Mudflat vegetation beach fixing and wave dissipating effect simulation system based on multi-scale numerical values

Abstract

The invention relates to the technical field of computer-aided engineering, in particular to a multi-scale numerical value-based beach vegetation beach fixing and wave dissipating effect simulation system, which comprises the following steps of: acquiring terrain basic data, boundary condition data, vegetation physical parameters and vegetation ecological parameters; establishing a numerical calculation grid and a terrain by using terrain basic data, and updating a landform elevation and a hydrodynamic condition based on a wave flow sediment coupling mechanism by using boundary condition data as a driving condition; according to a vegetation mechanical model, vegetation flexible deformation is calculated in combination with hydrodynamic conditions and vegetation physical parameters, and hydraulic drag force parameters are updated based on the vegetation flexible deformation; according to the habitat suitability model, evaluating the probabilistic trend of vegetation growth, death and expansion; physical process simulation and landform evolution calculation are combined, and adaptive verification and resetting steps are introduced, so that accelerated simulation of beach fixation and wave dissipation of beach vegetation is realized. According to the invention, through multi-scale simulation, accurate prediction of beach fixing and wave dissipation is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of computer-aided engineering, and in particular to a multi-scale numerical value-based simulation system for beach vegetation consolidation and wave elimination effects. Background Art

[0002] Tidal flat vegetation is a crucial ecological barrier, playing a key role in stabilizing coastlines and dissipating waves. To scientifically evaluate its protective effectiveness, numerical simulation techniques are often used to recreate physical processes such as hydrodynamics, sediment transport, and landform evolution.

[0003] Existing procedural numerical simulations, typically based on two- or three-dimensional digital models, can effectively simulate short-term physical processes such as wave propagation, water flow, and the sediment transport driven by these processes, providing an important technical reference for engineering design. However, the long-term evolution of tidal flat landforms is essentially the result of the coupling of rapid hydrodynamic processes with slower biological and geomorphological processes at multiple spatiotemporal scales. Current simulation methods are generally limited in their effectiveness in improving the accuracy of long-term predictions. The reliability of their predictions is insufficient to fully meet the needs of refined ecological assessments and engineering design, resulting in a failure to accurately capture the evolutionary patterns of ecological and geomorphological systems.

[0004] Therefore, a multi-scale numerical simulation system for the beach consolidation and wave-breaking effect of tidal flat vegetation is proposed. Summary of the Invention

[0005] The purpose of the present invention is to provide a multi-scale numerical simulation system for the effect of tidal flat vegetation on beach consolidation and wave elimination, which can achieve accurate prediction of beach consolidation and wave elimination through multi-scale simulation. The system obtains basic terrain data, boundary condition data, vegetation physical parameters, and vegetation ecological parameters; uses the basic terrain data to establish a numerical calculation grid and terrain; uses the boundary condition data as driving conditions, and updates the landform elevation and hydrodynamic conditions based on the wave-flow-sediment coupling mechanism; calculates the flexible deformation of vegetation based on the vegetation mechanics model, combining the hydrodynamic conditions and vegetation physical parameters, and updates the hydraulic drag force parameters based on the flexible deformation of vegetation; evaluates the probabilistic trends of vegetation growth, decline, and expansion based on the habitat suitability model; combines physical process simulation with landform evolution calculation, and introduces adaptive verification and reset steps to achieve accelerated simulation of tidal flat vegetation in beach consolidation and wave elimination.

[0006] To achieve the above object, the present invention provides the following technical solutions: A multi-scale numerical simulation system for the beach vegetation consolidation and wave dissipation effect, including: Data acquisition module: obtains terrain basic data, boundary condition data, vegetation physical parameters and vegetation ecological parameters; Simulation construction module: Use the basic terrain data to establish the numerical calculation grid and terrain, use the boundary condition data as the driving condition, and update the landform elevation and hydrodynamic conditions based on the wave-current-sediment coupling mechanism; Vegetation response module: Calculates vegetation flexible deformation based on vegetation mechanical model, combined with hydrodynamic conditions and vegetation physical parameters, and updates hydraulic drag force parameters based on the vegetation flexible deformation through a predetermined functional relationship; Landform Evolution Module: Based on the habitat suitability model, the module evaluates the probabilistic trends of vegetation growth, decline, and expansion based on landform elevation, hydrodynamic conditions, and vegetation ecological parameters, and feeds the updated vegetation status back to the Vegetation Response Module. Accelerated simulation module: combines physical process simulation with landform evolution calculation, introduces adaptive verification and reset steps, and realizes accelerated simulation of tidal flat vegetation in beach consolidation and wave dissipation.

[0007] Preferably, when the simulation construction module establishes the initial numerical calculation grid and terrain, the high-precision water depth data in the terrain basic data is used to establish a digital elevation model. For complex coastlines containing tidal gullies and mudflats, an unstructured triangular grid covering the entire study area is generated, and local grid encryption processing is performed on tidal gullies, vegetation-covered areas and areas of focus in the engineering area. The elevation data contained in the digital elevation model is interpolated onto the grid nodes to form the initial terrain required for model calculation.

[0008] Preferably, in the simulation construction module, the calculation process executed by the wave-flow-sediment coupling mechanism within a coupling time step is: the hydrodynamic solver calculates the average water flow field at the current moment and transmits it to the wave solver, and the wave solver calculates the wave propagation, deformation and wave-induced radiation stress on this basis, and then feeds the wave-induced radiation stress back to the hydrodynamic solver to be superimposed with the bed shear stress of the water flow to obtain the bottom bed shear stress, and uses the bottom bed shear stress to drive the sediment transport formula to calculate and update the landform elevation change of the current time step.

[0009] Preferably, the simulation construction module is further configured to call the data acquisition module to obtain physical property data including vegetation elastic modulus, and pass the elastic modulus as a core parameter to the cantilever beam mechanical model in the vegetation response module for calculating flexible bending deformation.

[0010] Preferably, the Morrison equation is used to calculate the hydrodynamic load acting on the vegetation, the cantilever beam theory is applied to mechanically model the vegetation, the load is applied to the model to obtain the vegetation bending displacement, and the vegetation bending displacement is used as the dynamic deformation parameter; the bending displacement and the geometric and mechanical properties of the vegetation are input into an empirical function model that characterizes the relationship between the dynamic bending of vegetation and the water flow drag force to update the hydraulic drag force parameter; the empirical function model is established by fitting and inverse calculation based on the dynamic deformation of vegetation and the total drag force data measured in the controlled physical model experiment.

[0011] Preferably, the landform evolution module evaluates the evolution trend of vegetation through a habitat suitability model. The habitat suitability model takes multiple environmental factors of the vegetation location, including average beach elevation, average scouring rate and average flooding frequency, as comprehensive inputs, and calculates a comprehensive habitat suitability index based on the comprehensive inputs; according to the value of the habitat suitability index, combined with the probability model, the growth, decline and spatial expansion trends of the vegetation in the next landform evolution time step are determined.

