A method for establishing a forest carbon sequestration model based on the dynamic change process of forest stands

Abstract

The present invention relates to a method for establishing a forest carbon fixation model based on a dynamic change process of a forest stand, and belongs to the technical field of forestry. The method comprises the following steps: sample plot investigation and data collection; calculating carbon changes in a dynamic change process of a forest stand; obtaining relevant driving factors, variable screening, and fitting the optimal carbon model of each process; and constructing a structural equation model of net carbon fixation. The method of the present invention uses a structural equation model to establish a forest carbon dynamic change model, and selects the optimal model through standardized root mean square residual, goodness of fit index, comparative fit index, and Akaike information criterion, which clarifies the different effects of biological and non-biological factors on forest carbon fixation under dynamic changes of forest stands, and reveals that in the restoration and management of natural secondary forests, targeted afforestation measures should be taken for different management purposes, while focusing on the optimization and management of forest stand structure to maximize forest carbon fixation, and has wide applicability and promotion value for analyzing carbon fixation in natural secondary forests in Northeast China.

Description

Technical Field

[0001] The present invention relates to the technical field of forestry, and in particular to a method for establishing a forest carbon sequestration model based on a dynamic change process of a forest stand. Background Art

[0002] Forests cover about 30% of the world's land surface and are an important part of the global carbon cycle. Increasing the rate and storage capacity of forest carbon sequestration has been recognized as an important natural solution to help mitigate climate change, and this carbon strategy has been a major focus of ecologists. However, carbon storage varies greatly across biomes, and the persistence of carbon sinks depends on how forests respond to changing biomes and environmental conditions. Natural secondary forests have significant carbon capture capacity and play a vital role in maintaining biodiversity and providing ecosystem services, but little is known about what factors drive forest carbon dynamics and net carbon sequestration. At the stand level, net carbon sequestration is mainly affected by three dynamic change processes (entry, growth, and dieback). Because the main biotic and abiotic drivers of each growth process may be different, they should be analyzed separately to more accurately determine the net carbon sequestration of forests. Understanding the dynamic changes in forest carbon and its driving factors will help us take targeted silviculture measures in the restoration and management of natural secondary forests to maximize forest carbon sequestration. Summary of the invention

[0003] The present invention aims to solve the technical problems in the prior art and provide a method for establishing a forest carbon sequestration model based on the dynamic change process of the forest stand.

[0004] In order to solve the above technical problems, the technical solutions of the present invention are as follows:

[0005] A method for establishing a forest carbon sequestration model based on a dynamic change process of a forest stand comprises the following steps:

[0006] Step i: Sample site survey and data collection;

[0007] Step ii: Calculate carbon changes during forest stand dynamics;

[0008] Step iii: Obtain relevant driving factors, screen variables, and fit the optimal carbon model for each process;

[0009] Step iv: Construct the structural equation model of net carbon sequestration.

[0010] In the above technical solution, step i specifically includes:

[0011] (1) Determine the research site;

[0012] (2) Survey of each tree;

[0013] All woody plants with a breast diameter greater than or equal to a certain value in the sample plots were marked and located, the plant species of each plant was recorded, its breast diameter, tree height, crown width and height under branches were measured, and the woody plants in the sample plots were re-surveyed at regular intervals; topographic data were measured in each plot;

[0014] (3) Collection of environmental factors;

[0015] Soil was collected from the sample plot for elemental analysis to obtain soil variables: total nitrogen, total phosphorus, total potassium, available nitrogen, available phosphorus, available potassium, organic matter content and pH value; the altitude, slope, aspect and convexity data of the plot were measured.

[0016] In the above technical solution, in step ii, the dynamic change process of the forest stand includes: entry, growth, and death.

[0017] In the above technical solution, in step ii, the carbon changes in the dynamic change process of the forest stand are calculated, and the calculation contents include: incoming charcoal, retained charcoal and dead and damaged charcoal.

[0018] In the above technical solution, the relevant driving factors obtained in step iii include: biological factors and abiotic factors.

