Method and system for dynamically evaluating influence of engineering construction on biodiversity

Abstract

The invention relates to the technical field of biodiversity assessment, and discloses a dynamic assessment method and system for the influence of engineering construction on biodiversity, and the method comprises the steps: obtaining a basic ecological data set, and carrying out the construction of an ecological patch graph structure and the edge building of hydrological connectivity according to the basic ecological data set, and obtaining a composite ecological graph model; performing plaque function heterogeneity aggregation and disturbance link simulation processing to obtain an ecological propagation path and a sensitivity pedigree of each node; performing spectral domain decomposition and key propagation node identification processing to obtain ecological intermediary nodes and propagation weights; performing reversible modeling and path entropy backtracking analysis processing to obtain a community reconstruction trend index and a function stability attenuation curve; and carrying out ecological response persistence analysis processing to obtain a time sequence biodiversity influence index. According to the method, the problems of static state, splitting, low coupling and the like in the prior art are solved, and the precision, timeliness and predictive capacity of ecological disturbance dynamic response evaluation are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of biodiversity assessment, and in particular to a method and system for dynamically assessing the impact of engineering construction on biodiversity. Background Art

[0002] In existing technologies, biodiversity impact assessments often rely on static monitoring data, macro-indicator systems, or species population statistics, supplemented by expert experience for disturbance intensity classification and regional sensitivity division. Such methods have difficulty revealing the dynamic transmission mechanism of disturbances within ecological networks, and are particularly unable to depict the spatiotemporal breakdown process of wetland patch community structure caused by changes in water connectivity. In addition, traditional methods fail to fully integrate multi-source heterogeneous information such as high-resolution topography, hydrological flow, community function, and species interactions. Assessment results are often one-sided and delayed, making it difficult to serve targeted ecological restoration designs, and unable to predict potential ecological chain reactions before geological disturbances occur.

[0003] Therefore, there is an urgent need for a dynamic assessment method and system for the impact of engineering construction on biodiversity to solve the above problems. Summary of the Invention

[0004] The present invention aims to provide a method and system for dynamically assessing the impact of engineering construction on biodiversity, thereby addressing the aforementioned issues. To achieve this objective, the present invention employs the following technical solutions: First, this application provides a dynamic assessment method for the impact of engineering construction on biodiversity, including: Obtaining basic ecological data sets, including water system distribution maps, forest wetland patch structure change maps, and multi-time period sample plot community data; Based on the basic ecological dataset, ecological patch graph structure construction and hydrological connectivity edge processing are performed to obtain a composite ecological graph model that reflects spatial structural variation; Based on the composite ecological graph model, patch functional heterogeneity aggregation and disturbance link simulation were performed to obtain the ecological transmission path and the sensitivity spectrum of each node; According to the ecological propagation path and the sensitivity spectrum of each node, spectral domain decomposition and key propagation node identification are performed to obtain the ecological intermediary nodes and propagation weights in the disturbance response network; According to the ecological intermediary nodes and propagation weights, reversibility modeling and path entropy back-analysis were performed to obtain the community reconstruction trend index and functional stability attenuation curve; Based on the community reconstruction trend index and the functional stability decay curve, an ecological response persistence analysis is performed to obtain a time series biodiversity impact index.

[0005] Secondly, this application also provides a dynamic assessment system for the impact of engineering construction on biodiversity, including: An acquisition unit is used to acquire a basic ecological data set, wherein the basic ecological data set includes a water system distribution map, a forest wetland patch structure change map, and multi-time period sample community data; A processing unit is used to construct an ecological patch graph structure and perform hydrological connectivity edge processing based on the basic ecological dataset to obtain a composite ecological graph model that reflects spatial structural variation; A simulation unit is used to perform patch functional heterogeneity aggregation and disturbance link simulation processing according to the composite ecological graph model to obtain the ecological transmission path and the sensitivity spectrum of each node; an identification unit, configured to perform spectrum domain decomposition and key propagation node identification processing based on the ecological propagation path and the sensitivity spectrum of each node, and obtain ecological intermediary nodes and propagation weights in the disturbance response network; A backtracking unit is used to perform reversibility modeling and path entropy backtracking analysis based on the ecological intermediary nodes and propagation weights to obtain a community reconstruction trend index and a functional stability attenuation curve; The analysis unit is used to perform ecological response continuity analysis and processing based on the community reconstruction trend index and the functional stability decay curve to obtain a time series biodiversity impact index.

[0006] The beneficial effects of the present invention are: This paper uses the travertine-forest-wetland complex ecological zone as a representative scenario and proposes a full-process algorithm scheme from patch structure-hydrological connectivity-disturbance propagation-response identification to impact index construction. This method constructs a composite ecological graph model, integrates hydrological paths and heterogeneous edge weights of ecological functions, and combines graph neural networks to model patch structure evolution, thereby identifying ecological propagation paths and sensitive lineages, and ultimately outputs a time series biodiversity index that reflects the intensity and delay of disturbance effects. This technology overcomes the problems of static, fragmented, and low-coupling existing technologies, significantly improving the accuracy, timeliness, and predictive capabilities of dynamic response assessments of ecological disturbances. It has broad adaptability and practical value for ecological protection.

[0007] Other features and advantages of the present invention will be described in the following description, and in part will become apparent from the description, or understood by practicing the embodiments of the present invention. The purposes and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments. It should be understood that the following drawings only illustrate certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can be obtained based on these drawings without paying any creative work.

[0009] Figure 1 A schematic flow chart of a dynamic assessment method for the impact of engineering construction on biodiversity according to an embodiment of the present invention; Figure 2 Schematic diagram of the structure of a dynamic assessment system for the impact of engineering construction on biodiversity according to an embodiment of the present invention.

[0010] In the figure: 701, acquisition unit; 702, processing unit; 703, simulation unit; 704, identification unit; 705, backtracking unit; 706, analysis unit. DETAILED DESCRIPTION

[0011] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings herein can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0012] It should be noted that similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are used only to distinguish the description and should not be understood as indicating or implying relative importance.

[0013] Example 1

[0014] This embodiment provides a method for dynamically assessing the impact of engineering construction on biodiversity.

[0015] See also Figure 1 , the figure shows that the method includes step S1, step S2, step S3, step S4, step S5 and step S6.

[0016] Step S1, obtaining a basic ecological data set, wherein the basic ecological data set includes a water system distribution map, a forest wetland patch structure change map, and multi-time period sample community data; This step, as you can understand, first involves acquiring a basic ecological dataset with temporal characteristics and spatial heterogeneity, targeting the unique geomorphological and ecological patterns. Specifically, this includes maps of water system distribution, changes in forest wetland patch structure, and multi-period sample plot community data. Water system extraction relies on the fusion of high-precision LiDAR terrain data and multispectral remote sensing imagery to accurately reconstruct irregular water system structures formed by geomorphological processes such as travertine accumulation, faults, and closed depressions. These structures play a key role in hydrological connectivity and the migration of ecological functions. Forest wetland patch structure is dynamically identified through object-oriented remote sensing classification techniques and temporal change detection algorithms, resulting in fragmentation, marginalization, and migration trends caused by geological disaster control projects. Sample plot community data is derived from multi-period survey plots deployed in representative habitats, documenting species composition, functional grouping, and their response characteristics, providing empirical support for subsequent semantic modeling of ecological nodes. This step bridges the gap between spatial layer information and biological attribute data, laying a unified data foundation for multi-dimensional spatial, functional, and temporal ecological graph structure modeling. Its technical effect is reflected in its ability to accurately reflect the structural evolution potential of regional complex ecosystems under disturbance backgrounds, and provide high-quality input for subsequent dynamic ecological communication mechanisms based on graph construction, thereby improving the spatial resolution and ecological explanatory power of the full-process evaluation.

