after intervention but also to meet the needs of human beings [4, 5]. This concept has attracted the attention of many countries, e.g. the US Department of Agriculture Forest Service is responsible for implementing a nationwide field forest health monitoring work [6]. There are also many research results on forest health in academia, among which are the establishment of the evaluation system and the study of spatio-temporal evolution [7, 8, 9]. Some scholars have also studied the risk factors of forest health, such as forest pest disasters, forest fires, acid rain, forest soil erosion, climate disasters and excessive logging, that will affect the healthy development of forests [10]. Therefore, to achieve and maintain the health of the forest, forest resources should be abundant, should have the ability of pest control and disaster resistance and should have the ability of sustainable development.

Globally, between 2003 and 2012, approximately 67 million hectares (1.7%) of forest land were burned every year, and pests affected >85 million hectares of the forest [11]. Through research, although forest health in many places is gradually getting better [12, 13], forest diseases and insect pests and forest fires still affect the stability and health of the forest ecosystem [14, 15, 16]. Especially, in the context of climate change, the increase in temperature had a significant positive impact on the increase in pest and fire frequencies [17, 18]. Research on carbon sinks showed that fire and insect pests are the second and fourth most important factors in the increase of forest carbon emissions [19]. Diseases and insect pests will cause significant damage to forest resources. The high occurrence frequency of diseases and insect pests and the low ability to resist diseases and insect pests led to poor forest conditions in many regions of China, which affect the sustainable development of the forest [20]. Improper management of forest fire prevention is an important factor affecting forest health, and the occurrence of fire will have a serious impact on forest resources [21, 22]. Huang [23] believed that the occurrence of forest fires would lead to the imbalance of the ecological environment, then cause a large area of diseases and insect pests and eventually lead to the premature decline of forests. In turn, insect pests will also change the combustible load and stand structure, which would affect the risk of forest fires [24]. Han [25] took the incidence of forest fire as one of the indicators to evaluate the status of forest health. The forest resources in the study area were almost free from disasters, so the forest resources in the study area were in good health status. On the other hand, some scholars believed that wildfires could cause huge social and economic losses and prescribed fires can bring some benefits [26]. In short, effective prevention and containment of unexpected forest fires are the important prerequisites for maintaining forest health and natural ecological security [27].

Data source and analysis

3.1

Data sources

In this paper, 31 provincial-level administrative regions in China are the research target (except Hong Kong Special Administrative Region, Macao Special Administrative Region and Taiwan Province), and all data from 2011 to 2018 are collected from China Forestry Statistical Yearbook and China Statistical Yearbook.

Table 1 shows the descriptive statistics results in average, standard deviation, minimum and maximum values of explanatory and explained variables. Each variable has a large range of variation, which indicates that different regions of China better reflect the development differences in the investment of forest fire prevention funds, the number of fires in different regions, the amount of stock per unit area, etc., which provides a good database for the study of the impact of forest fire prevention policies on the health of forest resources.

Table 1

Descriptive statistics of variables

Variables	Mean	Var	Min	Max
Investment in forest fire and forest public security (10,000 yuan)	15327.45	14603.8	244	113912
Area affected by fire (ha)	539.98	1371.63	0	16780
Number of fires (times)	117.81	184.46	0	1555
Pest control rate (%)	72.42	22.49	8.51	100
Afforestation area (ha)	213559.3	171155.1	710	805156
Proportion of expenditure on agriculture, forestry and water conservancy (%)	11.42	3.31	0.68	20.043
Gross regional product (100 million yuan)	22836.18	18651.76	605.83	97277.77
Volume per unit area (m³/ha)	50.47	30.65	9.63	153.72

Source: China Forestry Statistical Yearbook and China Statistical Yearbook

Among the variables, the minimum investment on forest fire and forest public security was 2.44 million yuan (Tianjin in 2017), which was also the year with the least number of fires and the smallest disaster area in Tianjin in 8 years. The maximum value was 1139.12 million yuan (Guangxi Province in 2013), much higher than the rest of the data, and the second largest investment on forest fire and forest policies was 645.44 million yuan (Yunnan Province in 2018). In 2013, the number of forest fires in Guangxi Province was 261, the fewest in 8 years and the smallest in terms of the area affected.

