Dynamic classification and integration method of multimodal forest and grassland data

Abstract

The present invention discloses a multimodal forest and grassland data dynamic classification and integration method, which relates to the field of data processing technology. The multimodal forest and grassland data dynamic classification and integration method continuously acquires forest and grassland time series data and performs data analysis separately to obtain the forest and grassland ecological health index and site factor index, and performs integrated analysis to obtain the forest and grassland classification protection index. At the same time, the method continuously acquires human activity time series data and combines it with the forest and grassland classification protection index for integrated analysis to obtain a comprehensive forest and grassland classification protection index for each quarter of each year in a set area. The method also performs trend analysis to obtain a comprehensive forest and grassland classification protection change index for each quarter of several groups of adjacent years. The method classifies forest and grassland resource protection based on the comprehensive forest and grassland classification protection change index for each quarter and takes corresponding protection measures based on the classification results, thereby monitoring the changing trends of forest and grassland resources and helping managers accurately judge and take corresponding protection measures.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and specifically to a method for dynamic classification and integration of multimodal forest and grassland data. Background Art

[0002] In the management and protection of forest and grassland resources, with the changes in the ecological environment and the impact of human activities, the dynamic monitoring and evaluation of forest and grassland resources are particularly important. Forest and grassland resources not only play an important role in the ecosystem, but are also closely related to climate change, land use, agricultural production and other aspects. In recent years, traditional ground survey methods have been difficult to meet the needs of large-scale, real-time and precise monitoring due to their limitations in time, space, cost, etc. With the rapid development of remote sensing technology and Internet of Things technology, forest and grassland resource monitoring based on multimodal data sources has become a research hotspot. These data can provide rich spatial and temporal information to help achieve accurate classification and evaluation of forest and grassland resources. The integration of multimodal data combines different types of data with the application of intelligent algorithms, which can effectively improve the accuracy and adaptability of forest and grassland resource classification.

[0003] Existing technologies, such as the digital forest and grassland system disclosed in patent application publication number CN115587083A, relate to the technical field of digital forest and grassland systems and include: a basic support platform for data storage, calculation, and manipulation; a business application layer for user operations; infrastructure for hardware and database management; and bidirectional signal connections between the basic support platform and the business application layer, as well as between the basic support platform and the infrastructure. This digital forest and grassland system introduces big data into forest and grassland resource management, integrates and consolidates various types of forest and grassland data, promotes data sharing and openness, strengthens big data analysis and mining, and establishes a management mechanism that "uses data to speak, use data to make decisions, use data to manage, and use data to innovate." This will effectively promote the transition of forest and grassland resource management and protection from a qualitative management approach based on experience to a precise governance approach driven by data, and comprehensively enhance the supervision, management, analysis, evaluation, macro-control, and decision-making support capabilities of forest and grassland resources.

[0004] Based on the above scheme, it was found that the limitations of the existing technology include at least the following problems: the existing technology fails to effectively capture the dynamic changes of forest and grass resources in different seasons. Forest and grass resources show different growth and status changes in the four seasons, and the existing technology does not analyze these seasonal differences, making it difficult to timely and dynamically reflect the health status of forest and grass resources, resulting in decision makers finding it difficult to take appropriate ecological management measures for different seasons. Secondly, the existing technology lacks the ability to analyze and predict the dynamic trends of the status of forest and grass resources in each season. The forest and grass status in each season has its own unique change pattern, but the existing technology is difficult to accurately predict these changes, which easily leads to delays in management measures and difficulty in timely adjustments. Summary of the Invention

[0005] In response to the shortcomings of the existing technology, the present invention provides a dynamic classification and integration method for multimodal forest and grassland data, which solves the problem that the existing technology fails to effectively capture the dynamic changes of forest and grassland resources in different seasons and the dynamic trend analysis and prediction capabilities of the forest and grassland resource status in each season.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a multimodal forest and grassland data dynamic classification and integration method, comprising the following steps: continuously acquiring forest and grassland time series data within a set area, the forest and grassland time series data including forest and grassland growth time series data and site environment time series data; performing data analysis on the forest and grassland time series data within the set area respectively to obtain the forest and grassland ecological health index and site factor index for each quarter of each year in the set area, and performing integrated analysis to obtain the forest and grassland classification protection index for each quarter of each year in the set area; at the same time, continuously acquiring the human activities time series data within the set area, and performing integrated analysis in combination with the forest and grassland classification protection index to obtain the comprehensive forest and grassland classification protection index for each quarter of each year in the set area; performing trend analysis on the comprehensive forest and grassland classification protection index for each quarter of each year in the set area to obtain the comprehensive forest and grassland classification protection change index for each quarter of several groups of adjacent years in the set area; classifying forest and grassland resource protection based on the comprehensive forest and grassland classification protection change index for each quarter of several groups of adjacent years in the set area, and taking corresponding protection measures based on the classification results.

[0007] Furthermore, the forest and grassland ecological health time series data include the normalized vegetation index, forest and grassland coverage value, leaf area index, forest and grassland photosynthetically active radiation value, forest and grassland temperature anomaly value, forest and grassland texture index and relative abundance value and stomatal conductance value of each forest and grassland in each quarter of each year; the site environment time series data include the soil moisture content value, soil particle content index, soil microbial activity index and soil thermal diffusion index in each quarter of each year; the human activity time series data include the human activity frequency index, human land degradation rate index, human energy consumption index and human ecological restoration intervention frequency index in each quarter of each year.

[0008] Furthermore, the specific formula for calculating the forest and grassland classification protection index for each quarter of each year in the set area is as follows: Among them, LtF ij is the forest and grassland classification protection index of the jth quarter of the i-th year in the set area, ZdY ij 、DzY ijThey are the comprehensive forest and grassland ecological health index and site factor index of the jth quarter of the i-th year in the set area, respectively. α1 and α2 are the forest and grassland diversity coefficient and site factor coefficient stored in the database, respectively. β1, β2, and β3 are the forest and grassland adjustment coefficient, site adjustment coefficient, and interaction adjustment coefficient stored in the database, respectively. α1+α2=1, i=1, 2, 3,…, i0, i0 is the number of years, j=1, 2, 3,…, j0, j0 is the number of quarters.

[0009] Furthermore, the specific steps for obtaining the comprehensive forest and grass ecological health index of each quarter of each year in the set area are as follows: standardize the normalized vegetation index, forest and grass coverage value, leaf area index, forest and grass photosynthetically active radiation value, forest and grass temperature anomaly value, forest and grass texture index, and relative abundance value and stomatal conductance value of each forest and grass in each quarter of each year in the set area; conduct a comprehensive analysis based on the normalized normalized vegetation index, forest and grass coverage value, leaf area index, forest and grass photosynthetically active radiation value, and relative abundance value and stomatal conductance value of each forest and grass in each quarter of each year in the set area to obtain the initial forest and grass ecological health index of each quarter of each year in the set area; and conduct a comprehensive analysis of the forest and grass temperature anomaly value and forest and grass texture index of each quarter of each year in the set area after the standardized processing to obtain the disease index of each quarter of each year in the set area; and conduct a comprehensive analysis of the initial forest and grass ecological health index and disease index of each quarter of each year in the set area to obtain the comprehensive forest and grass ecological health index of each quarter of each year in the set area.