[0017] Step S2: constructing an ecological patch graph structure and building edges based on hydrological connectivity based on the basic ecological dataset to obtain a composite ecological graph model that reflects spatial structural variation; It can be understood that this step is based on the acquired basic ecological data set, integrating the spatial attributes of ecological patches with the dynamic relationship of hydrological processes to construct a composite ecological map structure with spatiotemporal semantics. It transforms the traditional static patch distribution map into an ecological map that can dynamically evolve and has the ability to integrate multimodal information. It effectively captures the structural variation trends driven by engineering disturbances, hydrological reconstruction, and community migration within the scenic area, and significantly enhances the modeling ability and prediction accuracy of the ecological response process. In this step, step S2 includes step S21, step S22, step S23, and step S24.

[0018] Step S21: perform hydrological connectivity identification processing based on the water system distribution map in the basic ecological dataset and the terrain data of the preset digital elevation model, and calculate the minimum cumulative cost path of surface runoff based on the CostDistance model to obtain the hydrological connectivity attenuation matrix between patches; It's understandable that the water system in this step represents a unique geomorphic phenomenon resulting from travertine biodeposition. Its water distribution is controlled by surface calcareous crusts, microtopographic undulations, and subsurface undercurrents. Traditional hydrological models based on mainline water system extraction struggle to accurately reflect these local hydrological processes. Therefore, this step innovatively introduces a hydrological modeling approach based on the minimum cumulative cost path. This method couples the aspect and slope gradient of the digital elevation model's terrain data with the geomorphic structure of the travertine formation area to construct a high-resolution surface flow resistance field. Based on this, the system calculates the cumulative hydrological cost of reaching each other via the minimum resistance path between any two ecological patches, generating an asymmetric, directional hydrological connectivity attenuation matrix. This matrix not only quantifies the accessibility of potential transmission pathways for water or disturbance factors (such as pollution and sediment) between ecological units, but also reflects the decline in flow capacity due to factors such as travertine barriers and slope closure. This process effectively addresses the difficulty of modeling hydrological accessibility in highly complex geomorphic areas. This step provides a connectivity expression method that integrates microtopography, special travertine structure, and ecological patch layout. This ensures that the edge weights in subsequent ecological graph structure modeling are no longer simple spatial distances or artificial settings, but rather dynamic coupling indicators reflecting real geomorphological and hydrological processes, thereby enhancing the structural credibility and evolutionary explanatory power of the graph model in ecological process modeling.

[0019] This step first uses terrain preprocessing algorithms (such as depression filling, flow direction calculation, and slope weight extraction) to establish hydrological consistency in the digital elevation model. Hydrological consistency refers to ensuring that flow paths, watershed delineation, and hydrological process simulations conform to the actual surface topography and accurately reflect natural water flow behavior through terrain data and appropriate model parameter settings. Each ecological patch in the patch layer is then defined as a source-sink pair, with the actual watershed divisions in the river system distribution map serving as the constraint boundaries to form local flow units. The system then uses the CostDistance algorithm to establish a minimum resistance path model for water propagation. This model assigns a composite resistance value to each grid cell, composed of slope, travertine backwater factor, and landform roughness. A variant of the Dijkstra algorithm is then used to calculate a minimum "cost" path between each pair of patches, where the cost is the sum of the accumulated hydrological resistance.

[0020] Ultimately, each patch corresponds to a minimum hydrological path, and the cost values ​​of all paths form a hydrological connectivity attenuation matrix, where smaller matrix elements indicate greater path accessibility and smoother potential water flow, while smaller elements indicate significant blockage or isolation. Furthermore, to enhance ecological interpretability, this matrix can be further normalized and integrated with attributes such as patch area and boundary shape, enhancing its adaptability to ecological processes (such as nutrient transport and wetland restoration connectivity).

[0021] Step S22: constructing an ecological node graph based on the hydrological connectivity attenuation matrix, fusing the forest wetland patch structure change map and multi-period sample community data through a regional fusion method, extracting a set of spatial nodes representing different functional groups, and obtaining an ecological node graph with community function labels; It's understandable that this step not only identifies the location and boundaries of patches but also aggregates spatial units with similar ecological functions, allowing them to participate in subsequent propagation modeling as semantically consistent and functionally synergistic nodes within the graph. This process utilizes a regional fusion (graph-based regional fusion) approach. Unlike traditional object-oriented patch classification, it dynamically adjusts the ecological semantic units constructed by nodes while preserving the continuity of ecological structural boundaries.

[0022] This step first treats all spatial patches in the forest wetland patch structure change map as nodes for initial subdivision. Primary attribute characteristics are assigned to each node, including patch morphological index (fractal dimension), physical location, adjacent hydrological pathways, and evolutionary trends (obtained through multi-period patch overlap analysis). Simultaneously, multi-period sample community data are mapped to the corresponding patch areas, annotated with functional attribute labels such as species composition and functional groups (e.g., nitrogen-fixing plants, emergent plants, and key nodes in the food chain).

[0023] Subsequently, a regional fusion strategy was introduced, using both functional similarity (based on the Tanimoto coefficient or attribute distance) and structural connectivity (based on low-cost edges in the hydrological attenuation matrix) as dual constraints. This strategy gradually merged patches with similar functions and a tendency for hydrological connectivity within the original graph. This process not only preserved the true spatial distribution and hydrological coupling between ecological units within the micro-geomorphology, but also achieved precise merging of the ecological functions of graph nodes through joint structural-attribute optimization. The resulting node set not only reflects spatial clustering characteristics but also embeds biodiversity function labels derived from sample data, forming a functional ecological node graph with the ability to express ecological semantics.

[0024] Step S23: performing edge weight modeling based on the ecological node graph and the hydrological connectivity attenuation matrix, constructing multivariate heterogeneous edges through a spectral clustering algorithm, and obtaining an ecological relationship graph with structure-function joint information; This step first constructs an initial edge-connection framework using the cost values ​​from the hydrological connectivity decay matrix, selecting node pairs with significant accessibility along the hydrological path as candidate edge targets. Subsequently, three core characteristics are extracted for each node pair: flow direction consistency (determining whether the path conforms to the overall basin flow trend based on the flow direction grid), geomorphic resistance differences (such as slope difference and factors affecting travertine accumulation), and community functional redundancy (the degree of overlap in the functional structure of the two node communities, calculated using the Jaccard coefficient). These characteristics are then standardized and converted into a multidimensional attribute vector, representing the multiple ecological coupling characteristics of each candidate edge.

[0025] The system then applies a spectral clustering model to perform a preliminary cluster analysis of the node graph, identifying subgraph regions with locally dense structural connections or functionally distinct group characteristics. These clustering results serve as structural guidance for edge weight construction. Furthermore, the three core attributes are synthesized into edge weights using a multivariate weighted combination function. The weight coefficients are determined using information gain or the minimum redundancy maximum relevance (mRMR) algorithm to enhance the ecological representativeness of the edges. The resulting ecological relationship graph not only preserves the basic topological connections in the graph structure but also encodes the complexity of ecological processes. The heterogeneity of the edges is reflected in the presence of a mixture of hydrophysical edges, functional compensation edges, and coupling edges.