[0012] Preferably, in the accelerated simulation module, the execution process of the adaptive time step control is as follows: when each calculation time step is completed, the Courant-Friedrich-Law dimension used to judge the numerical stability is calculated in the entire calculation domain; if the Courant-Friedrich-Law dimension is greater than a preset safety threshold, the calculation result of the current time step is rejected, and the step size of the next time step is shortened by a preset ratio and recalculated.

[0013] Preferably, the method of combining physical process simulation with landform evolution calculation includes: performing high-resolution physical process simulation of a representative period to calculate the net change trend field of the landform; in a cyclic calculation step, applying the net change trend field to update the terrain within a macroscopic landform evolution time step; after the time step ends, verifying the preset system key indicators, and when the verification results of the key indicators meet the preset stability conditions, continuing to use the current net change trend field to repeat the cyclic calculation steps, and when the verification results do not meet the stability conditions, re-performing the high-resolution physical process simulation to update the net change trend field.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention calculates the flexible deformation of vegetation in real time and dynamically updates the hydraulic drag force parameters based on the deformation, so that the model can more realistically describe the resistance change process of vegetation, thereby improving the simulation accuracy of physical effects such as wave dissipation and siltation promotion.

[0015] 2. This study uses an assessment method based on a habitat suitability index and a probabilistic model. This method comprehensively considers the coupled impacts of multiple environmental factors and provides a probabilistic forecast of vegetation evolution trends. This makes the simulation of ecosystem responses over long-term evolution more closely aligned with natural laws, and the prediction results more valuable.

[0016] 3. The "simulate-verify-reset" mechanism proposed in this invention ensures that the evolution trend of the system state can be updated in a timely manner when the system state changes significantly by adaptively verifying the system state, thereby greatly improving computing efficiency while ensuring the stability and reliability of long-term simulation results. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 A schematic diagram of the structure of a system for simulating the effects of beach vegetation consolidation and wave dissipation based on multi-scale numerical values ​​provided by an embodiment of the present invention; Figure 2 A schematic diagram of a simulation building module flow chart provided by an embodiment of the present invention; Figure 3 A schematic diagram of the landform evolution module flow provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0019] Example 1: See also Figures 1 to 3 The present invention provides a multi-scale numerical simulation system for the beach vegetation consolidation and wave-breaking effect. The technical solution is as follows: A multi-scale numerical simulation system for the beach vegetation consolidation and wave dissipation effect, including: Data acquisition module: obtains basic terrain data, boundary condition data, vegetation physical parameters and vegetation ecological parameters.

[0020] Topographic data was used to construct the initial numerical computational grid and digital elevation model (DEM). This data includes the elevation, topographic features, and sediment information of the mudflat area. Data types include mudflat surface elevation, the location of tidal gullies and vegetation-covered areas, and sediment type (e.g., sandy or silty). Topographic data was obtained through a combination of field measurements and publicly available data. Field measurements used conventional distance measuring tools and poles to measure mudflat elevations at a grid spacing of approximately 5 m x 5 m, covering the entire study area. Tidal gully locations and depths were recorded manually, with a spacing of approximately 2 m between measurement points. Sediment samples were collected manually to record sediment types. For publicly available data, shoreline and vegetation distribution were extracted using freely available satellite imagery to supplement the field measurements. The collected elevation data was manually collated and converted to a unified planar coordinate system. Open-source GIS software was used to interpolate the elevation points to generate a DEM. Sediment data were classified to generate a distribution map. Outputs include a NetCDF DEM (including elevation and topographic features) and a Shapefile-based DEM (including sediment type).

[0021] Boundary condition data provide physical driving conditions for the numerical model, including hydrodynamic boundary conditions (tides, waves, wind fields) and sediment boundary conditions (suspended sediment concentration). Tidal data include tide level and flow velocity, wave data include wave height and wave period, wind field data include wind speed, and sediment data include offshore sediment concentration. Boundary condition data are obtained through a combination of field observations, public data, and auxiliary experiments to ensure that the data covers the dynamic characteristics of the study area.

[0022] Tidal data: Historical tidal data covering at least two complete tidal cycles were obtained from the National Oceanographic Data Center or local oceanographic stations. Field observations were conducted at 5–7 observation points at the boundary of the study area to measure tide levels and tidal current speeds, with records recorded six times daily for 25 days.

[0023] Wave data: Historical wave data (wave height, wave period, and wave direction) were obtained from oceanographic stations for 30 days, and field observations were conducted four times a day to record wave height, wave period, and wave direction for 20 days.

[0024] Wind data: Wind speed and direction data were collected from a nearby weather station every two hours for 30 days. On-site observations were conducted using weather stations deployed at 5-7 points along the border, continuously recording wind speed and direction every hour for 25 days.

[0025] Sediment boundary conditions: Seven sampling points were set up at the boundary of the study area, and water samples were collected four times a week (for 25 days). The suspended sediment concentration was measured and calibrated by filtration. The particle size distribution was recorded, and the sediment settling rate was determined by laboratory sedimentation experiments.

[0026] Tidal and wave data were interpolated using spline interpolation to generate continuous time series with a temporal resolution of 30 minutes. Wind data were averaged over one hour and smoothed to ensure temporal continuity. Sediment concentration and sedimentation rate data were organized into spatially distributed datasets and assigned to model boundary nodes using Kriging interpolation. The output format is: Hydrodynamic boundary conditions: stored in CSV format, including time, tide level, tidal velocity, wave height, wave period, wave direction, wind speed and wind direction.

[0027] Sediment boundary conditions: Stored in NetCDF format, including the spatial distribution of suspended sediment concentration and settling rate.

[0028] Model validation data is used to calibrate and verify the accuracy of the numerical model, including hydrodynamic validation data (tidal level, flow velocity, wave parameters), sediment transport validation data (sedimentation rate, landform elevation change) and vegetation status data (distribution, density, height). Hydrodynamic validation: 3-5 validation points are selected in the study area, high and low tide levels are recorded daily, flow velocity and wave parameters are measured, and the data are cross-validated with public tide tables and wave records to ensure consistency.

[0029] Verification of sediment transport: Three sediment collection points were set up in the vegetated area and three in the non-vegetated area. The sedimentation was measured once a week. The elevation changes were measured by poles and the changes in beach elevation were recorded.

[0030] Vegetation status verification: Vegetation data was obtained through ground surveys and image analysis. The ground survey covered 30 sampling points, recording vegetation type, coverage, and average height. Hydrodynamic data were organized into time series, and mean and standard deviations were calculated for comparison with model output. Sediment transport data were used to calculate sedimentation rates and elevation changes, generating spatial distribution maps. Vegetation status data were organized into coverage and height distribution maps and stored in Shapefile format.

[0031] Simulation construction module: using the terrain basic data to establish the initial numerical calculation grid and terrain; using the boundary condition data as the driving condition for the physical process calculation; in a unified numerical model, based on the wave-current-sediment coupling mechanism, calculating the sediment transport caused by the combined action of waves and water flow, and updating the landform elevation according to the spatial gradient of the sediment transport flux; using the model verification data to calibrate and verify the calculation results.