[0019] In the above technical solution, the variable screening in step iii is specifically as follows:

[0020] Variables with variance inflation factor VIF>5 were deleted in the entire model;

[0021] The Pearson correlation coefficient was used to test the binary relationship between carbon changes in each process and a single predictor variable, as well as the relationship between all explanatory variables, and the significantly correlated explanatory variables were selected.

[0022] In the above technical solution, the optimal carbon model for fitting each process in step iii is specifically:

[0023] The carbon changes in each process were logarithmically transformed, and the explanatory variables were standardized using the Z-score method to improve the interpretability of the regression coefficients.

[0024] Structural equation modeling was used to examine the direct and indirect effects of biotic and abiotic factors on biomass changes;

[0025] According to the known relationship between variables, the significantly correlated explanatory variables were respectively combined with incoming carbon, retained carbon, and damaged carbon to construct a structural equation model.

[0026] The chi-square test was used to evaluate and adjust the model fit. Models with insignificant chi-square test statistics (P>0.05), SRMR values ​​less than 0.08, and CFI values ​​greater than 0.95 were considered to have passed the test. The parameter indicators of the models that passed the test were compared, and the model with the highest GFI and CFI values, and the lowest SRMR and AIC values ​​was selected as the optimal model.

[0027] The structural equation model was constructed and evaluated using the lavaan package of R software.

[0028] In the above technical solution, the construction of the net carbon fixation structural equation model in step iv is specifically as follows:

[0029] The calculation formula for the net carbon fixation of the plot is:

[0030] C N =C R +C G -C M

[0031] In the formula, C N is the net carbon fixation, C R For imported charcoal, C G To retain carbon and C M For dead wood charcoal;

[0032] Relative contribution of incoming carbon, retained carbon and dead carbon to the change of net carbon sequestration R , R G , R M They are:

[0033] R R =[var(C R )+cov(C R ,C G )-cov(C R ,C M )] / var(C N )

[0034] R G =[var(C G )+cov(C R ,C G )-cov(C G ,C M )] / var(C N )

[0035] R M =[var(C M )-cov(C R ,C M )-cov(C G ,C M)] / var(C N )

[0036] Where var() is the variance of the three carbon changes, and cov() is the covariance of the two carbon changes.

[0037] The present invention has the following beneficial effects:

[0038] The method of the present invention for establishing a forest carbon sequestration model based on the dynamic change process of the forest stand utilizes the structural equation model to establish the forest carbon dynamic change model, and selects the optimal model through the standardized root mean square residual (SRMR), goodness of fit index (GFI), comparative fit index (CFI), and Akaike information criterion (AIC), clarifies the different effects of biotic and abiotic factors on forest carbon sequestration under the dynamic change of the forest stand, and suggests that in the restoration and management of natural secondary forests, attention should be paid to the optimization and management of the forest stand structure, and targeted silviculture measures should be taken for different management purposes to maximize forest carbon sequestration, which has wide applicability and promotion value for analyzing carbon sequestration in natural secondary forests in Northeast China. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0040] Figure 1 It is a schematic diagram of the implementation process of the method for establishing a forest carbon sequestration model based on the dynamic change process of the forest stand of the present invention.

[0041] Figure 2 Schematic diagram of the Pearson correlation coefficient between variables in Example 1 of the present invention.

[0042] Blue indicates positive correlation, orange indicates negative correlation. The square shaded area indicates the correlation coefficient (r), and the blank area indicates no significant correlation (P>0.05).

[0043] Figure 3 is the structural equation model of the relative contribution of carbon in each process to the net carbon fixation in Example 1 of the present invention; wherein:

[0044] (a) is the structural equation model of the effects of biotic and abiotic factors on the changes in incoming wood carbon in Example 1 of the present invention;

[0045] (b) is the structural equation model of the effects of biotic and abiotic factors on the changes in retained wood carbon in Example 1 of the present invention;

[0046] (c) is the structural equation model of the effects of biotic and abiotic factors on the carbon changes of dead wood in Example 1 of the present invention. DETAILED DESCRIPTION

[0047] The following is combined with Figure 1-3 The present invention is described in detail.

