A refined numerical prediction method for global earthquake and tsunami

Through the global earthquake tsunami refined numerical forecasting method, the progressive calculation of low resolution in large areas, medium resolution in potentially affected areas and high resolution in target areas is solved, and the problem of insufficient timeliness forecasting in the existing technology is achieved. Efficient and refined forecasting and real-time assessment of global earthquake tsunamis.

CN118377057BActive Publication Date: 2024-11-01NAT MARINE ENVIRONMENTAL FORECASTING CENT
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410510261.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-04-26
Publication Date
2024-11-01
Estimated Expiration
2044-04-26

AI Technical Summary

Technical Problem

The existing technology lacks a globally covered numerical forecasting system for earthquake tsunamis, resulting in insufficient timeliness of tsunami forecasting and the inability to assess the threat of tsunamis caused by earthquakes in real time, especially in the level of international cooperation, which is difficult to achieve refined forecasting.

Method used

The global earthquake tsunami fine numerical forecasting method is adopted to obtain basic earthquake parameters and tsunami source parameters, and progressive calculations of low resolution in large areas, medium resolution in potentially affected areas and high resolution in target areas are carried out, and GPU acceleration calculation method is used. Quickly obtain forecasts for tsunami propagation time and amplitude, and then achieve refined tsunami forecasts.

Benefits of technology

It has achieved efficient and refined forecasting of global earthquake tsunamis, improved timeliness, and able to provide real-time assessment results before stakeholder areas suffer losses, and rationally use computing resources.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118377057B_ABST
    Figure CN118377057B_ABST
Patent Text Reader

Abstract

The present invention relates to a method for fine-grained numerical prediction of global earthquake tsunamis, comprising: obtaining basic earthquake parameters and most dangerous tsunami source parameters of the latest earthquake; calculating low-resolution numerical prediction information of tsunamis in a large area; determining the potential impact area of ​​tsunamis and the maximum tsunami propagation time ETA according to the grid point information exceeding the set wave height threshold in the maximum amplitude prediction field in the prediction information Max ; Obtain the source mechanism of the earthquake, update the tsunami source parameters, and calculate the potential impact area information and ETA Max , calculate the medium-resolution tsunami numerical prediction information of the potential tsunami impact area; determine whether the potential tsunami impact area contains at least one target area. If so, extract the maximum tsunami propagation time of the target area, and use different processes to calculate the high-resolution refined tsunami prediction results of the target area according to its relationship with the earthquake location. The above method realizes the progressive and refined prediction of global earthquake tsunamis with high timeliness.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to earthquake tsunami prediction technology, and in particular to a global earthquake tsunami refined numerical prediction method. Background Art

[0002] Tsunamis are one of the most unpredictable marine disasters with great destructive power. At present, there is no rapid earthquake tsunami fine-grained numerical prediction process covering coastal areas around the world. The existing solution is that each country surrounding an ocean or semi-enclosed sea area jointly operates and maintains a tsunami numerical prediction system for a fixed area. Countries and regions within the region are responsible for numerical forecasts and warnings for their own coasts, and tsunami threats outside the country rely on numerical forecast results from other countries. If you want to independently assess the tsunami hazard, you need to establish a tsunami numerical prediction model covering a fixed range for a specific coastal area (or "target area"); if you want to obtain refined forecast results, you also need to combine regional nesting solutions.

[0003] At present, the commercial tsunami numerical prediction technology mainly includes two categories: one is the tsunami scenario database method; the other is the method of using the source mechanism solution to activate the quasi-real-time tsunami numerical model. Both have their own advantages and disadvantages.

[0004] The database method is to perform massive scenario calculations based on the division of tsunami sources, and then store the tsunami simulation results in the database. Once a strong earthquake that may trigger a tsunami occurs, the database can quickly match the corresponding tsunami scenario based on the earthquake location and magnitude in a short period of time, and calculate the tsunami amplitude through spatial interpolation. Obviously, this method is only applicable to tsunami scenarios that occur in the database earthquake catalog. Once an earthquake occurs outside the catalog, the database method will not be able to respond; moreover, the database method cannot consider the earthquake's source and focal mechanism at the first time. In the long run, with the update of scientific knowledge of earthquakes, geology and tsunamis, the tsunami scenarios in the database need to be recalculated.

[0005] The tsunami numerical model method calculates the tsunami source parameters and the tsunami propagation process based on the earthquake source parameters. Its advantage is that it can respond to submarine earthquakes at any location, and the tsunami simulation results are closer to the actual situation. For refined numerical prediction results for small areas, a multi-layer grid nesting method is often used. However, the tsunami numerical model method requires real-time inversion of source parameters, and the fastest stable results can only be completed 10-20 minutes after the earthquake. In addition, the numerical model calculation also takes some time. Therefore, the timeliness of this method is slightly insufficient compared to the database method.

[0006] There is a lack of a refined numerical prediction system for earthquakes and tsunamis that is common to all oceans around the world. Tsunami forecasts and warning results for any sea area need to be obtained through communication and agreement at the international cooperation level. Summary of the invention

[0007] 1. Technical issues to be resolved

[0008] In view of the above-mentioned shortcomings and deficiencies of the prior art, the present invention provides a method for refined numerical prediction of global earthquakes and tsunamis.

