Beach vegetation habitat dynamic partition determination method based on hydrological interference intensity

Abstract

The invention relates to the technical field of hydrology and ecology, in particular to a method and a system for determining a dynamic partition of a beach vegetation habitat based on hydrology interference intensity and a medium, and discloses the method for determining the dynamic partition of the beach vegetation habitat based on the hydrology interference intensity. Comprising the following steps: dividing grids for a beach effective range of a research area to generate a grid layer; calculating eight hydrological interference indexes of each grid unit in the grid layer; carrying out normalization processing on the hydrological interference indexes obtained by calculation, calculating the weight of each index by adopting an entropy weight method, and generating a weighted feature matrix; performing principal component dimensionality reduction on the weighted feature matrix, and reserving a principal component matrix with an accumulated variance contribution rate greater than or equal to 85%; and based on the principal component matrix, determining an optimal clustering number in a clustering number range of 3-7 through a Gaussian mixture model, and dividing the grid layer into a plurality of functional areas according to a clustering result.

Description

Technical Field

[0001] This disclosure relates to the field of hydrological and ecological technology, specifically to a method for determining the dynamic zoning of riparian vegetation habitats based on the intensity of hydrological disturbance. Background Technology

[0002] River and lake banks are key components of river and lake ecosystems, situated in transitional zones between water and land, exhibiting typical characteristics of aquatic-terrestrial ecotones. Their formation and evolution are controlled by various natural processes, particularly hydrodynamic factors such as flow velocity, water level, inundation frequency, and wave disturbance, which play a decisive role in the geomorphology, sedimentary structure, vegetation distribution, and ecosystem stability of the banks. Because bank areas possess both fluidity and ecological characteristics, serving as crucial interfaces for material migration, energy exchange, and biological diffusion, they play an irreplaceable role in maintaining the integrity of river and lake ecosystems and enhancing their aquatic ecosystem services.

[0003] River and lake banks are not only ecologically sensitive areas and biodiversity hotspots, but also important carriers for flood control, protection and buffering, landscape ecology, and resource development. In actual management and restoration, riverbank areas often face multiple intertwined needs, such as ecological protection, water security, land use, and landscape development. How to achieve refined management and efficient and intensive utilization of riverbank space has become the core issue of current river and lake system governance and ecological restoration.

[0004] Riparian vegetation, as an important component of river and lake ecosystems, encompasses a wide range of communities, from emergent, floating, and submerged plants to riparian terrestrial plants. These plants exhibit a clear spatial distribution gradient from the riverbanks to the main channel, reflecting their varying adaptations to water level fluctuations, hydrodynamic disturbances, and substrate characteristics. Through mechanisms such as intercepting water flow, slowing flow velocity, stabilizing substrate, promoting sedimentation, and regulating water quality, these plants not only directly influence the hydrodynamic structure and material transport processes of rivers but also construct diverse habitat types, making them key elements in maintaining the stability and functional diversity of river ecosystems.

[0005] In riparian areas, hydrodynamic processes are the dominant factors controlling the spatial distribution pattern of vegetation and community succession. Differences in hydrodynamic parameters such as flow velocity, water depth, flood frequency, inundation duration, shear intensity, and wave disturbance lead to significant variations in vegetation composition, cover, root development, and reinforcement capacity across different riparian sections. Hydrodynamic conditions further determine the suitable habitats, growth status, and recovery potential of various plant species by influencing factors such as substrate particle size, substrate stability, and wetting cycles. Therefore, establishing a dynamic zoning method based on hydrodynamic disturbances is a crucial foundation for identifying suitable vegetation habitats, constructing stable ecological zones, and optimizing riparian space utilization.

[0006] However, existing methods for riparian zoning often rely on elevation zones, land use types, or administrative boundaries, making it difficult to reflect the spatial variability of hydrodynamic disturbances and their actual driving effect on vegetation habitats. This hinders the development of differentiated and ecological governance strategies. In practical ecological engineering, riparian areas are often designed and managed as a whole, neglecting the impact of hydrodynamic gradients on habitat supply and community stability. This leads to homogenized governance solutions, extensive habitat configuration, slow ecological function recovery, and unstable results.

[0007] Therefore, there is an urgent need to establish a dynamic zoning method based on hydrodynamic processes, integrating topographic evolution patterns and vegetation habitat response mechanisms, in order to scientifically guide the differentiated protection and restoration of riparian ecosystems. Summary of the Invention

[0008] To address the problems in related technologies, this disclosure provides a method for determining the dynamic zoning of riparian vegetation habitats based on the intensity of hydrological disturbance.

[0009] In a first aspect, embodiments of this disclosure provide a method for determining the dynamic zoning of riparian vegetation habitats based on the intensity of hydrological disturbance, comprising the following steps: To divide the effective area of ​​the beach in the study area into grids and generate a grid layer; Calculate eight hydrological disturbance indicators for each grid cell in the grid layer: maximum flow velocity, annual inundation frequency, maximum inundation depth, bank slope, erosion rate, median grain size, annual deposition rate, and connectivity index. The calculated hydrological disturbance indices are normalized, and the weights of each index are calculated using the entropy weight method to generate a weighted feature matrix. Principal component dimensionality reduction is performed on the weighted feature matrix to retain the principal component matrix with a cumulative variance contribution rate ≥ 85%; Based on the principal component matrix, the optimal number of clusters is determined within the range of 3-7 using a Gaussian mixture model, and the grid layer is divided into multiple functional areas according to the clustering results.

[0010] According to embodiments of this disclosure, the step of dividing the effective area of ​​the beach in the study region into grids and generating a grid layer includes: Generate a polygon representing the effective extent of the beach area in the study region; Establish a standard coordinate system based on the geometric features of the beach; Based on the effective area polygon of the beach, an initial mesh surface layer is generated in a standard coordinate system; The area outside the effective range of the beach in the initial grid surface layer is masked to obtain an effective grid layer; A grid layer is generated by assigning attributes to each grid cell in the effective grid layer; the assigned attributes include center point coordinates, spatial boundary coordinates, and average elevation.