[0048] Example 1

[0049] The basic data are the forest resource survey data of two phases (2015 and 2020) of large sample plots in Northeast China. The biological and abiotic factors are: 2 species diversity factors ( 1 D. 2 D), 5 stand structure factors (coefficient of variation of breast diameter, coefficient of variation of tree height, Gini coefficient of breast diameter, Gini coefficient of tree height and structural diversity index), 1 stand density factor (basal area density at breast height), 4 terrain factors (altitude, slope, aspect and concavity), 8 soil factors (total nitrogen, total phosphorus, total potassium, etc.), a method for establishing a forest carbon sequestration model based on the dynamic change process of the stand, comprising the following steps:

[0050] (1) Sample site survey and data collection:

[0051] (11) Determine the study site. We established several hectares of sample plots in natural secondary forests in Northeast China. The study area spans about 1500 km from east to west (39.71°N-53.37°N), about 1000 km from north to south (119.80°E-134.02°E), and the altitude span is about 1100 meters (79-1255 meters). Our plots span a large area of ​​warm temperate, mid-temperate and cold temperate zones and have a temperate continental climate. The average annual temperature ranges from -5.6℃ to 9.8℃, and the annual precipitation ranges from 363.8mm to 1073.7mm. The forests are natural secondary forests, including pure forests and mixed forests. The main tree species include Pinus koraiensis, Acer mandshuricum, Ulmus laciniata, Carpinus cordata, Acerpictum subsp. mono, etc. The main shrubs include Syringa reticulata subsp. amurensis, Corylus mandshurica, etc. Before the 1990s, the forest stand structure was severely damaged due to over-logging and unsustainable use.

[0052] (12) Tree-by-tree survey. We marked and located all woody plants with a DBH ≥ 1 cm in the sample plots, recorded the species of each plant, and measured its DBH, tree height, crown width (east-west crown width, north-south crown width), and height under branches. We re-surveyed the woody plants in the sample plots every 5 years.

[0053] (13) Collection of environmental factors. The upper layer (0-10 cm), middle layer (10-20 cm) and lower layer (>20 cm) of the soil were collected for elemental analysis. The average values ​​of the three layers were taken to obtain soil variables: total nitrogen, total phosphorus, total potassium, available nitrogen, available phosphorus, available potassium, organic matter content and pH value. The altitude, slope, aspect and concavity data of the plot were measured.

[0054] (2) Calculate the carbon changes in the three dynamic changes of the forest stand (entry, growth, and death):

[0055] Calculated from the perspective of forest stand dynamics, it includes incoming wood charcoal, retained wood charcoal (trees in the growth process are generally defined as retained wood) and dead and damaged wood charcoal.

[0056] This study only considered tree species, and used a species-specific allometric equation to calculate the total biomass (TB) of a single tree, which was applicable to about 80% of the studied tree species. The other tree species were calculated using the equation of the same genus (see Table 1). The composition of total biomass includes leaf biomass, branch biomass, stem biomass and root biomass.

[0057] Table 1 Allometric growth equations for tree species

[0058]

[0059] We converted the total biomass of individual trees into total carbon storage (TC) by multiplying the average biomass carbon content (0.488 for broadleaved species and 0.508 for coniferous trees). -1 ) is calculated as follows:

[0060]

[0061] Among them, TC0 is the total carbon storage of a single tree at the initial measurement, TC1 is the total carbon storage of a single tree at the first re-measurement, and A is the number of years between two adjacent forest surveys.

[0062] Annual carbon fixation at the plot level (C, Mg·ha -1 ·year -1 ) is calculated as follows:

[0063]

[0064] Among them, ΔTC i is the annual carbon sequestration of the ith tree in the plot, and S is the area of ​​the plot.

[0065] The entry DBH is 5 cm. The entry trees refer to the living trees with a DBH < 5 cm in the initial measurement and a DBH ≥ 5 cm in the latest re-measurement. The retained trees refer to the trees with a DBH ≥ 5 cm in the initial measurement and still alive in the re-measurement. The dead and damaged trees refer to the trees that died naturally between the two surveys. According to the above calculation steps, the entry trees carbon (C R ), retaining charcoal (C G ) and dead wood carbon (C M ).

