Seabed landslide mass stability evaluation method based on multi-source geophysical data

Abstract

The invention relates to the technical field of geological disaster assessment, in particular to a seafloor landslide mass stability assessment method based on multi-source geophysical data, which comprises the following steps: S1, acquiring multi-beam sounding, side-scan sonar, seafloor seismograph and sediment columnar sample data, completing space-time registration, and performing equivalent density conversion on acoustic wave impedance and shear wave velocity to obtain an acoustic wave impedance and shear wave velocity; constructing a three-dimensional grid model; s2, based on the three-dimensional grid model, calculating a curvature mutation index by adopting sliding window Fourier analysis, calculating a pressure gradient in combination with pore water pressure data, and generating a characteristic parameter matrix composed of density, curvature and pressure gradient; and S3, based on the characteristic parameter matrix, calculating shear strain energy and gravitational potential energy superposition field intensity, and generating a landslide mass stability map taking the instability probability as an index. According to the invention, a visual three-dimensional landslide mass stability map is constructed, and high-precision identification and three-dimensional early warning of submarine landslide risks are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of geological disaster assessment, and in particular to a submarine landslide stability assessment method based on multi-source geophysical data. Background Art

[0002] With the continuous advancement of deep-sea resource development and submarine infrastructure construction, submarine landslides have become one of the major geological hazards restricting marine engineering safety and submarine pipeline layout. Once a landslide occurs, it may cause serious consequences such as secondary surges, equipment burial, and communication interruption. Therefore, it is urgent to establish a high-precision landslide stability assessment method based on geophysical observations to achieve early identification of potential slip areas and risk warning.

[0003] Existing submarine landslide stability analysis mostly relies on two-dimensional seismic profiles and limited physical parameter inference. It lacks joint modeling of the three-dimensional structural characteristics of the stratum and hydrodynamic factors, and it is difficult to effectively integrate multi-source data such as multi-beam bathymetry, side-scan sonar, and seafloor seismometers. This leads to coarse granularity and poor adaptability in stability assessment, and is unable to support spatial continuity modeling of landslide bodies and quantitative expression of instability probability. In addition, most existing methods use empirical indicators or static rules, which make it difficult to characterize the dynamic coupling process of shear energy and gravitational potential energy. Summary of the Invention

[0004] The present invention provides a submarine landslide stability assessment method based on multi-source geophysical data, which realizes high-precision quantitative assessment and three-dimensional visual identification of landslide risks, and significantly improves the landslide prediction and disaster prevention and control capabilities in complex environments.

[0005] A method for assessing the stability of a submarine landslide based on multi-source geophysical data comprises the following steps: S1, collects multi-source data including multi-beam bathymetry data, side-scan sonar data, seafloor seismometer data, and sediment core sampling data, and performs spatiotemporal registration on the collected multi-source data. The acoustic impedance and shear wave velocity values ​​are converted to density dimensions through the equivalent dynamic density conversion mechanism, and a three-dimensional grid model with spatial coordinates, density values, and bedding plane dip angle values ​​is generated. S2, based on the three-dimensional grid model, uses the sliding window Fourier curvature analysis method to calculate the bedding surface curvature mutation index of each grid cell, and simultaneously calculates the pore pressure gradient in combination with the pore water pressure monitoring data, and outputs a characteristic parameter matrix consisting of equivalent dynamic density, bedding surface curvature mutation index and pore pressure gradient; S3, based on the characteristic parameter matrix, generates a landslide stability map with the instability probability value as the strength index by calculating the superposition field strength of shear strain energy and gravitational potential energy.

[0006] Optionally, the S1 includes: S11, acquire multi-beam bathymetric data, side-scan sonar data, seafloor seismic data, and sediment core sampling data covering the target submarine landslide area, and pre-process the collected multi-source data, including format conversion, outlier removal, and preliminary coordinate calibration; S12, aligning the pre-processed multi-source data according to a unified time and space reference to construct a unified geographic reference frame. Based on the alignment, the acoustic impedance and shear wave velocity values ​​of each depth unit in the target area are extracted from the seafloor seismometer data and cross-checked with the sediment core sample data; S13, based on seismic inversion and lithologic conversion theory, uses an equivalent dynamic density conversion mechanism to convert acoustic impedance and shear wave velocity values ​​into density dimensions, forming a density attribute dataset for geological modeling. Combined with multi-beam bathymetric data and sediment columnar sampling data, a three-dimensional grid model with spatial coordinates, density values, and bedding plane dip angle values ​​is constructed.