[0011] According to the embodiments of this disclosure, the maximum flow velocity is taken as the maximum value of the measurement data from the dry season, the normal water season, and the wet season; The annual inundation frequency was obtained by comparing the water levels of hydrological stations with grid elevations on a daily basis. The maximum inundation depth is taken as the maximum value of the difference between the daily water level and the grid elevation throughout the year; The slope of the bank is obtained by extracting the slope of the neighboring area through the elevation raster and then taking the arithmetic mean. The erosion rate was calculated using high-resolution remote sensing of shoreline changes. The median particle size was obtained by laser particle size analysis of surface sediments. The annual deposition rate is calculated using the year-end depth increment of the deposition scale. The connectivity index is obtained by statistically analyzing the percentage of days when the daily water level exceeds the grid elevation plus a threshold.

[0012] According to embodiments of this disclosure, the step of determining the optimal number of clusters within the range of 3-7 using a Gaussian mixture model based on the principal component matrix, and dividing the grid layer into multiple functional areas according to the clustering results, includes: Based on the principal component matrix, the optimal number of clusters was determined to be 5 within the range of 3-7 using a Gaussian mixture model, and the beach area was divided into five functional zones: scour core zone, active erosion zone, periodic flooding zone, sedimentation buffer zone, and land margin zone.

[0013] According to embodiments of this disclosure, dividing the grid layer into multiple functional areas based on clustering results includes: Perform clustering with the optimal number of clusters as a parameter, and output the cluster label for each grid cell; Based on the cluster labels, isolated patches with an area of ​​less than 3 grid cells in the grid layer are identified, and then the isolated patches are reclassified according to the cluster label value with the most adjacent clusters. Calculate the weighted average of the eight hydrological disturbance indicators within each cluster, sort the results from largest to smallest, and divide the functional area of ​​each cluster according to the first two indexes of the sort.

[0014] According to embodiments of this disclosure, it further includes: The spatial structure of the multiple functional zones is quantified based on the landscape pattern index.

[0015] According to embodiments of this disclosure, it further includes: At least three representative indicators for each functional area are selected to calculate the weighted average. After determining the weights using the entropy weight method, the health index of the area is calculated.

[0016] According to embodiments of this disclosure, it further includes: The overall health index of the entire beach is calculated based on the area proportion of each functional zone and the health index of the respective zones.

[0017] In a second aspect, embodiments of this disclosure provide a system for implementing any of the methods of the first aspect, comprising: The mesh generation module is used to divide the effective area of ​​the beach in the study area into mesh layers. The index acquisition module is used to calculate eight hydrological disturbance indices for each grid cell in the grid layer: maximum flow velocity, annual inundation frequency, maximum inundation depth, bank slope, erosion rate, median grain size, annual deposition rate, and connectivity index. The weight calculation module is used to normalize the calculated hydrological disturbance indicators, calculate the weight of each indicator using the entropy weight method, and generate a weighted feature matrix. The dimensionality reduction module is used to perform principal component dimensionality reduction on the weighted feature matrix, retaining the principal component matrix with a cumulative variance contribution rate ≥ 85%. The partition generation module is used to determine the optimal number of clusters in the range of 3-7 based on the principal component matrix using a Gaussian mixture model, and to divide the grid layer into multiple functional areas according to the clustering results.

[0018] According to embodiments of this disclosure, the step of dividing the effective area of ​​the beach in the study region into grids and generating a grid layer includes: Generate a polygon representing the effective extent of the beach area in the study region; Establish a standard coordinate system based on the geometric features of the beach; Based on the effective area polygon of the beach, an initial mesh surface layer is generated in a standard coordinate system; The area outside the effective range of the beach in the initial grid surface layer is masked to obtain an effective grid layer; A grid layer is generated by assigning attributes to each grid cell in the effective grid layer; the assigned attributes include center point coordinates, spatial boundary coordinates, and average elevation.

[0019] According to the embodiments of this disclosure, the maximum flow velocity is taken as the maximum value of the measurement data from the dry season, the normal water season, and the wet season; The annual inundation frequency was obtained by comparing the water levels of hydrological stations with grid elevations on a daily basis. The maximum inundation depth is taken as the maximum value of the difference between the daily water level and the grid elevation throughout the year; The slope of the bank is obtained by extracting the slope of the neighboring area through the elevation raster and then taking the arithmetic mean. The erosion rate was calculated using high-resolution remote sensing of shoreline changes. The median particle size was obtained by laser particle size analysis of surface sediments. The annual deposition rate is calculated using the year-end depth increment of the deposition scale. The connectivity index is obtained by statistically analyzing the percentage of days when the daily water level exceeds the grid elevation plus a threshold.

[0020] According to embodiments of this disclosure, the step of determining the optimal number of clusters within the range of 3-7 using a Gaussian mixture model based on the principal component matrix, and dividing the grid layer into multiple functional areas according to the clustering results, includes: Based on the principal component matrix, the optimal number of clusters was determined to be 5 within the range of 3-7 using a Gaussian mixture model, and the beach area was divided into five functional zones: scour core zone, active erosion zone, periodic flooding zone, sedimentation buffer zone, and land margin zone.

[0021] According to embodiments of this disclosure, dividing the grid layer into multiple functional areas based on clustering results includes: Perform clustering with the optimal number of clusters as a parameter, and output the cluster label for each grid cell; Based on the cluster labels, isolated patches with an area of ​​less than 3 grid cells in the grid layer are identified, and then the isolated patches are reclassified according to the cluster label value with the most adjacent clusters. Calculate the weighted average of the eight hydrological disturbance indicators within each cluster, sort the results from largest to smallest, and divide the functional area of ​​each cluster according to the first two indexes of the sort.

[0022] According to embodiments of this disclosure, it further includes: The spatial structure of the multiple functional zones is quantified based on the landscape pattern index.

