A storm flood disaster evaluation method and system based on multi-scale precipitation data

Abstract

The present application relates to the technical field of storm flood disaster assessment, and specifically provides a storm flood disaster assessment method and system based on multi-scale precipitation data, which comprises the following steps: obtaining a digital elevation model of a target region, calculating a terrain curvature distribution field and an elevation mutation characteristic field, obtaining a terrain vertical gradient zone and arranging a multi-layer ground precipitation observation site group; generating a vertical weight factor according to the altitude distribution of the observation site, combining the radar beam direction with the terrain slope to generate a slope-shielding coefficient; fusing ground-based radar, satellite remote sensing and ground site data, generating a terrain-optimized precipitation field through the vertical weight factor and the slope-shielding coefficient; inputting the precipitation field into a terrain-coupled flood model, combining a water flow path network and a soil infiltration correction mechanism to deduce the spatial distribution of disasters. The present application solves the problem of observation distortion caused by terrain shielding and precipitation vertical differentiation in high mountain and canyon areas, and significantly improves the accuracy and reliability of storm flood disaster assessment.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of storm flood disaster assessment, and particularly relates to a storm flood disaster assessment method and system based on multi-scale precipitation data. BACKGROUND

[0002] Storm flood disaster assessment is a core technical means in the field of disaster prevention and reduction. In complex terrain areas, the lack of true precipitation data can affect the assessment accuracy. The existing technology cannot solve the following problems in high mountain and canyon areas with severe terrain cutting: first, the mountain physically blocks the ground radar beam, causing persistent blind areas in the canyon and steep slope areas; second, the existing technology does not layer according to the terrain gradient, and cannot capture the vertical precipitation mutation, resulting in underestimation of storm intensity at low-altitude sites.

[0003] For this special topography of high mountains and canyons, a storm flood disaster assessment method and system based on multi-scale precipitation data are urgently needed to solve the problem of insufficient precipitation observation accuracy caused by terrain shielding and vertical differentiation. SUMMARY

[0004] (1) Technical problem to be solved

[0005] The purpose of the present application is to provide a storm flood disaster assessment method and system based on multi-scale precipitation data to solve the problem of precipitation observation distortion caused by terrain shielding and vertical differentiation in high mountain and canyon areas.

[0006] (2) Technical scheme

[0007] To achieve the above purpose, on the one hand, the present application provides a storm flood disaster assessment method based on multi-scale precipitation data, which comprises:

[0008] Step S1: Obtain the digital elevation model of the target area and extract the grid point elevation value; calculate the terrain curvature distribution field and the elevation mutation characteristic field according to the grid point elevation value; perform terrain segmentation according to the spatial coupling relationship of the terrain curvature distribution field and the elevation mutation characteristic field, and identify the terrain vertical gradient zone; and arrange a multi-layer ground precipitation observation station group in the terrain vertical gradient zone.

[0009] Step S2: obtaining site scale precipitation observation values and their three-dimensional spatial coordinates through the multi-layer ground precipitation observation site group; generating terrain three-dimensional spatial information according to the digital elevation model and the three-dimensional spatial coordinates to obtain terrain shielding parameters; obtaining ground-based radar monitoring data and satellite remote sensing precipitation data, combining the site scale precipitation observation values to construct a multi-scale precipitation data set; generating a vertical weight factor according to the elevation distribution of the three-dimensional spatial coordinates; generating a slope shielding coefficient according to the geometric relationship between the radar beam direction vector and the digital elevation model slope normal vector; combining the vertical weight factor and the slope shielding coefficient to perform weighted fusion on the multi-scale precipitation data set to generate a terrain optimized precipitation field.

[0010] Step S3: inputting the terrain optimized precipitation field into a terrain coupled flood model to perform flood deduction calculation according to the water flow path network of the digital elevation model and the soil infiltration correction mechanism; outputting a disaster spatial distribution map of the target area.

[0011] Further, the method of calculating the terrain curvature distribution field and the elevation mutation feature field according to the grid point elevation values comprises:

[0012] obtaining the grid point elevation values of the digital elevation model, calculating the terrain curvature values of the grid points through a discrete Laplace operator; and constructing the terrain curvature distribution field by the terrain curvature values of all grid points.

[0013] In a preset analysis window, calculating the terrain structure feature quantity of the center grid point of the window .

[0014] wherein, is the maximum elevation in the window, is the minimum elevation in the window, is the gradient mean value in the direction of the maximum precipitation gradient; marking a continuous region with a terrain curvature value greater than a preset curvature threshold as a candidate gradient zone; and marking a grid point with a terrain structure feature quantity greater than a preset structure feature threshold as an elevation mutation feature point.