[0010] Furthermore, the specific formulas for calculating the initial forest and grassland ecological health index and the comprehensive forest and grassland ecological health index for each quarter of each year in the set area are as follows: Among them, CsT ij is the initial forest and grassland ecological health index of the jth quarter of the i-th year in the set area, GzB′ ij , LcF′ ij 、YmZ′ ij , LcG′ ij are the normalized vegetation index, forest and grass coverage value, leaf area index, and forest and grass photosynthetically active radiation value of the jth quarter of the i-th year in the set area after standardization, QkZ′ ijq , XdG′ ijq is the relative abundance value and stomatal conductance value of the qth forest and grass species in the jth quarter of the i-th year in the set area after standardization. η1, η2, and η3 are the forest and grass coverage influence coefficient, the first interaction coefficient, and the second interaction coefficient stored in the database. ZdY ijis the comprehensive forest and grassland ecological health index of the jth quarter of the i-th year in the set area, λ1 is the initial coefficient stored in the database, WhB ij is the disease index of the jth quarter of the ith year in the set area, λ2 is the disease coefficient stored in the database, λ1+λ2=1, i=1, 2, 3,…, i0, i0 is the number of years, j=1, 2, 3,…, j0, j0 is the number of quarters, q=1, 2, 3,…, q0, q0 is the number of forest and grass species, and e is a natural constant.

[0011] Furthermore, the specific steps for obtaining the site factor index for each quarter of each year in the set area are as follows: obtaining the soil moisture reference value, soil particle size content reference index, soil microbial activity reference index, and soil thermal diffusion reference index for each quarter in the set area; and comprehensively analyzing the soil moisture reference value, soil particle size content reference index, soil microbial activity reference index, and soil thermal diffusion reference index for each quarter in the set area, as well as the soil moisture value, soil particle size content index, soil microbial activity index, and soil thermal diffusion index for each quarter of each year, to obtain the site factor index for each quarter of each year in the set area.

[0012] Furthermore, the specific formula for calculating the site factor index for each quarter of each year in a set area is as follows: Among them, DzY ij is the forest and grassland site factor index of the jth quarter of the ith year in the set area, TdL ij 、TkY ij 、TwH ij TrK ij The soil moisture content, soil particle size index, soil microbial activity index, soil thermal diffusion index, CdL j , CkY j 、CwH j CrK j They are the reference values ​​of soil moisture content, soil particle size content, soil microbial activity, and soil thermal diffusion in the jth quarter in the set area, respectively. δ1, δ2, δ3, and δ4 are the moisture coefficient, particle size coefficient, microbial activity coefficient, and thermal diffusion coefficient stored in the database, respectively. i = 1, 2, 3, …, i0, where i0 is the number of years, and j = 1, 2, 3, …, j0, where j0 is the number of quarters.

[0013] Furthermore, the specific steps for obtaining the comprehensive forest and grassland classification protection index for each quarter of each year in the set area are as follows: normalize the human activity frequency index, human land degradation rate index, human energy consumption index, and human ecological restoration intervention frequency index for each quarter of each year in the set area; and conduct a comprehensive analysis based on the human activity frequency index, human land degradation rate index, human energy consumption index, and human ecological restoration intervention frequency index for each quarter of each year in the set area after the normalization process to obtain the human activity forest and grassland correction index for each quarter of each year in the set area; and conduct a comprehensive analysis of the human activity forest and grassland correction index and forest and grassland classification protection index for each quarter of each year in the set area to obtain the comprehensive forest and grassland classification protection index for each quarter of each year in the set area.

[0014] Furthermore, the specific formulas for calculating the human activities forest and grassland correction index and the comprehensive forest and grassland classification protection index for each quarter of each year in the region are as follows:

[0015]

[0016] Among them, RwX ij KqW′ is the forest and grassland correction index of human activities in the jth quarter of the i-th year in the set area, ij 、FqW′ ij , RgW′ ij , TdC′ ij are the human activity frequency index, human land degradation rate index, human energy consumption index, and human ecological restoration intervention frequency index of the jth quarter of the i-th year in the set area after normalization. θ1, θ2, θ3, and θ4 are the activity coefficient, degradation coefficient, energy consumption coefficient, and intervention coefficient stored in the database, respectively. θ1+θ2+θ3+θ4=1, and ZcF ij is the comprehensive forest and grassland classification protection index of the jth quarter of the i-th year in the set area, is the humanities correction adjustment coefficient stored in the database, LtF ij is the forest and grassland classification protection index of the jth quarter of the i-th year in the set area, is the classification adjustment coefficient stored in the database, is the interaction classification adjustment coefficient stored in the database, i = 1, 2, 3, ..., i0, i0 is the number of years, j = 1, 2, 3, ..., j0, j0 is the number of quarters.

[0017] Furthermore, the specific steps of classifying forest and grassland resource protection based on the comprehensive forest and grassland classification protection change index of each quarter of several groups of adjacent years in the set area, and taking corresponding protection measures based on the classification results are as follows: the comprehensive forest and grassland classification protection change index of each quarter of each group of adjacent years in the set area are judged and analyzed respectively with the preset comprehensive forest and grassland classification protection change index threshold range; if the comprehensive forest and grassland classification protection change index of each quarter of each group of adjacent years in the set area is higher than the preset upper limit of the comprehensive forest and grassland classification protection change index threshold range, the set area is marked as the first level protection, and the first protection measure is taken; if the comprehensive forest and grassland classification protection change index of each quarter of each group of adjacent years in the set area is within the preset comprehensive forest and grassland classification protection change index threshold, the set area is marked as the second level protection, and the second protection measure is taken; if the comprehensive forest and grassland classification protection change index of each quarter of each group of adjacent years in the set area is lower than the preset lower limit of the comprehensive forest and grassland classification protection change index threshold range, the set area is marked as the third level protection, and the third protection measure is taken.

[0018] The present invention has the following beneficial effects:

[0019] (1) This multimodal dynamic classification and integration method for forest and grassland data continuously obtains time series data on forest and grassland growth, site environment, and human activities, and combines it with dynamic analysis to monitor the changing trends of forest and grassland resources in real time, thereby reflecting the actual status of forest and grassland resources and quickly intervening and classifying them for protection when the health of forest and grassland is threatened. This helps managers accurately judge and take appropriate protection measures, thereby avoiding missing protection opportunities due to delayed judgment and effectively reducing the risk of ecosystem degradation.

[0020] (2) This multimodal forest and grassland data dynamic classification and integration method conducts comprehensive analysis of multimodal data, thereby being able to more comprehensively reflect the current ecological and environmental status of the region, incorporate human activities into the analysis, and combine the impact of human activities on the ecological environment to scientifically adjust the priority of protection measures, thereby improving the accuracy of protection measures, and balancing the relationship between ecological protection and human activities, thereby achieving long-term sustainable protection of forest and grassland resources.

