A simulation method for multi-source compound disaster inundation risk of complex tide-influenced river network

Abstract

The application discloses a kind of complex tide river network's multi-source compound disaster inundation risk simulation method, contains the construction method of astronomic tide-storm surge-flood-salinity-falling multi-scale dynamic process full coupling model, can analyze complex river network in climate change scenario mountain torrent, storm surge, salinity and rainstorm etc. extreme compound disaster's inundation and salinity upstream risk.Through potential inundation area determination, model grid optimization, numerical format setting and parallel acceleration etc. technical means, on the basis of guaranteeing momentum and material flux conservation in river network, maximum degree optimizes calculation efficiency, can realize in estuary-coastal cross-scale spatial region high-precision simulation of the purpose of giving priority to multi-source disaster inundation and saltwater stratification.Based on the application, the problem that water dynamics, salinity, relatively isolated nearshore complex topography cannot interact in traditional simulation can be solved, technical support is provided for scientific evaluation of estuary area multi-source disaster risk and return period, with outstanding social benefits and disaster prevention and mitigation economic benefits.

Description

Technical Field

[0001] This invention belongs to the field of coastal disaster risk simulation and assessment technology, and in particular relates to a method for simulating the inundation risk of multi-source composite disasters in complex tidal river networks. Background Technology

[0002] Low-lying coastal areas are vulnerable to disasters such as storm surges, and are generally highly susceptible to such hazards. Meanwhile, estuaries are also prone to river flooding. Under the influence of climate change, the risk of complex chain reactions from extreme events such as extreme floods, high-intensity tropical cyclones, storm surges, torrential rains, and saltwater intrusion is increasing, while sea-level rise, land subsidence, and the frequency of high-intensity tropical cyclones are also rising. [1-6] This further exacerbates the risk of inundation in estuaries and coastal areas, as well as the probability of secondary geological disasters. Particularly in complex tidal river networks at estuaries, the combined effects of climate change and human activities have led to a more concentrated trend of uneven riverbed erosion, watershed-induced floods, and urban waterlogging. Therefore, a deeper understanding of the interaction mechanisms of these multi-source composite disasters and the urgent need for disaster risk assessments of their affected areas, impact levels, and corresponding return periods require sophisticated numerical models that fully couple multi-scale dynamic processes. These models will reveal the interaction mechanisms of driving factors in composite disaster processes, the overlapping characteristics of multiple disaster-causing processes, and their response patterns to climate change. Summary of the Invention

[0003] Purpose of the Invention: To address the shortcomings of the aforementioned technologies, the purpose of this invention is to provide a simulation method for the multi-source composite disaster inundation risk in complex tidal river networks. By establishing a multi-scale dynamic process coupling model that considers the convergence of astronomical tides, storm surges, floods, saltwater intrusion, and precipitation in consideration of climate change, this method solves the problem of the relative isolation and lack of interaction between hydrodynamics, salinity, and complex nearshore topography in traditional simulations. This allows for a deeper understanding of the combination, risk, and evolution patterns of composite disasters in complex environments such as complex tidal river networks in estuaries.

[0004] Technical Solution: To achieve the above-mentioned objectives, the present invention employs a method for simulating the multi-source composite disaster inundation risk of complex tidal river networks, comprising the following steps:

[0005] Step 1: Build a grid model for complex tidal river networks;

[0006] Step 2: Establish and set up a fully coupled mathematical model of multi-scale dynamic processes;

[0007] Step 3: Simulate and calculate the dynamic coupling process of multi-source disasters;

[0008] Step 4: Extract information on inundation and saltwater intrusion disasters, and complete the analysis and simulation of inundation risks from multiple sources of compound disasters.

[0009] Furthermore, step 1, concerning the grid model of a complex tidal river network, includes determining the computational domain, mesh generation, data collection, and determining grid elevation values. Step 1 includes:

[0010] Step 1.1: Determine the computational region of the grid model based on the characteristics of the study area;

[0011] Step 1.2: Mesh the computational domain;

[0012] Step 1.3: Calculate the water depth and topographic values ​​of the grid.

[0013] Furthermore, the calculation area in step 1.1 includes tidal river networks, potentially inundated areas, and near-continental shelf waters outside river mouths. Step 1.1 includes:

[0014] Collect topographic, meteorological, and hydrological data of the study area;

[0015] Based on multi-year water level data from hydrological and tidal gauge stations, the location of the tidal zone boundary in the estuary and river network area is determined, and the tidal section must be included within the modeling scope.

[0016] Based on the DEM (Digital Elevation Model) elevation data, historical flood peak data, and historical storm surge data of the study area, potential inundation areas are determined in the tidal river network and estuary regions using contour thresholds. The specific selection of the contour threshold should be greater than the linear superposition of the historical maximum water level height and sea-level rise in the river network and estuary areas, and should consider a margin to cover the nonlinear effects of multiple-source disasters; this margin can be taken as 5 m. This method of determining the potential inundation area effectively covers low-lying coastal areas without increasing the number of invalid computational units by including areas that cannot be inundated. The computational area includes not only the tidal river network and potential inundation areas but also the near-continental shelf waters outside the estuary to ensure that the far-field effects of tropical cyclones on storm surge increases during their migration towards the coast are considered.