[0032] First, the digital elevation model and sediment distribution data generated by the data acquisition module are imported. The digital elevation model, stored in NetCDF format, contains the elevation and topographic features of the tidal flat area. The sediment distribution data, stored in Shapefile format, records sediment type information. The system parses the NetCDF file, extracts the elevation data, and checks its integrity to ensure coordinate system consistency. The sediment data is loaded using the spatial query tool to verify the data format and extent. Based on the spatial extent of the elevation data, the system automatically demarcates the computational domain boundaries and generates a polygonal boundary file containing shorelines and open boundaries. This boundary file is stored in Shapefile format for subsequent mesh generation.

[0033] For complex coastlines containing tidal gullies and mudflats, an open source mesh generation tool is used to generate an unstructured triangular mesh covering the entire computational domain. The boundary file is loaded, the initial mesh parameters are set, and a uniform triangular mesh is generated. The tidal gully location in the elevation data is identified, and the tidal gully centerline is extracted using spatial analysis tools. Based on the centerline and the tidal gully location recorded by manual measurement, the tidal gully area is locally meshed and the mesh side length is reduced to improve the resolution. Spatial queries are performed on vegetation-covered areas and engineering areas (such as breakwaters) to determine their boundary ranges. Local mesh encryption is also applied. The encryption process is achieved by setting the mesh size control field to ensure a smooth transition between the encrypted and non-encrypted areas. The mesh quality is checked, the triangle internal angles and skewness are optimized, and the node positions are adjusted to eliminate deformed units. The generated mesh is stored as a compressed file containing node coordinates, unit connection information, and encrypted area markers.

[0034] High-precision bathymetric topography data was interpolated onto the nodes of an unstructured triangular mesh to form an initial digital elevation model. The digital elevation model was loaded in NetCDF format, and elevation data was extracted. Kriging interpolation was used to assign elevation values ​​to the grid nodes. Bilinear interpolation was additionally applied to tidal creek areas to ensure smooth depth variations. Sediment distribution data was loaded in Shapefile format, and a spatial query was used to assign a corresponding sediment type to each grid cell. Sediment-related physical parameters, including sediment particle size, density, and critical starting stress, were extracted from a preset parameter library and associated with the grid cells. The initial digital elevation model was checked for continuity and accuracy, and outliers were corrected. The resulting initial digital elevation model was stored in NetCDF format, containing node coordinates, elevation values, and sediment parameters, ensuring a perfect match with the unstructured triangular mesh.

[0035] By generating an unstructured triangulated mesh and locally refining the mesh in tidal gullies, vegetated areas, and project areas, the model accurately adapts to the topographical characteristics of complex coastlines, improving simulation resolution in key areas and more accurately capturing the spatial variations in water flow and sediment transport. The initial digital elevation model is formed through high-precision interpolation, ensuring the long-term stability of topographic details. This provides a reliable spatial foundation for predicting the evolution of ecological and geomorphic systems and a foundation for improving the accuracy of long-term predictions.

[0036] Load the boundary condition data provided by the data acquisition module, including hydrodynamic boundary conditions and sediment boundary conditions. The hydrodynamic boundary conditions are stored in CSV format and contain time series data on tide level, tidal velocity, wave height, wave period, wave direction, wind speed, and wind direction. Parse the CSV file, extract the data line by line, generate a continuous time series using spline interpolation, and assign it to the nodes at the open boundary of the model to ensure that the time step is synchronized with the model calculation. The sediment boundary conditions are stored in NetCDF format and contain spatial distribution data on suspended sediment concentration and sedimentation rate. Load the NetCDF file, assign the concentration and sedimentation rate to the boundary nodes using the inverse distance weighted interpolation method, set the shoreline boundary as a no-slip boundary, use the Neumann condition to handle wave reflections, convert the wind field data into surface stress, and calculate it using the Charnock formula. The boundary condition dataset is stored in NetCDF format and contains time series and spatial distribution to drive subsequent physical process calculations.

[0037] In a unified numerical model, a two-dimensional computational framework is constructed based on the finite volume method, using an unstructured triangular mesh. An adaptive time step is set to ensure numerical stability. Boundary condition data is loaded and the hydrodynamic module is started. The two-dimensional shallow water equations are solved to calculate the water level and velocity distribution of the entire computational domain. A Godunov-type format is used to handle convection terms, and the PISO algorithm is used to solve the pressure-velocity coupling. Input data includes tide level, tidal velocity, and wind field stress. The calculation process considers the influence of the bottom boundary of the initial digital elevation model. The calculation results are stored as temporary variables, including water depth and velocity fields, for use in subsequent wave modules. The output water flow field data is stored in NetCDF format, recording the dynamic changes of each grid node.

[0038] The water level and velocity fields calculated by the hydrodynamic module are transferred to the wave module. To account for the significant timescale differences between the wave and flow fields, the coupling mechanism of this embodiment adopts a step-by-step approach. Specifically, the computational time step used by the wave solver is set to less than one-tenth of the hydrodynamic solver's time step. This ensures that the rapid wave oscillations are fully resolved within each hydrodynamic computation step, providing the hydrodynamic module with a stable and accurate, time-averaged wave-induced radiation stress field, ensuring the stability and physical realism of the entire coupled system during long-term simulations. Based on the wave equilibrium equations, wave propagation, refraction, breakup, and energy dissipation are calculated. Input data includes boundary wave height, wave period, and wave direction. A spectral method is used to solve the wave field and calculate the wave-induced radiation stress. The wave height and wave period distributions are updated, taking into account the influence of the water depth and flow velocity of the initial digital elevation model on wave propagation. The wave computation results are fed back to the hydrodynamic module to adjust the flow field. The output wave field data, including wave height, wave period, and radiation stress, is stored in NetCDF format for subsequent bed shear force calculations.

[0039] The total bed shear force is calculated by integrating the velocity field from the hydrodynamic module and the wave-induced radiation stress from the wave module. The bed shear stress due to flow is calculated using the Manning formula, based on the bottom roughness parameters of the grid cells in the initial digital elevation model. The bottom orbital velocity from the wave module is extracted to calculate the wave-induced shear stress. A nonlinear superposition method is used to combine the flow and wave shear stresses to generate the total bed shear force. The shear force is corrected using the Soulsby formula based on wave-current interaction to account for nonlinear effects. This ensures that the superposition process reflects the dynamic characteristics of wave-current coupling. The output total bed shear force data is stored in NetCDF format, containing the shear force value for each grid cell, for use in sediment transport calculations. To calculate the sediment transport rate driven by total bed shear stress, a composite sediment transport calculation module is employed. This module integrates transport formulas tailored to different bed types and performs targeted calculations for each grid cell within the computational domain based on the bed distribution map obtained from the data acquisition module. For non-cohesive sediment regions on sandy beds, the system uses the van Rijn transport formula; for cohesive sediment regions on silty beds, a transport formula based on the Partheniades-Krone theory is used to simulate erosion and sedimentation, respectively. This computational strategy, based on bed zoning and the use of different physical mechanism formulas, ensures the physical realism and accuracy of the landform evolution simulation. To ensure the numerical stability and physical immediacy of this coupled calculation, the wave-current-sediment coupling mechanism also includes an intra-step iterative feedback loop. After calculating the sediment transport flux for the current time step, the system does not immediately update the landform; instead, it first generates a temporary landform change based on this flux. This temporary landform will react to the hydrodynamic calculations, forming an iterative cycle until the hydrodynamic and landform changes converge within a single time step. This iterative mechanism captures the immediate negative feedback of landform changes on water and sediment dynamics, avoiding overestimation of erosion and deposition rates under strong dynamic conditions, and significantly enhancing the model's stability and physical realism when simulating extreme events.

