Management decision-making method for artificial broadleaf mixed forest in South China

Abstract

The invention discloses a South China artificial broadleaf mixed forest management decision-making method. The method comprises the following steps: constructing a sample plot initial forest stand database containing spatial information; calculating the important value of each tree species of the sample plot according to the initial forest stand database of the sample plot so as to determine the target tree species of the sample plot; performing spatial structure parameter calculation on the trees in the sample plot, performing optimization, and then analyzing and checking the spatial relevance between the target tree species and the associated tree species by adopting a bivariate point pattern; and providing an optimal operation decision scheme for adjusting tree species composition and spatial configuration through structure optimization simulation by combining a forest space structure and a spatial correlation result and combining a stand current situation and a set operation target. According to the method, spatial structure parameter analysis and point pattern analysis are combined, large-diameter timber cultivation for the target tree is focused on, the inter-species spatial relevance between the target tree species and the associated tree species in the stand is comprehensively considered, and therefore a scientific basis is provided for stand spatial structure optimization of an existing artificial broad-leaved mixed forest.

Description

Technical Field

[0001] This invention relates to the field of forest management technology, specifically to a decision-making method for the management of artificial broadleaf mixed forests in South China. Background Technology

[0002] Broadleaf mixed forests possess advantages such as complex structure, rich biodiversity, and stable ecosystem function, making them crucial for water conservation, soil and water conservation, biodiversity protection, and timber production. Scientifically and rationally optimizing their tree species composition and spatial configuration is key to improving forest quality and maximizing multiple benefits. It is also an important goal of modern forest management and for promoting sustainable forest management in South China.

[0003] During a specific historical stage of afforestation in South China, the "inter-row mixed planting" and "cluster planting" models reflected the forestry management philosophy at the time, which focused on rapid growth and high yield. This model tended towards monoculture in tree species selection, resulting in a limited composition of target tree species and significantly insufficient species diversity. From the dynamic perspective of community succession, this initial configuration artificially presupposes spatial isolation between tree species. As natural competition and succession progress, it not only fails to form a uniformly interspersed and structurally stable mixed forest, but also intensifies intraspecific competition, causing the forest to develop towards a clustered, isolated structure. This structure is inferior to naturally formed uneven-aged, multi-layered mixed forests in both ecological function and stability.

[0004] To quantify forest stand structure, scholars both domestically and internationally have developed various methods. Among them, the spatial structure parameter system based on adjacent tree relationships (including angular scale, mixing degree, and size ratio) proposed by Hui Gangying et al. has gained widespread recognition and application in my country due to its suitability to the actual situation of forest management. This system reveals the spatial relationships of trees at the micro-unit level:

[0005] Table 1 Spatial Structure Parameters

[0006] Spatial structure parameters definition formula Parameter meaning Angular scale Describe the horizontal distribution pattern of the target tree and its adjacent trees. For a discrete variable, when the j-th... Angle smaller than standard angle hour, ;on the contrary, . degree of mixing Describes the degree of species isolation between the target tree and its adjacent trees. As a discrete variable, when the reference number When the adjacent tree is not of the same species as the j-th tree, ;on the contrary, . Size ratio Describe the competitive growth advantage of the target tree compared to its neighboring trees. For a discrete variable, when the first... The ratio of adjacent trees to reference trees Hour, ;on the contrary, .

[0007] Point pattern analysis (such as O-ring statistics) reveals the spatial distribution patterns and interspecific relationships of populations at a macro scale. It is generally believed that when the target tree species and its companion tree species are negatively associated at a specific scale, there is competitive inhibition; when there is no association, it is a neutral effect; and when there is a positive association, there may be mutualism or shelter-promoting effects.

[0008] However, existing technologies still have shortcomings: on the one hand, management optimization based on spatial structure parameters often focuses on cultivating large-diameter individual trees, which may neglect the interspecific synergistic relationships of the entire stand; on the other hand, while point pattern analysis can diagnose macroscopic correlations, it is difficult to directly guide specific, individual-level management measures. These two methods differ in scale and have failed to be organically integrated, making it difficult for management decisions to simultaneously achieve the dual goals of "cultivating high-quality individuals" and "building harmonious communities." Summary of the Invention

[0009] The purpose of this invention is to overcome the shortcomings of the prior art and provide a management decision-making method for artificial broad-leaved mixed forests in South China, so as to solve the problems of single objectives and disconnected measures in the management of existing artificial broad-leaved mixed forests.

