Ecological restoration effect evaluation method and system based on satellite remote sensing

Abstract

The invention discloses an ecological restoration effect evaluation method and system based on satellite remote sensing, and the method comprises the steps: collecting the satellite remote sensing data and field soil data of a preset region, and taking the remote sensing data of the preset region within a specified time as to-be-analyzed data; calculating a time sequence deviation and a contrast ratio based on the satellite remote sensing data to obtain a grading region; performing deviation correction on the satellite remote sensing data according to the field soil data in the grading region, and inputting the corrected satellite remote sensing data into a space-time ecological development prediction model to obtain ecological development prediction data; performing restoration uncertainty analysis according to the ecological development prediction data and the field soil data to obtain an ecological restoration effect confidence coefficient, and performing elastic capability analysis according to the ecological restoration effect confidence coefficient and the satellite remote sensing data to obtain an ecological elastic index; and constructing a time-varying ecological restoration effect evaluation model according to the ecological elasticity index, and outputting an evaluation result based on the to-be-analyzed data.

Description

Technical Field

[0001] The present invention relates to the field of ecological assessment, and particularly to a method and system for evaluating the effect of ecological restoration based on satellite remote sensing. Background Art

[0002] With the combined influence of global climate change and human activities, the problem of ecosystem degradation has become increasingly prominent, and ecological restoration has become an important starting point for regional sustainable development. Traditional evaluation of ecological restoration effects mostly relies on small-scale field surveys, which have bottlenecks such as long data collection cycles, limited spatial coverage, and insufficient dynamic monitoring capabilities, and it is difficult to meet the evaluation requirements of large-scale, multi-factor, and time-series. Although satellite remote sensing technology can provide continuous observations of wide-area surface information, its application in the ecological restoration scenario still faces three challenges: First, the fusion mechanism of multi-source remote sensing data and ground measured data has not been perfected, resulting in the inversion accuracy of ecological parameters being limited by the inherent bias of a single data source; Second, the evolution of the ecosystem has significant spatio-temporal heterogeneity, and conventional static evaluation models are difficult to capture the non-linear response and elastic recovery threshold during the restoration process; Third, the ecological effects of restoration projects are often affected by uncertain factors such as climate fluctuations and human disturbances, and there is a theoretical lack in the existing methods for quantifying the confidence level of restoration effects and the anti-interference ability of the system.

[0003] In recent years, dynamic monitoring technologies based on time-series remote sensing data have provided new ideas for the above problems. Research shows that by fusing multi-dimensional ecological parameters such as vegetation coverage, surface temperature, and remote sensing indices of biodiversity, a comprehensive characterization system of ecosystem health can be constructed. However, existing research mostly focuses on the state evaluation of a single time node, lacking comparative analysis of the time-series evolution laws of the restored area and the non-restored area, resulting in insufficient accuracy in identifying key restoration driving factors. In addition, as the material carrier of ecological restoration, the spatial coupling relationship between its physical and chemical properties and remote sensing inversion parameters has not been fully explored, restricting the construction of the "sky-earth" collaborative evaluation model. How to lock in ecological sensitive areas through time-series deviation analysis, correct remote sensing inversion errors by combining ground soil data, and then establish a dynamic evaluation model integrating elastic recovery force has become the key technical breakthrough direction for improving the quantification accuracy of ecological restoration effects. Summary of the Invention

[0004] The object of the present invention is to provide a method for evaluating the effect of ecological restoration based on satellite remote sensing.

[0005] To achieve the above object, the present invention is implemented according to the following technical solution:

[0006] The present invention includes the following steps:

[0007] Collect satellite remote sensing data and on-site soil data in the preset area, and use the remote sensing data in the preset area within the specified time as the data to be analyzed; the data to be analyzed includes vegetation information, land use and land cover data, topographic and geomorphic data, hydrological data, thermal environment data, and biodiversity data;

[0008] Calculate the temporal deviation based on the satellite remote sensing data, and use the preset area with the temporal deviation greater than 0 as the research area, and vice versa as the non-research area. Obtain the graded area by grading the changes within the research area according to the contrast; including:

[0009] Calculate the temporal deviation:

[0010]

[0011] where the temporal deviation is θ, the upper limit of the monitoring time is T s , the a-th remote sensing data at the s2-th moment is h a (s2), the a-th remote sensing data at the s1-th moment is h a (s1), the starting moment of monitoring is s o , the height of the remote sensing satellite is D, the radius of the earth is R, and the viewing angle is

[0012] Perform deviation correction on the satellite remote sensing data according to the on-site soil data within the graded area, and input the corrected satellite remote sensing data into the spatio-temporal ecological development prediction model to obtain ecological development prediction data;

[0013] Conduct uncertainty analysis of restoration based on the ecological development prediction data and the on-site soil data to obtain the confidence level of ecological restoration effect, and conduct elasticity ability analysis based on the confidence level of ecological restoration effect and the satellite remote sensing data to obtain the ecological elasticity index;

[0014] Construct a time-varying ecological restoration effect evaluation model based on the ecological elasticity index, and input the data to be analyzed into the time-varying ecological restoration effect evaluation model to output the evaluation result.

