A method and system for simulating and predicting biogeomorphic evolution of multi-vegetation salt marshes

Abstract

The present invention discloses a multi-vegetation salt marsh bio-geomorphological evolution prediction method and system, constructs a coupling model based on a water-sand dynamics model and a vegetation dynamics model; uses the water-sand dynamics model to simulate water-sand flow, obtains water-sand flow simulation results after the simulation, calculates the characteristic values ​​of stress factors, and interpolates the characteristic values ​​of three stress factors from the water-sand dynamics grid into the vegetation grid of the vegetation dynamics model; uses the vegetation dynamics model to simulate vegetation distribution, completes the colonization, succession, stress death, annual death and expansion process of vegetation on the vegetation grid, and after the simulation, weighted averages the simulation results of vegetation distribution from the vegetation grid to the water-sand dynamics grid, and inputs them into the water-sand dynamics model; achieves long-term simulation of salt marsh bio-geomorphology through several cycles. The simulation prediction method of the present invention can solve the problems that the existing model can only consider a single vegetation species and the tidal current dynamics are oversimplified, and ensures the scientificity and accuracy of the simulation process.

Description

Technical Field

[0001] The invention belongs to the field of salt marsh biogeomorphic evolution, and in particular relates to a multi-vegetation salt marsh biogeomorphic evolution prediction method and system. Background Art

[0002] Salt marshes are located in the intertidal zone of estuaries, bays and tidal river sections. They are dynamic areas between land and sea and are considered to be one of the most productive natural ecosystems in the world, with important ecological functions and extremely high social value. Numerous studies have shown that salt marsh vegetation can weaken currents, break waves and promote siltation, and has great potential in coastal protection and disaster reduction.

[0003] At present, large-scale salt marsh wetland restoration projects with tidal dynamics restoration as the core are receiving more and more attention, mainly including strategies such as "managed realignment", "managed retreat" and "regulated tidal exchange".

[0004] In order to ensure that the ideal restoration goals are achieved, strict numerical modeling calculations are usually required before the implementation of such projects. The models involve the interaction of multiple processes such as vegetation, water flow, and sediment. However, the complex dynamic mechanism poses a huge challenge to the modeling work. Although relevant research has received increasing attention in recent years, a few scholars have attempted to construct a two-dimensional salt marsh biogeomorphological coupling model, but such research is more about generalized models to study the evolution mechanism of salt marsh biogeomorphology under natural conditions. It can only consider a single vegetation species and over-simplified tidal dynamics. The lack of scientificity and accuracy will restrict its further promotion and application. Summary of the invention

[0005] In view of the problems existing in the prior art, the present invention provides a method and system for predicting the biogeomorphic evolution of multi-vegetation salt marshes, which can solve the problems that the existing models can only consider a single vegetation species and the tidal dynamics are overly simplified, and ensure the scientificity and accuracy of the simulation process.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a method for predicting biogeomorphological evolution of multi-vegetation salt marsh, comprising the following steps:

[0007] S1. Taking the topographic data, vegetation data, water-sediment dynamics grid data, closed boundary data, and tidal open boundary data of the target area as input, combined with the preset hydrodynamic parameters and sediment dynamics parameters, a water-sediment dynamics model with the characteristics of three stress factors, namely, flooding time, maximum bottom shear stress, and bottom scouring and silting, as output is constructed;

[0008] S2. Taking the vegetation expansion process and vegetation extinction process as input and combining the spatial scale of the vegetation grid, a vegetation dynamics model with the distribution of vegetation as output is constructed; the vegetation expansion process includes: seedling colonization, vegetation succession and vegetation lateral expansion, and the vegetation extinction process includes: stress death and annual death;

[0009] The occurrence probability of vegetation expansion and vegetation extinction is determined by the pressure function S k Calculate, where

[0010] Seedling colonization refers to the process in which a grid unit changes from a bare beach state to any vegetation state, that is, the probability of colonization of vegetation type i is P when 0 changes to i. i est The calculation of is as follows:

[0011]

[0012] Where P i est is the basic probability of colonization;

[0013] Vegetation succession means that vegetation is replaced from one vegetation type to another, and the grid cell state is transformed from i to j, i≠j≠0. The calculation is as follows:

[0014]

[0015] Where: is the basic probability of the preset vegetation succession from species i to species j;

[0016] Stress death refers to the transformation of any vegetation species state i to beach state 0 due to environmental pressure. Death is triggered by the thresholds of the characteristic values ​​of the three stress factors. The probability of death of vegetation species i is The calculation of is as follows:

[0017]

[0018] Annual mortality refers to the transformation of any vegetation type state i to bare beach state 0 due to the natural cycle of annual vegetation. The calculation method of the annual mortality probability of vegetation type i is as follows:

[0019]

[0020] Where: P i ann The basic probability of vegetation death per year set by the user;

[0021] The horizontal expansion of vegetation requires several iterations in an ecological process. Assume that the vegetation grid size is Δx and the annual expansion rate of a vegetation is R. i , and Ri >Δx,P i exp The following relationship is satisfied:

[0022]

[0023] Where: N is the number of iterations, R i is the average annual expansion distance of vegetation i, and the corresponding As follows:

[0024]

[0025] Where: N neighbour is the number of vegetation i in the neighboring cells of the grid cell.