[0017] Further, the meshing in step 1.2 includes horizontal meshing and / or vertical meshing, and step 1.2 includes:

[0018] The horizontal meshing employs unstructured triangular and / or quadrilateral grids. To better simulate the exchange of momentum and mass fluxes between adjacent grids in the river channel, the horizontal meshing is partitioned into zones such as the river channel, distributary basins, open sea, and potentially inundated areas. These zones are separated by control lines, and the grid density is controlled by adjusting the node spacing on these control lines. Specifically, the river channel in the river network area uses triangular and / or quadrilateral grids along the river's direction, while distributary basins, open sea, and potentially inundated areas use triangular grids. Furthermore, the grid resolution of each zone must gradually transition to ensure computational stability.

[0019] Vertical meshing uses LSC 2 The hybrid stratification scheme (Localized Sigma Coordinates with Shaved Cells) uses a two-dimensional grid for the shallow upstream river section and a three-dimensional grid for other areas, especially those with significant vertical stratification. This grid partitioning scheme can maximize computational efficiency and improve the stability of the model calculation.

[0020] When there is no need to couple the upstream dynamic process of saltwater intrusion in the simulation of multi-source disasters, horizontal mesh partitioning is sufficient, and the mathematical model can be reduced to a two-dimensional plane mode for calculation.

[0021] Further, step 1.3 includes:

[0022] Submergence simulation requires high-resolution DEM elevation data. Therefore, it is essential to collect all available data, including local high-precision aerial survey water depth data, radar and / or laser ground elevation data, and publicly downloadable global DEM elevation data. Data from different sources and at different resolutions are then fused to obtain a merged water depth or elevation data. This fusion process involves unifying all data to an elevation datum, followed by data cropping, stitching, and other processing using tools such as Python and GIS software. Finally, the data is resampled and fused into a single rasterized water depth and topographic data file.

[0023] Using the fused water depth or elevation data, interpolation is performed at the nodes of the grid model through methods such as bilinear interpolation and nearest neighbor interpolation to obtain the water depth or terrain values ​​at each node of the model grid.

[0024] Furthermore, the multi-scale dynamic processes in step 2 include astronomical tides, storm surges, floods, saltwater intrusion, and precipitation. The fully coupled mathematical model of these multi-scale dynamic processes includes the governing equations, discretization methods, initial and boundary conditions, computational parameters, and related data for solving the multi-scale dynamic processes. The established mathematical model can be used to calculate and solve the complex tidal river network grid model created in step 1. Step 2 includes:

[0025] The SCHISM (Semi-implicit Cross-scale Hydroscience Integrated System Model), based on unstructured mesh finite element and finite volume discretization methods, is used to solve surface water flow under the combined effects of storm surge, flood, and other physical processes. The model employs the Eulerian–Lagrangian algorithm to solve the Navier–Stokes equations with the Boussinesq and hydrostatic assumptions. Simulation of the saltwater intrusion process simultaneously solves the mass transport equations with salinity and temperature as state variables.

[0026] Furthermore, step 2 also includes: using tidal data extracted from the TOPEX / POSEIDON satellite altimeter inversion model TPXO for the open sea boundary of the model; using measured or generalized flow data for the upstream river boundary; and applying meteorological fields such as wind, pressure, and precipitation to the mathematical model through sea surface momentum and flux exchange. Wind and pressure fields during the tropical cyclone's influence on the study area can be constructed using tropical cyclone parameterization models, atmospheric dynamic model simulations, or reanalysis data. The initial salinity and temperature fields, and the temporal profile distribution at the open sea boundary, can be provided using observational data or numerical products from the global temperature and salinity model.

[0027] Furthermore, to better simulate the flooding process, the SCHISM mode in step 2 employs a wet-dry algorithm that tracks the interface. By setting a very small water depth threshold, computational units are classified as wet: a unit is considered wet when all nodes and sides are wet, and dry when any node or side becomes dry. These computational units are the grid cells in step 1.2. The discretization scheme for the transport equations uses a second-order precision TVD (Total Variation Denoising) scheme to ensure the conservation of mass flux.

[0028] Before conducting simulation calculations, numerical models need to be developed using measured data such as wind speed and air pressure collected from meteorological stations, and water level, flow velocity, and salinity collected from hydrological stations. This data is then used to refine model parameters (such as the roughness coefficient Manning coefficient and the B parameter in the Holland parametric model, and the maximum wind speed radius). Adjustments and calibrations (etc.) are made to ensure the model is fully validated.

[0029] Furthermore, step 3 includes:

[0030] By setting up historical multi-source complex disaster events such as flood processes, precipitation processes, and tropical cyclone meteorological fields, or the encounter combinations of flood processes with different peak intensities and durations under different return periods, precipitation processes with different intensities and durations, and tropical cyclone processes with different intensities and speeds, the inundation process formed in tidal river networks and estuary areas due to the individual or combined effects of disaster sources such as flash floods, storm surges, and heavy precipitation is simulated.

[0031] When studying the response of multi-source, multi-scale dynamic processes to future climate change, reasonable predictions of sea-level change are necessary. Relative sea-level changes under different emission scenarios can be predicted using the following methods: extrapolation based on observational data of sea surface height (SSH), including satellite altimeters, GPS, and tide gauges; prediction based on the integrated results of the CMIP5 / 6 climate system model; or prediction considering both sources simultaneously. When using prediction methods based on the CMIP5 / 6 climate system model, corrections for land subsidence must be applied. These corrections can be estimated using satellite altimeter and long-term tide gauge data, or determined using glacial isostatic adjustment (GIA) results.

[0032] The simulation employs CPU-accelerated parallel computing.

[0033] Furthermore, step 4 includes:

[0034] The mathematical model outputs results such as water level, flow rate, and salinity hour by hour according to the set output time step.