[0015] Further, the method of performing terrain segmentation according to the spatial coupling relationship between the terrain curvature distribution field and the elevation mutation feature field to identify terrain vertical gradient zones comprises:

[0016] calculating the curvature gradient value of each grid point according to the terrain curvature distribution field; extracting a connected region with a curvature gradient value greater than a preset candidate gradient threshold and marking it as a candidate gradient zone; and performing spatial overlay analysis on the candidate gradient zone and the set of elevation mutation feature points, and if the density of the elevation mutation feature points in the candidate gradient zone exceeds a preset density threshold, determining it as a terrain vertical gradient zone. ​

[0017] Furthermore, the method of deploying a multi-layer surface precipitation observation station group within the terrain vertical gradient zone includes:

[0018] All grid points within the vertical gradient zone of the terrain are obtained, and an ordered grid point sequence is generated from low to high altitude; the vertical stratification boundary is determined according to the ratio of the cumulative value of the terrain curvature value, and the ordered grid point sequence is divided into L layers.

[0019] The average terrain curvature of each vertical layer is calculated, and the number of sites in each layer is dynamically allocated according to the ratio of the average terrain curvature value to the reference terrain curvature value; if the total number of sites exceeds the preset maximum number of sites, the number of sites in each layer is compressed proportionally.

[0020] Furthermore, the method of generating terrain three-dimensional spatial information based on the digital elevation model and the three-dimensional spatial coordinates to obtain terrain shielding parameters includes:

[0021] Taking the three-dimensional coordinates of the ground observation site P as the reference point, connect the radar station position S and P to generate the beam propagation path ; Extract the area with P as the center and radius on the surface of the digital elevation model The circular analysis area is discretized into radial paths, each path with The angle is ,in is the radar beam half-power angle.

[0022] Check the intersection of each radial path and the digital elevation model surface. If the intersection exists on the propagation path between point P and the radar station position S, and the elevation value of the intersection is greater than the elevation value of the radar beam straight propagation path at that location, it is marked as a blocked path; count the number of blocked paths ; Calculate the terrain shielding parameters ,in, is the standard deviation of elevation within the circular analysis area, is the standard deviation of the benchmark terrain, Terrain complexity adjustment parameter.

[0023] Furthermore, the method of generating a vertical weight factor according to the altitude distribution of the three-dimensional space coordinates; and generating a slope shielding coefficient according to the geometric relationship between the radar beam direction vector and the slope normal vector of the digital elevation model includes:

[0024] Get the measured altitude of the site , the vertical weight factor is obtained by the weight calculation formula ,in, is the base weight, is the vertical attenuation coefficient, The regional base altitude.

[0025] Calculate the terrain tangent plane equation at the site according to the digital elevation model and obtain the normal vector And normalized to a unit vector ,in, is the central differential gradient in the x direction, is the central differential gradient in the y direction.

[0026] Calculate radar beam direction vector ;calculate With unit vector Sine of the angle ; Generate aspect shielding coefficient ,in, is the masking nonlinear adjustment index.

[0027] Furthermore, the method of weightedly fusing the multi-scale precipitation dataset by combining the vertical weight factor and the aspect shielding coefficient to generate a terrain-optimized precipitation field includes:

[0028] Resample the ground-based radar monitoring data and satellite remote sensing precipitation data in the multi-scale precipitation dataset to the target spatial resolution grid to generate a spatially aligned radar precipitation field. and satellite precipitation fields ; Perform inverse distance weighted interpolation on station-scale precipitation observations to generate ground precipitation fields ; According to the vertical weight factor Slope aspect shielding coefficient Calculate terrain physical weight coefficient ; According to the weight relationship Perform fusion calculation to output terrain optimized precipitation field; among them, is the radar reliability coefficient, is the satellite timeliness coefficient.

[0029] Furthermore, the method of inputting the terrain-optimized precipitation field into a terrain-coupled flood model and performing flood deduction calculations based on a water flow path network and a soil infiltration correction mechanism of a digital elevation model includes:

[0030] The surface slope value of each grid cell is calculated based on the grid point elevation value, and the runoff equation is selected according to the comparison result between the surface slope value and the preset slope threshold.

[0031] Obtain soil type spatial distribution data and calculate the infiltration capacity correction coefficient ,in, is the saturated moisture content, is the soil water suction, is the saturated hydraulic conductivity, is the soil volume moisture content, is the cumulative infiltration volume.

[0032] Iterative calculations are performed within a preset time step. Each time step includes a runoff evolution calculation stage and an infiltration correction stage. The runoff evolution calculation stage updates the grid water depth by adopting the motion wave equation or diffusion wave equation. The infiltration correction stage adjusts the infiltration water volume according to the infiltration capacity correction coefficient. When the change in grid water depth between two adjacent iterations is less than the preset tolerance threshold, the calculation is terminated and the final grid water depth is output.