In all the data, the area affected by fire and the number of fires are 0 in 10 data. Except for Beijing in 2015 and Tibet in 2016, the remaining 8 data are from Shanghai. There has not been a single case of forest fire in Shanghai in 8 years. In addition to fire incidents, except in 2015, Shanghai’s investment on forest fires and forest policies was <10 million yuan, which was far below the average level.

In Hainan Province, the lowest pest control rate was only 8.51% in 2011, with a total of 13 data <30%. Except for Tibet, where the pest control rate was 19.3% in 2018, the other 12 data were all from Hainan and Guangxi Provinces. Although the maximum pest control rate was 100%, only 11 data values were >99.99%. All of them were from Beijing and Tianjin, and the mean value was 72.42%, indicating that China should improve the pest control level.

In this paper, the volume per unit area was used to represent the health of forest resources, with a minimum value of 9.63 m³/ha (Qinghai Province in 2012) and a maximum value of 153.72 m³/ha (Tibet Province in 2017). From the perspective of the overall dataset, Qinghai and Ningxia Provinces have the lowest stock volume per unit area, while Sichuan, Jilin and Tibet Autonomous Region are the three provinces with highest stock volume per unit area.

Empirical model construction and analysis

4.1

Empirical model construction

The panel data model established in this paper is as follows: $y_{it} = α_{1} x_{1 it} + α_{2} x_{2 it} + α_{3} x_{3 it} + \dots + α_{7} x_{7 it} + ε_{it}$ {y_{it}} = {\alpha _1}{x_{1it}} + {\alpha _2}{x_{2it}} + {\alpha _3}{x_{3it}} + \cdots + {\alpha _7}{x_{7it}} + {\varepsilon _{it}} where y is the forest storage volume per unit area, x₁ is the investment in forest fires and forest public security, x₂ is the area affected by forest fires, x₃ is the number of forest fires, x₄ is the pest control rate, x₅ is the afforestation area, x₆ is the proportion of agriculture, forestry and water expenditure in the general public budget support, x₇ is the gross regional product, α₁, α₂, …, α₇ are the coefficient of each independent variable, ε is the error term, i represents the region and t represents the year.

To prevent the occurrence of heteroscedasticity and to solve the problem of different data units, the variable was logarithmic, so the model was modified as follows: ${lny}_{it} = α_{1} \ln x_{1 it} + α_{2} \ln x_{2 it} + α_{3} \ln x_{3 it} + \dots + α_{7} \ln x_{7 it} + ε_{it}$ ln{y_{it}} = {\alpha _1}ln{x_{1it}} + {\alpha _2}ln{x_{2it}} + {\alpha_3}ln{x_{3it}} + \cdots + {\alpha _7}ln{x_{7it}} + {\varepsilon _{it}}

4.2

Analysis of empirical results

4.2.1

Stationary test

To avoid false regression, the panel data model should conduct a stationarity test on the data before regression. In this paper, the LLC test is chosen to test the unit root. If the null hypothesis is rejected, it indicates that the sequence is stationary; otherwise, it is non-stationary.

Table 2 shows that the five explanatory variables of forest fires and forest public security investment, fire-affected area, number of fires, pest control rate and afforestation area, and the explained variable of unit area accumulation are all stationary sequences. The proportion of agriculture, forestry and water expenditures in general public budget expenditures and the GDP are in a non-stationary sequence. To prevent the emergence of pseudo-regression, we conduct a co-integration test, and the test result is that there is a co-integration relationship between panel units, so there will be no pseudo-regression phenomenon.

Table 2

Panel unit root test

Variable	t	p	Stationarity
Volume per unit area	−94.333	0	Stationarity
Investment in forest fire and forest public security	−13.356	0	Stationarity
Area affected by fire	−11.866	0	Stationarity
Number of fires	−13.448	0	Stationarity
Pest control rate	−10.218	0.001	Stationarity
Afforestation area	−11.763	0	Stationarity
Proportion of expenditure on agriculture, forestry and water conservancy	−7.037	0.9382	Non-Stationarity
Gross regional product	2.088	1	Non-Stationarity

4.2.2

Multicollinearity test

From the theoretical point of view, there is a strong multicollinearity between the investment of forest fire and forest public security and the control variables such as the area affected by fire and the frequency of fire. Before empirical regression, multicollinearity tests were conducted between policy variables and control variables. In this paper, VIF was selected for the multicollinearity test.

As can be seen from Table 3, the maximum VIF value is 2.6, which is far <10. Therefore, we do not worry about multicollinearity.