[0010] To achieve the above objectives, the technical solution of the present invention is as follows:

[0011] A management decision-making method for artificial broadleaf mixed forests in South China includes:

[0012] Construct a database of initial forest stands for sample plots containing spatial information;

[0013] The dominance of each tree species is calculated based on the initial stand database of the sample plots to clarify the composition of the target tree species and associated tree species within the sample plots;

[0014] The spatial structure of forest trees was analyzed, and bivariate point pattern analysis was used to examine the spatial correlation between the target tree species and its associated tree species.

[0015] Based on the results of forest stand spatial structure and spatial correlation, combined with the current status of the forest stand and the established management objectives, the optimal management decision scheme for adjusting tree species composition and spatial configuration is proposed through structural optimization simulation.

[0016] Optionally, the analysis of forest spatial structure includes:

[0017] For each sample plot, based on the coordinates and diameter at breast height (DBH) data of individual trees, the size ratio, angular scale, and degree of mixing of each reference tree are calculated; and the average size ratio, average angular scale, and average degree of mixing of each tree species are calculated.

[0018] Optionally, the structural optimization simulation includes: adjusting based on the analysis results of the average size ratio, average angular scale and average mixing degree of each tree species; identifying areas with unreasonable spatial structure based on the size ratio, angular scale and mixing degree of individual trees; and selectively felling individual trees.

[0019] Optionally, the structural optimization simulation further includes:

[0020] High-intensity clustering unit identification: Based on the angular scale parameter, forest units with angular scale values ​​greater than a threshold are identified as high-intensity clustering units.

[0021] Internal status diagnosis of the unit: Calculate the average degree of mixing and the size ratio of trees within the unit for each high-intensity clustered unit, and adjust the degree of clustering;

[0022] Point pattern analysis and iterative verification: After adjusting the degree of clustering, bivariate point pattern analysis was used to examine the spatial correlation between the target tree species and the associated tree species.

[0023] Optionally, the method of using bivariate point pattern analysis to examine the spatial association between the target tree species and its associated tree species includes:

[0024] The Monte Carlo simulation generates a 95% confidence interval. Within the target spatial scale, points where the observed curve is above the upper envelope line indicate a positive correlation, points within the confidence interval indicate no significant correlation, and points below the lower envelope line indicate a negative correlation. If a negative correlation still exists, the adjustment continues, and the next round of adjustment is carried out based on the calculation results of the new spatial structure parameters until the spatial correlation between the target tree species and the associated tree species reaches a statistically non-negative correlation.

[0025] Optionally, the calculation method for the point pattern analysis is as follows:

[0026]

[0027]

[0028]

[0029]

[0030] In the formula: Let be the spatial association value between object 1 and object 2 at point i, with radius r and annular width w; n1 is the number of points of object 1 in the bivariate statistics. It is a ring in object 1 with point i as the center, r as the radius, and w as the width; Point s2 (X) calculates the number of points of object 2 within region X; Area(X) is the area of ​​region X; (x i , y i Let (x, y) be the coordinates of point i in object 1; S(x, y) is a binary variable. If the coordinates (x, y) are within the study area X, then S(x, y) = 1, otherwise S(x, y) = 0; z 2 This indicates the area of ​​a single cell. It is a variable that changes with the circle in object 1 centered at point i and with radius r; This indicates the number of points distributed within the cell for object 2; object 1 is the target tree species, and object 2 is another tree species.

[0031] Optionally, the optimal management decision includes: for large-diameter timber cultivation management decisions, achieving a uniform or random spatial distribution, with tree species correlation tending to be unrelated or positively correlated.

[0032] Optionally, the optimal management decision also includes: for the management decision of creating uneven-aged multi-layered forests, achieving a reasonable clustered distribution as the main goal.

[0033] Optionally, an initial forest stand database containing spatial information can be constructed using the following methods;

[0034] Within the pre-defined sample plot area, each living tree with a diameter at breast height (DBH) greater than the target diameter is measured, and the tree parameters and geographic coordinates (based on RTK positioning technology, GPS, or total station) are recorded. The geographic coordinates are then converted into relative coordinates with the southwest point of the sample plot as the origin, and the initial forest stand data containing spatial information is constructed accordingly.

[0035] Optionally, the forest parameters include tree species, diameter at breast height (DBH), tree height, height below the branch, crown width, and health status.