[0009] (II) Technical solution

[0010] In order to achieve the above object, the main technical solutions adopted by the present invention include:

[0011] In a first aspect, an embodiment of the present invention provides a method for fine-grained numerical prediction of global earthquakes and tsunamis, comprising:

[0012] S10, obtaining basic earthquake parameters of the latest earthquake and the most dangerous tsunami source parameters corresponding to the basic earthquake parameters; the basic earthquake parameters include: epicenter location, earthquake time, and magnitude;

[0013] S20, obtaining large-area low-resolution tsunami numerical prediction information according to the pre-divided ocean area information, basic earthquake parameters, and most dangerous tsunami source parameters; the large-area low-resolution tsunami numerical calculation information includes: tsunami propagation time prediction field and maximum amplitude prediction field;

[0014] S30, determining the potential tsunami impact area according to the grid point information exceeding the set wave height threshold in the maximum amplitude forecast field in the tsunami numerical forecast information; and obtaining the maximum tsunami propagation time ETA corresponding to the potential tsunami impact area from the tsunami propagation time forecast field Max ;

[0015] S40, obtain the earthquake source mechanism and the updated most dangerous tsunami source parameters, and then according to the potential impact area range information and the ETA Max , calculate medium-resolution numerical tsunami prediction information for tsunami potential impact areas; and,

[0016] Determine whether the tsunami potential impact area contains at least one target area. If so, extract the maximum tsunami propagation time RETA of the target area. Max ,According to the basic information of the target area, the relationship between the target area and the earthquake location is judged, so that different processes are used to calculate the refined tsunami forecast results of the target area.

[0017] Specifically, the S10 includes:

[0018] Determine whether the epicenter is within a distance of P from the seabed or the land side of the coastline and the magnitude is above M. If not, end the process; P is greater than 100km and M is greater than 7.0;

[0019] Otherwise, the ocean area and the large forecast area to which the current earthquake belongs are determined based on the epicenter location. Each ocean area corresponds to a fixed large forecast area, and the spatial range of the large forecast area is a known boundary.

[0020] According to the epicenter location, the strike angle, dip angle, fault plane depth, 90° slip angle, rupture length and width, and slip amount of tsunami source parameters are obtained from the global subduction zone slab system.

[0021] Specifically, the S20 includes:

[0022] According to the determined large forecast area, the known boundary information of the large forecast area, the spherical orthogonal grid resolution, time step, forecast duration, and wave height threshold of potential impact during numerical calculation are obtained, and the parameter file required for the tsunami numerical model is obtained; the spherical orthogonal grid resolution during numerical calculation is 2 to 5 arc-min;

[0023] According to the tsunami source parameters and tsunami model parameter files, a large-area low-resolution tsunami numerical model is used to execute a numerical calculation method based on GPU acceleration to obtain large-area low-resolution tsunami numerical forecast information;

[0024] Specifically, in the numerical calculation process, the tsunami wave height evolution process of all grid points is calculated step by step based on the time step, and it is determined whether each grid point reaches the set wave height threshold. If it reaches, the grid point information, i.e., the time, is recorded;

[0025] In addition, the tsunami wave height obtained at each time step is compared with the tsunami wave height at the previous integration moment, and the maximum value information of the tsunami wave height is recorded; after completing the calculation of the set forecast time, the large-area low-resolution tsunami numerical forecast information is obtained.

[0026] Specifically, the S30 includes: counting the positions of grid points exceeding the wave height threshold in the maximum amplitude forecast field, and obtaining the potential tsunami impact area in a point-to-surface manner;

[0027] The maximum tsunami propagation time (ETA) of the potential affected area is obtained from the tsunami propagation time forecast field. Max .

[0028] Specifically, the S40 includes:

[0029] S41. Obtain the boundary range information of the potential tsunami impact area, spherical orthogonal grid resolution, time step, forecast duration, and wave height threshold that causes impact;

[0030] S42, obtaining the source mechanism information of the latest earthquake, updating the most dangerous tsunami source parameters and obtaining the tsunami initial field information;

[0031] S43, calculating the tsunami forecast result of the tsunami potential impact area according to the boundary range information, source mechanism information, and tsunami initial field information of the tsunami potential impact area;

[0032] S43, determining whether the tsunami potential impact area includes at least one target area, wherein the target area is an area where there is a threat to property or life;

[0033] S44. If there is at least one target area, for the target area to be refined, the refined tsunami forecast result of the target area is calculated based on the basic information of the target area and the updated earthquake source parameters, or the refined tsunami forecast result of the target area is calculated based on the basic information of the target area and the updated earthquake basic parameters and source mechanism information.

[0034] Specifically, the S44 includes:

[0035] For the first target area to be finely analyzed, it is determined whether the epicenter position is within the first specified range from the first target area. If so, based on the range of the first target area, the resolution of the spherical orthogonal grid, the forecast duration, the source mechanism information at the latest moment, and the tsunami initial field at the latest moment, a high-resolution close-range tsunami numerical calculation is performed to obtain a fine tsunami forecast result for the first target area;

[0036] If the epicenter position is not located in the first specified range of the first target area, but is located in the second specified range, acquiring boundary driving data of the first target area;

[0037] According to the spherical orthogonal grid resolution, forecast duration and boundary driving data of the first target area, high-resolution close-range tsunami numerical calculations are performed to obtain refined tsunami forecast results for the first target area.