[0035] Based on the water level results at each time point and the topographic elevation information at the grid nodes, the inundation depth and inundation range distribution at each time point within the computational domain can be calculated, and the inundation hydrological process line at each node location can then be obtained.

[0036] Based on this, further calculations and analyses can be performed to obtain information such as the inundation duration, maximum inundation depth (or peak height), and corresponding peak arrival time at each location, i.e., the inundation analysis results. Based on the inundation analysis results, an inundation risk map can be further drawn.

[0037] Based on the salinity results, the planar salinity distribution at each time point and the vertical salinity distribution at different cross-sectional locations can be analyzed. Furthermore, the upstream location of the extracted brine can be further analyzed based on the 0.5 salinity contour lines.

[0038] Beneficial effects: This invention targets complex tidal river networks in estuaries. Through techniques such as determining potential inundation areas, partitioning the model's horizontal and vertical grids, setting numerical formats and parameters, and parallel acceleration, it optimizes model computation efficiency to the greatest extent while ensuring the conservation of momentum and mass fluxes in the river network. This enables high-precision simulation of complex disaster inundation and saline stratification in the estuary-nearshore multi-scale spatial region.

[0039] This invention addresses the problem of relatively isolated interaction between hydrodynamics, salinity, and complex nearshore topography in traditional simulations, providing technical support for the scientific assessment of the occurrence and recurrence period of multi-source complex disasters in estuaries. It offers significant social and economic benefits in disaster prevention and mitigation. Furthermore, by studying the interaction mechanism between multi-source disasters and saltwater intrusion, it reduces the damage of corresponding complex disasters to the ecosystem, thus also demonstrating significant ecological benefits. Attached Figure Description

[0040] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments, and the advantages of the present invention in the above and / or other aspects will become clearer.

[0041] Figure 1 This is a schematic diagram illustrating the relationships between multiple dynamic processes addressed by the simulation method for the inundation risk of complex tidal river networks with multiple sources of combined disasters, as presented in this invention.

[0042] Figure 2 This is a calculation area and a local magnified view of a multi-source composite disaster inundation model for the complex tidal river network in the Pearl River Estuary.

[0043] Figure 3 This is a schematic diagram of the horizontal and vertical mesh subdivision of the local mesh of the model.

[0044] Figure 4 This shows the water depth and topography of the tidal river network area after interpolation.

[0045] Figure 5 Instantaneous wind field of tropical cyclones calculated for the Holland model and its validation.

[0046] Figure 6 This is a simulation of the instantaneous vertical distribution of salinity during the upstream intrusion of saltwater in the Modaomen Channel during Super Typhoon Mangkhut in 2018.

[0047] Figure 7 This is a map showing the maximum inundation depth distribution under different disaster combination scenarios. The left figure shows the simulation results of the June 2005 flood; the right figure shows the storm surge inundation in Hong Kong, China, when a super typhoon traveling in the WNW direction encounters an astronomical high tide and makes landfall at an unfavorable location along the Pearl River Estuary, considering a sea level rise of 1 m. Detailed Implementation

[0048] The following, in conjunction with the accompanying drawings and specific embodiments, further clarifies the implementation scheme of the simulation method for multi-source composite disaster inundation risk in complex tidal river networks proposed in this invention. It should be understood that these embodiments are for illustrative purposes only and are not intended to limit the scope of the invention. After reading this invention, any modifications of the invention in various equivalent forms by those skilled in the art fall within the scope defined by the appended claims.

[0049] like Figure 1 As shown, watershed floods, astronomical tides, storm surges, heavy precipitation, and sea-level rise are all driving factors leading to the inundation risk of complex river networks. These factors differ in both temporal and spatial scales, spatially covering complex surface flows from streams to estuaries and then to the ocean, and temporally spanning processes from hours to centuries. Simultaneously coupling these multi-scale dynamic processes of multiple sources of disasters in mathematical models places high demands on simulation techniques. Furthermore, the further coupling of the saltwater intrusion process requires the model to accurately solve the diffusion processes of mass fluxes (such as salt flux and heat flux), demanding a high degree of conservation of mass fluxes. For simulating the saltwater intrusion process and its vertical salinity stratification (or density stratification), two-dimensional planar models based on shallow water equations with depth integrals cannot compare with fully three-dimensional models in terms of simulation performance and accuracy.

[0050] Although conceptual models of disaster chains and numerous studies on single-hazard events such as storm surges, floods, and geological disasters exist, research on "multi-convergence" complex disasters considering multiple processes such as astronomical tides, storm surges, floods, saltwater intrusion, and precipitation is rare. Currently, inundation simulation mainly employs two methods: one is a surface water flow model based on rasterized DEM elevation data, using the LISFLOOD-FP model as an example. [7,8] For example, the equations to be solved are generally one-dimensional momentum equations based on the quasi-linearization of the Saint–Venant equations:

[0051]

[0052] in, For unit width flow, Because of the water depth, For bed surface elevation, It is the acceleration due to gravity. This is the Manning coefficient. The hydraulic radius is used. The advantage of this type of model is its ease of coupling with distributed hydrological models and integration with GIS (Geographic Information System) systems. It can efficiently simulate flood evolution over large watersheds and facilitate visualization on GIS platforms. However, one of the main reasons for its high efficiency is that its equations are solved based on the water surface elevation difference between adjacent units, without considering convection and diffusion terms. This simplified equation is not suitable for solving problems such as steep terrain slopes and discontinuous water flow. Furthermore, the effective simulation of this type of model heavily relies on the accuracy of DEM elevation data.

