Method for regulating and controlling tree canopy structure

Abstract

The invention belongs to the technical field of statistical methods, and particularly relates to a method for regulating and controlling a tree canopy structure. According to the method, a nonlinear approximate ircorrelation equation of the tree height, the branch height and the effective crown height of the man-made forest and environmental factors of the average temperature in spring, the moisture loss in autumn and the moisture loss in spring is constructed, and the equation quantifies an interaction mechanism of the environmental factors and the relative sectional area on the dynamic change of the vertical structure of the crown; therefore, the canopies can be accurately managed.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of statistical methods, and particularly relates to a method for regulating tree crown structure. BACKGROUND

[0002] As a key functional unit in forest ecosystem, the spatiotemporal heterogeneity of crown structure controls the productivity and ecosystem services of forest stand, especially in regulating light interception efficiency and resource allocation pattern. The dynamic process of tree crown development has certain influence on wood partization and knot formation, and is the biological basis for forming wood quality. Among the parameters for quantifying crown structure, tree height, branch height and effective crown height constitute an important index system. The tree height is denoted as HT, the branch height is denoted as HCB, and the effective crown height is denoted as HEC. As the primary index reflecting the growth potential and competitive ability of trees, the tree height directly determines the vertical structure and biomass distribution pattern of the forest stand. The branch height is defined as the vertical distance from the ground to the base of the first surviving branch, which not only reflects the ecological adaptation strategy of trees, but also mechanically affects the development of tree taper through changing the stress distribution pattern of the tree trunk. Due to the vertical gradient change of light conditions within the crown, the functional contribution of branches at different heights is significantly different. Based on this understanding, researchers divide the vertical structure of the crown into different functional regions. According to the demarcation of whether the branches can achieve excess photosynthesis to promote tree growth, the concepts of "functional crown" and "non-functional crown" are proposed. Based on the difference in functional contribution, researchers divide the vertical structure of the crown into different functional regions. Subsequently, researchers further develop the classification system of "sun crown and shade crown" or "effective crown and ineffective crown" by analyzing the distribution pattern of the increment of trunk cross-sectional area in the vertical direction. This region is usually located in the upper part of the crown and can fully accept light to achieve efficient photosynthesis. Correspondingly, the region between the branch height and the effective crown height is defined as the non-effective crown, and the photosynthetic products of these non-effective crown branches are often insufficient to maintain their own growth and consumption due to insufficient light. These non-effective crown branches not only have limited contribution to volume growth, but also form large knots to reduce wood quality, and even may become a source of infection, threatening the overall health of the forest stand. Crown recession, i.e., the process of the base branches of the crown gradually dying due to the shade of the upper branches and leaves, is the main driving factor for the dynamic change of the branch height and the effective crown height. Accurate measurement and monitoring of the three key indicators of tree height, branch height and effective crown height are crucial for scientific evaluation of forest stand structure and function. By removing the non-effective crown part through artificial pruning, the crown structure can be effectively regulated, the resource allocation can be optimized, the wood quality can be improved, and the health status of the forest stand can be improved, which has important significance for achieving sustainable forest management and improving the economic value of wood. This precise management strategy based on the functional zoning of the crown provides a scientific basis for sustainable forest management, and has important significance for improving the economic value of wood and the function of ecosystem services.

[0003] Field measurement of tree height, branch height, and effective crown height is time-consuming, labor-intensive, and costly, especially in dense forest ecosystems. The accuracy and precision of measurement are often affected by human factors. Therefore, developing reliable statistical models is an effective way to estimate the dynamic changes of these tree crown parameters. Current research on tree crown modeling has two limitations: first, most studies have only constructed static crown size models, which cannot assess the evolution of tree crown over time; second, although existing dynamic models have better predictive performance than static models, obtaining long-term continuous observation data of individual tree crown characteristics still faces many limitations in practice. Knot analysis technology has been proven to be an effective alternative method for reconstructing the longitudinal development history of tree crown. Existing technologies have developed a generalized linear tree crown dynamic prediction model for Pseudotsuga taxifolia by analyzing the height of dead branches and their age of death; developed a dynamic model for crown base height and crown length in northeast larch plantation based on branch mortality technology; and constructed a mixed effect model for tree height, branch height, and effective crown height in Korean pine plantation based on knot analysis technology. However, due to the endogenous correlation between tree height, branch height, and effective crown height, the traditional ordinary least squares method cannot minimize the error of each equation simultaneously, violating the statistical assumption premise. The seemingly unrelated regression estimation method can make the covariance matrix of each equation error asymptotically effective without the limitation of error independence, thereby reducing the standard error of parameter estimation, providing a statistical optimization approach to solve such complex system modeling problems.

[0004] Tree crown growth and development dynamics are regulated by multiple biological and non-biological factors. Tree age, as a key indicator of tree physiological development stage, determines the time trajectory of tree crown morphogenesis. Tree crown dynamic models based on tree age can quantitatively describe the time series changes of vertical and horizontal structure parameters of tree crown. Intraspecific competition accelerates the decline of lower branches by limiting resource acquisition, especially intensifying light competition, thereby changing the pattern of tree crown morphology over time. Site conditions and climate factors, as external driving variables of tree growth, also significantly regulate the tree crown dynamic process. Studies have shown that the better the site conditions, the faster the tree growth rate, and the more significant the allocation of photosynthetic organ biomass and vertical extension within the crown. Under extreme climate conditions, tree crown can exhibit adaptive adjustments through morphological plasticity, such as systematic changes in crown morphology with age. The interaction between competition intensity and climate factors has a more complex regulatory mechanism on the time series dynamics of tree crown. Quantifying this interactive effect has important value for assessing tree sensitivity to climate change and optimizing forest management measures.

