A Method for Assessing the Structure and Stability of Regional Habitat Ecological Networks

Abstract

This invention relates to a method for assessing the structure and stability of regional habitat ecological networks. It addresses the lack of existing technologies for dynamically and comprehensively assessing the structure and stability of habitat ecological networks. The method comprises: 1. Constructing a regional habitat ecological network; 2. Assessing the structural characteristics of the regional habitat ecological network; 3. Assessing the connectivity of the regional habitat ecological network; 4. Constructing multiple scenarios for assessing the stability of the regional habitat ecological network; 5. Assessing the resilience stability of the regional habitat ecological network based on natural and anthropogenic disturbance scenarios; 6. Assessing the recovery stability of the regional habitat ecological network based on natural and anthropogenic recovery scenarios. This invention achieves a comprehensive and dynamic assessment of the structure and stability of habitat ecological networks, providing a scientific basis for ecological protection decisions and enhancing the adaptability of habitat ecological networks under natural and anthropogenic disturbances.

Description

TECHNICAL FIELD

[0001] The present application relates to a habitat ecological network structure and a method for evaluating the stability thereof. BACKGROUND

[0002] As an effective tool for protecting biodiversity, the habitat ecological network has been widely applied in the management and restoration of ecosystems. The core of the habitat ecological network is to maintain the structure and function of the ecosystem by enhancing the connectivity of habitats, and to reduce the adverse effects of landscape fragmentation on biodiversity. The structure and stability of the habitat ecological network directly affect the survival ability of the biological community and the resistance of the ecosystem, and play an irreplaceable role in maintaining global biodiversity.

[0003] The existing ecological network evaluation methods have the following shortcomings: first, the ecological network evaluation is limited to the topological characteristics of the network, and often focuses on a single network topological index, evaluating the density and spatial layout of the ecological source and the ecological corridor, and failing to comprehensively evaluate the structure and function of the ecological network. Second, the ecological network evaluation usually uses landscape indexes (landscape area, patch number, etc.) and ecosystem service indexes to evaluate the structure and function, but ignores the interaction between the elements of the ecological network and its influence on connectivity, and ignores the dynamics of material and energy flow in the ecological network. Third, the stability evaluation of the ecological network mostly takes the elasticity or resilience of the network as a single index, and fails to systematically distinguish between resistance and resilience, thereby reducing the scientificity and accuracy of the stability evaluation of the ecological network. SUMMARY

[0004] In order to solve the technical problem of lacking dynamic and comprehensive evaluation of the structure and stability of the habitat ecological network in the prior art, the present application provides a method for evaluating the structure and stability of a regional habitat ecological network, which is a method for dynamically and comprehensively evaluating the structure and stability of a regional habitat ecological network.

[0005] The method for evaluating the structure and stability of a regional habitat ecological network according to the present application is performed according to the following steps:

[0006] Step 1: Construct a regional habitat ecological network, the structural elements of the ecological network including ecological sources and ecological corridors;

[0007] Step 2: Evaluate the structural characteristics of the regional habitat ecological network, calculate the comprehensive centrality of the habitat patch, the important priority of the ecological corridor, and the compactness of the network structural elements;

[0008] Step 3: Evaluate the connectivity of the regional habitat ecological network, calculate the connectivity index of the habitat patch, the unobstructed index of the ecological corridor, and the connectivity of the network structure;

[0009] Step four, construct the regional habitat ecological network stability assessment scenario, set the interference, recovery scenario mode, construct the natural and human intervention strategy, simulate the stability state of the habitat ecological network under different scenarios;

[0010] Step five, based on the natural disturbance scenario and the human disturbance scenario, the resistance stability of the regional habitat ecological network is evaluated, and the ability of the habitat ecological network to maintain its structure and function when it is destroyed is quantified;

[0011] Step six, based on the natural recovery scenario and the human recovery scenario, the recovery stability of the regional habitat ecological network is evaluated, and the potential ability of the habitat ecological network to recover its structure and function after being destroyed is quantified.

[0012] In a single scenario, through the parameter setting of the intervention strategy, the change process of the habitat ecological network stability can be evaluated from multiple angles such as the number, proportion and importance of habitat patches or ecological corridors. This not only considers the dynamic scenario of the habitat ecological network due to changes in external environmental conditions, but also further analyzes the internal ecological process of the change of the habitat ecological network stability in a single scenario. The present application makes up for the shortcomings of the existing habitat ecological network evaluation technology, provides a multi-level stability evaluation framework refined to individual patches and corridors, and more comprehensively reveals the ecological dynamics within the network.

[0013] The present application proposes a dynamic and comprehensive regional habitat ecological network structure and its stability evaluation method, which innovatively combines multi-scenario simulation to realize quantitative analysis of the network structure, connectivity and stability, and provides decision support for biodiversity protection and restoration and sustainable management of ecological systems. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 is a flowchart of example 1;

[0015] Figure 2 is a regional habitat ecological network construction result graph of example 1;

[0016] Figure 3 is a regional habitat ecological network structure feature evaluation result graph of example 1;

[0017] Figure 4 is a regional habitat ecological network connectivity result graph of example 1;

[0018] Figure 5 is a regional habitat ecological network stability assessment scenario graph of example 1;

[0019] Figure 6 is a regional habitat ecological network resistance stability and recovery stability result graph of example 1. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without creative work fall within the scope of the present application.

[0021] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict.

[0022] Specific implementation method one: the regional habitat ecological network structure and the stability evaluation method thereof in the present embodiment are performed according to the following steps:

[0023] Step one, constructing a regional habitat ecological network, the structural elements of the ecological network including ecological sources and ecological corridors;

[0024] Step two, evaluating the structural characteristics of the regional habitat ecological network, calculating the comprehensive centrality of habitat patches, the important priority of ecological corridors and the compactness of network structural elements;

[0025] Step three, evaluating the connectivity of the regional habitat ecological network, calculating the connectivity index of habitat patches, the unobstructed index of ecological corridors and the connectivity of network structure;

[0026] Step four, constructing a regional habitat ecological network stability evaluation scenario, setting interference and recovery scenario modes, constructing natural and artificial intervention strategies, and simulating the stability state of the habitat ecological network under different scenarios;

[0027] Step five, based on the natural interference scenario and the artificial interference scenario, evaluating the resistance stability of the regional habitat ecological network, and quantifying the ability of the habitat ecological network to maintain its structure and function when it is destroyed;

[0028] Step six, based on the natural recovery scenario and the artificial recovery scenario, evaluating the recovery stability of the regional habitat ecological network, and quantifying the potential ability of the habitat ecological network to recover its structure and function after being destroyed.

[0029] In step one of the present embodiment, the regional habitat ecological network is constructed, and the structural elements thereof include ecological sources and ecological corridors. The ecological sources are extracted based on habitat area, habitat structure, habitat quality, habitat sensitivity and habitat protection planning; the influence of natural environmental changes and human disturbance is comprehensively considered to construct a comprehensive migration resistance surface, and then the habitat ecological network is constructed by coupling the minimum cost corridor method, the circuit theory and the moving window analysis method.

[0030] In step two of the embodiment, the habitat ecological network is extracted as the network structure elements: nodes and edges, the structural characteristics of the habitat ecological network are analyzed, the comprehensive centrality of habitat patches, the important priority of ecological corridors and the compactness of network structure elements are calculated, and the structural characteristics and complexity of the habitat ecological network are determined.