[0007] Optionally, the S11 includes: S111: Deploy a multibeam bathymetry system to acquire multibeam bathymetry data, deploy a side-scan sonar device to acquire side-scan sonar data, set up an ocean bottom seismometer (OBS) to acquire ocean bottom seismic data, and collect sediment core sampling data at points in the target area; S112, converting the collected multi-source data into a unified processing format, converting the multi-beam bathymetric data into an XYZ coordinate point cloud format, converting the side-scan sonar data into a grayscale matrix, converting the seafloor seismometer data into a depth profile reflection coefficient sequence, and converting the sediment columnar sample data into a layered parameter table; S113, detect and eliminate abnormal points in multi-source data, use median filtering algorithm to perform denoising, and set abnormal detection threshold , if satisfied , then the multi-source data is marked as abnormal and removed, where is the i-th observation value in the multi-source data, is the median of multi-source data, is the anomaly detection threshold (set to 2 for multibeam bathymetry data, 30 for sidescan sonar data, 300 for acoustic impedance and 100 for shear wave velocity in seafloor seismometer data, and 200 for sediment core data); S114, based on a unified reference coordinate system (such as WGS-84), the multi-beam bathymetric data, side-scan sonar image positions, seafloor seismometer array sampling positions and columnar sample points are uniformly spatially reprojected, and the acquisition time records are synchronously aligned.

[0008] Optionally, the S12 includes: S121, multi-beam bathymetric data, side-scan sonar data, seafloor seismic data and sediment core data are uniformly registered in space and time. The spatial registration adopts collinearity constraint and coordinate reprojection method. All multi-source data are reprojected to a unified geographic reference frame (WGS-84 + UTM projection). The original spatial coordinates of the observation point are set as , the spatial coordinates after registration are , the registration is achieved through the rigid body transformation matrix T, and the time registration is achieved through the track timestamp synchronization algorithm to correct the device clock offset and use linear time interpolation to convert the timestamps of different sources into Mapping to a standard reference time ; S122, extract the acoustic impedance AI and shear wave velocity corresponding to each depth unit (1 meter thickness) from the seafloor seismometer data in a unified coordinate system ; S123, the layer information and physical property parameters (measured density, particle size) in the sediment column sample data are compared with the extracted AI, Match the corresponding depth and build a verification function , when the depth unit When AI Verified by actual sample measurement, is the density error threshold, and the verification index is constructed through all valid checkpoints .

[0009] Optionally, the S13 includes: S131, based on seismic inversion and lithologic conversion theory, uses dynamic weight coupling to convert acoustic impedance and shear wave velocity into density dimensions to obtain equivalent density values; S132, using multi-beam bathymetric data to obtain the seabed topography point cloud, and through surface fitting and normal vector calculation, obtain the bedding plane dip value of each grid cell; S133, based on a unified geographic reference coordinate system, a three-dimensional grid model is constructed according to a regular voxel division method, where each grid cell includes spatial coordinates, an equivalent density value, and a bedding plane dip value.

[0010] Optionally, the S2 includes: S21, in the 3D grid model, a sliding window is constructed with each grid cell as the center to extract the bedding plane dip value in the neighborhood , and perform Fourier transform to identify the local curvature variation pattern and define the bedding plane curvature mutation index in the area ; S22, pore water pressure monitoring point data arranged in the target area Perform three-dimensional interpolation at each grid center point Calculate the pore water pressure gradient at ; S23, the equivalent density value, bedding surface curvature mutation index and pore water pressure gradient are integrated according to the spatial grid index m to form the characteristic parameter matrix T.

[0011] Optionally, the S21 includes: S211, in the three-dimensional grid model, with each target grid cell index m as the center, construct The sliding window area, where 、 、 are the window widths in the X, Y, and Z directions, respectively. Within this window, the bedding plane dip values ​​of all neighboring units are extracted to construct the dip tensor block. ; S212, tilt angle tensor block Perform a three-dimensional discrete Fourier transform to obtain a frequency domain representation tensor , the frequency domain modulus is ,in, is the complex coefficient of the nth frequency point in the Fourier transform result, is the total prime number of the window; S213, defines the curvature mutation index of the mth grid cell is the mean absolute deviation of the Fourier frequency domain modulus relative to the mean.