[0021] (3) This multimodal forest and grassland data dynamic classification and integration method compares and analyzes the comprehensive forest and grassland classification protection change index of each quarter with the preset threshold range, and automatically marks the area as different protection levels according to different ecological changes. This ensures that the set area is appropriately protected according to its current ecological status and change trend, thereby effectively preventing ecological degradation and improving the targeted nature of protection measures.

[0022] Of course, any product implementing the present invention does not necessarily need to achieve all of the advantages described above at the same time. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 This is a flow chart of the dynamic classification and integration method of multimodal forest and grassland data of the present invention.

[0024] Figure 2 This is a flow chart of the steps for obtaining the comprehensive forest and grassland ecological health index for each quarter of each year in a set area in the multimodal forest and grassland data dynamic classification and integration method of the present invention.

[0025] Figure 3 This is a flow chart of the steps for obtaining the comprehensive forest and grassland classification protection index for each quarter of each year in a set area in the multimodal forest and grassland data dynamic classification and integration method of the present invention. DETAILED DESCRIPTION

[0026] The overall approach to the problems in the embodiments of this application is as follows:

[0027] Continuously obtain forest and grassland time series data in the set area, and perform data analysis separately to obtain the forest and grassland ecological health index and site factor index of each quarter of each year in the set area, and perform integrated analysis to obtain the forest and grassland classification protection index of each quarter of each year in the set area. At the same time, continuously obtain time series data of human activities, and perform integrated analysis in combination with the forest and grassland classification protection index to obtain the comprehensive forest and grassland classification protection index of each quarter of each year in the set area, and perform trend analysis to obtain the comprehensive forest and grassland classification protection change index of each quarter of several groups of adjacent years in the set area, classify forest and grassland resource protection, and take corresponding protection measures based on the classification results.

[0028] See also Figure 1, the embodiment of the present invention provides a technical solution: a multimodal forest and grassland data dynamic classification and integration method, comprising the following steps: continuously acquiring forest and grassland time series data in a set area, wherein the forest and grassland time series data include forest and grassland growth time series data and site environment time series data; performing data analysis on the forest and grassland time series data in the set area respectively, obtaining the forest and grassland ecological health index and site factor index of each quarter of each year in the set area, and performing integrated analysis to obtain the forest and grassland classification protection index of each quarter of each year in the set area; at the same time, continuously acquiring the human activities time series data in the set area, and performing integrated analysis in combination with the forest and grassland classification protection index to obtain the set The comprehensive forest and grassland classification protection index of each quarter of each year in the region; trend analysis of the comprehensive forest and grassland classification protection index of each quarter of each year in the set region (it should be noted here that three consecutive years are grouped as a group, for example, the first quarter of the first year, the first quarter of the second year, and the first quarter of the third year are analyzed), to obtain the comprehensive forest and grassland classification protection change index of each quarter of several groups of adjacent years in the set region; classification of forest and grassland resource protection is carried out based on the comprehensive forest and grassland classification protection change index of each quarter of several groups of adjacent years in the set region, and corresponding protection measures are taken based on the classification results.

[0029] The specific formula for calculating the comprehensive forest and grassland classification protection change index for each quarter of several groups of adjacent years within a set area is as follows: Among them, BzF a is the comprehensive forest and grassland classification protection change index of the jth quarter of the ath group of consecutive years in the set area, ZcF( i+2 ) j is the comprehensive forest and grassland classification protection index of the jth quarter of the i+2th year in the set area, ZcF ij is the comprehensive forest and grassland classification protection index of the jth quarter of the i-th year in the set area, φ1 is the first trend adjustment coefficient stored in the database, ZcF( i+1 ) j is the comprehensive forest and grassland classification protection index of the jth quarter of the i+1th year in the set area, φ2 is the second trend adjustment coefficient stored in the database, and φ1+φ2=1, a=1, 2, 3, …, a0, a0 is the number of adjacent year groups, i=1, 2, 3, …, i0, i0 is the number of years, j=1, 2, 3, …, j0, j0 is the number of quarters.

[0030] It should be explained that φ1 and φ2 can be obtained through the following steps: read the comprehensive forest and grassland classification protection index of each quarter of each year in the set area, and divide it into three consecutive years starting from the initial acquisition year, and perform mean analysis on the comprehensive forest and grassland classification protection index of each quarter of the third year of three consecutive years and the first year (i.e. the initial acquisition year) to obtain the mean change of the first comprehensive forest and grassland classification protection index, and perform mean analysis on the comprehensive forest and grassland classification protection index of each quarter of the second year of three consecutive years and the first year (i.e. the initial acquisition year) to obtain the mean change of the second comprehensive forest and grassland classification protection index, sum up and analyze the mean change of the first comprehensive forest and grassland classification protection index and the mean change of the second comprehensive forest and grassland classification protection index to obtain the sum of changes, and perform proportion analysis on the mean change of the first comprehensive forest and grassland classification protection index and the mean change of the second comprehensive forest and grassland classification protection index with the sum of changes, and use the proportion analysis results as the corresponding coefficients.

[0031] The forest and grassland ecological health time series data include the normalized vegetation index, forest and grassland coverage value, leaf area index, forest and grassland photosynthetically active radiation value, forest and grassland temperature anomaly value, forest and grassland texture index and the relative abundance value and stomatal conductance value of each forest and grassland in each quarter of each year; the site environment time series data include the soil moisture content value, soil particle content index, soil microbial activity index and soil thermal diffusion index in each quarter of each year; the human activity time series data include the human activity frequency index, human land degradation rate index, human energy consumption index and human ecological restoration intervention frequency index in each quarter of each year.

[0032] Among them, the normalized vegetation index is the vegetation coverage and photosynthesis intensity, which can be obtained through the sampling method, that is, selecting vegetation on multiple soils for sampling, and using the MSI (multispectral imager) sensor to obtain the red light band reflectance and near-red light band reflectance of the vegetation for analysis, that is, (near red light band reflectance - red light band reflectance) / (near red light band reflectance - red light band reflectance), and averaging the analysis results of multiple vegetations, and the result is the parameter.

[0033] The leaf area index is the sum of the total area of ​​all leaves of forest and grass, which represents the photosynthetic capacity of forest and grass. The total area of ​​all leaves per unit area can be obtained through the LAI meter.

[0034] The photosynthetically active radiation value of forest and grass is the light intensity that can be utilized by forest and grass, which characterizes the photosynthesis rate. It can be obtained through the sampling method, that is, selecting multiple soils for sampling, and using ground light sensors to collect the light intensity of forest and grass on each soil, and performing averaging processing. The result is this parameter.

[0035] The abnormal value of forest and grass temperature is the absolute value of the difference between the temperature value of forest and grass under pests and diseases and the temperature value under normal growth. It can be obtained through the sampling method, that is, selecting multiple soils for sampling, and using temperature sensors to collect the temperature of forest and grass on each soil, and performing difference processing on the temperature value under normal growth, and performing average processing on the difference processing results. The result obtained is the parameter.