[0031] In step three of the embodiment, the connectivity of the habitat ecological network directly affects the reproduction, survival of species and sustainability of ecosystem services; based on the structure of the habitat ecological network, the connectivity index of habitat patches and the unobstructed index of ecological corridors are calculated, and the overall connectivity of the habitat ecological network is measured according to the structural relationship of the habitat ecological network, and then the possibility of species migration and gene exchange in the ecological network is analyzed, and the efficiency of species migration and gene exchange of the habitat ecological network is quantified from a macro perspective.

[0032] In step four of the embodiment, the regional habitat ecological network stability evaluation scenarios are constructed, in order to represent the influence of natural changes and human factors on the habitat ecological network, four evaluation scenario elements of natural disturbance, human disturbance, natural recovery and human recovery are constructed by using double-factor simulation experiment, the stability of regional habitat ecological network under different disturbance and recovery scenarios is quantified, the resistance of habitat ecological network to habitat loss is measured, and the ability of habitat ecological network to restore its structure and function after disturbance is measured.

[0033] In step five of the embodiment, the resistance stability of the regional habitat ecological network is evaluated based on the natural disturbance scenario and the human disturbance scenario, wherein the resistance stability refers to the ability of the habitat ecological network to maintain its normal structure and function when it is destroyed; by randomly or artificially removing habitat patches in the ecological network, the ability of the habitat ecological network to maintain connectivity is quantified, and the resistance stability is calculated; and the artificial removal rules can be set according to the number, proportion and importance of habitat patches or ecological corridors.

[0034] In step six of the embodiment, the recovery stability of the regional habitat ecological network is evaluated based on the natural recovery scenario and the human recovery scenario, and the recovery stability refers to the potential recovery ability of the ecological network itself after the habitat ecological network is destroyed by nature or human beings; by gradually recovering the habitat patches that once existed but have been lost, the recovery ability of the maximum connected subgraph of the ecological network is quantified, and the recovery stability is calculated.

[0035] Specific implementation method two: the difference between this embodiment and specific implementation method one is that in step one, the regional habitat ecological network is constructed: the structural elements include ecological sources and ecological corridors, the ecological sources are extracted based on habitat area, habitat structure, habitat quality, habitat sensitivity and habitat protection planning; the species migration resistance is quantified based on land use, elevation, slope, nitrogen and phosphorus pollution, distance from road and distance from water body; the ecological corridors are generated based on the spatial layout of the ecological sources, the moving threshold is set according to the migration ability of different species, the ecological pinch points and ecological barrier points on the corridors are identified, and the regional habitat ecological network is constructed. The other steps and parameters are the same as those in specific implementation method one.

[0036] Specific implementation method three: the difference between this embodiment and specific implementation method one is that in step two, the comprehensive centrality evaluation of the habitat patch: the degree centrality, the closeness centrality, the eigenvector centrality and the betweenness centrality of the habitat patch are calculated, the dimension effect is eliminated by normalizing, the comprehensive centrality is calculated by equal weight addition, and the importance of the habitat patch is comprehensively evaluated.

[0037] Among them,

[0038] 1) Degree centrality calculation formula: C D (v)=deg(v) (1)

[0039] In the formula, C D (v) is the degree centrality of habitat patch v, and deg(v) is the connection number of habitat patch v.

[0040] 2) Closeness centrality calculation formula:

[0041] In the formula, C C (v) is the closeness centrality of habitat patch v, and d(v, t) is the shortest path length from habitat patch v to habitat patch t.

[0042] 3) Eigenvector centrality calculation formula:

[0043] In the formula, N(v) is all habitat patches connected with habitat patch v, and λ is the maximum eigenvalue of the eigenvector.

[0044] 4) Betweenness centrality calculation formula:

[0045] In the formula, C B (v) is the betweenness centrality of habitat patch v, σ st is the number of shortest paths from habitat patch s to habitat patch t, and σ st (v) is the number of shortest paths through habitat patch v.

[0046] Step two: the important priority evaluation of ecological corridors

[0047] The smaller the resistance index value of the ecological corridor, the smaller the resistance of species migration, the higher the quality of the corridor, and the higher the importance and protection priority of the corridor. The calculation formula is as follows:

[0048]

[0049] In the formula, R(l) is the resistance index of ecological corridor l; Cwd is the cost-weighted distance value of the ecological corridor; and Lcp is the minimum cost path length.

[0050] Step two: the compactness evaluation of network structure

[0051] The average clustering coefficient of the habitat ecological network is calculated to measure the local connectivity between structural elements in the network, reflecting the overall structural characteristics of the network. The calculation formula is as follows:

[0052]

[0053] In the formula, C Cluster (w) is the average clustering coefficient of habitat ecological network w; N is the total number of habitat patches in the network; C v is the clustering coefficient of habitat patch v, indicating the ratio of the actual number of connections between neighbors of habitat patch v to the possible number of connections.

[0054] Specific implementation four: this implementation is different from specific implementation one in that step three: the connectivity index evaluation of habitat patches: the connectivity integral index and the connectivity probability are added with equal weight to calculate the connectivity index of habitat patches;

[0055] The calculation formula of the connectivity integral index is as follows:

[0056] In the formula, C IIC (v) is the connectivity integral index of habitat patch v; N is the total number of habitat patches in the network; P ij is the connection strength between habitat patch i and habitat patch j, which is usually based on habitat quality or similarity; d ij is the distance between habitat patch i and habitat patch j; and k is the attenuation parameter, which determines the degree of influence of distance on connection strength.

[0057] The calculation formula of the connectivity probability is as follows:

[0058] In the formula, C PC (v) is the connectivity probability of habitat patch v; N is the total number of habitat patches in the network; P ij is the connection probability between habitat patch i and habitat patch j.

[0059] The flow index of the ecological corridor in step three is evaluated: the flow and the substitution of the corridor are calculated by grading the ecological barrier points and the pinch points, and the flow index of the ecological corridor is calculated by equally adding the two;

[0060] The flow index of the ecological corridor in step three is evaluated: the flow and the substitution of the corridor are calculated by grading the ecological barrier points and the pinch points, and the flow index of the ecological corridor is calculated by equally adding the two;

[0061]

[0062] The higher the barrier density is, the more barriers exist in the corridor, the more difficult the species migration is, and the lower the flow of the corridor is. The closer the flow of the corridor is to 1, the stronger the flow capacity of the corridor is. The closer the flow of the corridor is to 0, the more serious the barrier of the corridor is.

[0063] The flow index of the ecological corridor in step three is evaluated: the flow and the substitution of the corridor are calculated by grading the ecological barrier points and the pinch points, and the flow index of the ecological corridor is calculated by equally adding the two;

[0064] The flow index of the ecological corridor in step three is evaluated: the flow and the substitution of the corridor are calculated by grading the ecological barrier points and the pinch points, and the flow index of the ecological corridor is calculated by equally adding the two;

[0065] The higher the barrier density is, the more barriers exist in the corridor, the more difficult the species migration is, and the lower the flow of the corridor is. The closer the flow of the corridor is to 1, the stronger the flow capacity of the corridor is. The closer the flow of the corridor is to 0, the more serious the barrier of the corridor is.

