High-cold arid slope ecological restoration adaptation method based on two-stage fuzzy matter element decision

Abstract

The present application relates to the technical field of slope or slope stability, disclose a kind of high-cold arid slope ecological restoration adaptation method based on two-stage fuzzy matter-element decision-making, in the present application, by the environmental factor of all influence vegetation growth is classified as matrix factor and non-matrix factor, and based on the coupling relationship of matrix factor and non-matrix factor, the variation range of matrix factor is limited;The various different plant communities of the marker shrub are regarded as the same kind of plant community, so that the relationship of different plants in plant community is decoupled;Then first fuzzy matter-element decision-making is carried out based on non-matrix factor, selects the marker shrub most suitable for local climate and topography, and the associated herbaceous plant can constitute the adapted plant community;After the adapted plant community is selected, under the premise that the coupling relationship of matrix factor and non-matrix factor has been integrated into the original data matrix, fuzzy matter-element decision-making can be carried out based on matrix factor, and the matrix suitable for plant community, local climate and topography is selected.

Description

Technical Field

[0001] This invention relates to the field of slope or sloping slope stability technology, and in particular to an adaptation method for ecological restoration of high-altitude and arid slopes based on two-level fuzzy matter-element decision-making. Background Technology

[0002] Slope ecosystems in high-altitude, cold, and arid regions exhibit significant uniqueness and fragility, characterized by low temperatures, strong radiation, water scarcity, thin and infertile soil, and nutrient depletion, resulting in extremely poor natural vegetation recovery capabilities. This is particularly true for steep rock slopes formed during infrastructure construction projects, which lack stable topsoil and the basic substrate conditions necessary for vegetation growth. Traditional vegetation restoration methods generally suffer from low survival rates, poor community stability, and difficulty in long-term maintenance.

[0003] Studies in the field of environmental quality assessment have shown that evaluation methods relying solely on physicochemical indicators often fail to fully reflect the actual impact of environmental conditions on organisms, especially in environments with compound pollution or adverse conditions. Similarly, in the ecological restoration of high-altitude, arid slopes, focusing only on the nutrient or structural indicators of the substrate itself while neglecting the physiological and ecological responses of vegetation makes it difficult to truly reflect the synergistic suitability between the substrate and vegetation. While existing slope ecological restoration techniques can temporarily improve site conditions, the lack of scientific evaluation of the vegetation-substrate relationship often leads to problems such as poor plant growth, population degradation, and unstable system functions. More critically, high-altitude environments are characterized by harsh features such as low temperatures, strong radiation, drought, and poor soil, posing multiple stresses on vegetation survival and growth. Vegetation that can grow under multiple stresses often possesses a series of specialized traits evolved to adapt to the environment, limiting its adaptability to a relatively narrow range of environments. However, existing vegetation selection techniques often rely on single climatic conditions and lack systematic research on the physiological and ecological characteristics of high-altitude species, resulting in significant blind selection of species.

[0004] These factors are inherent to the characteristics of high-altitude, arid slopes. For flatlands, climate, substrate (i.e., soil, gravel, and other materials that allow plants to take root), and topography are generally consistent within the same region, with changes typically occurring over distances of thousands of miles. However, high-altitude, arid slopes differ significantly from flatlands. Their climate and topography vary greatly with spatial distribution, leading to substantial changes in the substrate. The vegetation adapted to the local microenvironment also undergoes significant changes due to long-term co-evolution. A readily understandable effect is that flatland vegetation can often find a suitable substrate (such as fertile black soil) that allows almost all types of flatland vegetation to thrive. However, for alpine vegetation, a series of specialized characteristics result in almost every type of high-altitude, arid slope vegetation having its most suitable substrate, climate, and topography.

[0005] When restoring vegetation on slopes, it is often necessary to simultaneously restore the local substrate (because if the substrate has not been significantly damaged, vegetation restoration can usually be achieved through methods such as forest closure and reforestation without human intervention), especially on high-altitude, arid slopes (where the natural re-formation of the substrate is almost impossible). Existing substrate and vegetation selection methods are rather crude, generally following the approach of selecting local native species and relatively fertile substrates. However, for high-altitude, arid slopes, the selected vegetation and substrate are generally not suitable. There is no universally compatible substrate for vegetation that can grow on high-altitude, arid slopes, and even native species require a specific number of species within their corresponding microenvironments to be suitable.