[0038] Specifically, a nonlinear equation or a Bausch-Nessker equation is used to perform high-resolution short-range tsunami numerical calculations to obtain a refined tsunami forecast result for the first target area;

[0039] The first designated range is less than or equal to 500km, and the second designated range is greater than 500km;

[0040] When the epicenter is within the first specified range from the first target area, the spherical orthogonal grid resolution is 0.125 to 0.25 arc-min, and the forecast duration is 6 hours + Roundup (RETA Max ); Roundup means rounding up to an integer.

[0041] RETA Max Extract the maximum propagation time in the target area from the tsunami propagation time forecast field of the potential impact area;

[0042] When the distance between the epicenter and the first target area is greater than the first specified range and less than the second specified range, the spherical orthogonal grid resolution is also 0.125 to 0.25 arc-min, and the forecast duration is 6h+(Roundup(RETA Max )-Rounddown(RETA Min )); Rounddown means round down to an integer;

[0043] RETA Max To extract the maximum propagation time in the target area from the medium-resolution tsunami propagation time forecast field in the potential impact area, RETA Min To extract the minimum propagation time in the target area from the medium-resolution tsunami propagation time forecast field in the potential impact area;

[0044] The boundary driving data is the process data of the tsunami propagation calculation in the potential impact area to provide a time period for the first target area in [Rounddown (RETA Min ),6h+Roundup(RETA Max )] within the boundary conditions.

[0045] Specifically, the S43 includes:

[0046] The medium-resolution spherical orthogonal grid resolution in the potential tsunami impact area is 0.5 to 1 arc-min, and the forecast time is 6h + Roundup (ETA Max ),

[0047] High-resolution short-range tsunami numerical calculations are performed using nonlinear equations or the Bausch-Nessker equation to obtain tsunami forecast results for areas potentially affected by tsunamis.

[0048] In a second aspect, an embodiment of the present invention further provides a computing device, comprising: a memory, a CPU, a GPU and a bus, wherein the CPU, the GPU and the memory are all connected via the bus;

[0049] The memory is used to store a control program, and the CPU and GPU cooperate to run the control program, wherein when the control program is run, a global earthquake and tsunami refined numerical prediction method described in any one of the first aspects is executed.

[0050] (III) Beneficial effects

[0051] The purpose of the present invention is to provide a method for refined numerical prediction of global earthquake tsunamis. In the numerical prediction process, a progressive method of "low resolution in large areas - medium resolution in potential impact areas - high resolution in target areas" is adopted, so that real-time assessment results of tsunami threats can be obtained in a shorter time before the areas of interest suffer losses.

[0052] The method of the present invention realizes the refined forecast of global earthquake tsunamis, has high timeliness and reasonable use of computing resources. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 The schematic diagram of the layer-by-layer forecast range in the Pacific region is shown;

[0054] Figure 2 and Figure 3 All of them are flow charts of a method for refined numerical prediction of global earthquakes and tsunamis provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0055] In order to better explain the present invention and facilitate understanding, the present invention is described in detail below through specific implementation modes in conjunction with the accompanying drawings.

[0056] In order to better understand the above technical solution, exemplary embodiments of the present invention will be described in more detail below with reference to the accompanying drawings. Although exemplary embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to enable a clearer and more thorough understanding of the present invention and to fully convey the scope of the present invention to those skilled in the art.

[0057] Drawing on the existing national tsunami warning program, it is necessary to establish a tsunami numerical prediction system for a specific coastal area that covers the tsunami threat to the area. As the number of coastal areas increases, it is time-consuming and laborious both from the perspective of use and maintenance.

[0058] With the continuous development of high-performance computing technology and algorithm updates, the time required for the second scheme of tsunami numerical prediction is being shortened. If the business process can be further optimized, the entire numerical prediction process can be carried out in a progressive manner of "low resolution in large areas-medium resolution in potential impact areas-high resolution in target areas", and the real-time assessment results of tsunami threats can be obtained in a shorter time before the areas of interest suffer losses.

[0059] Embodiment 1

[0060] The embodiment of the present invention provides a method for fine-grained numerical prediction of global earthquakes and tsunamis. The method of this embodiment may include the following steps:

[0061] S10. Obtain basic earthquake parameters of the latest earthquake and the most dangerous tsunami source parameters corresponding to the basic earthquake parameters; the basic earthquake parameters include: epicenter location, earthquake time, and magnitude.

[0062] For example, determine whether the epicenter is within a distance of P from the seabed or the land side of the coastline and the magnitude is above M. If not, end the event; P is greater than 100km and M is greater than 7.0;

[0063] Otherwise, the ocean area and the large forecast area to which the current earthquake belongs are determined based on the epicenter location. Each ocean area corresponds to a fixed large forecast area, and the spatial range of the large forecast area is a known boundary.

[0064] According to the epicenter location, the strike angle, dip angle, fault plane depth, slip angle, depth, rupture length, rupture width and slip amount of tsunami source parameters are obtained from the global subduction zone slab system (the existing global subduction zone slab model or system).

[0065] For example, the sliding angle corresponding to the most dangerous situation may be 90°. The above-mentioned rupture length and rupture width may be determined according to an empirical formula, and the slip amount may be calculated according to the empirical formula.

[0066] S20. Obtaining large-area low-resolution tsunami numerical prediction information based on pre-divided ocean area information, basic earthquake parameters, and most dangerous tsunami source parameters; the large-area low-resolution tsunami numerical prediction information includes: tsunami propagation time prediction field and maximum amplitude prediction field.