[0066] The grading of the ecological pinch points and the barrier points is as follows:

[0067] The first level (the smallest impact area): almost no ecological pinch points or almost no migration barriers;

[0068] The second level (the slightly affected area): the pinch points have a slight impact, or the species migration is affected by some local factors;

[0069] The third level (the moderately affected area): the pinch points are obvious, or the migration path is still passed through by the barrier;

[0070] The fourth level (the severely affected area): the pinch points are obvious, or most of the migration paths are blocked by the barrier;

[0071] The fifth level (the extremely severely affected area): the pinch points are obvious and have a great impact, or the species migration path is completely blocked and even seriously affects the network function;

[0072] The connectivity of the network structure in step three is evaluated: the α index, the β index, the γ index, the C r index and the CC r comprehensive connectivity index are calculated.

[0073] The α index is calculated by the following formula:

[0074] L represents the number of ecological corridors between habitat patches in the network; V represents the number of habitat patches in the network;

[0075] The formula for calculating the β index is:

[0076] L represents the number of ecological corridors between habitat patches in the network; V represents the number of habitat patches in the network;

[0077] The formula for calculating the γ index is:

[0078] The denominator represents the maximum possible number of ecological corridors in a completely connected network. The closer this value is to 1, the closer the network is to complete connectivity, and the lower the migration resistance of species.

[0079] C r The formula for calculating the index is:

[0080] C represents the length of the ecological corridor, and L represents the number of ecological corridors.

[0081] CC r The formula for calculating the comprehensive connectivity index is:

[0082] The other steps and parameters are the same as in the first embodiment.

[0083] Embodiment Five: The difference between this embodiment and the first embodiment is that in step four, the habitat ecological network stability assessment scenarios are based on random and target intervention strategies, representing the influence of natural changes and human factors on the habitat ecological network, setting two intervention modes of disturbance and recovery, using a two-factor simulation experiment to construct four assessment scenario elements of natural disturbance, human disturbance, natural recovery, and human recovery, and a variety of stability assessment scenarios, quantifying the stability of the regional habitat ecological network under different disturbance and recovery scenarios, measuring the resistance of the habitat ecological network to habitat loss and the ability of the habitat ecological network to restore its original structure and function after disturbance;

[0084] The intervention strategies: random intervention removes or increases habitat patches randomly, simulating the state of damaged or restored habitat ecological networks under natural conditions; target intervention removes or increases habitat patches according to the set rules, simulating the state of the habitat ecological network after human intervention; the set rules are: removing or increasing habitat patches according to importance, quantity, and proportion;

[0085] Assessment scenario elements:

[0086] The natural interference scenario adopts a random intervention strategy to remove habitat patches randomly. The artificial interference scenario adopts a target intervention strategy to remove habitat patches according to a removal rule. The natural recovery scenario adopts a random intervention strategy to increase habitat patches that have been lost. The artificial recovery scenario adopts a target intervention strategy to increase habitat patches that have been lost according to an increase rule. Other steps and parameters are the same as in the first embodiment.

[0087] The target intervention setting rule in this embodiment is to remove or increase habitat patches according to importance, quantity, or proportion. Specifically, n patches are selected and numbered according to natural numbers. The patches can be removed or increased according to importance, quantity, or proportion. The setting rule in this embodiment can also remove or increase specific patches with certain numbers.

[0088] The sixth embodiment is different from the first embodiment in that the habitat ecological network resistance stability evaluation in step five is to quantify the ability of the habitat ecological network to maintain connectivity by removing habitat patches in the ecological network randomly or artificially, and to calculate the resistance stability.

[0089] The calculation formula is:

[0090] In the formula, N represents the total number of habitat patches in the habitat ecological network, S0 represents the size of the largest connected subgraph in the initial habitat ecological network, Si represents the size of the largest connected subgraph in the remaining habitat ecological network after the removal of the ith habitat patch, and S i

[0091] In the natural interference scenario in step five, a random intervention strategy is used to simulate the scenario in which the habitat ecological network is disturbed by nature. A random seed is set in each iteration to ensure that the results of each random selection are different. To avoid the randomness of random removal, the average resistance stability value of 100 iteration cycles is used as the final result to reflect the stability of the entire network when the habitat ecological network faces random loss of habitat patches.

[0092] ​​​Step five: human disturbance scenario: adopt target intervention strategy to simulate the scenario that important habitat patches in habitat ecological network are lost due to human disturbance; sort the comprehensive centrality obtained by normalizing and weightedly summing the degree centrality, eigenvector centrality, betweenness centrality and closeness centrality of the habitat patches, remove the habitat patches from high to low according to the comprehensive centrality score, and calculate the resistance stability of the habitat ecological network under the human disturbance scenario. The other steps and parameters are the same as those in embodiment one.

[0093] Embodiment seven: different from embodiment one, in step six, the habitat ecological network recovery stability evaluation: the recovery stability refers to the potential recovery ability of the habitat ecological network after being damaged by nature or human; the recovery ability of the maximum connected subgraph of the ecological network is quantified by gradually recovering the habitat patches that once existed but have been lost, and the recovery stability is calculated.

[0094] The calculation formula is:

[0095] In the formula, N represents the total number of damaged habitat patches in the habitat ecological network; S0 represents the size of the maximum connected subgraph in the complete habitat ecological network; Si represents the size of the maximum connected subgraph after recovering the ith habitat patch; i S represents the size of the maximum connected subgraph after recovering the ith habitat patch; S represents the sum of the sizes of the maximum connected subgraphs in all habitat patch recovery steps;

[0096] Step six: natural recovery scenario:

[0097] Random intervention strategy is adopted to simulate the natural recovery scenario of the habitat ecological network. In each iteration, a random seed is set to ensure that the results of each random selection are different. In order to avoid the randomness of the random increase, the average recovery stability value of 100 iteration cycles is taken as the final result, which reflects the stability of the entire network when the habitat ecological network is naturally recovered.

[0098] Step six: human recovery scenario:

[0099] Target intervention strategy is adopted to simulate the scenario of artificially recovering important habitat patches in the habitat ecological network. The comprehensive centrality obtained by normalizing and weightedly summing the degree centrality, eigenvector centrality, betweenness centrality and closeness centrality of the habitat patches is sorted, and the habitat patches are recovered from high to low according to the comprehensive centrality score, and the recovery stability of the habitat ecological network under the human recovery scenario is calculated. The other steps and parameters are the same as those in embodiment one.

[0100] The following examples are used to verify the effect of the present application:

[0101] The embodiment 1 takes the Songnen Plain habitat ecological network structure and its stability evaluation scheme as an example, comprehensively evaluates the habitat ecological network structure characteristics, connectivity and stability of the region, dynamically quantifies the resistance and resilience of the region in response to natural environmental changes and human disturbances, and an evaluation method of the habitat ecological network structure and its stability of the region is as follows:

[0102] S1. Construct a regional habitat ecological network, and the structural elements of the ecological network include ecological sources and ecological corridors;

[0103] S2. Evaluate the structural characteristics of the regional habitat ecological network, and quantify the comprehensive centrality of habitat patches, the important priority of ecological corridors, and the compactness of network structural elements;

[0104] S3. Evaluate the connectivity of the regional habitat ecological network, and calculate the connectivity index of habitat patches, the unobstructed index of ecological corridors, and the connectivity of the network structure;

[0105] S4. Construct a regional habitat ecological network stability evaluation scenario, set the disturbance and recovery scenario mode, construct the natural and human intervention strategy, and simulate the stability state of the habitat ecological network under different scenarios.

[0106] S5. Based on the natural disturbance scenario and the human disturbance scenario, evaluate the resistance stability of the regional habitat ecological network, and quantify the ability of the habitat ecological network to maintain its structure and function when it is destroyed.