[0006] In research on vegetation restoration, the fuzzy comprehensive evaluation matrix is ​​a powerful tool. However, due to the strong coupling relationship between vegetation, climate, topography, and substrate on high-altitude, cold, and arid slopes, this tool cannot be used effectively. For example, various parameters in the substrate are closely related to the climate and cannot change independently; for instance, the available potassium content in the soil on high-altitude, cold, and arid slopes is constrained by rainfall.

[0007] However, the inventors discovered that there are also some favorable conditions for vegetation restoration research on high-altitude and arid slopes. Because the vegetation is relatively sparse, the interaction between different plants is weak. When conducting research, the interaction between different plants is easy to decouple, and the evaluation of growth status is relatively simple. Summary of the Invention

[0008] This invention provides an adaptation method for ecological restoration of arid and cold slopes based on two-level fuzzy matter-element decision-making.

[0009] The technical problem to be solved is that there are many stress factors on high-altitude, cold and arid slopes, which leads to a series of specialized traits in vegetation. It often can only adapt to specific substrates and microenvironments, making vegetation restoration difficult.

[0010] To solve the above-mentioned technical problems, the present invention adopts the following technical solution: a two-level fuzzy matter-element decision-making method for ecological restoration of arid and cold slopes, used to determine whether the substrate and vegetation used for vegetation restoration are compatible with the slope when vegetation restoration is carried out on arid and cold slopes where both vegetation and substrate have been destroyed. The slopes that need vegetation restoration are recorded as slopes to be treated. The adaptation method includes the following steps: Step 1: Collect and classify all environmental factors that affect vegetation growth. Environmental factors that can be changed by changing the substrate are classified as substrate factors, and a feature set is constructed, denoted as the substrate factor set; the remaining environmental factors are classified as non-substrate factors, and a feature set is constructed, denoted as the non-substrate factor set. Step 2: Collect data on representative plant communities on high-altitude, cold, and arid slopes, and decouple the relationships between different plants within these communities; Representative matrices on high-altitude, cold, and arid slopes were statistically analyzed, and the range of variation of matric factors was restricted based on the coupling relationship between matric factors and non-matric factors. Step 3: For each representative plant community, with the set of non-matrix factors as the independent variable and vegetation cover as the dependent variable, calculate the entropy weight of each non-matrix factor. Step 4: Based on the non-matrix factor set of the slope to be treated, calculate the proximity of the slope to be treated under different representative plant communities. The higher the proximity, the more suitable the representative plant community is for the slope to be treated. The representative plant community with the highest proximity is recorded as the most suitable community. Step 5: Under the premise of selecting the most suitable community for vegetation restoration on the slope to be treated, the entropy weight of each matrix factor is calculated with the matrix factor set as the independent variable and the vegetation cover as the dependent variable. Step 6: Based on the matrix factor set of each representative matrix, calculate the closeness of each representative matrix. The higher the closeness, the more suitable the representative matrix is ​​for the slope to be treated. The representative matrix with the highest closeness is recorded as the most suitable matrix. Step 7: Restore vegetation based on the optimal community and optimal substrate.

[0011] Furthermore, in steps three and five, the entropy weight is calculated using the composite fuzzy matter-entropy weight method; in steps four and six, the obtained fit is the Euclidean fit.

[0012] Furthermore, in steps three and five, the specific steps are as follows: constructing the original data matrix, standardizing the data and constructing membership degree fuzzy matter elements, and determining the weights of environmental factors. In step three, an original data matrix is ​​constructed based on multiple slope quadrat data with different non-matrix factors but consistent matrix obtained from literature and actual surveys. The original data matrix includes the non-matrix factor set and the corresponding vegetation cover of each slope quadrat. In step five, an original data matrix is ​​constructed based on multiple slope quadrat data with different matrices but consistent non-matric factors obtained from literature and actual surveys. The original data matrix includes the matrix factor set of each matrix and the corresponding vegetation cover.