[0023] According to embodiments of this disclosure, it further includes: At least three representative indicators for each functional area are selected to calculate the weighted average. After determining the weights using the entropy weight method, the health index of the area is calculated.

[0024] According to embodiments of this disclosure, it further includes: The overall health index of the entire beach is calculated based on the area proportion of each functional zone and the health index of the respective zones.

[0025] Thirdly, embodiments of this disclosure provide a computer-readable storage medium having computer instructions stored thereon, which, when executed by a processor, implement the method as described in any of the first aspects.

[0026] The technical effects provided by the embodiments of this disclosure may include the following beneficial effects: According to the technical solution provided in this disclosure, a method for determining dynamic zoning of riparian vegetation habitat based on hydrological disturbance intensity includes the following steps: dividing the effective range of the riparian area of ​​the study area into grids to generate a grid layer; calculating eight hydrological disturbance indicators for each grid cell in the grid layer: maximum flow velocity, annual inundation frequency, maximum inundation depth, bank slope, erosion rate, median grain size, annual deposition rate, and connectivity index; normalizing the calculated hydrological disturbance indicators, calculating the weight of each indicator using the entropy weight method, and generating a weighted feature matrix; performing principal component dimensionality reduction on the weighted feature matrix, retaining the principal component matrix with a cumulative variance contribution rate ≥ 85%; based on the principal component matrix, determining the optimal number of clusters within the range of 3-7 using a Gaussian mixture model, and dividing the grid layer into multiple functional zones according to the clustering results.

[0027] The aforementioned technical solution divides the riparian area into rectangular grid units and extracts eight key hydrological disturbance indicators, including maximum flow velocity and annual inundation frequency, thus constructing for the first time a multi-dimensional quantitative evaluation system for hydrodynamic, geomorphological, sedimentary, and vegetation responses. This design effectively overcomes the limitations of traditional zoning methods that rely on static elevation zones or administrative boundaries, accurately captures the spatial heterogeneity of dynamic processes such as flow velocity, inundation frequency, and erosion intensity, and scientifically reveals the driving mechanism of hydrological disturbances on vegetation habitat formation.

[0028] Based on the objective weighting method of entropy weighting and principal component dimensionality reduction technology, this invention significantly reduces the computational complexity of multi-dimensional indicators while retaining more than 85% of the feature information. By combining a Gaussian mixture model to adaptively determine the optimal number of partitions within a cluster size range of 3-7, it achieves intelligent identification of five functional zones, including scour core areas and active erosion areas. This data-driven partitioning method completely changes the "one-size-fits-all" extensive management model of existing technologies, providing precise spatial basis for differentiated ecological governance.

[0029] By optimizing patch-level reclassification in spatial smoothing, this invention successfully eliminates isolated patches with an area of ​​less than 3 grids, ensuring the continuity and rationality of the spatial distribution of functional zones and significantly improving the morphological integrity and stability of the ecosystem.

[0030] The final zoning results deeply integrate hydrodynamic processes, topographic evolution, and vegetation response mechanisms, providing a scientific decision-making basis for riverbank ecological restoration projects. This method supports the refined governance needs such as the delineation of ecological protection red lines, site-appropriate planting design, and restoration priority ranking, promoting the transformation of river and lake banks from homogeneous governance to precise regulation based on habitat functional units.

[0031] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0032] Other features, objects, and advantages of this disclosure will become more apparent from the following detailed description of non-limiting embodiments, taken in conjunction with the accompanying drawings.

[0033] Figure 1 A flowchart illustrating a method for determining dynamic zoning of riparian vegetation habitats based on hydrological disturbance intensity, according to an embodiment of the present disclosure.

[0034] Figure 2 A schematic diagram of mesh division according to a specific embodiment of the present disclosure is shown.

[0035] Figure 3 A schematic diagram of elevation distribution in a grid layer according to a specific embodiment of the present disclosure is shown.

[0036] Figure 4 This diagram illustrates the slope distribution of a bank slope in a grid layer according to a specific embodiment of the present disclosure.

[0037] Figure 5 This diagram illustrates the annual flooding frequency distribution in a grid layer according to a specific embodiment of the present disclosure.

[0038] Figure 6 This illustration shows a habitat zoning distribution cloud map based on hydrological disturbance intensity according to a specific embodiment of the present disclosure.

[0039] Figure 7 A structural block diagram of a system for determining dynamic zoning of riparian vegetation habitats based on hydrological disturbance intensity, according to an embodiment of the present disclosure, is shown. Detailed Implementation

[0040] In the following, exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings to enable those skilled in the art to readily implement them. Furthermore, for clarity, portions unrelated to the description of exemplary embodiments have been omitted from the drawings.

[0041] In this disclosure, it should be understood that terms such as “comprising” or “having” are intended to indicate the presence of features, figures, steps, behaviors, components, parts or combinations thereof disclosed in this specification, and are not intended to exclude the possibility of the presence or addition of one or more other features, figures, steps, behaviors, components, parts or combinations thereof.

[0042] It should also be noted that, unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other. This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0043] Existing methods for zoning riverbanks often rely on elevation zones, land use types, or administrative boundaries, making it difficult to reflect the spatial variability of hydrodynamic disturbances and their actual driving effect on vegetation habitats. This hinders the development of differentiated and ecological governance strategies. In practical ecological engineering, riverbank areas are often designed and managed as a whole, neglecting the impact of hydrodynamic gradients on habitat supply and community stability. This results in homogenized governance solutions, extensive habitat configuration, slow ecological function recovery, and unstable outcomes.

[0044] In view of the above-mentioned shortcomings, the method for determining the dynamic zoning of riparian vegetation habitat based on hydrological disturbance intensity provided in this disclosure divides the riparian area into rectangular grid units and extracts eight key hydrological disturbance indicators, such as maximum flow velocity and annual inundation frequency, thus constructing for the first time a multi-dimensional quantitative evaluation system coupling hydrodynamics, geomorphology, and vegetation. This design effectively overcomes the limitations of traditional zoning methods that rely on static elevation zones or administrative boundaries, accurately captures the spatial heterogeneity of dynamic processes such as flow velocity, inundation frequency, and erosion intensity, and scientifically reveals the driving mechanism of hydrological disturbance on vegetation habitat formation.