[0005] As an important stand density regulation measure, the core scientific problem of artificial pruning is how to prolong the effective crown ratio while optimizing the allocation pattern of photosynthetic products among tree organs by precisely regulating the longitudinal development dynamics of the crown, so as to realize the synergistic improvement of wood quality and carbon storage efficiency. However, there is a problem of insufficient crown management precision in regulating the effective crown structure. SUMMARY

[0006] In order to solve the above technical problems, the present application provides a method for regulating the crown structure of a tree.

[0007] A method for regulating the crown structure of a tree, comprising the following steps: Obtaining data: the data includes the single tree diameter at breast height, tree height, and branch height of the tree; Tree trunk and branch analysis and knot section analysis: using the single tree diameter at breast height to select an analysis tree, performing trunk analysis on the analysis tree to obtain disc data, performing branch analysis to obtain branch penetration depth, and performing knot section analysis to obtain the age of the tree; Using the tree height to determine the site index; Variable calculation: obtaining the tree height of the tree at different age stages according to the branch penetration depth; obtaining the bark-free diameter at breast height of the tree trunk at different ages according to the annual ring width of the disc data; obtaining the bark diameter at breast height at different growth stages according to the bark-free diameter at breast height of the tree trunk at different ages; obtaining the target tree cross-sectional area according to the bark diameter at breast height at different growth stages; calculating the stand cross-sectional area according to the diameter at breast height of the competing tree; obtaining the relative cross-sectional area according to the target tree cross-sectional area and the stand cross-sectional area; obtaining the average spring temperature, the water deficit in autumn, and the water deficit in spring; Obtaining a non-linear seemingly unrelated equation of the predicted tree effective crown according to the tree height at different age stages, the branch height, the relative cross-sectional area, the average spring temperature, the water deficit in autumn, the water deficit in spring, and the site index; Determining the pruning age of the tree according to the non-linear seemingly unrelated equation, and performing pruning at the age to achieve the target effective crown, thereby regulating the crown structure.

[0008] Preferably, the non-linear seemingly unrelated equation is as follows: ; In the formula, H, Hb and He respectively represent the tree height, the branch height and the effective crown height, A represents the age of the tree, R represents the relative cross-sectional area of the single tree competition advantage, , , Hs, Hs and Hs respectively represent the climate indicators of the average spring temperature, the water deficit in autumn and the water deficit in spring, I represents the site index, , 、 、 The intercept term in the HT model of tree height, the model parameters of age, the model parameters of relative basal area, the model parameters of spring average temperature, the model parameters of the interaction term of relative basal area and spring average temperature, and the model parameters of site index; 、 、 、 、 、 The intercept term in the HCB model for underbranch height, the model parameters for age, the model parameters for relative basal area, the model parameters for autumn water deficit, the model parameters for the interaction term between relative basal area and spring average temperature, and the model parameters for site index; 、 They are the intercept term in the effective crown HEC model, the model parameter of age, and the model parameter of spring water loss, respectively.

[0009] Preferably, when selecting analytical trees using the acquired single tree DBH, the minimum and maximum values ​​of the tree DBH D in the plot data are obtained, and then the range from the minimum DBH to the maximum DBH is divided into 5 diameter grades according to the principle of equal intervals, and analytical trees are selected from the 5 diameter grades.

[0010] Preferably, when performing trunk analysis, the trunk is divided into 1-meter-long segments from the base to the treetop, and 3-5 cm thick discs are cut at the stump, each segment, and 1.3 m apart.

[0011] Preferably, the average spring temperature, autumn water deficit and spring water deficit data of each sample plot are obtained based on a spatial interpolation method of the longitude, latitude and altitude of the sample plot.

[0012] Preferably, the relative cross-sectional area is a static relative cross-sectional area.

[0013] Preferably, trees within a radius of 8 m from the target tree are selected as competing trees.

[0014] The method is applied to the regulation of canopy structure of Korean pine plantations.

[0015] Preferably, the effective crown value is set according to the site quality and climate pressure, and the age of the pruning tree is obtained according to the nonlinear seemingly unrelated equation.

[0016] Compared with the prior art, the present invention has the following beneficial effects: The present invention constructs a nonlinear seemingly unrelated equation for the environmental factors of artificial forest tree height, branch height, effective crown height and average spring temperature, autumn water loss and spring water loss. This equation quantifies the interaction mechanism between environmental factors and relative basal area on the dynamic changes of the vertical structure of the crown, so that the canopy can be managed accurately.

[0017] The present application uses the equation to determine the tree pruning tree age, in which the pruning intervention is implemented, the crown and stem ratio is systematically optimized, the formation of high-quality wood is continuously promoted, and finally the precise regulation of the crown structure of the Korean pine plantation is realized.