[0067] For example, based on the determined large forecast area, the known boundary information of the large forecast area, the spherical orthogonal grid resolution, time step, forecast duration, and wave height threshold with potential impact during numerical calculation can be obtained to obtain the tsunami model parameter file of the ocean area information; the spherical orthogonal grid resolution during numerical calculation is 2 to 5 arc-min;

[0068] According to the tsunami source parameters and tsunami model parameter files, a large-area low-resolution tsunami numerical model is used to execute a numerical calculation method based on GPU acceleration to obtain large-area low-resolution tsunami numerical forecast information;

[0069] Specifically, in the numerical calculation process, the tsunami wave height evolution process of all grid points is calculated step by step based on the time step, and it is determined whether each grid point reaches the set wave height threshold. If it reaches, the grid point information, i.e., the time, is recorded;

[0070] In addition, the tsunami wave height obtained at each time step is compared with the tsunami wave height at the previous integration moment, and the information of the larger tsunami wave height is recorded; after completing the calculation of the set forecast time, the low-resolution tsunami numerical forecast information for a large area is obtained.

[0071] S30, determining the potential tsunami impact area according to the grid point information exceeding the set wave height threshold in the maximum amplitude forecast field in the tsunami numerical forecast information; and obtaining the maximum tsunami propagation time ETA corresponding to the potential tsunami impact area from the tsunami propagation time forecast field Max .

[0072] In this embodiment, the potential tsunami impact area refers to the range given by the low-resolution tsunami numerical prediction results. Max The maximum value of the tsunami propagation time is calculated within this range.

[0073] S40, obtain the earthquake source mechanism and the updated most dangerous tsunami source parameters, and then according to the potential impact area range information and the ETA Max , calculate medium-resolution numerical tsunami forecast information for the potential tsunami impact area; and,

[0074] Determine whether the tsunami potential impact area contains at least one target area. If so, extract the maximum tsunami propagation time RETA of the target area. Max ,According to the basic information of the target area, the relationship between the target area and the earthquake location is judged, so that different processes are used to calculate the refined tsunami forecast results of the target area.

[0075] In this embodiment, the basic earthquake parameters need to be updated before each numerical calculation.

[0076] Regarding the above step S30, it may include:

[0077] Count the grid points that exceed the wave height threshold in the maximum amplitude forecast field, and obtain the potential tsunami impact area from points to surfaces;

[0078] The maximum tsunami propagation time (ETA) of the potential affected area is obtained from the tsunami propagation time forecast field. Max .

[0079] Regarding the above step S40, it includes:

[0080] S41. Obtain the boundary range information of the potential tsunami impact area, spherical orthogonal grid resolution, time step, forecast duration, and wave height threshold that causes impact;

[0081] S42, obtaining the source mechanism information of the latest earthquake, updating the most dangerous tsunami source parameters and obtaining the tsunami initial field information;

[0082] S43, obtaining a tsunami forecast result for the tsunami potential impact area according to the boundary range information, source mechanism information, and tsunami initial field information of the tsunami potential impact area;

[0083] S43, judging whether the tsunami potential impact area includes at least one target area, and if so, determining whether each target area requires a refined tsunami forecast result; the target area is an area where there is a threat to property or life, that is, an area of ​​interest.

[0084] The medium-resolution spherical orthogonal grid resolution in the potential tsunami impact area is 0.5 to 1 arc-min, and the forecast time is 6h + Roundup (ETA Max );

[0085] High-resolution short-range tsunami numerical calculations are performed using nonlinear equations or the Bausch-Nessker equation to obtain tsunami forecast results for areas potentially affected by tsunamis.

[0086] S44. If there is at least one target area that requires a refined forecast result, then for the target area to be refined, based on its relationship with the earthquake location, different processes are used to calculate the refined tsunami forecast result of the target area.

[0087] Specifically, the S44 includes:

[0088] For the first target area to be finely analyzed, it is determined whether the epicenter position is within the first specified range from the first target area. If so, based on the range of the first target area, the resolution of the spherical orthogonal grid, the forecast duration, the source mechanism information at the latest moment, and the tsunami initial field at the latest moment, a high-resolution close-range tsunami numerical calculation is performed to obtain a fine tsunami forecast result for the first target area;

[0089] If the epicenter position is not located in the first specified range of the first target area, but is located in the second specified range, acquiring boundary driving data of the first target area;

[0090] According to the spherical orthogonal grid resolution, forecast duration and boundary driving data of the first target area, high-resolution close-range tsunami numerical calculations are performed to obtain refined tsunami forecast results for the first target area.

[0091] In this embodiment, nonlinear equations or Bauschnesk equations can be used to perform high-resolution short-range tsunami numerical calculations to obtain refined tsunami forecast results for the first target area;

[0092] The first designated range is less than or equal to 500km, and the second designated range is greater than 500km;

[0093] When the epicenter is within the first specified range from the first target area, the spherical orthogonal grid resolution is 0.125 to 0.25 arc-min (200 to 400 m), and the forecast duration is 6 hours + Roundup (RETA Max ); Roundup means rounding up to an integer;

[0094] RETA Max To extract the maximum propagation time in the target area from the medium-resolution tsunami propagation time forecast field in the potential impact area;

[0095] When the distance between the epicenter and the first target area is greater than the first specified range and less than the second specified range, the spherical orthogonal grid resolution is also 0.125 to 0.25 arc-min, and the forecast duration is 6h+(Roundup(RETA Max )-Rounddown(RETA Min ));

[0096] RETA Max To extract the maximum propagation time in the target area from the medium-resolution tsunami propagation time forecast field in the potential impact area, RETA Min To extract the minimum propagation time in the target area from the medium-resolution tsunami propagation time forecast field in the potential impact area;

[0097] The boundary driving data is the process data of the tsunami propagation calculation in the potential impact area to provide a time period for the first target area in [Rounddown (RETA Min ),6h+Roundup(RETA Max )] within the boundary conditions.