[0036] Compared with the prior art, the advantages of this invention are as follows:

[0037] This invention combines spatial structure parameter analysis with point pattern analysis, focusing not only on the cultivation of large-diameter timber for the target tree, but also comprehensively considering the interspecific spatial relationships between the target tree species and associated tree species within the stand, thus providing a scientific basis for optimizing the spatial structure of existing artificial broad-leaved mixed forest stands. Attached Figure Description

[0038] Figure 1 The main flowchart of the management decision-making method for artificial broadleaf mixed forests in South China provided in the embodiments of this application;

[0039] Figure 2 A flowchart illustrating the optimal state for management decisions in artificial broadleaf mixed forests in South China, provided in this application embodiment;

[0040] Figure 3 A complete technical roadmap for the management decision-making method of artificial broadleaf mixed forests in South China provided in the embodiments of this application;

[0041] Figure 4 To optimize the spatial relationship between the target tree species and associated tree species in the large-diameter timber cultivation plots;

[0042] Figure 5 Spatial correlation between target tree species and associated tree species after optimization of large-diameter timber cultivation plots. Detailed Implementation

[0043] Example:

[0044] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0045] See Figure 1 As shown, the management decision-making method for broadleaf mixed forests provided in this embodiment mainly includes the following steps:

[0046] 110. Construct an initial forest stand database for sample plots that includes spatial information.

[0047] Specifically, a research plot of no less than 0.04 hectares (20m×20m) is set up. Within the pre-set plot area, the starting measurement of diameter at breast height (DBH) is 5cm. Data such as tree species, DBH, tree height, branch height, crown width, and health status of each individual tree are collected, along with precise coordinates (accurate to 0.1m) obtained (based on RTK positioning technology, GPS, or total station). The geographic coordinates are converted into relative coordinates with the southwest point of the plot as the origin, and an initial stand database containing spatial information is constructed accordingly.

[0048] 120. Based on the initial forest stand database of the sample plot, calculate the dominance of each tree species in the sample plot and clarify the composition of the target tree species and associated tree species in the sample plot.

[0049] Specifically, the percentage of the basal area at breast height (BHH) of each tree species in the sample plot is calculated, i.e., the relative significance, and the composition of the target tree species and associated tree species in the sample plot is determined according to the degree of dominance.

[0050] 130. Analyze the spatial structure of forest trees and use bivariate point pattern analysis to examine the spatial correlation between the target tree species and its associated tree species.

[0051] 140. Based on the results of forest spatial structure and spatial correlation, combined with the current status of the forest stand and the established management objectives, the optimal management decision scheme for adjusting tree species composition and spatial configuration is proposed through structural optimization simulation.

[0052] Therefore, this method combines spatial structure parameter analysis with point pattern analysis, focusing not only on the cultivation of large-diameter timber for the target trees, but also comprehensively considering the interspecific spatial relationships between the target tree species and associated tree species within the stand, thus providing a scientific basis for optimizing the spatial structure of existing artificial broad-leaved mixed forest stands.

[0053] In one specific embodiment, the calculation of spatial structure parameters for trees in a sample plot includes: for each sample plot, based on the coordinates and diameter at breast height (DBH) data of individual trees, calculating the size ratio, angular scale, and mixing degree of each reference tree, and calculating the average size ratio, average angular scale, and average mixing degree of each tree species. To create a good spatial structure, optimization based on the three spatial structure parameters is performed, including: adjusting based on the analysis results of the average size ratio, average angular scale, and average mixing degree of each tree species; identifying areas with unreasonable spatial structure based on the size ratio, angular scale, and mixing degree of individual trees, and selectively felling individual trees.

[0054] For forest land with good site conditions (such as slope less than 25°, deep and fertile soil), convenient transportation, a high proportion of dominant target tree species (such as precious timber species such as Castanopsis fargesii and Phoebe zhennan) (e.g., >65%), and a relatively simple stand structure (such as the same age and single layer), intensive management measures aimed at cultivating large-diameter timber (diameter at breast height ≥25cm) can be given priority.

[0055] For forest land with sensitive or complex site conditions (such as slope greater than 25°, high risk of soil erosion), good natural regeneration of forest stands, rich tree species composition (more than 10 species), and forest stand structure that has shown or is ready to develop into multi-layered, heterogeneous forests (such as obvious forest gaps and good natural regeneration, or continuous diameter distribution), the management goal should be to protect and create heterogeneous multi-layered forests, and to give full play to their functions of ecological protection and biodiversity maintenance.

