Evaluation method for monitoring pushing type landslide risk by using InSAR deformation

Abstract

The invention discloses an evaluation method for monitoring a pushing type landslide risk by using InSAR deformation, and belongs to the technical field of geological disaster dynamic identification and monitoring. In the method, the thickness of a soil landslide mass is inverted by using a mass conservation equation, and the landslide form is obtained by combining topographic information, so that the cost can be greatly reduced, and the working efficiency can be greatly improved. The method is of great significance to large-range landslide monitoring and identification. And meanwhile, the mass conservation equation can consider the material composition and deformation characteristics of the landslide mass, so that the slip surface thickness can be inverted more accurately, the risk grade rise of the pushing type landslide can be found through formulating an early warning scheme and carrying out real-time monitoring, and corresponding disaster prevention and reduction measures can be taken. The early warning scheme is based on the risk assessment result. The deformation data of the pushing type landslide can be obtained regularly through real-time monitoring, and when the monitoring data shows that the risk level of the pushing type landslide is increased, early warning information can be issued in time to remind related personnel to take disaster prevention and reduction measures, so that the efficiency of disaster prevention and reduction work is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of dynamic identification and monitoring of geological disasters, and specifically relates to an evaluation method for the risk of translational landslides using InSAR deformation monitoring. Background Technique

[0002] A translational landslide, also known as a translational or lateral landslide, is a type of landslide in which the landslide mass moves by translation along an approximately horizontal or gently inclined plane. Its sliding surface is usually parallel or nearly parallel to the terrain surface, and the sliding direction is inconsistent with the slope inclination direction. The formation of translational landslides is mainly related to specific geological structures, topographic conditions, hydrogeological environments, and external triggering factors, and is common in slope areas with weak structural planes or weak interlayers. The sliding mass of translational landslides is mostly soil or loose accumulations, and its sliding mechanism is usually due to the weakening of the support force at the bottom or one side of the slope, resulting in the upper soil mass sliding downward along the weak plane under the action of gravity. Accurately identifying the movement characteristics and deformation patterns of translational landslides is of great significance for geological disaster prevention and control work. For the identification of translational landslides, the existing main technical means include field geological surveys, GNSS monitoring, geophysical exploration, InSAR deformation calculation, etc.

[0003] However, the existing evaluation methods cannot obtain the overall deformation information of the landslide surface from single-point monitoring data. Increasing the monitoring points is less economical. In the process of field geological surveys, high geological experience is required for the participants. In addition, it cannot obtain the deep deformation characteristics and can only be used as an auxiliary verification means for landslide pattern recognition. Summary of the Invention

[0004] The purpose of the present invention is to provide an evaluation method for the risk of translational landslides using InSAR deformation monitoring to solve the above-mentioned problems.

[0005] The technical solution adopted by the present invention is as follows: An evaluation method for the risk of translational landslides using InSAR deformation monitoring, the method comprising the following steps:

[0006] S1: Collect radar remote sensing data covering the study area, including SAR images at different times and different orbits, and obtain high-precision aerial flight data for calculating landslide terrain information, including slope and aspect, to provide basic data for the deformation analysis of translational landslides;

[0007] S2: Preprocess the SAR images, including radiometric correction, geometric correction, and registration, to ensure data quality and provide accurate data for the deformation analysis of translational landslides; calculate the annual average deformation amount in different line-of-sight directions in the landslide area to obtain surface deformation information;

[0008] S3: Using SAR data with different incident angles, combined with landslide topographic information, decompose the line-of-sight displacement into horizontal and vertical displacements, and analyze the distribution characteristics of the horizontal and vertical displacements;

[0009] S4: Perform two-dimensional displacement decomposition, and use the two-dimensional deformation decomposition formula to convert the annual deformation information of different incident angles into the actual two-dimensional deformation field of the ground surface, and obtain the deformation rates in the horizontal and vertical directions;

[0010] S5: Using high-precision aerial flight data, combined with the information of the landslide body thickness, invert the slip surface morphology of the translational landslide. Based on the mass conservation equation, perform the inversion of the slip surface morphology and invert the thickness of the soil landslide body; use the landslide topographic data and elevation information, subtract the landslide body thickness from the current elevation to obtain the landslide morphology; according to the maximum thickness of the landslide and the distribution characteristics of the landslide body thickness on both sides, divide the slip surface into concave and planar types;

[0011] S6: According to the geomorphic distribution characteristics, divide the landslide body into three parts: the landslide rear edge, the landslide front edge, and the landslide body. Comprehensively judge the landslide deformation mode by combining the average horizontal deformation rate and the picture morphology. The identification characteristics of the translational landslide deformation mode are: the picture morphology of the landslide body is planar, along the direction of the maximum slope drop, the change of the average annual horizontal deformation rate is not obvious, and the average annual vertical deformation rate shows a linear decreasing characteristic; the difference between the maximum and minimum values of the average annual horizontal and vertical deformation rates is relatively large, and the maximum vertical deformation is located at the landslide rear edge;

[0012] S7: Establish a risk assessment model to evaluate the potential risk level of the translational landslide;

[0013] S8: According to the risk assessment results, formulate an early warning plan and conduct real-time monitoring, paying attention to the change trend of the horizontal displacement of the translational landslide and the linear decreasing characteristic of the vertical displacement; when the monitoring data shows that the risk level of the translational landslide increases, issue early warning information in a timely manner and take corresponding disaster prevention and mitigation measures, with emphasis on the early warning and protection measures for the landslide rear edge area;

[0014] S9: Continuously optimize the risk assessment model according to the actual monitoring data to improve the accuracy and reliability of the assessment results.