[0036] The forest and grass texture index is the leaf strength of forest and grass when attacked by pests and diseases. It can be obtained through the sampling method, that is, multiple soils are selected for sampling, and the stiffness (the ability of leaves to resist deformation when bent or stretched when subjected to external forces, the stiffness of multiple leaves is obtained using a material testing machine, and the average is processed) and hardness (the compressive strength of leaves, that is, the ability of the leaf surface to resist indentations or scratches by external forces, the hardness of multiple leaves is obtained using a hardness tester, and the average is processed) of the leaves of forest and grass on each soil are obtained. The stiffness and hardness of the leaves on each soil are weighted, and the weighted processing results on each soil are averaged, and the result is the parameter.

[0037] The soil moisture value can be obtained through a soil moisture sensor.

[0038] The soil particle size content index is a comprehensive value of the proportion of soil particles of different particle sizes in the soil, which affects many aspects such as soil permeability, water retention, aeration, and plant growth. It can be obtained through sampling, that is, selecting multiple soil locations for sampling and using a laser particle size analyzer (which uses the principle of laser beam passing through soil particles and scattering to measure the size of particles based on the scattering intensity and angle of the particles) to obtain the proportion of each particle, perform weighted processing, and perform mean processing based on the weighted processing results. The result obtained is the parameter.

[0039] The soil microbial activity index is the metabolic intensity of microorganisms in the soil (which can be expressed by the carbon dioxide release rate), reflecting the health of the soil ecosystem. It can be obtained through sampling, that is, sampling soil from multiple locations and using a portable soil respirometer to obtain the carbon dioxide release rate, and then averaging the result to obtain this parameter.

[0040] The soil thermal diffusivity index is the diffusion rate of heat in the soil, which affects soil temperature changes and the adaptability of vegetation roots. It can be obtained through the sampling method, that is, selecting multiple soil locations for sampling, and obtaining the soil thermal conductivity (obtained by a heat flow meter), soil density, and specific heat capacity (obtained by a thermal analyzer) of each soil location, and calculating them, that is, soil thermal diffusivity index = soil thermal conductivity / (soil density*specific heat capacity), and performing average processing. The result is the parameter.

[0041] The human activity frequency index is the frequency of human activities (such as mining, logging, construction, etc.) in that quarter, which can be obtained through reports from relevant management departments stored in the database.

[0042] The human-induced land degradation rate index is the total area of ​​land degraded due to human activities (such as agricultural activities, urban expansion, mining, deforestation, etc.) in the quarter, that is, the human-induced land degradation rate index = total area of ​​land degradation / quarter length. The total area of ​​land degradation can be obtained through the monitoring reports on land degradation released by relevant departments and stored in the database.

[0043] The anthropogenic energy consumption index is an indicator of the comprehensive energy consumed in the quarter, which can be obtained through the following steps: obtaining the consumption of electricity, natural gas, and water resources in the set area (all of which can be obtained through energy consumption reports stored in the energy management platform), standardizing them, and performing weighted processing based on the standardized processing results. The result obtained is the parameter.

[0044] The frequency index of human ecological restoration intervention is the sum of the areas of each human ecological restoration (such as afforestation, wetland restoration, grassland restoration, etc.) within the quarter, and the area of ​​each restoration can be obtained through the ecological restoration project report of the environmental protection department stored in the database.

[0045] Specifically, the specific formula for calculating the forest and grassland classification protection index for each quarter of each year in a set area is as follows: Among them, LtF ij is the forest and grassland classification protection index of the jth quarter of the i-th year in the set area, ZdY ij is the forest and grassland ecological health index of the jth quarter of the i-th year in the set area, α1 is the forest and grassland diversity coefficient stored in the database, β1 is the forest and grassland adjustment coefficient stored in the database, DzY i is the forest and grassland site factor index of the jth quarter of the i-th year in the set area, α2 is the site factor coefficient stored in the database, β2 is the site adjustment coefficient stored in the database, α1+α2=1, β3 is the interaction adjustment coefficient stored in the database, α1+α2=1, i=1, 2, 3,…, i0, i0 is the number of years, j=1, 2, 3,…, j0, j0 is the number of quarters.

[0046] It needs to be explained that the β3*ZdY in the formula ij *DzY ij This item is used to represent the superposition effect between the forest and grassland ecological health index and the site factor index, and to avoid the forest and grassland classification protection index being too large or too small.

[0047] And α1 and α2 can be obtained through the following steps: read the forest and grassland ecological health index and site factor index of each quarter of each year in the set area, and perform mean analysis, perform sum analysis based on the mean analysis results to obtain the classification and value, and perform proportion analysis on the mean analysis results and the classification and value respectively, and use the proportion analysis results as the corresponding coefficients.

[0048] β1, β2, and β3 can be obtained through the following steps: using historical data, combined with indicators such as the forest and grassland ecological health index and the site factor index, to conduct statistical regression analysis, quantify the specific impact of each factor on the forest and grassland classification protection index, and thus fit the initial weight value; secondly, using the sensitivity analysis method, adjust the value range of each coefficient, observe its impact on the evaluation results of the forest and grassland classification protection index, ensure the stability and rationality of the model, and based on regional characteristics and actual conditions, correct and optimize the preliminary fitting coefficients, and finally determine the coefficient value applicable to the region.

[0049] In this implementation plan, through mean analysis, sum analysis and proportion analysis, the superposition effect of forest and grassland ecological health index and site factor index can be comprehensively considered, so that the final forest and grassland classification protection index can more accurately reflect the ecological status of the region. Secondly, through regression analysis and sensitivity analysis, the specific impact of factors such as forest and grassland ecological health index and site factor index on forest and grassland classification protection index can be quantified, which helps to clarify the role of each factor in the overall evaluation and avoid errors in human estimation. Through sensitivity analysis, the value range of the coefficient can also be adjusted to verify its impact on the results, so as to ensure the stability and reliability of the model in practical applications, so as to dynamically identify which factors have a greater impact on forest and grassland resource protection, and timely adjust model parameters to improve the accuracy and practicality of the evaluation results. Finally, through the optimization of the coefficients, the flexibility and adaptability of the evaluation method are ensured, thereby providing tailored solutions for forest and grassland protection in different seasons and improving the accuracy and efficiency of resource management.

[0050] Specifically, if Figure 2As shown, the specific steps for obtaining the comprehensive forest and grass ecological health index of each quarter of each year in the set area are as follows: standardize the normalized vegetation index, forest and grass coverage value, leaf area index, forest and grass photosynthetically active radiation value, forest and grass temperature anomaly value, forest and grass texture index, and the relative abundance value and stomatal conductance value of each forest and grass in each quarter of each year in the set area; conduct a comprehensive analysis based on the normalized normalized vegetation index, forest and grass coverage value, leaf area index, forest and grass photosynthetically active radiation value, and the relative abundance value and stomatal conductance value of each forest and grass in each quarter of each year in the set area to obtain the initial forest and grass ecological health index of each quarter of each year in the set area; and conduct a comprehensive analysis of the forest and grass temperature anomaly value and forest and grass texture index of each quarter of each year in the set area after standardization to obtain the disease index of each quarter of each year in the set area; and conduct a comprehensive analysis of the initial forest and grass ecological health index and disease index of each quarter of each year in the set area to obtain the comprehensive forest and grass ecological health index of each quarter of each year in the set area.