[0018] Based on the constructed simultaneous equation model, the optimization scheme of Korean pine plantation pruning technology is formulated for different site conditions and climate change scenarios, providing a theoretical basis and decision support tool for the directional cultivation of Korean pine large-diameter non-knot timber, and establishing a scientific prediction model for the adaptability adjustment of forest management scheme under the background of climate change. Based on the method of the present application, the effective crown height of artificial Korean pine can be predicted, the consistency and reliability of the prediction are improved, and the dynamic precise prediction of the time series of the crown structure parameters is realized DETAILED DESCRIPTION

[0019] The specific embodiments of the present application will be described in detail below, but it should be understood that the protection scope of the present application is not limited by the specific embodiments. Based on the examples in the present application, all other examples obtained by those of ordinary skill in the art without creative labor belong to the scope of protection of the present application. The experimental methods described in the embodiments of the present application are conventional methods unless otherwise specified.

[0020] 1. Overview of the study area The present application selects two artificial forest sample plots in the temperate continental monsoon climate zone, located in Mengjiagang Forest Farm of Jiamusi City, Heilongjiang Province and Qinghecheng Forest Farm of Benxi City, Liaoning Province in Northeast China. The climate conditions in these two regions are temperate continental monsoon climate, characterized by cold winters and hot summers, moderate precipitation, and relatively short frost-free periods. The annual average precipitation in Mengjiagang Forest Farm is 550 mm, the annual average temperature is 2.7℃, the winter minimum temperature can reach-34.7℃, the annual average frost-free period is 120 days, and the soil is mainly dark brown forest soil. The annual average temperature in Qinghecheng Forest Farm is 6℃-8℃, the annual average frost-free period is 120-125 days, the annual average precipitation is 800-1000 mm, the winter minimum temperature is-37.7℃, the summer maximum temperature is 36.5℃, and the soil type is mainly brown soil. The significant differences between the two regions provide an ideal environment for comparative analysis of the growth characteristics of artificial Korean pine forests.

[0021] 2. Data acquisition and processing (1) Data acquisition In Mengjiagang Forest Farm and Qinghecheng Forest Farm, 12 plots of 30-46 year-old and 18 plots of 28-70 year-old artificial Korean pine forests were set up in 2010 and 2023, respectively, with an area of 20m x 30m. The present application measures the diameter at breast height and tree height of each tree in all sample plots.

[0022] The breast diameter is abbreviated as D, and the unit is cm; The single tree height is abbreviated as HT, and the unit is m; According to the equal cross-sectional area diameter class standard tree method, the present application divides the trees into five grades in total, and the division standards of each grade are as follows: first, the minimum value and the maximum value of the breast diameter D of the sample plot data are obtained, and then the range from the minimum breast diameter to the maximum breast diameter is divided into five diameter class grades according to the equal interval principle, that is, the first grade to the fifth grade, and each grade corresponds to a specific breast diameter range.

[0023] For each tree in the sample plot, the grade to which the tree belongs is determined according to the size of the breast diameter, and then the number of trees in each diameter class grade is counted to determine the average breast diameter.

[0024] The single tree cross-sectional area of each tree is calculated using the breast diameter data, as shown in the following formula: Further, the total cross-sectional area of each diameter class grade and the total cross-sectional area of the whole sample plot are calculated. Next, the total number of standard trees required to be selected in each diameter class grade is determined, and then the number of standard trees to be selected in each diameter class grade is allocated according to the proportion of the cross-sectional area of each grade to the total cross-sectional area of the sample plot, that is, the more the proportion of the cross-sectional area, the more the number of standard trees allocated in the grade. Finally, in each diameter class grade, the representative trees with normal growth and no disease and insect infestation are selected as the standard trees of the grade by using the method of random sampling or systematic sampling. After the selection of the standard trees is completed, the average breast diameter and the average tree height of all the standard trees in each grade are calculated using D and H in the sample plot data, respectively, as the representative data of the trees in the grade.

[0025] Based on the above-mentioned tree grade division standard, in each sample plot of Mengjiagang Forest Farm, five artificial Korean pines with basically the same breast diameter and tree height are selected as the analysis trees for each grade, and in Qinghecheng Forest Farm, the present application adopts a different tree selection strategy, specifically, the secondary average breast diameter value of the first grade, the third grade and the fifth grade is taken as the selection standard, and the trees with corresponding breast diameter range are selected as the analysis trees, a total of 114 trees are selected for trunk analysis, branch analysis and knot sectioning, wherein the secondary average breast diameter value refers to the average value obtained by first calculating the sum of the squares of the breast diameters of the trees in each grade, and then dividing the sum by the number of trees and taking the square root.

[0026] (2) Trunk and branch analysis and knot sectioning Trunk analysis: when felling the sample tree, careful operation is performed to avoid branch breakage as much as possible to ensure data integrity. After felling, the trunk is divided into 1m-long sections from the base to the top, and 3cm-5cm-thick discs are cut at the root cutting place, each section and 1.3m.

[0027] Branch analysis: each round of branches is systematically numbered according to the age and spatial position of the branches. Further, the branch penetration depth of all the branches is measured.