[0026] S3. Couple the water-sediment dynamics model and the vegetation dynamics model. Specifically, the stress factor characteristics calculated by the water-sediment dynamics model are input into the vegetation dynamics model. The vegetation in the vegetation dynamics model completes the expansion and extinction process according to the characteristic values ​​to obtain the vegetation distribution change. The changed vegetation distribution is then input into the water-sediment dynamics model again to complete a cycle. After a preset number of iterations, the salt marsh biogeomorphological evolution results are obtained.

[0027] Furthermore, the aforementioned hydrodynamic parameters include: resistance coefficient, and the sediment dynamic parameters include: geomorphic acceleration factor, critical scour shear stress, and critical siltation shear stress.

[0028] Furthermore, the aforementioned flooding time FLT is calculated as follows:

[0029]

[0030] Where, T flood is the time when the bottom is flooded, which is equal to the number of time steps with water depth greater than 0 in the water depth time series output by the model, T sim is the model simulation time, which is equal to the total number of time steps of the model;

[0031] The maximum bed shear stress MaxBSS is calculated as follows:

[0032] MaxBSS = max(τ)

[0033] Where, τ is the time series of bed shear stress output by the model;

[0034] The BLC of the bottom elevation change is calculated as follows:

[0035] BLC=h1-h0

[0036] Where h1 is the elevation of the bottom bed after the model simulation, and h0 is the initial elevation of the bottom bed.

[0037] Furthermore, the aforementioned pressure functions are of three types, as follows:

[0038] The first type is the constant type. When the pressure value s satisfies s min <s<s max When S k =1, otherwise S k =0;

[0039] The second type is linear, when s<s min When S k = 0, when s min <s<s max When S k =(ss min ) / (s max -s min ), when s>s max When S k =1;

[0040] The third type is parabolic, when s<s min or s>s max When S k = 0, when s min <s<s max hour, where s min and max They respectively represent the minimum and maximum values ​​of the pressure value range when different pressure functions are in effect.

[0041] Furthermore, the aforementioned water and sand dynamics grid is a triangular unstructured grid, the vegetation grid is a square unstructured grid with a smaller spatial scale than the water and sand dynamics grid, and the number of water and sand dynamics grid cells is smaller than that of the vegetation grid; the characteristic value distribution of vegetation distribution and stress factors is saved in both the water and sand dynamics grid and the vegetation grid, and the data in the two sets of grids need to be transmitted in each cycle to ensure effective synchronization and storage of data in the two sets of grids.

[0042] Furthermore, the aforementioned one cycle includes the following sub-steps:

[0043] S3.1, the water-sediment dynamics model runs, the model simulation results are output, the characteristic values ​​of the three stress factors are calculated, and the results are stored in the water-sediment dynamics grid;

[0044] S3.2, using the interpolation method to convert the characteristic values ​​of the three stress factors stored in the water and sediment dynamics grid into data in the vegetation grid;

[0045] S3.3, the vegetation dynamics model completes the five processes of vegetation colonization, succession, stress death, annual death and lateral expansion according to the characteristic values ​​of the stress factors stored in the vegetation grid, and stores the changed vegetation distribution results in the vegetation grid;

[0046] S3.4. The vegetation distribution stored in the vegetation grid is accumulated into the water-sediment dynamics model using a weighted average method.

[0047] Another aspect of the present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the steps of any one of the methods described in the present invention when executing the computer program.

[0048] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program implements the steps of any one of the methods described in the present invention when executed by a processor.

[0049] Compared with the prior art, the beneficial technical effects of the above technical solutions adopted in the present invention are as follows: the present invention provides a method and system for predicting the biogeomorphological evolution of multi-vegetation salt marshes. The coupling model includes a water-sediment dynamics model constructed based on Delft3D-FM software and a vegetation dynamics model constructed based on Python language. Multiple vegetations can be used in the vegetation dynamics model to solve the problem that the prior art model can only consider a single vegetation species. In addition, the water-sediment dynamics model is simulated by Delft3D-FM and the accuracy is verified to construct a complex water dynamics field, which can be applied in a real environment to solve the problem of over-simplification of tidal dynamics. The vegetation dynamics model is constructed based on the real life history of salt marsh vegetation, taking into account water and sand. The impact of dynamic conditions on vegetation is studied by correcting vegetation parameters through measured data to ensure the scientificity and rigor of the results; the model cycle is automated throughout using Python, and parameter settings and overall control are performed through external text files, which is convenient for general technicians to use, and the code can also be easily modified to optimize or add functions; the vegetation dynamics model is written in Python and coupled with the mature open source water and sand dynamics model Delft3D-FM. Through technologies such as Voronoi diagrams and RTree, an efficient grid data interaction channel is established. By establishing a model of the interaction between water and sand dynamics and salt marsh vegetation, the purpose of long-term simulation of salt marsh biogeomorphology is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 The present invention provides a schematic diagram of the coupling framework and technical flow of the biogeomorphology model. In the figure, (a) is a diagram of the overall framework of the model, (b) is a diagram of the grid data transmission method, and (c) is a diagram of the model calculation flow.