[0040] By accurately calculating the wave-current-sediment coupling mechanism based on the Soulsby formula, we can realistically simulate the changes in bottom shear stress under the combined effects of waves and currents, improving the reliability of sediment transport calculations. This approach, by accounting for nonlinear wave-current interactions, improves the accuracy of long-term geomorphological evolution predictions, meeting the needs of refined ecological assessments and engineering design.

[0041] Load model validation data, including hydrodynamic validation data, sediment transport validation data, and vegetation status data, and run the model, covering the observation period of the validation data. Extract the time series of tide level, flow velocity, and wave height output by the model at internal validation points and compare them with the hydrodynamic validation data. Adjust the bed roughness coefficient to optimize the hydrodynamic simulation results. Compare the model's sedimentation rate and elevation changes with the sediment transport validation data, and adjust the sedimentation rate and starting flow velocity parameters in the sediment transport formula to ensure consistent scouring and deposition trends.

[0042] Through the multi-source data fusion verification method, the model performance was comprehensively evaluated using hydrodynamics, sediment transport and vegetation status data. Through weighted average error analysis, the optimization effect of model parameters was enhanced, ensuring that the simulation results were more stable and reliable in long-term predictions, meeting the requirements of refined ecological assessment and engineering design.

[0043] An independent validation dataset was used to compare the model outputs of tide level, current velocity, wave height, sedimentation rate, and elevation change with the measured data. Correlation coefficients were calculated to evaluate the model's predictive ability. The distribution of sediment deposition in vegetated areas was examined to verify whether the simulation results reflected the retardation of sediment transport by vegetation. Statistical analysis was performed to generate time series comparison charts and spatial distribution comparison charts. The validation results, error indicators, and correlation coefficients were stored in CSV format.

[0044] Vegetation response module: receives verified real-time hydrodynamic conditions and calculates the flexible bending deformation of the vegetation based on a preset vegetation mechanical model, and updates the hydraulic drag force parameters that characterize the comprehensive resistance effect of the vegetation on the water flow based on the deformed posture.

[0045] The Vegetation Response Module is launched, loading the verified real-time hydrodynamic condition data output from the Simulation Construction Module. The hydrodynamic condition data, stored in NetCDF format, contains time series of water level, velocity, wave height, wave period, and wave-induced radiation stress. The system parses the hydrodynamic data file, extracting velocity and wave field data for each time step and checking the data for integrity and temporal continuity. Using spatial query tools, the hydrodynamic condition data is assigned to nodes of an unstructured triangular mesh, ensuring that they match the spatial locations of the vegetation distribution area. The velocity and wave data are temporally interpolated using spline interpolation to generate a time series dataset synchronized with the vegetation mechanics calculations and stored as temporary variables.

[0046] Load the vegetation status data provided by the data acquisition module, including vegetation distribution, density, initial height and type information. The vegetation status data is stored in Shapefile format. Parse the vegetation status data file to extract the boundary, type (such as reed, salt marsh grass), coverage rate and initial height data of the vegetation coverage area. Through spatial query, assign the vegetation data to the cells of the unstructured triangular mesh and generate a vegetation attribute table for each grid cell. Load the vegetation mechanical model parameters and extract the biomechanical properties, including elastic modulus, section inertia moment, stem diameter, bending stiffness and leaf area, from the parameter library. The parameter library is stored in JSON format and contains parameters of common vegetation types, such as the typical mechanical properties of reed and salt marsh grass. Check the integrity of vegetation data and parameters, correct missing or outliers, and generate a vegetation initial state dataset containing vegetation attributes and mechanical parameters of grid cells. The dataset is stored in NetCDF format.

[0047] The vegetation mechanics model is activated using beam theory, treating the plant stem as a flexible cantilever beam subjected to current and wave action, fixed at the bottom and free at the top. Input data includes the flow velocity field from the hydrodynamic module, the bottom orbital velocity from the wave field, and plant mechanical parameters (such as elastic modulus, section moment of inertia, and leaf area). For each time step, the Morison equation is used to calculate the hydrodynamic loads acting on the plant stem and leaves, decomposing them into drag and inertial forces. The drag force is based on the flow velocity and the projected plant area (combined stem and leaf areas), while the inertial force is based on the wave acceleration and plant mass. The beam theory model is solved using the finite difference method to calculate the bending deformation of the stem along the height direction, accounting for the added drag effect of the leaves. The deformation curve for each grid cell is output and stored in NetCDF format, recording the stem top displacement, bending angle, and leaf pitch angle.

[0048] Based on the spatial distribution of vegetation, the spatial density coefficient between vegetation was calculated. Dense vegetation areas were iteratively merged using the DBSCAN clustering algorithm to generate synthetic vegetation areas. The neighborhood radius of the DBSCAN algorithm ranged from 5 to 15 times the average interplant spacing, and the minimum neighborhood data point ranged from 10 to 30. The neighborhood radius was determined by analyzing the k-distance graph and selecting the inflection point of the curve. The k value was set based on the typical vegetation patch size in the study area. A typical k value for a reed community was 20. The spatial density coefficient was determined by analyzing the spatial distance matrix of vegetation cover between grid cells. The distance threshold was dynamically adjusted based on vegetation type and water depth. The DBSCAN algorithm merged dense vegetation cells into bounding boxes, which were defined as synthetic vegetation areas. These were stored in NetCDF format and contained the bounding coordinates (minimum x, y and maximum x, y) and the average vegetation density within the area.

[0049] After determining the synthetic vegetation areas, the system dynamically adjusts and updates the comprehensive hydraulic drag parameter, or drag coefficient, Cd, within each area. This drag coefficient, Cd, is determined as a function of multiple vegetation morphological and hydrodynamic parameters. Its inputs include geometric and mechanical properties such as spatial density, stem diameter, and elastic modulus, as well as dynamic deformation parameters of stem bending displacement calculated in real time by the model. To establish a functional model characterizing the relationship between the hydraulic drag coefficient and the dynamic response of vegetation, basic data were first obtained through controlled flume physical model experiments. Water flow conditions and vegetation physical parameters (such as stiffness and density) were systematically varied. High-speed video and force sensors were used to synchronously measure the dynamic bending deformation of the vegetation and the total flow drag. Second, a functional model was constructed based on the experimental data. Specifically, for each operating condition, the standard drag definition in fluid dynamics was applied. The hydraulic drag coefficient describing the deformation state was reversely solved based on the measured total drag, current flow velocity, water density, and the vegetation's headwater area. Finally, the relative curvature of multiple groups of vegetation is associated with the hydraulic drag coefficient calculated by the above method. Through curve fitting, an empirical function model is established that can reflect the change of this coefficient with the bending deformation of vegetation. Subsequently, based on the updated drag coefficient Cd and the real-time water velocity, the system follows the standard drag equation to calculate the hydraulic impact index. This index is clearly defined as the spatially unevenly distributed momentum sink term (i.e., resistance term) generated by vegetation on the water flow, which is used to accurately quantify the dissipation effect of vegetation on water flow energy.