[0015] Furthermore, the method for obtaining the graded area by grading the changes within the research area according to the contrast includes:

[0016] Conduct grid division on the research area according to its size and shape to obtain sub-grids, extract the change characteristics within the sub-grids according to the change ratio of the remote sensing satellite data, and use the change characteristics with the change ratio greater than 0.013 as the key characteristics;

[0017] Take the sub-grid as a unit, construct a clustering feature tree based on the sub-grid according to the contrast and the key characteristics, use the key characteristics as the clustering feature nodes, and calculate the abnormal degree index of the clustering feature nodes;

[0018] Input the anomaly degree index into the student psychological optimization algorithm, eliminate the abnormal clustering feature nodes according to the optimization characteristics, obtain the optimized key feature set, integrate and adjust the clustering feature nodes in the optimized key feature set, and update the clustering feature tree;

[0019] Calculate the contrast of the sub-grid based on the spatio-temporal change according to the updated clustering feature tree:

[0020]

[0021] Among them, the contrast of the w-th sub-grid is l w , the number of key features in the optimized key feature set within the w-th sub-grid is N w , the v-th key feature at the s2 moment is H v (s2), the v-th discriminative feature at the s1 moment is H v (s1), the weight coefficient of the v-th key feature is ζ v , the sign function is sgn(·);

[0022] When the contrast is less than 0, the sub-grid is a poor-quality repair area; when the contrast is greater than 0 and less than 0.0051, the sub-grid is a substandard repair area; when the contrast is greater than 0.0051 and less than 0.015, the sub-grid is a medium repair area; when the contrast is greater than 0.015 and less than 0.0297, the sub-grid is a good repair area; when the contrast is greater than 0.0297, the sub-grid is an excellent repair area;

[0023] Connect the neighboring sub-grids of the same level according to the contrast, and output the connected result as a graded area.

[0024] Furthermore, the method of the spatio-temporal ecological development prediction model includes:

[0025] Collect the climate environment data in the preset area, and perform time series impact analysis on the climate environment data to obtain environmental factors;

[0026] The spatio-temporal ecological development prediction model includes a graph network, time-guided embedding, and a long short-term memory neural network;

[0027] The graph network represents the pixels, regions, or objects in the remote sensing data as graph nodes, represents the relationships between the graph nodes as edges, uses the graph neural network architecture to capture the complex relationships between the graph nodes, and extracts the complex spatio-temporal features of the remote sensing data according to the complex relationships;

[0028] Time-guided embedding integrates time information into the embedding vector by learning the spatio-temporal ecological development, and captures the time cycle pattern of ecological development in combination with the complex spatio-temporal features;

[0029] The long short-term memory neural network captures long-term dependencies in time series through gating mechanisms and cell states, and predicts ecological development based on time cycling patterns and long-term dependencies.

[0030] Furthermore, a method for obtaining the confidence level of ecological restoration effect through uncertainty analysis of restoration based on the ecological development prediction data and the on-site soil data includes:

[0031] Convert the ecological development prediction data and the on-site soil data into multi-dimensional vectors, and construct an ecological feature matrix based on the multi-dimensional vectors;

[0032] Obtain the ecological restoration effect indicators and ecological features of the graded area, and construct a least absolute shrinkage and selection operator (Lasso) regression model based on the ecological restoration effect indicators and ecological features;

[0033] In the Lasso regression model, take the sum of the squares of the standardized coefficients of the ecological features as the sensitivity coefficient, obtain the non-zero frequency of the weights of the ecological features in the Lasso regression model, and calculate the importance of the ecological features:

[0034]

[0035] Where the sensitivity coefficient of the w-th ecological feature is χ w , the non-zero frequency of the w-th ecological feature is F w , and the importance of the w-th ecological feature is φ w ;

[0036] Calculate the correlation between ecological features using the Pearson correlation coefficient, take the ecological features with a correlation greater than 0.673 as key features, use the elbow method to determine the optimal number of clusters of key features, and divide the graded area into multiple ecological homogeneous groups based on the relationship features;

[0037] Randomly sample the ecological homogeneous groups according to a proportion, use traditional ecological indicators for random sampling, set a control group with the same area, obtain the ecological environment quality scores, soil quality indices, and vegetation coverage of the ecological homogeneous groups and the control group, and calculate the ecological restoration effects of the ecological homogeneous groups and the control group:

[0038]

[0039] Where the ecological restoration effect of the c-th group is The experimental area of the c-th group is S c , the vegetation coverage at the s2 moment is The vegetation coverage at the s1 moment is The ecological environment quality score at the s2 moment is The ecological environment quality score at the s1 moment is The soil quality index at time s2 is The soil quality index at time s1 is The weight coefficients are β1, β2, and β3 respectively;

[0040] Draw the time-series ecological restoration effect reference curve based on ecological characteristics and the time-series ecological restoration effect curve based on key characteristics, and calculate the difference degree between the time-series ecological restoration effect reference curve and the time-series ecological restoration effect curve:

[0041]

[0042] where the number of characteristics is N c , the i-th characteristic is b i , the ecological restoration effect of characteristic b at time s2 i is The ecological restoration effect reference of characteristic b at time s1 i is The difference degree of the c-th ecological homogeneous group is K c ;

[0043] Take the ratio of the difference degree of the sampling result to the mean of the difference degrees as the coefficient of variation. When the coefficient of variation is greater than 0.279, output the difference between 1 and the coefficient of variation as the ecological restoration effect confidence level.