[0098] The numerical forecasting process of this embodiment adopts a progressive approach of "low resolution for large areas - medium resolution for potential impact areas - high resolution for target areas", which enables real-time assessment results of tsunami threats to be obtained in a shorter time before the areas of interest suffer losses.

[0099] Embodiment 2

[0100] A global earthquake tsunami refined numerical prediction method of this embodiment may include the following steps:

[0101] Step 1: Obtain basic earthquake parameters in real time.

[0102] The basic earthquake parameters of this embodiment include: the location of the earthquake, i.e., the epicenter, the time of the earthquake, and the magnitude.

[0103] For example, if the earthquake is determined to be located within 100 km of the seabed or the land side of the coastline and the magnitude is above 7.0, the ocean area to which the epicenter belongs will be judged, such as whether it is located within the five independent ocean areas of the Pacific Ocean, Indian Ocean, Atlantic Ocean, Mediterranean Sea or Caribbean Sea.

[0104] In this embodiment, each ocean area corresponds to a fixed large forecast area, and the specific spatial range is fixed.

[0105] like Figure 1 As shown, the area inside the first wireframe is the large forecast area corresponding to the Pacific Ocean among the five independent ocean regions in the world. The large forecast area is an area with a fixed range. In practice, low-resolution grids are used for numerical calculation of tsunamis. Figure 1 The area covered by the second wireframe is the potential impact area calculated based on the wave height threshold, and the tsunami numerical calculation is performed using a medium-resolution grid. Figure 1 The area covered by the third wireframe is the specific coastal area, i.e., the target area, where high-resolution grids are used for numerical calculation of tsunamis.

[0106] Figure 1 The five-pointed star in the middle indicates the location where the earthquake occurred.

[0107] Step 2: According to the epicenter location, query the known global subduction zone slab model to obtain the tsunami source parameters at the epicenter location.

[0108] The tsunami source parameters may include: earthquake location, fault strike angle, fault plane depth, dip angle, slip angle, depth, rupture length and width, and slip amount, etc. The most dangerous tsunami source slip angle in this example is set to 90°. The tsunami source parameters are stored.

[0109] It is understandable that it is also necessary to obtain the rupture length, rupture width, and slip amount of the fault plane.

[0110] The rupture length and width of the fault can also be estimated based on the calculation of the initial tsunami field, such as using the "magnitude-rupture scale empirical formula"; and the slip of the fault plane can be estimated by setting the elastic modulus of the fault medium. Usually, the faults that trigger tsunamis are located on the oceanic crust (oceanic crust) by default. The composition of the oceanic crust (that is, the medium) at different locations varies significantly with depth, and the elastic modulus can be regarded as a function of the focal depth.

[0111] The elastic modulus of the fault varies with the location, composition and geological structure of the oceanic crust.

[0112] Considering that the epicenters of disastrous tsunamis are relatively shallow, this embodiment takes the minimum value of the global ocean crust average as 30 GPa.

[0113] Furthermore, based on the above tsunami source parameters, the most dangerous tsunami scenario caused by a pure earthquake process can be obtained.

[0114] Step 3: Based on the tsunami source parameter information of the earthquake location and the basic information of the earthquake location, the required parameter conditions are calculated and a large-area low-resolution tsunami model parameter file is generated.

[0115] For example, the forecast area boundary corresponding to the ocean where the earthquake is located, the spherical orthogonal grid resolution, time step, forecast duration, wave height threshold causing impact (0.1m and above), etc. are written into the file to form a large-area low-resolution tsunami model parameter file.

[0116] The grid resolution here can be set to 2arc-min or 5arc-min (low resolution). The larger the resolution value, the coarser the grid, and the shorter the forecast time for the same duration. The forecast duration should be set to ensure that the tsunami can propagate to all grid points in the forecast area with a water depth of more than 100m. The forecast duration for the five large forecast areas can be determined through testing. Taking the Pacific Ocean as an example, it takes at least 36 hours for a tsunami to cover the entire area.

[0117] Step 4: Based on the large-area low-resolution tsunami model parameter file, start the large-area low-resolution tsunami numerical model.

[0118] The model running environment of this embodiment is a high-performance computer with a CPU+GPU processor. It needs to be completed using a fast execution method based on GPU acceleration of the tsunami numerical model. Specifically, the OKADA elastic deformation model is used to calculate the crustal deformation caused by the earthquake and directly mapped to the surface as the tsunami initial field; the linear equation is used to calculate the tsunami propagation process, which has the advantage of being time-saving.