[0033] Furthermore, the method of outputting the disaster spatial distribution map of the target area includes:

[0034] Obtain the grid cell flooding depth output by the terrain-coupled flood model With water flow speed ;when When It is determined to be the second risk level: It is determined to be the third risk level: When the risk level is determined to be the fourth level, is the first water depth threshold, is the second water depth threshold, is the first flow rate threshold, is the second flow rate threshold.

[0035] The risk level determination rules are applied to all grid cells to generate a disaster risk level coding matrix, which is then converted into a disaster risk level map with geographic coordinate references; the disaster risk level map includes four attribute fields: spatial location, inundation depth, water flow velocity, and risk level.

[0036] Based on the same inventive concept, on the other hand, the present invention also provides a rainstorm flood disaster assessment system based on multi-scale precipitation data, the system comprising:

[0037] The terrain perception observation network construction module is used to obtain the digital elevation model of the target area and extract the grid point elevation values; calculate the terrain curvature distribution field and the elevation mutation characteristic field based on the grid point elevation values; segment the terrain based on the spatial coupling relationship between the terrain curvature distribution field and the elevation mutation characteristic field, and identify the terrain vertical gradient belt; and deploy a multi-layer ground precipitation observation station group within the terrain vertical gradient belt.

[0038] The multi-source precipitation physical fusion module is used for obtaining site scale precipitation observation values and three-dimensional space coordinates thereof through the multi-layer ground precipitation observation site group; terrain three-dimensional space information is generated according to the digital elevation model and the three-dimensional space coordinates, and terrain shielding parameters are obtained; ground-based radar monitoring data and satellite remote sensing precipitation data are obtained, and combined with the site scale precipitation observation values, a multi-scale precipitation data set is constructed; a vertical weight factor is generated according to the altitude elevation distribution of the three-dimensional space coordinates; a slope shielding coefficient is generated according to the geometric relationship between the radar beam direction vector and the digital elevation model slope normal vector; the multi-scale precipitation data set is weighted and fused combined with the vertical weight factor and the slope shielding coefficient, and a terrain optimized precipitation field is generated.

[0039] The terrain coupling deduction module is used for inputting the terrain optimized precipitation field into a terrain coupling flood model, performing flood deduction calculation according to the water flow path network of the digital elevation model and the soil infiltration correction mechanism; and outputting a disaster spatial distribution map of the target region.

[0040] (3) Advantageous effects

[0041] Compared with the prior art, the advantageous effects of the present application are:

[0042] 1. The present application accurately identifies the key gradient zone of precipitation vertical mutation by terrain curvature segmentation, and dynamically arranges observation sites in the curvature mutation area, solving the problem of insufficient detection accuracy caused by rough altitude layering in the traditional method.

[0043] 2. The present application establishes a geometric shielding model of radar beam and slope, physically quantifies the shielding effect of mountains on radar signals, significantly eliminates the radar blind area caused by canyon terrain, and realizes the global accurate capture of precipitation data under complex terrain. BRIEF DESCRIPTION OF DRAWINGS

[0044] Figure 1 A rainstorm flood disaster evaluation method flow chart based on multi-scale precipitation data for embodiment 1 of the present application;

[0045] Figure 2 A rainstorm flood disaster evaluation system composition schematic diagram based on multi-scale precipitation data for embodiment 2 of the present application. DETAILED DESCRIPTION

[0046] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0047] Before the examples are given, the application scenarios of the present application concept need to be described. The present application is particularly suitable for high mountain and canyon regions with dramatic terrain and significant vertical precipitation differentiation. These regions are obviously affected by terrain shielding. When the angle between the canyon direction and the radar beam is less than 30 degrees, the proportion of observation blind area is more than 80%. In high mountain and canyon regions with a vertical differentiation rate of precipitation greater than 0.5, the traditional observation network has a large error in disaster assessment due to the failure to arrange stations according to the terrain gradient. The present application uses a vertical gradient zone recognition technology based on terrain curvature segmentation, and uses geometric modeling of the radar beam and the slope surface to eliminate the influence of the radar blind zone in the canyon region. The selected canyon region in the embodiment is a typical scene. The terrain undulation intensity coefficient of the canyon region belongs to a strong undulation region, which verifies the effectiveness of the present application in extreme terrain.

[0048] Embodiment 1: As shown in the following table, the present embodiment provides a heavy rain and flood disaster assessment method based on multi-scale precipitation data, which comprises the following steps: Figure 1

[0049] Step S1: Obtain the digital elevation model of the target region, extract the grid point elevation value, calculate the terrain curvature distribution field and the elevation mutation characteristic field according to the grid point elevation value, perform terrain segmentation according to the spatial coupling relationship of the terrain curvature distribution field and the elevation mutation characteristic field, and identify the terrain vertical gradient zone. A multi-layer ground precipitation observation station group is arranged in the terrain vertical gradient zone.