[0056] Whether cultivating large-diameter timber or establishing uneven-aged multi-layered forests, structural regulation aims to promote the growth of the target tree species and optimize the stand's spatial structure, achieved through structural selective felling. However, due to their different cultivation objectives, the directions of their structural adjustments differ fundamentally:

[0057] Large-diameter timber cultivation aims to provide optimal independent growth space for the target individual tree. Therefore, its structural adjustment direction is to make the target tree species tend to be evenly or randomly distributed to avoid excessive competition.

[0058] The creation of multi-layered forests of different ages aims to simulate the community structure of natural forests. Therefore, the direction of structural adjustment is to form a reasonable clustered distribution with the dominant species as the core, so as to promote natural regeneration and ecosystem stability.

[0059] To achieve the aforementioned regulatory objectives, it is necessary to first identify and diagnose high-intensity clustering units within the forest stand and then implement targeted measures accordingly. Therefore, bivariate point pattern analysis is used to examine the spatial association between the target tree species and associated tree species, including:

[0060] High-intensity clustering unit identification: Based on the angular scale parameter, forest units with angular scale values ​​greater than a threshold are identified as high-intensity clustering units.

[0061] Internal status diagnosis of each unit: For each high-intensity clustered unit, the average degree of mixing and the size ratio of trees within the unit are calculated, and the degree of clustering is adjusted, as shown in Table 2:

[0062] Table 2. Rationality Judgment of Optimized Aggregation Degree in Uneven-aged Multi-layered Forests

[0063] Unreasonable gatherings Reasonable gathering Judgment criteria The mixed degree is 0 (pure forest) and the size ratio is close to 1 (all are compressed timber). The structure is rigid and competition is fierce. The degree of mixing is higher than 0.75, with a dominant tree (size ratio ≤ 0.25) as the core. The diverse structure is conducive to development. Regulatory measures For unreasonable clusters, thinning and transformation should be carried out, prioritizing the removal of diseased and weak trees, suppressed trees, and non-target tree species to break the ineffective competition pattern; if the entire cluster consists of non-target tree species, intensive thinning can be carried out and precious native tree species can be replanted. Protective and promotional measures should be implemented for reasonable cluster units, resolutely preserving the mother trees and dominant trees, and only lightly thinning out the surrounding overly dense companion trees, in order to release the growth space and light resources of the target trees.

[0064] Point pattern analysis and iterative verification: After adjusting the clustering level, bivariate point pattern analysis was used to examine the spatial correlation between the target tree species and its associated tree species, specifically including:

[0065] Monte Carlo simulations generate 95% confidence intervals. Within the target spatial scale, points on the observed curve above the upper envelope indicate a positive correlation, points within the confidence interval indicate no significant correlation, and points below the lower envelope indicate a negative correlation. If a negative correlation still exists, adjustments are made, and another round of adjustments is conducted based on the new spatial structure parameters until the spatial correlation between the target tree species and its companion tree species reaches a statistically non-negative correlation. This achieves synergistic optimization of tree species configuration at both the macro (spatial correlation) and micro (structural unit) levels, until the stand structure reaches its optimal state. Figure 2-3 As shown, for the management decision of large-diameter timber cultivation, the goal is to achieve a uniform or random spatial distribution, with tree species having little or no correlation; for the management decision of creating uneven-aged multi-layered forests, the goal is to achieve a reasonable clustered distribution.

[0066] Finally, the sample plots were implemented according to the optimal solution that achieved the best state of business decision-making.

[0067] Specifically, the calculation method for the point pattern analysis is as follows:

[0068]

[0069]

[0070]

[0071]

[0072] In the formula: Let be the spatial association value between object 1 and object 2 at point i, with radius r and annular width w; n1 is the number of points of object 1 in the bivariate statistics. It is a ring in object 1 with point i as the center, r as the radius, and w as the width; Point s2 (X) calculates the number of points of object 2 within region X; Area(X) is the area of ​​region X; (x i , y i Let (x, y) be the coordinates of point i in object 1; S(x, y) is a binary variable. If the coordinates (x, y) are within the study area X, then S(x, y) = 1, otherwise S(x, y) = 0; z 2 This indicates the area of ​​a single cell. It is a variable that changes with the circle in object 1 centered at point i and with radius r; This indicates the number of points distributed within the cell for object 2; object 1 is the target tree species group, and object 2 is other tree species.