[0012] Optionally, the S3 includes: S31, for each three-dimensional grid cell, calculate the shear strain energy based on its equivalent density value, bedding plane inclination value and pore water pressure gradient and gravitational potential energy ; S32, defines the total energy field strength of each grid cell is the shear strain energy and gravitational potential energy The sum of is used to characterize the potential driving force of instability; S33, normalize the total energy field strength and map it to the instability probability value , and the instability probability values ​​of all grid cells Spatial splicing to finally generate a landslide stability map , used for 3D visualization of potential slip areas and their degree of instability.

[0013] Optionally, the S33 includes: S331, the total energy field strength value of all grid cells is normalized using the linear method Perform normalization processing; S332: Input the normalized energy field intensity value into the nonlinear risk response function and use the sigmoid function to calculate the corresponding landslide instability probability value. ; S333, all spatial grid cells Splice according to their three-dimensional spatial positions to generate a landslide stability map , the landslide stability map supports three-dimensional visualization, and the color mapping is based on Gradient rendering enables three-dimensional risk perception and decision-making assistance in potential unstable areas of landslides.

[0014] Beneficial effects of the present invention: The present invention achieves high-precision fusion of multi-source geophysical data by collecting multi-beam bathymetric data, sidescan sonar data, seafloor seismometer data and sediment columnar sampling data, and performing spatiotemporal alignment based on a unified geographic reference framework. Furthermore, combined with the equivalent dynamic density conversion mechanism, a three-dimensional structured grid model with spatial coordinates, density values ​​and bedding surface dip values ​​is constructed.

[0015] The present invention uses the sliding window Fourier curvature analysis method in a three-dimensional grid model to extract the bedding surface curvature mutation index, and combines it with pore water pressure monitoring data to calculate the pore pressure gradient, thereby effectively constructing a multi-dimensional characteristic parameter matrix to characterize the susceptibility of landslides, enhancing the comprehensive expression ability of potential sliding surface discontinuities, stratum disturbance and hydraulic disturbance effects, and improving the physical relevance and explanatory power of landslide formation mechanism modeling.

[0016] The present invention calculates the superimposed field strength of shear strain energy and gravitational potential energy through a characteristic parameter matrix, and uses normalization and sigmoid functions to complete instability probability mapping, generating a three-dimensional landslide stability map with probability values ​​as intensity indicators. This can achieve three-dimensional identification and dynamic perception of high-risk areas, improving the intelligence and refinement of submarine geological disaster warning, project site selection, and risk decision-making. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only for the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 Schematic diagram of the evaluation method flow in an embodiment of the present invention; Figure 2 This is a schematic diagram of the multi-source data registration and 3D mesh model construction process according to an embodiment of the present invention. DETAILED DESCRIPTION

[0019] The present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. Those skilled in the art may also adopt other alternatives to implement some known technologies; and the accompanying drawings are only for more specific description of the embodiments and are not intended to specifically limit the present invention.

[0020] like Figure 1-Figure 2 As shown, a submarine landslide stability assessment method based on multi-source geophysical data includes the following steps: S1, collects multi-source data including multi-beam bathymetry data, side-scan sonar data, seafloor seismometer data, and sediment core sampling data, and performs spatiotemporal registration on the collected multi-source data. The acoustic impedance and shear wave velocity values ​​are converted to density dimensions through the equivalent dynamic density conversion mechanism, and a three-dimensional grid model with spatial coordinates, density values, and bedding plane dip angle values ​​is generated. S2, based on the three-dimensional grid model, uses the sliding window Fourier curvature analysis method to calculate the bedding surface curvature mutation index of each grid cell, and simultaneously calculates the pore pressure gradient in combination with the pore water pressure monitoring data, and outputs a characteristic parameter matrix consisting of equivalent dynamic density, bedding surface curvature mutation index and pore pressure gradient; S3, based on the characteristic parameter matrix, generates a landslide stability map with the instability probability value as the strength index by calculating the superposition field strength of shear strain energy and gravitational potential energy.