[0050] Step S2: Obtain the station scale precipitation observation value and its three-dimensional spatial coordinates through the multi-layer ground precipitation observation station group; generate terrain three-dimensional spatial information according to the digital elevation model and the three-dimensional spatial coordinates to obtain terrain shielding parameters; obtain ground-based radar monitoring data and satellite remote sensing precipitation data, combine the station scale precipitation observation value, and construct a multi-scale precipitation data set; generate a vertical weight factor according to the altitude elevation distribution of the three-dimensional spatial coordinates; generate a slope shielding coefficient according to the geometric relationship between the radar beam direction vector and the digital elevation model slope normal vector; combine the vertical weight factor and the slope shielding coefficient to weight and fuse the multi-scale precipitation data set to generate a terrain optimized precipitation field.

[0051] Step S3: Input the terrain optimized precipitation field into a terrain coupled flood model, perform flood deduction calculation according to the water flow path network and soil infiltration correction mechanism of the digital elevation model; and output the disaster spatial distribution map of the target region.

[0052] For example, a certain canyon is taken as an example to obtain a digital elevation model of the target region with a resolution of 30m.

[0053] ​​​​Further, the method for calculating the terrain curvature distribution field and the terrain elevation mutation feature field according to the grid point elevation values comprises:

[0054] Obtaining the grid point elevation values of the digital elevation model, calculating the terrain curvature values of the grid points at the positions of the grid points through a discrete Laplace operator, and constructing the terrain curvature distribution field by using the terrain curvature values of all the grid points.

[0055] In a preset analysis window, calculating the terrain structure feature quantity of the grid point at the center of the window. .

[0056] Among them, is the maximum elevation in the window, is the minimum elevation in the window, is the average gradient along the direction of the maximum precipitation gradient; marking the continuous region with the terrain curvature value greater than the preset curvature threshold as a candidate gradient zone; marking the grid point with the terrain structure feature quantity greater than the preset structure feature threshold as an elevation mutation feature point.

[0057] Exemplarily, a 3*3 grid window is taken , the center point coordinate is , and the elevation value is .

[0058] Calculating the terrain curvature value .

[0059] The elevation range value , the average slope value ; .

[0060] Particularly, the curvature threshold is set to 400, and the structure feature threshold is set to 5.0. In this example, , , the feature point is marked.

[0061] Further, the method for terrain segmentation and identification of the terrain vertical gradient zone according to the spatial coupling relationship between the terrain curvature distribution field and the terrain elevation mutation feature field comprises:

[0062] Calculating the curvature gradient value of each grid point according to the terrain curvature distribution field; extracting the connected region with the curvature gradient value greater than the preset candidate gradient threshold, and marking the connected region as a candidate gradient zone; performing spatial overlay analysis on the candidate gradient zone and the set of elevation mutation feature points, and if the density of the elevation mutation feature points in the candidate gradient zone exceeds the preset density threshold, the terrain vertical gradient zone is determined. ​

[0063] For example, the central difference method calculates the curvature gradient value G = 104.2, and identifies 3 connected regions where G > 100.

[0064] Area of region A , containing 82 mutation points, the density is 16.4 points / m2 , greater than the preset density threshold of 10 points / m2 , retained.

[0065] Further, the method of arranging a plurality of ground precipitation observation sites in the terrain vertical gradient zone comprises:

[0066] All grid points in the terrain vertical gradient zone are obtained, and an ordered grid point sequence is generated from low to high in altitude; the vertical stratification boundary is determined according to the proportion of the cumulative value of the terrain curvature value, and the ordered grid point sequence is divided into L layers.

[0067] The average terrain curvature of each vertical stratification layer is calculated, and the number of sites in each layer is dynamically allocated according to the ratio of the average terrain curvature value to the reference terrain curvature value; if the total number of sites exceeds the preset maximum number of sites, the number of sites in each layer is compressed in proportion.

[0068] For example, 6 grid points in the terrain vertical gradient zone are arranged in ascending order of altitude [2850, 2920, 3150, 3200, 3420, 3520]; the corresponding terrain curvature values are [180, 290, 380, 410, 380, 420], and the total number of layers L = 3, .

[0069] Layer 1 boundary: , ; .

[0070] Layer 2 boundary: , ; .

[0071] Layer 1 average , Layer 2 average , Layer 3 average , the average of the three layers is ; , , . The total number of sites is 31 > 30, and the new .