[0028] Node analysis: The position below the base of the crown is segmented at intervals of 10-30 cm, the node-containing part is analyzed and numbered, and the node height is recorded. At the same time, the position of the node is judged according to the scar left on the trunk after the branch dies, the trunk is vertically longitudinally sectioned along the center line of the pith, the node longitudinal section image is obtained according to the node height and the node position, and the node longitudinal section image is digitized by a scanner.

[0029] Based on the digitization result of the node longitudinal section image, the node development process is divided into four key time nodes: node B: the age of the branch growing outward from the pith, i.e. the appearance age of the node, node C: the age of the branch completely stopping growing, i.e. the growth stopping age of the node, node D: the age of the branch losing growth vitality and dying, i.e. the death age of the node, and node O: the age of the branch being completely covered by the trunk, i.e. the covering age of the node.

[0030] Based on these time nodes, the node measurement software is used to analyze the number of annual rings from each time node to the pith to obtain the age.

[0031] Since the photosynthetic capacity of the branch is weakened when it stops growing, it no longer contributes to the growth of the trunk, therefore, all branches above the height at which the branch stops growing are defined as "effective crown", i.e. effective crown at different ages. Similarly, after the branch dies, it cannot provide nutrients for the trunk and itself, and all these branches constitute the "live crown", and the position of the branch death in the trunk at different ages can determine the under-canopy height at different ages. It should be particularly noted that when determining the effective crown height and the under-canopy height by using the node analysis data, the following principles should be followed: when the growth stopping age and the death age of the node at a certain height are both greater than the corresponding ages of other nodes at the same height, the height can be determined as the effective height of the crown or the under-canopy height at the age of the tree. This hierarchical analysis method based on the development stage of the node can ensure the longitudinal continuity and time consistency of the crown change.

[0032] (3) Determining site index by using the tree height The average diameter at breast height and the tree height of the dominant tree are represented by the average value of the six thickest trees in each sample plot. The site index is calculated according to the established guide curve, and the height of the dominant tree at the age of 40 years is taken as the standard, and the calculation formula is as follows:

[0033] Formula (1.1); Formula (1.2); In the formula, is the site index, is the stand age, is the predicted height of the dominant tree at the age of , Predicted dominant height for the reference age, Measured dominant height of the stand.

[0034] (4) Variable calculation a. Dynamic variable calculation Dynamic variables are the height of trees at different ages, the bark diameter of tree trunks at different ages, the relative cross-sectional area at different ages, the average spring temperature at different ages, the water deficit in autumn at different ages, and the water deficit in spring at different ages.

[0035] 1) Calculation of tree height at different ages: Korean pine, as a needle evergreen plant, usually has a clear uniaxial growth pattern, with the main stem growing upward year by year and forming a whorl of branches at the end of a certain growth cycle, which means that the end of each annual growth cycle is marked by the formation of a new branch whorl, i.e. a number of lateral branches spiraling from a specific part of the main stem. This growth pattern can serve as a theoretical basis for reconstructing the growth history of tree height. This whorl feature can confirm the significant correlation between tree height growth and branch attachment depth. The present invention utilizes this relationship by measuring the resolved branch attachment depth of all branches in each whorl, calculating the average value, and taking this value as the estimated value of the tree height in the formation year of the branch whorl. In this way, the present invention can trace the tree height of the sample at different ages.

[0036] 2) Calculation of bark diameter of tree trunk at different ages: For the historical reconstruction of the diameter of tree trunk at different ages, the present invention uses the bark coefficient method combined with annual ring analysis technology to achieve accurate estimation. Specifically, first, a tree trunk disc is cut at breast height, i.e. 1.3 meters from the ground, and the width of each annual ring from the tree core to the bark is precisely measured, and the bark-free diameter DBH1 of the tree trunk at each year from the seedling stage to the current age is reconstructed, i.e. the diameter of the pure wood part without including the thickness of the bark; At the same time, the bark diameter DBH0 at breast height of the tree at the time of felling is measured as the reference data, which includes the thickness of the bark. Next, in order to accurately estimate the bark diameter of each year in history, a mathematical relationship model between the bark coefficient and the age is established, where the bark coefficient KB is defined as the ratio of the bark-free diameter to the bark diameter, i.e. KB = DBH1 / DBH0, which reflects the proportion of the thickness of the bark relative to the diameter of the entire tree trunk. Through experimental data analysis, it is found that there is a stable power function relationship between the bark coefficient KB and the age Age, and the mathematical expression is KB = a x Age b, where a and b are the model parameters to be solved, and the tree age is calculated as the number of trunk rings or branch rings. Using the three known benchmark data of the year, namely the tree age Age, the diameter at breast height with bark DBH0, and the diameter at breast height without bark DBH1, the bark coefficient KB of the year is first calculated. Then, the KB value and the corresponding tree age Age are substituted into the power function formula, and the specific values ​​of the parameters a and b are solved by mathematical fitting methods. After obtaining the model parameters, according to the tree age in any historical year, KB=a×Age is used. b The bark coefficient of the tree is calculated. Combined with the known DBH1 of the tree from the tree ring analysis, the DBH0 of the tree with bark is calculated as DBH1 / KB. This completes the reconstruction of the DBH0 of the tree from the seedling stage to the current age. The DBH0 of the tree with bark at different ages is obtained, providing accurate basic data support for the subsequent calculation of the tree's basal area. The calculation formula is formula (2).