[0107] S6. Based on the natural recovery scenario and the human recovery scenario, evaluate the recovery stability of the regional habitat ecological network, and quantify the potential ability of the habitat ecological network to restore its structure and function after being destroyed.

[0108] The flowchart of the embodiment is shown in Figure 1 .

[0109] Step S1 constructs a regional habitat ecological network, and the structural elements thereof include ecological sources and ecological corridors. Ecological sources are extracted based on habitat area, habitat structure, habitat quality, habitat sensitivity and habitat protection planning; and the comprehensive resistance surface of species migration is constructed by comprehensively considering the influence of natural environmental changes and human disturbances, and then the habitat ecological network is constructed by coupling the minimum cost corridor method, the circuit theory and the moving window analysis method.

[0110] Based on the vegetation coverage, air temperature, precipitation, growth season drought index, land use and protected area planning data, the key habitat areas are extracted as ecological sources; based on the land use, elevation, slope, nitrogen and phosphorus pollution, distance from road and distance from water data, the species migration resistance is quantified; based on the spatial layout of the ecological sources, the moving threshold is set according to the migration ability of different species, the ecological corridors are generated, the ecological pinch points and ecological barrier points on the corridors are identified, and the regional habitat ecological network is constructed.

[0111] Figure 2 This is a diagram showing the results of the regional habitat ecological network construction in this embodiment. Through the assessment of habitat characteristics within the region, ecological patches with the highest habitat quality, of high importance and sensitive to climate change, contiguous landscape core areas, and areas implementing ecological protection policies were selected as ecological source areas. This allows for the accurate identification of the most suitable habitats for species and the origin and destination areas of their migration behaviors, such as... Figure 2 As shown in Figure a, there are 36 ecological source areas in the Songnen Plain, accounting for 19.02% of the total area. These areas are mainly distributed in the eastern forests and along the Nenjiang River. These forests, water bodies, and wetlands provide important habitats for the survival and reproduction of aquatic, amphibian, mammal, and bird species in the Songnen Plain. Based on environmental conditions affecting species migration (topographical changes, land use types, road injury risks, and pollution emission risks), a comprehensive resistance surface for species migration in the Songnen Plain can be constructed. This allows for the scientific quantification of the migration difficulty of different species within the region. Figure 2 As shown in b, urban expansion (such as in Changchun and Harbin) and road construction (such as the Beijing-Harbin Expressway and railway) have the most significant impacts on species migration. Farmland expansion and soil pollution are also factors that hinder species migration. Based on the migration corridor lengths of amphibians (0.5-5 km), small and medium-sized mammals (5-20 km), and birds (100-1000 km) and the area of ​​regional habitats, ecological corridors can be identified. Furthermore, crowded areas (ecological pinch points) and obstructive areas (ecological barrier points) within these corridors can be identified. This allows for a hierarchical consideration of the migration characteristics of different species, enabling the construction of a more realistic regional habitat ecological network. Figure 2 As shown in Figure c, there are 80 ecological corridors in the Songnen Plain, covering most of the plain. This indicates that organisms in the region can access most habitat patches through these corridors, obtaining sufficient resources and facilitating effective information and gene exchange. There are 78 ecological choke points along the ecological corridors, mainly distributed at the corridor endpoints, narrow sections of large habitat patches, and fragmented habitat patches. They are particularly densely distributed around the Zhalong and Chagan Lake Nature Reserves, indicating that nature reserves and good habitat patches are frequent transit points for species. There are 96 ecological barrier points along the ecological corridors, mainly distributed in cultivated land, construction land, unused land, and around towns between habitat patches, indicating that these areas hinder species migration. By considering the combined effects of natural changes and human disturbances, this method extracts ecological source areas by integrating multiple habitat characteristics, constructs migration resistance surfaces by combining the behaviors of multiple species, sets thresholds based on the migration capabilities of multiple species, and identifies ecological corridors, ecological pinch points, and ecological barrier points. This enables the accurate and scientific construction of regional habitat ecological networks from the perspective of habitat characteristics and species behavior, thus overcoming the shortcomings of existing construction technologies that do not consider the interaction between natural and human factors from the perspective of habitat characteristics and species behavior.

[0112] Step S2 extracts the ecological sources and ecological corridors of the habitat ecological network as the elements of the network structure: nodes and edges, analyzes the structural characteristics of the habitat ecological network, calculates the comprehensive centrality of the habitat patches, the important priority of the ecological corridors, and the compactness of the network structure elements, to understand the structural complexity of the habitat ecological network.

[0113] (1) Habitat patch comprehensive centrality evaluation

[0114] Considering multiple centrality indicators, the degree centrality, closeness centrality, eigenvector centrality, and betweenness centrality of habitat patches are calculated, normalized to eliminate the influence of dimensions, and added equally to calculate the comprehensive centrality, which comprehensively evaluates the importance of habitat patches and more accurately identifies key habitat patches that have a greater impact on network structure and function.

[0115] 1) Degree centrality: Degree centrality reflects the number of connections of each habitat patch in the habitat ecological network, i.e., the number of direct connections of a habitat patch to other habitat patches. Habitat patches with high degree centrality are directly connected to many other habitat patches and are generally considered key habitat patches in the habitat ecological network. The calculation formula is as follows:

[0116] C D (v) = deg(v) (1)

[0117] In the formula, C D (v) is the degree centrality of habitat patch v, and deg(v) is the number of connections of habitat patch v.

[0118] 2) Closeness centrality: Closeness centrality evaluates the average distance of a habitat patch to all other habitat patches in the network, measuring the ease of reaching other habitat patches. The higher the closeness centrality, the faster the material and energy flow between the habitat patch and other habitat patches in the network. The calculation formula is as follows:

[0119]

[0120] In the formula, C C (v) is the closeness centrality of habitat patch v, and d(v, t) is the shortest path length from habitat patch v to habitat patch t.

[0121] 3) Eigenvector centrality: Eigenvector centrality not only considers the direct number of connections of a habitat patch, but also considers the importance of the habitat patches connected to it, reflecting the comprehensive influence of the habitat patch. The calculation formula is as follows:

[0122]

[0123] where N(v) is all habitat patches connected with habitat patch v; λ is the largest eigenvalue of the eigenvector.

[0124] 4) Intermediate centrality: Intermediate centrality measures the bridge role of a habitat patch in the network, representing the number of shortest paths passing through the habitat patch, measuring the number of times a species must pass through the habitat patch to reach another habitat patch in a path. Habitat patches with high intermediate centrality control the flow of information or resources between many habitat patches, and are "bridge" patches in the network. The calculation formula is as follows:

[0125]

[0126] where C B (v) is the intermediate centrality of habitat patch v; σ st is the number of shortest paths from habitat patch s to habitat patch t; σ st (v) is the number of shortest paths passing through habitat patch v.

[0127] (2) Evaluation of the importance priority of ecological corridors

[0128] The ratio of the cost-weighted distance of the ecological corridor to the minimum cost path length can effectively indicate their contribution to the overall network species mobility. The smaller the resistance index value of the ecological corridor, the smaller the resistance to migration, the higher the quality of the corridor, and the higher the importance and protection priority of the corridor. The calculation formula is as follows:

[0129]

[0130] where R(l) is the resistance index of ecological corridor l; Cwd is the cost-weighted distance value of the ecological corridor; Lcp is the minimum cost path length.