[0051] The specific formula for calculating the disease index for each quarter of each year in the set area is as follows: Among them, WhB ij is the disease index of the jth quarter of the i-th year in the set area, LcW′ ij is the abnormal temperature value of forest and grass in the jth quarter of the i-th year in the set area after standardization, μ1 is the temperature coefficient stored in the database, LcZ′ ij is the forest and grass texture index of the jth quarter of the ith year in the set area after standardization, μ2 is the texture coefficient stored in the database, μ1+μ2=1, i=1, 2, 3,…, i0, i0 is the number of years, j=1, 2, 3,…, j0, j0 is the number of quarters.

[0052] It should be explained that μ1, μ2, and μ3 can be obtained through the following steps: read the forest and grass temperature anomaly values ​​and forest and grass texture index of each quarter of each year in the set area after standardization, and perform mean analysis to obtain the mean forest and grass temperature anomaly values ​​and the mean forest and grass texture index values ​​in the set area after standardization, and perform sum analysis to obtain the hazard sum value, and perform a proportion analysis on the mean forest and grass temperature anomaly values ​​and the mean forest and grass texture index values ​​in the set area after standardization with the hazard sum value, and use the proportion results as the corresponding coefficients.

[0053] The specific formulas for calculating the initial forest and grassland ecological health index and the comprehensive forest and grassland ecological health index for each quarter of each year in the set area are as follows: Among them, CsT ijis the initial forest and grassland ecological health index of the jth quarter of the i-th year in the set area, GzB′ ij LcF′ is the normalized vegetation index of the jth quarter of the i-th year in the set area after standardization. ij is the forest and grass coverage value of the jth quarter of the ith year in the set area after standardization, η1 is the forest and grass coverage impact coefficient stored in the database (i.e., the impact of forest and grass coverage on the normalized vegetation index), YmZ′ ij LcG′ is the leaf area index of the jth quarter of the i-th year in the set area after standardization. ij is the photosynthetically active radiation value of forest and grass in the jth quarter of the ith year in the set area after standardization, η2 is the first interaction coefficient stored in the database (i.e., the interaction between leaf area index and photosynthetically active radiation value of forest and grass), QkZ′ ijq is the relative abundance value of the qth forest and grass species in the jth quarter of the ith year in the set area after standardization, XdG′ ijq is the stomatal conductance value of the qth forest and grass species in the jth quarter of the ith year in the set area after standardization, η3 is the second interaction coefficient stored in the database (i.e., the interaction between relative abundance value and stomatal conductance value), ZdY ij is the comprehensive forest and grassland ecological health index of the jth quarter of the i-th year in the set area, λ1 is the initial coefficient stored in the database, WhB ij is the disease index of the jth quarter of the ith year in the set area, λ2 is the disease coefficient stored in the database, λ1+λ2=1, i=1, 2, 3,…, i0, i0 is the number of years, j=1, 2, 3,…, j0, j0 is the number of quarters, q=1, 2, 3,…, q0, q0 is the number of forest and grass species, e is a natural constant, and in this implementation example, the value is 2.71.

[0054] It should be explained that η1, η2, and η3 can be obtained through the following steps: using historical data, combined with normalized vegetation index, forest and grass coverage value, leaf area index, forest and grass photosynthetic active radiation value, and relative abundance value, stomatal conductance value and other indicators of each forest and grass, statistical regression analysis is carried out to quantify the impact of forest and grass coverage on the normalized vegetation index, the interaction between leaf area index and forest and grass photosynthetic active radiation value, and the specific impact of the interaction between relative abundance value and stomatal conductance value on the forest and grass classification protection index, so as to fit the initial weight value. Then, the sensitivity analysis method is used to adjust the value range of the coefficient to evaluate the stability and applicability of these parameters to the formula output. Next, the weights are further fitted through model optimization (such as machine learning algorithm) to ensure that the formula can accurately reflect the actual initial health status of forest and grass, and the coefficients are fine-tuned based on the characteristics of different regions to ensure that they are suitable for specific ecological and environmental assessment needs.

[0055] λ1 and λ2 can be obtained through the following steps: read the initial forest and grassland ecological health index and disease index of each quarter of each year in the set area, perform mean analysis, perform sum analysis based on the mean analysis results to obtain the comprehensive forest and grassland ecological health value, perform proportion analysis on the mean analysis results and the comprehensive forest and grassland ecological health value, and use the proportion results as the corresponding coefficients.

[0056] In this implementation plan, through the integrated processing of multiple parameters such as normalized vegetation index, forest and grass coverage rate, leaf area index, photosynthetically active radiation value, etc., the status of forest and grass ecological environment can be fully reflected, avoiding the one-sidedness of a single factor. In addition, the calculation of disease index not only focuses on the health status of forest and grass, but also introduces temperature anomalies, stress hormones, texture and other factors, which helps to provide early warning of possible ecological problems, thereby improving the pertinence and timeliness of forest and grass protection. Secondly, the interaction between different ecological indicators is taken into account, which helps to accurately reflect the actual health status of the ecosystem and reduce the error caused by a single factor. At the same time, through regression analysis, sensitivity analysis and machine learning optimization coefficients, the model can continuously adapt to changes in the ecological environment in the region, thereby ensuring the stability and long-term applicability of the model. Finally, the various coefficients are calculated through standardized data, so that the impact of different factors on the final result is quantified and optimized, thereby accurately reflecting the actual situation of forest and grass ecological health.

[0057] Specifically, the specific steps for obtaining the site factor index for each quarter of each year in the set area are as follows: obtain the soil moisture reference value, soil particle size content reference index, soil microbial activity reference index, and soil thermal diffusion reference index for each quarter in the set area; and comprehensively analyze the soil moisture reference value, soil particle size content reference index, soil microbial activity reference index, and soil thermal diffusion reference index for each quarter in the set area, as well as the soil moisture value, soil particle size content index, soil microbial activity index, and soil thermal diffusion index for each quarter of each year to obtain the site factor index for each quarter of each year in the set area.

[0058] The soil moisture reference value can be obtained by the following steps: obtaining historical soil moisture values ​​of several historical quarters (the same as the quarter), and performing weighted averaging processing to obtain the soil moisture reference value.

[0059] At the same time, the steps for obtaining the soil particle content reference index, soil microbial activity reference index, and soil thermal diffusion reference index are consistent with the soil moisture reference value.

[0060] The specific formula for calculating the site factor index for each quarter of each year in a given area is as follows: Among them, DzY ij is the forest and grassland site factor index of the jth quarter of the ith year in the set area, TdL ij is the soil moisture content value of the jth quarter of the ith year in the set area, CdL j is the reference value of soil moisture in the jth quarter in the set area, δ1 is the moisture coefficient stored in the database, TkY ij CkY is the soil particle content index of the jth quarter of the ith year in the set area, j is the reference index of soil particle size content in the jth quarter in the set area, δ2 is the particle size coefficient stored in the database, TwH ij is the soil microbial activity index of the jth quarter of the i-th year in the set area, CwH j is the soil microbial activity reference index in the jth quarter in the set area, δ3 is the microbial activity coefficient stored in the database, TrK ij is the soil thermal diffusion index of the jth quarter of the ith year in the set area, CrK j is the soil thermal diffusion reference index of the jth quarter in the set area, δ4 is the thermal diffusion coefficient stored in the database, i = 1, 2, 3, ..., i0, i0 is the number of years, and j = 1, 2, 3, ..., j0, j0 is the number of quarters.