[0072] Further, the method for generating terrain three-dimensional space information according to the digital elevation model and the three-dimensional space coordinates to obtain a terrain shielding parameter comprises:

[0073] Connecting the radar station position S and P to generate a beam propagation path ; extracting a circular analysis region with P as the center and a radius from the digital elevation model surface; discretizing the analysis region into radial paths, and the angle between each path and is , wherein is the half-power angle of the radar beam.

[0074] Checking the intersection of each radial path with the digital elevation model surface, if the intersection exists on the propagation path between points P and radar station position S, and the elevation value of the intersection is greater than the elevation value of the radar beam straight-line propagation path at this position, it is marked as a shielding path; counting the number of shielding paths ; calculating the terrain shielding parameter , wherein is the standard deviation of the elevation within the circular analysis region, is the standard deviation of the reference terrain, is the terrain complexity adjustment parameter.

[0075] Illustratively, a three-dimensional terrain surface is generated based on a 30m digital elevation model, and a station coordinate is used as a control point; the propagation path SP from the radar station S to the station P is simulated; it is found that the path SP is intercepted by a ridge at the P point 565m, the intersection elevation 3250m is greater than the theoretical elevation 3010m of the straight line SP, and it is marked as a shielding path. Within the radius r=565m, the total path is 360, and the shielding path is 331.

[0076] The regional elevation standard deviation , is a dimensionless coefficient calibrated by radar echo experiments, is the standard deviation of the reference terrain, and the terrain shielding parameter .

[0077] Further, the method for generating a vertical weight factor according to the elevation distribution of the three-dimensional space coordinates; and generating a slope shielding coefficient according to the geometric relationship between the radar beam direction vector and the digital elevation model slope normal vector comprises:

[0078] Obtaining the measured elevation of the station , and obtaining the vertical weight factor by a weight calculation formula, wherein is the reference weight, is the vertical attenuation coefficient, is the regional reference elevation.

[0079] The terrain tangent plane equation at the site is calculated according to the digital elevation model, and the normal vector is obtained and normalized to a unit vector , wherein, is the x-direction central difference gradient, is the y-direction central difference gradient.

[0080] The radar beam direction vector is calculated; The sine value of the included angle between the unit vector is calculated ; The slope shielding coefficient is generated, wherein, is the shielding non-linear adjustment index.

[0081] Exemplarily, the regional reference elevation is calculated by calculating the arithmetic mean of the digital elevation model , the elevation of a certain site is , and the absolute height difference is .

[0082] The reference weight is set, the vertical attenuation coefficient is set, and the vertical weight factor .

[0083] , wherein the vertical attenuation coefficient is the correction strength of the elevation deviation on the weight of the precipitation observation.

[0084] , The terrain tangent plane normal vector is normalized , the beam direction vector is calculated, and the included angle sine value , .

[0085] The shielding non-linear adjustment index is calibrated through radar echo attenuation experiments, and the value range is 1.0~1.5, and the preferred value in this scene is ; The slope shielding coefficient is obtained.

[0086] Further, the method of combining the vertical weight factor and the slope shielding coefficient to weight and fuse the multi-scale precipitation data set to generate a terrain-optimized precipitation field comprises:

[0087] Resample ground-based radar monitoring data in the multi-scale precipitation dataset and satellite remote sensing precipitation data to a target spatial resolution grid to generate spatially aligned radar precipitation fields and satellite precipitation fields and satellite precipitation fields ; inverse distance weighted interpolation of site-scale precipitation observations to generate a ground precipitation field ; calculation of terrain physical weight coefficients according to vertical weight factors and slope shading coefficients ; fusion calculation according to weight relationship formula to output terrain-optimized precipitation fields; wherein, is a radar reliability coefficient, is a satellite timeliness coefficient.

[0088] For example, radar and satellite data are resampled to a 1km grid, and site data is used to generate a ground precipitation field by inverse distance weighted interpolation.

[0089] Set the radar reliability coefficient when the real-time radar signal-to-noise ratio is greater than 35dB , and set the satellite timeliness coefficient when the GPM data delay is less than 30 minutes , and the terrain physical weight coefficient ;

[0090] Radar precipitation value , satellite precipitation value , ground interpolated precipitation value ; then ; output terrain-optimized precipitation fields.

[0091] Further, the method of inputting the terrain-optimized precipitation field into a terrain-coupled flood model to perform flood deduction calculation according to the water flow path network of the digital elevation model and the soil infiltration correction mechanism comprises:

[0092] Calculate the surface slope value of each grid unit based on the elevation value of the grid point, and select the runoff equation according to the comparison result of the surface slope value and the preset slope threshold.

[0093] Obtain soil type spatial distribution data and calculate the infiltration capacity correction coefficient , wherein, is the saturated water content, is the soil water suction, is the saturated hydraulic conductivity, is the soil volume water content, is the cumulative infiltration amount.