[0131] (3) Evaluation of network structure compactness

[0132] Calculate the average clustering coefficient of the habitat ecological network to measure the local connectivity between structural elements in the network, reflecting the overall structural characteristics of the network, and better interpret the overall structure of the habitat ecological network. The calculation formula is as follows:

[0133]

[0134] where C Cluster (w) is the average clustering coefficient of habitat ecological network w; N is the total number of habitat patches in the network; C v is the clustering coefficient of habitat patch v, representing the ratio of the number of actual connections between neighbors of habitat patch v to the number of possible connections.

[0135] Figure 3Fig. 4 is a diagram of the evaluation results of the structural characteristics of the regional habitat ecological network in this embodiment. From the perspective of the importance of nodes, the comprehensive centrality evaluation results of the habitat patches show that the 20th and 30th habitat patches of the Songnen Plain have the highest comprehensive centrality (>0.8), and are in the most important position among all habitat patches, and the comprehensive centrality of the other 34 habitat patches is relatively balanced (about 0.3). From the perspective of the importance of edges, the priority results of the ecological corridors show that the important priority of the 80 ecological corridors of the Songnen Plain fluctuates within a certain interval, and except for the 71st corridor, the important priority of each corridor is not as extreme as the importance results of the habitat patches. From the perspective of the structural relationship between nodes and edges, the structural compactness results of the habitat ecological network show that the habitat patches and ecological corridors of the Songnen Plain are distributed compactly, and the average clustering coefficient is close to 0.7, indicating that the habitat patches and ecological corridors in the network form a relatively close group, realizing the comprehensive evaluation of the structural characteristics of the regional habitat ecological network from the perspectives of structural elements and structural whole, clarifying the importance distribution difference of the habitat patches and ecological corridors and their structural relationship, and revealing the complexity of the habitat ecological network structure.

[0136] Step S3 The connectivity of the habitat ecological network directly affects the reproduction, survival of species, and sustainability of ecological service functions. Based on the results of the habitat ecological network, the connectivity index of the habitat patch and the patency index of the ecological corridor are calculated, the overall connectivity of the habitat ecological network is measured according to the structural relationship of the habitat ecological network, and the possibility of species migration and gene exchange in the ecological network is analyzed, quantifying the efficiency of species migration and gene exchange of the habitat ecological network from a macro perspective.

[0137] (1) Habitat patch connectivity index evaluation

[0138] The connectivity integral index and the connectivity probability based on graph theory can indicate the connectivity of each habitat patch and its importance to landscape connectivity, and the connectivity index of the habitat patch is obtained by equally adding the two indices, which evaluates the migration potential of species between habitat patches, promotes gene flow, and reduces the risk of isolated populations.

[0139] 1) Connectivity integral index: The connectivity integral index comprehensively considers the connection between all habitat patches in the network, and is an important index for quantifying the connectivity of habitat patches in the ecological network, with a value range of 0-1, and the larger the value, the better the connectivity. The calculation formula is as follows:

[0140]

[0141] In the formula, C IIC (v) is the connectivity integral index of habitat patch v; N is the total number of habitat patches in the network; P ijis the connection strength between habitat patch i and habitat patch j, usually based on habitat quality or similarity; d ij is the distance between habitat patch i and habitat patch j; k is the decay parameter, determining the degree of influence of distance on connection strength.

[0142] 2) Connectivity probability: Connectivity probability reflects the connection probability between any two habitat patches in the network, which can evaluate the accessibility of habitats and the possibility of biological migration. Connectivity probability considers a more comprehensive connection model, which is not affected by adjacent habitat patches when analyzing data, ranging from 0 to 1 and increasing with the improvement of connectivity. The calculation formula is as follows:

[0143]

[0144] In the formula: C PC (v) is the connectivity probability of habitat patch v; N is the total number of habitat patches in the network; P ij is the connection probability between habitat patch i and habitat patch j.

[0145] (2) Ecological corridor passability index evaluation

[0146] The passability of ecological corridors has an important influence on the structure and function of ecological networks. The passable ecological corridors promote the migration of species between habitats, enhance gene flow, reduce the risk of isolated populations, and maintain and improve biodiversity. When the habitat ecological network is disturbed or destroyed, the passable ecological corridors can accelerate the recovery process and promote the recovery of network structure and function. From the state of biological flow of ecological corridors (congestion state and blockage state), the passability of ecological corridors is evaluated, and the passability index of ecological corridors is obtained by equal-weighting the flow-through and corridor substitution.

[0147] 1) Corridor flow-through: The flow-through of the corridor reflects the efficiency and smoothness of species migration through the ecological corridor. Corridors with high flow-through can effectively connect habitat patches and allow species to migrate freely, while corridors with low flow-through will be affected by obstacles and limit species migration. Focus on corridors with high barrier density, which can be reduced by ecological restoration or pathway optimization to reduce their impact on species migration. The calculation formula is as follows:

[0148] T = 1 - Barrier density (9)

[0149]

[0150] In the formula, the higher the barrier density, the more obstacles there are in the corridor, the more difficult it is for species to migrate, and the lower the flow-through of the corridor. The closer the corridor flow-through is to 1, the stronger the flow-through ability of the corridor, and the closer to 0 indicates that the corridor is severely blocked.

[0151] 2) Corridor substitutability: Substitutability of a corridor refers to whether other corridors can provide the same function when a certain corridor fails, maintaining the migration and gene flow of species. By analyzing the substitutability of corridors, we can continue to maintain connectivity through alternative corridors when key corridors fail, reducing the vulnerability of the ecosystem.

[0152] R = 1 - pinch point density (11)

[0153]

[0154] In the formula, the higher the pinch point density, the lower the substitutability of the corridor, and the function is more easily constrained by a single corridor; the substitutability value tends to 0, indicating that the corridor has low functional redundancy; and the substitutability tends to 1, indicating that the network has high redundancy.

[0155] 3) Classification of ecological pinch points and barrier points: In order to assess the connectivity and substitutability in more detail, we classified the ecological pinch points and barrier points into five levels according to their impact, with the levels representing different degrees of impact from low to high. We calculated the ratio of the area of the fifth level to the total area to quantify the impact of the density on the overall connectivity of the corridor.

[0156] First level (minimal impact area): almost no pinch points or almost no migration barriers;

[0157] Second level (slight impact area): pinch points have a slight impact, or species migration is affected by some local factors;

[0158] Third level (moderate impact area): pinch points are more obvious, or the migration path is still affected by barriers;

[0159] Fourth level (heavy impact area): pinch points are obvious, or most of the migration paths are blocked by barriers;

[0160] Fifth level (extremely heavy impact area): pinch points are obvious and have a large impact, or the species migration path is completely blocked and even severely affects the network function.

[0161] (3) Network structure connectivity assessment

[0162] Considering the number, distribution, aggregation degree, and connectivity of habitat patches and ecological corridors in the ecological network structure, we used the α index, β index, and γ index to quantitatively describe the overall connectivity of the habitat ecological network from three aspects: the circulation relationship, spatial location, and connection efficiency of the habitat ecological network structure.

[0163] 1) Alpha index: The alpha index is used to describe the number of independent cycles in the habitat ecological network, indicating the complexity of the internal loops in the habitat ecological network. The more loops in the habitat ecological network, the more redundant paths there are within the habitat ecological network, which is important for improving the migration ability of species. The calculation formula is:

[0164]

[0165] In the formula, L represents the number of ecological corridors between habitat patches in the network; V represents the number of habitat patches in the network.