[0061] The specific implementation example of calculating the site factor index for each quarter of each year in a set area is as follows. The existing data include the soil moisture content, soil particle content index, soil microbial activity index, and soil thermal diffusion index for the four quarters of the first year. The specific data are shown in Tables 1 and 2:

[0062] Table 1 Example of site environment data within the specified area

[0063]

[0064]

[0065] Table 2 Example of site environment reference data within the specified area

[0066]

[0067] The water content coefficient δ1 stored in the database is approximately: 0.58;

[0068] The particle size coefficient δ2 stored in the database is approximately: 0.61;

[0069] The microbial activity coefficient δ3 stored in the database is approximately: 0.69;

[0070] The thermal diffusivity δ4 stored in the database is approximately: 0.76;

[0071] Substituting the above coefficients and the data in Tables 1 and 2 into the specific formula for calculating the site factor index for each quarter of each year in the set area, we obtain:

[0072] The site factor index of the first quarter of the first year in the region is set to ln(1+

[0073] (0.17 / 0.18) 0.58 *(0.28 / 0.33) 0.61 *(0.75 / 0.70) 0.69 *(0.11 / 0.12) 0.76 )≈0.63;

[0074] The site factor index of the second quarter of the first year in the region is set to ln(1+

[0075] (0.11 / 0.10) 0.58 *(0.34 / 0.29) 0.61 *(0.62 / 0.65) 0.69 *(0.25 / 0.20) 0.76 )≈0.85;

[0076] The site factor index of the third quarter of the first year in the region is set to ln(1+

[0077] (0.14 / 0.15) 0.58 *(0.27 / 0.35) 0.61 *(0.83 / 0.80) 0.69 *(0.14 / 0.15) 0.76 )≈0.60;

[0078] The site factor index of the fourth quarter of the first year in the region is set to ln(1+

[0079] (0.18 / 0.20) 0.58 *(0.31 / 0.27) 0.61 *(0.52 / 0.50) 0.69 *(0.08 / 0.06) 0.76 )≈0.84.

[0080] It should be explained that δ1, δ2, δ3, and δ4 can be obtained through the following steps: using historical data, evaluating the influence of each variable on the site factor index through statistical modeling and regression analysis, and thus fitting the initial weight values, and then adjusting the value range of these coefficients based on sensitivity analysis to ensure that the formula has good adaptability to site factors under different environmental conditions, and reasonably modifying the weight coefficients for specific regions.

[0081] In this implementation plan, by combining multiple indicators such as soil moisture content, particle size force, microbial activity, and thermal diffusion, the physical, chemical, and biological properties of the soil can be fully reflected, thereby ensuring that the site factor index is more accurate. By using historical data for weighted averaging, the reference value is ensured to be representative, avoiding the impact of short-term data fluctuations on the analysis results, thereby improving the stability of the site factor index. Secondly, through regression analysis and sensitivity analysis, the influence of each factor can be quantified and adjusted, making the formula more adaptable, so that it can cope with changes in different environmental conditions and regional characteristics, and thus has important significance for application in different soil types or ecological environments. At the same time, by integrating the reference values ​​and actual measured values ​​of multiple indicators such as soil moisture content, particle size force, and microbial activity, the influence of different soil characteristics on the site factor can be more flexibly reflected, thereby improving the accuracy of environmental management and monitoring. Finally, by accurately calculating the site factor index, a scientific basis can be provided for soil protection, restoration, and ecological environment management, thereby helping decision makers understand the soil condition and then evaluate the potential impact of site factors, thereby formulating more reasonable management strategies.

[0082] Specifically, if Figure 3 As shown, the specific steps for obtaining the comprehensive forest and grassland classification protection index for each quarter of each year in the set area are as follows: normalize the human activity frequency index, human land degradation rate index, human energy consumption index, and human ecological restoration intervention frequency index for each quarter of each year in the set area (i.e., remove the unit); and conduct a comprehensive analysis based on the human activity frequency index, human land degradation rate index, human energy consumption index, and human ecological restoration intervention frequency index for each quarter of each year in the set area after normalization to obtain the human activity forest and grassland correction index for each quarter of each year in the set area; and conduct a comprehensive analysis of the human activity forest and grassland correction index and forest and grassland classification protection index for each quarter of each year in the set area to obtain the comprehensive forest and grassland classification protection index for each quarter of each year in the set area.

[0083] The specific formulas for calculating the human activities forest and grassland correction index and the comprehensive forest and grassland classification protection index for each quarter of each year within the region are as follows:

[0084]

[0085] Among them, RwX ij KqW′ is the forest and grassland correction index of human activities in the jth quarter of the i-th year in the set area, ij is the human activity frequency index of the jth quarter of the ith year in the set area after normalization, θ1 is the activity coefficient stored in the database, and FqW′ ij is the normalized anthropogenic land degradation rate index in the jth quarter of the ith year in the set area, θ2 is the degradation coefficient stored in the database, RgW′ ij is the normalized human energy consumption index for the jth quarter of the ith year in the set area, θ3 is the energy consumption coefficient stored in the database, TdC′ ij is the human ecological restoration intervention frequency index in the jth quarter of the i-th year in the set area after normalization, θ4 is the intervention coefficient stored in the database, θ1+θ2+θ3+θ4=1, ZcF ij is the comprehensive forest and grassland classification protection index of the jth quarter of the i-th year in the set area, is the humanities correction adjustment coefficient stored in the database, LtF ij is the forest and grassland classification protection index of the jth quarter of the i-th year in the set area, is the classification adjustment coefficient stored in the database, is the interaction classification adjustment coefficient stored in the database, i = 1, 2, 3, ..., i0, i0 is the number of years, j = 1, 2, 3, ..., j0, j0 is the number of quarters.

[0086] It should be explained that θ1, θ2, θ3, and θ4 can be obtained through the following steps: read the human activity frequency index, human land degradation rate index, human energy consumption index, and human ecological restoration intervention frequency index of each quarter of each year in the set area after normalization processing, and perform mean analysis. Based on the mean analysis results, a sum analysis is performed to obtain a comprehensive classification and value, and the mean analysis results are respectively analyzed with the comprehensive classification and value, and the proportion analysis results are used as the corresponding coefficients.

[0087] In the formula This item is used to represent the superposition effect between the human activities forest and grassland correction index and the forest and grassland classification protection index, and to avoid the comprehensive forest and grassland classification protection index being too large or too small.

[0088] and It can be obtained through the following steps: using historical data, through the dynamic change analysis of variables, using statistical regression methods to quantify the initial impact of each variable on the comprehensive forest and grassland classification protection index, so as to obtain the initial coefficient value; then, based on the sensitivity analysis technology, adjust the value range of these coefficients under different environmental scenarios to ensure the applicability of the formula to diverse forest and grassland environments; then, calibrate the model through scenario analysis to optimize the rationality and stability of the parameter values.