[0013] Furthermore, in step two, the following method is used to decouple the relationships between different plants in the plant community: Investigate typical plant communities on high-altitude, cold, and arid slopes. Extract one dominant shrub from each typical plant community as a representative of the plant community and designate it as a marker shrub. If the marker shrub is a shade-loving plant, the slope quadrat should be selected on the shady slope; if the marker shrub is a sun-loving plant, the slope quadrat should be selected on the sunny slope. In the literature and in actual slope quadrats, different plant communities with marker shrubs as the dominant shrubs are considered as the same plant community. Different quadrats considered as the same plant community are selected as evaluation objects to calculate the entropy weights of each non-matrix factor.

[0014] Furthermore, in step two, the following method is used to limit the range of variation of the matrix factor based on the coupling relationship between the matrix factor and the non-matrix factor: In steps two, three, and five, when selecting slope quadrats, only slope quadrats with native matrix should be selected, and no human intervention, including fertilization and irrigation, is allowed during the data collection process.

[0015] Furthermore, during the data standardization process, environmental factors are divided into three categories: positive indicators, negative indicators, and variable indicators, and are processed separately. Variable indicators are environmental factors that are neither better the larger they are nor better the smaller they are. Variable indicators can be obtained from literature and / or experiments to find the most favorable value for improving vegetation cover, and are denoted as the most favorable value. As a directional indicator, environmental factors are considered positive indicators when they are below the optimal value, and negative indicators when they are above the optimal value.

[0016] Furthermore, the non-matrix factor set includes average annual temperature, average annual rainfall, altitude, and slope; the matrix factor set includes sand content, soil organic matter content, total nitrogen content, available phosphorus, available potassium, and pH value. Among them, average annual temperature, average annual rainfall, altitude, sand content, and pH value are inverse indicators, soil organic matter content, total nitrogen, available phosphorus, and available potassium are positive indicators, and slope is an inverse indicator.

[0017] Compared with existing technologies, the ecological restoration adaptation method for high-altitude and arid slopes based on two-level fuzzy matter-element decision-making in this invention has the following advantages: In this invention, all environmental factors affecting vegetation growth are classified into matrix factors and non-matrix factors, and the range of variation of matrix factors is limited based on the coupling relationship between matrix factors and non-matrix factors. Different plant communities with dominant shrubs as marker shrubs are regarded as the same plant community (the relationship between different plants in vegetation on high-altitude arid slopes is weak, and as long as the marker shrub with high growth requirements can grow normally, the growth of the accompanying herbaceous plants will not be a problem), thereby decoupling the relationship between different plants in the plant community. Then, based on non-matrix factors, fuzzy matter-element decision-making is performed (if the relationships between different plants in the community are not decoupled, this step cannot even construct the original data matrix). The marker shrubs that are most adapted to the local climate and terrain are selected, and paired with the accompanying herbaceous plants, a suitable plant community can be formed. After the suitable plant community is selected, under the premise that the coupling relationship between matrix factors and non-matrix factors has been integrated into the original data matrix, fuzzy matter-element decision-making can be carried out based on matrix factors to select a matrix that is compatible with the previously selected plant community as well as the local climate and topography. Moreover, the selected matrix will not have a situation that is theoretically feasible but practically impossible due to the coupling relationship. Attached Figure Description

[0018] Figure 1 This is a flowchart of the ecological restoration adaptation method for high-altitude and arid slopes based on two-level fuzzy matter-element decision-making, as described in this invention. Detailed Implementation Note that in the "high-altitude cold and arid slope" of this invention, "arid" does not mean that the precipitation on the slope is absolutely insufficient. The "arid" here is largely due to the large slope and poor water retention capacity of the substrate, which prevents the effective retention of precipitation and causes the vegetation to be under drought stress.

