JP2016223956A - Early coastal tsunami prediction method using tsunami propagation characteristics - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an early coastal tsunami prediction method using tsunami propagation characteristics, which early predicts a tsunami waveform in the vicinity of a coast and the maximum tsunami water level at the coast right after the incident of the tsunami waveform in a sea area, by using a tsunami incident waveform observed in the sea area and tsunami propagation characteristics according to seabed topography previously prepared.SOLUTION: An early coastal tsunami prediction method using tsunami propagation characteristics includes: a first step (a) of predicting a tsunami waveform in the vicinity of a coast by using a tsunami incident waveform observed in a sea area and tsunami propagation characteristics according to seabed topography previously prepared; and a second step (b) of performing small-scale two-dimensional tsunami simulation including target points between the vicinity of the coast and the coast by using the tsunami waveform in the vicinity of the coast, which is predicted in the first step (a), so as to predict a coastal tsunami waveform and a coastal wave height.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a coastal early tsunami prediction method using tsunami propagation characteristics.

  Conventionally, no coastal early tsunami prediction method using tsunami propagation characteristics has been proposed.

Toshihiko Kanazawa: About the Japan Trench Submarine Earthquake Tsunami Observation Network, Earthquake Journal, Vol. 55, No. 6, pp. 28-33, 2013 Eiji Sugino, Nagatake Kure, Riko Korenaga, Shin Nemoto, Yoko Iwabuchi, Katsuzo Serizawa: Verification of the 2011 Tohoku Earthquake Tsunami at the Nuclear Site, Proceedings of the Japan Earthquake Engineering Society, 13, 2 (Special Issue), pp. 2-21 2013 Satake, K., Fujii, Y., Harada, T. and Namegaya, Y.:.... Time and Space Distribution of Coseismic Slip of the 2011 Tohoku Earthquake as Inferred from Tsunami Waveform Data, Bull Seismol Soc Am, 103, pp. 1473-1492, 2013 Cabinet Office: Nankai Trough Earthquake Review Committee, 12th meeting reference materials, http://www.bousai.go.jp/jishin/chubou/nankai_trou/12/sub_1.pdf, 2012 Japan Coast Guard: Digital undersea terrain data "M7000 series" Kenichi Hasegawa, Takao Suzuki, Kazuo Inagaki, Nobuo Shudo: Research on lattice spacing and time integration interval in tsunami numerical experiments, Journal of Japan Society of Civil Engineers, 381 / II-7, No. 2, pp. 111-120, 1987 Geographical Survey Institute: “5m mesh elevation data”, http://www.gsi.go.jp/ Okada, Y .: Internal Deformation due to Shear and Tensile Faults in a Half-Space, Bull. Seismol. Soc. Am., Vol. 82, No. 2, 1992. Imamura, F., Shuto, N. and Goto, C .: Numeric Simulations of the Transnational Propagation of Tsunamis, 6th Congress APD-IAHR, pp. 26 pp. 26 Tomoaki Goto: Tsunami Numerical Calculation, 1986 (22nd) Summer Workshop Lecture on Water Engineering, B Course, Japan Society of Civil Engineers, Hydraulic Committee, B-3-1 to B-3-21, 1986 Naufus: Ministry of Land, Infrastructure, Transport and Tourism Port Bureau National Port Ocean Wave Information Network, http://www.mlit.go.jp/kouwan/nowphas/ Ryo Fujiwara, Takahiro Tamiya, Mariko Korenaga, Norihiko Hashimoto: Evaluation of differences in tsunami wave power due to differences in tsunami wave source models, 2013 Seismological Society of Japan Fall Meeting, 2013 Yoshiro Ehara: Digital signal processing, Tokyo Denki University Press, p. 38, 1991 Isa Isamu: Simulation of an old tsunami off Sanriku, Bulletin of Earthquake Research Institute, 52, pp. 71-101, 1977

  In view of the above situation, the present invention uses the tsunami incident waveform observed in the sea area and the tsunami propagation characteristics by the seabed topography prepared in advance, so that the tsunami waveform near the coast immediately after the tsunami waveform incidence in the sea area and the coast The purpose of this project is to provide an early coastal tsunami prediction method using tsunami propagation characteristics, which predicts the maximum tsunami water level at an early stage.