[0066] (3) Obtain relevant driving factors:

[0067] Includes biotic factors (species diversity, tree size differentiation, stand density) and abiotic factors (topography, soil).

[0068] Hill (1937) integrated species abundance and richness into a parameterized diversity measure, the Hill number calculation formula is:

[0069]

[0070] In the formula, p i It represents the proportion of the number of the i-th species in the community, S represents the number of species or richness of the entire community, and the parameter q determines the sensitivity of the measurement to relative abundance, and q≠1 is defined. When the q limit approaches 1, it is in the form of Shannon diversity index, and when q=2, the Hill number is equivalent to the inverse form of Gini-Simpson diversity. 1 D Treat all tree species equally. 2 D is more sensitive to high abundance and dominant tree species.

[0071] In order to quantify structural diversity, the Gini coefficient, coefficient of variation and structural diversity index of DBH and tree height are introduced. Taking 20m×20m sample plots as the basic unit, the Gini coefficient indicates the uniformity of distribution. The larger its value, the more uneven the growth of DBH or tree height. The coefficient of variation measures the degree of dispersion of DBH or tree height. The larger its value, the greater the dispersion. The structural diversity index is the average of the Shannon diversity index of DBH and tree height, which indicates the vertical and horizontal structural changes of the forest stand. There are many dependent variables, so the principal component analysis method is used for dimensionality reduction. The first axis principal component with the highest explanatory value is retained, and multiple variables are converted into comprehensive indicators (tree size differentiation) under the premise of very little loss. Stand density (BA) is also an important indicator affecting the structure of the forest stand, which is expressed by the sum of the cross-sectional area at breast height of each 20m×20m sample plot. The statistical results of each biological factor are shown in Table 2, and the specific calculation formula is as follows:

[0072]

[0073] In the formula, CVD is the coefficient of variation of diameter at breast height; DBH Std is the standard deviation of DBH; is the average breast diameter; CV H is the coefficient of variation of tree height; H Std is the standard deviation of tree height; is the average tree height; G D is the Gini coefficient of breast diameter; N d is the breast diameter grade number; ba k is the chest height cross-sectional area of ​​the kth individual arranged from small to large; N is the total number of individuals in the sample plot; G H is the Gini coefficient of tree height; N h is the tree height level; h m is the tree height of the mth individual arranged from small to large; H′ is the structural diversity index; H′ dbh is the Shannon diversity index of breast diameter; H′ h is the tree height Shannon diversity index.

[0074] Table 2 Statistics of biological factors

[0075]

[0076]

[0077] In order to simplify the soil factor information, the principal component analysis method is also used, and the first principal component axis is used to express the soil condition. The elevation values ​​of the four corners of the small sample plot are measured and averaged. The average angle between the plane formed by the vertices of any three of the four corners of the sample plot and the horizontal plane is the slope. The slope direction is obtained by the average angle between the plane formed by the vertex angles of the sample plot and the vertical plane. The concavity is obtained by subtracting the elevation of the sample plot from the elevation of the center of the sample plot.

[0078] (4) Variable screening:

[0079] Variable screening. In order to reduce the number of predictive variables and avoid the strong influence of multicollinearity in the model, we deleted the variables with variance inflation factor (VIF)>5 in the whole model (see Table 3). To reduce the explanatory variable data in the equation, the Pearson correlation coefficient was used to test the binary relationship between each carbon fixation amount and a single predictive variable, as well as the relationship between all explanatory variables, and the significantly correlated explanatory variables were selected (see Figure 2 ).

[0080] Table 3 Variable screening results

[0081]

[0082] Note: STR PC1 is the first axis principal component of size differentiation; ELE is altitude; ASP is aspect; SLO is slope; CON is concavity; SOILPC1 It is the first axis principal component of soil variables.