[0015] In a preferred embodiment, in the step S1, during the process of collecting radar remote sensing data covering the study area, it is first necessary to determine the boundary of the study area and obtain the geographical coordinate information of this area; then, contact the satellite remote sensing data provider to obtain the SAR images covering this area;

[0016] Collect SAR images at different time intervals to analyze the variation trend of landslides over time; select SAR images obtained every few months within a year to capture the seasonal variations of landslides; while collecting SAR images, it is also necessary to obtain high-precision aerial flight data; this is achieved by contacting professional aerial survey companies or using drones for aerial photography; the aerial flight data provides high-resolution landslide terrain information, including slope and aspect, which is very important for analyzing the deformation characteristics of translational landslides.

[0017] In a preferred embodiment, in step S2, the processing flow includes:

[0018] Set the terrain factors in the Range direction and the Azimuth direction to 5 and 1 respectively to generate an intensity map reflecting the actual terrain ratio of the study area;

[0019] Select the radar intensity image obtained in autumn and winter as the main image, register the intensities of the remaining images to this main image, and require the registration error to be less than one-quarter of a pixel; using the polynomial coefficients obtained by intensity map registration iteration as the standard, register the original single-look complex raw data to obtain the resampled single-look complex data (rslc); according to the resampled single-look complex data, calculate the resampled intensity map (rmli);

[0020] Set the temporal baseline and the spatial baseline to construct an interferometric combination; to best meet the interferometric conditions, it is necessary to shorten the intervals of the interferometric temporal and spatial baselines, so set the maximum temporal baseline to 24d and the spatial baseline to 100m;

[0021] Perform pixel registration on the rslc that constitutes the interferometric combination, then substitute the pixel values for conjugate multiplication to obtain the interferometric calculation result, and further substitute the terrain parameters of the study area in the radar coordinate format to obtain the differential interferometric phase;

[0022] Use the minimum cost flow method to unwrap the differential interferometric data, and superimpose multiple differential interferometric phases to calculate the phase within the superimposed time period. The superimposition calculation formula is Finally, calculate the phase information as deformation information to obtain the annual average deformation amount of the study area;

[0023] Collect radar data at different incident angles, and calculate according to the above method in turn to obtain the deformation values in different line-of-sight directions covering the same time period.

[0024] In a preferred embodiment, in step S3, the specific method includes: resampling the slope aspect and slope raster files to make the resolutions of different raster data consistent, using the georeferencing tool in Arcgis to georeference the terrain raster data and the deformation raster data with consistent resolutions, and further using the extractbymask tool for double registration verification to ensure that multiple raster files coincide precisely; then defining the maximum slope descent direction as the horizontal direction and the direction perpendicular to the maximum slope descent direction as the vertical direction to establish a coordinate system.

[0025] In a preferred embodiment, in step S4, the annually averaged deformation information at different incident angles obtained by calculation is denoted as dLos1 and dlos2 respectively. Using the raster calculator tool in Arcgis, substitute the dLos1, dlos2, Aspect, and slope raster data into the following formula to calculate the actual two-dimensional deformation field of the surface;

[0026]

[0027] where d H and d V are the surface deformation rates in the horizontal and vertical directions respectively, α represents the angle between the satellite flight direction and the true north direction, β is the angle between the maximum slope descent direction of the slope and the true north direction, and θ represents the satellite incident angle.

[0028] In a preferred embodiment, in step S5, the thickness of the landslide mass of the accumulation landslide tapers off at both side boundaries, and the landslide mass of the soil landslide exhibits uniform rheological characteristics in space. The landslide depth and the vertical deformation obtained by InSAR are solved using the following formula

[0029]

[0030] where h is the landslide thickness, t is the time, f is the rheological coefficient, and Vsurf is the surface deformation rate. Substitute the decomposed two-dimensional displacement field information, and the landslide mass slope thickness information can be obtained using the least squares method.

[0031] In a preferred embodiment, in step S6, during the drawing of the mountain shadow map, use a high-precision surface elevation model to draw the mountain shadow map of the landslide mass; the mountain shadow map can intuitively reflect the terrain undulation and geomorphic features of the landslide mass, which helps to better understand the spatial distribution and deformation characteristics of the landslide mass;

[0032] During the landslide zoning process, the rear edge of the landslide is located at the highest point of the landslide mass, the front edge of the landslide is located at the lowest point of the landslide mass, and the landslide mass is located between the rear edge and the front edge of the landslide;

[0033] Analyze the vertical and horizontal deformation rates of the three parts of the landslide to understand the deformation characteristics and differences in different regions; this helps to more accurately judge the deformation mode and potential risks of the landslide.