[0053] For complex tidal river networks, the connectivity between river morphology control nodes is intricate. Furthermore, due to the tidal influence of the open sea, the water flow in the river network exhibits a periodic reciprocating flow under the combined effects of runoff and tidal currents. Detailed simulation of the dynamic processes requires the ability to reproduce the flow distribution ratio characteristics of different key water-dividing nodes within the river network. The second type of model is based on hydrodynamic simulation, which is the approach adopted in this invention. However, existing simulation methods have seen limited research on multi-source, multi-dynamic process simulations. Generally, hydrodynamic simulation models solve the Navier-Stokes equations based on the Boussinesq assumption and the hydrostatic assumption, which can comprehensively reflect the interactions between different dynamic factors. However, the computational resource requirements are generally higher than the first type of method. For this reason, flood and inundation process simulation studies typically employ two-dimensional planar models. Even so, when it is necessary to consider the multi-source complex disasters and their mutual feedback effects of processes such as basin floods, astronomical tides, storm surges, heavy precipitation, saltwater intrusion and sea-level rise in the simulation, especially when considering the high-precision simulation of inundation simulation and saltwater intrusion process in a large spatial area, the three-dimensional model is superior to the two-dimensional model. This requires balancing the relationship between solving physical processes and computational efficiency, and realizing the multi-dynamic process coupling simulation of complex river network multi-source complex disasters through model construction and optimization.

[0054] Therefore, a simulation method for the multi-source composite disaster inundation risk of complex tidal river networks is needed to solve the above problems.

[0055] References:

[0056] [1] Beckley BD, Callahan PS, Hancock III DW, et al. On the “Cal-Mode” Correction to TOPEX Satellite Altimetry and Its Effect on the Global Mean SeaLevel Time Series. J Geophys Res Oceans. 2017;122:8371–8384.

[0057] [2] Frederikse T, Landerer F, Caron L, et al. The causes of sea-levelrise since 1900. Nature. 2020;584:393–397.

[0058] [3] DeConto RM, Pollard D. Contribution of Antarctica to past andfuture sea-level rise. Nature. 2016;531:591–597.

[0059] [4] Webster PJ, Holland GJ, Curry JA, et al. Changes in TropicalCyclone Number, Duration, and Intensity in a Warming Environment. Science.2005;309:1844–1846.

[0060] [5] Elsner JB, Kossin JP, Jagger TH. The increasing intensity of thestrongest tropical cyclones. Nature. 2008;455:92–95.

[0061] [6] Song J, Klotzbach PJ, Tang J, et al. The increasing variabilityof tropical cyclone lifetime maximum intensity. Scientific Reports. 2018;8:16641.

[0062] [7] Bates PD, De Roo APJ. A simple raster-based model for floodinundation simulation. Journal of Hydrology. 2000;236:54–77.

[0063] [8] Bates PD, Horritt MS, Fewtrell TJ. A simple inertial formulation of the shallow water equations for efficient two-dimensional flood inundationmodelling. Journal of hydrology (Amsterdam). 2010;387:33–45.

[0064] Example:

[0065] This embodiment takes the Pearl River Delta estuary as an example to conduct simulation studies on single or compound disaster sources under different scenarios.

[0066] This application discloses a method for simulating the multi-source composite disaster inundation risk of complex tidal river networks, comprising the following steps:

[0067] Step 1: Build a grid model for the Pearl River Estuary based on the characteristics of the study area;

[0068] Step 1, concerning the grid model of the Pearl River Estuary, includes the following steps:

[0069] Step 1.1: Determine the computational region of the grid model based on the characteristics of the study area;

[0070] The model calculation area includes tidal river network sections, adjacent potential inundation areas, and near-continental shelf waters outside the river mouth.

[0071] Collect topographic, meteorological, and hydrological data of the study area;

[0072] Based on multi-year water level data from hydrological and tidal gauge stations, the location of the tidal zone boundary in the estuary and river network area is determined, and the tidal section must be included within the modeling scope.

[0073] Based on DEM elevation data, historical flood peak data, and historical storm surge data of the study area, a potential inundation area is determined in the tidal river section using a certain contour threshold. The specific selection of the contour threshold should be greater than the linear superposition of the historical maximum water level height and sea-level rise in the river network and estuary, and a margin should be considered to cover the nonlinear effects when multiple disasters occur. This margin can be taken as 5 m, and the contour threshold is generally recommended to be no less than 15-20 m. This method of determining the potential inundation area effectively covers low-lying coastal areas without increasing the number of invalid computational units by including areas that cannot be inundated. The model calculation range includes not only the tidal river network section and adjacent potential inundation areas, but also the near-continental shelf sea area outside the estuary to ensure that the far-field effects of tropical cyclones on storm surge increases during their migration towards the coast are considered. A fully coupled model of multi-source disaster, multi-scale dynamic processes is established for the above calculation area.

[0074] Figure 2 The diagram shows the calculation area determined for the Pearl River Estuary river network. It covers the South China Sea region centered on the Pearl River Estuary (105.2–120.4°E, 14.8–23.6°N), with the upstream boundaries of the tidal river network mainly including Boluo, Gaoyao, Shijiao, Laoyagang, and Shizui. The potential inundation area is defined by the envelope of 20 m contour lines.