[0045] Figure 1 A flowchart illustrating a method for determining dynamic zoning of riparian vegetation habitats based on hydrological disturbance intensity, according to an embodiment of the present disclosure.

[0046] like Figure 1 As shown, the method for determining the dynamic zoning of riparian vegetation habitats based on hydrological disturbance intensity includes the following steps: S110: Generate a grid layer by dividing the effective area of ​​the beach in the study area into grids.

[0047] The goal of this step is to divide the beach area into regular rectangular grid cells (or functional process units, FPUs), which includes the following sub-steps: First, remote sensing images and elevation data of the study area are acquired to define the boundaries of the shoreline area that needs to be gridded. For example, the width range of the shoreline is determined by combining remote sensing images and elevation data. Multi-temporal water level lines, such as those for the dry season (low water level) and the wet season (high water level), are overlaid and analyzed. The area between the outermost high water level line and the lowest water level line of the year is extracted as the effective range of the shoreline.

[0048] Then, the number of grid rows and columns is calculated. Based on the spatial scale of the effective shoreline area, such as width and length, the number of grid divisions is determined in the horizontal direction (e.g., perpendicular to the river) and the vertical direction (e.g., along the river). For example, based on the determined shoreline width, the number of grid rows required in the horizontal direction is calculated. The number of grid columns in the vertical direction can be flexibly set according to the length of the study area and the required grid size.

[0049] Next, within the defined effective area of ​​the beach, a regular rectangular grid with a specified number of rows and columns is generated. For example, using the centerline of the shoreline or a main axis determined according to the flow direction as a reference baseline, a regular rectangular grid with a specified number of rows and columns covering the effective area of ​​the beach is generated.

[0050] Finally, the grid boundaries and invalid regions are processed, the effective shoreline grid range is precisely defined, invalid grid cells outside the boundaries are masked, and key attribute information, such as center point coordinates and average elevation, is assigned to each effective grid cell. The output is a grid layer containing the spatial geometric attributes of the effective grid cells.

[0051] S120: Calculate 8 hydrological disturbance indicators for each grid cell in the grid layer: maximum flow velocity, annual inundation frequency, maximum inundation depth, bank slope, erosion rate, median grain size, annual deposition rate, and connectivity index.

[0052] In this disclosed method, for each FPU unit, eight key process variables can be extracted from existing observation data or simulation results to construct a complete multidimensional feature set.

[0053] Specifically, this includes: maximum flow velocity ( : Represents the maximum hydrodynamic intensity experienced by a unit, reflecting the maximum scour capacity at that location, and is a key indicator for identifying the core scour zone; annual flooding frequency ( : Measures the frequency of grid flooding, determining the category of plant and sediment response; maximum flooding depth ( ): Reflects the most extreme inundation conditions and has important constraints on vegetation carrying capacity and siltation processes; bank slope ( ): Determines slope stability and stormwater runoff convergence patterns, influencing the difficulty of vegetation restoration; erosion rate ( ): Quantifies shoreline stability to identify the extent of ongoing degradation; Annual sedimentation rate (SAR), indicating siltation intensity, is a direct measure for identifying sedimentary buffer zones; Median grain size ( ): 50% cumulative particle size of the sediment sample volume distribution; Connectivity Index (CCI): measures ecological connectivity and material exchange potential.

[0054] The aforementioned eight hydrological disturbance indicators can be categorized into hydrodynamic indicators (e.g., flow velocity / inundation-related indicators), geomorphological indicators (e.g., slope / erosion-related indicators), sedimentary indicators (e.g., grain size / SAR-related indicators), and vegetation response indicators (e.g., connectivity index, CCI). By integrating these hydrodynamic, geomorphological, sedimentary, and vegetation response indicators, the problem of extensive zoning based on empirical thresholds or single indicators is solved, providing a reliable technical means for the design and refined management of shoreline ecological engineering.

[0055] Among the above eight hydrological disturbance indicators, the maximum flow velocity is the most significant. Annual flooding frequency Maximum flooding depth Annual sedimentation rate (SAR) and connectivity index (CCI) are positive indicators; higher values ​​indicate better habitat conditions or more active processes. Bank slope... erosion rate and median particle size It is a negative indicator; the larger the value, the worse the habitat conditions or the higher the risk.

[0056] The above 8 indicators are recorded in matrix form to construct an n×8 original feature matrix, where n is the number of effective functional units and 8 is the indicator dimension.

[0057] S130: The calculated hydrological disturbance indices are normalized, and the weights of each index are calculated using the entropy weight method to generate a weighted feature matrix.

[0058] In this disclosure method, positive indicators ( , , SAR, CCI) and negative indicators ( , , The range normalization method is used for each indicator. Then, the entropy weight method is introduced to assign weights to each indicator, and the weight vector w j The matrix is ​​automatically generated by the information entropy function, avoiding interference from subjective factors. The normalized matrix is ​​then multiplied by the weights to form the weighted feature matrix v. i .

[0059] Specifically, the eight extracted indicators ( (where i is the grid number and j is the index number) are sequentially normalized, weighted using the entropy weight method to calculate the weights, and a weighted feature vector is generated, specifically including: The positive indicator is calculated using the following formula: .

[0060] The negative index is calculated using the following formula: .

[0061] Where min i and max i These are the minimum and maximum values ​​of the j-th index across all grids, respectively. These are the indicators after normalization.

[0062] Calculate the weight of the j-th index across all grids: .

[0063] Calculate information entropy: .

[0064] Calculate the coefficient of variation of the indicators: .

[0065] Normalization yields the entropy weight. .

[0066] For each grid cell i, construct an 8-dimensional weighted vector. .

[0067] All v i Form an n×8 weighted feature matrix.

[0068] S140: Perform principal component dimensionality reduction on the weighted feature matrix and retain the principal component matrix with a cumulative variance contribution rate ≥ 85%.