[0051] Figure 21 is a schematic diagram of the technical implementation of the subordinate relationship between the water and sand dynamics grid and the vegetation grid provided by the present invention, in which (a) is a schematic diagram of determining the vegetation model calculation area, (b) is a schematic diagram of the unstructured grid nodes in the selected area, (c) is a schematic diagram of constructing an overlapping grid with a mask, (d) is a schematic diagram of calculating the Voronoi of each unstructured node, and (e) is a schematic diagram of finding the corresponding structured grid unit through the RTree algorithm.

[0052] Figure 3 It is the model grid and the initial and designed topographic maps in Example 2 provided by the present invention, in which (a) is the model grid and the initial topographic map, (b) is the map of the dam with two openings + the excavated tidal ditch, and (c) is the map of the dam with three openings + the excavated tidal ditch.

[0053] Figure 4 It is a diagram of the evolution process of the salt marsh biogeomorphology after the whole dike is removed in Example 2 provided by the present invention, in which (a) is the evolution diagram in the first year, (b) is the evolution diagram in the fifth year, (c) is the evolution diagram in the tenth year, and (d) is the evolution diagram in the twentieth year.

[0054] Figure 5 This is a diagram showing the evolution of salt marsh biogeomorphology under various working conditions in Example 2 provided by the present invention. DETAILED DESCRIPTION

[0055] In order to better understand the technical content of the present invention, specific embodiments are given and described as follows in conjunction with the accompanying drawings.

[0056] Various aspects of the invention are described herein with reference to the accompanying drawings, in which many illustrative embodiments are shown. The embodiments of the invention are not limited to those described in the accompanying drawings. It should be understood that the invention is implemented by any of the various concepts and embodiments described above, as well as the concepts and embodiments described in detail below, because the concepts and embodiments disclosed in the invention are not limited to any implementation. In addition, some aspects disclosed in the invention may be used alone or in any appropriate combination with other aspects disclosed in the invention.

[0057] Embodiment 1:

[0058] like Figure 1 As shown in (a), (b), and (c), the present invention provides a method for predicting biogeomorphic evolution of multi-vegetation salt marshes, comprising the following steps:

[0059] S1. Taking the topographic data, vegetation data, water-sediment dynamics grid data, closed boundary data, and tidal open boundary data of the target area as input, combined with the preset hydrodynamic parameters and sediment dynamics parameters, a water-sediment dynamics model with the characteristics of three stress factors, namely, flooding time, maximum bottom shear stress, and bottom scouring and silting, as output is constructed based on Delft3D-FM software;

[0060] S2. Taking the vegetation expansion process and vegetation extinction process as input and combining the spatial scale of the vegetation grid, a vegetation dynamics model with the vegetation distribution result as output is constructed based on Python language; the vegetation expansion process includes: seedling colonization, vegetation succession and vegetation lateral expansion, and the vegetation extinction process includes: stress death and annual death;

[0061] S3. Set the running time of the coupling model, couple the water-sediment dynamics model and the vegetation dynamics model, use the water-sediment dynamics model to simulate the water-sediment flow according to the simulation time, and shorten the simulation time of the water-sediment dynamics model through the landform acceleration factor. After the simulation, the water-sediment flow simulation results are obtained. The characteristic values ​​of the three stress factors, namely, the flooding time, the maximum bed shear stress and the bed scouring and silting, are calculated through the water-sediment flow simulation results. The characteristic values ​​of the three stress factors are then stored in the water-sediment dynamics grid of the water-sediment dynamics model. Finally, the characteristic values ​​of the three stress factors are interpolated from the water-sediment dynamics grid into the vegetation grid of the vegetation dynamics model.

[0062] The vegetation dynamics model is used to simulate vegetation distribution according to the simulation time. Through the set parameters, the vegetation colonization, succession, stress death, annual death and expansion processes are completed on the vegetation grid. The stress death process uses three stress factor characteristic values. After the simulation is completed, the simulation results of vegetation distribution are saved, and then the simulation results of vegetation distribution are weighted averaged from the vegetation grid to the water and sediment dynamics grid, and then sent to the water and sediment dynamics model;

[0063] The above cycle process is repeated to achieve a long-term simulation of salt marsh biogeomorphology through several cycles. After the simulation, the result files of the water-sediment dynamics model and the result files of the vegetation dynamics model are read through Python, and the matplotlib library is used to overlay the result files of the water-sediment dynamics model and the result files of the vegetation dynamics model to draw the elevation distribution map and multi-vegetation distribution map of the regional prediction in any year.