[0026] Step S24: perform graph structure perturbation expression processing based on the ecological relationship graph, introduce a gated graph neural network to dynamically modulate and learn the edge weight changes, capture the changing trend of the connection strength of forest wetland patches before and after geological disaster control, and obtain a composite ecological graph model that reflects spatial structure variation.

[0027] It can be understood that this step first inputs the ecological relationship graph constructed under multiple time slices (corresponding to key node time points before, during, and after the governance) into a dynamic graph sequence, and retains the weight changes of each edge at different times. In the gated graph neural network, the retention of historical states and the introduction of new states are dynamically controlled through a gating mechanism (the reset gate and update gate in the GRU structure), allowing the system to effectively learn the edge weight change patterns before and after the disturbance event. For example, for node pairs whose hydrological flow is interrupted due to governance, their edge weights will undergo a sudden decrease over time, while for areas with artificially restored water connectivity, their edge weights will show an upward trend after the disturbance. This network not only updates the node state itself but also propagates and modulates the structure of its adjacent edges, enabling the entire graph to achieve "disturbance-sensitive reconstruction" in terms of structure.

[0028] The output is a time-enhanced composite ecological graph model, whose node attributes retain ecological semantics, edge weights express the strength of connections after temporal evolution, and the overall structure reflects the changing connectivity trends of spatial patches under the background of governance disturbance. This spatial graph structure will serve as the basis for subsequent disturbance path extraction and sensitivity analysis.

[0029] Step S3: performing patch functional heterogeneity aggregation and disturbance link simulation processing according to the composite ecological graph model to obtain the ecological transmission path and the sensitivity spectrum of each node; It can be understood that this step leverages the established composite ecological graph model, focusing on the propagation pathways of disturbances within the ecosystem and modeling the differences in node responses, revealing the transmission pathways and impact diffusion mechanisms between functionally heterogeneous patches. This step, for the first time, integrates ecological function clustering with graph-structured path propagation models, breaking the limitations of separate modeling of spatial structure and ecological attributes and enhancing the ecological semantic expression of disturbance pathways. In this step, step S3 includes steps S31, S32, S33, and S34.

[0030] Step S31: performing patch function index extraction processing according to the composite ecological graph model, and obtaining a node function feature tensor by applying a feature coding mechanism based on community function redundancy, species sensitivity, and habitat rarity to the graph nodes; It can be understood that in this step, the system first uses sample community data and combines it with the plant functional classification system (Grime's triangulation model) to functionally classify species within each node. Community functional redundancy is calculated by the balance of species numbers within the same functional group. Higher values ​​indicate greater functional substitutability and improved resistance to disturbances. Species sensitivity is calculated by weighting the list of known environmentally sensitive species and their frequency of occurrence in the sample. The weights can be set based on local stress response thresholds. Higher values ​​indicate a stronger response to micro-disturbance and greater ecological vulnerability. Habitat rarity is derived by calculating the standard deviation of the geographic distribution density and patch area of ​​nodes within the entire ecological map. Rare, small, isolated nodes score higher on this metric, reflecting their irreplaceable nature within the community network.

[0031] These three indicators are encoded as three-dimensional feature channels at the node level, constructing a node functional feature tensor that structurally corresponds to ecological map nodes and semantically integrates information from multiple ecological processes. This tensor serves as the core input variable for subsequent propagation path identification and response pedigree analysis, supporting multi-scale graph neural networks to efficiently model inter-node influence and functional heterogeneity.

[0032] Step S32: performing patch functional heterogeneity aggregation processing based on the node functional feature tensor, dividing the nodes into functional response subgroups by using an adaptive fuzzy C-means clustering algorithm, and adjusting the aggregation boundaries in combination with the edge density and local connectivity in the graph structure to obtain a functional heterogeneity subgraph set; This step, understandably, relies on the node functional feature tensor constructed in the previous step. This tensor encompasses high-dimensional ecological indicators such as community functional redundancy, species sensitivity, and habitat rarity, creating a multidimensional ecological state description vector for each node. Taking into account the fuzzy boundaries and local cross-influences between these ecological indicators, the system introduces an adaptive fuzzy C-means clustering method to perform soft classification and aggregation of nodes.

[0033] The adaptive fuzzy C-means clustering method used in this step introduces a structural guidance term and a semantic entropy adjustment factor into the distance function. This ensures that the membership of nodes assigned to multiple cluster centers is constrained by ecological functional similarity while also dynamically adjusting weights based on sensitivity to structural disturbances. This mechanism is particularly useful in ecosystems with similar functions but distinct responses. For example, if two wetland nodes have nitrogen-fixing vegetation, but one is structurally isolated and more sensitive to disturbances, the adaptive fuzzy C-means clustering method can cluster them into a response-prioritized subcluster.

[0034] The guided clustering process of the structural guidance term considers graph structural features of the disturbance propagation path, such as path betweenness, coupling, and centrality, making it easier for adjacent high-propagation nodes to be grouped into the same functional response subset, thereby improving the propagation consistency of the results and the explanatory power of ecological connectivity. The semantic entropy adjustment factor measures the change in global membership entropy caused by cluster assignment, limiting the excessive divergence or convergence of a node's membership distribution across all clusters, maintaining a reasonable range of ecological ambiguity, preventing "structural jump clustering" or "excessive ecological heterogeneity within clusters," and improving the semantic stability of ecological clusters.

[0035] Among them, habitat rarity is an important indicator used to measure the spatial distribution scarcity and ecological substitutability of a certain ecological unit (such as species habitat, functional patch, ecological node) in the entire ecological network. It is a key factor in evaluating its ecological protection priority, response vulnerability and sensitivity to disturbance propagation. It is expressed by functional substitutability. Habitat rarity and functional substitutability are opposite numbers. Functional substitutability indicates the similarity with other habitats and is calculated using the cosine function.

[0036] After initial clustering is complete, the system further fine-tunes the boundaries by combining edge density and edge weight (determined by hydrological connectivity and functional similarity) within the ecological graph structure with local connectivity metrics (such as the local clustering coefficient). This process ensures that subgraph divisions are not only rationally clustered in the functional space but also form cohesive regions along the propagation paths within the graph structure. The resulting set of functionally heterogeneous subgraphs, each representing a relatively consistent ecological response unit in both the functional and structural spaces, exhibits strong cohesion and a clear transmission path for external disturbances.

[0037] Step S33: performing disturbance link simulation processing based on the functional heterogeneity subgraph set, constructing a structural disturbance model based on random walk, simulating path crossing probability in combination with construction disturbance time series data, and obtaining a potential disturbance propagation path network; It can be understood that this step first treats each functionally heterogeneous subgraph as a propagation unit. All nodes in the graph that overlap with the construction disturbance space or have drastically changed edge weights are identified as the set of disturbance source nodes. These nodes correspond to patches where construction disrupted travertine hydrological connectivity, squeezed wetland boundaries, or triggered landform transformation. The system then performs a weighted biased random walk on each source node within its subgraph and adjacent subgraphs. The random walk transition probability not only considers the edge weights of the graph structure (determined by an ecological relationship graph with joint structure-function information), but also integrates the following two disturbance-aware factors: Temporal weight of disturbance influencing factors: Based on the timestamps and disturbance intensity records of the construction stages (e.g., early blasting, mid-term slope support, and late water diversion), a disturbance temporal attenuation function (e.g., exponential decay or memory gated regulation) is assigned to each transfer. Functional imbalance propagation tendency: The path jumping probability is adjusted by the functional distance between nodes (derived from the tensor characteristics of S31), so that edges with strong functional heterogeneity are more likely to spread influence, reflecting the weighted strengthening of the ecological "imbalance channel".