[0166] 2) Beta index: The beta index reflects the density of the network, which is used to measure the density of connections between habitats, and to determine whether some habitats are too isolated or lack interconnectivity. The formula is:

[0167]

[0168] In the formula, L represents the number of ecological corridors between habitat patches in the network; V represents the number of habitat patches in the network.

[0169] 3) Gamma index: The gamma index measures the ratio of the actual number of ecological corridors in the network to the maximum number of ecological corridors that could exist in the network, assessing whether the connectivity of the entire ecological network is close to the ideal state. The higher the gamma index, the tighter the connection between habitat patches, and the smoother the migration and material flow in the ecosystem. The calculation formula is:

[0170]

[0171] In the formula, the denominator represents the maximum number of ecological corridors that could exist in a completely connected network. The closer this value is to 1, the closer the connection in the network is to complete connectivity, and the lower the migration resistance of species.

[0172] 4) C r index: C r index measures the cost ratio of constructing corridors, ranging from 0 to 1, with a higher value indicating a higher cost of passing through corridors. Ecological networks with low cost ratios are more likely to form migration corridors. The calculation formula is:

[0173]

[0174] In the formula, C represents the length of the ecological corridor, and L represents the number of ecological corridors.

[0175] 5) CC r comprehensive connectivity index: CC r is a comprehensive measure of network connectivity, combining alpha, beta, gamma, and C rFour indices are used to reflect the overall performance of network connectivity. The index integrates the contribution of multiple parameters, providing a comprehensive evaluation of the overall connectivity and redundancy of the network. The higher the index value, the better the connectivity of the network, the stronger the feasibility of species migration, and the higher the priority of the network. The calculation formula is:

[0176]

[0177] Figure 4 The results of the regional habitat ecological network connectivity in this embodiment are shown. From the perspective of patch connectivity, the connectivity index results of habitat patches show that habitat patch No. 30 in the Songnen Plain has the highest connectivity index (close to 1), habitat patches No. 20 and No. 35 have higher connectivity index (about 0.6), and the connectivity index of other habitat patches is mostly <0.4, achieving quantitative evaluation of the connectivity degree of habitat patches. At the same time, the connectivity index of the most important habitat patch No. 20 is higher, while the connectivity index of the habitat patch with higher comprehensive centrality No. 30 is the highest, which shows that the comprehensive centrality of habitat patches is related to the connectivity index, but not completely consistent, revealing the role of habitat patches in the network structure from multiple aspects. From the perspective of corridor connectivity, the replaceability of the ecological corridor in the Songnen Plain is high, indicating that there are sufficient corridors for biological flow, and it is not easy to be affected by a single corridor, while the flowability of the ecological corridor in the Songnen Plain is low, indicating that although there are many corridors, there are many blocked areas, resulting in limited degree of smoothness of the ecological corridor, achieving quantitative evaluation of the smoothness of the corridor (congestion state and blocked state). From the perspective of the structural relationship between patches and corridors, the results of habitat ecological network structure connectivity show that the complexity of the internal loop of the habitat network in the Songnen Plain is moderate (α index = 0.67), the network structure is relatively complete (β index = 2.22), the network connection is close to the ideal state (γ index = 0.78), but the cost of corridor construction is high (Cr index = 0.97); Overall, the overall connectivity of the habitat ecological network in the Songnen Plain is good, and the priority is high (CCr index = 1.2), achieving comprehensive evaluation of regional habitat ecological network connectivity from the perspective of structural elements and overall structure.

[0178] Step S4 constructs a regional habitat ecological network stability evaluation scenario, based on random and target intervention strategies, representing the influence of natural changes and human factors on the habitat ecological network, setting two intervention modes of disturbance and recovery, and using a two-factor simulation experiment to construct four evaluation scenario elements of natural disturbance, human disturbance, natural recovery, and human recovery, and various stability evaluation scenarios, to quantify the stability of the regional habitat ecological network under different disturbance and recovery scenarios, measure the resistance of the habitat ecological network to habitat loss, and the ability of the habitat ecological network to restore its original structure and function after disturbance. In this embodiment, four representative scenarios of natural disturbance, human disturbance, natural recovery after disturbance, and human recovery are selected to evaluate the resistance stability and recovery stability of the habitat ecological network in the Songnen Plain.

[0179] (1) Intervention strategy

[0180] Random intervention removes or increases habitat patches randomly, simulating the state of habitat ecological network under natural conditions. Target intervention removes or increases habitat patches according to certain rules (such as habitat importance ranking, quantity and proportion, etc.), simulating the state of habitat ecological network after human intervention.

[0181] (2) Evaluation scenario elements

[0182] The natural disturbance scenario uses random intervention strategy to randomly remove habitat patches; the human disturbance scenario uses target intervention strategy to remove habitat patches according to human-set removal rules; the natural recovery scenario uses random intervention strategy to randomly increase habitat patches that have been lost; and the human recovery scenario uses target intervention strategy to increase habitat patches that have been lost according to ecological restoration management requirements or other standards.

[0183] Figure 5 The regional habitat ecological network stability evaluation scenario in this embodiment is shown. According to the steady-state characteristics of the ecosystem and its change law, the resistance to disturbance and self-recovery ability of the habitat ecological network are evaluated, considering four factors of nature, human, disturbance, and recovery, which cross to produce 14 stability evaluation scenarios, fully representing the stability dynamic changes of the habitat ecological network under different environmental conditions. Through the four typical evaluation scenarios of natural disturbance, human disturbance, natural recovery, and human recovery, the feedback of habitat ecological network stability to environmental condition changes can be intuitively displayed, which makes up for the shortcomings of existing technologies that only consider a single aspect of ecological network stability (resilience or robustness) or confuse the two.

[0184] Step S5 evaluates the resistance stability of the regional habitat ecological network based on natural disturbance scenarios and human disturbance scenarios. Resistance stability refers to the ability of the habitat ecological network to maintain its normal structure and function when the habitat ecological network is destroyed. By randomly or artificially removing habitat patches in the ecological network, the ability of the habitat ecological network to maintain connectivity is quantified, and the resistance stability is calculated.

[0185] The original habitat ecological network is initialized, the size of the maximum connected subgraph is calculated, and habitat patches are removed step by step according to a random order or artificial intervention rule. After each habitat patch is removed, the size of the maximum connected subgraph in the remaining habitat ecological network is recalculated until all habitat patches are removed. In this process, the size of the maximum connected subgraph in the ecological network after each habitat patch is removed is calculated step by step, and divided by the total number of habitat patches in the initial ecological network to obtain the proportion of the size of the connected subgraph removed at each step to the size of the initial ecological network, and the sum is accumulated. The accumulated value is divided by the total number of habitat patches to obtain the resistance stability of a single simulation, reflecting the "average performance" of the connectivity of the ecological network during the entire removal process. The resistance stability ranges from 0 to 1, and the higher the value, the more the habitat ecological network can maintain a larger connected subgraph during habitat removal, indicating a stronger anti-interference ability. The calculation formula is as follows:

[0186]

[0187] In the formula, N represents the total number of habitat patches in the habitat ecological network; S0 represents the size of the maximum connected subgraph in the initial habitat ecological network; Si represents the size of the maximum connected subgraph after the removal of the ith habitat patch; and S represents the resistance stability of the habitat ecological network. i

[0188] (1) Natural disturbance scenario

[0189] A random intervention strategy is used to simulate the scenario of the habitat ecological network being disturbed by nature. In each iteration, a random seed is set to ensure that the results of each random selection are different. To avoid the randomness of random removal, the average resistance stability value of 100 iteration cycles is used as the final result, reflecting the stability of the habitat ecological network as a whole when facing the random loss of habitat patches.