[0044] Furthermore, a method for obtaining the ecological resilience index by performing resilience capacity analysis based on the ecological restoration effect confidence level and the satellite remote sensing data includes:

[0045] Construct a multi-dimensional ecological index system based on satellite remote sensing data, and screen the multi-dimensional ecological indexes based on the 4R theory refinement indexes to obtain resilience evaluation indexes; the screening includes: extracting indexes of key ecological functions according to the types of damage risks, removing redundant indexes according to the substitutability of species within the ecosystem, and screening intelligent indexes according to the self-regulation and mobilization and restoration capabilities when facing damage risks;

[0046] Divide the ecological resilience change process into a preparation stage, a resistance and absorption stage, an adaptive regulation stage, and a recovery stage according to time, and construct an ecological resilience comprehensive measurement model based on the performance indexes of each stage of the ecosystem and the 4R theory refinement indexes;

[0047] Calculate the ecological resilience index:

[0048]

[0049] where the ecological resilience index of the k-th classification area is The number of ecological resilience evaluation indexes is M r , the weight of the r-th index in the t-th stage is ρ r(t), the elasticity evaluation index in the t-th stage is f r (t), the preparation stage is t1, the recovery stage is t4, and the confidence level of the ecological restoration effect in the k-th classification area in the t-th stage is The index weight in the preparation stage t1 is ρ r (t1), the elasticity evaluation index in the preparation stage t1 is f r (t1), the confidence level of the ecological restoration effect in the k-th classification area in the preparation stage t1 is

[0050] Furthermore, the method for constructing a time-varying ecological restoration effect evaluation model based on the ecological elasticity index includes:

[0051] Obtain the ecological performance evaluation index and corresponding weight of the research area, and construct an objective function according to the ecological elasticity index and the loss function. The expression is:

[0052]

[0053]

[0054] Among them, the loss function is The initial ecosystem performance of the k-th classification area is HO k , the number of classification areas in the research area is M k , the r-th index weight at the s-th moment is η r (s), the ecological performance evaluation index at the s-th moment is g r (s), the ecosystem performance of the k-th classification area at the s-th moment is H k (s), the ecological elasticity index of the k-th classification area at the s-th moment is The objective function at the s-th moment is

[0055] The time-varying ecological restoration effect evaluation model includes random forest algorithm, time series analysis algorithm, and spatio-temporal graph neural network;

[0056] The random forest algorithm conducts ensemble learning on multiple decision trees of satellite remote sensing data, constructs multiple decision trees by randomly selecting features and samples, and determines the classification result through a voting mechanism;

[0057] The time series analysis algorithm based on spatio-temporal spectral features extracts the time, space, and spectral features of pixel points, combines a support vector machine classifier for change detection, and extracts the temporal change law of surface features according to the change detection;

[0058] The spatio-temporal graph neural network is constructed into a spatio-temporal graph according to the temporal variation law, and the spatio-temporal module of the graph neural network is used to model the spatial and temporal dependence relationships respectively. The ecological restoration effect is predicted according to the dependence relationships, and the predicted ecological restoration effect is output.

[0059] In a second aspect, an ecological restoration effect evaluation system based on satellite remote sensing includes:

[0060] A data acquisition module: used to acquire satellite remote sensing data and field soil data of a preset area, and use the remote sensing data of the preset area within a specified time as the data to be analyzed; the data to be analyzed includes vegetation information, land use and land cover data, topographic and geomorphic data, hydrological data, thermal environment data, and biodiversity data;

[0061] A zoning and grading module: used to calculate the temporal deviation based on the satellite remote sensing data, use the preset area with the temporal deviation greater than 0 as the research area, and vice versa as the non-research area, and perform change grading on the research area according to the contrast to obtain graded areas;

[0062] A correction and prediction module: in the graded area, correct the deviation of the satellite remote sensing data according to the field soil data, and input the corrected satellite remote sensing data into the spatio-temporal ecological development prediction model to obtain ecological development prediction data;

[0063] A confidence elasticity module: used to perform restoration uncertainty analysis based on the ecological development prediction data and the field soil data to obtain the confidence of the ecological restoration effect, and perform elasticity ability analysis based on the confidence of the ecological restoration effect and the satellite remote sensing data to obtain the ecological elasticity index;

[0064] A modeling and output module: used to construct a time-varying ecological restoration effect evaluation model according to the ecological elasticity index, input the data to be analyzed into the time-varying ecological restoration effect evaluation model, and output the evaluation result.

[0065] The present invention is an ecological restoration effect evaluation method and system based on satellite remote sensing. Compared with the prior art, the present invention has the following technical effects:

[0066] Through steps such as deviation correction, obtaining ecological development prediction data, restoration uncertainty analysis, elasticity ability analysis, and model construction, the present invention can improve the accuracy of ecological restoration effect evaluation, thereby improving the precision of ecological restoration effect evaluation. Optimizing the ecological restoration effect evaluation can greatly save resources and improve work efficiency. It can realize the intelligent evaluation of ecological restoration effects, perform restoration uncertainty analysis and elasticity ability analysis on ecological restoration effect evaluation in real time, which is of great significance for ecological restoration effect evaluation, and can adapt to different standards of ecological restoration effect evaluation and different ecological restoration effect evaluation requirements, having a certain universality. Description of the Drawings

[0067] Figure 1 This is a flowchart showing the steps of an ecological restoration effect evaluation method based on satellite remote sensing according to the present invention. Detailed Embodiments

[0068] The present invention will be further described below through specific embodiments. The illustrative embodiments and descriptions of the present invention are used to explain the present invention, but do not limit the present invention.