[0050] To specifically define the empirical function model, in this embodiment, the function is constructed as a second-order polynomial function that relates the relative curvature of vegetation to the hydraulic drag force. The specific form of the function is as follows: ; represents the updated hydraulic drag coefficient after flexible deformation correction, Represents the initial hydraulic drag coefficient of the vegetation when no bending deformation occurs, Represents the bending displacement of the vegetation tip calculated in real time by the vegetation mechanics model, represents the natural height of vegetation, and represents the dimensionless empirical coefficient obtained by fitting and back-calculating the data obtained from the controlled physical model experiment. When the simulation object is a typical mudflat reed, based on the experimental data, the empirical coefficient can be taken as =-0.5, =0.15.

[0051] The prediction accuracy of the relationship of the multivariate polynomial function is closely related to the range of experimental calibration data used when establishing the relationship. In this embodiment, the preferred applicable range of the relationship of the function is limited to a specific interval of experimental calibration conditions (such as flow rate and water depth). For example, when the hydrodynamic conditions encountered in the simulation exceed 30% of the calibration range, the system will use a linear interpolation or extrapolation strategy based on the boundary value to estimate the drag force parameters and output a warning message at the same time. In addition, the system will monitor the duration of the period exceeding the applicable range. If conditions exceeding the applicable range occur in multiple consecutive macro-geomorphological evolution time steps, the reset mechanism in the accelerated simulation module will be forcibly triggered to return to executing high-resolution physical process simulation to ensure that the model regains a reliable parameterization scheme under the new dynamic environment.

[0052] By calculating the spatial density coefficient between vegetation and using dynamic clustering to generate synthetic vegetation areas, it is possible to accurately reflect the actual dense distribution of vegetation in space and generate regional boundaries closely related to hydrodynamics and sediment transport. By calculating the hydraulic influence index, the drag force parameters are dynamically adjusted to accurately quantify the resistance effect of vegetation on water flow, truly reflecting the adaptive behavior of vegetation in complex hydrodynamic environments, enhancing the long-term stability of hydrodynamic and sediment transport simulations, and providing reliable support for capturing the evolution laws of ecological geomorphological systems and refined ecological assessments.

[0053] The flexible bending deformation of vegetation, the boundaries of the synthetic vegetation area, the updated hydraulic drag coefficient, and the calculated hydraulic influence index are integrated to generate a vegetation dynamic response dataset. This dataset is stored in NetCDF format and contains key information such as the time series deformation curve, area boundaries, drag coefficient, and hydraulic influence index. The system will check the consistency of the data to ensure that it fully matches the time step of the hydrodynamic conditions and can generate corresponding visualization files, such as plotting the spatial distribution map and time series map of the vegetation bending state, synthetic vegetation area, and hydraulic influence index. The hydraulic influence index in this vegetation dynamic response dataset will be directly fed back to the fluid dynamics equation of the simulation construction module for calculation. At the same time, the entire dataset is also delivered to the landform evolution module for analysis of the long-term impact of vegetation on landform evolution.

[0054] Load the vegetation status verification data provided by the data acquisition module and store it in Shapefile format, including measured vegetation height, coverage, leaf status, and hydraulic resistance data. Run the vegetation response module to cover the time period of the verification data. Extract the synthetic vegetation area, hydraulic drag force parameters, and hydraulic influence index output by the model and compare them with the measured data. Calculate the deviation between the simulated and measured values, adjust the parameters of the vegetation mechanics model, including the elastic modulus, drag force coefficient, and blade resistance coefficient, and iterate and optimize until the deviation is minimized. Verify that the synthetic vegetation area accurately reflects the dense distribution. Check the impact of the hydraulic influence index on hydrodynamics and sediment transport. Confirm that the velocity field and sedimentation rate output by the simulation construction module are consistent with the measured values. Verification results are stored in CSV format, including error indicators and correlation coefficients. The adjusted parameters are stored in JSON format. Update the parameter library to ensure model reliability.

[0055] Landform evolution module: Based on the habitat suitability model, the probabilistic trends of vegetation growth, decline and expansion are evaluated based on landform elevation, hydrodynamic conditions and vegetation ecological parameters, and the updated vegetation status is fed back to the vegetation response module.

[0056] After the geomorphic evolution module is started, the geomorphic elevation and hydrodynamic condition data output by the simulation construction module are loaded and stored in NetCDF format. The geomorphic elevation data contains the elevation values ​​of the unstructured triangular mesh nodes, reflecting the changes in scouring and deposition caused by sediment transport. The hydrodynamic condition data includes the time series of water level, flow velocity, wave height, wave period, and wave-induced radiation stress. The system parses the NetCDF file, extracts the geomorphic elevation and hydrodynamic field data for each time step, checks the data integrity and temporal continuity, and uses spatial query tools to assign the data to the unstructured triangular mesh nodes to ensure that they match the spatial location of the vegetation distribution area. A synchronized time series dataset is generated and stored as a temporary variable for subsequent habitat suitability assessment.

[0057] Load the initial state data of vegetation provided by the data acquisition module and store it in Shapefile format, including vegetation distribution boundaries, coverage, initial height, stem density and type information. Parse the Shapefile file, extract vegetation attributes, assign them to unstructured triangular grid cells through spatial query, and generate a vegetation attribute table. Load the preset habitat suitability model parameters and store them in JSON format. The parameters mainly contain the weights and coefficients corresponding to each environmental factor used in the logistic regression model. Check the integrity of vegetation data and habitat parameters and correct outliers. Generate a dataset of vegetation initial state and habitat suitability parameters, including the vegetation attributes of the grid cells and the model weights and coefficients, and store these data in NetCDF format.

[0058] The habitat suitability model based on multiple environmental factors is used to evaluate the evolution trend of vegetation. The output probability of the logistic regression model is an S-shaped function value based on a linear combination of environmental factors. The linear combination includes a basic intercept term and weight coefficients corresponding to the beach elevation, scouring rate and flooding frequency respectively. In a typical embodiment for reeds, by fitting the historical data of the region, the basic intercept term obtained is 0.5, the weight coefficient of the beach elevation is 2.5, the weight coefficient of the scouring rate is -1.5, and the weight coefficient of the flooding frequency is -0.8.