[0034] In a preferred embodiment, in step S7, establishing a risk assessment model specifically includes:

[0035] Data preparation: Collect deformation data of translational landslides, including horizontal displacement, vertical displacement, landslide body thickness, and slip surface morphology; in addition, data on the topographic information, geological conditions, and hydrological conditions of the landslide also need to be collected;

[0036] Model selection: Select a risk assessment model based on fuzzy logic. This model divides the risk levels into different fuzzy sets according to different input data and conducts risk assessment through fuzzy inference rules;

[0037] Determination of model parameters: Determine model parameters, including the definition of fuzzy sets, the selection of membership functions, and the formulation of inference rules; use horizontal displacement, vertical displacement, landslide body thickness, and slip surface morphology as input variables and define different fuzzy sets, such as "small", "medium", and "large"; at the same time, according to expert experience and actual situations, determine the membership functions and inference rules of each fuzzy set;

[0038] Model training: Use the collected translational landslide data to train the model to determine the values of model parameters; use the least squares method to minimize the error between the model prediction value and the actual observed value;

[0039] Model evaluation: Evaluate the trained model to verify its accuracy and reliability; use the cross-validation method to divide the data set into a training set and a test set and calculate the prediction accuracy of the model on the test set;

[0040] Risk assessment: Use the trained model to conduct risk assessment on new translational landslides to determine their potential risk levels; input the deformation data of the new translational landslide into the model and judge its potential risk level according to the output result of the model.

[0041] In a preferred embodiment, in step S8, according to the risk assessment result, formulating an early warning plan and conducting real-time monitoring specifically include the following methods:

[0042] Setting of early warning threshold: According to the risk assessment model and the deformation characteristics of translational landslides, set the early warning threshold; when the horizontal displacement change rate exceeds 1 mm / year or the vertical displacement change rate exceeds 0.5 mm / year, trigger an early warning;

[0043] Real-time monitoring: Use InSAR technology for real-time monitoring to regularly obtain deformation data of translational landslides; obtain SAR images once every month and calculate the change rates of horizontal displacement and vertical displacement.

[0044] Early warning information release: When the monitoring data indicates that the risk level of translational landslides increases and reaches the early warning threshold, release early warning information in a timely manner; release early warning information through text messages, emails, and social media channels to remind relevant personnel to take disaster prevention and mitigation measures.

[0045] Disaster prevention and mitigation measures: Take corresponding disaster prevention and mitigation measures according to the early warning information; organize personnel to evacuate from the landslide danger area, set up warning signs, and conduct engineering treatment.

[0046] In a preferred embodiment, in step S9, use field geological drilling or trenching survey methods to conduct on-site investigations on the structure, material composition, and deformation characteristics of the landslide body to verify the accuracy of the identification results.

[0047] In summary, due to the adoption of the above technical solutions, the beneficial effects of the present invention are as follows:

[0048] 1. In the present invention, the two-dimensional deformation decomposition formula is the core of the entire identification process, which can convert InSAR observation data with different incident angles into the actual deformation rates in the horizontal and vertical directions of the ground surface. By introducing terrain information and incident angle parameters, the line-of-sight displacement is converted into horizontal and vertical displacements, thus more accurately reflecting the true deformation characteristics of the landslide. Using the mass conservation equation to invert the thickness of the soil landslide sliding mass and combining with terrain information to obtain the landslide morphology can greatly reduce costs and improve work efficiency. This is of great significance for large-scale landslide monitoring and identification. At the same time, the mass conservation equation can consider the material composition and deformation characteristics of the landslide body, thereby more accurately inverting the thickness of the sliding surface. Inverting the landslide mode by integrating the shape of the sliding surface and the two-dimensional displacement distribution characteristics can more accurately judge the deformation mode of the landslide, thus providing a more reliable basis for landslide early warning and disaster prevention. It can effectively reduce the risk of landslide disasters.

[0049] 2. In the present invention, by formulating an early warning plan and conducting real-time monitoring, it is possible to promptly detect an increase in the risk level of translational landslides and take corresponding disaster prevention and mitigation measures. The early warning plan can set different early warning thresholds according to the risk assessment results, such as yellow warning, orange warning, red warning, etc., so as to promptly issue early warning information. Real-time monitoring can regularly obtain the deformation data of translational landslides, thereby grasping the real-time deformation situation of translational landslides. When the monitoring data indicates an increase in the risk level of translational landslides, early warning information can be promptly issued to remind relevant personnel to take disaster prevention and mitigation measures, such as organizing personnel to evacuate dangerous landslide areas, setting warning signs, and carrying out engineering treatment. This helps to reduce the threat of translational landslides to people's lives and property safety and improve the efficiency of disaster prevention and mitigation work.