[0083] (5) Fitting the optimal carbon model for each process:

[0084] The carbon changes of inbounded wood, retained wood and dead wood were logarithmically transformed, and the explanatory variables were standardized using the Z-score method to improve the interpretability of the regression coefficients. The structural equation model was used to test the direct and indirect effects of biotic factors (species diversity, tree size differentiation, and stand density) and abiotic factors (topography and soil factors) on the three carbon changes. According to the known relationships between the variables, the significantly correlated explanatory variables were used to construct the structural equation model with inbounded wood carbon, retained wood carbon, and dead wood carbon. The chi-square test was used to evaluate and adjust the model fit. Models with insignificant chi-square test statistics (P>0.05), SRMR values ​​less than 0.08, and CFI values ​​greater than 0.95 were considered to have passed the test. The parameter indicators of the models that passed the test were compared, and the model with the highest GFI and CFI values ​​and the lowest SRMR and AIC values ​​was selected as the optimal model. The construction and evaluation of the structural equation model were calculated using the lavaan package of R software.

[0085] (6) Constructing the structural equation model of net carbon sequestration:

[0086] The calculation formula of net carbon fixation in the plot is as follows:

[0087] C N =C R +C G -C M (9)

[0088] The relative contributions of incoming wood carbon, retained wood carbon and dead wood carbon to the change in net carbon sequestration are calculated using the following three formulas:

[0089] R R =[var(C R )+cov(C R ,C G )-cov(C R ,C M )] / var(C N ) (10)

[0090] R G =[var(C G )+cov(C R ,C G )-cov(C G ,C M )] / var(C N ) (11)

[0091] R M =[var(CM )-cov(C R ,C M )-cov(C G ,C M )] / var(C N ) (12)

[0092] In the formula, R R , R G , R M are the relative contributions of incoming carbon, retained carbon and damaged carbon to the change of net carbon fixation; var() is the variance of the three carbon changes, and cov() is the covariance of the two carbon changes.

[0093] Using equations (10)-(12), a structural equation model of forest net carbon sequestration was constructed (see Figure 3 ).

[0094] Figure 2 The Pearson correlation coefficients between the variables are given. As can be seen from the figure, forests with large tree size differentiation (i.e., high coefficient of variation and Gini coefficient) promote the growth of inbound trees, but inhibit the growth of retained trees. The negative impact of forest structure on the carbon fixation change of retained trees may be related to asymmetric competition for light. For retained trees, the larger the coefficient of variation (Gini coefficient) of DBH and tree height, the higher the degree of inequality in tree size. In addition, the degree of size differentiation in the figure is significantly positively correlated with the density of large trees (P<0.05), indicating that plots with complex forest structure may have more large and small trees. Since strong light can cause leaves to reach light saturation and cannot photosynthesize, there is a threshold for the light interception ability of large trees in the top canopy, while small trees in the understory cannot reach light saturation, so the light utilization efficiency of large trees is lower than that of small trees. This conclusion has been confirmed in the study of Cordonnier et al. On the other hand, compared with small trees, large trees need to use more of the absorbed carbon to maintain supporting organs (trunks and branches) rather than growth, so the growth of plots with more large trees may be lower than that of plots with more small trees. The positive correlation between tree size differentiation and carbon changes in bordering trees may be due to the fact that the complex forest structure enables more site resources to be utilized and increases the light interception efficiency of understory vegetation, thereby promoting the growth of bordering trees.

[0095] Figure 3 (a) The structural equation model of the effects of biotic and abiotic factors on the carbon changes of the bordering wood is given. As can be seen from the figure, species diversity has no significant effect on the carbon of the bordering wood. It may be that the sample sites with high species diversity are more likely to have inferior trees. The slower growth characteristics of inferior trees offset the advantages of the rapid growth of dominant trees, making the effect of species diversity on carbon changes insignificant. Concavity and convexity are significantly positively correlated with the carbon of the bordering wood. The reason for this is that the light conditions are better at the ridges with raised terrain, which is more conducive to the growth of bordering wood.