[0019] Taking the integrated slope protection and ecological restoration technology project for the disturbance area of ​​major hydropower projects in the upper reaches of the rivers in Southwest China (project code 2024YFF1307803), which is proposed to adopt the technical solution of this invention, as an example, Figure 1 As shown, a two-level fuzzy matter-element decision-making-based ecological restoration adaptation method for arid and cold slopes is used to determine whether the substrate and vegetation used for vegetation restoration are compatible with the slope when vegetation restoration is carried out on arid and cold slopes where both vegetation and substrate have been destroyed. Slopes requiring vegetation restoration are denoted as slopes to be treated. The adaptation method includes the following steps: Step 1: Collect and classify all environmental factors that affect vegetation growth. Environmental factors that can be changed by changing the substrate are classified as substrate factors (i.e., various parameters of the soil) and a feature set is constructed, denoted as the substrate factor set. The remaining environmental factors are classified as non-substrate factors and a feature set is constructed, denoted as the non-substrate factor set (i.e., parameters related to climate and topography). This step forms the basis for subsequent two-level fuzzy matter-element decision-making. Selecting suitable vegetation and substrate through single-level fuzzy matter-element decision-making is unrealistic because many substrate factors and non-matrix factors are strongly coupled. If single-level fuzzy elements are used for decision-making, these coupling relationships are not considered, and the so-called optimal choice may be completely unattainable in reality. Here, two sets are constructed where there is no obvious coupling between substrate factors and no obvious coupling between non-matrix factors, allowing for two-level fuzzy matter-element decision-making processes.

[0020] Step 2: Collect data on representative plant communities on high-altitude, cold, and arid slopes, and decouple the relationships between different plants within these communities; This decoupling step is to construct the original data matrix required for the first-level fuzzy matter-element decision-making. Completely identical plant communities in different slope quadrats can only be achieved under artificial cultivation. However, in the second-level decision-making process of this invention, to ensure that the coupling relationship is integrated into the original data matrix required for the second-level fuzzy matter-element decision-making, slope quadrat data with any trace of artificial interference cannot be used. If completely identical plant communities are required to construct the original data matrix for the first-level fuzzy matter-element decision-making, the necessary data cannot be collected.

[0021] The representative plant communities selected in this embodiment are as follows: Table 1: Representative plant communities Representative matrices on high-altitude, cold, and arid slopes were statistically analyzed, and the range of variation of matric factors was restricted based on the coupling relationship between matric factors and non-matric factors. On high-altitude, arid slopes, many climatic and topographical conditions can affect certain parameters in the matrix. For example, excessive rainfall can limit the available potassium content in the matrix. Therefore, when making second-level fuzzy matter-element decisions, this coupling relationship must be taken into account to avoid selecting what appears to be the optimal option, which is actually unattainable.

[0022] Step 3: For each representative plant community, with the set of non-matrix factors as the independent variable and vegetation cover as the dependent variable, calculate the entropy weight of each non-matrix factor. Step 4: Based on the non-matrix factor set of the slope to be treated, calculate the proximity of the slope to be treated under different representative plant communities. The higher the proximity, the more suitable the representative plant community is for the slope to be treated. The representative plant community with the highest proximity is recorded as the most suitable community. This step can be used not only for decision-making in a project where vegetation restoration has not yet been carried out, but also for evaluating the rationality of decisions made in projects where decisions have already been made.

[0023] Step 5: Under the premise of selecting the most suitable community for vegetation restoration on the slope to be treated, the entropy weight of each matrix factor is calculated with the matrix factor set as the independent variable and the vegetation cover as the dependent variable. Step 6: Based on the matrix factor set of each representative matrix, calculate the closeness of each representative matrix. The higher the closeness, the more suitable the representative matrix is ​​for the slope to be treated. The representative matrix with the highest closeness is recorded as the most suitable matrix. Similarly, this step can also evaluate whether the decisions made in projects have been made are reasonable.

[0024] Step 7: Restore vegetation based on the optimal community and optimal substrate.

[0025] In steps three and five, the entropy weight is calculated using the composite fuzzy matter-entropy weight method; in steps four and six, the obtained fit is the Euclidean fit.

[0026] Steps three and five involve constructing the original data matrix, standardizing the data and constructing membership fuzzy matter-element pairs, and determining the weights of environmental factors. The specific implementation process of this method is as follows: 1. Original data matrix The main workload in this step involves investigating the database and creating on-site slope quadrats.