[0190] (2) Human disturbance scenario

[0191] ​​​The target intervention strategy is used to simulate the scenario of important habitat patches in the habitat ecological network being lost due to human disturbance. The habitat patches are sorted according to the comprehensive centrality obtained by normalizing and weighting the degree centrality, eigenvector centrality, betweenness centrality and closeness centrality, and the habitat patches are removed from high to low according to the comprehensive centrality score, and the resistance stability of the habitat ecological network under the scenario of human disturbance is calculated.

[0192] Step S6 evaluates the restoration stability of the regional habitat ecological network based on the natural restoration scenario and the artificial restoration scenario. The restoration stability refers to the potential restoration ability of the ecological network itself after the habitat ecological network is naturally or artificially destroyed. By gradually restoring the once-existing but lost habitat patches, the restoration ability of the maximum connected subgraph of the ecological network is quantified, and the restoration stability is calculated.

[0193] The size of the maximum connected subgraph of the habitat ecological network after being naturally destroyed is recorded, and the habitat patches that need to be restored are added back to the ecological network one by one, and the size of the maximum connected subgraph after each restoration is recorded. After each habitat patch is restored, the migration corridor between the original adjacent habitats is reconnected to reflect a more realistic restoration scenario, and then the restored habitat patch and the migration corridor record are erased from the to-be-restored list to prevent the same habitat patch and migration corridor from being selected and restored repeatedly in subsequent cycles, so as to avoid meaningless restoration and result in incorrect results. The connectivity size of all restoration steps is accumulated, and normalized by the initial total number of habitat patches and the size of the maximum connected subgraph after the habitat ecological network is destroyed, to calculate the restoration stability of the habitat ecological network after being disturbed. The restoration stability ranges from 0 to 1, and the higher the value, the stronger the restoration ability of the habitat ecological network after being destroyed, as it can still maintain a larger connected subgraph during the habitat restoration process. The calculation formula is as follows:

[0194]

[0195] In the formula, N represents the total number of destroyed habitat patches in the habitat ecological network; S0 represents the size of the maximum connected subgraph in the complete habitat ecological network; Si represents the size of the maximum connected subgraph after the i-th habitat patch is restored; and S represents the sum of the sizes of the maximum connected subgraphs in all habitat patch restoration steps. i In the formula, N represents the total number of destroyed habitat patches in the habitat ecological network; S0 represents the size of the maximum connected subgraph in the complete habitat ecological network; Si represents the size of the maximum connected subgraph after the i-th habitat patch is restored; and S represents the sum of the sizes of the maximum connected subgraphs in all habitat patch restoration steps. In the formula, N represents the total number of destroyed habitat patches in the habitat ecological network; S0 represents the size of the maximum connected subgraph in the complete habitat ecological network; Si represents the size of the maximum connected subgraph after the i-th habitat patch is restored; and S represents the sum of the sizes of the maximum connected subgraphs in all habitat patch restoration steps.

[0196] (1) Natural restoration scenario

[0197] The random intervention strategy is used to simulate the natural restoration scenario of the habitat ecological network. A random seed is set in each iteration to ensure that the results of each random selection are different. To avoid the randomness of accidental increase, the average restoration stability value of 100 iteration cycles is used as the final result to reflect the stability of the entire network during the natural restoration of the habitat ecological network.

[0198] (2) Human restoration scenarios

[0199] The target intervention strategy was used to simulate the scenario of artificially restoring important habitat patches in the habitat ecological network. The habitat patches were ranked according to the comprehensive centrality, which was the weighted sum of the degree centrality, eigenvector centrality, betweenness centrality and closeness centrality. The habitat patches were restored from high to low according to the comprehensive centrality score, and the restoration stability of the habitat ecological network under the human restoration scenario was calculated.

[0200] Figure 6 The results of the resistance stability and restoration stability of the regional habitat ecological network were shown. Under the natural disturbance scenario, the resistance stability of the habitat ecological network in the Songnen Plain was 0.65, and under the human disturbance scenario, it was 0.27, indicating that high-intensity human activities reduced the resistance stability of the habitat ecological network in the region. Under the natural restoration scenario, the restoration stability of the habitat ecological network in the Songnen Plain was 0.64, and under the human restoration scenario, it was 0.83, indicating that human restoration measures significantly increased the restoration capacity of the ecosystem and improved the restoration stability of the regional habitat ecological network. These evaluation results showed that under different scenarios, the results of this evaluation method were consistent with ecological principles and actual logic, and achieved dynamic and reliable evaluation of the resistance and restoration of habitat ecological network stability. Through the parameter setting of the intervention strategy, this method could evaluate the change process of habitat ecological network stability from multiple angles such as the number, proportion and importance of habitat patches or ecological corridors. This method not only considered the various scenario combinations of the habitat ecological network due to changes in external environmental conditions, but also further analyzed the internal ecological processes of the change in habitat ecological network stability under a single scenario, making up for the shortcomings of existing habitat ecological network evaluation techniques.

Claims

1. A regional habitat ecological network structure and a method for evaluating stability thereof, characterized by The method for evaluating the structure and stability of regional habitat ecological network is carried out according to the following steps: Step one, constructing the regional habitat ecological network, the structural elements of the ecological network include ecological sources and ecological corridors; Step two, evaluating the structural characteristics of the regional habitat ecological network, calculating the comprehensive centrality of habitat patches, the important priority of ecological corridors and the compactness of network structural elements; Step three, evaluating the connectivity of the regional habitat ecological network, calculating the connectivity index of habitat patches, the unobstructed index of ecological corridors and the connectivity of network structure; in step three, the connectivity index of habitat patches is evaluated: the connectivity integral index and the connectivity probability are added with equal weight to calculate the connectivity index of habitat patches; where the connectivity integral index is calculated by the formula: (7), where: is the connectivity integral index of habitat patch v; is the total number of habitat patches in the network; is the habitat patch is the habitat patch is the connection strength between habitat patch and habitat patch ; is the distance between habitat patch is the decay parameter, determining the degree of distance effect on connection strength; Connectivity probability calculation formula: (8), where: is the connectivity probability of habitat patch v; is the total number of habitat patches in the network; is the habitat patch and habitat patch is the connection probability between the habitat patches In step three, the unobstructed index of ecological corridors is evaluated: by grading the ecological obstacle points and the pinch points, the corridor flowability and the corridor substitutability are calculated, and the unobstructed index of ecological corridors is calculated by adding the two with equal weight; wherein the gallery flowability calculation formula is: (9), (10), In the formula, the higher the barrier density, the more obstacles exist in the corridor, the more difficult the species migration, and the lower the flowability of the corridor; the closer the corridor flowability to 1, the stronger the flowability of the corridor, and the closer to 0, the more serious the obstruction of the corridor; Corridor alternative calculation formula: (11), (12), In the formula, the higher the pinch point density, the lower the substitutability of the corridor, and the more easily the function is restricted by a single corridor; the substitutability value tends to 0, indicating that the functional redundancy of the corridor is low; and the substitutability tends to 1, indicating that the network has high redundancy; Step four, constructing the scenario for evaluating the stability of the regional habitat ecological network, setting the interference and recovery scenario mode, constructing the natural and human intervention strategy, and simulating the stability state of the habitat ecological network under different scenarios; Step five, based on the natural interference scenario and the human interference scenario, evaluating the resistance stability of the regional habitat ecological network, quantifying the ability of the habitat ecological network to maintain its structure and function when it is destroyed; in step five, the resistance stability of the habitat ecological network is evaluated: by randomly or artificially removing habitat patches in the ecological network, the ability of the habitat ecological network to maintain connectivity is quantified, and the resistance stability is calculated; The calculation formula is: (18), In the formula: N represents the total number of habitat patches in the habitat ecological network; This represents the size of the largest connected subgraph in the initial habitat ecological network; Represents removing the first After each habitat patch, the size of the largest connected subgraph in the remaining habitat ecological network; Represents removing the first The relative size of the largest connected subgraph in the habitat ecological network after each habitat patch; This represents the average connectivity ratio across all habitat patch removal steps. Step six, based on the natural recovery scenario and the human recovery scenario, evaluating the recovery stability of the regional habitat ecological network, quantifying the potential ability of the habitat ecological network to recover its structure and function after being destroyed; in step six, the recovery stability of the habitat ecological network is evaluated: the recovery stability refers to the potential recovery ability of the habitat ecological network after being naturally or artificially destroyed; by gradually recovering the habitat patches that once existed but have been lost, the recovery ability of the maximum connected subgraph of the ecological network is quantified, and the recovery stability is calculated; The calculation formula is: (19), where N represents the total number of destroyed habitat patches in the habitat ecological network; represents the size of the largest connected subgraph in the intact habitat ecological network; represents the size of the largest connected subgraph in the habitat ecological network after the restoration of the first habitat patch; represents the total sum of the size of the largest connected subgraph in all habitat patch restoration steps.