[0096] Figure 3 (b) The structural equation model of the effects of biotic and abiotic factors on the changes in retained wood carbon is given. As can be seen from the figure, stand density is the most important driving factor for the growth of retained wood carbon. The reason for this may be that high stand density allows for greater canopy coverage, which can absorb more light energy. At the same time, high-density stands have a niche complementarity effect, and trees occupy more space to obtain more resources. The significant effect of species diversity on retained wood carbon can also be explained by the niche complementarity effect.

[0097] Figure 3 (c) A structural equation model of the effects of biotic and abiotic factors on the changes in wood carbon loss is given. As can be seen from the figure, altitude is the only factor that inhibits wood carbon loss. Altitude is a common and complex abiotic factor, which is subject to the combined constraints of climate and soil factors. High altitudes are usually accompanied by increased precipitation, and the higher the altitude, the weaker the human disturbance in the forest, which may be more conducive to tree growth and lower tree loss. In addition, the model explanation of wood carbon loss is low (R 2 =0.02), the key factors affecting tree loss still need further exploration and analysis.

[0098] entire Figure 3 The dynamic changes of forest carbon and their relative contributions to net carbon sequestration are integrated. As can be seen from the figure, abiotic factors can directly affect forest carbon changes, and there is also a large part of indirect effects. The impact of abiotic factors on the carbon sequestration of standing trees depends largely on the stand structure and species diversity. Topographic factors have a greater impact on community carbon sequestration and species diversity. With increasing altitude, the growth of trees shows a trend of shifting from water restriction to heat restriction. The combination of environmental characteristics such as soil, climate and biology in the region will produce heterogeneity according to different altitude gradients, resulting in changes in species diversity and forest carbon sequestration. Slope can affect forest distribution by affecting many environmental factors, such as solar radiation, soil texture, soil erosion, soil moisture and soil nutrients. Generally, gentler slopes can provide more soil moisture and nutrients. Concavity is also an important environmental factor affecting plant community structure, reflecting the intensity of light absorption and wind interference. Soil chemical properties have no significant effect on stand structure. The needs of different species for soil resources may be complementary, and tree species in different growth groups have different habitat preferences.

[0099] Among the relative contributions of the three types of carbon to net carbon sequestration, retained wood carbon accounts for the largest proportion (46%). It is worth noting that the relative contribution of dead wood carbon (26%) is very close to that of incoming wood (28%). At the same time, considering that the dead wood carbon model has a low explanatory capacity, it is suggested that while we pay attention to the carbon sequestration capacity of living trees, the ability of dead wood to release carbon should not be underestimated, and the biological and abiotic factors that affect forest death also need our focus.

[0100] In summary, compared with species diversity, stand structure and stand density have a stronger effect on the dynamic changes of forest carbon. By changing the management measures of the stand structure, the efficiency of forest carbon sequestration will be effectively improved. However, the way to change the stand structure is different for different natural secondary forest succession stages and management purposes. If the main management purpose of the forest is the growth and renewal of small trees, it is recommended to appropriately reduce the stand density and increase the degree of differentiation of tree size; if the main purpose is to increase the forest's carbon sequestration capacity, it is recommended to appropriately increase the stand density while avoiding excessive differentiation of tree size. In addition, not only biological factors such as species and structural composition in the stand should be considered, but also the direct effects of terrain and soil factors, as well as their indirect effects through affecting species composition and distribution cannot be ignored.

[0101] The method of the present invention for establishing a forest carbon sequestration model based on the dynamic change process of the forest stand utilizes the structural equation model to establish the forest carbon dynamic change model, and selects the optimal model through the standardized root mean square residual (SRMR), goodness of fit index (GFI), comparative fit index (CFI), and Akaike information criterion (AIC), clarifies the different effects of biotic and abiotic factors on forest carbon sequestration under the dynamic change of the forest stand, and suggests that in the restoration and management of natural secondary forests, attention should be paid to the optimization and management of the forest stand structure, and targeted silviculture measures should be taken for different management purposes to maximize forest carbon sequestration, which has wide applicability and promotion value for analyzing carbon sequestration in natural secondary forests in Northeast China.