Table 3

Multicollinearity test among independent variables

Variable	VIF	1/VIF
Investment in forest fire and forest public security	2.06	0.484321
Area affected by fire	3.76	0.265954
Number of fires	4.20	0.238306
Pest control rate	1.38	0.722797
Afforestation area	2.44	0.409427
Proportion of expenditure on agriculture, forestry and water conservancy	1.88	0.532277
Gross regional product	2.46	0.406037
Mean VIF	2.60

4.2.3

Regression results and analysis

According to Table 4, 5 of the 7 independent variables passed the test of 1%, 4 passed the test of 5% and 5 passed the test of 10%.

Table 4

Regression results

Health of forest resources	Coef.	Robust Std. Err.	t	P>\|t\|	[95% Coef. Interval]
Investment in forest fire and forest public security	0.195	0.043	4.50	0.00	0.109	0.280
Area affected by fire	−0.025	0.029	−0.87	0.385	−0.082	0.032
Number of fires	0.117	0.040	2.93	0.004	0.038	0.196
Pest control rate	0.270	0.095	2.85	0.005	0.083	0.456
Afforestation area	−0.081	0.041	−2.00	0.047	−0.161	−0.001
Proportion of expenditure on agriculture, forestry and water conservancy	0.149	0.119	1.26	0.211	−0.085	0.382
Gross regional product	−0.140	0.054	−2.57	0.011	−0.247	−0.033
Constant	2.476	0.611	4.05	0.00	1.272	3.680

Table 5

Regression results after excluding the two control variables

Health of forest resources	Coef.	Robust Std. Err.	t	P>\|t\|	[95% Coef. Interval]
Investment in forest fire and forest public security	0.133	0.039	3.43	0.001	0.057	0.209
Area affected by fire	−0.019	0.029	−0.67	0.505	−0.076	0.038
Number of fires	0.074	0.033	1.99	0.048	0.01	0.146
Pest control rate	0.106	0.086	1.23	0.219	−0.063	0.275
Afforestation area	−0.017	0.037	−0.44	0.658	−0.090	0.057
Constant	2.111	0.506	4.17	0.000	1.113	3.108

The key policy variables in this paper are forest fires and forest public security investment funds; the regression coefficient is positive and passed the 1% level test, indicating that the implementation of forest fire protection policy and forest health was significantly positively correlated. It can be seen that the investment in forest fires and forest public security is effective. An increase of 1 unit of investment in forest fires and forest public security will increase the volume of forest per unit area by 0.195 units. Investing funds in forest fires and forest public security can improve the soundness and advanced level of forest fire prevention facilities, upgrade forest fire prevention personnel and forest public security supporting facilities, conduct more training, speed up the firefighting when a fire occurs, reduce the number of forest fires and the affected area, so as to reduce the damage suffered by the forest, increase the accumulation per unit area and improve the health of the forest.

The regression coefficient of the number of forest fires is positive, which also passed the 1% level test, indicating that the number of fires had a significant positive correlation with forest health. Forest fires occurred once more, and the accumulation of forest per unit area increased by 0.117 unit. This result has deviated from daily cognition.

The regression coefficient of the pest control rate is positive, and it also passed the test of 1% level, indicating that the implementation of pest control had a significant positive correlation with forest health. If the pest control rate increases by 1 unit, the stock volume per unit area of the forest will increase by 0.27 unit. It can be seen that the implementation of pest control is one of the important factors to effectively improve the health of the forest. However, the level of pest control in China is not high, and the average pest control rate is only 73.42%. Therefore, improving the level of pest control and the rate of pest control can provide a good environment for the healthy growth of forest trees, improve the health of forest resources and realise the sustainable development of forest resources.

The regression coefficient of afforestation area is negative, which only passed the test of 10% level, indicating that there is a negative correlation between afforestation area and forest health. If the afforestation area increases by 1 unit, the stock of forest per unit area will decrease by 0.081 unit. This result shows that it is not that the more afforestation area is, the stock will increase. The increase of afforestation area will lead to the increase of forest area and forest stock, but if the increase of forest area is larger than the increase of forest stock, it will lead to the decline of stock volume per unit area.

4.2.4

Robust test

To verify the robustness of the conclusion, the empirical results of the full sample were tested from two aspects.