[0064] In each cycle, the first step is to run the water and sediment dynamics model, output the model simulation results, use Python scripts to automatically read the result files, and calculate the characteristic values ​​of the three stress factors, and store the results in the water and sediment dynamics grid; the second step is to use interpolation methods (such as linear interpolation or cubic spline interpolation) to convert the characteristic values ​​of the three stress factors stored in the water and sediment dynamics grid into data in the vegetation grid; the third step is that the vegetation model completes the five processes of vegetation colonization, succession, stress death, annual death and lateral expansion according to the characteristic values ​​of the stress factors stored in the vegetation, and stores the changed vegetation distribution results in the vegetation grid; the fourth step is to use the weighted average method to accumulate the vegetation distribution stored in the vegetation grid to the water and sediment dynamics vegetation grid, and create a vegetation resistance file that can be recognized by Delft3D-FM, and input it into the next cycle. The specific steps are as follows:

[0065] The first step of running the water-sediment dynamics model is as follows: the Delft3D-FM nc result file is read through the netCDF4 library, and the time series of its parameters is saved in the center of the grid (face). The horizontal and vertical coordinates of the center are numpy arrays with m rows and 1 column, and the time series of parameters is stored as a numpy array with m rows and n columns (m is the number of center points, and n is the number of simulation steps). The characteristic value of the stress factor calculated according to formulas (1) to (3) is a numpy array with m rows and 1 column, which is still stored in the water-sediment dynamics grid.

[0066] The flooding time FLT is calculated as follows:

[0067]

[0068] Where, T flood is the time when the bottom is flooded, which is equal to the number of time steps with water depth greater than 0 in the water depth time series output by the model, T sim is the model simulation time, which is equal to the total number of time steps of the model;

[0069] The maximum bed shear stress MaxBSS is calculated as follows:

[0070] MaxBSS = max(τ)

[0071] Where, τ is the time series of bed shear stress output by the model;

[0072] The BLC of the bottom elevation change is calculated as follows:

[0073] BLC=h1-h0

[0074] Where h1 is the elevation of the bottom bed after the model simulation, and h0 is the initial elevation of the bottom bed.

[0075] The second step uses the interpolation method: use Python's interpolate function in the scipy library to interpolate the characteristic values ​​of the stress factors stored in the water and sediment dynamics grid in the numpy array of m rows and 1 column into the vegetation grid. The result is a numpy array of x rows and 1 column (x is the number of vegetation grid cells), which realizes the transmission of the characteristic values ​​of the three stress factors (flooding time, maximum bed shear stress and bed elevation change) from the water and sediment dynamics grid to the vegetation dynamics grid.

[0076] The third step is as follows: the stress factor characteristic values ​​stored in the vegetation grid are fed into formula (6), and the probability of stress death of vegetation in each grid cell is calculated according to the resistance threshold of vegetation to each stress factor set by the user and the selected pressure function; other processes have no direct relationship with the stress factor characteristic values, and the user needs to select the appropriate threshold and pressure function. In each cycle, the vegetation has an initial distribution, and after five processes, the changed vegetation distribution is generated and saved as a numpy array. Probability of death of vegetation type i The calculation of is as follows:

[0077]

[0078] Annual mortality refers to the transformation of any vegetation type state i to bare beach state 0 due to the natural cycle of annual vegetation. The calculation method of the annual mortality probability of vegetation type i is as follows:

[0079]

[0080] Where: The basic probability of vegetation death per year set by the user;

[0081] The fourth step is to accumulate the vegetation distribution stored in the vegetation grid to the water-sand dynamics vegetation grid and create a vegetation resistance file. By identifying the several vegetation grid cells corresponding to each water-sand dynamics grid cell, the simulation results of the vegetation distribution are weighted averaged from the vegetation grid to the water-sand dynamics grid. The identification method is to call Python's geovoronoi and shapely libraries, and find the correspondence between the water-sand dynamics grid and the vegetation grid through the Voronoi diagram and RTree algorithm. The vegetation resistance files that Delft3D-FM can recognize are arl and ttd files, where the arl file stores the horizontal and vertical coordinates of the center point of the water-sand dynamics grid, the serial number and the vegetation coverage rate, and the ttd file stores the serial number, the height of the vegetation stem, the product of the diameter and density of the vegetation stem, the Cd resistance coefficient and the resistance formula. The two files are associated by the serial number. The vegetation distribution in the vegetation grid can be obtained by weighted averaging the vegetation coverage stored in the water and sand dynamics grid, while the stem height, diameter and density are fixed parameters set by the user, the Cd resistance coefficient is an empirical coefficient calibrated according to the experiment (recommended to be 2 to 5), and the resistance formula is recommended to choose the widely used Baptist formula. The processing result is used as the bottom bed roughness provided by the vegetation and is automatically sent to the water and sand dynamics simulation in the next cycle as an input condition. The above content can complete the transmission of spatial vegetation distribution and vegetation resistance to water flow from the vegetation grid to the water and sand dynamics grid.