[0050] 3. In the present invention, by establishing a risk assessment model, it is possible to effectively evaluate the potential risk level of translational landslides. This model comprehensively considers factors such as the deformation characteristics, slide mass thickness, and slip surface morphology of translational landslides, and can more comprehensively reflect the actual situation of translational landslides. Through the risk assessment model, the risk level of translational landslides can be divided into different levels, such as low risk, medium risk, high risk, etc., thereby providing a scientific basis for landslide prevention and control work. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a schematic diagram of the process principle of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0052] In order to make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0053] Refer to Figure 1 ,

[0054] Embodiment:

[0055] An evaluation method for monitoring the risk of translational landslides using InSAR deformation, the method comprising the following steps:

[0056] S1: Collect radar remote sensing data covering the study area, including SAR images at different times and different orbits, and obtain high-precision aerial flight data for calculating landslide terrain information, such as slope, aspect, etc., to provide basic data for translational landslide deformation analysis;

[0057] S2: Preprocess the SAR images, including radiometric correction, geometric correction, registration, etc., to ensure data quality and provide accurate data for translational landslide deformation analysis. Calculate the annual average deformation amount in different line-of-sight directions of the landslide area to obtain planar deformation information;

[0058] S3: Using SAR data with different incident angles and combining with landslide terrain information, decompose the line-of-sight displacement into horizontal and vertical displacements, and analyze the distribution characteristics of the horizontal and vertical displacements;

[0059] S4: Conduct two-dimensional displacement decomposition, use the two-dimensional deformation decomposition formula to convert the annual deformation information of different incident angles into the actual two-dimensional deformation field of the ground surface, and obtain the deformation rates in the horizontal and vertical directions;

[0060] S5: Using high-precision aerial flight data and combining with the landslide thickness information, invert the slip surface morphology of the translational landslide. Based on the mass conservation equation, conduct the inversion of the slip surface morphology and invert the thickness of the soil landslide mass; Use the landslide terrain data and elevation information, subtract the landslide mass thickness from the current elevation to obtain the landslide morphology; According to the maximum thickness of the landslide and the distribution characteristics of the landslide mass thickness on both sides, divide the slip surface into concave and planar types;

[0061] S6: According to the geomorphic distribution characteristics, divide the landslide body into three parts: the landslide rear edge, the landslide front edge, and the landslide body. Comprehensively judge the landslide deformation mode by combining the average horizontal deformation rate and the picture morphology. The identification characteristics of the translational landslide deformation mode are: the picture morphology of the landslide body is planar, along the direction of the maximum slope drop, the change of the average annual horizontal deformation rate is not obvious, and the average annual vertical deformation rate shows a linear decrease characteristic; The difference between the maximum and minimum values of the average annual horizontal and vertical deformation rates is relatively large, and the maximum vertical deformation is located at the landslide rear edge;

[0062] S7: Establish a risk assessment model to evaluate the potential risk level of the translational landslide;

[0063] S8: According to the risk assessment results, formulate a warning plan and conduct real-time monitoring, pay attention to the change trend of the horizontal displacement of the translational landslide and the linear decrease characteristic of the vertical displacement. When the monitoring data shows that the risk level of the translational landslide increases, issue a warning message in a timely manner and take corresponding disaster prevention and mitigation measures, focusing on the warning and protection measures in the landslide rear edge area;

[0064] S9: Continuously optimize the risk assessment model according to the actual monitoring data to improve the accuracy and reliability of the assessment results.

[0065] In step S1, during the process of collecting radar remote sensing data covering the study area, it is first necessary to determine the boundary of the study area and obtain the geographic coordinate information of this area. Then, it is possible to contact satellite remote sensing data providers, such as the European Space Agency (ESA) or the Canadian Space Agency (CSA), to obtain SAR images covering this area.

[0066] Collect SAR images at different time intervals to analyze the change trend of landslides over time. SAR images can be obtained every few months within a year to capture the seasonal changes of landslides. When collecting SAR images, high-precision aerial flight data also needs to be obtained. This can be achieved by contacting professional aerial survey companies or using drones for aerial photography. Aerial flight data can provide high-resolution landslide terrain information, including slope, aspect, etc., which is very important for analyzing the deformation characteristics of translational landslides.

[0067] In step S2, the processing flow includes:

[0068] Set the terrain factors in the Range direction and Azimuth direction to 5 and 1 respectively to generate an intensity map reflecting the actual terrain ratio of the study area;

[0069] Select the radar intensity image obtained in autumn and winter as the main image, and register the intensities of the remaining images to this main image, requiring the registration error to be less than one-quarter of a pixel. Using the polynomial coefficients obtained by intensity map registration iteration as the standard, register the original single-look complex raw data to obtain the resampled single-look complex data (rslc). Calculate the resampled intensity map (rmli) based on the resampled single-look complex data;

[0070] Set the temporal baseline and spatial baseline to construct an interferometric combination. To best meet the interferometric conditions, the intervals of the interferometric temporal and spatial baselines need to be shortened. Therefore, set the maximum temporal baseline to 24d and the spatial baseline to 100m;

[0071] Perform pixel registration on the rslc that constitutes the interferometric combination, then substitute the pixel values for conjugate multiplication to obtain the interferometric calculation result, and further substitute the terrain parameters of the study area in radar coordinate format to obtain the differential interferometric phase;

[0072] Use the minimum cost flow method to unwrap the differential interferometric data, and superimpose multiple differential interferometric phases to calculate the phase within the superimposed time period. The superimposition calculation formula is Finally, calculate the phase information as deformation information to obtain the average annual deformation of the study area;

[0073] Collect radar data at different incident angles and calculate according to the above method in turn to obtain the deformation values in different line-of-sight directions covering the same time period.

[0074] In step S3, the specific method includes: resampling the slope aspect and slope raster files to make the resolutions of different raster data consistent, using the georeferencing tool in Arcgis to georegister the terrain raster data and the deformation raster data with consistent resolutions, and further using the extractbymask tool for double registration verification to ensure that multiple raster files coincide precisely. After that, define the maximum slope descent direction as the horizontal direction, and the direction perpendicular to the maximum slope descent direction as the vertical direction to establish a coordinate system.