[0069] An ecological restoration effect evaluation method and system based on satellite remote sensing according to the present invention includes the following steps:

[0070] As Figure 1 shown, in this embodiment, it includes the following steps:

[0071] Collect satellite remote sensing data and on-site soil data of a preset area, and use the remote sensing data of the preset area within a specified time as the data to be analyzed; the data to be analyzed includes vegetation information, land use and land cover data, topographic and geomorphic data, hydrological data, thermal environment data, and biodiversity data;

[0072] In actual evaluation, an ecological restoration area is used as the research object, and the satellite remote sensing data and on-site soil data from 2019 to 2023 are used as historical data for training a time-varying ecological restoration effect evaluation model, and the satellite remote sensing data of Area A and Area B in the spring of 2024 are used as the data to be analyzed;

[0073] Area A: The vegetation information is that the vegetation coverage rate is 85%, mainly broad-leaved forests and coniferous forests, mixed with a small amount of shrubs and herbaceous plants, and the vegetation health index is 0.8; the land use and land cover data is that the forest coverage area is 70%, the grassland area is 15%, the water area is 10%, and the construction land area is 5%; the topographic and geomorphic data is that the average altitude is 800 meters, the average slope is 15°, and most of the areas have slopes between 10° and 20°, mainly mountains and hills, with the mountain proportion being 60% and the hill proportion being 40%; the hydrological data is that the river density is 0.3 km / km², the average annual precipitation is 1200 mm, and the average depth of the groundwater level is 5 meters; the thermal environment data is that the average surface temperature in summer is 25°C, the heat island effect intensity is weak, and the average temperature difference is less than 2°C; the biodiversity data is that the species richness index is 12 and the number of rare species is 10;

[0074] Region B: Vegetation information shows a vegetation coverage rate of 30%, mainly sparse herbaceous plants and a small amount of drought-tolerant shrubs, with herbaceous plants accounting for 71% and shrubs accounting for 29%, and a vegetation health index of 0.3; Land use and land cover data show a forest coverage area of 10%, grassland area of 40%, water area of 5%, construction land area of 25%, and wasteland area of 20%; Topographic and geomorphic data show an average altitude of 500 meters, an average slope of 5°, with most areas having slopes between 0° - 10°, mainly plains and a small amount of low hills, plains accounting for 80% and low hills accounting for 20%; Hydrological data show a river density of 0.1 km / km², an average annual precipitation of 600 mm, and an average groundwater depth of 15 meters; Thermal environment data show an average surface temperature of 32°C in summer, a strong heat island effect intensity, and an average temperature difference greater than 5°C; Biodiversity data show a species richness index of 4 and the number of rare species of 2;

[0075] The on-site soil data of Region A in March are as follows: sand content 30%, silt content 40%, clay content 30%, organic matter content 3.5%, total nitrogen content 0.2%, available phosphorus content 25 mg / kg, available potassium content 180 mg / kg, pH value 6.8, average water content 25%, microbial biomass carbon 1500 mg / kg, soil bulk density 1.2 g / cm³;

[0076] The on-site soil data of Region B in March are as follows: sand content 60%, silt content 20%, clay content 20%, organic matter content 1.0%, total nitrogen content 0.08%, available phosphorus content 10 mg / kg, available potassium content 80 mg / kg, pH value 8.2, average water content 10%, microbial biomass carbon 500 mg / kg, soil bulk density 1.5 g / cm³;

[0077] Calculate the temporal deviation based on the satellite remote sensing data, take the preset area with the temporal deviation greater than 0 as the research area, and vice versa as the non-research area, and obtain the graded area by grading the changes within the research area according to the contrast; including:

[0078] Calculate the temporal deviation:

[0079]

[0080] Where the temporal deviation is θ, the upper limit of the monitoring time is T s , the a-th remote sensing data at the s2 moment is h a (s2), the a-th remote sensing data at the s1 moment is h a (s1), the starting moment of monitoring is s o , the height of the remote sensing satellite is D, the radius of the earth is R, and the viewing angle is

[0081] In the actual evaluation, the research areas are Area A and Area B; the sub-areas of Area A are A1, A2, A3, and A4; the sub-areas of Area B are B1, B2, and B3; the classified areas of Area A are A1, A2 and A3, A4; the classified areas of Area B are B1, B3;

[0082] In the classified area, the satellite remote sensing data is corrected for deviation according to the field soil data, and the corrected satellite remote sensing data is input into the spatio-temporal ecological development prediction model to obtain ecological development prediction data;

[0083] Uncertainty analysis of restoration is carried out based on the ecological development prediction data and the field soil data to obtain the confidence level of ecological restoration effect, and elastic capacity analysis is carried out based on the confidence level of ecological restoration effect and the satellite remote sensing data to obtain the ecological elasticity index;

[0084] In the actual evaluation, the confidence levels of ecological restoration effects of Area A and Area B are 0.877 and 0.861 respectively; the ecological elasticity indices of Area A and Area B are 0.429 and 0.218 respectively;

[0085] A time-varying ecological restoration effect evaluation model is constructed based on the ecological elasticity index, and the data to be analyzed is input into the time-varying ecological restoration effect evaluation model to output the evaluation result;

[0086] In the actual evaluation, the restoration effects of Area A and Area B are 0.607 and 0.211 respectively.