[0082] This embodiment 1 provides a method for simulating and predicting the biogeomorphological evolution of multi-vegetation salt marshes. The coupling model includes a water-sediment dynamics model constructed based on Delft3D-FM software and a vegetation dynamics model constructed based on Python language. Multiple vegetations can be used in the vegetation dynamics model to solve the problem that the existing technical model can only consider a single vegetation species. In addition, the water-sediment dynamics model is simulated by Delft3D-FM and the accuracy is verified to construct a complex water dynamics field, which can be applied in a real environment to solve the problem of over-simplification of tidal dynamics.

[0083] Embodiment 2:

[0084] like Figure 1 As shown in (a)(b)(c), this embodiment 2 provides a multi-vegetation salt marsh biogeomorphological evolution simulation prediction method, which focuses on the construction of a water-sand dynamics model and a vegetation dynamics model based on the Python language. The simulation of the salt marsh biogeomorphology is achieved through the coupling and circulation of the water-sand dynamics model and the vegetation dynamics model, which can guide technicians to complete the code in a similar way, specifically including the following steps:

[0085] Step 1: Construction and verification of water-sediment dynamics model. Based on terrain data, grid data, closed boundary data, tidal open boundary data, hydrodynamic parameters and sediment dynamics parameters, the water-sediment dynamics model is constructed according to conventional steps, including determination of the study area, drawing of the grid and insertion of terrain, selection of model parameters, etc. The setting of model parameters is debugged with reference to the default values. At the same time, the accuracy of the model is verified according to conventional steps, including verification of tide height, flow velocity, direction and suspended sediment concentration of several time series at several measuring points, and the relevant parameters are calibrated according to the verification results until the model accuracy meets the requirements. The constructed model is called a water-sediment dynamics model. The model grid is aligned with the initial and design terrain maps as shown in the figure. Figure 3 As shown in the figure. The terrain data comes from the measured and local nautical chart data, the grid is constructed by the Delft3D-FM model or SMS, the tidal boundary data comes from the TPXO model, and the measured water and sediment dynamics data (tidal level, flow velocity, flow direction and suspended sediment data) comes from the long-term or short-term observations of various hydrological stations. At this time, the Trachytopes module for simulating vegetation resistance is not used in the Delft3D-FM model.

[0086] Step 2: Selection of basic parameters of vegetation dynamics model. The relevant parameters of vegetation dynamics model should be calibrated according to measured data and related research, and the information required such as the spatial scale of vegetation grid and the number of simulation years should be selected according to actual needs and time cost. In the vegetation dynamics model, each vegetation needs to specify the maximum rhizome density, rhizome diameter and height respectively. The reference values ​​are shown in Table 1. In Table 1, SA refers to Spartina alterniflora, SS refers to Suaeda salsa, and PA refers to Phragmites australis. All three are typical dominant species of salt marsh vegetation in Jiangsu tidal flats. Vegetation parameters are saved separately as txt files in text format, waiting to be read by the vegetation dynamics model.

[0087] Table 1 Parameter settings of vegetation dynamics model

[0088] Parameter name Reference value unit Maximum root density of vegetation m 1200(SA),100(SS),300(PA) <![CDATA[Company m -2 > Vegetation root diameter d 5(SA),10(SS),10(PA) mm Vegetation height k 1.0(SA),0.4(SS),2.0(PA) m

[0089] Step 3: Start the formal simulation of the water-sediment dynamic model. When the coupled model is started, the vegetation dynamic model needs to be initialized, which mainly includes reading parameters, building vegetation grids, and arranging the initial vegetation state. The initial vegetation state needs to be given by the user.

[0090] (1) Setting up the water-sediment dynamics model. The simulation time of the water-sediment dynamics model is one year, which is consistent with the time step of the vegetation dynamics model. The water-sediment dynamics model at this time is basically the same as the model in step 1, but a larger geomorphic factor is required to speed up the simulation. The geomorphic factor selected at this time is 24, which corresponds to a geomorphic simulation time of one year and a water-sediment dynamics simulation time of half a month. During this half-month period, the model fully simulates the changes in water-sediment dynamics on the tidal flat under a complete tidal process (large and small tides). The simulation results are saved as nc files.

[0091] (2) Reading simulation results. The Delft3D-FM model uses the Trachytopes module. The simulation time is usually one year. It is possible to consider using a larger geomorphic coefficient to reduce the model's calculation time. The result file of Delft3D-FM is an nc file. The result file is read through Python's netCDF4 library. The main function used is the load_variable function. The position parameters that need to be read include the position of each node (node) of the grid and the position of the center point (face) of each grid unit. The corresponding variable names of the result file are mesh2d_node_x, mesh2d_node_y, mesh2d_face_x and mesh2d_face_y; the time series of the parameters that need to be read include water depth, bottom shear stress and elevation. The corresponding variable names of the result file are mesh2d_waterdepth, mesh2d_taus and mesh2d_mor_bl. The time series of the three parameters are corresponding.