[0089] In this implementation plan, by normalizing indicators such as the human activity frequency index, human land degradation rate index, human energy consumption index, and human ecological restoration intervention frequency index, it is possible to quantify and comprehensively consider the various impacts of human activities on the forest and grassland ecological environment, thereby providing more accurate data support for forest and grassland protection. Secondly, through sensitivity analysis, regression analysis, and dynamic change analysis of historical data, the weight value of each coefficient can be flexibly adjusted so that the model can adapt to changes in different environments, improving the applicability and stability of the formula in diverse scenarios. Through reasonable proportion analysis and interactive adjustment of coefficients, it is possible to avoid the superposition effect between the human activity forest and grassland correction index and the forest and grassland classification protection index being too large or too small, thereby avoiding overestimation or underestimation of the protection effect, thereby ensuring that the final comprehensive forest and grassland classification protection index truly reflects the ecological health status.

[0090] Specifically, the specific steps of classifying forest and grassland resource protection based on the comprehensive forest and grassland classification protection change index of each quarter of several groups of adjacent years in a set area, and taking corresponding protection measures based on the classification results are as follows: the comprehensive forest and grassland classification protection change index of each quarter of each group of adjacent years in the set area is respectively compared with the preset comprehensive forest and grassland classification protection change index threshold interval; if the comprehensive forest and grassland classification protection change index of each quarter of each group of adjacent years in the set area is higher than the preset upper limit of the comprehensive forest and grassland classification protection change index threshold interval (i.e., the maximum value of the interval), the set area (of that quarter) is marked as the first-level protection, and the first protection measure is taken (i.e., providing suggestions to relevant departments on strengthening environmental monitoring, increasing investment in protection resources, taking measures such as grazing ban, closed restoration, and vegetation restoration; prohibiting development activities, restricting human activities, increasing ecological restoration projects, strengthening soil and water conservation, and reducing pollution source input) If the comprehensive forest and grassland classification protection change index of each quarter of each group of adjacent years in the set area is within the preset comprehensive forest and grassland classification protection change index threshold, the set area (of that quarter) will be marked as the second-level protection, and the second protection measures will be taken (i.e., providing ecological restoration measures, artificial intervention, regular inspection and restoration management to relevant departments; restricting certain types of development activities, but allowing limited resource development in accordance with ecological protection requirements); if the comprehensive forest and grassland classification protection change index of each quarter of each group of adjacent years in the set area is lower than the lower limit of the preset comprehensive forest and grassland classification protection change index threshold interval (i.e., the minimum value of the interval), the set area (of that quarter) will be marked as the third-level protection, and the third protection measures will be taken (i.e., providing relevant departments with suggestions on continuing to maintain ecological monitoring, implementing moderate sustainable utilization, appropriately reducing ecological intervention, and strengthening regional management to prevent possible ecological degradation).

[0091] In this implementation plan, by comparing the comprehensive forest and grassland classification protection change index of each quarter of several groups of years with the preset threshold interval, the health status and change trend of forest and grassland resources in the region can be dynamically and regularly evaluated, thereby avoiding the deviation that may be caused by the assessment at a single time point, and graded protection is implemented according to different change indices, and the forest and grassland resources in the region are managed according to different protection levels to ensure the rational use and protection of resources. Secondly, through dynamic analysis of quarterly data, protection measures are adjusted in real time to ensure that the measures always match the changes in the ecological environment, thereby avoiding excessive or insufficient protection, and enhancing the adaptability and flexibility of regional management. Finally, through continuous monitoring and reasonable intervention, the ecological quality of the region is gradually restored and improved, thereby ensuring the long-term stability and health of the ecosystem, and preventing ecological degradation or irreversible damage.

[0092] In summary, this application has at least the following effects:

[0093] By continuously acquiring time-series data on forest and grass growth, site environment, and human activities, and combining it with dynamic analysis, we can monitor the changing trends of forest and grass resources in real time, thereby reflecting the actual status of forest and grass resources. When the health of forest and grass is threatened, we can quickly intervene and classify them for protection, thereby helping managers to accurately judge and take appropriate protection measures, thereby avoiding missing protection opportunities due to delayed judgment and effectively reducing the risk of ecosystem degradation.

[0094] Through comprehensive analysis of multimodal data, we can more comprehensively reflect the current ecological and environmental status of the region, incorporate human activities into the analysis, and combine the impact of human activities on the ecological environment to scientifically adjust the priority of protection measures, thereby improving the accuracy of protection measures, and balancing the relationship between ecological protection and human activities, thereby achieving long-term sustainable protection of forest and grassland resources.

[0095] By comparing and analyzing the comprehensive forest and grassland classification protection change index of each quarter with the preset threshold range, and automatically marking the area as different protection levels according to different ecological changes, it is ensured that the set area is appropriately protected according to its current ecological status and change trends, thereby effectively preventing ecological degradation and improving the targeted nature of protection measures.

[0096] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0097] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A dynamic classification and integration method for multimodal forest and grassland data, characterized by: The following steps are involved: Continuously acquiring forest and grass time series data within a set area, wherein the forest and grass time series data includes forest and grass growth time series data and site environment time series data; Data analysis was performed on the forest and grassland time series data within the set area to obtain the forest and grassland ecological health index and site factor index for each quarter of each year within the set area. An integrated analysis was then performed to obtain the forest and grassland classification protection index for each quarter of each year within the set area. The specific formula is as follows: Among them, LtF ij is the forest and grassland classification protection index of the jth quarter of the i-th year in the set area, ZdY ij 、DzY ij are the comprehensive forest and grassland ecological health index and site factor index of the jth quarter of the i-th year in the set area, α1 and α2 are the forest and grassland diversity coefficient and site factor coefficient stored in the database, β1, β2, and β3 are the forest and grassland adjustment coefficient, site adjustment coefficient, and interaction adjustment coefficient stored in the database, α1+α2=1, i=1, 2, 3, …, i0, i0 is the number of years, j=1, 2, 3, …, j0, j0 is the number of quarters; At the same time, the time series data of human activities in the set area are continuously obtained, and integrated analysis is performed in combination with the forest and grassland classification protection index to obtain the comprehensive forest and grassland classification protection index for each quarter of each year in the set area; The specific formulas for calculating the human activities forest and grassland correction index and the comprehensive forest and grassland classification protection index for each quarter of each year within the region are as follows: Among them, RwX ij KqW′ is the forest and grassland correction index of human activities in the jth quarter of the i-th year in the set area, ij 、FqW′ ij , RgW′ ij , TdC′ ij are the human activity frequency index, human land degradation rate index, human energy consumption index, and human ecological restoration intervention frequency index of the jth quarter of the i-th year in the set area after normalization. θ1, θ2, θ3, and θ4 are the activity coefficient, degradation coefficient, energy consumption coefficient, and intervention coefficient stored in the database, respectively. θ1+θ2+θ3+θ4=1, and ZcF ij is the comprehensive forest and grassland classification protection index of the jth quarter of the i-th year in the set area, is the humanities correction adjustment coefficient stored in the database, LtF ij is the forest and grassland classification protection index of the jth quarter of the i-th year in the set area, is the classification adjustment coefficient stored in the database, Adjust coefficients for interaction classifications stored in the database; Conduct trend analysis on the comprehensive forest and grassland classification protection index of each quarter of each year in the set area to obtain the comprehensive forest and grassland classification protection change index of each quarter of several groups of adjacent years in the set area; Forest and grassland resource protection classification is carried out based on the comprehensive forest and grassland classification protection change index of each quarter of several groups of adjacent years in the set area, and corresponding protection measures are taken based on the classification results.