[0069] In this disclosure, principal component analysis (PCA) is used to extract the principal structure, retain the first three principal components, and construct a principal component matrix P with a dimensionality reduction matrix of n×3, achieving a cumulative explanatory power of over 85%.

[0070] S150: Based on the principal component matrix, the optimal number of clusters is determined within the range of 3-7 using a Gaussian mixture model, and the grid layer is divided into multiple functional areas according to the clustering results.

[0071] In this disclosure, a Gaussian Mixture Model (GMM) is introduced for unsupervised automatic clustering of the Free Function Process (FPU). To avoid dispersion and oversimplification, GMMs are constructed based on the principal component matrix P and the Bayesian Information Criterion (BIC) value is calculated within the range of 3 to 7 cluster numbers k. The k value that minimizes the BIC value is selected. As the optimal number of clusters, k The Gaussian mixture model is initialized with cluster numbers, and the principal component matrix P is fitted. Based on the clustering results, all functional units are divided into multiple functional regions, corresponding to different shoreline process characteristics such as scour core areas and sedimentation buffer zones.

[0072] The following section explains the technical details of the above process steps using ArcGIS software.

[0073] According to an embodiment of this disclosure, step S110, which involves dividing the effective area of ​​the beach in the study area into grids and generating a grid layer, includes: Generate a polygon representing the effective extent of the beach area in the study region; Establish a standard coordinate system based on the geometric features of the beach; Based on the effective area polygon of the beach, an initial mesh surface layer is generated in a standard coordinate system; The area outside the effective range of the beach in the initial grid surface layer is masked to obtain an effective grid layer; A grid layer is generated by assigning attributes to each grid cell in the effective grid layer; the assigned attributes include center point coordinates, spatial boundary coordinates, and average elevation.

[0074] In this disclosed method, based on remote sensing images and elevation data of the study area, hydrological station data can also be combined to extract the area between the outermost high water level line and the lowest water level line of the year as the effective range of the beach. GIS spatial operations such as creating envelopes, converting lines to polygons, or merging buffer zones are used to generate polygons of the effective range of the beach.

[0075] In this disclosed method, a baseline shoreline, typically a stable low water level, can be extracted from remote sensing imagery or surveying data of a suitable time period. Then, using the central axis of the baseline shoreline or the effective polygon of the shoreline as a reference baseline, and the direction parallel to the river's main axis / shoreline central axis as the longitudinal direction, and the direction perpendicular to the longitudinal direction / river's main axis as the transverse direction, a standard coordinate system is constructed. This provides a unified local coordinate reference for subsequent spatial positioning of grid cells, attribute calculation, and process analysis.

[0076] In this disclosure, the original discrete elevation data, such as LiDAR point clouds and measurement points, are used to generate a continuous elevation grid covering the entire effective area of ​​the beach using a scatter interpolation algorithm. This provides a high-quality, continuous base elevation surface for assigning average elevation values ​​to subsequent grid cells.

[0077] In this disclosed method, the number of vertical grid columns and the number of horizontal grid rows are determined based on the spatial resolution and the longitudinal length and transverse width of the shoreline. In GIS software, based on the bounding box of the effective polygonal area of ​​the shoreline and the defined standard coordinate system, a regular rectangular grid surface layer with a specified number of rows and columns covering the bounding box area is generated as the initial grid surface layer.

[0078] In this disclosed method, based on the geometry of the effective beach area polygon or its contained elevation point data, a convex hull boundary identification algorithm is used to calculate the smallest convex polygon boundary that can tightly enclose the effective beach area, thus obtaining the effective beach boundary polygon. Then, the calculated effective beach boundary polygon is used to clip the initial mesh layer. For mesh cells that fall outside the effective beach boundary polygon after clipping, a special mark is made in their attribute table (e.g., assigning NaN (Not a Number) values ​​to all subsequent calculation fields) to mask invalid areas, resulting in an effective mesh layer. This ensures that all subsequent analysis calculations are performed only on the marked effective mesh cells. Finally, for each effective mesh cell, the center point coordinates, spatial boundary coordinates (i.e., the polygon vertex coordinates of the mesh cell), and average elevation (i.e., the average of the elevation values ​​of all cells falling within the geometric range of each effective mesh cell) are calculated and recorded. Ultimately, a standard mesh layer containing the spatial geometric boundaries and core attributes (center point coordinates, average elevation) of the effective mesh cells is generated.

[0079] According to an embodiment of this disclosure, in step S120: The maximum flow velocity is taken from the maximum values ​​of the measurement data points during the dry season, normal water season, and wet season. Specifically, it can be obtained through actual measurement or numerical simulation methods. Five equally spaced data points need to be set up on the cross section of the grid centerline during each of the dry, normal, and wet seasons, and the maximum value of all measurement points in the three seasons is taken.

[0080] The annual inundation frequency is obtained by comparing the water levels of hydrological stations with the grid elevation on a daily basis. Specifically, the daily average water level can be obtained by interpolation or numerical simulation using data from hydrological stations near the riverbank. This average water level is then compared with the grid center elevation to determine whether inundation occurs on a daily basis. The number of inundated days throughout the year is then counted to calculate the annual inundation frequency.

[0081] The maximum inundation depth is taken as the maximum value of the difference between the daily water level and the grid elevation throughout the year.

[0082] The slope of the bank is obtained by extracting the slope of the neighboring area from the elevation raster and then taking the arithmetic mean.

[0083] The erosion rate is calculated using high-resolution remote sensing of shoreline changes. Specifically, the shoreline boundary at the same water level can be extracted using remote sensing images, and high-resolution shoreline polygons at intervals of 1 to 3 years can be compared to calculate the difference in centerline distance and divide by the time interval.

[0084] The median particle size is obtained by laser particle size analysis of surface sediments. Specifically, surface sediment samples of 0–10 cm can be collected and measured using a laboratory laser particle size analyzer.