[0073] This method is applicable to middle-aged artificial broad-leaved mixed forests in South China. The age group is determined based on the average age of the dominant tree species (group) in the sample plot.

[0074] Each tree is taken as the object tree, and its four nearest neighboring trees are found to form a basic structural unit for spatial structure analysis.

[0075] Table 3. Tree species selection for artificially promoting natural regeneration in mixed broadleaf forests in a certain region.

[0076] Tree species Examples of recommended tree species Precious and major group-forming tree species Red-flowered privet, fragrant magnolia, camphor, red cone, iron holly, fragrant rosewood, cinnamon, bamboo tree, Zhejiang nanmu, persimmon, gray magnolia, pointed-leaved euonymus, etc. Shade-tolerant and understory tree species Hainan red bean, *Castanopsis fargesii* and other Fagaceae species, *Spathiphyllum*, *Ilex chinensis*, etc.

[0077] The following examples of application scenarios will further verify and illustrate this method:

[0078] Application Example 1: Cultivation of Large-Diameter Timber

[0079] 1. Sample plots and data

[0080] Location: A forest farm in South China

[0081] Sample plot: a fixed sample plot of 50m × 50m

[0082] Data: A total of 403 trees with a diameter at breast height (DBH) greater than 5 cm were surveyed, including species such as Magnolia officinalis, Michelia champaca, Schima superba, Evodia rutaecarpa, Zanthoxylum bungeanum, and Schefflera heptaphylla.

[0083] Table 4 Initial data of quadrats

[0084] Serial Number species Breast diameter / cm x-coordinate / m y coordinate / m 1 Manglietia glauca 18.5 48.31 2.09 2 Manglietia glauca 16.5 47.29 3.14 3 Michelia macclurei, the intoxicatingly fragrant smile 17.9 45.09 0.74 4 Michelia macclurei, the intoxicatingly fragrant smile 26.1 44.15 1.62 5 Michelia macclurei, the intoxicatingly fragrant smile 10.5 42.59 3.09 6 Michelia macclurei, the intoxicatingly fragrant smile 10.5 44.15 4.92 7 Michelia macclurei, the intoxicatingly fragrant smile 10.2 45.31 4.51 8 Michelia macclurei, the intoxicatingly fragrant smile 15.1 49.74 4.02 9 Michelia macclurei, the intoxicatingly fragrant smile 30.3 48.72 5.17 10 Michelia macclurei, the intoxicatingly fragrant smile 20.9 34.64 4.33

[0085] 2. Business Objectives

[0086] Through structured forest management, cultivate high-quality, efficient, and structurally stable large-diameter timber in broadleaf mixed forests.

[0087] 3. Initial Analysis Results

[0088] Given that the combined relative significance of *Magnolia denudata* and *Michelia champaca* accounted for 89.7% of the total sample area, they were identified as the dominant tree species in this sample area. With the management goal of cultivating high-quality broad-leaved large-diameter timber, *Magnolia denudata* and *Michelia champaca* were defined as target tree species, while all other tree species were designated as companion species.

[0089] Initial spatial structure:

[0090] Table 5 Spatial parameters of the target tree species in the large-diameter timber cultivation plots before optimization.

[0091] species Mean angular scale (W) Average size ratio (U) Average degree of mixing (M) Grey Magnolia 0.531 0.373 0.398 Drunken Fragrance Smiling 0.501 0.461 0.361

[0092] Priority strategy:

[0093] Magnolia ash: The primary task is to reduce the angular scale and increase the degree of interbreeding while loosening the clusters.

[0094] Drunken Fragrance and Smiling: Remove inferior trees within the cluster to increase the degree of mixed planting.

[0095] Point pattern analysis:

[0096] like Figure 4 As shown, in the bivariate O-ring point pattern analysis, all target tree species (Magnolia grahami and Michelia champaca) in the sample plot were marked as type 1 point events, and all associated tree species were marked as type 0 point events, in order to examine the spatial correlation between the two before and after the management intervention.

[0097] The target tree species and associated tree species show a negative correlation in distribution at a small scale (r<2m). The target tree species and associated tree species also exhibit competitive behavior at a small scale. The target tree species and associated tree species rarely appear in close proximity; therefore, when creating artificial mixed forests, tree species with strong competitive relationships should not be simply planted closely together at a small scale.