[0059] Based on the value of the habitat suitability index, the system combines the probability model of the random forest algorithm to predict the growth, decline and spatial expansion trends of vegetation in the next landform evolution time step. The random forest model contains 100 decision trees with a maximum depth of 10. The main features used for division are the suitability index of the current unit and the average vegetation coverage of 8 neighboring units. The system then selects the trend with the highest probability as the final evolution direction of the unit and updates its vegetation status accordingly. Specifically, if the predicted trend is growth, the vegetation coverage in the unit will increase by a fixed growth rate percentage extracted from the parameter library and preset for the species; accordingly, if the predicted trend is decline, its coverage will decrease by a fixed decline rate percentage extracted from the parameter library and preset for the species; for expansion trends, the system will establish new vegetation in adjacent blank cells that meet the habitat suitability conditions with a preset initial seed coverage extracted from the parameter library. The fixed growth rate, fixed decline rate and initial seed coverage are all physical parameters that can be calibrated through routine experiments in this field. To further enhance the realism of the simulation, for growth trends, vegetation cover and stem density were increased based on preset growth rates extracted from a parameter library. For decline trends, cover was reduced or vegetation was removed, and height was adjusted to zero. For expansion trends, a spatial diffusion model was used to expand vegetation distribution to adjacent grid cells, setting the initial cover and height of each new grid cell. Through specific physical parameters and spatial models, the dynamic processes of vegetation growth, decline, and expansion were meticulously depicted. This makes the ecological succession simulation more closely aligned with natural laws, improving the realism and predictive accuracy of the entire system.

[0060] The updated spatial distribution and physiological parameters (cover, height, and stem density) of vegetation are fed back to the vegetation response module to recalculate its hydraulic impact index, achieving bidirectional feedback of biophysical processes. To ensure the reliability of the probabilistic model and avoid non-physical predictions near ecological tipping points, the historical observational or experimental dataset used to train the random forest model should preferably include a sufficient number of samples covering the critical range of the habitat suitability index. For example, samples with a suitability index between 0.3 and 0.7 are recommended to comprise no less than 15% of the total sample size to ensure the model has sufficient ability to learn and identify critical behaviors of vegetation expansion and decline. To enhance the long-term adaptability of ecological evolution predictions, the random forest model is configured to automatically recalibrate after a specified simulation duration. This recalibration process uses landform and vegetation state data generated during the previous simulation year as a new training set to reoptimize the internal parameters of the random forest model, enabling the probabilistic model to adapt to changes in weight relationships caused by the evolution of the vegetation-landform system.

[0061] A habitat suitability model based on multiple environmental factors integrates average beach elevation, scour rate, and flooding frequency to calculate a comprehensive suitability index. This is combined with a random forest probabilistic model to assess vegetation growth, decline, and spatial expansion trends, simulating the evolution of vegetation in a dynamic environment. Incorporating a two-way feedback mechanism, the updated vegetation distribution and parameters are fed back to the vegetation response module, significantly improving the realism of coupled biophysical processes and enhancing the accuracy of long-term predictions of eco-geomorphic evolution, providing reliable support for ecological assessments and engineering design.

[0062] The updated spatial distribution of vegetation, physiological parameters (cover, height, stem density), synthetic vegetation areas, and hydraulic impact indices were integrated to generate a biogeomorphic evolution dataset. The data were verified for consistency with the time steps of hydrodynamic conditions and geomorphic elevations, and visualizations were generated, including a spatial distribution map of vegetation cover, a distribution map of the suitability index, and a probabilistic time series plot of evolutionary trends (growth, decline, and expansion). These results were fed back into the simulation construction module to update hydrodynamic and sediment transport parameters and store them for subsequent ecogeomorphic analysis.

[0063] Load the vegetation status verification data provided by the Data Acquisition Module, stored in Shapefile format, including measured vegetation cover, height, and stem density. Run the Landform Evolution Module, covering the verification data time period, extract the output vegetation distribution, suitability index, and physiological parameters, compare them with the measured data, calculate the deviation, adjust the factor weights in the habitat suitability model and the parameters of the random forest model, iterate and optimize to minimize the error, verify the impact of vegetation distribution on hydrodynamics and sediment transport, and confirm the consistency of the velocity field and sedimentation rate output by the Simulation Construction Module. Verification results, error indicators, and correlation coefficients are stored in CSV format, and the adjusted parameters are stored in JSON format. Update the parameter library to ensure model reliability.

[0064] Accelerated simulation module: combines physical process simulation with landform evolution calculation, introduces adaptive verification and reset steps, and realizes accelerated simulation of tidal flat vegetation in beach consolidation and wave dissipation.

[0065] Start the accelerated simulation module, load and integrate the current state output data from the simulation construction module, vegetation response module and landform evolution module, including the current landform elevation, vegetation spatial distribution, physiological parameters and hydraulic drag force parameters; initialize the control parameters of the accelerated simulation, which are stored in JSON format and include the total time scale of the simulation, the macroscopic landform evolution time step, and the key system indicators used for state verification and their respective stable state thresholds.

[0066] The accelerated simulation begins with a high-resolution physical process simulation. This simulation continuously runs the simulation building module, vegetation response module, and landform evolution module in a fully coupled mode over a representative hydrodynamic cycle. The simulation calculates the net landform change trend field caused by the combined effects of physical and biological processes over that cycle. The representative hydrodynamic cycle is preferably one that reflects the main tidal dynamic characteristics, such as a complete tidal cycle (approximately 15 days) encompassing both spring and neap tides, to ensure that the calculated landform change trends are statistically representative. To ensure numerical stability, a Courant-Friedrich-Löss number (CFL number) check is performed within each computational time step of this high-resolution simulation to adaptively adjust the step size.

[0067] To ensure numerical stability for high-resolution physical process simulations, the system performs adaptive step-size control based on the Courant-Friedrich-Ludwig number (CFL) condition within each computational time step. The control logic is as follows: At the beginning of each time step, the system calculates the CFL number for each computational grid cell based on the cell size, the current flow velocity, and the selected time step. The CFL number represents the ratio of the distance information propagates within a single time step to the grid size. The system then finds the maximum CFL number in the entire computational domain and compares it with a preset safety threshold that is strictly less than 1.

[0068] If the maximum CFL number does not exceed the safety threshold, the current time step is considered stable and the calculation will continue. Conversely, if the maximum CFL number exceeds the safety threshold, it means that the current time step is too long, which may cause the numerical solution to diverge. In this case, the system will automatically reject the calculation of the current step, reduce the time step by a preset reduction ratio, and then recalculate the step using the new, shorter time step. This process will be repeated until the maximum CFL number meets the stability condition. This adaptive step size control method can effectively avoid the occurrence of numerical instability and ensure the reliability of the simulation results.

[0069] After obtaining the net change trend field, the system enters a cyclical geomorphological evolution process, performing adaptive state verification and reset after each macroscopic time step. This verification logic is implemented by monitoring a set of preset key system indicators, including the average beach elevation change across the entire region, the total vegetation cover area, and the wave height attenuation rate at key sections. The stability condition is determined by a dynamic relative change threshold. Specifically, after each macro-time step, the system calculates the value of each key indicator and compares it with the value of the previous macro-time step. If the relative change rate of any indicator exceeds a preset threshold percentage, the system state is considered to have changed significantly and the current net change trend field is no longer applicable. At this point, the system automatically interrupts the acceleration cycle and returns to re-execute the high-resolution physical process simulation based on the latest topography to generate an updated net change trend field. The preset threshold percentage is determined by selecting a specific reset threshold percentage based on the preset interval of a characteristic parameter representing landform activity (for example, the historical average annual erosion and deposition rate). The mapping relationship between the interval and the threshold is predefined: if the average annual erosion and deposition rate is less than 5 cm / year, the threshold is 5%; if the rate is between 5 and 20 cm / year, the threshold is 15%; if the rate is greater than 20 cm / year, the threshold is 25%. In the adaptive verification and reset steps, in addition to determining the amount of change in the key indicator, the direction of change in the indicator is also determined. The system records whether key indicators increase or decrease over the most recent consecutive time steps, forming a sequence of change directions. When this sequence exhibits a persistent alternating reversal pattern (for example, a continuous pattern of "increase-decrease-increase"), the simulation is judged to contain unphysical numerical oscillations, triggering a reset step. This method can identify and suppress numerical oscillations and monitor changing behavioral patterns, enabling earlier and more accurate detection of signs of simulation instability, ensuring the convergence and physical validity of long-term evolution predictions.