[0038] Each path accumulates disturbance crossing probabilities as it moves, forming a disturbance propagation probability field radiating outward from the source node. The system performs multiple rounds of random sampling and aggregation on propagation paths within each subgraph, ultimately summarizing them into a network of potential disturbance propagation paths. Its structure manifests as a disturbance diffusion map with strong directionality, uneven weighting, and high path overlap.

[0039] Compared with traditional methods based on the shortest path or full-graph walk, this step introduces a structural disturbance perception mechanism to match path selection with the ecological vulnerability between patches. In addition, by integrating functional distance and propagation tendency, the path network output by the system is not only a connectivity graph in terms of graph structure, but also an explicit expression of potential ecological collapse chains, providing key support for subsequent sensitive lineage identification and regulatory strategy design.

[0040] Step S34: perform sensitivity spectrum identification processing based on the disturbance propagation path network, jointly rank the maximum incremental rate of node edge weight evolution and functional contribution, use spectral clustering to divide the high-responsiveness node subsets, and obtain the set of highly sensitive nodes in the ecological propagation path and their sensitivity spectrum.

[0041] It's understandable that this step first analyzes the evolution of edge weights for all nodes in the perturbation propagation network. Based on the time series of edge weights learned by the graph neural network, the maximum incremental rate of each node's adjacent edges—the maximum amplitude or slope of the change in edge weights before and after the perturbation—is calculated as an indicator of the node's perturbation response strength during structural propagation. Simultaneously, the system uses the encoded node functional feature tensor and calculates each node's functional contribution using principal component analysis. This value reflects the node's ecological maintenance weight within its functional module.

[0042] ,in, represents the maximum incremental rate of the adjacent edges of node i, Indicates traversing all adjacent edges of node i, Indicates finding the position of the maximum increment in the time series (from moment 2), represents the edge weight between node i and adjacent node j at time t (output by the graph neural network), ε represents a small constant to prevent zero, and T represents the total time step (such as the number of perturbation time series frames).

[0043] The system then constructs a joint ranking matrix based on these two dimensions, ranking nodes by dual-index scores to identify the top-scoring candidate sensitive node sets. Building on this foundation, the system further introduces a spectral clustering method. Based on the Laplace matrix of the perturbation propagation path network, its eigenvectors are calculated and an embedding space is constructed. The candidate sensitive nodes are projected into this space and clustered by response strength and relative topological position, thereby identifying a subset of highly responsive nodes with strong response synergy and similar functional locations.

[0044] The final output is a set of highly sensitive nodes and their sensitivity spectrum: each highly sensitive node has clear functional attributes (such as high dependence on wetland stability or hydrological connectivity), structural attributes (located at the incoming / outgoing bottleneck of the disturbance path) and its position in the cluster spectrum, constituting the most vulnerable and critical response unit in the ecological network.

[0045] Step S4: performing spectrum domain decomposition and key propagation node identification processing based on the ecological propagation path and the sensitivity spectrum of each node to obtain the ecological intermediary nodes and propagation weights in the disturbance response network; It can be understood that this step, based on the previously obtained ecological transmission path network and its corresponding node sensitivity spectrum, further identifies ecological intermediary nodes in the disturbance transmission process, namely, the structural cores that bridge, guide, or diffuse multiple transmission paths. It also assigns precise transmission weights to each intermediary node to characterize its true ecological niche and control power in the diffusion of system-level disturbances. In this step, step S4 includes steps S41, S42, S43, and S44.

[0046] Step S41: Based on the ecological propagation path and the sensitivity spectrum of each node, Laplace spectrum decomposition of the propagation path graph is performed. By constructing a normalized Laplace matrix of the perturbation propagation graph and calculating its eigenvalue-eigenvector pairs, a spectral domain feature set for path energy distribution identification is obtained; It can be understood that this step first represents the ecological transmission path network G = (V, E) as a weighted graph, where nodes V represent functional patches and edges E represent disturbance transmission paths. Edge weights incorporate hydrological connectivity, functional redundancy, and transmission probability (derived from disturbance link simulation). Next, a normalized Laplacian matrix of this network is constructed. This Laplacian matrix helps maintain the stability of spectral domain features despite changes in node number or graph density, making it suitable for structural analysis of heterogeneous patch networks.

[0047] Next, the system performs eigenvalue decomposition on the Laplace matrix, obtaining the first k eigenvalue-eigenvector pairs. The eigenvalues ​​represent the frequency characteristics of the perturbation's propagation path in the graph, while the eigenvectors represent the structural weight of each node in that propagation mode. Low-order eigenvalues ​​correspond to full-graph connectivity, reflecting slow diffusion and globally stable processes. High-order eigenvalues ​​capture boundary fractures and localized catastrophic paths, providing a crucial basis for identifying chained perturbations and structural imbalances.

[0048] The system combines the projections of all nodes onto these spectral vectors to form a graph domain feature set, namely a node × mode spectral matrix. This set can be used for subsequent analysis, such as spectral energy calculation, identification of propagation energy focus, and optimization of intermediary nodes. In particular, the system also retains the magnitude of the change in spectral features before and after the disturbance occurs, measuring the "frequency activation effect" of the disturbance on the network propagation substrate.

[0049] This step projects the disturbance propagation path from graph space to frequency domain space, providing pattern recognition capabilities for structural transmission mechanisms, improving the visualization and quantitative interpretation of system-level infectious paths, and enhancing the analytical power of the ecological graph structure to the dynamic response of disturbances. It avoids relying solely on superficial indicators such as node degree and adjacency, and instead uses spectral morphology to comprehensively characterize the propagation resilience and instability threshold of the ecosystem.

[0050] Step S42: Calculate the energy concentration of propagation nodes based on the graph domain feature set, identify propagation path segments with significantly concentrated spectral energy through the maximum spectral energy density projection method, and obtain a local path subgraph with a disturbance aggregation effect greater than a preset threshold; It can be understood that in this step, the system first calculates the energy of the spectral projection value of each node in the spectral domain feature set in each order feature vector (corresponding to different propagation modes). The algorithm used is the maximum spectral energy density projection method, that is, the spectral energy density of each node in the spectral vector group is calculated as: , where E(i) represents the maximum spectral energy density of node i, represents the component value at node i in the l-th eigenvector (i.e., the projection of the node on the l-th mode), and k represents the number of spectral modes taken.

[0051] The system then uses a sliding window across the entire graph to extract propagation path segments and aggregates the mean total energy density of the nodes in each path segment as a disturbance concentration indicator for that segment. All path segments are sorted and an adaptive energy threshold (e.g., mean + 1.5 × standard deviation) is set based on the statistical characteristics of the energy distribution of the entire graph. Path segments with significantly higher energy densities than the background are selected to form a set of highly disturbed paths.

[0052] These high-disturbance segments are reaggregated into local path subgraphs with continuous structure and smooth edge weight transfer. These are regions in the propagation graph where strong disturbance responses are concentrated in the frequency domain. This step constructs local structural units with the strongest disturbance effects, the largest response amplitudes, and the most significant control significance, providing a structural enclosure and energy focusing foundation for key node identification.