[0075] In step S4, the annual deformation information at different incident angles obtained by calculation is denoted as dLos1 and dlos2 respectively. Using the raster calculator tool in Arcgis, substitute the dLos1, dlos2, Aspect, and slope raster data into the following formula to calculate the actual two-dimensional deformation field of the ground surface;

[0076]

[0077] where d H and d V are the ground surface deformation rates in the horizontal and vertical directions respectively, α represents the angle between the satellite flight direction and the true north direction, β is the angle between the maximum slope descent direction of the slope and the true north direction, and θ represents the satellite incident angle.

[0078] In step S5, the thickness of the sliding mass of the accumulation landslide tapers off at both side boundaries, and the sliding mass of the soil landslide shows uniform rheological characteristics in space. The landslide depth and the vertical deformation obtained by InSAR are solved using the following formula

[0079]

[0080] where h is the landslide thickness, t is the time, f is the rheological coefficient, and Vsurf is the ground surface deformation rate. Substitute the decomposed two-dimensional displacement field information, and the information on the thickness of the landslide slope can be obtained using the least squares method.

[0081] In step S6, during the process of drawing the mountain shadow map, use a high-precision ground surface elevation model to draw the mountain shadow map of the landslide body; the mountain shadow map can visually reflect the terrain undulation and geomorphic features of the landslide body, which helps to better understand the spatial distribution and deformation characteristics of the landslide body;

[0082] During the landslide zoning process, the rear edge of the landslide is located at the highest point of the landslide body, the front edge of the landslide is located at the lowest point of the landslide body, and the landslide body is located between the rear edge and the front edge of the landslide;

[0083] Analyze the vertical and horizontal deformation rates of the three parts of the landslide to understand the deformation characteristics and differences in different regions; this helps to more accurately judge the deformation mode and potential risks of the landslide.

[0084] In step S7, establishing the risk assessment model specifically includes:

[0085] Data preparation: Collect the deformation data of translational landslides, including horizontal displacement, vertical displacement, landslide body thickness, slip surface morphology, etc. In addition, data such as the topographic information, geological conditions, and hydrological conditions of the landslide also need to be collected.

[0086] Model selection: Select a risk assessment model based on fuzzy logic. This model can divide the risk levels into different fuzzy sets according to different input data and conduct risk assessment through fuzzy inference rules.

[0087] Determining model parameters: Determine model parameters, including the definition of fuzzy sets, the selection of membership functions, the formulation of inference rules, etc. The horizontal displacement, vertical displacement, landslide body thickness, slip surface morphology, etc. can be used as input variables, and different fuzzy sets such as "small", "medium", and "large" can be defined. At the same time, according to expert experience and actual situations, the membership functions and inference rules of each fuzzy set need to be determined.

[0088] Model training: Use the collected translational landslide data to train the model to determine the values of model parameters. Methods such as the least squares method can be used to minimize the error between the model prediction value and the actual observation value.

[0089] Model evaluation: Evaluate the trained model to verify its accuracy and reliability. Methods such as cross-validation can be used to divide the data set into a training set and a test set and calculate the prediction accuracy of the model on the test set.

[0090] Risk assessment: Use the trained model to conduct risk assessment on new translational landslides to determine their potential risk levels. The deformation data of the new translational landslide can be input into the model, and based on the output results of the model, its potential risk level can be judged.

[0091] Suppose a risk assessment model based on fuzzy logic is used to evaluate the potential risk level of translational landslides. The horizontal displacement, vertical displacement, landslide body thickness, slip surface morphology, etc. can be used as input variables, and the following fuzzy sets can be defined:

[0092] Horizontal displacement: "small", "medium", "large"

[0093] Vertical displacement: "small", "medium", "large"

[0094] Landslide body thickness: "thin", "medium", "thick"

[0095] Slip surface morphology: "plane shape", "concave shape"

[0096] Then, based on expert experience and actual situations, the membership functions and inference rules of each fuzzy set can be determined. The "small" horizontal displacement can be defined as a displacement less than 1 mm / year, the "medium" as 1 - 5 mm / year, and the "large" as greater than 5 mm / year.

[0097] Finally, the collected data of translational landslides can be used to train the model, and the trained model can be used to conduct risk assessment on new translational landslides. If a translational landslide has a horizontal displacement of 2 mm / year, a vertical displacement of 3 mm / year, a landslide body thickness of 2 m, and a slip surface shape of planar, then according to the output result of the model, the potential risk level of this landslide may be "medium".

[0098] In step S8, according to the risk assessment results, an early warning plan is formulated and real-time monitoring is carried out, which specifically includes the following methods:

[0099] Early warning threshold setting: According to the risk assessment model and the deformation characteristics of translational landslides, an early warning threshold is set. It can be set that when the horizontal displacement change rate exceeds 1 mm / year or the vertical displacement change rate exceeds 0.5 mm / year, an early warning is triggered.

[0100] Real-time monitoring: The InSAR technology is used for real-time monitoring to regularly obtain the deformation data of translational landslides. SAR images can be obtained once every month, and the change rates of horizontal displacement and vertical displacement are calculated.