[0092] (3) Calculation of characteristic values ​​of stress factors. The characteristic values ​​of stress factors are obtained by simply calculating the time series of the above parameters through the numpy library. The calculation results still correspond to the position of the center point of each grid unit. The characteristic values ​​of stress factors include flooding time, maximum bottom bed shear stress and bottom bed scouring and silting. Specifically,

[0093] The calculation method of submergence time FLT is as follows:

[0094]

[0095] Among them, T flood is the time when the bottom is flooded, which is equal to the number of time steps with water depth greater than 0 in the water depth time series output by the model, T sim is the model simulation time, which is equal to the total number of time steps of the model;

[0096] The calculation method of the maximum bed shear stress MaxBSS is as follows:

[0097] MaxBSS = max(τ)

[0098] Where, τ is the time series of bed shear stress output by the model;

[0099] The calculation method of the bed elevation change BLC is as follows:

[0100] BLC=h1-h0

[0101] Among them, h1 is the elevation of the bottom bed after the model simulation is completed, and h0 is the initial bottom bed elevation.

[0102] Step 4: The characteristic values ​​of the stress factors are transferred from the water-sediment dynamics model to the vegetation dynamics model. Figure 1 As shown in the figure, the grid of the vegetation model is a small-scale square structured grid, the water and sand dynamics grid is a triangular unstructured grid, and the vegetation grid is a square unstructured grid slightly smaller than the water and sand dynamics grid, in order to achieve a more accurate simulation. The vegetation grid is created by numpy according to the maximum and minimum values ​​in the x and y directions and the spatial size of each grid, and is stored and changed in the form of numpy.array. The interpolate.griddata function of the scipy library can complete the data transmission from the water and sand dynamics grid to the vegetation grid.

[0103] Step 5, start simulating the vegetation dynamics model. The vegetation dynamics model is written based on the vegetation life history, and the key stages of the vegetation life history include colonization, stress death, succession and expansion, all of which occur within one year. It mainly relies on Python's numpy library for vegetation modeling, data transmission and scientific computing. In the vegetation dynamics model, the most critical information is the state of each vegetation grid cell, where 0 represents the beach state and natural numbers starting from 1 represent each vegetation. The state of each vegetation cell will be determined by the dynamic conditions and the states of the 8 cells nearby. Based on this, the processes of vegetation colonization, succession, stress death, annual death and expansion are defined by the pressure function method.

[0104] Planting process: Planting refers to the process in which a grid unit changes from a bare beach state to any vegetation state (i.e., 0 changes to i). The probability of planting vegetation type i is P i est The calculation method is as follows:

[0105]

[0106] Where: P i est is the basic probability of colonization, S k is a function of pressure.

[0107] Succession process: Vegetation succession means that vegetation is replaced from one vegetation type to another, and the grid unit state is transformed from i to j (i≠j≠0). The calculation method is:

[0108]

[0109] Where: The base probability of vegetation succession from species i to species j set by the user.

[0110] Pressure death process, pressure death refers to the transformation of any vegetation species state i to beach state 0 due to environmental pressure, and death is triggered by exceeding the threshold of the characteristic values ​​of the three stress factors. The calculation method is as follows:

[0111]

[0112] Annual death process, annual death refers to the transformation of any vegetation type state i to bare beach state 0 due to the natural cycle of annual vegetation. Annual vegetation such as Suaeda salsa needs to adopt annual death. The calculation method of the annual death probability of vegetation type i is as follows:

[0113]

[0114] Where: Pi ann The annual basic probability of vegetation death set by the user.

[0115] Vegetation expansion is different from other processes. Other processes directly obtain the state probability through the pressure function, while expansion is a process that requires multiple iterative calculations in an ecological process. For example, if the vegetation grid size is Δx and the annual expansion rate of a vegetation is R i , usually R i >Δx,P i exp The following relationship must be satisfied:

[0116]

[0117] Where: N is the number of iterations, R i is the average annual expansion distance of vegetation i, N needs to be determined through iteration, and then the corresponding

[0118]

[0119] Where: N neighbour is the number of vegetation i in the neighboring cells of the grid cell, with a maximum of 8. The more the number of surrounding vegetation cells, the greater the probability that the current grid will be expanded and occupied.

[0120] Pressure function. This embodiment includes three types of pressure functions. The first type is a constant type. When the pressure value s satisfies s min <s<s max When S k =1, otherwise S k =0;

[0121] The second type is linear, when s<s min When S k = 0, when s min <s<s max When S k =(ss min ) / (s max -s min ), when s>s max When S k =1;

[0122] The third type is parabolic, when s<s min or s>s max When S k = 0, when s min <s<s max hour, where s min and maxThey represent the minimum and maximum values ​​of the pressure range when different pressure functions are in effect. The three pressure functions are suitable for different vegetation evolution processes.

[0123] The simulation of vegetation status is completed by performing five processes in sequence. The pressure function and process function are constructed by creating new functions through Python's numpy library. The data transmission and state change are based on variables in the format of numpy.array, and the final vegetation status simulation results are saved as npy files.

[0124] Step six: The vegetation state distribution is transferred from the vegetation dynamics model to the water-sediment dynamics model.