[0119] Step 5: In the process of numerically solving the equation, the model will calculate the evolution of the tsunami wave height of all grid points at each time step, and determine whether any grid point reaches the set wave height threshold. If reached, the current moment of tsunami propagation will be recorded as the tsunami propagation time of the corresponding point (that is, how long after the earthquake the tsunami begins to affect this place); in the subsequent integration process, the wave height threshold judgment and propagation time update of the grid point will no longer be performed. At the same time, the tsunami wave height calculated at each time step will also be compared with the wave height at the previous integration moment, and the larger wave height value will be replaced and saved. The above process is repeated at each integration moment during the forecast period. After the calculation of the set forecast time is completed, the tsunami propagation time forecast field (corresponding to all grid points) and the maximum amplitude forecast field are output and saved to the file.

[0120] Step 6: Count the points in the maximum amplitude forecast field that exceed the wave height threshold, and delineate the potential tsunami impact area from points and surfaces; obtain the maximum tsunami propagation time ETA from the tsunami propagation time forecast results Max , which represents the minimum time for a tsunami to cover the potential impact area.

[0121] Step 7: Obtain medium-resolution tsunami model parameter information in the potential impact area.

[0122] Extract the four boundaries of the potential tsunami impact area in the southeast, northwest, and northeast regions, and set the spherical orthogonal grid resolution, time step, forecast duration, and wave height threshold for whether it causes impact. The grid resolution here can be set to 0.5 arc-min or 1 arc-min (medium resolution), and the forecast duration is set to 6h+Roundup (ETA Max ). The purpose of this setting is to ensure that the point with the longest tsunami propagation time can also obtain at least 6 tsunami wave cycles (except for super tsunamis, the tsunami wave cycle in the high-energy frequency band is usually no more than 1h), that is, to avoid blindly setting the forecast time, resulting in insufficient tsunami calculations, or wasting computing resources due to too long a time. In addition, if the earthquake source mechanism based on real-time data inversion can be obtained at this time, the source parameters used in the most theoretically dangerous tsunami scenario can be replaced to avoid overestimation of tsunami danger.

[0123] Step 8: Use the model parameter file set in step 7 and the updated source parameters to start medium-resolution tsunami numerical prediction in the potential impact area.

[0124] It is understandable that the source parameters need to be updated before each tsunami numerical calculation, so the updated source parameters are used in step 8.

[0125] For specific implementation methods, please refer to step 4. The slight change is that here, the nonlinear equation or the Bausinesq equation will be selected according to the spatial scale of the potential impact area. The difference between the two is that when the tsunami propagates over a long distance, the Bausinesq equation will be able to obtain relatively more accurate tsunami forecast results due to the addition of high-order dispersion terms. After the calculation of the set forecast time is completed, the tsunami propagation time forecast field and the maximum amplitude forecast field are output and saved to the file.

[0126] Step 9: Combined with the medium-resolution tsunami forecast results in the potential impact area, if there are key areas of concern or high-value targets facing a greater tsunami risk (depending on the specific conditions of the carrier), then the refined tsunami numerical forecast for the target area should be initiated. There are two specific situations: the first is that the target area is close to the earthquake location (less than 500km); the second is that the target area is far away from the earthquake location (more than 500km).

[0127] Step 10: For the first case in step 9, a similar solution to the numerical prediction of tsunami in the potential impact area can be adopted (refer to steps 7 and 8). The differences include the improvement of grid resolution and the different forecast durations. The specific process is to define the calculation range according to the target area, increase the grid resolution to 0.125 arc-min or 0.25 arc-min (it can be higher according to actual conditions); the forecast time is set by extracting the maximum propagation time RETA in the target area from the tsunami propagation time forecast field in the potential impact area. Max , round up and add 6h to get Roundup (RETA Max )+6h.

[0128] Step 11: For the second case of step 9, it is necessary to perform high-resolution numerical forecasts for the target area through boundary-driven methods. The specific solution is to define the calculation range according to the target area, set the grid resolution to 0.125 arc-min or 0.25 arc-min; extract the minimum tsunami arrival time RETA of the target area from the medium-resolution potential impact area tsunami forecast results Min and the maximum tsunami arrival time RETA Max , the propagation time of the high-resolution tsunami model in the target area is 6h+(Roundup(RETA Max )-Rounddown(RETA Min )); The so-called boundary drive refers to the use of process data of tsunami propagation calculation in the potential impact area to provide [Rounddown(RETA Min ),6h+Roundup(RETA Max )]The boundary conditions of this time period.

[0129] Step 12: After the calculation of the set forecast time is completed, the tsunami propagation time forecast field and the maximum amplitude forecast field are output and saved to a file. The refined tsunami forecast results for the target area also include the tsunami wave process forecast for a specific location or one or more forecast points around the carrier.

[0130] Taking the 7.3 magnitude earthquake tsunami event in Taiwan on April 3, 2024 as an example, the beneficial effects of the present invention are introduced. First, according to the location of the earthquake, the numerical prediction area is determined to be the Pacific Ocean. After determining the basic parameters of the earthquake, the Taiping area is selected for low-resolution numerical prediction. On the Intel (R) Xeon (R) Gold 5218CPU and NVIDIA A100-PCIE GPU heterogeneous high-performance computing platform, the numerical calculation of the 36h tsunami in the Pacific region takes about 15s. The amplitude judgment standard for the potential impact area is 0.1m. Strictly speaking, the potential impact area is an irregular surface. In practice, the southeast, northwest and northwest boundary values ​​can be taken to draw a rectangle for the convenience of the next step of calculation. At the same time, the tsunami propagation time of the potential impact area is obtained. Statistics show that the maximum propagation time for the potential impact area to reach 0.1m is 3.2h.