First, the proportion of agriculture, forestry and water expenditure and the regional difference of GDP were eliminated, and the remaining variables were regressed with the forest health degree. The estimated results are shown in Table 5. The key variables of forest fire prevention and forest public security investment have not undergone subversive and fundamental changes.

From Table 5, the key policy variables of forest fire prevention and forest public security investment funds still passed the test of 1% level, and the regression coefficient was positive, indicating that forest fire prevention and forest public security investment funds still showed a significant positive correlation with forest health relationship. With an additional investment of 1 unit, the volume of forest per unit area will increase by 0.133 unit. The number of forest fires has passed the test of 5% level and the regression coefficient is positive, indicating that the number of fires has a significant positive correlation with forest health. If a forest fire occurs once more, the accumulation of forest per unit area will increase by 0.074 unit. These two variables not only pass the test, but the signs of the regression coefficients are also the same as before.

Second, the research data in this paper are from 2011 to 2018. To make the research results robust, the 5-year data from 2014 to 2018 are regressive. The estimated results are shown in Table 6, and the estimated results are still not subversive or fundamentally changed.

Table 6

Analysis of regression results from 2014 to 2018

Health of forest resources	Coef.	Robust Std. Err.	t	P>\|t\|	[95% Coef. Interval]
Investment in forest fire and forest public security	0.168	0.572	2.93	0.004	0.546	0.281
Area affected by fire	−0.023	0.038	−0.61	0.542	−0.097	0.051
Number of fires	0.081	0.053	1.53	0.128	−0.024	0.186
Pest control rate	0.219	0.133	1.65	0.100	−0.043	0.481
Afforestation area	−0.081	0.054	−1.50	0.135	−0.188	0.025
Proportion of expenditure on agriculture, forestry and water conservancy	0.254	0.160	1.59	0.115	−0.062	0.570
Gross regional product	−0.070	0.073	−0.97	0.334	−0.214	0.073
Constant	2.160	0.836	2.58	0.011	0.509	3.813

As can be seen from Table 6, the key variables, investment in forest fire prevention and forest public security, still passed the test at the 1% level and the regression coefficient was positive, indicating that the investment in forest fire prevention and forest public security still showed a significant positive correlation with forest health. An additional 1 unit of investment would increase the forest stock per unit area by 0.168 unit. The pest control rate also passed the test of 10% level and the regression coefficient was positive, indicating that the pest control rate was also positively correlated with forest health. If the pest control rate increased by 1 unit, the forest stock per unit area would increase by 0.219 unit.

Therefore, by excluding the difference between the two regions and shortening the sample period for regression, it is concluded that the key policy variables of forest fire prevention and forest public security investment funds have always shown a significant positive correlation with forest health. In the regression result after excluding the difference between the two regions, the number of forest fires is also positively correlated with forest health. In the results of regression by shortening the number of years, the pest control rate also showed a positive correlation. It can be seen that the estimated results of the two samples are consistent with those of the full sample without subversive or fundamental changes. Therefore, it is considered that the estimated results of this paper are robust.

Conclusions and discussions

5.1

Conclusion

This paper takes forest fire policy and forest health as the starting point to analyse the impact of investment status in forest fire policy on forest health. First, China’s policies related to forest fire prevention and forest health were summarised. Second, relevant studies conducted by the scholars were summarised. Finally, data from 31 provinces (except Hong Kong, Macao and Taiwan) from 2011 to 2018 were taken as samples for the study. The conclusions of this paper are as follows:

China gives great importance to forest fire prevention, especially after the Great Khingan Mountains fire on May 6, 1987; in addition to the corresponding policies issued by the state and governments at all levels, various provinces and cities have applied a large number of funds to forest fire prevention to prevent and reduce the occurrence of forest fires, for timely rescue when a fire occurs and to adopt a series of remedial measures to reduce the adverse consequences of forest fires. Shanghai is the area with the least amount of investment over the years, and the number of fires and the area affected by fire in this region are 0. It can be seen that the fire prevention measures in Shanghai are very effective.

The number of forest fires and the area affected by them are very serious in China. In the past 8 years, the area affected has reached 16,780 hectares at most, and there are 1555 fires a year at most. This has caused very serious damage to China’s forest resources, and the follow-up restoration work is very difficult and severe. In terms of provincial data, Inner Mongolia Autonomous Region, Sichuan Province and Guangxi Province are the largest affected areas, with an annual affected area of >1,000 hectares. The Inner Mongolia Autonomous Region, in particular, recorded a record of 16,780 hectares in 2017, causing huge losses to local forest resources. In recent years, the number of fires in Hubei, Hunan and other southwestern regions has exceeded 100 times. Although the total affected area has not exceeded 1,000 hectares, fire prevention measures are still needed to reduce the number of fires and reduce the fire situation, so as to minimise the loss to the health of forest resources.