[0085] The annual deposition rate is calculated by measuring the depth increment at the end of the year using a depositional scale. Specifically, depositional scale plates or marker tubes can be placed at representative grid locations, and the depositional depth increment is measured at the end of the year and divided by time.

[0086] The connectivity index is obtained by statistically analyzing the percentage of days when the daily water level exceeds the grid elevation plus a threshold. Specifically, the number of days when the daily water level exceeds the grid elevation plus a threshold of 0.5 to 1 m can be counted, and then divided by the total number of days in the year.

[0087] According to an embodiment of this disclosure, in step S150, based on the principal component matrix, the optimal number of clusters is determined within the range of 3-7 using a Gaussian mixture model, and the grid layer is divided into multiple functional areas according to the clustering results, including: Based on the principal component matrix, the optimal number of clusters was determined to be 5 within the range of 3-7 using a Gaussian mixture model, and the beach area was divided into five functional zones: scour core zone, active erosion zone, periodic flooding zone, sedimentation buffer zone, and land margin zone.

[0088] In this disclosure, dividing the grid layer into multiple functional areas based on the clustering results includes: Perform clustering with the optimal number of clusters as a parameter, and output the cluster label for each grid cell; Based on the cluster labels, isolated patches with an area of ​​less than 3 grid cells in the grid layer are identified, and then the isolated patches are reclassified according to the cluster label value with the most adjacent clusters. Calculate the weighted average of the eight hydrological disturbance indicators within each cluster, sort the results from largest to smallest, and divide the functional area of ​​each cluster according to the first two indexes of the sort.

[0089] In this disclosed method, the cluster label of each grid cell i is output. .

[0090] label c i Returning the data to the ArcGIS grid layer, the "Region Group" tool is used to identify isolated small patches. For isolated patches with an area less than 3 grid cells, the "Nearest Maximum Reclassification" operation is performed to ensure spatial continuity.

[0091] Calculate the weighted mean of 8 normalized indices for each cluster. .according to Sort by largest to smallest, and take the first two indices. .in, This represents the normalized index value within each cluster.

[0092] Among them, if If the maximum flow velocity and connectivity index are indicated, then this functional area belongs to the core scour zone. like If the indicators of erosion rate and bank slope are used, then the functional area belongs to the active erosion zone. like If the indicators of annual inundation frequency and maximum inundation depth are used, then the functional area belongs to the periodic inundation zone. like If the annual sedimentation rate and median grain size are considered, then this functional area belongs to the sedimentary buffer zone. like If the context is otherwise specified, the functional area belongs to the land edge area.

[0093] According to embodiments of this disclosure, it further includes: The spatial structure of the multiple functional zones is quantified based on the landscape pattern index.

[0094] In this disclosed method, within GIS software, the grid layer dividing functional zones is first converted into functional zone vector polygon features using the RasterToPolygon tool, and a Dissolve operation is performed with "cluster label" as the field to eliminate internal minor fragmentation. Then, the Zonal Statistics as Table tool is used to calculate the area and perimeter of each block for each functional zone polygon using the Calculate Geometry method. Subsequently, based on the functional zone vector polygon features, landscape pattern indices such as the number of patches (NP), average patch area (MPS), and edge density (ED) for each type of functional zone are calculated to quantify the spatial structure of each functional zone.

[0095] According to embodiments of this disclosure, it further includes: At least three representative indicators for each functional area are selected to calculate the weighted average. After determining the weights using the entropy weight method, the health index of the area is calculated.

[0096] In this disclosure, three representative indicators are selected for a specific functional area, such as average flow velocity, vegetation cover, and sediment stability index. Weighted average values ​​within each cluster are calculated based on annual monitoring data; for example, average flow velocity is 2.1, average vegetation cover is 15, and average sediment stability is 0.7. These average values ​​are then normalized by range, resulting in average flow velocity of 0.85, average vegetation cover of 0.40, and average sediment stability of 0.9. The weights are then determined using the entropy weight method; for example, the weight of average flow velocity is 0.5, the weight of vegetation cover is 0.20, and the weight of sediment stability is 0.3. The zonal health index H is then calculated as (0.85 × 0.5) + (0.40 × 0.2) + (0.90 × 0.3) = 0.745. After calculating the zonal health index, the zonal health can be further classified into three levels—"suitable," "generally suitable," and "unsuitable"—based on two thresholds of 0.8 and 0.6. Since 0.6 < 0.745 < 0.8, the health rating of this functional area is "generally suitable".

[0097] According to embodiments of this disclosure, it further includes: The overall health index of the entire beach is calculated based on the area proportion of each functional zone and the health index of the respective zones.

[0098] In this disclosure, after calculating the overall health index of the entire beach, similar to the zonal health rating, a threshold can also be set and a health rating of the entire area can be performed. This disclosure will not elaborate on this.

[0099] The following section uses a typical riverbank in the middle reaches of the Yangtze River as the research area to further illustrate the technical solution disclosed herein.

[0100] For the aforementioned study area, based on remote sensing imagery, hydrological station data, and elevation measurements, the scope of the study beach was first clarified through shoreline extraction and high / low water level boundary analysis. A standardized longitudinal and transverse coordinate system was constructed based on the shoreline's central axis. Several functional process units (FPUs) were then constructed within the beach area, divided along the longitudinal flow direction. The coordinates of their center points and spatial boundaries were recorded in a regular grid format. The grid division results are shown in [reference needed]. Figure 2 . Figure 2 The horizontal axis represents the lateral distance of the beach, and the vertical axis represents the longitudinal distance of the beach.

[0101] This disclosure employs scattered interpolation and boundary clipping algorithms to smooth the elevation data, which is used to assign average elevation values ​​to subsequent grid cells, ensuring that each grid cell obtains effective terrain foundation information. The processed elevation distribution is as follows: Figure 3 As shown in the figure, grayscale represents the elevation changes. The darker the grayscale, the higher the elevation, and the lighter the grayscale, the lower the elevation.