[0053] Step S43: performing ecological intermediary node identification processing based on the local path subgraph, extracting key nodes with bridging effects in multiple propagation paths through a joint screening mechanism of betweenness centrality and path coupling, and obtaining a set of ecological intermediary nodes in the disturbance response network; It can be understood that in this step, the system first calculates the betweenness centrality of the node in each local path subgraph. This indicator measures the frequency of the node as a transit node in all shortest paths. In the ecological communication graph, nodes with high betweenness centrality are often the connection hubs of different paths, controlling the main channel for the diffusion of disturbances from one functional subdomain to another. At the same time, the system also introduces a path coupling index to measure the degree of overlap of a node's participation in multiple disturbance paths. The specific definition is: for each node, the frequency of its simultaneous appearance in each independent disturbance path is counted, and the functional heterogeneity between these paths (such as the functional differences of the subgraphs connected by the paths) is weighted. The higher the value, the more significant its bridging effect in "cross-functional communication".

[0054] Next, the system normalizes and integrates betweenness centrality and path coupling, then uses a linear weighted approach to jointly rank all candidate nodes. Nodes with scores exceeding the network mean + standard deviation are selected as the final set of intermediary nodes. To enhance ecological explanatory power, the system also retains a list of participating pathways, the type of bridging function (e.g., forest-wetland, travertine-bare slope), and the rate of change in edge weight for each intermediary node, facilitating subsequent visualization modeling and influencing factor attribution. This allows accurate identification of key nodes with bridging effects in multipath propagation, providing precise control points for interrupting disturbed links or guiding ecological restoration pathways.

[0055] Step S44: Perform propagation weight modeling based on the set of ecological intermediary nodes, and assign a response weight to each intermediary node on the engineering construction disturbance path through a multi-factor data normalization method based on node redundancy and path coupling coefficient, to obtain the ecological intermediary nodes and propagation weights in the disturbance response network.

[0056] It can be understood that this step first starts from two core dimensions to construct the propagation effect factor of each intermediary node under the engineering construction path. On the one hand, the system calculates node redundancy to measure whether the node can be replaced in the functional subgraph. The calculation method is based on the redundancy of the functional group of the node (such as the number of nodes with the same functional type) and the similarity of ecological status (such as the overlap of the functional axis score). Nodes with low redundancy are usually indispensable control points on a certain functional path, and their propagation weight should be relatively high. Among them, engineering construction disturbance paths refer to these paths. These are propagation channels that affect certain nodes in the ecosystem (such as wetlands, forests, etc.) in the context of engineering activities (such as construction and infrastructure construction). These paths will vary according to the sensitivity and structural changes of the ecosystem.

[0057] The system also calculates the node's path coupling coefficient within the propagation path, quantifying the strength of the synergistic effect created by its role as a bridge node in the intersection of multiple propagation paths. The path coupling coefficient takes into account two factors: the number of functionally heterogeneous paths the node appears on (cross-functional coupling), and the synchronization of edge weight changes across these paths (disturbance co-movement).

[0058] ,in, represents the path coupling coefficient of node v, represents the set of all disturbance propagation paths containing node v, and belong Two different disturbance propagation paths.

[0059] The system then uses a multi-factor data normalization method to assign each intermediary node's propagation capacity along each perturbation path. This algorithm constructs a raw propagation impact score by multiplying the inverse of node redundancy (indicating irreplaceability) by the path coupling coefficient. This algorithm then uses local intra-path normalization and global graph-level mapping to ensure that propagation weights are comparable, nonlinearly scalable, and context-adaptive. Ultimately, each intermediary node is assigned a propagation weight between 0 and 1 on each participating path, forming a propagation weight matrix with the structure {node ID, path ID, weight}.

[0060] Finally, a set of ecological intermediary nodes and their propagation weight distribution on each path in the disturbance response network are obtained. This structure is used as the core input for subsequent disturbance backtracking modeling, structural reversibility learning, and stability trend quantitative analysis.

[0061] Step S5: performing reversibility modeling and path entropy backtracking analysis based on the ecological intermediary nodes and propagation weights to obtain a community reconstruction trend index and a functional stability attenuation curve; It can be understood that this step, based on the set of ecological intermediary nodes and their propagation weight matrix obtained in the previous step, further conducts disturbance reversibility modeling and path entropy backtracking analysis. This aims to quantitatively characterize the ecological network's recovery potential, reconstruction trends, and functional stability evolution under disturbance at a systemic level. This step not only reveals whether the disturbance can be spontaneously absorbed or repaired by the ecosystem but also outputs actionable evaluation indicators: the community reconstruction trend index and the functional stability decay curve, providing key support for ecological intervention design and time window optimization. In this step, step S5 includes steps S51, S52, S53, and S54.

[0062] Step S51: Based on the ecological intermediary nodes and propagation weights, a disturbance reversibility state modeling process is performed. A weighted graph convolutional autoencoder is constructed to simulate the disturbance node deletion and reconstruction process to obtain a reconstructed difference tensor of the ecological network structure before and after the disturbance. It can be understood that this step first takes the original graph structure of the disturbance-response network as input, where nodes represent ecological patches or functional units, edges are disturbance propagation paths, and edge weights are provided by the propagation weight matrix, which expresses the intensity and susceptibility of the disturbance signal propagating along each path. Each node in the set of ecological intermediary nodes is regarded as a key disturbance entry point, and its state of being weakened, invalidated, or blocked during a disturbance event is simulated to construct a disturbance graph. This is done by pruning or attenuating the connection structure of the intermediary nodes based on the original graph.

[0063] The original and perturbed graphs are then simultaneously fed into a weighted graph convolutional autoencoder framework. This model consists of a graph convolutional encoder and a graph decoder. The encoder extracts latent representations of nodes through multiple layers of graph convolution, integrating their functional attributes with structural adjacency features. The decoder attempts to reconstruct the adjacency matrix or edge weight tensor based on the encoded representations. The model is trained on two graphs separately, learning the structural differences before and after the perturbation and outputting the difference in connectivity reconstruction between each pair of nodes in the two states.

[0064] In order to more clearly express the spatial distribution characteristics of perturbation reversibility, the system defines the structural reconstruction difference tensor: , where the D structure reconstructs the difference tensor, a three-dimensional tensor structure with dimensions of N X N X T , N Indicates the total number of network nodes, T represents the number of time steps or perturbation simulation stages, represents the structural reconstruction difference between node i and node j at the tth perturbation time step, represents the adjacency matrix element value of the tth disturbance time step before the disturbance occurs, represents the value of the adjacency matrix element at the tth perturbation time step after the perturbation occurs. Larger differences in the tensor indicate that the structural connection is sensitive to perturbations and has poor recovery capabilities; smaller differences indicate that the system has strong functional resilience and structural reconfigurability.

[0065] Step S52: performing path perturbation information entropy evolution analysis based on the reconstructed difference tensor, constructing a path information entropy sequence based on node state transition probability, combining the perturbation cascade amplitude change, calculating the entropy diffusion rate of the perturbation propagation path, and obtaining a path entropy backtracking function set; It can be understood that this step first extracts the connection difference values ​​between node pairs in each disturbance propagation path based on the structural reconstruction difference tensor, and constructs the local state space formed by the structural changes before and after the disturbance.

[0066] Next, the system defines its information entropy evolution sequence based on the path: ,in, represents the structural perturbation information entropy of path p at the tth perturbation time step, represents all adjacent node pairs in path p, represents the normalized value of the connection difference between the node pair (i, j) in the perturbation step t, and p represents a certain perturbation propagation path.