[0037] Formula (2); Where, is the trunk's diameter at breast height with bark, The bark-less diameter at breast height of tree trunks at different ages, is the bark coefficient, Age is the age of the tree, =1.34, =-0.01.

[0038] 3) Dynamic calculation of relative cross-sectional area at different age stages: use Calculate the basal area of ​​target trees at different ages ,formula , where Are of different ages The diameter of a tree at breast height, Are of different ages The basal area of ​​a tree.

[0039] The selection of competing trees adopts a fixed radius method, based on the optimal competition radius of 8 meters or less confirmed by previous studies. The present invention sets the maximum competition radius to 8 meters, defines trees within the competition radius of the target tree as competing trees, and uses growth cones to drill east-west and south-north tree cores at the diameter at breast height of competing trees. The tree ring width is measured using a tree ring image analysis system to obtain the diameter at breast height of competing trees at different ages. The diameter at breast height of different ages is used to calculate the basal area of ​​the stand at different ages within 8 meters. ,formula , where It is The basal area of ​​a tree, It is the sum of the cross-sectional areas of each tree within 8m.

[0040] Dynamic relative basal area is calculated using target tree basal area and stand basal area at different ages, and the relative basal area is abbreviated , formula .

[0041] 4) Calculation of average spring temperature, autumn water deficit and spring water deficit: Climate AP provides detailed data on average spring temperature, autumn water deficit and spring water deficit in the baseline period and future scenarios as a reliable climate data source. This data source has shown good applicability and accuracy in related studies of forest growth models. Through spatial interpolation methods based on the longitude, latitude and altitude of the sample plot, the average spring temperature, autumn water deficit and spring water deficit data of each sample plot can be obtained.

[0042] Considering the physiological characteristics of Korean pine growth, the present invention particularly obtains the average spring temperature, autumn water deficit and spring water deficit data corresponding to the years of dynamic tree height, effective crown and branch height, in order to deeply analyze the influence mechanism of climate factors on the dynamic change process of Korean pine crown structure. This method enables the present invention to more accurately quantify the correlation between climate variables and the dynamic morphological development of Korean pine crown.

[0043] b, static variable calculation The present invention also calculates the static target tree basal area to stand basal area ratio based on the measurement of the state of the year, i.e. the static relative basal area. The calculation method refers to the calculation method of the dynamic target tree basal area to stand basal area ratio, and the corresponding diameter at breast height data in the formula is replaced by the diameter at breast height data of the year of data acquisition.

[0044] Although the overall framework of the present invention aims to establish a dynamic growth model, the accuracy comparison analysis after substituting different target tree basal area to stand basal area ratios into the basic model shows that the static relative basal area index is close to the dynamic relative basal area in terms of model fitting effect. Considering the time and economic cost-effectiveness, the present invention finally selects the static relative basal area to represent the competition between trees.

[0045] 3, establishment of basic and generalized models According to the tree height, branch height, relative basal area, average spring temperature, autumn water deficit and spring water deficit, and site index at different ages, a nonlinear non-correlation equation for predicting the effective crown of the tree is obtained, i.e. model establishment.

[0046] Establishment of basic model The present application fits the Logistic equation, the Mitscherlich equation, the Gompertz equation, the Richards equation and the Korf equation to the tree height, the branch height and the effective crown of the tree trunk of different ages to obtain a1 and a2. The optimal basic model of each growth index is determined by the model test shown in formulas (9)-(14), and the optimal basic model is the Mitscherlich equation and the Logistic equation. The Mitscherlich equation, that is, formula (3), shows the best fitting effect and is selected as the basic model for dynamic prediction of tree height; and the Logistic equation, that is, formula (4), has the optimal performance in describing the growth law of branch height.

[0047] Considering the constraint condition of the physiological characteristics of trees, the basic model of branch height, that is, formula (4), is mathematically deformed to construct formula (5) as the basic model of effective crown height, so that the predicted effective crown height is always greater than the branch height, thereby meeting the actual situation of tree growth.

[0048] Formula (3); Formula (4); Formula (5).

[0049] Establishment of a generalized model In order to improve the prediction accuracy and applicability of the model, a generalized model is constructed, and eight variables including tree height, effective crown, branch height, tree age, relative cross-sectional area, average spring temperature, water deficit in autumn and water deficit in spring are considered in the construction of the generalized model. In the model construction process, in order to ensure the robustness of the model and effectively avoid the problem of multicollinearity between variables, the variance inflation factor (VIF<5) is used as a statistical criterion for variable screening, and the candidate variable combination is systematically evaluated and optimally selected. After the above modeling process, the three optimal generalized models (tree height, branch height and effective crown height) constitute a seemingly unrelated regression model system, that is, formula (16) as follows.