[0125] (1) Data transmission. The data transmission in this step is more complicated. Unlike step 4, which relies only on simple interpolation, it relies on weighted average to complete data transmission. It is necessary to identify the number of vegetation grid cells corresponding to each water and sand dynamics grid cell. The identification method is to call Python's geovoronoi and shapely libraries, and find the corresponding relationship between the water and sand dynamics grid and the vegetation grid through the Voronoi diagram and RTree algorithm. The specific steps are as follows: Figure 2 As shown, it is necessary to use the Point class of shapely.geometry to instantiate the water and sand dynamics grid cell points, use the coords_to_points function of geovoronoi to make the location of the vegetation grid points recognizable by the shapely library, and use the STRtree function of shapely.strtree to build the RTree index. Figure 2 (a) is a schematic diagram of determining the vegetation model calculation area, (b) is a schematic diagram of the unstructured grid nodes in the selected area, (c) is a schematic diagram of the overlapping grid with a mask, (d) is a schematic diagram of calculating the Voronoi of each unstructured node, and (e) is a schematic diagram of finding the corresponding structured grid unit through the RTree algorithm. The vegetation distribution data is transmitted from the grid of a smaller spatial scale (vegetation grid) to the grid of a larger spatial scale (water and sand dynamics grid) by weighted averaging, and the vegetation resistance parameters in each water and sand dynamics grid unit are calculated. The purpose of data transmission is to weight the vegetation information stored in the vegetation grid unit, including height, root density and root diameter, to the corresponding water and sand dynamics grid unit, and the weight is the area of ​​each vegetation grid unit.

[0126] (2) File writing. The vegetation information recorded in the water and sediment dynamics grid cells is written into two control files, arl and ttd, which can be recognized by the Trachytopes module of Delft3D-FM. The arl file records the center coordinates and vegetation coverage ratio of each water and sediment dynamics grid cell boundary (edge) considered for vegetation, while the ttd file records the vegetation height h of these cell boundaries. v , the product n of the root density m and the root diameter D (n = mD) and the vegetation resistance formula used. It is recommended to use the Baptist formula. In the next cycle, Delft3D-FM can automatically read the arl and ttd files to complete the coupling of the water and sediment dynamics model and the vegetation dynamics model.

[0127] Step 7: Long-term simulation and post-processing.

[0128] (1) Long-term simulation. In steps 3 to 6, a cycle between different modules of the model was completed, simulating the biogeomorphology for one year. Through several cycles, the biogeomorphology simulation for several years can be completed. Figure 3 The model grid distribution and topographic map in the embodiment are shown, where (a) is the model grid and the initial topographic map, (b) is the dam with two openings + excavated tidal channels, and (c) is the dam with three openings + excavated tidal channels. These correspond to the six working conditions in Table 2. The vegetation in these six working conditions only includes Spartina alterniflora.

[0129] Table 2 Model working condition settings

[0130] Is vegetation considered? Is there a slot? Is there a dike and an opening? Number of openings Condition 1 no no no — Condition 2 yes no no — Condition 3 yes no yes 2 Condition 4 yes yes yes 2 Condition 5 yes no yes 3 Condition 6 yes yes yes 3

[0131] (2) Simulation results display. The terrain data in the Delft3D-FM result file is read through Python's netCDF4 library, that is, the mesh2d_mor_bl variable is read through the load_variable function. The vegetation distribution data saved in the vegetation dynamics model in npy format is directly read through numpy's numpy.load function. The terrain map and vegetation distribution map are drawn respectively through the tricontourf and pcolormesh functions of the matplotlib.pyplot library. The results are shown in Figure 2. Figure 4 and Figure 5 As shown. Among them, Figure 4 The changes of salt marsh biogeomorphology in the first, fifth, tenth and twentieth years under the condition of levee removal and the presence of Spartina alterniflora, Suaeda salsa and Phragmites australis. Figure 4 (a) is the evolution diagram for the first year, (b) is the evolution diagram for the fifth year, (c) is the evolution diagram for the tenth year, and (d) is the evolution diagram for the twentieth year, demonstrating the simulation capability of the model under multiple vegetation conditions. Figure 5These are the simulation results of salt marsh biogeomorphology in the 20th year under six working conditions, when only Spartina alterniflora exists, showing the differences in the development of biogeomorphology under different working conditions.

[0132] Although the present invention has been described above with preferred embodiments, it is not intended to limit the present invention. A person skilled in the art of the present invention may make various modifications and improvements without departing from the spirit and scope of the present invention. Therefore, the scope of protection of the present invention shall be determined by the definition of the claims.