On the whole, China’s forest fire prevention policy has a certain effect on forest health, and the investment funds have promoted the increase of forest volume per unit area. Although there is a positive correlation between the number of forest fires and the accumulation per unit area, the task of improving the health of forest resources cannot be assigned to the repair work after the fire occurs. ‘Blocking is worse than sparse’, so by reducing the number of fires we can really increase the accumulation of forest resources per unit area and ensure the health of forest resources. Pest control also has a positive correlation with the health of forest resources. Therefore, it is necessary to improve the pest control rate and reduce the damage of pests to forest resources, so as to increase the forest volume per unit area and thus improve the health of forest resources. There is a negative correlation between afforestation area and forest stock per unit area. Therefore, attention should be paid to the stock of new afforestation during afforestation activities to improve the health of forest resources.

5.2

Discussions

Based on the above theoretical and empirical analysis, this paper found that the forest fire prevention policy for health promotion forest resources has a certain effect, but still need to be improved to reduce the number of forest fires and the affected area, the forest fire and forest public security money for better use of it, improve the health degree of forest resources more efficiently.

Strengthen the construction of fire prevention teams, train and build high-quality fire prevention teams. First, we must improve the staff’s awareness of fire prevention, cultivate their enthusiasm and initiative in fire prevention work and actively prepare for a series of fire prevention work. Second, it is necessary to strengthen the training of business ability, improve the theoretical knowledge and operational level through relevant training courses and practical drills, and cultivate a group of high-quality forest fire rescue teams. Finally, the fire prevention unit should pay attention to cooperation with other related units, especially strengthening the training of forest police on fire prevention. The fire protection and rescue work cannot be delegated to the firefighting team alone. It is necessary to improve the ability of forest public security in fire protection, reduce the pressure on the firefighting team and improve the efficiency of firefighting and rescue.

Use fire prevention funds rationally and efficiently. First, use funds to improve monitoring equipment and methods and thereby improve monitoring efficiency. Use modern science and technology to monitor forest fires in an all-around way, like drones, video surveillance, etc., to reduce the subsequent investment in manpower and material resources in the firefighting process. Second, the use of funds to improve rescue equipment can not only improve the perfection and advanced nature of the equipment but also ensure the personal safety of rescue workers to a certain extent. Finally, use the money to improve infrastructure, improve roads and fire barriers in forest areas, control the spread of fires, reduce the spread of forest fires and reduce the damage caused. Instant communication facilities should also be built to ensure the flow of information within the forest area, and old facilities should be maintained and updated promptly to ensure the proper functioning of the infrastructure.

Strengthen the prevention and control of pests. Pest prevention teams can be set up according to the characteristics and actual conditions of the region to carry out targeted prevention and control work according to different diseases and pests. Pest control can be combined with forest fire prevention, so as to monitor the fire situation and pests at the same time, saving manpower, material resources and financial resources. It is also necessary to prevent the impact of foreign pests on national forests, strengthen border surveillance for epidemic prevention and ensure the health of forest resources in various ways.

Ilustracje i tabele

Multicollinearity test among independent variables

Variable	VIF	1/VIF
Investment in forest fire and forest public security	2.06	0.484321
Area affected by fire	3.76	0.265954
Number of fires	4.20	0.238306
Pest control rate	1.38	0.722797
Afforestation area	2.44	0.409427
Proportion of expenditure on agriculture, forestry and water conservancy	1.88	0.532277
Gross regional product	2.46	0.406037
Mean VIF	2.60

Regression results

Health of forest resources	Coef.	Robust Std. Err.	t	P>\|t\|	[95% Coef. Interval]
Investment in forest fire and forest public security	0.195	0.043	4.50	0.00	0.109	0.280
Area affected by fire	−0.025	0.029	−0.87	0.385	−0.082	0.032
Number of fires	0.117	0.040	2.93	0.004	0.038	0.196
Pest control rate	0.270	0.095	2.85	0.005	0.083	0.456
Afforestation area	−0.081	0.041	−2.00	0.047	−0.161	−0.001
Proportion of expenditure on agriculture, forestry and water conservancy	0.149	0.119	1.26	0.211	−0.085	0.382
Gross regional product	−0.140	0.054	−2.57	0.011	−0.247	−0.033
Constant	2.476	0.611	4.05	0.00	1.272	3.680