[0098] Post-logging spatial structure:

[0099] Table 6 Spatial parameters of the target tree species in the large-diameter timber cultivation plots before optimization

[0100] species Mean angular scale (W) Average size ratio (U) Average degree of mixing (M) Grey Magnolia 0.516 0.397 0.402 Drunken Fragrance Smiling 0.517 0.481 0.373

[0101] Point pattern analysis:

[0102] like Figure 5 As shown, the target tree species and associated tree species are unrelated. By selectively felling some trees, the shift from "negative association" to "no association" alleviates the intense direct competition between the target tree species and associated tree species. This transforms the forest from a highly competitive and simple structure to one with less competition, more accessible resources, and greater development potential, creating crucial conditions for the forest structure to develop in a more stable and complex direction.

[0103] Application Example 2: Creating Multi-Aged Forests

[0104] Location: A forest farm in South China

[0105] Sample plot: a fixed sample plot of 50m × 50m

[0106] Data: A total of 377 trees with a diameter at breast height (DBH) greater than 5 cm were surveyed. The tree species included Michelia champaca, Michelia alba, Cunninghamia lanceolata, Cunninghamia lanceolata, Ziziphus jujuba, Schima superba, Pinus massoniana, Cinnamomum camphora, Elaeocarpus longicornis, Castanopsis fargesii, Litsea cubeba, Schefflera heptaphylla, etc.

[0107] Due to the diverse tree species composition and vertical layers, uneven-aged multi-layered forests have more complex intra-forest competition relationships, highly heterogeneous stand structure, and generally clustered tree distribution patterns. The effect of homogenization management is limited. It is recommended to start by optimizing spatial structural units and creating reasonably clustered small units to make the stand spatial structure more stable.

[0108] Native tree species with valuable timber, excellent trunk shape, and outstanding ecological functions, as well as rare and endangered species, were selected as target trees, such as Michelia champaca, Caryophyllum indicum, Schima superba, Castanopsis fargesii, and Elaeocarpus longicornis. The target trees in the sample plots formed the main forest layer in the multi-layered forest.

[0109] The target trees are adjusted based on size ratio, angular scale, and mixing degree. Using a single target tree species as the core, the spatial structural units are adjusted to achieve a reasonable clustering.

[0110] Taking the two tree species with the highest importance values ​​in the main canopy as examples:

[0111] Table 7. Spatial parameters of tree species for the purpose of creating uneven-aged multi-layered forests in sample plots.

[0112] species Mean angular scale (W) Average size ratio (U) Average degree of mixing (M) Drunken Fragrance Smiling 0.52 0.45 0.49 Shell Vegetable Fruit 0.59 0.43 0.11

[0113] The purpose of this application example is to cultivate uneven-aged multi-layered forests. Therefore, it is necessary to adjust the stand structure to develop the sample plots into multi-layered forests dominated by the target tree species.

[0114] Optimization strategy for the top two tree species in importance within the sample plot:

[0115] Michelia champaca: In areas where Michelia champaca is abundant, the dominant tree with a size ratio of less than 0.25 is used as the target tree for spatial structure. The competing trees around the dominant tree are lightly thinned. This creates a structure with the dominant Michelia champaca as the core, surrounded by various other tree species.

[0116] Shellfish: The average mixing degree of shellfish is 0.11, indicating that the aggregation of shellfish is serious. Identify the spatial structural units dominated by shellfish with an angular scale greater than 0.5, remove the crushed trees, dying trees, and poorly growing shellfish (size ratio of 1), create small forest gaps, and replant shade-tolerant native tree seedlings suitable for planting in South China, such as Phoebe bournei and Castanopsis fargesii.

[0117] For large forest gaps, forest land clearing combined with replanting of shade-tolerant native tree seedlings can be used to artificially promote forest regeneration.

[0118] Point pattern analysis confirms that the spatial relationship between the target tree species and its companion tree species is mainly unrelated or negatively related (competitive), and it is difficult to form a wide range of positive relationships. In view of this relationship characteristic, when managing uneven-aged multi-layered broad-leaved mixed forests with a main canopy layer of middle age, the relationship between tree species can be temporarily disregarded as a core consideration.

[0119] The above embodiments are merely illustrative of the technical concept and features of the present invention, and are intended to enable those skilled in the art to understand the content of the present invention and implement it accordingly. They should not be construed as limiting the scope of protection of the present invention. All equivalent changes or modifications made based on the essence of the content of the present invention should be covered within the scope of protection of the present invention.