In order to achieve the above object, the present invention provides
[1] In coastal early tsunami prediction method using tsunami propagation characteristics,
(A) a first step of predicting a tsunami waveform in the vicinity of the coast by using a tsunami incident waveform observed in the sea area and a tsunami propagation characteristic by a seabed topography prepared in advance; and (b) the first step. The second step of predicting the coastal tsunami waveform and wave height by performing a small two-dimensional tsunami simulation from the coastal region to the coastal target point using the coastal tsunami waveform predicted in (a). It is characterized by giving.

  [2] In the coastal early tsunami prediction method using the tsunami propagation characteristic described in [1] above, the tsunami propagation characteristic G2 (τ) of the first step is a simulation waveform S2 ( t) and the simulation waveform X2 (t) of the near-shore prediction point are calculated by deconvolution in the frequency domain, and the tsunami propagation characteristic G2 (τ) and the observed tsunami waveform S1 (t) at the sea area incident point for all fault slips It is characterized by convolution.

  [3] In the coastal early tsunami prediction method using the tsunami propagation characteristic described in [1] above, the second step is to perform coastal tsunami waveform and wave height prediction using a small-scale two-dimensional tsunami simulation. It is characterized by.

  [4] In the coastal early tsunami prediction method using the tsunami propagation characteristics described in [1] above, when using a tsunami incident waveform at a short distance and a long-time tsunami incident waveform, the coastal tsunami waveform and wave height It is expected to improve prediction accuracy.

  According to the present invention, the following effects can be achieved.

  By preparing the tsunami propagation characteristics of the sea area incident point and the coastal vicinity prediction point (water depth of 50m or more where the linear theory is established) in advance, the target when the tsunami incident waveform is recorded at the sea area after the tsunami occurs It is possible to immediately predict the tsunami waveform near the coastal area.

  In addition, a detailed two-dimensional tsunami simulation (less than 50m in depth using nonlinear theory) is carried out using the predicted tsunami waveform near the coast, thereby predicting the detailed tsunami height distribution on the coastline shoreline immediately. Can do.

It is a schematic diagram of the coastal early tsunami prediction method using the tsunami propagation characteristic of the present invention. It is a figure which shows the fault slip of the Tohoku-Pacific Ocean Earthquake. It is a figure which shows a seafloor topographic model. It is a figure which shows the comparison of the observed tsunami waveform in the offing of central Miyagi Prefecture and the calculated tsunami waveform using the linear long wave theory. It is a figure which shows the calculation procedure of the schematic diagram (FIG. 1) of the coastal early tsunami prediction method using the tsunami propagation characteristic of this invention. It is a figure which shows the evaluation point (● is an interval of about 10 km) and the main fault slip region (□) of the tsunami simulation according to the present invention. It is a figure which shows the tsunami waveform of a sea area incident point (846,895). It is a figure which shows the tsunami waveform of a coast vicinity prediction point (893). It is a figure which shows the comparison of the maximum tsunami water level rise amount in the Oshika Peninsula shoreline. It is a figure which shows the tsunami waveform of four coastal vicinity (944,919,893,866) estimated from the tsunami incident waveform of the sea area point 846. FIG. It is a figure which shows the error evaluation of the maximum tsunami water level of the coast vicinity prediction point 919 with respect to a different sea area incident point. It is a figure which shows the comparison of the maximum tsunami water level rise amount in the Oshika Peninsula shoreline with respect to a different sea area incident point. It is a figure which shows the error evaluation of the maximum tsunami water level of the coast vicinity prediction point 919 with respect to the data length from which a sea area incident waveform (point 846) differs. It is a figure which shows the comparison of the maximum tsunami water level rise amount in the Oshika Peninsula shoreline with respect to the data length from which a sea area incident waveform (point 846) differs.