[0094] ​The iterative calculation is performed in preset time steps, each time step including a runoff evolution calculation phase and a infiltration correction phase, wherein the runoff evolution calculation phase updates the grid water depth by using a motion wave equation or a diffusion wave equation, and the infiltration correction phase adjusts the infiltration water amount according to the infiltration capacity correction coefficient; the calculation is terminated and the final grid water depth is output when the grid water depth variation of adjacent two iterations is less than a preset tolerance threshold.

[0095] Exemplarily, a precipitation field is input into the flood model, the iterative calculation is performed, and the result is output.

[0096] The grid slope is calculated The motion wave equation is used:

[0097] .

[0098] The soil type is sandy loam, which is determined according to the USDA soil texture classification, the water suction of sandy soil , the saturated water content , the real-time water content , the saturated water content , and then:

[0099] The iteration is solved, and the infiltration amount decreases with the increase of F.

[0100] The tolerance threshold is set as , and the iteration content is as follows:

[0101]

[0102] The grid water depth after termination of the iteration The water flow velocity is calculated by the motion wave equation .

[0103] Further, the method of outputting the disaster spatial distribution map of the target area comprises:

[0104] The grid cell inundation water depth output by the terrain-coupled flood model is obtained and the water flow velocity ; when , it is determined as the first risk level; when , it is determined as the second risk level; when , it is determined as the third risk level; and when , it is determined as the fourth risk level; wherein, is the first water depth threshold, is the second water depth threshold, is the first flow velocity threshold, is the second flow velocity threshold.

[0105] The risk level determination rule is applied to all grid cells to generate a disaster risk level coding matrix, and the coding matrix is converted into a disaster risk level map with geographic coordinate reference; the disaster risk level map includes four attribute fields of spatial position, inundation depth, water flow velocity and risk level.

[0106] For example, a first water depth threshold , a second water depth threshold , a first flow velocity threshold , and a second flow velocity threshold .

[0107] And , the third risk level is determined.

[0108] The output comprehensive evaluation: [longitude, latitude, , , third risk level].

[0109] Embodiment 2: Based on the same inventive concept, as shown in Figure 2 , this embodiment also provides a rainstorm flood disaster evaluation system based on multi-scale precipitation data, which comprises:

[0110] A terrain-aware observation network construction module is configured to obtain a digital elevation model of a target area and extract grid point elevation values; a terrain curvature distribution field and an elevation mutation feature field are calculated based on the grid point elevation values; terrain segmentation is performed based on the spatial coupling relationship between the terrain curvature distribution field and the elevation mutation feature field, and a terrain vertical gradient zone is identified; and a plurality of multi-layer ground precipitation observation site groups are arranged in the terrain vertical gradient zone.

[0111] A multi-source precipitation physical fusion module is configured to obtain site-scale precipitation observation values and their three-dimensional spatial coordinates through the multi-layer ground precipitation observation site groups; a terrain three-dimensional spatial information is generated based on the digital elevation model and the three-dimensional spatial coordinates to obtain a terrain shielding parameter; ground-based radar monitoring data and satellite remote sensing precipitation data are obtained, and combined with the site-scale precipitation observation values, a multi-scale precipitation data set is constructed; a vertical weight factor is generated based on the altitude elevation distribution of the three-dimensional spatial coordinates; a slope shielding coefficient is generated based on the geometric relationship between the radar beam direction vector and the digital elevation model slope normal vector; and the multi-scale precipitation data set is weighted and fused by combining the vertical weight factor and the slope shielding coefficient to generate a terrain-optimized precipitation field.

[0112] A terrain coupling deduction module is configured to input the terrain-optimized precipitation field into a terrain coupling flood model, perform flood deduction calculation based on the water flow path network of the digital elevation model and the soil infiltration correction mechanism, and output a disaster spatial distribution map of the target area.

[0113] It should be noted that, as to the system in the above-mentioned embodiments, the specific manner in which each module performs an operation has been described in detail in the embodiments related to the method, and will not be described in detail here.

[0114] Finally, it should be noted that although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions recorded in the foregoing embodiments, or equivalently replace some of the technical features, as long as they are within the spirit and principles of the present application. Any modification, equivalent replacement, improvement, etc. made shall be included in the protection scope of the present application.