Panel unit root test

Variable	t	p	Stationarity
Volume per unit area	−94.333	0	Stationarity
Investment in forest fire and forest public security	−13.356	0	Stationarity
Area affected by fire	−11.866	0	Stationarity
Number of fires	−13.448	0	Stationarity
Pest control rate	−10.218	0.001	Stationarity
Afforestation area	−11.763	0	Stationarity
Proportion of expenditure on agriculture, forestry and water conservancy	−7.037	0.9382	Non-Stationarity
Gross regional product	2.088	1	Non-Stationarity

Regression results after excluding the two control variables

Health of forest resources	Coef.	Robust Std. Err.	t	P>\|t\|	[95% Coef. Interval]
Investment in forest fire and forest public security	0.133	0.039	3.43	0.001	0.057	0.209
Area affected by fire	−0.019	0.029	−0.67	0.505	−0.076	0.038
Number of fires	0.074	0.033	1.99	0.048	0.01	0.146
Pest control rate	0.106	0.086	1.23	0.219	−0.063	0.275
Afforestation area	−0.017	0.037	−0.44	0.658	−0.090	0.057
Constant	2.111	0.506	4.17	0.000	1.113	3.108

Analysis of regression results from 2014 to 2018

Health of forest resources	Coef.	Robust Std. Err.	t	P>\|t\|	[95% Coef. Interval]
Investment in forest fire and forest public security	0.168	0.572	2.93	0.004	0.546	0.281
Area affected by fire	−0.023	0.038	−0.61	0.542	−0.097	0.051
Number of fires	0.081	0.053	1.53	0.128	−0.024	0.186
Pest control rate	0.219	0.133	1.65	0.100	−0.043	0.481
Afforestation area	−0.081	0.054	−1.50	0.135	−0.188	0.025
Proportion of expenditure on agriculture, forestry and water conservancy	0.254	0.160	1.59	0.115	−0.062	0.570
Gross regional product	−0.070	0.073	−0.97	0.334	−0.214	0.073
Constant	2.160	0.836	2.58	0.011	0.509	3.813

Descriptive statistics of variables

Variables	Mean	Var	Min	Max
Investment in forest fire and forest public security (10,000 yuan)	15327.45	14603.8	244	113912
Area affected by fire (ha)	539.98	1371.63	0	16780
Number of fires (times)	117.81	184.46	0	1555
Pest control rate (%)	72.42	22.49	8.51	100
Afforestation area (ha)	213559.3	171155.1	710	805156
Proportion of expenditure on agriculture, forestry and water conservancy (%)	11.42	3.31	0.68	20.043
Gross regional product (100 million yuan)	22836.18	18651.76	605.83	97277.77
Volume per unit area (m³/ha)	50.47	30.65	9.63	153.72

Referencje

[1] Zhao Y. (2019) Analysis of forest fire prevention and control ability in Hebei province from the perspective of emergency management. Hebei University of Science and Technology. ZhaoY 2019 Analysis of forest fire prevention and control ability in Hebei province from the perspective of emergency management Hebei University of Science and Technology Search in Google Scholar

[2] Yu H. (2019) Assessment of the ecosystem health of Pinus tabulaeformis forests in the source area of the Qinhe River. Shanxi University. YuH 2019 Assessment of the ecosystem health of Pinus tabulaeformis forests in the source area of the Qinhe River Shanxi University Search in Google Scholar

[3] Chen W. (2012) Forest Ecosystem Health Assessment and Early Warning Research in East Dongting Lake Region. Central South University of Forestry and Technology. ChenW 2012 Forest Ecosystem Health Assessment and Early Warning Research in East Dongting Lake Region Central South University of Forestry and Technology Search in Google Scholar

[4] Gao C, Qing-Li W, Hong-Bing D, et al. (2002) On forest ecosystem health and its Connotations. Journal of Forestry Research, 13(2):147–150. GaoC Qing-LiW Hong-BingD 2002 On forest ecosystem health and its Connotations Journal of Forestry Research 13 2 147 150 10.1007/BF02857242 Search in Google Scholar