  The coastal early tsunami prediction method using the tsunami propagation characteristics of the present invention includes (a) a tsunami waveform near the coast by using the tsunami incidence waveform observed in the sea area and the tsunami propagation characteristics prepared in advance in the seabed. And (b) a small-scale two-dimensional tsunami simulation including a coastal target point from the vicinity of the coast using the tsunami waveform near the coast predicted in the first step (a). By performing a second step of predicting the coastal tsunami waveform and wave height.

  Hereinafter, embodiments of the present invention will be described in detail.

  Under the experience of the devastating tsunami damage on the Pacific coast of eastern Japan caused by the 2011 off the Pacific coast of Tohoku Earthquake, the Japan Trench Submarine Earthquake Tsunami Observation is aimed at predicting appropriate earthquakes and tsunamis immediately after the earthquake that occurred in the sea area. A network (Seafloor Observing Network for Earthquakes and Tsunami miss the Japan Trench: S-net) (see Non-Patent Document 1 above) is being developed off the Pacific coast of East Japan. In the future, since tsunamis will be observed immediately after the occurrence of an earthquake in areas far from the coast, they will not be reduced to seafloor displacement corresponding to the epicenter area or magnitude, and tsunami waveforms along the coast using tsunami records observed in the sea area.・ Development of a direct prediction method for wave height is expected.

  Therefore, in the present invention, by using the tsunami incident waveform observed in the sea area and the tsunami propagation characteristics by the seabed topography prepared in advance, the tsunami waveform near the coast immediately after the tsunami waveform incidence in the sea area and the coastal A method for predicting tsunami wave height was developed.

  FIG. 1 is a schematic diagram of a coastal early tsunami prediction method using the tsunami propagation characteristics of the present invention.

Specifically, the coastal instantaneous tsunami waveform and wave height are predicted by the two steps shown in FIG.
(1) Predict the tsunami waveform 5 in the vicinity of the coast 4 by using the tsunami incident waveform 3 observed in the sea area 2 and the tsunami propagation characteristics of the seabed topography prepared in advance by the crustal deformation at the seismogenic fault plane 1 (1st step: direct prediction using tsunami propagation characteristics).
(2) By using the tsunami waveform 5 of the coastal vicinity 4 predicted in the 1st step, a small-scale two-dimensional tsunami simulation 7 including the target points from the coastal vicinity 4 to the coastal 6 is performed. The wave height 9 is predicted (2nd step: small-scale numerical calculation).

Details of the 1st step and the 2nd step will be described below.
(Waveform prediction near the coast using tsunami propagation characteristics)
(1) Two-dimensional tsunami simulation of the 2011 Tohoku-Pacific Ocean Earthquake For the tsunami source model of the 2011 Tohoku-Pacific Ocean Earthquake, several models (the above non-patent documents [2], [3], (Refer to [4]). However, the difference in the linear tsunami waveform between each wave source model is small, and the observation record can be generally explained from the two-dimensional tsunami simulation result using any wave source model.

  FIG. 2 is a diagram showing a fault slip of the Tohoku-Pacific Ocean Earthquake (see Non-Patent Document [4] above), and □ indicates a main fault slip region.

  Here, the tsunami wave source model of the Cabinet Office in which the wave source model is expressed in the most detail (see the non-patent document [4] above) is used.

  FIG. 3 is a diagram showing a seafloor topographic model (see Non-Patent Document [5] above) (lattice interval 1215 m, contour line interval 500 m).