[0021] S1 includes: S11, acquire multi-beam bathymetric data, side-scan sonar data, seafloor seismic data, and sediment core sampling data covering the target submarine landslide area, and pre-process the collected multi-source data, including format conversion, outlier removal, and preliminary coordinate calibration; S12, aligning the pre-processed multi-source data according to a unified time and space reference to construct a unified geographic reference frame. Based on the alignment, the acoustic impedance and shear wave velocity values ​​of each depth unit in the target area are extracted from the seafloor seismometer data and cross-checked with the sediment core sample data; S13, based on seismic inversion and lithologic conversion theory, uses an equivalent dynamic density conversion mechanism to convert acoustic impedance and shear wave velocity values ​​into density dimensions, forming a density attribute dataset for geological modeling. Combined with multi-beam bathymetric data and sediment columnar sampling data, a three-dimensional grid model with spatial coordinates, density values, and bedding plane dip angle values ​​is constructed.

[0022] The S11 includes: S111: Deploy a multibeam bathymetry system to acquire multibeam bathymetry data, deploy a side-scan sonar device to acquire side-scan sonar data, set up an ocean bottom seismometer (OBS) to acquire ocean bottom seismic data, and collect sediment core sampling data at points in the target area; S112, converting the collected multi-source data into a unified processing format, converting the multi-beam bathymetric data into an XYZ coordinate point cloud format, converting the side-scan sonar data into a grayscale matrix, converting the seafloor seismometer data into a depth profile reflection coefficient sequence, and converting the sediment columnar sample data into a layered parameter table; S113, detect and eliminate abnormal points in multi-source data, use median filtering algorithm to perform denoising, and set abnormal detection threshold , if satisfied , then the multi-source data is marked as abnormal and removed, where is the i-th observation value in the multi-source data, is the median of multi-source data, is the anomaly detection threshold (set to 2 for multibeam bathymetry data, 30 for sidescan sonar data, 300 for acoustic impedance and 100 for shear wave velocity in seafloor seismometer data, and 200 for sediment core data); S114, based on a unified reference coordinate system (such as WGS-84), the multi-beam bathymetric data, side-scan sonar image positions, seafloor seismometer array sampling positions and columnar sample points are uniformly spatially reprojected, and the acquisition time records are synchronously aligned.

[0023] S12 includes: S121, multi-beam bathymetric data, side-scan sonar data, seafloor seismic data and sediment core data are uniformly registered in space and time. The spatial registration adopts collinearity constraint and coordinate reprojection method. All multi-source data are reprojected to a unified geographic reference frame (WGS-84 + UTM projection). The original spatial coordinates of the observation point are set as , the spatial coordinates after registration are , the registration is achieved through the rigid body transformation matrix T, and the time registration is achieved through the track timestamp synchronization algorithm to correct the device clock offset and use linear time interpolation to convert the timestamps of different sources into Mapping to a standard reference time , expressed as: ; Where T is a homogeneous transformation matrix including rotation and translation; ; in, 、 is the time offset and scaling factor obtained by fitting the synchronization points; S122, extract the acoustic impedance AI and shear wave velocity corresponding to each depth unit (1 meter thickness) from the seafloor seismometer data in a unified coordinate system , expressed as: ; in, is the acoustic impedance value of the kth layer, is the density of the kth layer, is the longitudinal wave velocity of the kth layer; ; in, , b are empirical coefficients obtained by fitting regional geological conditions; S123, the layer information and physical property parameters (measured density, particle size) in the sediment column sample data are compared with the extracted AI, Match the corresponding depth and build a verification function , when the depth unit When AI Verified by actual sample measurement, =100 is the density error threshold, and the verification index is constructed through all valid checkpoints , expressed as: ; in, is the measured density value of the kth depth unit, is the calculated density value of k depth units; ; Where N is the total number of matching layers, is the indicator function.