[0067] To further measure the rate at which disturbances propagate and dissipate along different paths, the system introduces the entropy diffusion rate metric, performing first-order differences and trend fitting on the entropy sequence of each path. A high perturbation information entropy indicates rapid diffusion along the path and high recovery difficulty, while a low perturbation information entropy indicates slow dissipation of path information or a certain degree of structural resilience.

[0068] Ultimately, the system packages each path's entropy sequence, diffusion rate, stability volatility, and other parameters, outputting a set of path entropy backtest functions. This set will serve as the basis for identifying community reconstruction trends, fitting functional decay curves, and zoning risk in subsequent steps.

[0069] Step S53: performing community response trend clustering processing according to the path entropy backtracking function set, and jointly embedding the entropy change rate and community structure redundancy within the time window into a preset sparse representation model to extract the structural response pattern and obtain the community reconstruction trend index; It can be understood that this step first performs time window warping on the information entropy evolution sequence of each path to ensure temporal comparability across different paths. To this end, a dynamic time warping algorithm is introduced to align the entropy curves across multiple windows, addressing the scale differences caused by hysteresis and asynchrony in path perturbation responses.

[0070] The system then uses the entropy evolution sequence of each path as a feature vector input and employs a pre-defined sparse representation model to perform low-rank encoding on all paths. By optimizing the reconstruction error between paths and the sparsity of the dictionary, the system extracts the most discriminative perturbation trend subspace. Each principal component axis in this subspace represents a typical perturbation evolution pattern, such as "rapid diffusion-slow recovery," "persistent high-entropy oscillation," and "local rebound stability."

[0071] The system then projects each path onto this perturbation trend subspace and performs spectral clustering, classifying them into several perturbation response categories based on the inter-path similarity matrix. For each path category, the system reverse-maps the node community area it covers to generate a community response trend label. The system then calculates the fluctuations in structural connectivity, residual functional stability, and reversibility of the response during the perturbation process to form a community reconstruction trend index.

[0072] Among them, the community reconstruction trend index is as follows: ,in, CRTI (C k ) represents community C k The reconstruction trend index, Represents community C k The fluctuation amplitude of the structural connectivity during the disturbance process, Res (C k ) represents community C kThe functional stability residual ratio, Rev (C k ) represents community C k is the response reversibility index, α, β, and γ are weight parameters, indicating the relative importance of the three factors.

[0073] This index is comprehensively assigned based on the disturbance intensity at the path level, functional redundancy at the community level, and structural volatility. The higher the value, the greater the possibility that the community in the area will undergo reorganization, breakage, or irreversible migration during disturbance, and the more complex the ecological evolution trend; the lower the value, the greater the potential for the system to return to steady state or functional conservation.

[0074] Step S54: quantify the ecological function stability according to the community reconstruction trend index, and obtain a functional stability attenuation curve by constructing a multi-scale response amplitude function and performing wavelet reconstruction analysis.

[0075] It can be understood that this step first collects the functional variable time series data of each community node during the disturbance simulation process, including but not limited to species functional group (such as nitrogen fixation, water regulation) abundance changes, functional redundancy, system interaction (such as species coexistence, energy flow density) and other dimensions, and uses them as a multi-channel time series tensor , where n is the number of communities, d is the functional dimension, and T is the number of perturbation time steps.

[0076] The system uses principal component analysis to reduce the dimensionality of each community's functional time series, extracting the disturbance-dominated functional fluctuation pattern and constructing a one-dimensional representative variable. Wavelet transform is then introduced to analyze the series, extracting its changing trends and mutation points on multiple time scales, thereby constructing a continuous function curve describing the evolution of functional status over time, defined as: , where S k (t) represents community C k The continuous curve of functional state evolution, represents the fluctuation sequence of the dominant function of the community, represents the wavelet basis function, with scale a and translation b. Represents the integrand time variable, which is used to represent the historical information of the original sequence integration, t represents the current time step, and T is the number of disturbance time steps.

[0077] Step S6: Perform ecological response sustainability analysis based on the community reconstruction trend index and the functional stability attenuation curve to obtain a time series biodiversity impact index.

[0078] It can be understood that this step integrates the community reconstruction trend index and the functional stability attenuation curve generated in the previous step to carry out an ecological response continuity analysis of the dynamic change trend of the ecosystem after the disturbance. The goal is to ultimately construct an indicator function that can continuously reflect the intensity, duration and functional carrying capacity of the disturbance impact, that is, the time series biodiversity impact index. In this step, step S6 includes step S61, step S62, step S63 and step S64.

[0079] Step S61: performing ecological response delay modeling based on the community reconstruction trend index and the functional stability attenuation curve, constructing a variable structure state equation based on a time-delay nonlinear dynamic system to capture the delayed response time period of the biological community after the disturbance, and obtaining a multi-time-scale ecological response delay function set; It can be understood that this step first constructs a two-dimensional input tensor based on the reconstruction trend index of each community and its corresponding stability decay curve, which serves as the time-space function input of the ecosystem dynamic response. On this basis, the system introduces a time-delay nonlinear dynamic system modeling framework to construct the response function: ,in, Represents community C k The response function, Represents community C k The continuous curve of functional state evolution, λ represents the natural attenuation coefficient, the rate at which the control system returns to the stable state, μ represents the response inertia weight, which affects the feedback strength of the time lag term on the current state, and σ represents the sigmoid activation function, which is used for nonlinear modeling. is the integrand time variable, which represents the historical information of the original sequence integration, η represents the external driving strength coefficient, which controls the direct impact of the input tensor on the response, X k，t Represents community C k The combined trend-decay input value at time step t.

[0080] Then, the delayed response functions of all nodes are unified and organized to form a multi-time-scale ecological response delay function set, which is used to support the disturbance impact prediction window, key turning point identification and recovery time estimation.

[0081] Step S62: performing a disturbance effect persistence estimation process based on the set of ecological response delay functions, and obtaining a distribution map of the ecological effect persistence at different stages after the disturbance by calculating the multi-window alignment error between the trend index and the stability curve under dynamic time warping; It can be understood that in this step, the system first performs a multi-time window sliding analysis on each delay function, introduces the window length and disturbance response threshold, and calculates whether the delayed response value of each community in different time periods exceeds the ecological disturbance sensitivity threshold. The following function is constructed for each community node: , where Dk (t) represents community C k Whether it is in the high-sensitivity disturbance response state at time t, 1 means yes, 0 means no, ω means the sliding window length (unit: time step), z means the number of offset steps in the sliding window, S k (tz) represents community C k The response delay state value at time step tz, θ represents the sensitive threshold value of the ecological disturbance response (such as the preset structure-function tolerance upper limit).

[0082] This function outputs whether the disturbance effect is persistently active for each time period. After the sliding window is shifted, a binary sequence representing persistent disturbance activity is formed, which can then be used to calculate time features such as duration, frequency, and maximum span.

[0083] Next, the system introduced a dynamic time warping method to align and compare the disturbance delay functions between different communities, identify which communities have similar disturbance effect persistence patterns, and construct a clustering heat map based on function similarity to reveal the synchronous / asynchronous diffusion pattern of disturbance effects in space.

[0084] Step S63: reconstructing species functional contributions based on the ecological effect persistence distribution map, identifying key functional groups in each persistence stage by constructing a community response clustering model based on fuzzy hierarchical weighted clustering, and obtaining a set of biodiversity structure vectors under each time slice; It can be understood that this step is based on the disturbance response persistence data and the species functional performance matrix to construct a synergistic change correlation matrix between each species and the community disturbance response.