[0101] Early warning information release: When the monitoring data indicates that the risk level of the translational landslide increases and reaches the early warning threshold, early warning information is promptly released. The early warning information can be released through channels such as text messages, emails, and social media to remind relevant personnel to take disaster prevention and mitigation measures.

[0102] Disaster prevention and mitigation measures: According to the early warning information, corresponding disaster prevention and mitigation measures are taken. Personnel can be organized to evacuate from the landslide dangerous area, warning signs can be set up, and engineering treatment can be carried out.

[0103] Suppose a risk assessment model based on fuzzy logic is used to evaluate the potential risk level of translational landslides. The horizontal displacement, vertical displacement, landslide body thickness, slip surface shape, etc. can be used as input variables, and different fuzzy sets such as "small", "medium", "large", etc. are defined. At the same time, based on expert experience and actual situations, the membership functions and inference rules of each fuzzy set need to be determined.

[0104] Then, the collected translational landslide data can be used to train the model, and the trained model can be used to conduct risk assessment on new translational landslides. If a translational landslide has a horizontal displacement of 2 mm / year, a vertical displacement of 3 mm / year, a landslide body thickness of 2 m, and a landslide surface shape of planar, then according to the output results of the model, the potential risk level of this landslide may be "medium".

[0105] Based on the risk assessment results, early warning plans can be formulated. The following early warning thresholds can be set:

[0106] Yellow warning: The horizontal displacement change rate exceeds 1 mm / year, or the vertical displacement change rate exceeds 0.5 mm / year.

[0107] Orange warning: The horizontal displacement change rate exceeds 2 mm / year, or the vertical displacement change rate exceeds 1 mm / year.

[0108] Red warning: The horizontal displacement change rate exceeds 3 mm / year, or the vertical displacement change rate exceeds 1.5 mm / year.

[0109] When the monitoring data indicates that the risk level of the translational landslide rises and reaches the early warning threshold, early warning information shall be issued in a timely manner. The early warning information can be issued through channels such as text messages, emails, and social media to remind relevant personnel to take disaster prevention and mitigation measures.

[0110] According to the early warning information, corresponding disaster prevention and mitigation measures can be taken. Personnel can be organized to evacuate from the landslide dangerous area, warning signs can be set up, and engineering treatment can be carried out.

[0111] In step S9, field geological drilling or trenching investigation methods are used to conduct on-site investigations on the structure, material composition, and deformation characteristics of the landslide body to verify the accuracy of the identification results.

[0112] In the present invention, the two-dimensional deformation decomposition formula is the core of the entire identification process, which can convert InSAR observation data with different incident angles into the actual deformation rates in the horizontal and vertical directions of the ground surface. By introducing terrain information and incident angle parameters, the line-of-sight displacement is converted into the displacement in the horizontal and vertical directions, so as to more accurately reflect the true deformation characteristics of the landslide. Using the mass conservation equation to invert the thickness of the soil landslide body and combining with terrain information to obtain the landslide morphology can greatly reduce costs and improve work efficiency. This is of great significance for large-scale landslide monitoring and identification. At the same time, the mass conservation equation can consider the material composition and deformation characteristics of the landslide body, so as to more accurately invert the thickness of the landslide surface. Inverting the landslide mode by comprehensively considering the landslide surface shape and two-dimensional displacement distribution characteristics can more accurately judge the deformation mode of the landslide, thereby providing a more reliable basis for landslide early warning and disaster prevention and control. The risk of landslide disasters can be effectively reduced.

[0113] In the present invention, by formulating an early warning plan and conducting real-time monitoring, it is possible to promptly detect an increase in the risk level of translational landslides and take corresponding disaster prevention and mitigation measures. The early warning plan can set different early warning thresholds according to the risk assessment results, such as yellow warning, orange warning, red warning, etc., so as to promptly issue early warning information. Real-time monitoring can regularly obtain the deformation data of translational landslides, thereby grasping the real-time deformation situation of translational landslides. When the monitoring data indicates an increase in the risk level of translational landslides, early warning information can be promptly issued to remind relevant personnel to take disaster prevention and mitigation measures, such as organizing personnel to evacuate from the landslide danger area, setting warning signs, and carrying out engineering treatment. This helps to reduce the threat of translational landslides to the safety of people's lives and property and improve the efficiency of disaster prevention and mitigation work.

[0114] In the present invention, by establishing a risk assessment model, it is possible to effectively evaluate the potential risk level of translational landslides. This model comprehensively considers factors such as the deformation characteristics, slide mass thickness, and slide surface morphology of translational landslides, and can more comprehensively reflect the actual situation of translational landslides. Through the risk assessment model, the risk level of translational landslides can be divided into different levels, such as low risk, medium risk, high risk, etc., thereby providing a scientific basis for landslide prevention and control work. This helps relevant departments to formulate targeted disaster prevention and mitigation measures and reduce the threat of translational landslides to the safety of people's lives and property.

[0115] It should be noted that in this article, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the term "comprising" or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or device comprising a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or device. Without further limitation, an element defined by the phrase "comprising a..." does not exclude the existence of additional identical elements in the process, method, article, or device comprising the element.

[0116] The above description enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown herein, but rather to the broadest scope consistent with the principles and novel features disclosed herein.