2. The regional habitat ecological network structure and its stability evaluation method according to claim 1, characterized in that In step one, the regional habitat ecological network is constructed: the structural elements include ecological sources and ecological corridors, the ecological sources are extracted based on habitat area, habitat structure, habitat quality, habitat sensitivity and habitat protection planning; based on land use, elevation, slope, nitrogen and phosphorus pollution, distance from road and distance from water body data, the species migration resistance is quantified; based on the spatial layout of ecological sources, the moving threshold is set according to the migration ability of different species, the ecological corridors are generated, the ecological pinch points and ecological obstacle points on the corridors are identified, and the regional habitat ecological network is constructed.

3. The regional habitat ecological network structure and its stability evaluation method according to claim 1, characterized in that In step two, the comprehensive centrality of habitat patches is evaluated by calculating the degree centrality, closeness centrality, eigenvector centrality and betweenness centrality of the habitat patches, normalizing them to eliminate the influence of dimension, and then equally adding them to comprehensively evaluate the importance of the habitat patches; wherein, 1) Centrality calculation formula: (1), wherein, is the degree centrality of habitat patch v, is the number of connections of habitat patch v; 2) The closeness centrality formula: (2), wherein is the closeness centrality of habitat patch v; is the shortest path length from habitat patch v to habitat patch t; 3) Eigenvector centrality formula: (3), wherein is all habitat patches connected to habitat patch v; is the largest eigenvalue of the eigenvector 4) Mediation centrality formula: (4), wherein is the intermediate centrality of habitat patch v; is the number of shortest paths from habitat patch s to habitat patch t; is the number of shortest paths through habitat patch v; In step two, the important priority of the ecological corridor is evaluated: The smaller the resistance index value of the ecological corridor, the smaller the resistance of species migration, the higher the quality of the corridor, and the higher the importance and protection priority of the corridor; the calculation formula is as follows: (5), wherein is a resistance index of the ecological corridor ; is a cost-weighted distance value of the ecological corridor is a minimum cost path length; In step two, the compactness of the network structure is evaluated: The average clustering coefficient of the habitat ecological network is calculated to measure the local connectivity between structural elements in the network, reflecting the overall structural characteristics of the network; the calculation formula is as follows: (6), where: is the average clustering coefficient of the habitat ecological network w; is the total number of habitat patches in the network; is the clustering coefficient of habitat patch v, representing the ratio of the number of edges that actually connect neighbors of habitat patch v to the number of edges that could connect them.

4. The regional habitat ecological network structure and its stability evaluation method according to claim 1, characterized in that In step four, the stability evaluation scenarios of the habitat ecological network are based on random and target intervention strategies to represent the influence of natural changes and human factors on the habitat ecological network, set two intervention modes of disturbance and recovery, and use double-factor simulation experiments to construct four evaluation scenario elements of natural disturbance, human disturbance, natural recovery and human recovery and various stability evaluation scenarios, to quantify the stability of the regional habitat ecological network under different disturbance and recovery scenarios, and measure the resistance of the habitat ecological network to habitat loss and the ability of the habitat ecological network to restore its original structure and function after disturbance; The intervention strategies are: random intervention removes or adds habitat patches randomly, simulating the state of the habitat ecological network damaged or recovered under natural conditions; target intervention removes or adds habitat patches according to the set rules, simulating the state of the habitat ecological network after human intervention; the set rules are: remove or add habitat patches according to importance, quantity and proportion; Evaluation scenario elements: The natural disturbance scenario adopts the random intervention strategy to randomly remove habitat patches; the human disturbance scenario adopts the target intervention strategy to remove habitat patches by setting removal rules; the natural recovery scenario adopts the random intervention strategy to randomly add habitat patches that once existed but have been lost; the human recovery scenario adopts the target intervention strategy to add habitat patches that once existed but have been lost according to the requirements of ecological restoration management or other standards.

5. The regional habitat ecological network structure and its stability evaluation method according to claim 1, characterized in that In step five, the natural disturbance scenario: the random intervention strategy is used to simulate the scenario of the habitat ecological network being disturbed by nature, a random seed is set in each iteration to ensure that the results of each random selection are different; to avoid the randomness of random removal, the average resistance stability value of 100 iteration cycles is used as the final result to reflect the stability of the entire network when the habitat ecological network faces random loss of habitat patches; In step five, the human disturbance scenario: the target intervention strategy is used to simulate the scenario of important habitat patches in the habitat ecological network being lost due to human disturbance; the comprehensive centrality of the habitat patches is sorted by normalizing and weighting the sum of the degree centrality, eigenvector centrality, betweenness centrality and closeness centrality; the habitat patches are removed from high to low according to the comprehensive centrality score, and the resistance stability of the habitat ecological network under the human disturbance scenario is calculated.

6. The regional habitat ecological network structure and its stability evaluation method according to claim 1, characterized in that The natural recovery scenario in step six: the random intervention strategy is used to simulate the natural recovery of the habitat ecological network. A random seed is set in each iteration to ensure that the results of each random selection are different. To avoid the randomness of the random increase, the average recovery stability value of 100 iteration cycles is used as the final result, reflecting the stability of the entire network during the natural recovery of the habitat ecological network. The artificial recovery scenario in step six: The target intervention strategy is used to simulate the scenario of artificially recovering important habitat patches in the habitat ecological network. The habitat patches are sorted according to the comprehensive centrality obtained by normalizing and weighting the degree centrality, eigenvector centrality, betweenness centrality, and closeness centrality. The habitat patches are recovered from high to low according to the comprehensive centrality score, and the recovery stability of the habitat ecological network under the artificial recovery scenario is calculated.

Images