[0085] Then, the functional contribution function of each species during the disturbance duration period is defined as: , where G w represents the overall functional contribution of species w during the disturbance duration period, C k,w Indicates that species w is present in community C k Functional correlation during perturbation response, represents the functional performance vector of species w, represents the functional response intensity of the community at time step t, represents the disturbance-sensitive time period of the community, and k represents the kth community.

[0086] By clustering the functional contribution function curves of each species, species with high functional response, high stability or delayed functional recovery are identified; and key functional supporting species are screened out by combining indicators such as betweenness centrality and functional redundancy in the community structure diagram to form a restoration priority intervention species pool (key functional groups). Then, the collective response of the member species under each functional group in the disturbance scenario is integrated into a vector representation, thereby generating a set of biodiversity structure vectors under the time slice.

[0087] Step S64: construct a time series index based on the biodiversity structure vector set to obtain a biodiversity impact index.

[0088] It's understandable that this system defines a weighted aggregation function at each time step, fusing the structural state vectors from several previous time slices. It also introduces a disturbance sensitivity factor as a tuning factor, allowing the index construction process to adaptively adjust the smoothing window weights when the disturbance intensity fluctuates significantly. This mechanism ensures that the generated exponential curve reflects both the system's short-term dramatic response and the long-term structural trend, ultimately outputting the resulting biodiversity impact index.

[0089] The biodiversity impact index is as follows: , Where BII(t) represents the biodiversity impact index, which indicates the functional stability of the ecosystem relative to the reference state at time step t, with a value range of [0,1]. t represents the normalization factor, s represents any time step index in the time window, represents the time window length, α s Denotes the disturbance sensitivity weighting factor, V s represents the biodiversity structure vector, V ref represents the reference state structure vector.

[0090] Example 2

[0091] like Figure 2 As shown, this embodiment provides a dynamic assessment system for the impact of engineering construction on biodiversity, which includes an acquisition unit 701 , a processing unit 702 , a simulation unit 703 , an identification unit 704 , a backtracking unit 705 and an analysis unit 706 .

[0092] An acquisition unit 701 is used to acquire a basic ecological data set, wherein the basic ecological data set includes a water system distribution map, a forest wetland patch structure change map, and multi-time period sample community data; Processing unit 702 is used to construct an ecological patch graph structure and perform hydrological connectivity edge processing based on the basic ecological dataset to obtain a composite ecological graph model that reflects spatial structural variation; The simulation unit 703 is used to perform patch functional heterogeneity aggregation and disturbance link simulation processing according to the composite ecological graph model to obtain the ecological transmission path and the sensitivity spectrum of each node; The identification unit 704 is used to perform spectrum domain decomposition and key propagation node identification processing based on the ecological propagation path and the sensitivity spectrum of each node, and obtain the ecological intermediary nodes and propagation weights in the disturbance response network; The backtracking unit 705 is used to perform reversibility modeling and path entropy backtracking analysis based on the ecological intermediary nodes and propagation weights to obtain a community reconstruction trend index and a functional stability attenuation curve; The analysis unit 706 is configured to perform ecological response persistence analysis based on the community reconstruction trend index and the functional stability decay curve to obtain a time series biodiversity impact index.

[0093] It should be noted that, regarding the system in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0094] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

[0095] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A dynamic assessment method for the impact of engineering construction on biodiversity, characterized by: include: Obtaining basic ecological data sets, including water system distribution maps, forest wetland patch structure change maps, and multi-time period sample plot community data; Based on the basic ecological dataset, ecological patch graph structure construction and hydrological connectivity edge processing are performed to obtain a composite ecological graph model that reflects spatial structural variation; Based on the composite ecological graph model, patch functional heterogeneity aggregation and disturbance link simulation were performed to obtain the ecological transmission path and the sensitivity spectrum of each node; According to the ecological propagation path and the sensitivity spectrum of each node, spectral domain decomposition and key propagation node identification are performed to obtain the ecological intermediary nodes and propagation weights in the disturbance response network; According to the ecological intermediary nodes and propagation weights, reversibility modeling and path entropy back-analysis were performed to obtain the community reconstruction trend index and functional stability attenuation curve; Based on the community reconstruction trend index and the functional stability decay curve, an ecological response persistence analysis is performed to obtain a time series biodiversity impact index.

2. The dynamic evaluation method according to claim 1, characterized in that ,Based on the basic ecological dataset, the ecological patch graph structure ,and hydrological connectivity edge processing are carried out, ,including: Based on the water system distribution map in the basic ecological dataset and the terrain data of the preset digital elevation model, hydrological connectivity identification processing is carried out. The minimum cumulative cost path of surface runoff is calculated based on the CostDistance model to obtain the hydrological connectivity attenuation matrix between patches. An ecological node graph is constructed based on the hydrological connectivity attenuation matrix. The forest wetland patch structure change map and multi-period sample community data are fused through a regional fusion method to extract spatial node sets representing different functional groups, thereby obtaining an ecological node graph with community function labels. Based on the ecological node graph and the hydrological connectivity attenuation matrix, edge weight modeling is performed, and multivariate heterogeneous edges are constructed through the spectral clustering algorithm to obtain an ecological relationship graph with structure-function joint information; The graph structure disturbance expression processing is performed according to the ecological relationship graph. By introducing a gated graph neural network to dynamically modulate the edge weight changes, the changing trend of the connection strength of forest wetland patches before and after geological disaster control is captured, and a composite ecological graph model reflecting spatial structure variation is obtained.

3. The dynamic evaluation method according to claim 1, characterized in that ,According to the composite ecological graph model, patch functional heterogeneity ,aggregation and disturbance link simulation processing are carried out, ,including: Extracting patch function indicators based on the composite ecological graph model, and applying a feature coding mechanism based on community function redundancy, species sensitivity, and habitat rarity to graph nodes to obtain node function feature tensors; Based on the node functional feature tensor, patch functional heterogeneity aggregation processing is performed, and the nodes are divided into functional response subgroups by using an adaptive fuzzy C-means clustering algorithm, and the aggregation boundary is adjusted in combination with the edge density and local connectivity in the graph structure to obtain a functional heterogeneity subgraph set; A disturbance link simulation process is performed based on the functional heterogeneity subgraph set. By constructing a random walk-based structural disturbance model and combining it with construction disturbance time series data to simulate path crossing probabilities, a potential disturbance propagation path network is obtained. The sensitivity spectrum identification processing is performed based on the disturbance propagation path network. The maximum incremental rate of node edge weight evolution and functional contribution are jointly ranked, and the high-responsiveness node subsets are divided using spectral clustering to obtain the set of highly sensitive nodes in the ecological propagation path and their sensitivity spectrum.

4. The dynamic evaluation method according to claim 1, characterized in that ,According to the ecological propagation path and the sensitivity spectrum of each node, spectral domain decomposition and key propagation node identification processing are carried out, including: According to the ecological propagation path and the sensitivity spectrum of each node, the Laplace spectrum decomposition of the propagation path graph is performed. By constructing the normalized Laplace matrix of the perturbation propagation graph and calculating its eigenvalue-eigenvector pairs, a spectral domain feature set for path energy distribution identification is obtained; Calculating the energy concentration of propagation nodes based on the graph domain feature set, identifying propagation path segments with significantly concentrated spectral energy through the maximum spectral energy density projection method, and obtaining a local path subgraph with a disturbance aggregation effect greater than a preset threshold; According to the local path subgraph, ecological intermediary nodes are identified and processed, and through the joint screening mechanism of betweenness centrality and path coupling, key nodes with bridging effects in multiple propagation paths are extracted to obtain the set of ecological intermediary nodes in the disturbance response network; Propagation weight modeling is performed based on the set of ecological intermediary nodes. Through a multi-factor data normalization method based on node redundancy and path coupling coefficient, the response weight of each intermediary node on the engineering construction disturbance path is assigned, and the ecological intermediary nodes and propagation weights in the disturbance response network are obtained.