[0075] Step 1.2: Mesh the computational domain;

[0076] Mesh generation includes horizontal and / or vertical mesh generation. The horizontal mesh generation of the computational domain uses unstructured triangular and / or quadrilateral meshes. To better simulate the exchange of momentum and mass fluxes between adjacent meshes in the river channel, the horizontal mesh generation is divided into zones such as the river channel, distributary areas, open sea, and potentially inundated areas. These zones are separated by control lines, and the mesh density of different local areas is controlled by adjusting the node spacing on the control lines. Specifically, the river channel in the river network area uses triangular and / or quadrilateral meshes along the river channel direction, while distributary areas, open sea, and potentially inundated areas use triangular meshes. In this embodiment, the mesh is generated using SMS (Surface-water Modeling System) software, and the mesh form for different polygonal regions is achieved by setting patches or paving.

[0077] Vertical meshing uses LSC 2 The hybrid stratification scheme uses 2D meshes for the shallower upstream river sections and 3D meshes for other areas, especially those with significant vertical stratification effects. This meshing scheme can maximize computational efficiency and improve the stability of the model calculation. Figure 3 The diagram shows the local arrangement of the horizontal grid and the vertical stratification at different water depths along the Modaomen waterway.

[0078] When there is no need to couple the upstream dynamic process of saltwater intrusion in the simulation of multi-source disasters, horizontal meshing is sufficient, and the dimensionality can be reduced to use the planar 2D mode of the model for calculation.

[0079] Step 1.3: Calculate the water depth and topographic values ​​of the grid.

[0080] Submergence simulation requires high resolution DEM elevation data. Therefore, it is essential to collect all available data, including local high-precision aerial survey water depth data, radar and / or laser ground elevation data, and publicly downloadable global DEM elevation data. Data from different sources and at different resolutions are then fused to obtain fused water depth or elevation data. This fusion process involves unifying all data at an elevation datum, followed by data trimming and stitching using tools such as Python and GIS software, and then resampling the data to create a single rasterized water depth and topographic data file. Using the fused water depth or elevation data, bilinear interpolation and nearest neighbor interpolation methods are applied to the grid model nodes to obtain the water depth or topographic values ​​at each node of the model grid. In this embodiment, the fused data includes ETOPO1 data (https: / / ngdc.noaa.gov / mgg / global / global.html), nautical charts from the Navigation Assurance Department of the PLA Navy Headquarters, local aerial survey water depth data, and global ASTER GDEM2 data (https: / / lpdaac.usgs.gov / products / astgtmv002 / ). Figure 4 The figure shows the water depth and topographic distribution of each grid node in the river network area after interpolation.

[0081] Step 2: Establish and set up a fully coupled mathematical model of multi-scale dynamic processes;

[0082] Multi-scale dynamic processes include astronomical tides, storm surges, floods, saltwater intrusion, and precipitation. A fully coupled mathematical model of the multi-scale dynamic processes of astronomical tides, storm surges, floods, saltwater intrusion, and precipitation is established, including the governing equations, discretization methods, initial and boundary conditions, computational parameters, and related data for solving the multi-dynamic processes. The established mathematical model can be used to solve the grid model of the complex tidal river network in the Pearl River Estuary created in step 1. Step 2 includes:

[0083] Step 2.1 Governing Equations

[0084] The SCHISM model, based on an unstructured mesh finite element / finite volume discretization method, solves surface water flow under the combined effects of storm surge, flood, and other physical processes. This model employs the Eulerian–Lagrangian algorithm to solve the Navier–Stokes equations with the Boussinesq and hydrostatic assumptions, where the continuity equation is:

[0085] (1)

[0086] The momentum equation is:

[0087] (2)

[0088] Mass transport equations:

[0089] (3)

[0090] Seawater state equation:

[0091] (4)

[0092] in, The three axes that represent spatial coordinates; For time; It is the horizontal velocity vector; The vertical velocity component; The height of the water surface; For water depth; It is the acceleration due to gravity; For the Laplace operator; The concentration of a substance (which can be salinity, temperature, etc.); and These are salinity and temperature, respectively. These are the diffusion and source / sink terms for matter. This refers to hydrostatic pressure. The density of water; It is the molecular dynamic viscosity coefficient; The diffusion coefficient is denoted as . The additional volume force term includes physical processes such as Coriolis force and Earth's potential, as well as terms related to sea surface and bottom forces, and takes the following form:

[0093] (5)

[0094] in, It is a unit vector, directed vertically upwards; The Coriolis force coefficient; Reference density for seawater; Atmospheric pressure; For the Earth's potential, For the effective spherical elastic coefficient; Water level elevation; is the horizontal eddy viscosity coefficient.

[0095] Step 2.2 Construction of the tropical cyclone meteorological field

[0096] The meteorological field of a tropical cyclone includes wind and pressure fields, which can be constructed through parametric models of tropical cyclones, simulations of atmospheric dynamic models, or reanalysis of data. One of the objectives of this case study is to identify the feedback mechanisms between the multi-source risks of astronomical tides, storm surges, floods, saltwater intrusion, and precipitation, and their dynamic processes. To facilitate the construction of complex scenarios involving multiple convergences, this embodiment uses a parametric model to construct the meteorological field accompanying the tropical cyclone. Since the Holland model was proposed in 1980, various models such as the Jelesnianski model, the Emanuel & Rotunno model, the Fujita & Takahashi model, and the Ueno model have emerged and been widely used. Given the excellent performance of the Holland model in reproducing the wind and pressure fields of tropical cyclones, this embodiment adopts this model, with the specific formula as follows:

[0097] (6)

[0098] (7)

[0099] (8)

[0100] in, This is the distance to the center of the tropical cyclone. for Air pressure at the location; The pressure at the center of the cyclone; Background air pressure; For gradient wind speed; Maximum wind speed; radius of maximum wind speed Calculated using the Willoughby & Rahn formula; For Holland B parameters; This refers to the cyclone's movement speed; This is the boundary layer adjustment factor. air density; This is the Coriolis force coefficient.