[0027] In this embodiment, 10 environmental factors are used as independent variables, and vegetation cover is used as a dependent variable. The non-matrix factor set includes average annual temperature, average annual rainfall, altitude, and slope; the matrix factor set includes sand content, soil organic matter content, total nitrogen, available phosphorus, available potassium, and pH value.

[0028] Assuming that each level of fuzzy matter-element decision-making uses 10 slope quadrats, the original data matrix for each level must be constructed along with the vegetation cover and independent variables. That is, the original data matrix used for the first level of fuzzy matter-element decision-making is a 10x5 matrix (10 rows because of the 10 slope quadrats; 5 columns because of the 4 independent variables and 1 dependent variable, totaling 5). The original data matrix used for the second level of fuzzy matter-element decision-making is a 10x7 matrix.

[0029] 2. Composite fuzzy matter element 2.1 Fuzzy Matter Element and Composite Fuzzy Matter Element In matter-element analysis, a matter element R(M, c, x) consists of the name M of the object, its characteristic c, and its magnitude. (Kang Wenni et al., 2020) If an object M has n characteristics and their corresponding fuzzy values, then these m n-dimensional fuzzy matter elements can constitute a composite fuzzy matter element R. mn (Tang Shuo et al., 2012), the specific formula is as follows: (1) 2.2 Based on the principle of preferential membership, describe the degree of membership of the corresponding fuzzy values ​​of each evaluation indicator.

[0030] Larger is better (positive indicator): (2) Smaller is better (contrarian indicator): (3) In the formula: u ij For the first i The first type j The value corresponding to each indicator; for i Type No. jThe maximum and minimum values ​​among all the values ​​corresponding to each indicator.

[0031] This allows us to construct a fuzzy matter element R' with superior membership. mn (4) 2.3 Standard Fuzzy Matter Element and Sum-of-Difference Composite Fuzzy Matter Element R' mn The standard fuzzy matter element R is established by using the maximum or minimum value of the preferred membership degree of each indicator. 0n If the square of the differences between the indices of the standard fuzzy matter-element and the preferred membership fuzzy matter-element is Δ... ij , that is, Δ ij =(μ 0j -μ ij ) 2 Then, the composite fuzzy matter element R that makes up the difference square is... Δ , can be represented as: (5) 3. Entropy weight method for weighting 2.1 Construct the original judgment matrix R for m evaluation objects and n evaluation indicators. R= (i=1, 2,…m; j=1, 2,…n)(6) 3.2 Normalize the judgment matrix to obtain the normalized judgment matrix A.

[0032] Positive indicators (7) contrarian indicators (8) In the formula, a ij For r ij After normalization, a ij ∈[0,1] 3.3 Calculate the entropy value of the index (9) in (10) 3.4 Calculate the entropy weight W of the index (11) In the formula, w j ≥0 3.5 European Style Proximity and Overall Evaluation Proximity ρ Hi This indicates how closely a sample resembles a standard sample; a larger value indicates that the sample is closer to the optimal value. (12) By cleaning, handling missing values, and standardizing the data, and then calculating the weights of each indicator using the standardized data, the information entropy and entropy weight are further calculated to finally obtain the weights of each indicator. Table 2: Weight Summary Table Weights can reflect the degree of influence of each indicator on the evaluation process. The larger the weight, the greater the impact of its fluctuation on the result.

[0033] In step three, an original data matrix is ​​constructed based on multiple slope quadrat data with different non-matrix factors but consistent matrix obtained from literature and actual surveys. The original data matrix includes the non-matrix factor set and the corresponding vegetation cover of each slope quadrat. The term "consistent matrix" here refers to a matrix that is of the same type, such as gravelly, rocky, sandy, or limestone. Naturally formed matrices, if of the same type, will also have similar properties.

[0034] In step five, an original data matrix is ​​constructed based on multiple slope quadrat data with different matrices but consistent non-matric factors obtained from literature and actual surveys. The original data matrix includes the matrix factor set of each matrix and the corresponding vegetation cover.