[0102] Next, step S120 is executed to calculate the eight hydrological disturbance indicators for each grid cell. Taking the bank slope and annual inundation frequency as examples, the existing raster data can be used to predict the trend surface of the scattered attributes through the random forest model. Then, ordinary kriging interpolation is performed on the regression residuals. Finally, the regression prediction values ​​are added to the kriging residual field to obtain the continuous raster field data. Figure 4 The diagram shows the slope distribution of the bank in the grid layer. The grayscale value represents the change in slope magnitude. The darker the grayscale, the greater the slope, and the lighter the grayscale, the smaller the slope. Figure 5 The diagram shows the annual flooding frequency distribution in the grid layer. The grayscale value in the diagram represents the flooding frequency. The darker the grayscale value, the higher the flooding frequency, and the lighter the grayscale value, the lower the flooding frequency.

[0103] By executing steps S130 to S150, a habitat zoning distribution cloud map based on hydrological disturbance intensity is finally generated, such as... Figure 6 As shown in the figure, different gray levels represent different habitat functional zones, such as the scour core zone, the active erosion zone, the periodic flooding zone, the sedimentary buffer zone, and the terrestrial margin zone.

[0104] Figure 7 This diagram illustrates a structural block diagram of a system for determining dynamic zoning of riparian vegetation habitats based on hydrological disturbance intensity, according to an embodiment of the present disclosure. This system can be implemented as part or all of an electronic device through software, hardware, or a combination of both.

[0105] like Figure 7 As shown, the riparian vegetation habitat dynamic zoning determination system 700 based on hydrological disturbance intensity includes: Mesh generation module 710 is used to divide the effective area of ​​the beach in the study area into mesh layers; The index acquisition module 720 is used to calculate eight hydrological disturbance indices for each grid cell in the grid layer: maximum flow velocity, annual inundation frequency, maximum inundation depth, bank slope, erosion rate, median grain size, annual deposition rate, and connectivity index. The weight calculation module 730 is used to normalize the calculated hydrological disturbance indexes, calculate the weight of each index using the entropy weight method, and generate a weighted feature matrix. The dimensionality reduction module 740 is used to perform principal component dimensionality reduction on the weighted feature matrix and retain the principal component matrix with a cumulative variance contribution rate ≥ 85%. The partition generation module 750 is used to determine the optimal number of clusters in the range of 3-7 based on the principal component matrix and a Gaussian mixture model, and to divide the grid layer into multiple functional areas according to the clustering results.

[0106] The riparian vegetation habitat dynamic zoning system based on hydrological disturbance intensity provided in this disclosure divides the riparian area into rectangular grid units and extracts eight key hydrological disturbance indicators, such as maximum flow velocity and annual inundation frequency, thus constructing for the first time a multi-dimensional quantitative evaluation system coupling hydrodynamics, geomorphology, and vegetation. This design effectively overcomes the limitations of traditional zoning methods that rely on static elevation zones or administrative boundaries, accurately captures the spatial heterogeneity of dynamic processes such as flow velocity, inundation frequency, and erosion intensity, and scientifically reveals the driving mechanism of hydrological disturbance on vegetation habitat formation.

[0107] For specific technical details of the embodiments disclosed herein, please refer to the above method embodiments, which will not be repeated here.

[0108] According to embodiments of this disclosure, the step of dividing the effective area of ​​the beach in the study region into grids and generating a grid layer includes: Generate a polygon representing the effective extent of the beach area in the study region; Establish a standard coordinate system based on the geometric features of the beach; Based on the effective area polygon of the beach, an initial mesh surface layer is generated in a standard coordinate system; The area outside the effective range of the beach in the initial grid surface layer is masked to obtain an effective grid layer; A grid layer is generated by assigning attributes to each grid cell in the effective grid layer; the assigned attributes include center point coordinates, spatial boundary coordinates, and average elevation.

[0109] According to the embodiments of this disclosure, the maximum flow velocity is taken as the maximum value of the measurement data from the dry season, the normal water season, and the wet season; The annual inundation frequency was obtained by comparing the water levels of hydrological stations with grid elevations on a daily basis. The maximum inundation depth is taken as the maximum value of the difference between the daily water level and the grid elevation throughout the year; The slope of the bank is obtained by extracting the slope of the neighboring area through the elevation raster and then taking the arithmetic mean. The erosion rate was calculated using high-resolution remote sensing of shoreline changes. The median particle size was obtained by laser particle size analysis of surface sediments. The annual deposition rate is calculated using the year-end depth increment of the deposition scale. The connectivity index is obtained by statistically analyzing the percentage of days when the daily water level exceeds the grid elevation plus a threshold.

[0110] According to embodiments of this disclosure, the step of determining the optimal number of clusters within the range of 3-7 using a Gaussian mixture model based on the principal component matrix, and dividing the grid layer into multiple functional areas according to the clustering results, includes: Based on the principal component matrix, the optimal number of clusters was determined to be 5 within the range of 3-7 using a Gaussian mixture model, and the beach area was divided into five functional zones: scour core zone, active erosion zone, periodic flooding zone, sedimentation buffer zone, and land margin zone.

[0111] According to embodiments of this disclosure, dividing the grid layer into multiple functional areas based on clustering results includes: Perform clustering with the optimal number of clusters as a parameter, and output the cluster label for each grid cell; Based on the cluster labels, isolated patches with an area of ​​less than 3 grid cells in the grid layer are identified, and then the isolated patches are reclassified according to the cluster label value with the most adjacent clusters. Calculate the weighted average of the eight hydrological disturbance indicators within each cluster, sort the results from largest to smallest, and divide the functional area of ​​each cluster according to the first two indexes of the sort.

[0112] According to embodiments of this disclosure, it further includes: The spatial structure of the multiple functional zones is quantified based on the landscape pattern index.

[0113] According to embodiments of this disclosure, it further includes: At least three representative indicators for each functional area are selected to calculate the weighted average. After determining the weights using the entropy weight method, the health index of the area is calculated.