[0101] Step 2.3 Calculation setup and model parameter calibration

[0102] The open sea boundary of the model uses tidal data extracted from the TPXO model retrieved by the TOPEX / POSEIDON satellite altimeter inversion model, while the upstream river boundary uses measured flow data. Meteorological fields such as wind, air pressure, and precipitation are applied to the mathematical model through sea surface momentum and flux exchange. The initial salinity and temperature fields, as well as their time-varying profile distributions at the open sea boundary, are given using global temperature and salinity numerical products provided by HYCOM. The sea surface boundary conditions are expressed in the following form:

[0103] (9)

[0104] Among them wind stress The calculation uses , The coefficient of friction on the sea surface. The wind speed is 10 meters above the sea surface. The seabed boundary conditions are expressed in the following form:

[0105] (10)

[0106] in The seabed frictional stress is in the form of... , The coefficient of friction on the seabed. This represents the flow velocity at the top of the boundary layer.

[0107] To better simulate the flooding process, a wet-dry algorithm based on interface tracking is adopted. A very small water depth threshold is set to classify computational units: a unit is considered wet when all nodes and sides are wet, and a unit is considered dry when any node or side becomes dry. The discretization scheme of the transport equations adopts a second-order precision TVD scheme to ensure the conservation of mass flux. In this embodiment, the water depth threshold is set to 0.01 m.

[0108] Before conducting simulation calculations, numerical models need to be developed using measured data such as wind speed and air pressure collected from meteorological stations, and water level, flow velocity, and salinity collected from hydrological stations. This data is then used to refine model parameters (such as the roughness coefficient Manning coefficient and the B parameter in the Holland parametric model, and the maximum wind speed radius). Adjustments and calibrations were performed to ensure the model was fully validated. Due to the wide coverage of the river network area and the influence of different factors such as topography, slope, and bottom sediment, the roughness coefficient varies in different river sections. Based on previous research experience and calibration, the Manning coefficient values ​​in this study are as follows: the Manning coefficient for the entire study area is 0.012~0.03, gradually increasing from the open sea to the estuary and then to the upper reaches of the river network; the average value in the Lingdingyang is 0.016; the upper reaches of the Xijiang River in the river network area are 0.028, the upper reaches of the Beijiang River are 0.03, and the average value in the middle reaches is 0.025; the value in the land area is 0.03. Figure 5The figure shows the verification results of the instantaneous wind field and wind speed process of Typhoon Mangkhut in 2018 using the Holland parametric model after calibration.

[0109] Step 3: Simulate and calculate the dynamic coupling process of multi-source disasters;

[0110] By setting up historical multi-source complex disaster events such as flood processes, precipitation processes, and tropical cyclone meteorological fields, or the encounter combinations of flood processes with different peak intensities and durations under different return periods, precipitation processes with different intensities and durations, and tropical cyclone processes with different intensities and speeds, the inundation process formed by the individual or combined effects of disaster sources such as flash floods, storm surges, and heavy precipitation in tidal river networks and estuary areas is calculated.

[0111] When studying the response of multi-source, multi-scale dynamic processes to future climate change, reasonable predictions of sea-level change are necessary. Relative sea-level changes under different emission scenarios can be estimated based on sea-surface height (SSH) observation data, including multi-satellite fusion altimeter, GPS (Global Positioning System), and tide gauge data. Alternatively, they can be based on the integrated results of the CMIP (Coupled Model Intercomparison Project) 5 / 6 climate system models, or a reasonable estimate considering multiple data sources simultaneously. When using prediction methods based on the CMIP 5 / 6 climate system models, corrections for land subsidence are required. These corrections can be estimated using satellite altimeter and long-term tide gauge data, or determined through Glacial Isostatic Adjustment (GIA) results. Reasonable predictions of sea-level rise under climate change scenarios are a crucial component of studying the response of multi-source, multi-scale dynamic processes to future climate change. In this embodiment, the assumptions regarding sea-level rise are based on the integrated results of satellite altimeter and tide gauge observation data with the CMIP 5 climate system model. Specifically, based on sea level height data from AVISO (Archiving, Validation and Interpretation of Satellite Oceanographic data) multi-satellite fusion altimeter and tide gauge station data from 1993 to 2021, statistical methods such as EOF (Empirical Orthogonal Function) decomposition were used to evaluate and screen 15 models that closely match the measured characteristics of sea level height, including ACCESS1-0 (Australian Community Climate and Earth System Simulator coupled model), FGOALS-g2 (Flexible Global Ocean–Atmosphere–Land System Model gridpoint, version 2, Institute of Atmospheric Physics, Chinese Academy of Sciences), and GFDL-ESM2G (Geophysical Fluid Dynamics Laboratory Earth System Model with Generalized Ocean Layer Dynamics).These model ensemble results are used to predict sea level changes under scenarios such as RCP4.5 and RCP8.5. On the other hand, using satellite altimeter and glacial isostatic adjustment (GIA) results, combined with long-term tidal data from representative stations along the Chinese coast, the characteristics of relative sea level changes along the coast and the upward trend over the past 30 years are analyzed, and the distribution of land subsidence rates along the Chinese coast is estimated. Combining the CMIP5 model ensemble results and the estimated land subsidence rates, the distribution of relative sea level rise along the Chinese coast can be predicted, for example, by 2100. According to the estimation results under the RCP8.5 scenario, the relative sea level changes in the waters near the Pearl River Estuary in 2050 and 2100 are 0.4 m and 0.9 m, respectively. The scenario assumptions of sea level rise are applied to the mathematical model using the orographic uplift method.