[0024] S13 includes: S131, based on seismic inversion and lithologic conversion theory, uses dynamic weight coupling to convert acoustic impedance and shear wave velocity into density dimensions, and obtains the equivalent density value, which is expressed as: ; in, is the equivalent density value of the k-th depth unit, is the acoustic impedance of the kth layer, is the longitudinal wave velocity of the kth layer, is the shear wave velocity of the kth layer, 、 、 is the empirical fitting parameter; S132 uses multi-beam bathymetric data to obtain the seabed topography point cloud. Through surface fitting and normal vector calculation, the bedding plane dip value of each grid cell is obtained, which is expressed as: ; in, is the bedding plane dip angle of the kth grid cell, is the normal vector of the kth grid obtained by fitting the bathymetric point cloud, is the vertical upward unit vector, for Length of the module; S133, based on the unified geographic reference coordinate system, a three-dimensional grid model is constructed according to the regular voxel division method. Each grid cell includes spatial coordinates, equivalent density value, and bedding plane dip value, which are expressed as: .

[0025] S2 includes: S21, in the 3D grid model, a sliding window is constructed with each grid cell as the center to extract the bedding plane dip value in the neighborhood , and perform Fourier transform to identify the local curvature variation pattern and define the bedding plane curvature mutation index in the area ; S22, pore water pressure monitoring point data arranged in the target area Perform three-dimensional interpolation at each grid center point Calculate the pore water pressure gradient at , expressed as: ; in, 、 、 are the local pressure change rates in the corresponding directions respectively; ; ; ; in, is the pore water pressure value at a point in three-dimensional space, is the coordinate of the center point of the mth grid cell, 、 、 are the grid spacing in the X, Y, and Z directions respectively; S23, the equivalent density value, bedding surface curvature mutation index and pore water pressure gradient are integrated according to the spatial grid index m to form the characteristic parameter matrix T, which is expressed as: ; Where M is the total number of grid cells.

[0026] S21 includes: S211, in the three-dimensional grid model, with each target grid cell index m as the center, construct The sliding window area, where 、 、 are the window widths in the X, Y, and Z directions, respectively. Within this window, the bedding plane dip values ​​of all neighboring units are extracted to construct the dip tensor block. , expressed as: ; in, is located at the grid coordinates The bedding plane dip value at 、 、 is the half width of the sliding window; S212, tilt angle tensor block Perform a three-dimensional discrete Fourier transform to obtain a frequency domain representation tensor , the frequency domain modulus is ,in, is the complex coefficient of the nth frequency point in the Fourier transform result, is the total prime number of the window; S213, defines the curvature mutation index of the mth grid cell is the mean absolute deviation of the Fourier frequency domain modulus relative to the mean, expressed as: ; in, is the mean of the frequency modulus values ​​in the mth window.

[0027] S3 includes: S31, for each three-dimensional grid cell, calculate the shear strain energy based on its equivalent density value, bedding plane inclination value and pore water pressure gradient and gravitational potential energy , expressed as: ; in, is the shear strain, is the bedding plane dip value, is the shear modulus, is the shear wave velocity of the unit; ; Where g is the acceleration due to gravity, is the depth of the seabed at the center of the unit; S32, defines the total energy field strength of each grid cell is the shear strain energy and gravitational potential energy The sum of is used to characterize the potential driving force of instability and is expressed as: ; in, is the gravitational potential energy weight factor; S33, normalize the total energy field strength and map it to the instability probability value , and the instability probability values ​​of all grid cells Spatial splicing to finally generate a landslide stability map , used for 3D visualization of potential slip areas and their degree of instability.

[0028] S33 includes: S331, the total energy field strength value of all grid cells is normalized using the linear method Normalization is performed to prevent the incomparability of instability probabilities caused by different energy dimensions and absolute values, and it is expressed as: ; in, is the normalized energy field strength value of the mth grid unit, 、 are the minimum and maximum values ​​of the energy field strength of all grids in the entire area; S332: Input the normalized energy field intensity value into the nonlinear risk response function and use the sigmoid function to calculate the corresponding landslide instability probability value. , expressed as: ; in, is the instability probability value of the mth grid unit, is the function steepness adjustment coefficient, is the energy response threshold, when hour , indicating stability, when hour , indicating high risk; Function steepness adjustment coefficient According to the normalized energy field strength distribution The standard deviation setting is expressed as: ; in, is the normalized energy field intensity distribution The standard deviation of is the response sensitivity factor, ranging from 5 to 15; S333, all spatial grid cells Splice according to their three-dimensional spatial positions to generate a landslide stability map , the landslide stability map supports three-dimensional visualization, and the color mapping is based on Gradient rendering can realize three-dimensional risk perception and decision-making assistance in the potential unstable area of ​​the landslide body, which can be expressed as: ; in, It is a stability spectrum function defined in three-dimensional space. Each point represents the instability probability of the corresponding grid unit. The splicing method is based on the voxel grid index. , mapped to the spatial coordinate system .