[5] Shi Y. (2017) Evaluation of Mountain Forest Health Resources in Changping District, Beijing. Greening and Life, (06):48–49. ShiY 2017 Evaluation of Mountain Forest Health Resources in Changping District, Beijing Greening and Life 06 48 49 Search in Google Scholar

[6] Woodall C W, Amacher MC, Bechtold W A, et al. (2011) Status and future of the forest health indicators program of the USA. Environmental Monitoring and Assessment, 177(1–4):419–436. WoodallCW AmacherMC BechtoldWA 2011 Status and future of the forest health indicators program of the USA Environmental Monitoring and Assessment 177 1–4 419 436 10.1007/s10661-010-1644-820711860 Search in Google Scholar

[7] Lu S, Zhou Y, Sun H, et al. (2021) Examining the influencing factors of forest health, its implications on rural revitalization: A case study of five forest farms in Beijing. Land Use Policy, 102: 105265. LuS ZhouY SunH 2021 Examining the influencing factors of forest health, its implications on rural revitalization: A case study of five forest farms in Beijing Land Use Policy 102 105265 10.1016/j.landusepol.2020.105265 Search in Google Scholar

[8] Kumar M, Kalra N, Singh H, et al. (2021) Indicator-based vulnerability assessment of forest ecosystem in the Indian Western Himalayas: An analytical hierarchy process integrated approach. Ecological Indicators, 125. KumarM KalraN SinghH 2021 Indicator-based vulnerability assessment of forest ecosystem in the Indian Western Himalayas: An analytical hierarchy process integrated approach Ecological Indicators 125 10.1016/j.ecolind.2021.107568 Search in Google Scholar

[9] Trumbore S, Brando P, Hartmann H. (2015) Forest health and global change [J]. Science, 349(6250): 814–818. TrumboreS BrandoP HartmannH 2015 Forest health and global change [J] Science 349 6250 814 818 10.1126/science.aac675926293952 Search in Google Scholar

[10] Gui Z, Zhang G. (2012) Research progress on forest health risks in China. Chinese Agricultural Science Bulletin, 28(13): 1–5. GuiZ ZhangG 2012 Research progress on forest health risks in China Chinese Agricultural Science Bulletin 28 13 1 5 Search in Google Scholar

[11] Pieter L, Erik L, Shiroma S, et al. (2015) Global forest area disturbance from fire, insect pests, diseases and severe weather events. Forest Ecology and Management, 352:78–88. PieterL ErikL ShiromaS 2015 Global forest area disturbance from fire, insect pests, diseases and severe weather events Forest Ecology and Management 352 78 88 10.1016/j.foreco.2015.06.010 Search in Google Scholar

[12] Dong J, Cui Y. (2020) Research on Forest Health Evaluation Based on Analytic Hierarchy Process. Forestry Inventory and Planning, 45(01): 15–18. DongJ CuiY 2020 Research on Forest Health Evaluation Based on Analytic Hierarchy Process Forestry Inventory and Planning 45 01 15 18 Search in Google Scholar

[13] Li Y, Wang S. (2020) Analysis and evaluation of forest quality changes in northwestern Yunnan in the past 10 years. Forestry Inventory and Planning, 45(02): 97–102. LiY WangS 2020 Analysis and evaluation of forest quality changes in northwestern Yunnan in the past 10 years Forestry Inventory and Planning 45 02 97 102 Search in Google Scholar

[14] Sun S, Gu Z, Li X, et al. (2013) On the prevention and control of forest diseases and insect pests in nature reserves. Jilin Forestry Science and Technology, 42(01): 53–54+57. SunS GuZ LiX 2013 On the prevention and control of forest diseases and insect pests in nature reserves Jilin Forestry Science and Technology 42 01 53 54+57 Search in Google Scholar

[15] Zhao H. (2019) Discussion on strengthening the prevention and control of forest diseases and insect pests and promoting the healthy development of forestry [J]. Rural Practical Technology, (10): 83. ZhaoH 2019 Discussion on strengthening the prevention and control of forest diseases and insect pests and promoting the healthy development of forestry [J] Rural Practical Technology 10 83 Search in Google Scholar

[16] Tkacz B, Moody B, Castillo J V, et al. (2008) Forest health conditions in North America. Environmental Pollution, 155(3):409–425. TkaczB MoodyB CastilloJ V 2008 Forest health conditions in North America Environmental Pollution 155 3 409 425 10.1016/j.envpol.2008.03.00318479794 Search in Google Scholar