[0087] In this embodiment, the method for obtaining the classified area by grading the changes in the research area according to the contrast includes:

[0088] Grid division is carried out according to the size and shape of the research area to obtain sub-grids, the change characteristics within the sub-grids are extracted according to the change ratio of the remote sensing satellite data, and the change characteristics with a change ratio greater than 0.013 are used as key characteristics;

[0089] Taking the sub-grid as a unit, a clustering feature tree based on the sub-grid is constructed according to the contrast and the key characteristics, the key characteristics are used as clustering feature nodes, and the abnormal degree index of the clustering feature nodes is calculated;

[0090] The abnormal degree index is input into the student psychological optimization algorithm, the abnormal clustering feature nodes are eliminated according to the optimization characteristics to obtain an optimized key feature set, and the clustering feature nodes in the optimized key feature set are integrated and adjusted to update the clustering feature tree;

[0091] According to the updated clustering feature tree, the contrast of the sub-grid is calculated based on the spatio-temporal changes:

[0092]

[0093] where the contrast of the w-th sub-grid is l w and the number of key features in the optimized key feature set within the w-th sub-grid is N w and the v-th key feature at the s2 moment is H v (s2), and the v-th discriminative feature at the s1 moment is H v (s1), and the weight coefficient of the v-th key feature is ζ v and the sign function is sgn(·);

[0094] When the contrast is less than 0, the sub-grid is a poor repair area; when the contrast is greater than 0 and less than 0.0051, the sub-grid is a substandard repair area; when the contrast is greater than 0.0051 and less than 0.015, the sub-grid is a medium repair area; when the contrast is greater than 0.015 and less than 0.0297, the sub-grid is a good repair area; when the contrast is greater than 0.0297, the sub-grid is an excellent repair area;

[0095] Connect adjacent sub-grids of the same level according to the contrast, and output the connected result as a graded area.

[0096] In this embodiment, the method of the spatio-temporal ecological development prediction model includes:

[0097] Collect climate environment data within a preset area, and perform time-series impact analysis on the climate environment data to obtain environmental factors;

[0098] The spatio-temporal ecological development prediction model includes a graph network, time-guided embedding, and a long short-term memory neural network;

[0099] The graph network represents pixels, regions, or objects in remote sensing data as graph nodes, represents the relationships between graph nodes as edges, uses a graph neural network architecture to capture the complex relationships between graph nodes, and extracts complex spatio-temporal features of remote sensing data according to the complex relationships;

[0100] Time-guided embedding incorporates time information into the embedding vector by learning spatio-temporal ecological development, and captures the time cycle pattern of ecological development in combination with complex spatio-temporal features;

[0101] The long short-term memory neural network captures long-term dependencies in the time series through a gating mechanism and a cell state, and predicts ecological development based on the time cycle pattern and long-term dependencies.

[0102] In this embodiment, the method for obtaining the confidence level of ecological restoration effect through repair uncertainty analysis based on the ecological development prediction data and the field soil data includes:

[0103] Convert the ecological development prediction data and the field soil data into multi-dimensional vectors, and construct an ecological feature matrix according to the multi-dimensional vectors;

[0104] Obtain the ecological restoration effect indicators and ecological characteristics of the hierarchical area, and construct a Least Absolute Shrinkage and Selection Operator (Lasso) regression model based on the ecological restoration effect indicators and ecological characteristics;

[0105] In the Lasso regression model, take the sum of the squares of the standardized coefficients of the ecological characteristics as the sensitivity coefficient, obtain the non-zero frequency of the weights of the ecological characteristics in the Lasso regression model, and calculate the importance of the ecological characteristics:

[0106]

[0107] Among them, the sensitivity coefficient of the \(w\)th ecological characteristic is \(\chi\) w , the non-zero frequency of the \(w\)th ecological characteristic is \(F\) w , the importance of the \(w\)th ecological characteristic is \(\varphi\) w ;

[0108] Calculate the correlation between ecological characteristics using the Pearson correlation coefficient, take the ecological characteristics with a correlation greater than 0.673 as key characteristics, use the elbow method to determine the optimal number of clusters of key characteristics, and divide the hierarchical area into multiple ecological homogeneous groups based on the relationship characteristics;

[0109] Randomly sample the ecological homogeneous groups according to a proportion, use traditional ecological indicators for random sampling, set a control group with the same area, obtain the ecological environment quality scores, soil quality indices, and vegetation coverage of the ecological homogeneous groups and the control group, and calculate the ecological restoration effects of the ecological homogeneous groups and the control group:

[0110]

[0111] Among them, the ecological restoration effect of the \(c\)th group is The experimental area of the \(c\)th group is \(S\) c , the vegetation coverage at the \(s2\) moment is The vegetation coverage at the \(s1\) moment is The ecological environment quality score at the \(s2\) moment is The ecological environment quality score at the \(s1\) moment is The soil quality index at the \(s2\) moment is The soil quality index at the \(s1\) moment is The weight coefficients are \(\beta1\), \(\beta2\), and \(\beta3\) respectively;

[0112] Draw the time-series ecological restoration effect reference curve based on ecological characteristics and the time-series ecological restoration effect curve based on key characteristics, and calculate the difference degree between the time-series ecological restoration effect reference curve and the time-series ecological restoration effect curve:

[0113]

[0114] where the number of features is N c and the i-th feature is b i and the ecological restoration effect of feature b at time s2 i is the reference of the ecological restoration effect of feature b at time s1 i is the difference degree of the c-th ecological homogeneous group is K c ;

[0115] Take the ratio of the difference degree of the sampling result and the mean value of the difference degree as the coefficient of variation. When the coefficient of variation is greater than 0.279, output the difference between 1 and the coefficient of variation as the confidence level of the ecological restoration effect.