Claims

1. A method for predicting biogeomorphic evolution of multi-vegetation salt marshes, characterized in that: The following steps are involved: S1. Taking the topographic data, vegetation data, water-sediment dynamics grid data, closed boundary data, and tidal open boundary data of the target area as input, combined with the preset hydrodynamic parameters and sediment dynamics parameters, a water-sediment dynamics model with the characteristics of three stress factors, namely, flooding time, maximum bottom shear stress, and bottom scouring and silting, as output is constructed; S2. Taking the vegetation expansion process and vegetation extinction process as input and combining the spatial scale of the vegetation grid, a vegetation dynamics model with the distribution of vegetation as output is constructed; the vegetation expansion process includes: seedling colonization, vegetation succession and vegetation lateral expansion, and the vegetation extinction process includes: stress death and annual death; The occurrence probability of vegetation expansion and vegetation extinction is determined by the pressure function S k Calculate, where Seedling colonization refers to the process in which a grid unit changes from a bare beach state to any vegetation state, that is, the probability of colonization of vegetation type i changes from 0 to i. The calculation of is as follows: In the formula, is the basic probability of colonization; Vegetation succession means that vegetation is replaced from one vegetation type to another, and the grid cell state is transformed from i to j, i≠j≠0. The calculation is as follows: Where: is the basic probability of the preset vegetation succession from species i to species j; Stress death refers to the transformation of any vegetation species state i to beach state 0 due to environmental pressure. Death is triggered by the thresholds of the characteristic values ​​of the three stress factors. The probability of death of vegetation species i is The calculation of is as follows: Annual mortality refers to the transformation of any vegetation type state i to bare beach state 0 due to the natural cycle of annual vegetation. The calculation method of the annual mortality probability of vegetation type i is as follows: Where: The basic probability of vegetation death per year set by the user; The horizontal expansion of vegetation requires several iterations in an ecological process. Assume that the vegetation grid size is Δx and the annual expansion rate of a vegetation is R. i , and R i >Δx, The following relationship is satisfied: Where: N is the number of iterations, R i is the average annual expansion distance of vegetation i, and the corresponding As follows: Where: N neighbour is the number of vegetation i in the neighboring cells of the grid cell; S3. Couple the water-sediment dynamics model and the vegetation dynamics model. Specifically, the stress factor characteristics calculated by the water-sediment dynamics model are input into the vegetation dynamics model. The vegetation in the vegetation dynamics model completes the expansion and extinction process according to the characteristic values ​​to obtain the vegetation distribution change. The changed vegetation distribution is then input into the water-sediment dynamics model again to complete a cycle. After a preset number of iterations, the salt marsh biogeomorphological evolution results are obtained.

2. A method for predicting biogeomorphic evolution of multi-vegetation salt marshes according to claim 1, characterized in that: The hydrodynamic parameters include: resistance coefficient, and the sediment dynamic parameters include: geomorphic acceleration factor, critical scour shear stress, and critical sedimentation shear stress.

3. The method for predicting biogeomorphic evolution of multi-vegetation salt marsh according to claim 1, characterized in that: In step S1, the flooding time FLT is calculated as follows: Where, T flood is the time when the bottom is flooded, which is equal to the number of time steps with water depth greater than 0 in the water depth time series output by the model, T sim is the model simulation time, which is equal to the total number of time steps of the model; The maximum bed shear stress MaxBSS is calculated as follows: MaxBSS = max(τ) Where, τ is the time series of bed shear stress output by the model; The BLC of the bottom elevation change is calculated as follows: BLC=h1-h0 Where h1 is the elevation of the bottom bed after the model simulation, and h0 is the initial elevation of the bottom bed.

4. The method for predicting biogeomorphological evolution of multi-vegetation salt marshes according to claim 1, characterized in that: There are three types of pressure functions, as follows: The first type is the constant type. When the pressure value s satisfies s min <s<s max When S k =1, otherwise S k =0; The second type is linear, when s<s min When S k = 0, when s min <s<s max When S k =(ss min ) / (s max -s min ), when s>s max When S k =1; The third type is parabolic. When s<s min or s>s max When S k = 0, when s min <s<s max hour, where s min and max They respectively represent the minimum and maximum values ​​of the pressure value range when different pressure functions are in effect.

5. The method for predicting biogeomorphological evolution of multi-vegetation salt marshes according to claim 1, characterized in that: The water and sediment dynamics grid is a triangular unstructured grid, and the vegetation grid is a square unstructured grid with a smaller spatial scale than the water and sediment dynamics grid. The number of water and sediment dynamics grid cells is smaller than that of the vegetation grid. The characteristic value distribution of vegetation distribution and stress factors is saved in both the water and sediment dynamics grid and the vegetation grid. The data in the two sets of grids need to be transmitted in each cycle to ensure effective synchronization and storage of data in the two sets of grids.

6. The method for predicting biogeomorphological evolution of multi-vegetation salt marshes according to claim 1, characterized in that: In step S3, one cycle includes the following sub-steps: S3.1, the water-sediment dynamics model runs, the model simulation results are output, the characteristic values ​​of the three stress factors are calculated, and the results are stored in the water-sediment dynamics grid; S3.2, using the interpolation method to convert the characteristic values ​​of the three stress factors stored in the water and sediment dynamics grid into data in the vegetation grid; S3.3, the vegetation dynamics model completes the five processes of vegetation colonization, succession, stress death, annual death and lateral expansion according to the characteristic values ​​of the stress factors stored in the vegetation grid, and stores the changed vegetation distribution results in the vegetation grid; S3.

4. The vegetation distribution stored in the vegetation grid is accumulated into the water-sediment dynamics model using a weighted average method.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images