[0050] Formula (16); In the formula, tree height, branch height and effective crown height, tree age, the relative cross-sectional area representing the competitive advantage of single trees, , , representing the climate indexes of average spring temperature, water deficit in autumn and water deficit in spring, representing the site index. , , , Intercept term in the height above ground (HT) model, model parameter of age, model parameter of relative basal area, model parameter of average temperature in spring, model parameter of the interaction term of relative basal area and average temperature in spring, model parameter of site index, respectively; , , , , , Intercept term in the height under branch (HCB) model, model parameter of age, model parameter of relative basal area, model parameter of water deficit in autumn, model parameter of the interaction term of relative basal area and average temperature in spring, model parameter of site index, respectively; , Intercept term in the effective crown (HEC) model, model parameter of age, model parameter of water deficit in spring, respectively.

[0051] The model system can simultaneously predict multiple growth indicators of artificial Korean pine, and improve the consistency and reliability of prediction.

[0052] Seemingly unrelated model introduction Seemingly unrelated regression, English name, is a system of equation groups composed of multiple regression equations, which allows different combinations of independent variables for each equation, providing great flexibility for statistical modeling. The core statistical advantage of the SUR model lies in the innovation of its parameter estimation strategy: unlike the traditional single equation independent estimation method, the SUR model can handle heteroscedasticity and fully utilize the correlation between error terms in different equations during parameter estimation. This joint estimation strategy makes the parameter estimation efficiency significantly better than the traditional method of separate equation estimation under certain statistical conditions. The model assumes is the sample size, is the dependent variable, independent variables ( ), the SUR model can be constructed as a system of nonlinear equations, which is shown in formulas (6)-(8).

[0053] The height, height under branch and effective crown height prediction model composed of formula (16) in the application is a specific application based on the above SUR theoretical framework. Through joint modeling, the internal correlation between the growth indicators of artificial Korean pine is effectively captured, thereby realizing the simultaneous prediction and system optimization of multiple growth indicators, and reflecting the important application value of the method in forest growth modeling.

[0054] Equation (6); Equation (7); Equation (8); where, and is N a 1 N × k matrix, is a k i 1 dimensional vector, , and i =1,2,…, m . Assume that the model error terms are independent over time series, but have correlation between equations, where ir denotes a number consisting of two digits, i =1,2,…, j , m , r =1,2,…, s , N where i and j are positive integers. Thus, the present invention assumes that when r ≠ s , E ( e ir , e js | X )=0, and otherwise E ( e ir , e jr | X )= σ ij . ∑ =[ σ ij ] represents the m × m conditional variance matrix of each observation, e i The covariance matrix of .

[0055] 4. Model checking and evaluation To comprehensively evaluate the fitting degree and prediction accuracy of the established basic model in the present application, the present application adopts multiple statistical indexes to quantitatively evaluate the model performance. The model fitting effect evaluation adopts statistical indexes such as the determination coefficient, the root mean square error, the mean absolute error and the mean error, to comprehensively evaluate the fitting degree and deviation characteristics of the model to the original data. These indexes can reflect the fitting degree, the prediction accuracy and the systematic deviation of the model from different angles. Considering the prediction performance in the actual application of the model, the present application adopts the leave-one-out cross-validation method to systematically test the generalization ability of the model. This method is based on the "cutting method" principle, and each time randomly selects sample woods from sample woods for model fitting, and the remaining 1 sample wood is used as a verification sample to test the prediction accuracy. The process is repeated times to ensure that each sample participates in the test as a verification sample, so as to obtain an unbiased model prediction performance evaluation. Among them, the determination coefficient is denoted as , the root mean square error is denoted as , the mean absolute error is denoted as , and the mean error is denoted as .

[0056] Based on the cross-validation results, in addition to calculating , , , the present application also calculates the fitting index and the mean absolute percentage error, to comprehensively evaluate the prediction performance and robustness of the model. The calculation formula of is formula (9), is formula (10), is formula (11), is formula (12), is formula (13), and is formula (14). Among them, the fitting index is denoted as , and the mean absolute percentage error is denoted as .

[0057] Formula (9); Formula (10); Formula (11); Formula (12); Formula (13); Formula (14); Where, is the number of model parameters; ， and are the observed, predicted and average values ​​of tree height, effective crown height or under-branch height, respectively; It is the predicted value of tree height or effective crown height or underbranch height obtained by the "jackknife method"; is the sample size. and It reflects the degree to which the model explains the variation of the observed data. The value range is 0~1. The closer the value is to 1, the better the model fitting effect is. and The model prediction accuracy is measured from the perspective of square error and absolute error respectively. The smaller the value, the higher the prediction accuracy. Reflects the systematic deviation of the model prediction, with positive values ​​indicating overestimation and negative values ​​indicating underestimation; It reflects the relative error of the model prediction. The smaller the value, the higher the prediction reliability.

[0058] 6. Proposal of artificial pruning plan Based on the tree height, height under branches and effective crown of trees at different age stages, the bark-containing breast diameter of trunks at different ages, relative basal area, average spring temperature, autumn water loss and spring water loss, the present invention constructs seemingly unrelated independent models of tree height, height under branches and effective crown height, providing a theoretical basis for determining the timing of pruning under different sites and climatic conditions.