[0116] In this embodiment, the method for obtaining the ecological resilience index by performing elastic capacity analysis based on the confidence level of the ecological restoration effect and the satellite remote sensing data includes:[[]]

[0117] Construct a multi-dimensional ecological index system based on satellite remote sensing data, and refine the indicators based on the 4R theory to screen the elastic evaluation indicators; the screening includes: extracting the indicators of key ecological functions according to the types of damage risks, eliminating redundant indicators according to the substitutability of species in the ecosystem, and screening intelligent indicators according to the self-regulation and mobilization and restoration ability when facing damage risks.

[0118] Divide the ecological resilience change process into a preparation stage, a resistance and absorption stage, an adaptive regulation stage, and a recovery stage according to time, and construct an ecological resilience comprehensive measurement model based on the performance indicators of each stage of the ecosystem and the refined indicators of the 4R theory.

[0119] Calculate the ecological resilience index:[[]]

[0120]

[0121] where the ecological resilience index of the k-th classification area is the number of ecological resilience evaluation indicators is M r and the weight of the r-th indicator in the t-th stage is ρ r (t), the elastic evaluation indicator in the t-th stage is f r (t), the preparation stage is t1, the recovery stage is t4, and the confidence level of the ecological restoration effect of the k-th classification area in the t-th stage is the weight of the indicators in the preparation stage t1 is ρ r (t1), the elastic evaluation indicator in the preparation stage t1 is f r (t1), and the confidence level of the ecological restoration effect of the k-th classification area in the preparation stage t1 is

[0122] In this embodiment, a method for constructing a time-varying ecological restoration effect evaluation model based on the ecological resilience index includes:

[0123] Obtain the ecological performance evaluation indicators and corresponding weights of the research area, and construct an objective function according to the ecological resilience index and the loss function. The expression is:

[0124]

[0125] where the loss function is The initial ecosystem performance of the kth hierarchical area is HO k , the number of hierarchical areas in the research area is M k , the weight of the rth indicator at the s-th moment is η r (s), the ecological performance evaluation indicator at the s-th moment is g r (s), the ecosystem performance of the kth hierarchical area at the s-th moment is H k (s), the ecological resilience index of the kth hierarchical area at the s-th moment is The objective function at the s-th moment is

[0126] The time-varying ecological restoration effect evaluation model includes a random forest algorithm, a time series analysis algorithm, and a spatio-temporal graph neural network;

[0127] The random forest algorithm conducts ensemble learning on multiple decision trees of satellite remote sensing data, constructs multiple decision trees by randomly selecting features and samples, and determines the classification result through a voting mechanism;

[0128] The time series analysis algorithm, an algorithm based on spatio-temporal spectral features, extracts the time, space, and spectral features of pixel points, combines a support vector machine classifier for change detection, and extracts the temporal change law of surface features according to the change detection;

[0129] The spatio-temporal graph neural network constructs a spatio-temporal graph according to the temporal change law, uses the spatio-temporal modules of the graph neural network to model spatial and temporal dependencies respectively, predicts the ecological restoration effect according to the dependencies, and outputs the predicted ecological restoration effect.

[0130] Second, an ecological restoration effect evaluation system based on satellite remote sensing includes:

[0131] A data acquisition module: used to collect satellite remote sensing data and field soil data of a preset area, and use the remote sensing data of the preset area within a specified time as the data to be analyzed; the data to be analyzed includes vegetation information, land use and land cover data, topographic and geomorphic data, hydrological data, thermal environment data, and biodiversity data;

[0132] Partition and grading module: used to calculate the temporal deviation based on the satellite remote sensing data, take the preset area with the temporal deviation greater than 0 as the research area, and vice versa as the non-research area, and obtain the graded area by grading the changes within the research area according to the contrast;

[0133] Correction and prediction module: within the graded area, correct the deviation of the satellite remote sensing data according to the field soil data, and input the corrected satellite remote sensing data into the spatio-temporal ecological development prediction model to obtain ecological development prediction data;

[0134] Confidence elasticity module: used to perform uncertainty analysis of restoration based on the ecological development prediction data and the field soil data to obtain the confidence of ecological restoration effect, and perform elasticity ability analysis based on the confidence of ecological restoration effect and the satellite remote sensing data to obtain the ecological elasticity index;

[0135] Modeling and output module: used to construct a time-varying ecological restoration effect evaluation model according to the ecological elasticity index, input the data to be analyzed into the time-varying ecological restoration effect evaluation model, and output the evaluation result.