[0029] The present invention encompasses any alternatives, modifications, equivalents, and solutions that fall within the spirit and scope of the present invention. To provide a thorough understanding of the present invention, specific details are described in detail below in connection with the preferred embodiments of the present invention, but those skilled in the art will be able to fully understand the present invention without these detailed descriptions. Furthermore, to avoid unnecessary confusion regarding the essence of the present invention, well-known methods, processes, procedures, components, and circuits have not been described in detail.

[0030] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as within the scope of protection of the present invention.

Claims

1. A method for assessing submarine landslide stability based on multi-source geophysical data, characterized in that: The following steps are involved: S1, collects multi-source data including multi-beam bathymetry data, side-scan sonar data, seafloor seismometer data, and sediment core sampling data, and performs spatiotemporal registration on the collected multi-source data. The acoustic impedance and shear wave velocity values ​​are converted to density dimensions through the equivalent dynamic density conversion mechanism, and a three-dimensional grid model with spatial coordinates, density values, and bedding plane dip angle values ​​is generated. S2, based on the three-dimensional grid model, uses the sliding window Fourier curvature analysis method to calculate the bedding surface curvature mutation index of each grid cell, and simultaneously calculates the pore pressure gradient in combination with the pore water pressure monitoring data, and outputs a characteristic parameter matrix consisting of equivalent dynamic density, bedding surface curvature mutation index and pore pressure gradient; S3, based on the characteristic parameter matrix, generates a landslide stability map with the instability probability value as the strength index by calculating the superposition field strength of shear strain energy and gravitational potential energy.

2. The method for evaluating submarine landslide stability based on multi-source geophysical data according to claim 1, characterized in that: Said S1 comprises: S11, acquire multi-beam bathymetric data, side-scan sonar data, seafloor seismic data, and sediment core sampling data covering the target submarine landslide area, and pre-process the collected multi-source data, including format conversion, outlier removal, and preliminary coordinate calibration; S12, aligning the pre-processed multi-source data according to a unified time and space reference to construct a unified geographic reference frame. Based on the alignment, the acoustic impedance and shear wave velocity values ​​of each depth unit in the target area are extracted from the seafloor seismometer data and cross-checked with the sediment core sample data; S13, based on seismic inversion and lithologic conversion theory, uses an equivalent dynamic density conversion mechanism to convert acoustic impedance and shear wave velocity values ​​into density dimensions, forming a density attribute dataset for geological modeling. Combined with multi-beam bathymetric data and sediment columnar sampling data, a three-dimensional grid model with spatial coordinates, density values, and bedding plane dip angle values ​​is constructed.

3. The method for evaluating submarine landslide stability based on multi-source geophysical data according to claim 2, characterized in that: The S11 includes: S111: Deploy a multibeam bathymetry system to acquire multibeam bathymetry data, deploy a side-scan sonar device to acquire side-scan sonar data, set up a seafloor seismometer to acquire seafloor seismic data, and collect sediment core sampling data at points in the target area; S112, converting the collected multi-source data into a unified processing format, converting the multi-beam bathymetric data into an XYZ coordinate point cloud format, converting the side-scan sonar data into a grayscale matrix, converting the seafloor seismometer data into a depth profile reflection coefficient sequence, and converting the sediment columnar sample data into a layered parameter table; S113, detect and eliminate abnormal points in multi-source data, use median filtering algorithm to perform denoising, and set abnormal detection threshold , if satisfied , then the multi-source data is marked as abnormal and removed, where is the i-th observation value in the multi-source data, is the median of multi-source data; S114, based on a unified reference coordinate system, the multi-beam bathymetric data, side-scan sonar image positions, seafloor seismometer array sampling positions and columnar sample points are uniformly spatially reprojected, and the acquisition time records are synchronously aligned.