[0070] By repeating this loop until the set total simulation timescale is reached, the system completes the long-term evolution simulation. After each geomorphic evolution time step, the system stores the landform elevation, vegetation status, and key hydrodynamic conditions at that moment to record the complete dynamic evolution process. After the simulation is completed, the data from all time steps are integrated to generate a comprehensive dataset reflecting the long-term geomorphic evolution and vegetation effects. This dataset is stored in NetCDF format and can generate corresponding visualization results. The final data is fed back to other modules to support subsequent analysis and prediction.

[0071] By generating a net change trend field through high-resolution physical process simulation within a representative period, and combining key indicator verification in the cyclic calculation to dynamically update the terrain, it is possible to efficiently simulate the long-term landform evolution and the dynamic processes of vegetation beach consolidation and wave-breaking effects. The adaptive verification and reset mechanism ensures the applicability of the trend field under environmental changes, improves the prediction accuracy of multi-scale coupling, and provides reliable support for refined ecological assessment and engineering design.

[0072] The validation data provided by the data acquisition module is loaded and stored in Shapefile and NetCDF formats. This data includes measured landform elevation changes, vegetation distribution, and hydraulic effect data. The accelerated simulation module is run, covering the time period of the validation data. The model outputs for beach consolidation (based on the long-term trend of sedimentation) and wave dissipation (based on the long-term amplitude of wave height attenuation) are extracted and compared with the measured data. The deviation between the simulated and measured values ​​is calculated, and model accuracy is assessed using root mean square error (RMS) and correlation coefficients. The model is optimized by iteratively adjusting the landform evolution time step, the stability thresholds of key system indicators, the CFL safety threshold, and the time step reduction ratio until the error converges. The validation results are stored in CSV format to confirm that the model accurately captures the long-term beach consolidation and wave dissipation effects, ensuring the reliability of long-term predictions.

[0073] Through the collaborative work of data acquisition, simulation construction, vegetation dynamic response, biogeomorphic evolution and accelerated simulation modules, the multi-scale coupling of rapid hydrodynamic processes and slow bio-geomorphic processes is achieved, significantly improving the accuracy of long-term predictions. The system uses unstructured triangular grids and wave-current sediment coupling mechanisms to accurately simulate the sediment transport and geomorphic evolution of complex coastlines; based on the habitat suitability model and vegetation mechanics model, it truly reflects the growth, decline and hydraulic effects of vegetation; through adaptive time step control and key indicator verification, it optimizes the stability and efficiency of long-term evolution simulation. Compared with traditional methods, the present invention effectively captures the evolution laws of ecological geomorphic systems, meets the needs of refined ecological assessment and engineering design, and provides reliable support for the scientific evaluation of the protection effect of tidal flat vegetation.

[0074] Example 2: Launch a multi-scale numerical simulation system for the effects of reed vegetation on beach consolidation and wave dissipation on mudflats. Simulate the effects of reed vegetation on beach consolidation and wave dissipation in Area B of the mudflat under five years of tidal action. Load the terrain basic data, boundary condition data, and initial vegetation state data provided by the data acquisition module, all stored in NetCDF and Shapefile formats. The basic terrain data includes a digital elevation model (average elevation 0.5 m, silt bottom). The elevation and bottom distribution are extracted by parsing NetCDF files. An open-source mesh generation tool is used to generate an unstructured triangulated mesh. Local mesh encryption is performed on tidal gullies and reed-covered areas to improve resolution. The elevation data is assigned to grid nodes using the Kriging interpolation method to generate an initial digital elevation model. Boundary condition data include tidal flow velocity and wave height. CSV and NetCDF files are parsed, and a 30-minute resolution time series is generated through spline interpolation and assigned to grid boundary nodes. The initial vegetation state data is stored in Shapefile format, including the reed distribution. It is assigned to grid cells through spatial query and a vegetation attribute table is generated. The parameter library is loaded to obtain the factor weights and coefficients of the reed habitat suitability model. The obtained reed data is checked for consistency and processed to generate an initial dataset and stored in NetCDF format.

[0075] Furthermore, the vegetation response module is started, the hydrodynamic condition data output by the simulation construction module is loaded, the initial state data of the reed is loaded, the beam theory is used to regard the reed as a flexible cantilever beam, the Morison equation is used to calculate the hydrodynamic load, and the stem bending deformation is calculated based on the flow velocity and the wave track speed, and a deformation curve is generated. The DBSCAN clustering algorithm is used to generate a synthetic vegetation area based on the coverage rate and the distance between grids, and the spatial density coefficient is calculated. Based on the deformation curve and vegetation attributes, the hydraulic drag coefficient is calculated through a pre-calibrated function to generate a hydraulic impact index. In this embodiment, at a certain moment in the simulation, the system monitors the reeds in a certain grid unit, whose natural height H = 1.8 meters and the initial hydraulic drag coefficient is 1.1. According to calculations, the bending displacement of the top of the stem at this moment is 0.36 meters, so its relative curvature is 0.2. According to the empirical function model described in the present invention, and using the preset reed empirical coefficient ( =-0.5, =0.15), the updated hydraulic drag coefficient is calculated to be 0.997. Through this calculation, the system concludes that the drag coefficient of the reed unit has been dynamically updated from the initial 1.1 to approximately 0.997 at that moment. This result is then used to generate the hydraulic influence index. The generated data is stored in NetCDF format and fed back to the simulation building module to adjust the flow field calculation.

[0076] The geomorphic evolution module is launched, loading the geomorphic elevation and hydrodynamic conditions output by the simulation construction module, as well as the reed distribution and hydraulic impact index from the vegetation response module, all stored in NetCDF format. The reed niche parameters are loaded, and based on the habitat suitability model, the average beach elevation, average scour rate, and average flooding frequency of each grid cell are extracted. A comprehensive suitability index is calculated using a logistic regression method. A random forest model is used to input the suitability index and the coverage of adjacent grid cells to predict the growth, decline, or expansion trend of reeds. For growing grids, the coverage and height are increased; for declining grids, the coverage is reduced or vegetation is removed; and for expanding grids, a spatial diffusion model is used to expand to adjacent grids. The updated vegetation distribution and physiological parameters are fed back to the vegetation response module to generate a new synthetic vegetation area and hydraulic impact index, which are stored in NetCDF format.