Claims

1. An assessment method for the risk of translational landslides using InSAR deformation monitoring, characterized in that: The method includes the following steps: S1: Collect radar remote sensing data covering the study area, including SAR images at different times and different orbits, and obtain high-precision aerial flight data for calculating landslide terrain information, including slope and aspect, to provide basic data for the analysis of translational landslide deformation; S2: Preprocess the SAR images, including radiometric correction, geometric correction, and registration, to ensure data quality and provide accurate data for the analysis of translational landslide deformation; calculate the annual average deformation amount in different line-of-sight directions of the landslide area to obtain planar deformation information; S3: Use SAR data with different incident angles, combined with landslide terrain information, to decompose the line-of-sight displacement into horizontal and vertical displacements, and analyze the distribution characteristics of the horizontal and vertical displacements; S4: Perform two-dimensional displacement decomposition, and use the two-dimensional deformation decomposition formula to convert the annual average deformation information at different incident angles into the actual two-dimensional deformation field on the ground surface to obtain the deformation rates in the horizontal and vertical directions; S5: Use high-precision aerial flight data, combined with the landslide body thickness information, to invert the slip surface morphology of the translational landslide. Based on the mass conservation equation, perform slip surface morphology inversion to invert the thickness of the soil landslide body; use the landslide terrain data and elevation information to subtract the landslide body thickness from the current elevation to obtain the landslide morphology; according to the maximum thickness of the landslide and the distribution characteristics of the landslide body thickness on both sides, divide the slip surface into concave and planar types; S6: According to the geomorphic distribution characteristics, divide the landslide body into three parts: the landslide rear edge, the landslide front edge, and the landslide body. Comprehensively judge the landslide deformation mode by combining the average horizontal deformation rate and the slip surface morphology. The identification characteristics of the translational landslide deformation mode are as follows: the slip surface morphology of the landslide body is planar, along the direction of the maximum slope drop, the change in the annual average horizontal deformation rate is not obvious, and the annual average vertical deformation rate shows a linear decrease characteristic; the difference between the maximum and minimum values of the annual average horizontal and vertical deformation rates is relatively large, and the maximum vertical deformation is located at the landslide rear edge; S7: Establish a risk assessment model to evaluate the potential risk level of the translational landslide; S8: According to the risk assessment results, formulate an early warning plan and conduct real-time monitoring, paying attention to the change trend of the horizontal displacement of the translational landslide and the linear decrease characteristic of the vertical displacement; when the monitoring data indicates an increase in the risk level of the translational landslide, issue early warning information in a timely manner and take corresponding disaster prevention and mitigation measures, with a focus on the early warning and protection measures in the landslide rear edge area; S9: Continuously optimize the risk assessment model according to the actual monitoring data to improve the accuracy and reliability of the assessment results.

2. The method for evaluating the risk of translational landslides using InSAR deformation monitoring according to claim 1, wherein: In step S1, during the process of collecting radar remote sensing data covering the study area, it is first necessary to determine the boundary of the study area and obtain the geographical coordinate information of this area; then, contact the satellite remote sensing data provider to obtain the SAR images covering this area; Collect SAR images at different time periods to analyze the variation trend of landslides over time; select SAR images obtained every few months within a year to capture the seasonal changes of landslides; while collecting SAR images, high-precision aerial flight data also needs to be obtained; this is achieved by contacting professional aerial survey companies or using drones for aerial photography; the aerial flight data provides high-resolution landslide topographic information, including slope and aspect, which is very important for analyzing the deformation characteristics of translational landslides.

3. The method for evaluating the risk of translational landslides using InSAR deformation monitoring according to claim 1, characterized in that: In step S2, the processing flow includes: Set the terrain factors in the Range direction and Azimuth direction to 5 and 1 respectively to generate an intensity map reflecting the actual terrain ratio of the study area; Select the radar intensity image obtained in the autumn and winter seasons as the main image, register the intensities of the remaining images to this main image, and require the registration error to be less than one-quarter of a pixel; using the polynomial coefficients obtained by intensity map registration iteration as the standard, register the original single-look complex raw data to obtain the resampled single-look complex data (rslc); calculate the resampled intensity map (rmli) based on the resampled single-look complex data. Set up an interference combination by setting the temporal baseline and spatial baseline; to maximize the satisfaction of the interference conditions, it is necessary to shorten the intervals of the temporal and spatial baselines of the interference, so set the maximum temporal baseline to 24d and the spatial baseline to 100m; Perform pixel registration on the rslc that constitutes the interference combination, then substitute the pixel values for conjugate multiplication to obtain the interference calculation result, and further substitute the topographic parameters of the study area in the radar coordinate format to obtain the differential interferometric phase; The differential interferometry data is unwrapped using the minimum cost flow method, and multiple differential interferometry phases are superimposed to calculate the phase within the superimposed time period. The superimposition calculation formula is Finally, the phase information is calculated as deformation information, and the annual average deformation amount of the study area can be obtained; Collect radar data at different incident angles and calculate according to the above method in turn to obtain the deformation values in different line-of-sight directions covering the same time period.