  In the present invention, as shown in FIG. 3, a two-dimensional tsunami simulation is performed by a finite difference method, and the mesh size (lattice interval) is set to be gradually reduced from the sea area toward the inland (4 stages: 1215 m, 405 m, 135 m, 45 m lattice spacing). About 20 gratings were arranged for one wavelength of the tsunami (see Non-Patent Document [6] above), and the time step was set to 0.81 seconds in the linear theory and 0.09 seconds in the nonlinear theory. A topographic model was created by assigning water depth and elevation data to the grid centers of each region. For the water depth data, please refer to the Japan Coast Guard digital data “M7000 series” (see the above non-patent document [5]). For the elevation data, refer to the “5m mesh elevation data” from the Geospatial Information Authority of Japan (see the above non-patent document [7]) ) Was used. The initial water level of the tsunami was calculated from the crustal deformation of the vertical component in the semi-infinite elastic body (see Non-Patent Document [8] above). The nonlinear shallow water long wave theoretical formula used in the tsunami simulation (see the non-patent document [9]) is shown below.

  Here, x · y horizontal coordinate, z: vertical coordinate, t: time, g: gravitational acceleration, η: water level, h: water depth, MN: flow rate, D = (h + η): total water depth. .

  From the two-dimensional tsunami simulation using the linear long wave theory, the tsunami waveform from the sea area to the coastal area (water depth of about 50 m) (see Non-Patent Document [10] above) was calculated. In the linear analysis, the advection terms (second term and third term) and the seabed friction term (fifth term) are not considered in the above equations (2) and (3). In a linear wave field, the tsunami propagates at a depth-dependent wave velocity expressed as c = √gh. On the other hand, the coastal tsunami waveform was calculated from the vicinity of the coast by two-dimensional tsunami simulation using nonlinear shallow water long wave theory. In a nonlinear wave field, the tsunami propagates at a wave speed expressed by c = √g (h + η).

  FIG. 4 is a diagram showing a comparison between the observed tsunami waveform off the central part of Miyagi Prefecture and the calculated tsunami waveform using the linear long wave theory.

As an example, FIG. 4 compares a tsunami waveform calculated off the central part of Miyagi Prefecture with a tsunami waveform observed by a GPS wave meter (see Non-Patent Document [11] above) (Ministry of Land, Infrastructure, Transport and Tourism) Reference [12]). Since the calculated tsunami waveform reproduces the observed waveform well, the validity of the two-dimensional tsunami simulation used in the present invention can be understood.
(2) Prediction of the vicinity of the coast by the tsunami incident waveform and the tsunami propagation characteristics In the first step of the present invention, the tsunami incidence waveform observed in the sea area and the tsunami propagation characteristics by the seafloor topography prepared in advance are used. Develop a method for predicting the tsunami waveform. As the tsunami incident waveform, the two-dimensional tsunami simulation result by the all fault slip of the 2011 Tohoku-Pacific Ocean Earthquake was used. Regarding the tsunami propagation characteristics due to the seafloor topography, it is necessary to extract the tsunami propagation characteristics of the sea area incident point and the coastal vicinity prediction point {FIGS. 2 and 6 (described later)} for the smallest possible fault region. Therefore, a two-dimensional tsunami simulation based on the major fault slip of the 2011 Tohoku-Pacific Ocean Earthquake was performed based on the linear theory described above, and the tsunami propagation characteristics at the sea area incident point and the near coastal prediction point were prepared in advance using the result. The calculation procedure of the present invention is shown in FIG.

  FIG. 5 is a diagram showing a calculation procedure of a schematic diagram (FIG. 1) of a coastal early tsunami prediction method using the tsunami propagation characteristics of the present invention, and FIG. 6 is an evaluation point (● is an interval of about 10 km) according to the present invention. ) And the main fault slip region (□).

  In the verification of the present invention, the two-dimensional tsunami simulation result using all fault slips was regarded as a correct answer, and the comparison result of the calculation result according to the present invention was used.

  The tsunami propagation characteristic G2 (τ) was calculated by deconvolution in the frequency domain of the simulation waveform S2 (t) at the sea area incident point and the simulation waveform X2 (t) at the near-shore prediction point for the main fault slip (formula [4 ]). Convolution of the tsunami propagation characteristic G2 (τ) and the simulation waveform S1 (t) at the sea area incident point for all fault slips (equation [5]) allows the tsunami waveform Y (t) at the coastal vicinity prediction point to be easily and It can be calculated immediately.