Claims

1. A method for storm flood disaster assessment based on multi-scale precipitation data, characterized in that, The method comprises: Obtaining a digital elevation model of a target area, extracting grid point elevation values; calculating a terrain curvature distribution field and an elevation mutation feature field from the grid point elevation values; performing terrain segmentation according to the spatial coupling relationship of the terrain curvature distribution field and the elevation mutation feature field, and identifying a terrain vertical gradient zone; and arranging a multi-layer ground precipitation observation site group in the terrain vertical gradient zone; Obtaining site scale precipitation observation values and three-dimensional spatial coordinates thereof through the multi-layer ground precipitation observation site group; generating terrain three-dimensional spatial information according to the digital elevation model and the three-dimensional spatial coordinates, and obtaining terrain shielding parameters; obtaining ground-based radar monitoring data and satellite remote sensing precipitation data, combining the site scale precipitation observation values, and constructing a multi-scale precipitation data set; generating a vertical weight factor according to the elevation distribution of the three-dimensional spatial coordinates; generating a slope shielding coefficient according to the geometric relationship between the radar beam direction vector and the digital elevation model slope normal vector; and performing weighted fusion on the multi-scale precipitation data set by combining the vertical weight factor and the slope shielding coefficient, and generating a terrain optimized precipitation field; Inputting the terrain optimized precipitation field into a terrain coupled flood model, performing flood deduction calculation according to the water flow path network of the digital elevation model and the soil infiltration correction mechanism; and outputting a disaster spatial distribution map of the target area. 2.The storm flood disaster evaluation method based on multi-scale precipitation data according to claim 1, wherein, The method for calculating a terrain curvature distribution field and an elevation mutation feature field from the grid point elevation values comprises: obtaining the grid point elevation values of the digital elevation model, calculating the terrain curvature values of the grid points by a discrete Laplace operator ; forming the terrain curvature distribution field by the terrain curvature values of all the grid points; In a preset Within the analysis window, calculate the terrain structure feature quantity of the center grid point of the window ; wherein, is the maximum value of the elevation within the window, is the minimum value of the elevation within the window, is the average value of the gradient along the direction of the maximum precipitation gradient; the continuous region with the terrain curvature value greater than the preset curvature threshold is marked as a candidate gradient zone; and the grid point with the terrain structure feature value greater than the preset structure feature threshold is marked as an elevation mutation feature point. 3.The storm flood disaster evaluation method based on multi-scale precipitation data according to claim 2, characterized in that, The method for performing terrain segmentation according to the spatial coupling relationship of the terrain curvature distribution field and the elevation mutation feature field, and identifying a terrain vertical gradient zone comprises: Calculating the curvature gradient value of each grid point according to the terrain curvature distribution field; extracting a connected region with a curvature gradient value greater than a preset candidate gradient threshold value, and marking it as a candidate gradient zone; and performing spatial overlay analysis on the candidate gradient zone and the elevation mutation feature point set, and if the density of the elevation mutation feature points in the candidate gradient zone exceeds a preset density threshold value, the terrain vertical gradient zone is determined. 4.The storm flood disaster evaluation method based on multi-scale precipitation data according to claim 3, characterized in that, The method for arranging a multi-layer ground precipitation observation site group in the terrain vertical gradient zone comprises: Obtaining all grid points in the terrain vertical gradient zone, generating an ordered grid point sequence from low to high in terms of altitude; determining the vertical stratification boundary according to the proportion of the cumulative value of the terrain curvature value, and dividing the ordered grid point sequence into L layers; Calculating the average terrain curvature of each vertical stratification layer, and dynamically allocating the number of sites in each layer according to the ratio of the average terrain curvature value to the reference terrain curvature value; if the total number of sites exceeds a preset maximum number of sites, the number of sites in each layer is compressed in proportion.

5. The storm flood disaster assessment method based on multi-scale precipitation data according to claim 4, characterized in that, The method for generating terrain three-dimensional spatial information according to the digital elevation model and the three-dimensional spatial coordinates, and obtaining terrain shielding parameters comprises: With the three-dimensional coordinates of the ground observation site P as a reference point, the radar station position S and P are connected to generate a beam propagation path ; a circular analysis area with P as the center and a radius of is extracted on the surface of the digital elevation model; the analysis area is discretized into radial paths; Checking intersection of each radial path with the surface of the digital elevation model, if the intersection exists on the propagation path between point P and radar station position S, and the elevation value of the intersection is greater than the elevation value of the radar beam straight propagation path at the position, mark as a shielding path; count the number of shielding paths ; Calculate the terrain shielding parameter , wherein is the standard deviation of the elevation in the circular analysis area, is the standard deviation of the reference terrain, is the terrain complexity adjustment parameter.