[0077] The accelerated simulation module is launched, and the output data from the simulation construction module, vegetation response module, and landform evolution module are loaded. Simulation parameters are initialized for a five-year timescale, with a macroscopic time step of one month. A high-resolution physical process simulation is run over one tidal cycle to calculate the net landform change trend field. At each computational time step, the CFL number is solved. If it exceeds the safety threshold of 0.9, the step size is shortened by 0.8 and the calculation is repeated to ensure numerical stability. In the loop, the net change trend field is applied to the terrain using a macroscopic time step of one month. After each step, key indicators are verified, including the rate of change in elevation, rate of change in coverage, and rate of change in the hydraulic impact index. If stability conditions are met, the loop is continued; otherwise, the high-resolution simulation is rerun to update the trend field. After five years of simulation, the landform elevation change, reed distribution, and hydraulic impact index are output and stored in NetCDF format.

[0078] The accelerated simulation results were integrated to generate a comprehensive data set with a time scale of 5 years, including changes in landform elevation (reflecting scouring and deposition trends), reed distribution (coverage rate, height, reflecting the beach consolidation effect) and hydraulic impact index (reflecting the wave-breaking effect). The consistency of the data with the time steps of each module was verified, and visualization results were generated, including the spatial distribution map of landform elevation changes, the reed distribution map and the hydraulic impact index time series map. The results were fed back to the simulation construction module to update the terrain, the vegetation response module to update the drag force parameters, and the landform evolution module to update the vegetation distribution, and stored as input for ecological and geomorphological analysis.

[0079] Load measured validation data, including landform elevation changes, reed cover, beach consolidation, and wave dissipation. Run the accelerated simulation module, extract the output, compare it with the measured data, and calculate the root mean square error. Adjust the factor weights in the habitat suitability model, as well as the classification rule parameters, CFL threshold, step-length reduction ratio, and key indicator threshold in the probability model, and iterate to minimize the error. Store the validation results in CSV format (including error indicators and correlation coefficients), store the adjusted parameters in JSON format, and update the parameter library to ensure model reliability.

[0080] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A multi-scale numerical simulation system for the beach vegetation consolidation and wave-breaking effect, characterized in that: include: Data acquisition module: obtains terrain basic data, boundary condition data, vegetation physical parameters and vegetation ecological parameters; Simulation construction module: Use the basic terrain data to establish the numerical calculation grid and terrain, use the boundary condition data as the driving condition, and update the landform elevation and hydrodynamic conditions based on the wave-current-sediment coupling mechanism; Vegetation response module: Calculates vegetation flexible deformation based on vegetation mechanical model, combined with hydrodynamic conditions and vegetation physical parameters, and updates hydraulic drag force parameters based on the vegetation flexible deformation through a predetermined functional relationship; Landform Evolution Module: Based on the habitat suitability model, the module evaluates the probabilistic trends of vegetation growth, decline, and expansion based on landform elevation, hydrodynamic conditions, and vegetation ecological parameters, and feeds the updated vegetation status back to the Vegetation Response Module. Accelerated simulation module: combines physical process simulation with landform evolution calculation, introduces adaptive verification and reset steps, and realizes accelerated simulation of tidal flat vegetation in beach consolidation and wave dissipation.

2. The multi-scale numerical simulation system for beach vegetation consolidation and wave elimination according to claim 1 is characterized in that: When the simulation construction module establishes the initial numerical calculation grid and terrain, it uses the high-precision water depth data in the terrain basic data to establish a digital elevation model. For complex coastlines containing tidal gullies and mudflats, it generates an unstructured triangular mesh covering the entire study area, and performs local mesh encryption on the tidal gullies, vegetation-covered areas, and areas of key concern in the engineering area. The elevation data contained in the digital elevation model is interpolated onto the grid nodes to form the initial terrain required for model calculation.

3. The multi-scale numerical simulation system for beach vegetation consolidation and wave elimination according to claim 1 is characterized in that: In the simulation construction module, the calculation process executed by the wave-flow-sediment coupling mechanism within a coupling time step is as follows: the hydrodynamic solver calculates the average water flow field at the current moment and transmits it to the wave solver. The wave solver calculates the wave propagation, deformation and wave-induced radiation stress on this basis. The wave-induced radiation stress is then fed back to the hydrodynamic solver and superimposed with the bed shear stress of the water flow to obtain the bottom bed shear stress. The bottom bed shear stress is then used to drive the sediment transport formula to calculate and update the landform elevation change at the current time step.

4. The multi-scale numerical simulation system for beach vegetation consolidation and wave elimination according to claim 1 is characterized in that: The simulation construction module is further configured to call the data acquisition module to obtain physical property data including vegetation elastic modulus, and pass the elastic modulus as a core parameter to the cantilever beam mechanical model in the vegetation response module for calculating flexible bending deformation.

5. The multi-scale numerical simulation system for beach vegetation consolidation and wave elimination according to claim 1 is characterized in that: The hydrodynamic load acting on the vegetation is calculated using the Morrison equation, and the vegetation is mechanically modeled using the cantilever beam theory. The load is applied to the model to obtain the vegetation bending displacement, which is used as a dynamic deformation parameter. The bending displacement and the geometric and mechanical properties of the vegetation are input into an empirical function model that characterizes the relationship between the dynamic bending of vegetation and the water drag force to update the hydraulic drag force parameter. The empirical function model is established through fitting and inverse calculation based on the dynamic deformation and total drag force data of vegetation measured in a controlled physical model experiment.

6. The multi-scale numerical simulation system for beach vegetation consolidation and wave elimination according to claim 1 is characterized in that: The landform evolution module evaluates the evolution trend of vegetation through a habitat suitability model. The habitat suitability model takes multiple environmental factors of the vegetation location, including average beach elevation, average scour rate and average flooding frequency, as comprehensive inputs, and calculates a comprehensive habitat suitability index based on the comprehensive inputs. According to the value of the habitat suitability index, combined with the probability model, the growth, decline and spatial expansion trends of the vegetation in the next landform evolution time step are determined.

7. The multi-scale numerical simulation system for beach vegetation consolidation and wave elimination according to claim 1 is characterized in that: The accelerated simulation module also includes an adaptive time step control. The execution process of the adaptive time step control is as follows: when each calculation time step is completed, the Courant-Friedrich-Low dimension used to judge numerical stability is calculated in the entire calculation domain. If the Courant-Friedrich-Low dimension is greater than a preset safety threshold, the calculation result of the current time step is rejected, and the step size of the next time step is shortened by a preset ratio and then recalculated.

8. The multi-scale numerical simulation system for beach vegetation consolidation and wave elimination according to claim 1 is characterized in that: The method for combining physical process simulation with landform evolution calculation includes: performing high-resolution physical process simulation of a representative period to calculate a net change trend field of the landform; in a cyclic calculation step, applying the net change trend field to update the terrain within a macroscopic landform evolution time step; after the time step ends, verifying the preset system key indicators; when the verification results of the key indicators meet the preset stability conditions, continuing to use the current net change trend field to repeat the cyclic calculation steps; when the verification results do not meet the stability conditions, re-performing the high-resolution physical process simulation to update the net change trend field.

Images