G2 (ω) = X2 (ω) / S2 (ω) (4)
Y (ω) = S1 (ω) × G2 (ω) (5)
The incident data, output waveform, and tsunami propagation function were subjected to the following data processing (see reference [13]).
・ Low-pass filter (180 second time window) by moving average is applied to the input waveform, the data length is shortened to about 48 minutes including the main motion to exclude the following wave, and then cosine taper is applied. .
-The high frequency signal of 7 minutes or less was cut off from the following formula.

M = 0.443 × fs / fc (6)
Here, fs: sampling frequency (1 Hz), M: number of samples, fc: cutoff frequency.
・ The output composite waveform and tsunami transfer function were low-pass filtered by moving average. The time window for the moving average was 60 seconds for the synthesized waveform and 200 seconds for the tsunami propagation function.

  FIG. 7 is a diagram showing the tsunami waveform at the sea area incident point (846, 895), and FIG. 8 is a diagram showing the tsunami waveform at the coastal vicinity predicted point (893).

  Fig. 7 shows two tsunami waveforms (simulated from the observed tsunami incident waveform in the present invention) calculated at the sea area incident points 846 and 895, calculated by a two-dimensional tsunami simulation using the whole fault slip of the Tohoku Earthquake. Show. Although no significant difference is observed in the maximum amplitude, it can be seen that the time when the maximum amplitude appears is different. On the other hand, regarding the predicted waveform of the coastal vicinity point 893 calculated by the 1st step of the present invention (see FIG. 8), there is a significant difference in the maximum amplitude appearance time of the two predicted waveforms using the sea area incident points 846 and 895. However, this is due to the fact that the tsunami propagation characteristics at the sea area incident point and the coastal vicinity prediction point are appropriately evaluated. As for the entire waveform, the one where the coastal vicinity prediction point and the sea area incident point are located at a short distance (sea area incident point 895) is highly consistent with the prediction result, and when the tsunami approaches the coast, It is suggested that the prediction accuracy of the present invention is improved.

As a result of the present invention, for the tsunami generated by the 2011 off the Pacific coast of Tohoku Earthquake, if there is a sea area incident point and a coastal near prediction point on a certain survey line connecting the main fault slip region and inland, This shows that the tsunami amplitude and phase can be predicted easily and appropriately.
(Coastal wave height prediction using small-scale tsunami simulation)
FIG. 9 is a diagram showing a comparison of the maximum amount of tsunami water level rise along the Oshika Peninsula Minato Line. FIG. 9A is a diagram showing a calculation region of a small-scale two-dimensional tsunami simulation, and FIG. The amount of tsunami water level rise. FIG. 10 is a diagram showing a tsunami waveform at four points near the coast (944, 919, 893, 866) predicted from the tsunami incident waveform at the sea area point 846.