[0136] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. An ecological restoration effect evaluation method based on satellite remote sensing, characterized in that, The following steps are involved: Collect satellite remote sensing data and field soil data of a preset area, and use the remote sensing data of the preset area within a specified time as data to be analyzed; the data to be analyzed includes vegetation information, land use and land cover data, topography data, hydrological data, thermal environment data, and biodiversity data; Calculating the time series deviation based on the satellite remote sensing data, taking the preset area where the time series deviation is greater than 0 as the study area, and otherwise as the non-study area, and classifying the changes in the study area according to the contrast to obtain a graded area; comprising: Calculate timing deviation: where the timing deviation is θ and the upper limit of the monitoring time is T s , the a-th remote sensing data at the s2-th moment is h a (s2), the a-th remote sensing data at the s1-th moment is h a (s1), the starting moment of monitoring is s o , the altitude of the remote sensing satellite is D, the radius of the earth is R, and the viewing angle is performing deviation correction on the satellite remote sensing data according to the field soil data in the grading area, and inputting the corrected satellite remote sensing data into a spatiotemporal ecological development prediction model to obtain ecological development prediction data; Performing restoration uncertainty analysis based on the ecological development forecast data and the field soil data to obtain an ecological restoration effect confidence level, and performing elasticity analysis based on the ecological restoration effect confidence level and the satellite remote sensing data to obtain an ecological resilience index; A time-varying ecological restoration effect evaluation model is constructed based on the ecological resilience index, the data to be analyzed is input into the time-varying ecological restoration effect evaluation model, and an evaluation result is output.

2. The ecological restoration effect evaluation method based on satellite remote sensing according to claim 1, wherein The method for obtaining graded regions by grading changes in the study area according to contrast comprises: The subgrids were obtained by dividing the grid according to the size and shape of the study area. The change features within the subgrids were extracted according to the change ratio of the remote sensing satellite data. The change features with a change ratio greater than 0.013 were taken as key features. Taking the subgrid as a unit, a subgrid-based clustering feature tree is constructed according to the contrast and key features. The key features are used as clustering feature nodes, and the abnormality degree index of the clustering feature nodes is calculated. The abnormality index is input into the student psychology optimization algorithm, abnormal clustering feature nodes are eliminated according to the optimization characteristics, the optimized key feature set is obtained, the clustering feature nodes in the optimized key feature set are integrated and adjusted, and the clustering feature tree is updated; Calculate the contrast of the subgrid based on the temporal and spatial changes according to the updated cluster feature tree: where the contrast of the w-th sub-grid is l w , the number of key features in the optimized key feature set within the w-th sub-grid is N w , the v-th key feature at the s2 moment is H v (s2), the v-th discriminative feature at the s1 moment is H v (s1), the weight coefficient of the v-th key feature is the sign function is sgn(·); When the contrast is less than 0, the subgrid is a poor-quality restoration area; when the contrast is greater than 0 and less than 0.0051, the subgrid is a poor-quality restoration area; when the contrast is greater than 0.0051 and less than 0.015, the subgrid is a medium-quality restoration area; when the contrast is greater than 0.015 and less than 0.0297, the subgrid is a good-quality restoration area; when the contrast is greater than 0.0297, the subgrid is an excellent-quality restoration area. Adjacent subgrids of the same level are connected according to the contrast, and the connected results are output as the graded area.

3. The ecological restoration effect evaluation method based on satellite remote sensing according to claim 1, characterized in that The method of the spatiotemporal ecological development prediction model comprises: Collect climate and environmental data within the preset area, and perform time series impact analysis on the climate and environmental data to obtain environmental factors; Spatiotemporal ecological development prediction models include graph networks, time-guided embeddings, and long short-term memory neural networks; The graph network represents pixels, regions, or objects in remote sensing data as graph nodes, represents the relationships between graph nodes as edges, uses a graph neural network architecture to capture the complex relationships between graph nodes, and extracts the complex spatio-temporal features of remote sensing data based on the complex relationships; The time-guided embedding integrates time information into the embedding vector by learning the spatio-temporal ecological development, and captures the time cycle pattern of ecological development in combination with the complex spatio-temporal features; The long short-term memory neural network captures the long-term dependencies in the time series through the gating mechanism and the cell state, and predicts the ecological development based on the time cycle pattern and the long-term dependencies.

4. The ecological restoration effect evaluation method based on satellite remote sensing according to claim 1, wherein A method for obtaining the confidence level of ecological restoration effect through the restoration uncertainty analysis based on the ecological development prediction data and the field soil data includes: Converting the ecological development prediction data and the field soil data into multi-dimensional vectors, and constructing an ecological feature matrix based on the multi-dimensional vectors; Obtaining the ecological restoration effect indicators and ecological features of the hierarchical regions, and constructing a least absolute shrinkage and selection operator Lasso regression model based on the ecological restoration effect indicators and ecological features; In the Lasso regression model, taking the sum of the squares of the standardized coefficients of the ecological features as the sensitivity coefficient, obtaining the non-zero frequency of the weights of the ecological features in the Lasso regression model, and calculating the importance of the ecological features: where the sensitivity coefficient of the w-th ecological feature is χ w , the non-zero frequency of the w-th ecological feature is F w , the importance of the w-th ecological feature is φ w ; Calculating the correlation between ecological features using the Pearson correlation coefficient, taking the ecological features with a correlation greater than 0.673 as key features, determining the optimal number of clusters of the key features using the elbow method, and dividing the hierarchical regions into multiple ecological homogeneous groups based on the relationship features; Randomly sampling the ecological homogeneous groups according to a proportion, randomly sampling using traditional ecological indicators, setting a control group with the same area, obtaining the ecological environment quality scores, soil quality indices, and vegetation coverage of the ecological homogeneous groups and the control group, and calculating the ecological restoration effects of the ecological homogeneous groups and the control group: Among them, the ecological restoration effect of the c-th group is The experimental area of the c-th group is S c , and the vegetation coverage at the s2 moment is The vegetation coverage at the s1 moment is The ecological environment quality score at the s2 moment is The ecological environment quality score at the s1 moment is The soil quality index at the s2 moment is The soil quality index at the s1 moment is The weight coefficients are β1, β2, and β3 respectively; Drawing a time-series ecological restoration effect reference curve based on ecological features and a time-series ecological restoration effect curve based on key features, and calculating the difference degree between the time-series ecological restoration effect reference curve and the time-series ecological restoration effect curve; where the number of features is N c , the i-th feature is b i , the ecological restoration effect of feature b at time s2 i is The reference for the ecological restoration effect of feature b at time s1 i is The difference degree of the c-th ecological homogeneous group is K c ; Taking the ratio of the difference degree of the sampling results to the mean of the difference degrees as the coefficient of variation, and when the coefficient of variation is greater than 0.279, outputting the difference between 1 and the coefficient of variation as the confidence level of the ecological restoration effect.