4. The method for evaluating the risk of translational landslides using InSAR deformation monitoring according to claim 1, wherein: In step S3, the specific method includes: resample the aspect and slope raster files so that the resolutions of different raster data are consistent, use the georeferencing tool in Arcgis to georegister the terrain raster data and deformation raster data with consistent resolutions, and further use the extractbymask tool for double registration verification to ensure the precise coincidence of multiple raster files; then define the direction of the maximum slope drop as the horizontal direction, and the direction perpendicular to the maximum slope drop as the vertical direction to establish a coordinate system.

5. The evaluation method for the risk of translational landslides using InSAR deformation monitoring according to claim 1, wherein: In step S4, the average annual deformation information obtained at different incident angles is denoted as dLos1 and dlos2 respectively. Use the raster calculator tool in Arcgis to substitute the dLos1, dlos2, Aspect, and slope raster data into the following formula to calculate the actual two-dimensional deformation field of the surface; where d H and d V are the horizontal and vertical ground deformation rates respectively, α represents the angle between the satellite flight direction and the true north direction, β is the angle between the maximum slope direction of the slope and the true north direction, and θ represents the satellite incidence angle.

6. The method for evaluating the risk of translational landslides using InSAR deformation monitoring according to claim 1, characterized in that: In step S5, the thickness of the sliding mass of the accumulation landslide tapers off at both side boundaries, and the sliding mass of the soil landslide shows uniform rheological characteristics in space. The landslide depth and the vertical deformation obtained by InSAR are solved by the following formula where h is the landslide thickness, t is the time, f is the rheological coefficient, and Vsurf is the surface deformation rate. Substitute the decomposed two-dimensional displacement field information and use the least squares method to obtain the information on the thickness of the sliding mass slope surface.

7. The method for evaluating the risk of translational landslides using InSAR deformation monitoring according to claim 1, wherein: In step S6, during the process of drawing the mountain shadow map, a high-precision surface elevation model is used to draw the mountain shadow map of the landslide body; the mountain shadow map can intuitively reflect the terrain undulation and geomorphic features of the landslide body, which helps to better understand the spatial distribution and deformation characteristics of the landslide body; During the landslide zoning process, the posterior edge of the landslide is located at the highest point of the landslide body, the anterior edge of the landslide is located at the lowest point of the landslide body, and the landslide body is located between the posterior edge and the anterior edge of the landslide; Analyze the vertical and horizontal deformation rates of the three parts of the landslide to understand the deformation characteristics and differences in different regions; This helps to more accurately judge the deformation mode and potential risks of the landslide.

8. The method for evaluating the risk of translational landslides using InSAR deformation monitoring according to claim 1, wherein: In step S7, establishing a risk assessment model specifically includes: Data preparation: Collect deformation data of translational landslides, including horizontal displacement, vertical displacement, landslide body thickness, and slip surface morphology; in addition, data on the topographic information, geological conditions, and hydrological conditions of the landslide also need to be collected; Model selection: Select a risk assessment model based on fuzzy logic. This model divides the risk levels into different fuzzy sets according to different input data, and conducts risk assessment through fuzzy inference rules; Determination of model parameters: Determine model parameters, including the definition of fuzzy sets, the selection of membership functions, and the formulation of inference rules; use horizontal displacement, vertical displacement, landslide body thickness, and slip surface morphology as input variables, and define different fuzzy sets, such as "small", "medium", "large"; at the same time, according to expert experience and actual situations, determine the membership functions and inference rules of each fuzzy set; Model training: Use the collected translational landslide data to train the model to determine the values of model parameters; use the least squares method to minimize the error between the model prediction value and the actual observed value; Model evaluation: Evaluate the trained model to verify its accuracy and reliability; use the cross-validation method to divide the data set into a training set and a test set, and calculate the prediction accuracy of the model on the test set; Risk assessment: Use the trained model to conduct risk assessment on new translational landslides to determine their potential risk levels; input the deformation data of the new translational landslide into the model, and judge its potential risk level according to the output result of the model.

9. The method for evaluating the risk of translational landslides using InSAR deformation monitoring according to claim 1, characterized in that: In step S8, according to the risk assessment results, formulate an early warning plan and conduct real-time monitoring, which specifically includes the following methods: Setting of early warning threshold: Set the early warning threshold according to the risk assessment model and the deformation characteristics of translational landslides; When the horizontal displacement change rate exceeds 1 mm / year, or the vertical displacement change rate exceeds 0.5 mm / year, trigger an early warning; Real-time monitoring: Use InSAR technology for real-time monitoring, and regularly obtain the deformation data of translational landslides; obtain SAR images once a month, and calculate the change rates of horizontal displacement and vertical displacement; Release of early warning information: When the monitoring data shows that the risk level of the translational landslide increases and reaches the early warning threshold, release early warning information in a timely manner; release early warning information through text messages, emails, and social media channels to remind relevant personnel to take disaster prevention and mitigation measures; Disaster prevention and mitigation measures: Take corresponding disaster prevention and mitigation measures according to the early warning information; Organize personnel to evacuate from the landslide danger area, set warning signs, and conduct engineering treatment.

10. The method for evaluating the risk of translational landslides using InSAR deformation monitoring according to claim 1, wherein: In the step S9, use the field geological drilling or trench investigation method to conduct on-site investigations on the structure, material composition, and deformation characteristics of the landslide body to verify the accuracy of the identification results.

Images