The coastal maximum tsunami wave height was predicted in the Oshika Peninsula Line (bold line in FIG. 9B). In the 2nd step shown in FIGS. 1 and 5, the predicted tsunami waveform at the coastal vicinity calculated in 1st step is incident on the east boundary surface (dotted line in FIG. 9). Since the predicted waveforms at the four points are approximately the same in both amplitude and phase, the predicted tsunami waveform at the coastal neighboring point 919 calculated at the 1st step as the incident waveform on the eastern boundary surface of 2nd step (dotted line in FIG. 9). Was used. The boundary areas on the north, south, and west sides of the analysis area were set as open boundaries. No tsunami was incident on these three boundaries, and the tsunami wave field excited by the incident wave was free to flow into and out of the boundary on the east side. This tsunami simulation was performed at a grid interval of 45 m and a time increment of 0.09 seconds based on the nonlinear shallow water long wave theory.
The maximum tsunami water level rise along the Oshika Peninsula shoreline was calculated from a small-scale two-dimensional tsunami simulation using the predicted tsunami waveform at the coastal point as the incident waveform (the predicted tsunami waveform from the sea area incident point 846). The result is shown in FIG. Compared with the two-dimensional tsunami simulation result using the whole fault slip (correct answer in the present invention), the result of the present invention shows a difference in the maximum tsunami water level rise, but the shape is almost consistent and the spatial distribution tends to be You can see that it is generally captured. The low consistency in the vicinity of the north boundary is because the contribution of a tsunami that is diffracted from outside the analysis range is not considered in the conditions of the present invention (incident only on the east boundary). Geometric mean K and geometric standard deviation k (K and k are 1 perfectly harmonized) are used as an index (see Non-Patent Document [14] above) indicating the spatial suitability between the tsunami trace height and the calculated value from a fault model. Yes, for reference, the values obtained in this case are K = 1.6 and k = 2.3. The present invention uses a tsunami waveform incident in the sea area to predict a maximum tsunami water level rise of more than 10 m along the coast, and has high utility value as an immediate coastal tsunami height prediction.
(Tsunami estimation accuracy)
(1) Examination of accuracy by tsunami incident distance Fig. 11 shows an error evaluation of the maximum tsunami water level rise at the near-shore prediction point 919 for different sea area incident points, and Fig. 12 shows the maximum tsunami at the Oshika Peninsula shoreline for different sea area incident points. It is a figure which shows the comparison of a water level raise amount.

For different sea area incidence points, we compared the maximum tsunami water level rise at the coastal vicinity prediction point and the Oshika Peninsula Minato Line, and examined the prediction accuracy by the tsunami incidence distance. The maximum tsunami water level rise at the coastal vicinity prediction point 919 (result of the present invention by 1st step) tends to increase as the distance increases, but within 50% even when tsunami incidence is performed at a sea area point of 100 km or more Can be predicted with the accuracy of. Moreover, the same tendency was recognized also with respect to the maximum amount of tsunami water level rise (result of this invention by 2nd step) in the Oshika Peninsula shoreline. If the point 20km away from the coastal vicinity prediction point is the sea area tsunami incidence, it can be predicted with an accuracy of about 20%, and the closer the distance between the coastal vicinity prediction point and the sea area incident point, the better the accuracy of the tsunami prediction result Confirmed to do.
(2) Examination of accuracy based on tsunami incident time Figure 13 shows an error evaluation of the maximum tsunami water level rise at the coastal vicinity prediction point 919 for different data lengths of the sea area incident waveform (point 846), and Fig. 14 shows the sea area incident waveform (point) It is a figure which shows the comparison of the maximum tsunami water level rise amount in the Oshika Peninsula shore line with respect to different data length of 846).

  For the data length of the sea area incident waveform, the maximum tsunami water level rise at the coastal vicinity prediction point and the Oshika Peninsula shore line was compared and the prediction accuracy by the tsunami incident time was examined. Although the error tends to increase as the data length is shorter, the maximum amount of tsunami water level can be predicted with the same degree of accuracy as when all the waveforms are used if a major tsunami motion of about 15 minutes (see Fig. 7) is incident. It was confirmed.

  As a feature or advantage of the present invention, by preparing tsunami propagation characteristics at the sea area incident point and the coastal vicinity prediction point (water depth of 50 m or more where linear theory is established) in advance, the tsunami incident at the sea area after the tsunami occurs When the waveform is recorded, it is possible to immediately predict the tsunami waveform at the target coastal point. In addition, a detailed two-dimensional tsunami simulation (less than 50m in depth using nonlinear theory) is carried out using the predicted tsunami waveform near the coast, thereby predicting the detailed tsunami height distribution on the coastline shoreline immediately. Can do. Actually, the time taken to calculate the tsunami wave height of the coastline shoreline from the point near the coast (2nd step) according to the present invention is about a time history waveform output of about 20 minutes with a 45 m grid interval. Used for about 20 minutes. If the number of cores increases to about 10 times the current number (80 to 100 cores), the calculation time required for a small-scale two-dimensional tsunami simulation is expected to be dramatically reduced. In addition, because the coast of the Oshika Peninsula, which is the tsunami prediction point, has a complex topography, it was necessary to calculate with detailed grid spacing, but in the Sendai Plain, where the coastline is relatively uniform, It is important to devise a method to shorten the calculation time based on considerations such as roughening the time when making an immediate prediction (if possible, within 5 minutes after tsunami detection).