5. The ecological restoration effect evaluation method based on satellite remote sensing according to claim 1, characterized in that, A method for obtaining an ecological resilience index through the resilience ability analysis based on the confidence level of the ecological restoration effect and the satellite remote sensing data includes: Constructing a multi-dimensional ecological index system based on the satellite remote sensing data, and screening the multi-dimensional ecological indices based on the refined indices of the 4R theory to obtain the resilience evaluation indices; the screening includes: extracting the indices of key ecological functions according to the types of damage risks, removing redundant indices according to the substitutability of species in the ecosystem, and screening intelligent indices according to the self-regulation and mobilization and restoration abilities when facing damage risks; Dividing the ecological resilience change process into a preparation stage, a resistance and absorption stage, an adaptive regulation stage, and a recovery stage according to time, and constructing an ecological resilience comprehensive measurement model based on the performance indices of each stage of the ecosystem and the refined indices of the 4R theory; Calculating the ecological resilience index: where the ecological resilience index of the k-th classification area is The number of ecological resilience assessment indicators is M r , the weight of the r-th indicator in the t-th stage is ρ r (t), the resilience assessment indicator in the t-th stage is f r (t), the preparation stage is t1, the recovery stage is t4, and the confidence level of the ecological restoration effect of the k-th classification area in the t-th stage is The indicator weight in the preparation stage t1 is ρ r (t1), the resilience assessment indicator in the preparation stage t1 is f r (t1), and the confidence level of the ecological restoration effect of the k-th classification area in the preparation stage t1 is 6. The ecological restoration effect evaluation method based on satellite remote sensing according to claim 1, characterized in that, A method for constructing a time-varying ecological restoration effect evaluation model based on the ecological resilience index, comprising: Obtaining the ecological performance evaluation indicators and corresponding weights of the research area, and constructing an objective function according to the ecological resilience index and the loss function. The expression is: where the loss function is The initial ecosystem performance of the k-th hierarchical region is HO k , the number of hierarchical regions in the study area is M k , the weight of the r-th index at the s-th moment is η r (s), the ecological performance evaluation index at the s-th moment is g r (s), the ecosystem performance of the k-th hierarchical region at the s-th moment is H k (s), the ecological resilience index of the k-th hierarchical region at the s-th moment is The objective function at the s-th moment is The time-varying ecological restoration effect evaluation model includes a random forest algorithm, a time series analysis algorithm, and a spatio-temporal graph neural network; The random forest algorithm performs ensemble learning on multiple decision trees of satellite remote sensing data, constructs multiple decision trees by using randomly selected features and samples, and determines the classification result through a voting mechanism; The time series analysis algorithm based on spatio-temporal spectral features extracts the time, space, and spectral features of pixel points, combines a support vector machine classifier for change detection, and extracts the temporal change law of surface features according to the change detection; The spatio-temporal graph neural network constructs a spatio-temporal graph according to the temporal change law, uses the spatio-temporal module of the graph neural network to model the spatial and temporal dependence relationships respectively, predicts the ecological restoration effect according to the dependence relationships, and outputs the predicted ecological restoration effect.

7. An ecological restoration effect evaluation system based on satellite remote sensing for implementing the method according to any one of claims 1-6, characterized in that, Including: Data acquisition module: used to collect satellite remote sensing data and field soil data of a preset area, and use the remote sensing data of the preset area within a specified time as the data to be analyzed; The data to be analyzed includes vegetation information, land use and land cover data, topographic and geomorphic data, hydrological data, thermal environment data, and biodiversity data; Partitioning and grading module: used to calculate the temporal deviation based on the satellite remote sensing data, use the preset area with the temporal deviation greater than 0 as the research area, and vice versa as the non-research area, and perform change grading on the research area according to the contrast to obtain a graded area; Correction and prediction module: correct the deviation of the satellite remote sensing data according to the field soil data in the graded area, and input the corrected satellite remote sensing data into the spatio-temporal ecological development prediction model to obtain ecological development prediction data; Confidence elasticity module: used to perform restoration uncertainty analysis according to the ecological development prediction data and the field soil data to obtain the confidence of the ecological restoration effect, and perform elasticity ability analysis according to the confidence of the ecological restoration effect and the satellite remote sensing data to obtain the ecological resilience index; Modeling and output module: used to construct a time-varying ecological restoration effect evaluation model according to the ecological resilience index, input the data to be analyzed into the time-varying ecological restoration effect evaluation model, and output the evaluation result.

Images