2. The multimodal forest and grassland data dynamic classification and integration method according to claim 1 is characterized in that: The forest and grassland ecological health time series data include the normalized vegetation index, forest and grassland coverage value, leaf area index, forest and grassland photosynthetically active radiation value, forest and grassland temperature anomaly value, forest and grassland texture index and the relative abundance value and stomatal conductance value of each forest and grassland in each quarter of each year; the site environment time series data include the soil moisture content value, soil particle content index, soil microbial activity index and soil thermal diffusion index in each quarter of each year; the human activity time series data include the human activity frequency index, human land degradation rate index, human energy consumption index and human ecological restoration intervention frequency index in each quarter of each year.

3. The multimodal forest and grassland data dynamic classification and integration method according to claim 2 is characterized in that: The specific steps to obtain the comprehensive forest and grassland ecological health index for each quarter of each year in the set area are as follows: The normalized vegetation index, forest and grass coverage value, leaf area index, forest and grass photosynthetically active radiation value, forest and grass temperature anomaly value, forest and grass texture index, and relative abundance value and stomatal conductance value of each forest and grass in each quarter of each year in the set area are standardized; Based on the standardized normalized vegetation index, forest and grass coverage value, leaf area index, forest and grass photosynthetically active radiation value, and relative abundance value and stomatal conductance value of each forest and grass in each quarter of each year in the set area, a comprehensive analysis is conducted to obtain the initial forest and grass ecological health index for each quarter of each year in the set area; And conduct a comprehensive analysis of the abnormal temperature values ​​and forest and grass texture index of each quarter of each year in the set area after standardization, and obtain the disease index of each quarter of each year in the set area; The initial forest and grassland ecological health index and disease index of each quarter of each year in the set area are comprehensively analyzed to obtain the comprehensive forest and grassland ecological health index of each quarter of each year in the set area.

4. The multimodal forest and grassland data dynamic classification and integration method according to claim 3 is characterized in that: The specific formulas for calculating the initial forest and grassland ecological health index and the comprehensive forest and grassland ecological health index for each quarter of each year in the set area are as follows: Among them, CsT ij is the initial forest and grassland ecological health index of the jth quarter of the i-th year in the set area, GzB′ ij , LcF′ iJ 、YmZ′ ij , LcG′ ij are the normalized vegetation index, forest and grass coverage value, leaf area index, and forest and grass photosynthetically active radiation value of the jth quarter of the i-th year in the set area after standardization, QkZ′ ijq , XdG′ ijq is the relative abundance value and stomatal conductance value of the qth forest and grass species in the jth quarter of the i-th year in the set area after standardization. η1, η2, and η3 are the forest and grass coverage influence coefficient, the first interaction coefficient, and the second interaction coefficient stored in the database. ZdY ij is the comprehensive forest and grassland ecological health index of the jth quarter of the i-th year in the set area, λ1 is the initial coefficient stored in the database, WhB ij is the disease index of the jth quarter of the ith year in the set area, λ2 is the disease coefficient stored in the database, λ1+λ2=1, i=1, 2, 3,…, i0, i0 is the number of years, j=1, 2, 3,…, j0, j0 is the number of quarters, q=1, 2, 3,…, q0, q0 is the number of forest and grass species, and e is a natural constant.

5. The multimodal forest and grassland data dynamic classification and integration method according to claim 2 is characterized in that: The specific steps to obtain the site factor index for each quarter of each year in the set area are as follows: Obtain soil moisture reference values, soil particle content reference index, soil microbial activity reference index, and soil thermal diffusion reference index for each quarter in the set area; The soil moisture reference value, soil particle content reference index, soil microbial activity reference index, soil thermal diffusion reference index of each quarter in the set area and the soil moisture value, soil particle content index, soil microbial activity index, soil thermal diffusion index of each quarter of each year are comprehensively analyzed to obtain the site factor index of each quarter of each year in the set area.

6. The multimodal forest and grassland data dynamic classification and integration method according to claim 5 is characterized in that: The specific formula for calculating the site factor index for each quarter of each year in a given area is as follows: Among them, DzY ij is the forest and grassland site factor index of the jth quarter of the ith year in the set area, TdL ij 、TkY ij 、TwH ij TrK ij The soil moisture content, soil particle size index, soil microbial activity index, soil thermal diffusion index, CdL j , CkY j 、CwH j CrK j They are the reference values ​​of soil moisture content, soil particle size content, soil microbial activity, and soil thermal diffusion in the jth quarter in the set area, respectively. δ1, δ2, δ3, and δ4 are the moisture coefficient, particle size coefficient, microbial activity coefficient, and thermal diffusion coefficient stored in the database, respectively. i = 1, 2, 3, …, i0, where i0 is the number of years, and j = 1, 2, 3, …, j0, where j0 is the number of quarters.

7. The multimodal forest and grassland data dynamic classification and integration method according to claim 2 is characterized in that: The specific steps to obtain the comprehensive forest and grassland classification protection index for each quarter of each year in the set area are as follows: Normalize the human activity frequency index, human land degradation rate index, human energy consumption index, and human ecological restoration intervention frequency index for each quarter of each year within the set area; A comprehensive analysis is conducted based on the normalized human activity frequency index, human land degradation rate index, human energy consumption index, and human ecological restoration intervention frequency index for each quarter of each year within the set area to obtain the human activity forest and grassland correction index for each quarter of each year within the set area. A comprehensive analysis is also conducted on the forest and grassland correction index for human activities and the forest and grassland classification protection index for each quarter of each year in the set area to obtain the comprehensive forest and grassland classification protection index for each quarter of each year in the set area.

8. The multimodal forest and grassland data dynamic classification and integration method according to claim 1 is characterized in that: The specific steps for classifying forest and grassland resources for protection based on the comprehensive forest and grassland classification protection change index for each quarter of several groups of consecutive years within a set area and taking corresponding protection measures based on the classification results are as follows: The comprehensive forest and grassland classification protection change index of each quarter of each group of adjacent years in the set area is compared with the preset comprehensive forest and grassland classification protection change index threshold range for judgment analysis; If the comprehensive forest and grassland classification protection change index of each quarter in each group of adjacent years in the set area is higher than the preset comprehensive forest and grassland classification protection change index threshold upper limit, the set area will be marked as the first level protection and the first protection measure will be taken; If the comprehensive forest and grassland classification protection change index of each quarter of each group of adjacent years in the set area is within the preset comprehensive forest and grassland classification protection change index threshold, the set area will be marked as the second-level protection and the second protection measure will be taken; If the comprehensive forest and grassland classification protection change index of each quarter of each group of adjacent years in the set area is lower than the preset lower limit of the comprehensive forest and grassland classification protection change index threshold range, the set area will be marked as the third level protection and the third protection measures will be taken.

Images