  In the present invention, a tsunami waveform prediction method in the vicinity of the coast using tsunami propagation characteristics and a coastal tsunami wave prediction method using a small-scale tsunami simulation have been developed as an immediate prediction method of the tsunami waveform and wave height. The effectiveness of the present invention was confirmed by comparing with the results of a two-dimensional tsunami simulation caused by the 2011 Tohoku-Pacific Ocean Earthquake (Mw 9.0). Also, the accuracy of tsunami estimation with respect to the tsunami incidence distance and time is examined, and it is shown that immediate prediction can be made with accuracy within 50% even when using a tsunami incident waveform with a distance of 100 km and half the data length. It was. Therefore, the immediate prediction of the tsunami waveform and the tsunami wave height in the vicinity of the coast and the coastline shoreline according to the present invention is very effective in grasping the margin time when taking evacuation action and the extent of the tsunami wave height.

  In addition, this invention is not limited to the said Example, Based on the meaning of this invention, a various deformation | transformation is possible and these are not excluded from the scope of the present invention.

  The coastal early tsunami prediction method using the tsunami propagation characteristics of the present invention uses the tsunami incident waveform observed in the sea area and the tsunami propagation characteristics based on the seafloor topography prepared in advance, immediately after the tsunami waveform incidence in the sea area. It can be used as an early tsunami prediction method for early prediction of the tsunami waveform near the coast and the maximum tsunami water level on the coast.

1 Seismic fault plane 2 Sea area incident point 3 Sea area tsunami incident waveform 4 Coastal point 5 Coastal predicted tsunami waveform 6 Coastal point 7 Coastal to coastal area 8 Coastal predicted tsunami waveform 9 Coastal predicted tsunami height

Claims (4)

(A) a first step of predicting a tsunami waveform in the vicinity of the coast by utilizing the tsunami incident waveform observed in the sea area and the tsunami propagation characteristics of the seabed topography prepared in advance;
(B) Using the tsunami waveform near the coast predicted in the first step (a), a small-scale two-dimensional tsunami simulation including the target point from the coast to the coast is performed. An early coastal tsunami prediction method using tsunami propagation characteristics, characterized by performing a second step of predicting wave height.   2. The coastal early tsunami prediction method using the tsunami propagation characteristic according to claim 1, wherein the tsunami propagation characteristic G2 (τ) of the first step is a simulation waveform S2 (t) of a sea area incident point with respect to a small fault slip and a coastal The simulation waveform X2 (t) at the vicinity prediction point is calculated by deconvolution in the frequency domain, and the tsunami propagation characteristic G2 (τ) and the observed tsunami waveform S1 (t) at the sea area incident point for all fault slips are convolved. An early coastal tsunami prediction method using tsunami propagation characteristics characterized by this.   2. The coastal early tsunami prediction method using tsunami propagation characteristics according to claim 1, wherein the second step performs coastal tsunami waveform and wave height prediction using a small-scale two-dimensional tsunami simulation. An early coastal tsunami prediction method using tsunami propagation characteristics.   In the coastal early tsunami prediction method using the tsunami propagation characteristic according to claim 1, when the tsunami incident waveform at a short distance and the tsunami incident waveform for a long time are used, the prediction accuracy of the coastal tsunami waveform and wave height is improved. A coastal early tsunami prediction method using tsunami propagation characteristics characterized by expectations.

Images