Claims

1. A management decision-making method for artificial broadleaf mixed forests in South China, characterized in that, include: Construct a database of initial forest stands for sample plots containing spatial information; The dominance of each tree species is calculated based on the initial stand database of the sample plots to clarify the composition of the target tree species and associated tree species within the sample plots; The spatial structure of forest trees was analyzed, and bivariate point pattern analysis was used to examine the spatial correlation between the target tree species and its associated tree species. Based on the results of forest stand spatial structure and spatial correlation, combined with the current status of the forest stand and the established management objectives, the optimal management decision scheme for adjusting tree species composition and spatial configuration is proposed through structural optimization simulation.

2. The method as described in claim 1, characterized in that, The analysis of the spatial structure of forest trees includes: For each sample plot, based on the coordinates and diameter at breast height (DBH) data of individual trees, the size ratio, angular scale, and degree of mixing of each reference tree are calculated; and the average size ratio, average angular scale, and average degree of mixing of each tree species are calculated.

3. The method as described in claim 1, characterized in that, The structural optimization simulation includes: adjusting based on the analysis results of the average size ratio, average angular scale and average mixing degree of each tree species; identifying areas with unreasonable spatial structure based on the size ratio, angular scale and mixing degree of individual trees; and selectively felling individual trees.

4. The method as described in claim 1, characterized in that, The structural optimization simulation also includes: High-intensity clustering unit identification: Based on the angular scale parameter, forest units with angular scale values ​​greater than a threshold are identified as high-intensity clustering units. Internal status diagnosis of the unit: Calculate the average degree of mixing and the size ratio of trees within the unit for each high-intensity clustered unit, and adjust the degree of clustering; Point pattern analysis and iterative verification: After adjusting the degree of clustering, bivariate point pattern analysis was used to examine the spatial correlation between the target tree species and its associated tree species.

5. The method as described in claim 4, characterized in that, The method of using bivariate point pattern analysis to examine the spatial association between the target tree species and its associated tree species includes: The Monte Carlo simulation generates a 95% confidence interval. Within the target spatial scale, points where the observed curve is above the upper envelope line indicate a positive correlation, points within the confidence interval indicate no significant correlation, and points below the lower envelope line indicate a negative correlation. If a negative correlation still exists, the adjustment continues, and the next round of adjustment is carried out based on the calculation results of the new spatial structure parameters until the spatial correlation between the target tree species and the associated tree species reaches a statistically non-negative correlation.

6. The method as described in claim 4, characterized in that, The calculation method for the point pattern analysis is as follows: In the formula: Let be the spatial association value between object 1 and object 2 at point i, with radius r and annular width w; n1 is the number of points of object 1 in the bivariate statistics. It is a ring in object 1 with point i as the center, r as the radius, and w as the width; Point s2 (X) calculates the number of points of object 2 within region X; Area(X) is the area of ​​region X; (x i , y i Let (x, y) be the coordinates of point i in object 1; S(x, y) is a binary variable. If the coordinates (x, y) are within the study area X, then S(x, y) = 1, otherwise S(x, y) = 0; z 2 This indicates the area of ​​a single cell. It is a variable that changes with the circle in object 1 centered at point i and with radius r; This indicates the number of points distributed within the cell for object 2; object 1 is the target tree species group, and object 2 is other tree species.

7. The method as described in claim 1, characterized in that, The optimal management decision includes: for the management decision of large-diameter timber cultivation, achieving a uniform or random spatial distribution, and the correlation between tree species tending to be unrelated or positively related.

8. The method as described in claim 7, characterized in that, The optimal management decision also includes: for the management decision of creating multi-layered forests of different ages, the goal is to achieve a reasonable clustered distribution.

9. The method as described in claim 1, characterized in that, The initial forest stand database containing spatial information for sample plots is constructed using the following method; Within the pre-defined sample plot area, each living tree with a diameter at breast height (DBH) greater than the target diameter is measured, and the tree parameters and geographic coordinates obtained based on RTK positioning technology, GPS, or total station are recorded. The geographic coordinates are then converted into relative coordinates with the southwest point of the sample plot as the origin, and the initial forest stand data containing spatial information is constructed accordingly.

10. The method as described in claim 9, characterized in that, Therefore, forest parameters include tree species, diameter at breast height (DBH), tree height, height below the branch, crown width, and health status.

Images