[0112] Simulations of different specific cases employ CPU-accelerated parallel computing, with corresponding computational tasks submitted on the cluster.

[0113] Step 4: Extract information on inundation and saltwater intrusion disasters, and complete the analysis and simulation of inundation risks from multiple sources of compound disasters.

[0114] The model outputs water level, flow velocity, and salinity results hourly according to a set output time step (1 h). Based on the water level results at each time step and the topographic elevation information at the grid nodes, the inundation depth and inundation range distribution within the computational domain at each time step can be calculated, and the inundation hydrological process line at each node location can be obtained. On this basis, the inundation duration, maximum inundation depth (or peak height), and corresponding peak arrival time at each location can be further calculated. Based on the salinity results, the planar salinity distribution at each time step and the vertical salinity distribution at different cross-sectional locations can be analyzed, and the upstream location of brine can be further analyzed and extracted based on the 0.5 salinity contour line.

[0115] For the specific study of this embodiment, Figure 6 This is a simulation of the instantaneous vertical salinity distribution of saltwater intrusion in the Modaomen Channel during Super Typhoon Mangkhut in 2018. Figure 7 The figures show the simulated maximum inundation depth distribution for different disaster combinations. The left figure shows the maximum inundation depth distribution extracted from the simulation of the June 2005 basin flood; the right figure shows the storm surge inundation in Hong Kong, China, when a super typhoon traveling in the WNW (West-Northwest) direction encounters an astronomical high tide and makes landfall at an unfavorable location along the Pearl River Estuary, considering a sea level rise of 1 m.

[0116] In its specific implementation, this application provides a computer storage medium and a corresponding data processing unit. The computer storage medium is capable of storing a computer program, which, when executed by the data processing unit, can run the invention's content regarding the simulation method for multi-source composite disaster inundation risk in complex tidal river networks, as well as some or all of the steps in various embodiments. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0117] Those skilled in the art will clearly understand that the technical solutions in the embodiments of the present invention can be implemented using computer programs and their corresponding general-purpose hardware platforms. Based on this understanding, the technical solutions in the embodiments of the present invention, or the parts that contribute to the prior art, can be embodied in the form of computer programs, i.e., software products. These computer program software products can be stored in a storage medium and include several instructions to cause a device containing a data processing unit (which may be a personal computer, server, microcontroller, MUU, or network device, etc.) to execute the methods described in various embodiments or certain parts of the embodiments of the present invention.

[0118] This invention provides a method for simulating the inundation risk of multi-source composite disasters in complex tidal river networks. Many methods and approaches exist for implementing this technical solution; the above description is merely a specific embodiment of this invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this invention, and these improvements and modifications should also be considered within the scope of protection of this invention. All components not explicitly stated in this embodiment can be implemented using existing technologies.

Claims

1. A method for simulating the multi-source composite disaster inundation risk of complex tidal river networks, comprising the following steps: Step 1: Build a grid model for complex tidal river networks; Step 2: Establish and set up a fully coupled mathematical model of multi-scale dynamic processes; Step 3: Simulate and calculate the dynamic coupling process of multi-source disasters; Step 4: Extract information on inundation and saltwater intrusion disasters, and complete the analysis and simulation of inundation risks from multiple sources of compound disasters; Step 2, described above, involves multi-scale dynamic processes including astronomical tides, storm surges, floods, saltwater intrusion, and precipitation. The fully coupled mathematical model of these multi-scale dynamic processes includes the governing equations, discretization methods, initial and boundary conditions, computational parameters, and related data for solving the multi-scale dynamic processes. The established mathematical model can solve the complex tidal river network grid model created in Step 1. Step 2 includes: The SCHISM model, based on unstructured mesh finite element and finite volume discretization methods, is used to solve surface water flow under the combined action of multiple physical processes, including storm surge and flood. The finite element and finite volume discretization methods employ the Eulerian–Lagrangian algorithm to solve the Navier–Stokes equations with the Boussinesq assumption and hydrostatic assumption. Simulation of the saltwater intrusion process simultaneously solves the material transport equations with salinity and temperature as state variables. Step 3 includes: By setting up historical multi-source composite disaster events, flood processes, precipitation processes, and tropical cyclone meteorological fields, or the encounter combinations of flood processes with different peak intensities and durations under different return periods, precipitation processes with different intensities and durations, and tropical cyclone processes with different intensities and speeds of movement, the inundation process formed in tidal river networks and / or estuary areas due to the separate or combined effects of flash floods, storm surges, and heavy precipitation disaster sources is simulated. When studying the response of multi-source disasters and multi-scale dynamic processes to future climate change, it is necessary to predict sea level changes. Relative sea level changes under different emission scenarios can be predicted in the following ways: by using observational data of sea surface height, including satellite altimeters, GPS, and tide gauges, to extrapolate the data; by using the integrated results of the CMIP5 / 6 climate system model; or by considering data from both sources simultaneously. When using prediction methods based on the CMIP5 / 6 climate system model, it is necessary to correct for land subsidence. The correction value can be estimated and determined using satellite altimeter and long-term tide gauge data, or determined using glacier isostatic adjustment results.