[0131] The tsunami propagation time is 3.2+6, which is rounded to 10h, and the resolution is set to 1arc-min. It takes about 5 seconds to make a medium-resolution regional tsunami numerical forecast, and the maximum tsunami amplitude and tsunami propagation time are obtained. Assume that there are two target areas, the set R1 target area and the set R2 target area.

[0132] For the R1 target area, the earthquake is located within the area. After updating the focal mechanism, the maximum tsunami propagation time of the target area is queried from the regional tsunami propagation time result, which is 0.4h. After adding 6, it is rounded to 7h. The resolution of the target area is set to 0.125arc-min (about 200m). The high-resolution and refined tsunami numerical forecast is expected to take about 15s. It can be found that the predicted tsunami amplitude in the coastal waters of the target area generally exceeds 1m.

[0133] For the R2 target area, the earthquake is outside the area, and a boundary-driven approach is required for numerical prediction. From the regional tsunami propagation time results, the maximum tsunami propagation time in the target area is 0.9h and the minimum propagation time is 0.6h. According to the algorithm, the simulation start time is 0:

[0134] The target area resolution is set to 0.125 arc-min (about 200 m), and a high-resolution and refined tsunami numerical forecast is conducted, which is expected to take about 25 seconds.

[0135] The entire numerical prediction process is carried out in a progressive manner of "large area low resolution - potential impact area medium resolution - target area high resolution". Under the above target area assumptions, the refined numerical prediction process of the 7.3 magnitude earthquake tsunami in Taiwan is expected to take about 70 seconds. The target area that is not in the potential impact area will no longer be refined numerically predicted because the amplitude does not reach the 0.1m threshold set for the large area. The entire numerical prediction process is highly timely and uses computing resources reasonably.

[0136] Embodiment 3

[0137] According to another aspect of an embodiment of the present invention, the embodiment of the present invention further uses a computing device to execute the methods described in the above-mentioned embodiments 1 and 2, and the computing device may include: a memory, a processor, and a bus, wherein the processor is connected to the memory through the bus; the processor includes multiple GPUs and CPUs;

[0138] The memory is used to store a control program, and the processor is used to run the control program, wherein the control program executes the steps of the method described in any of the above embodiments when it is run.

[0139] It should be noted that in the claims, any reference numerals placed between brackets shall not be construed as limiting the claims. The word "comprising" does not exclude the presence of components or steps not listed in the claims. The word "a" or "an" preceding a component does not exclude the presence of a plurality of such components. The invention may be implemented by means of hardware comprising several different components and by means of a suitably programmed computer. In the claims enumerating several means, several of these means may be embodied by the same hardware. The use of the words first, second, third, etc., is for convenience of expression only and does not indicate any order. These words may be understood as part of the component name.

[0140] In addition, it should be noted that, in the description of this specification, the description of the terms "one embodiment", "some embodiments", "embodiment", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples, unless they are contradictory.

[0141] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments after knowing the basic creative concept. Therefore, the claims should be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present invention.

[0142] Obviously, those skilled in the art can make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention should also include these modifications and variations.

Claims

1. A method for refined numerical prediction of global earthquake and tsunami, characterized in that: include: S10, obtaining basic earthquake parameters of the latest earthquake and the most dangerous tsunami source parameters corresponding to the basic earthquake parameters; The basic parameters of an earthquake include: epicenter location, earthquake time, and magnitude; S20, obtaining large-area low-resolution tsunami numerical prediction information according to the pre-divided ocean area information, basic earthquake parameters, and most dangerous tsunami source parameters; the large-area low-resolution tsunami numerical calculation information includes: tsunami propagation time prediction field and maximum amplitude prediction field; S30, determining the potential tsunami impact area according to the grid point information exceeding the set wave height threshold in the maximum amplitude forecast field in the tsunami numerical forecast information; and obtaining the maximum tsunami propagation time ETA corresponding to the potential tsunami impact area from the tsunami propagation time forecast field Max ; S40, obtain the earthquake source mechanism and the updated most dangerous tsunami source parameters, and then according to the potential impact area range information and the ETA Max , calculate medium-resolution numerical tsunami forecast information for the potential tsunami impact area; and, Determine whether the tsunami potential impact area contains at least one target area. If so, extract the maximum tsunami propagation time RETA of the target area. Max ,According to the basic information of the target area, the relationship between the target area and the earthquake location is judged, so that different processes are used to calculate the refined tsunami forecast results of the target area.

2. A global earthquake tsunami refined numerical prediction method according to claim 1, characterized in that: The S10 includes: Determine whether the epicenter is within a distance of P from the seabed or the land side of the coastline and the magnitude is above M. If not, end the process; P is greater than 100km and M is greater than 7.0; Otherwise, the ocean area and the large forecast area to which the current earthquake belongs are determined based on the epicenter location. Each ocean area corresponds to a fixed large forecast area, and the spatial range of the large forecast area is a known boundary. According to the epicenter location, the strike angle, dip angle, fault plane depth, 90° slip angle, rupture length and width, and slip amount of tsunami source parameters are obtained from the global subduction zone slab system.