[0059] Taking into account the significant impact of site quality and climate pressure on tree growth, the present invention establishes a graded environmental condition evaluation system. Site quality is divided into three levels: high, medium and low according to the site index value range, and the classification criteria are as follows: the scenario with high site quality is set to the maximum value of the site index; the scenario with medium site quality is set to the middle value of the site index; the scenario with low site quality is set to the minimum value of the site index. The climate pressure level evaluation is based on the key climate factors in the canopy parameter model, including spring average temperature, spring water deficit and autumn water deficit. The present invention sets three climate pressure levels: high climate pressure, medium climate pressure and low climate pressure, and the setting criteria are as follows: the high climate pressure scenario is set to the combination of the minimum spring average temperature and the maximum spring and autumn water deficit; the medium climate pressure scenario adopts the middle value of each climate factor; the low climate pressure scenario corresponds to the combination of the maximum spring average temperature and the minimum spring and autumn water deficit.

[0060] For nine combined scenarios of three levels of site quality and three levels of climate pressure, the present invention designs a phased pruning scheme based on the effective crown height threshold. The scheme divides the pruning process into four stages, which are implemented when the effective crown height reaches 2 meters, 4 meters, 6 meters, and 8 meters respectively. The specific implementation method is: by calculating the inverse function of the effective crown height model, the corresponding tree age when the effective crown height reaches 2 meters, 4 meters, 6 meters, and 8 meters is determined. , so as to clarify the specific time of pruning at each stage. The first pruning is when the effective crown height reaches 2 meters Implementation, by Substitute the branch height model to calculate the current branch height , the first pruning intensity is ( ) meters, that is, the height of the effective crown is 2 meters. The second, third and fourth pruning are respectively at the corresponding ages when the effective crown height reaches 4 meters, 6 meters and 8 meters. Pruning is completed when the effective crown height reaches 8 meters. This phased pruning method conforms to the biological laws of the transition from saplings to mature trees. Mathematical models accurately predict the optimal pruning timing at each stage, providing a scientific basis for pruning timing decisions in plantation management.

[0061] result 1. Nonlinearity seems irrelevant to model development Based on the age-related basic model, the present invention gradually introduced variables such as site index, competition, and climate through significance tests and the minimum AIC criterion. The optimal generalized model constituted the final nonlinear seemingly unrelated model, namely formula (16). There was no multicollinearity between the variables in the model, and all parameter estimates were statistically significant. The results are shown in Table 1. Considering that tree height has been included as an independent variable in the underbranch height and effective crown models, and underbranch height is also an independent variable in the effective crown model, in order to simplify the model and ensure convergence, the present invention avoids introducing too many variables in the effective crown model. For example, although competition-related variables do not appear directly in the effective crown model, they are included in the tree height and underbranch height models. The present invention will use the variable transfer relationship of the seemingly unrelated model to analyze these indirect effects. The parameter estimation results show that the parameters a3 and b3, which represent the relative basal area of ​​the competitive advantage of individual trees, are positively correlated with tree height, underbranch height, and effective crown. In terms of climate variables, a5, which represents the average spring temperature, is positively correlated with tree height, while b4 and c3, which represent the degree of drought, have a positive impact on underbranch height and effective crown. The site index showed a positive effect on all three dependent variables. In addition, parameters a5 and b5, as the interaction coefficients between competition and climate, indicate that climate and competition factors regulate each other and jointly affect tree growth characteristics.

[0062] Tables 2 and 3 show the fit statistics and leave-one-out cross-validation results. The analysis indicates that both the seemingly uncorrelated model and the base model exhibit good fit, with no significant difference in fitting accuracy between the two. However, the SUR model exhibits certain advantages in theoretical structure by accounting for the inherent correlations between equations. This model is able to capture the complex relationships between variables, making it more realistic.

[0063] Table 1 Parameter estimates of the model Note: In the P column, * represents P < 0.05, ** represents P < 0.01, *** represents P < 0.005, and . represents P < 0.1.

[0064] Table 2 Model fitting indicators Table 3 Model test indicators 2. Design of time and intensity plan for manual pruning In nine combination scenarios consisting of three levels of site quality and three levels of climate pressure, the present invention uses the SUR model system to simulate the graded pruning timing based on the effective crown height threshold. Through the inverse function calculation of the effective crown height model, the optimal implementation time of each pruning stage is accurately determined: the corresponding forest age when the effective crown height reaches 2 meters, 4 meters, 6 meters, and 8 meters is calculated respectively as the time nodes of the four pruning operations, see Table 4. This method of determining the timing of pruning based on the developmental state of effective crown height fully considers the differences in tree growth under different environmental conditions, and provides an accurate reference basis for pruning time for forestry production practice. Through this solution, operators can accurately grasp the best time for pruning according to the specific site conditions and climate environment, and realize the scientific management of artificial forests.

[0065] Table 4 Pruning schemes under different sites and climate pressure intensities Korean pine is one of the most ecologically and economically valuable plantation tree species in Northeast China, occupying a central role in the regional forest carbon sequestration. Improving the quality and biomass productivity of Korean pine wood is a strategic goal of forest management. However, current management practices lack precise canopy management, resulting in a yield of high-quality timber from Korean pine plantations that falls below theoretical potential, constraining the security of the country's strategic timber reserves.

[0066] The present application firstly integrates the branch analysis technology and the nonlinear seemingly unrelated equation model to establish a quantitative framework of the dynamic change of the crown of Korean pine under the joint action of climate factors and competition pressure, and uses the equation to realize accurate prediction of the time series dynamic of the crown structure parameters, thereby providing a theoretical basis for scientifically formulating the artificial pruning scheme, and the nonlinear seemingly unrelated equation model is denoted as NSUR.