[0114] According to embodiments of this disclosure, it further includes: The overall health index of the entire beach is calculated based on the area proportion of each functional zone and the health index of the respective zones.

[0115] In another aspect, this disclosure also provides a computer-readable storage medium storing program code that, when executed by a computer processor, can complete all the steps of the above-described method. This medium is suitable for intelligent analysis platforms such as ecological assessment systems and water conservancy information systems.

[0116] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0117] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0118] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0119] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0120] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0121] Memory may include non-persistent memory in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0122] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0123] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0124] The above description is merely a preferred embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

Claims

1. A method for determining the dynamic zoning of riparian vegetation habitats based on hydrological disturbance intensity, characterized in that, Includes the following steps: To divide the effective area of ​​the beach in the study area into grids and generate a grid layer; Calculate eight hydrological disturbance indicators for each grid cell in the grid layer: maximum flow velocity, annual inundation frequency, maximum inundation depth, bank slope, erosion rate, median grain size, annual deposition rate, and connectivity index. The calculated hydrological disturbance indices are normalized, and the weights of each index are calculated using the entropy weight method to generate a weighted feature matrix. Principal component dimensionality reduction is performed on the weighted feature matrix to retain the principal component matrix with a cumulative variance contribution rate ≥ 85%; Based on the principal component matrix, the optimal number of clusters is determined within the range of 3-7 using a Gaussian mixture model, and the grid layer is divided into multiple functional areas according to the clustering results.

2. The method for determining dynamic zoning of riparian vegetation habitat based on hydrological disturbance intensity according to claim 1, characterized in that, The process of dividing the effective area of ​​the beach in the study region into grid layers and generating a grid layer includes: Generate a polygon representing the effective extent of the beach area in the study region; Establish a standard coordinate system based on the geometric features of the beach; Based on the effective area polygon of the beach, an initial mesh surface layer is generated in a standard coordinate system; The area outside the effective range of the beach in the initial grid surface layer is masked to obtain an effective grid layer; A grid layer is generated by assigning attributes to each grid cell in the effective grid layer; the assigned attributes include center point coordinates, spatial boundary coordinates, and average elevation.

3. The method for determining dynamic zoning of riparian vegetation habitat based on hydrological disturbance intensity according to claim 1, characterized in that: The maximum flow velocity is taken from the maximum value of the measurement data during the dry season, normal water season, and wet season; The annual inundation frequency was obtained by comparing the water levels of hydrological stations with grid elevations on a daily basis. The maximum inundation depth is taken as the maximum value of the difference between the daily water level and the grid elevation throughout the year; The slope of the bank is obtained by extracting the slope of the neighboring area through the elevation raster and then taking the arithmetic mean. The erosion rate was calculated using high-resolution remote sensing of shoreline changes. The median particle size was obtained by laser particle size analysis of surface sediments. The annual deposition rate is calculated using the year-end depth increment of the deposition scale. The connectivity index is obtained by statistically analyzing the percentage of days when the daily water level exceeds the grid elevation plus a threshold.

4. The method for determining dynamic zoning of riparian vegetation habitat based on hydrological disturbance intensity according to claim 1, characterized in that, Based on the principal component matrix, the optimal number of clusters is determined using a Gaussian mixture model within the range of 3-7. Based on the clustering results, the grid layer is divided into multiple functional areas, including: Based on the principal component matrix, the optimal number of clusters was determined to be 5 within the range of 3-7 using a Gaussian mixture model, and the beach area was divided into five functional zones: scour core zone, active erosion zone, periodic flooding zone, sedimentation buffer zone, and land margin zone.

5. The method for determining dynamic zoning of riparian vegetation habitat based on hydrological disturbance intensity according to claim 4, characterized in that, The step of dividing the grid layer into multiple functional areas based on the clustering results includes: Perform clustering with the optimal number of clusters as a parameter, and output the cluster label for each grid cell; Based on the cluster labels, isolated patches with an area of ​​less than 3 grid cells in the grid layer are identified, and then the isolated patches are reclassified according to the cluster label value with the most adjacent clusters. Calculate the weighted average of the eight hydrological disturbance indicators within each cluster, sort the results from largest to smallest, and divide the functional area of ​​each cluster according to the first two indexes of the sort.

6. The method for determining dynamic zoning of riparian vegetation habitat based on hydrological disturbance intensity according to claim 1, characterized in that, Also includes: The spatial structure of the multiple functional zones is quantified based on the landscape pattern index.

7. The method for determining dynamic zoning of riparian vegetation habitat based on hydrological disturbance intensity according to claim 6, characterized in that, Also includes: At least three representative indicators for each functional area are selected to calculate the weighted average. After determining the weights using the entropy weight method, the health index of the area is calculated.

8. The method for determining dynamic zoning of riparian vegetation habitat based on hydrological disturbance intensity according to claim 7, characterized in that, Also includes: The overall health index of the entire beach is calculated based on the area proportion of each functional zone and the health index of the respective zones.

9. A system for determining the dynamic zoning of riparian vegetation habitats based on hydrological disturbance intensity, characterized in that, include: The mesh generation module is used to divide the effective area of ​​the beach in the study area into mesh layers. The index acquisition module is used to calculate eight hydrological disturbance indices for each grid cell in the grid layer: maximum flow velocity, annual inundation frequency, maximum inundation depth, bank slope, erosion rate, median grain size, annual deposition rate, and connectivity index. The weight calculation module is used to normalize the calculated hydrological disturbance indicators, calculate the weight of each indicator using the entropy weight method, and generate a weighted feature matrix. The dimensionality reduction module is used to perform principal component dimensionality reduction on the weighted feature matrix, retaining the principal component matrix with a cumulative variance contribution rate ≥ 85%. The partition generation module is used to determine the optimal number of clusters in the range of 3-7 based on the principal component matrix using a Gaussian mixture model, and to divide the grid layer into multiple functional areas according to the clustering results.

10. A computer-readable storage medium storing computer instructions thereon, characterized in that, When executed by a processor, the computer instructions implement the method described in any one of claims 1-8.

Images