[17] Nicholas AF, Scott RA, Matthew P, et al. (2014) Climate, trees, pests, and weeds: Change, uncertainty, and biotic stressors in eastern US national park forests. FOREST ECOL MANAG, 327:31–39. NicholasAF ScottRA MatthewP 2014 Climate, trees, pests, and weeds: Change, uncertainty, and biotic stressors in eastern US national park forests FOREST ECOL MANAG 327 31 39 10.1016/j.foreco.2014.04.033 Search in Google Scholar

[18] Rubin A, Saenz R, Lindig C, et al. (2015) Bark beetle pests in an altitudinal gradient of a Mexican managed forest. FOREST ECOL MANAG, 343:73–79. RubinA SaenzR LindigC 2015 Bark beetle pests in an altitudinal gradient of a Mexican managed forest FOREST ECOL MANAG 343 73 79 10.1016/j.foreco.2015.01.028 Search in Google Scholar

[19] Williams CA, Gu H, Maclean R, et al. (2016) Disturbance and the carbon balance of US forests: A quantitative review of impacts from harvests, fires, insects, and droughts. Global and Planetary Change, 143:66–80. WilliamsCA GuH MacleanR 2016 Disturbance and the carbon balance of US forests: A quantitative review of impacts from harvests, fires, insects, and droughts Global and Planetary Change 143 66 80 10.1016/j.gloplacha.2016.06.002 Search in Google Scholar

[20] Yue Z. (2020) Forest health assessment in Dalishugou forest area of Saihanba Forest Farm. Green Science and Technology, (01): 48–49. YueZ 2020 Forest health assessment in Dalishugou forest area of Saihanba Forest Farm Green Science and Technology 01 48 49 Search in Google Scholar

[21] Wei Y, Tang J, Zou R, et al. (2020) Exploring the current situation and development direction of forestry resource management and protection. Seed Science and Technology, 38(04): 112–113. WeiY TangJ ZouR 2020 Exploring the current situation and development direction of forestry resource management and protection Seed Science and Technology 38 04 112 113 Search in Google Scholar

[22] Huang T. (2019) Research on the Health Evaluation of Main Forest Types in the Baoku River Basin, Datong County, Qinghai Province. Beijing Forestry University. HuangT 2019 Research on the Health Evaluation of Main Forest Types in the Baoku River Basin, Datong County, Qinghai Province Beijing Forestry University Search in Google Scholar

[23] Huang J. (2018) Forest health status and control measures in nature reserves. Green Science and Technology, (13): 198–199. HuangJ 2018 Forest health status and control measures in nature reserves Green Science and Technology 13 198 199 Search in Google Scholar

[24] Xie H, Fawcett J E, Wang G G. (2020) Fuel dynamics and its implication to fire behavior in loblolly pine-dominated stands after southern pine beetle outbreak. Forest Ecology and Management, 466(118130). XieH FawcettJE WangGG 2020 Fuel dynamics and its implication to fire behavior in loblolly pine-dominated stands after southern pine beetle outbreak Forest Ecology and Management 466 118130 10.1016/j.foreco.2020.118130 Search in Google Scholar

[25] Han J. (2019) Distribution and quality evaluation of forest resources in Licheng District. Shandong Agricultural University. HanJ 2019 Distribution and quality evaluation of forest resources in Licheng District Shandong Agricultural University Search in Google Scholar

[26] Pereira P, Bogunovic I, Zhao W, et al. (2021) Short term effect of wildfires and prescribed fires on ecosystem services. Current Opinion in Environmental Science and Health, 202100266. PereiraP BogunovicI ZhaoW 2021 Short term effect of wildfires and prescribed fires on ecosystem services Current Opinion in Environmental Science and Health 202100266. 10.1016/j.coesh.2021.100266 Search in Google Scholar

[27] Sannigrahi S, Pilla F, Basu B, et al. (2020) Examining the effects of forest fire on terrestrial carbon emission and ecosystem production in India using remote sensing approaches. Science of The Total Environment, 725(38331). SannigrahiS PillaF BasuB 2020 Examining the effects of forest fire on terrestrial carbon emission and ecosystem production in India using remote sensing approaches Science of The Total Environment 725 38331 10.1016/j.scitotenv.2020.13833132302833 Search in Google Scholar