2. The method for simulating the multi-source composite disaster inundation risk of a complex tidal river network according to claim 1, characterized in that: Step 1, which involves a grid model for a complex tidal river network, includes determining the computational domain, mesh generation, data collection, and determining grid elevation values. Step 1 includes the following steps: Step 1.1: Determine the computational region of the grid model based on the characteristics of the study area; Step 1.2: Mesh the computational domain; Step 1.3: Calculate the water depth and topographic values ​​of the grid.

3. The method for simulating the multi-source composite disaster inundation risk of a complex tidal river network according to claim 2, characterized in that: The calculation area in step 1.1 includes tidal river networks, potentially inundated areas, and near-continental shelf waters outside river mouths; Step 1.1 includes: collecting topographic, meteorological, and hydrological data of the study area; determining the tidal boundary of the estuary and river network area based on multi-year water level data from hydrological and tidal gauge stations, ensuring that the tidal river section is included within the modeling scope; determining potential inundation areas in the tidal river network and estuary area using contour line thresholds based on DEM elevation data, historical flood peak data, and historical storm tide data of the study area; the selected contour line thresholds should be greater than the linear superposition of the historical maximum water level height of the river network and estuary area with the sea level rise, and should consider margins to cover the nonlinear effects when multiple disasters occur.

4. The method for simulating the multi-source composite disaster inundation risk of a complex tidal river network according to claim 3, characterized in that: The meshing in step 1.2 includes horizontal meshing and / or vertical meshing. Step 1.2 includes: horizontal meshing using unstructured triangular and / or quadrilateral meshes, with the horizontal meshing divided into zones according to river channels, diversion and confluence areas, open sea, and potential inundation areas. Each zone is separated by control lines, and the mesh density of different zones is controlled by adjusting the node spacing on the control lines; river channels in the river network area use triangular and / or quadrilateral meshes along the river channel direction, while diversion and confluence areas, open sea, and potential inundation areas use triangular meshes; simultaneously, the mesh resolution of each zone gradually transitions. Vertical meshing uses LSC 2 The hybrid stratification scheme uses a two-dimensional grid for the shallow upstream river section and a three-dimensional grid for other areas where vertical stratification has a significant effect. When there is no need to couple the upstream dynamic process of saltwater intrusion in the simulation of multi-source disasters, horizontal mesh partitioning is sufficient, and dimensionality reduction is performed using the two-dimensional planar mode of the mathematical model.

5. The method for simulating the multi-source composite disaster inundation risk of a complex tidal river network according to claim 4, characterized in that: Step 1.3 includes: Submersion simulation requires high resolution of DEM elevation data. It involves collecting all data, including local high-precision aerial survey water depth data, radar and / or laser ground elevation data, and publicly downloadable global DEM elevation data. Data from different sources and with different resolutions are fused to obtain fused water depth or elevation data. The fusion process includes unifying all data at an elevation datum, then cropping and stitching the data, and finally resampling it into a single rasterized water depth and topographic data file. The fused water depth or elevation data is then interpolated at the nodes of the grid model to obtain the water depth or topographic values ​​at each node.

6. The method for simulating the multi-source composite disaster inundation risk of a complex tidal river network according to claim 5, characterized in that: Step 2 further includes: the open sea boundary of the grid model uses tidal process data extracted from the TOPEX / POSEIDON satellite altimeter inversion model TPXO; the upstream river boundary uses measured or generalized flow process data; meteorological fields, including wind, pressure, and precipitation, are applied to the mathematical model through sea surface momentum and flux exchange; the wind and pressure fields during the tropical cyclone's influence on the study area are constructed using tropical cyclone parameterization models, atmospheric dynamic model simulations, or reanalysis data; the initial salinity and temperature fields, as well as the time-varying profile distribution at the open sea boundary, are given using observational data or global temperature and salinity model numerical products.

7. The method for simulating the multi-source composite disaster inundation risk of a complex tidal river network according to claim 6, characterized in that: In step 2, the SCHISM mode uses a wet / dry algorithm to track the interface. By setting a water depth threshold, the calculation unit is identified as wet when all nodes and sides of a calculation unit are wet, and dry when any node or side becomes dry. The computational unit is the grid cell in step 1.2; the discretization scheme of the transport equation adopts the second-order precision TVD scheme; Before conducting simulation calculations, the mathematical model needs to be adjusted and calibrated using measured data of wind speed, air pressure collected from meteorological stations, and water level, flow velocity, and salinity collected from hydrological stations. These model parameters include the roughness coefficient (Manning coefficient), the B parameter in the Holland parametric model, and the maximum wind speed radius. .

8. The method for simulating the multi-source composite disaster inundation risk of complex tidal river networks according to claim 7, characterized in that: In step 3, the dynamic coupling process of multi-source disasters is simulated using CPU-accelerated parallel computing.

9. The method for simulating the multi-source composite disaster inundation risk of a complex tidal river network according to claim 8, characterized in that: Step 4 includes the following steps: The mathematical model outputs numerical results, including water level, flow velocity and salinity, time by time according to the set output time step; based on the water level results at each time and the topographic elevation information at the grid nodes, the inundation depth and inundation range distribution at each time in the calculation area are calculated, and then the inundation hydrological process line at each node location is obtained. Based on this, the inundation duration, maximum inundation depth or peak height and corresponding peak arrival time at each location are calculated and analyzed; an inundation risk map is drawn. Based on the salinity results, the planar salinity distribution at each time point and the vertical salinity distribution at different cross-sectional locations were analyzed, and the upstream location of the brine was extracted based on the 0.5 salinity contour lines.

Images