[0067] Firstly, the research results clearly verify that the mediation effect of tree size on the response to different environmental factors has significant differences. Secondly, the model results of the present application strongly prove that the competition advantage plays a core regulation role in regulating the climate response of trees. In addition, the competition-climate interaction mechanism explains the formation reason of the individual development difference in the stand, and provides an important reference for predicting the forest structure dynamics under the background of future climate change. Finally, based on the simulation results of the NSUR model, the present application innovatively constructs a set of phased pruning decision-making framework considering site index, forest age and different climates. By implementing appropriate pruning intervention at the key nodes of tree growth, the crown-shoot ratio is systematically optimized, the formation of high-quality timber is continuously promoted, and finally the precise regulation of the crown structure of Korean pine plantation is realized. The methodological framework established by the present application provides a new theoretical tool and decision support for forest adaptive management. The research results not only deepen the scientific understanding of the dynamic mechanism of crown structure, but also provide a pruning scheme with practical guiding significance for forest managers.

[0068] It should be noted that when the present application claims involving numerical ranges, both endpoints of each numerical range and any number between the two endpoints can be selected. In order to prevent repetition, the present application describes preferred embodiments.

[0069] Although the preferred embodiments of the present application have been described, those skilled in the art can make further changes and modifications to the embodiments once they know the basic inventive concept. Therefore, the appended claims are intended to include the preferred embodiments and all changes and modifications falling within the scope of the present application.

[0070] Obviously, those skilled in the art can make various modifications and variations to the present application without departing from the spirit and scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the claims of the present application and their equivalent technologies, the present application also intends to include these modifications and variations.

Claims

1. A method for regulating tree canopy structure, characterized in that: The following steps are involved: Acquiring data: The data includes the tree's single tree diameter at breast height, tree height, and height below branches; Trunk and branch analysis and node analysis: using the single tree DBH to select parsed trees, performing trunk analysis on the parsed trees to obtain disk data, performing branch analysis to obtain branching depth, and performing node analysis to obtain tree age; determining a site index using the tree height; Variable calculation: obtaining the height of trees at different age stages based on the branching depth; obtaining the barked diameter at breast height of trunks at different ages based on the annual ring width of the disk data; obtaining the barked diameter at breast height of trunks at different ages based on the barked diameter at breast height of trunks at different ages; obtaining the target tree basal area based on the barked diameter at breast height of trunks at different growth stages; calculating the stand basal area based on the diameter at breast height of competing trees; obtaining the relative basal area based on the target tree basal area and the stand basal area; obtaining the average spring temperature, autumn water deficit, and spring water deficit; According to the tree height, height below the branches, relative basal area, average spring temperature, autumn water deficit, spring water deficit and site index at different age stages, a nonlinear seemingly unrelated equation was obtained to predict the effective crown of trees. The age of tree pruning is determined according to nonlinear seemingly unrelated equations, and pruning is performed at this age to achieve the target effective crown, thereby regulating the canopy structure.

2. The method according to claim 1, characterized in that The nonlinearity seems to be irrelevant to the equation below: ； Where, Represent tree height, branch height and effective crown height respectively. Represents the age of the tree, Relative basal area representing the competitive advantage of a single tree, 、 、 Represent the climate indicators of spring average temperature, autumn and spring water loss, represents the site index, 、 、 、 The intercept term in the HT model of tree height, the model parameters of age, the model parameters of relative basal area, the model parameters of spring average temperature, the model parameters of the interaction term of relative basal area and spring average temperature, and the model parameters of site index; 、 、 、 、 、 The intercept term in the HCB model for underbranch height, the model parameters for age, the model parameters for relative basal area, the model parameters for autumn water deficit, the model parameters for the interaction term between relative basal area and spring average temperature, and the model parameters for site index; 、 They are the intercept term in the effective crown HEC model, the model parameter of age, and the model parameter of spring water loss, respectively.

3. The method according to claim 1, characterized in that When using the obtained single tree DBH to select analytical trees, the minimum and maximum values ​​of the tree DBH D in the plot data are obtained, and then the range from the minimum DBH to the maximum DBH is divided into 5 diameter grades according to the principle of equal intervals, and analytical trees are selected from the 5 diameter grades.

4. The method according to claim 1, wherein When analyzing the trunk, the trunk is divided into 1-meter-long sections from the base to the top, and 3-5-cm-thick discs are cut at the stump, each section, and 1.3-meter apart.

5. The method according to claim 1, wherein Based on the spatial interpolation method of the longitude, latitude and altitude of the sample site, the average spring temperature, autumn water deficit and spring water deficit data of each sample site were obtained.

6. The method according to claim 1, characterized in that The relative cross-sectional area is a static relative cross-sectional area.

7. The method according to claim 1, characterized in that Trees within a radius of 8 m from the target tree were selected as competing trees.

8. The method according to claim 1, characterized in that According to the site quality and climate pressure, the effective crown value is set, and the age of the whole tree is obtained according to the nonlinear seemingly uncorrelated equation.

Images