[0035] In step two, the following method is used to decouple the relationships between different plants in the plant community: Investigate typical plant communities on high-altitude, cold, and arid slopes. Extract one dominant shrub from each typical plant community as a representative of the plant community and designate it as a marker shrub. If the marker shrub is a shade-loving plant, the slope quadrat should be selected on the shady slope; if the marker shrub is a sun-loving plant, the slope quadrat should be selected on the sunny slope. In the literature and in actual slope quadrats, different plant communities with marker shrubs as the dominant shrubs are considered as the same plant community. Different quadrats considered as the same plant community are selected as evaluation objects to calculate the entropy weights of each non-matrix factor.

[0036] When selecting the most suitable community, various plant communities with the indicator shrub as the dominant shrub are also considered as the same plant community. The "most suitable community" with the highest degree of similarity can be a variety of different plant communities. In practical applications, the community can be selected according to the current material conditions.

[0037] Naturally formed vegetation, unlike artificially cultivated vegetation, cannot have completely identical species types within the community, making it impossible to construct the original data matrix required for Level 1 fuzzy matter-element decision-making. However, this invention cannot use artificially intervened slope quadrat data. Therefore, it utilizes the weak interrelationships among different plants in the vegetation of arid alpine slopes. As long as the marker shrubs with high growth requirements can grow normally, the growth of the accompanying herbaceous plants will not be a problem. The marker shrubs are used to represent the plant community; if the marker shrubs are identical, then it is considered to be the same community.

[0038] In step two, the following method is used to limit the range of variation of the matrix factor based on the coupling relationship between the matrix factor and the non-matrix factor: In steps two, three, and five, when selecting slope quadrats, only slope quadrats with native matrix should be selected, and no human intervention, including fertilization and irrigation, is allowed during the data collection process.

[0039] Ideally, a constraint function should be constructed using the coupling relationship between the non-matrix factor set and the matrix factor set. This constraint function would then be used to limit the data range for the second-level fuzzy matter-element decision-making. However, this approach is very difficult. Therefore, a workaround is used here: the original data matrix required for the second-level fuzzy matter-element decision-making is directly constructed from data that already contains coupling relationships. In this way, the constraint inherently exists in the original data, eliminating the need to specifically fit a constraint function. Of course, given the available resources, a constraint function still has a wider range of applicability.

[0040] During the data standardization process, environmental factors are divided into three categories: positive indicators, negative indicators, and variable indicators, and are processed separately. Variable indicators are environmental factors that are neither better the larger they are nor better the smaller they are. Variable indicators can be obtained from literature and / or experiments to find the most favorable value for improving vegetation cover, and are denoted as the most favorable value. As a directional indicator, environmental factors are considered positive indicators when they are below the optimal value, and negative indicators when they are above the optimal value.

[0041] This illustrates some other characteristics of high-altitude, cold, and arid slopes: many environmental factors that are more favorable to higher elevations on flat land are not always favorable to high-altitude, cold, and arid slopes.

[0042] Among them, average annual temperature, average annual rainfall, altitude, sand content, and pH value are inverse indicators, soil organic matter content, total nitrogen, available phosphorus, and available potassium are positive indicators, and slope is an inverse indicator.

[0043] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Various modifications and improvements made by those skilled in the art to the technical solutions of the present invention without departing from the spirit of the present invention should fall within the protection scope defined by the claims of the present invention.