[0102] Obviously, the above embodiments are merely examples for the purpose of clear explanation, and are not intended to limit the implementation methods. For those skilled in the art, other different forms of changes or modifications can be made based on the above description. It is not necessary and impossible to list all the implementation methods here. The obvious changes or modifications derived from these are still within the scope of protection of the invention.

Claims

1. A method for establishing a forest carbon sequestration model based on the dynamic change process of forest stands, characterized in that: The following steps are involved: Step i: Sample site survey and data collection; Step ii: Calculate carbon changes during forest stand dynamics; Step iii: Obtain relevant driving factors, screen variables, and fit the optimal carbon model for each process; Step iv: construct the structural equation model of net carbon sequestration; In step ii, the dynamic change process of the forest stand includes: entry, growth, and death; In step ii, the carbon changes in the dynamic change process of the forest stand are calculated, and the calculation contents include: incoming carbon, retained carbon and dead and damaged carbon; The relevant driving factors obtained in step iii include: biotic factors and abiotic factors; The variable screening in step iii is specifically as follows: Variables with variance inflation factor VIF>5 were deleted in the entire model; Pearson correlation coefficient was used to test the binary relationship between carbon changes in each process and a single predictor variable, as well as the relationship between all explanatory variables, and the significantly correlated explanatory variables were selected; The optimal carbon model for each process in step iii is specifically: The carbon changes in each process were logarithmically transformed, and the explanatory variables were standardized using the Z-score method to improve the interpretability of the regression coefficients; The structural equation model was used to examine the direct and indirect effects of biotic and abiotic factors on carbon changes in each process; the carbon changes in each process included: carbon changes in incoming wood, carbon changes in retained wood, and carbon changes in dead and damaged wood; According to the known relationship between variables, the significantly correlated explanatory variables were respectively combined with incoming carbon, retained carbon, and damaged carbon to construct a structural equation model. The chi-square test was used to evaluate and adjust the model fit. Models with insignificant chi-square test statistics, P>0.05, SRMR values ​​less than 0.08, and CFI values ​​greater than 0.95 were considered to have passed the test. The parameter indicators of the models that passed the test were compared, and the model with the highest GFI and CFI values ​​and the lowest SRMR and AIC values ​​was selected as the optimal model. The structural equation model was constructed and evaluated using the lavaan package of R software.

2. The method for establishing a forest carbon sequestration model based on the dynamic change process of forest stands according to claim 1, characterized in that: Step i specifically includes: (1) Determine the research site; (2) Survey of each tree; All woody plants with a breast diameter greater than or equal to a certain value in the sample plots were marked and located, the plant species of each plant was recorded, its breast diameter, tree height, crown width and height under branches were measured, and the woody plants in the sample plots were re-surveyed at regular intervals; topographic data were measured in each plot; (3) Collection of environmental factors; Soil was collected from the sample plot for elemental analysis to obtain soil variables: total nitrogen, total phosphorus, total potassium, available nitrogen, available phosphorus, available potassium, organic matter content and pH value; the altitude, slope, aspect and convexity data of the plot were measured.

3. The method for establishing a forest carbon sequestration model based on the dynamic change process of forest stands according to claim 1, characterized in that: The specific construction of the net carbon sequestration structural equation model in step iv is: The calculation formula for the net carbon fixation of the plot is: C N =C R +C G -C M In the formula, C N is the net carbon fixation, C R For imported charcoal, C G To retain carbon and C M For dead wood charcoal; Relative contribution of incoming carbon, retained carbon and dead carbon to the change of net carbon sequestration R , R G , R M They are: R R =[var(C R )+cov(C R ,C G )-cov(C R ,C M )] / var(C N ) R G =[var(C G )+cov(C R ,C G )-cov(C G ,C M )] / var(C N ) R M =[var(C M )-cov(C R ,C M )-cov(C G ,C M )] / var(C N ) In the formula, var() is the variance of the three carbon changes, and cov() is the covariance of the two carbon changes; The relative contribution of incoming carbon, retained carbon and dead carbon to the change of net carbon sequestration was calculated using R R , R G , R M The structural equation model of forest net carbon sequestration was constructed.

Images