6. The storm flood disaster assessment method based on multi-scale precipitation data according to claim 5, characterized in that, Generating a vertical weight factor according to the elevation distribution of the three-dimensional spatial coordinates; The method for generating a slope shielding coefficient according to the geometric relationship between the radar beam direction vector and the digital elevation model slope normal vector comprises: Obtaining site measured elevation height , a vertical weight factor is obtained by a weight calculation formula , wherein, is a reference weight, is a vertical attenuation coefficient, is a regional reference elevation; The terrain tangent plane equation at the site is calculated according to a digital elevation model to obtain a normal vector and normalized to a unit vector wherein is a central difference gradient in the x direction, is a central difference gradient in the y direction; Calculate radar beam direction vector ;calculate With unit vector Sine of the angle ; Generate slope aspect shielding coefficient ,in, is the masking nonlinear adjustment index.

7. The method of claim 6, wherein the method is characterized by, The method for performing weighted fusion on the multi-scale precipitation data set by combining the vertical weight factor and the slope shielding coefficient, and generating a terrain optimized precipitation field comprises: Resample the ground-based radar monitoring data and satellite remote sensing precipitation data in the multi-scale precipitation dataset to the target spatial resolution grid to generate a spatially aligned radar precipitation field. and satellite precipitation fields ; Generate ground precipitation field by performing inverse distance weighted interpolation on station-scale precipitation observations ; According to the vertical weight factor Slope aspect shielding coefficient Calculate terrain physical weight coefficient ; According to the weight relationship Perform fusion calculation to output terrain optimized precipitation field; among them, is the radar reliability coefficient, is the satellite timeliness coefficient. 8.The storm flood disaster evaluation method based on multi-scale precipitation data according to claim 7, characterized in that, The method for inputting the terrain-optimized precipitation field into the terrain-coupled flood model to perform flood deduction calculation according to the water flow path network of the digital elevation model and the soil infiltration correction mechanism comprises the following steps: Calculate the surface slope value of each grid unit based on the grid point elevation value, and select the runoff equation according to the comparison result of the surface slope value and the preset slope threshold value; Obtain the soil type spatial distribution data, and calculate the infiltration capacity correction coefficient wherein, is the saturated water content, is the soil water suction, is the saturated hydraulic conductivity, is the soil volume water content, is the cumulative infiltration amount; Perform iterative calculation within a preset time step, and each time step includes a runoff evolution calculation phase and an infiltration correction phase. The runoff evolution calculation phase updates the grid water depth by using the motion wave equation or the diffusion wave equation, and the infiltration correction phase adjusts the infiltration water volume according to the infiltration capacity correction coefficient. When the grid water depth change amount of adjacent two iterations is less than a preset tolerance threshold, the calculation is terminated and the final grid water depth is output. 9.The storm flood disaster evaluation method based on multi-scale precipitation data according to claim 8, characterized in that, The method for outputting the disaster spatial distribution map of the target area comprises the following steps: a grid cell inundation water depth obtained from a terrain-coupled flood model output a water flow velocity a first risk level is determined when a second risk level is determined when a third risk level is determined when a fourth risk level is determined when wherein is a first water depth threshold value, is a second water depth threshold value, is a first flow velocity threshold value, is a second flow velocity threshold value. Apply the risk level determination rule to all grid units to generate a disaster risk level coding matrix, and convert the coding matrix into a disaster risk level map with geographic coordinate reference. The disaster risk level map includes four attribute fields of spatial position, flooding depth, water flow velocity and risk level. 10.A rainstorm and flood disaster assessment system based on multi-scale precipitation data, characterized in that, The system comprises: A terrain-aware observation network construction module is configured to obtain a digital elevation model of a target area, extract grid point elevation values, calculate a terrain curvature distribution field and an elevation mutation feature field based on the grid point elevation values, perform terrain segmentation based on the spatial coupling relationship between the terrain curvature distribution field and the elevation mutation feature field, and identify terrain vertical gradient zones. A plurality of multilayer ground precipitation observation site groups are arranged in the terrain vertical gradient zones. A multi-source precipitation physical fusion module is configured to obtain site-scale precipitation observation values and their three-dimensional spatial coordinates through the multilayer ground precipitation observation site groups, generate terrain three-dimensional spatial information based on the digital elevation model and the three-dimensional spatial coordinates to obtain terrain shielding parameters, obtain ground-based radar monitoring data and satellite remote sensing precipitation data, combine the site-scale precipitation observation values to construct a multi-scale precipitation data set, generate a vertical weight factor based on the elevation distribution of the three-dimensional spatial coordinates, generate a slope shielding coefficient based on the geometric relationship between the radar beam direction vector and the digital elevation model slope normal vector, and combine the vertical weight factor and the slope shielding coefficient to perform weighted fusion on the multi-scale precipitation data set to generate a terrain-optimized precipitation field. A terrain-coupled deduction module is configured to input the terrain-optimized precipitation field into a terrain-coupled flood model to perform flood deduction calculation according to the water flow path network of the digital elevation model and the soil infiltration correction mechanism, and output a disaster spatial distribution map of the target area.

Images