3. A global earthquake tsunami refined numerical prediction method according to claim 2, characterized in that: The S20 includes: According to the determined large forecast area, the known boundary information of the large forecast area, the spherical orthogonal grid resolution, time step, forecast duration, and wave height threshold of potential impact during numerical calculation are obtained, and the parameter file required for the tsunami numerical model is obtained; the spherical orthogonal grid resolution during numerical calculation is 2 to 5 arc-min; According to the tsunami source parameters and tsunami model parameter files, a large-area low-resolution tsunami numerical model is used to execute a numerical calculation method based on GPU acceleration to obtain large-area low-resolution tsunami numerical forecast information; Specifically, in the numerical calculation process, the tsunami wave height evolution process of all grid points is calculated step by step based on the time step, and it is determined whether each grid point reaches the set wave height threshold. If it reaches, the grid point information, i.e., the time, is recorded; In addition, the tsunami wave height obtained at each time step is compared with the tsunami wave height at the previous integration moment, and the maximum value information of the tsunami wave height is recorded; after completing the calculation of the set forecast time, the large-area low-resolution tsunami numerical forecast information is obtained.

4. A method for refined numerical prediction of global earthquake tsunamis according to any one of claims 1 to 3, characterized in that: The S30 includes: counting the positions of grid points exceeding the wave height threshold in the maximum amplitude forecast field, and obtaining the potential impact area of ​​the tsunami by means of points and surfaces; The maximum tsunami propagation time (ETA) of the potential affected area is obtained from the tsunami propagation time forecast field. Max .

5. A method for refined numerical prediction of global earthquake tsunamis according to any one of claims 1 to 3, characterized in that: The S40 includes: S41. Obtain the boundary range information of the potential tsunami impact area, spherical orthogonal grid resolution, time step, forecast duration, and wave height threshold that causes impact; S42, obtaining the source mechanism information of the latest earthquake, updating the most dangerous tsunami source parameters and obtaining the tsunami initial field information; S43, calculating the tsunami forecast result of the tsunami potential impact area according to the boundary range information, source mechanism information, and tsunami initial field information of the tsunami potential impact area; S43, determining whether the tsunami potential impact area includes at least one target area, wherein the target area is an area where there is a threat to property or life; S44. If there is at least one target area, for the target area to be refined, the refined tsunami forecast result of the target area is calculated based on the basic information of the target area and the updated earthquake source parameters, or the refined tsunami forecast result of the target area is calculated based on the basic information of the target area and the updated earthquake basic parameters and source mechanism information.

6. A global earthquake tsunami refined numerical prediction method according to claim 5, characterized in that: The S44 includes: For the first target area to be finely analyzed, it is determined whether the epicenter position is within the first specified range from the first target area. If so, based on the range of the first target area, the resolution of the spherical orthogonal grid, the forecast duration, the source mechanism information at the latest moment, and the tsunami initial field at the latest moment, a high-resolution close-range tsunami numerical calculation is performed to obtain a fine tsunami forecast result for the first target area; If the epicenter position is not located in the first specified range of the first target area, but is located in the second specified range, acquiring boundary driving data of the first target area; According to the spherical orthogonal grid resolution, forecast duration and boundary driving data of the first target area, high-resolution close-range tsunami numerical calculations are performed to obtain refined tsunami forecast results for the first target area.

7. A global earthquake tsunami refined numerical prediction method according to claim 6, characterized in that: Use nonlinear equations or Bauschneske equations to perform high-resolution short-range tsunami numerical calculations to obtain refined tsunami forecast results for the first target area; The first designated range is less than or equal to 500km, and the second designated range is greater than 500km; When the epicenter is within the first specified range from the first target area, the spherical orthogonal grid resolution is 0.125 to 0.25 arc-min, and the forecast duration is 6 hours + Roundup (RETA Max ); Roundup means rounding up to an integer; RETA Max To extract the maximum propagation time in the target area from the medium-resolution tsunami propagation time forecast field in the potential impact area; When the distance between the epicenter and the first target area is greater than the first specified range and less than the second specified range, the spherical orthogonal grid resolution is also 0.125 to 0.25 arc-min, and the forecast duration is 6h+(Roundup(RETA Max )-Rounddown(RETA Min )); Rounddown means round down to an integer; RETA Max To extract the maximum propagation time in the target area from the medium-resolution tsunami propagation time forecast field in the potential impact area, RETA Min To extract the minimum propagation time in the target area from the medium-resolution tsunami propagation time forecast field in the potential impact area; The boundary driving data is the process data of the tsunami propagation calculation in the potential impact area to provide a time period for the first target area in [Rounddown (RETA Min ),6h+Roundup(RETA Max )] within the boundary conditions.

8. A global earthquake tsunami refined numerical prediction method according to claim 5, characterized in that: The S43 includes: The medium-resolution spherical orthogonal grid resolution in the potential tsunami impact area is 0.5 to 1 arc-min, and the forecast time is 6h + Roundup (ETA Max ), High-resolution short-range tsunami numerical calculations are performed using nonlinear equations or the Bausch-Nessker equation to obtain tsunami forecast results for areas potentially affected by tsunamis.

9. A computing device, characterized in that include: A memory, a CPU, a GPU and a bus, wherein the CPU, the GPU and the memory are all connected via the bus; The memory is used to store control programs and data, and the CPU and GPU cooperate to run the control program, wherein the control program executes a global earthquake and tsunami refined numerical prediction method as described in any one of claims 1 to 8 when running.