Claims

1. A two-level fuzzy matter-element decision-making method for ecological restoration of arid and cold slopes, used to determine whether the substrate and vegetation used for vegetation restoration are compatible with the slope when restoring vegetation on arid and cold slopes where both vegetation and substrate have been destroyed. Slopes requiring vegetation restoration are designated as slopes to be treated. The method is characterized by: The adaptation method includes the following steps: Step 1: Collect and classify all environmental factors that affect vegetation growth. Environmental factors that can be changed by changing the substrate are classified as substrate factors, and a feature set is constructed, denoted as the substrate factor set; the remaining environmental factors are classified as non-substrate factors, and a feature set is constructed, denoted as the non-substrate factor set. Step 2: Collect data on representative plant communities on high-altitude, cold, and arid slopes, and decouple the relationships between different plants within these communities; Representative matrices on high-altitude, cold, and arid slopes were statistically analyzed, and the range of variation of matric factors was restricted based on the coupling relationship between matric factors and non-matric factors. Step 3: For each representative plant community, with the set of non-matrix factors as the independent variable and vegetation cover as the dependent variable, calculate the entropy weight of each non-matrix factor. Step 4: Based on the non-matrix factor set of the slope to be treated, calculate the proximity of the slope to be treated under different representative plant communities. The higher the proximity, the more suitable the representative plant community is for the slope to be treated. The representative plant community with the highest proximity is recorded as the most suitable community. Step 5: Under the premise of selecting the most suitable community for vegetation restoration on the slope to be treated, the entropy weight of each matrix factor is calculated with the matrix factor set as the independent variable and the vegetation cover as the dependent variable. Step 6: Based on the matrix factor set of each representative matrix, calculate the closeness of each representative matrix. The higher the closeness, the more suitable the representative matrix is ​​for the slope to be treated. The representative matrix with the highest closeness is recorded as the most suitable matrix. Step 7: Vegetation restoration based on optimal community and optimal substrate; In steps three and five, the entropy weight is calculated using the composite fuzzy matter-entropy weight method; in steps four and six, the obtained fit is the Euclidean fit. In steps three and five, the specific steps are as follows: constructing the original data matrix, standardizing the data and constructing membership degree fuzzy matter elements, and determining the weights of environmental factors. In step three, an original data matrix is ​​constructed based on multiple slope quadrat data with different non-matrix factors but consistent matrix obtained from literature and actual surveys. The original data matrix includes the non-matrix factor set and the corresponding vegetation cover of each slope quadrat. In step five, an original data matrix is ​​constructed based on multiple slope quadrat data with different matrices but consistent non-matric factors obtained from literature and actual surveys. The original data matrix includes the matrix factor set of each matrix and the corresponding vegetation cover. In step two, the following method is used to limit the range of variation of the matrix factor based on the coupling relationship between the matrix factor and the non-matrix factor: In steps two, three, and five, when selecting slope quadrats, only slope quadrats with native matrix should be selected, and no human intervention, including fertilization and irrigation, is allowed during the data collection process.

2. The method for ecological restoration adaptation of high-altitude and arid slopes based on two-level fuzzy matter-element decision-making as described in claim 1, characterized in that: In step two, the following method is used to decouple the relationships between different plants in the plant community: Investigate typical plant communities on high-altitude, cold, and arid slopes. Extract one dominant shrub from each typical plant community as a representative of the plant community and designate it as a marker shrub. If the marker shrub is a shade-loving plant, the slope quadrat should be selected on the shady slope; if the marker shrub is a sun-loving plant, the slope quadrat should be selected on the sunny slope. In the literature and in actual slope quadrats, different plant communities with marker shrubs as the dominant shrubs are considered as the same plant community. Different quadrats considered as the same plant community are selected as evaluation objects to calculate the entropy weights of each non-matrix factor.

3. The method for ecological restoration and adaptation of high-altitude and arid slopes based on two-level fuzzy matter-element decision-making as described in claim 1, characterized in that: During the data standardization process, environmental factors are divided into three categories: positive indicators, negative indicators, and variable indicators, and are processed separately. Variable indicators are environmental factors that are neither better the larger they are nor better the smaller they are. Variable indicators can be obtained from literature and / or experiments to find the most favorable value for improving vegetation cover, and are denoted as the most favorable value. As a directional indicator, environmental factors are considered positive indicators when they are below the optimal value, and negative indicators when they are above the optimal value.

4. The method for ecological restoration and adaptation of high-altitude and arid slopes based on two-level fuzzy matter-element decision-making as described in claim 3, characterized in that: The non-matrix factor set includes average annual temperature, average annual rainfall, altitude, and slope; the matrix factor set includes sand content, soil organic matter content, total nitrogen content, available phosphorus, available potassium, and pH value. Among them, average annual temperature, average annual rainfall, altitude, sand content, and pH value are inverse indicators, soil organic matter content, total nitrogen, available phosphorus, and available potassium are positive indicators, and slope is an inverse indicator.

Images