5. The dynamic evaluation method according to claim 1, characterized in that ,According to the ecological intermediary nodes and propagation weights, reversibility modeling and path entropy backtracking analysis are carried out, including: Based on the ecological intermediary nodes and propagation weights, the perturbation reversibility state modeling process is carried out. By constructing a weighted graph convolutional autoencoder to simulate the perturbation node deletion and reconstruction process, the reconstructed difference tensor of the ecological network structure before and after the perturbation is obtained; Performing path perturbation information entropy evolution analysis and processing based on the reconstructed difference tensor, constructing a path information entropy sequence based on node state transition probability, combining the perturbation cascade amplitude change, calculating the entropy diffusion rate of the perturbation propagation path, and obtaining a path entropy backtracking function set; Performing community response trend clustering processing based on the path entropy backtracking function set, jointly embedding the entropy change rate and community structure redundancy within the time window into a preset sparse representation model, extracting the structural response pattern therein, and obtaining a community reconstruction trend index; The ecological function stability is quantified based on the community reconstruction trend index, and the functional stability attenuation curve is obtained by constructing a multi-scale response amplitude function and performing wavelet reconstruction analysis.

6. A dynamic assessment system for the impact of engineering construction on biodiversity, characterized by: include: An acquisition unit is used to acquire a basic ecological data set, wherein the basic ecological data set includes a water system distribution map, a forest wetland patch structure change map, and multi-time period sample community data; A processing unit is used to construct an ecological patch graph structure and perform hydrological connectivity edge processing based on the basic ecological dataset to obtain a composite ecological graph model that reflects spatial structural variation; A simulation unit is used to perform patch functional heterogeneity aggregation and disturbance link simulation processing according to the composite ecological graph model to obtain the ecological transmission path and the sensitivity spectrum of each node; an identification unit, configured to perform spectrum domain decomposition and key propagation node identification processing based on the ecological propagation path and the sensitivity spectrum of each node, and obtain ecological intermediary nodes and propagation weights in the disturbance response network; A backtracking unit is used to perform reversibility modeling and path entropy backtracking analysis based on the ecological intermediary nodes and propagation weights to obtain a community reconstruction trend index and a functional stability attenuation curve; The analysis unit is used to perform ecological response continuity analysis and processing based on the community reconstruction trend index and the functional stability decay curve to obtain a time series biodiversity impact index.

7. The dynamic evaluation system according to claim 6, characterized in that: The processing unit includes: The first processing subunit is used to perform hydrological connectivity identification processing based on the water system distribution map in the basic ecological dataset and the terrain data of the preset digital elevation model, and calculate the minimum cumulative cost path of surface runoff based on the CostDistance model to obtain the hydrological connectivity attenuation matrix between patches; The second processing subunit is used to construct an ecological node graph based on the hydrological connectivity attenuation matrix, fuse the forest wetland patch structure change map and the multi-period sample community data through a regional fusion method, extract the spatial node sets representing different functional groups, and obtain an ecological node graph with community function labels; The third processing subunit is used to perform edge weight modeling based on the ecological node graph and the hydrological connectivity attenuation matrix, construct multivariate heterogeneous edges through a spectral clustering algorithm, and obtain an ecological relationship graph with structure-function joint information; The fourth processing sub-unit is used to perform graph structure disturbance expression processing based on the ecological relationship graph, and by introducing a gated graph neural network to dynamically modulate and learn the edge weight changes, it captures the changing trend of the connection strength of forest wetland patches before and after geological disaster control, and obtains a composite ecological graph model that reflects spatial structure variation.

8. The dynamic evaluation system according to claim 6, characterized in that: The simulation unit comprises: The first simulation subunit is used to extract and process patch function indicators based on the composite ecological graph model, and obtain node function feature tensors by applying a feature coding mechanism based on community function redundancy, species sensitivity and habitat rarity on the graph nodes; The second simulation subunit is used to perform patch functional heterogeneity aggregation processing based on the node functional feature tensor, divide the nodes into functional response subgroups by using an adaptive fuzzy C-means clustering algorithm, and adjust the aggregation boundary in combination with the edge density and local connectivity in the graph structure to obtain a functional heterogeneity subgraph set; The third simulation subunit is used to perform disturbance link simulation processing based on the functional heterogeneity subgraph set, construct a structural disturbance model based on random walk, and simulate the path crossing probability in combination with the construction disturbance time series data to obtain a potential disturbance propagation path network; The fourth simulation subunit is used to perform sensitivity spectrum identification processing based on the disturbance propagation path network, jointly rank the maximum incremental rate of node edge weight evolution and functional contribution, use spectral clustering to divide the high-responsiveness node subsets, and obtain the high-sensitive node set and its sensitivity spectrum in the ecological propagation path.

9. The dynamic evaluation system according to claim 6, characterized in that: The identification unit includes: The first identification subunit is used to perform Laplace spectrum decomposition of the propagation path graph according to the ecological propagation path and the sensitivity spectrum of each node, and obtain a spectral domain feature set for path energy distribution identification by constructing a normalized Laplace matrix of the perturbation propagation graph and calculating its eigenvalue-eigenvector pairs; The second identification sub-unit is used to calculate the energy concentration of the propagation nodes according to the graph domain feature set, identify the propagation path segments with significantly concentrated spectral energy through the maximum spectral energy density projection method, and obtain a local path subgraph with a disturbance aggregation effect greater than a preset threshold; The third identification subunit is used to identify ecological intermediary nodes based on the local path subgraph, extract key nodes with bridging effects in multiple propagation paths through a joint screening mechanism of betweenness centrality and path coupling, and obtain a set of ecological intermediary nodes in the disturbance response network; The fourth identification subunit is used to perform propagation weight modeling processing based on the set of ecological intermediary nodes, and to assign the response weight of each intermediary node on the engineering construction disturbance path through a multi-factor data normalization method based on node redundancy and path coupling coefficient, so as to obtain the ecological intermediary nodes and propagation weights in the disturbance response network.

10. The dynamic evaluation system according to claim 6, characterized in that: The tracing back unit includes: The first backtracking subunit is used to perform disturbance reversibility state modeling based on the ecological intermediary nodes and propagation weights, and to simulate the disturbance node deletion and reconstruction process by constructing a weighted graph convolutional autoencoder to obtain a reconstructed difference tensor of the ecological network structure before and after the disturbance; A second backtracking subunit is configured to perform path perturbation information entropy evolution analysis based on the reconstructed difference tensor, calculate the entropy diffusion rate of the perturbation propagation path by constructing a path information entropy sequence based on the node state transition probability and combining the perturbation cascade amplitude change, and obtain a path entropy backtracking function set; The third backtracking subunit is used to perform community response trend clustering processing based on the path entropy backtracking function set, and extract the structural response pattern by jointly embedding the entropy change rate and community structure redundancy within the time window into a preset sparse representation model to obtain the community reconstruction trend index; The fourth backtracking subunit is used to quantify the stability of ecological functions according to the community reconstruction trend index, and obtain a functional stability attenuation curve by constructing a multi-scale response amplitude function and performing wavelet reconstruction analysis.

Images