4. The method for evaluating submarine landslide stability based on multi-source geophysical data according to claim 3, characterized in that: The S12 includes: S121, the multi-beam bathymetric data, side-scan sonar data, seafloor seismic data and sediment core sample data are uniformly registered in space and time. The spatial registration adopts the collinearity constraint and coordinate reprojection method. All multi-source data are reprojected to a unified geographic reference frame. The original spatial coordinates of the observation point are set as , the spatial coordinates after registration are , the registration is achieved through the rigid body transformation matrix T, and the time registration is achieved through the track timestamp synchronization algorithm to correct the device clock offset and use linear time interpolation to convert the timestamps of different sources into Mapping to a standard reference time ; S122, extract the acoustic impedance AI and shear wave velocity corresponding to each depth unit from the seafloor seismometer data in a unified coordinate system ; S123, the layer information and physical property parameters in the sediment column sample data are compared with the extracted AI, Match the corresponding depth and build a verification function , when the depth unit When AI Verified by actual sample measurement, is the density error threshold, and the verification index is constructed through all valid checkpoints .

5. The method for evaluating submarine landslide stability based on multi-source geophysical data according to claim 4, characterized in that: The S13 includes: S131, based on seismic inversion and lithologic conversion theory, uses dynamic weight coupling to convert acoustic impedance and shear wave velocity into density dimensions to obtain equivalent density values; S132, using multi-beam bathymetric data to obtain the seabed topography point cloud, and through surface fitting and normal vector calculation, obtain the bedding plane dip value of each grid cell; S133, based on a unified geographic reference coordinate system, a three-dimensional grid model is constructed according to a regular voxel division method, where each grid cell includes spatial coordinates, an equivalent density value, and a bedding plane dip value.

6. The method for assessing submarine landslide stability based on multi-source geophysical data according to claim 5, characterized in that: The S2 includes: S21, in the 3D grid model, a sliding window is constructed with each grid cell as the center to extract the bedding plane dip value in the neighborhood , and perform Fourier transform to identify the local curvature variation pattern and define the bedding plane curvature mutation index in the area ; S22, pore water pressure monitoring point data arranged in the target area Perform three-dimensional interpolation at each grid center point Calculate the pore water pressure gradient at ; S23, the equivalent density value, bedding surface curvature mutation index and pore water pressure gradient are integrated according to the spatial grid index m to form the characteristic parameter matrix T.

7. The method for assessing submarine landslide stability based on multi-source geophysical data according to claim 6, characterized in that: The S21 includes: S211, in the three-dimensional grid model, with each target grid cell index m as the center, construct The sliding window area, where 、 、 are the window widths in the X, Y, and Z directions, respectively. Within this window, the bedding plane dip values ​​of all neighboring units are extracted to construct the dip tensor block. ; S212, tilt angle tensor block Perform a three-dimensional discrete Fourier transform to obtain a frequency domain representation tensor , the frequency domain modulus is ,in, is the complex coefficient of the nth frequency point in the Fourier transform result, is the total prime number of the window; S213, defines the curvature mutation index of the mth grid cell is the mean absolute deviation of the Fourier frequency domain modulus relative to the mean.

8. The method for evaluating submarine landslide stability based on multi-source geophysical data according to claim 7, characterized in that: The S3 includes: S31, for each three-dimensional grid cell, calculate the shear strain energy based on its equivalent density value, bedding plane inclination value and pore water pressure gradient and gravitational potential energy ; S32, defines the total energy field strength of each grid cell is the shear strain energy and gravitational potential energy The sum of is used to characterize the potential driving force of instability; S33, normalize the total energy field strength and map it to the instability probability value , and the instability probability values ​​of all grid cells Spatial splicing to finally generate a landslide stability map , used for 3D visualization of potential slip areas and their degree of instability.

9. The method for assessing submarine landslide stability based on multi-source geophysical data according to claim 8, characterized in that: The S33 includes: S331, the total energy field strength value of all grid cells is normalized using the linear method Perform normalization processing; S332: Input the normalized energy field intensity value into the nonlinear risk response function and use the sigmoid function to calculate the corresponding landslide instability probability value. ; S333, all spatial grid cells Splice according to their three-dimensional spatial positions to generate a landslide stability map , the landslide stability map supports three-dimensional visualization, and the color mapping is based on Gradient rendering enables three-dimensional risk perception and decision-making assistance in potential unstable areas of landslides.

Images