JP5816979B2 - Tsunami detection method, detection device, and detection program - Google Patents

Description

  The present invention relates to tsunami detection based on measurement of magnetic field fluctuations on land.

  Conventionally, tsunami observation methods have been known to detect and alert tsunamis using tide gauges and pressure sensors. However, for example, when a huge tsunami comes in, the tide gauge needle is not fully shaken. Or have failed (see Non-Patent Document 1). In addition, a pressure sensor was installed offshore, and information was transmitted to the ground via a submarine cable. However, if the sensor failed, maintenance was difficult.

Also, in these methods, the installation position of sensors and measuring units is restricted to the sea floor and the sea surface, which causes high installation costs and maintenance costs, damage due to collision with the ship and deterioration due to solar rays, There were problems such as the fact that a small wave or an instantaneous wave due to a gust of wind was detected as noise and the process of distinguishing it from a tsunami became complicated.

Therefore, the swell wave generated during the occurrence of a tsunami or high wave causes an eddy current to flow along the sea surface, generating a sinusoidal periodic magnetic field in a direction perpendicular to the eddy current. There is a tsunami detection method that makes it possible to easily identify the tsunami while storing it and making it easy to install it on the sea using dredging, etc., without installing the device on the sea floor or sea surface. It has been proposed (see Patent Document 1).
However, this detection method has not yet sufficiently obtained accurate tsunami information in real time, and has not yet reliably prevented tsunami damage.

JP 2009-198207 A

Earth Planets Space, 63, 0-0, 2011 “Significant tsunami observed at the ocean-bottom pressure gauges at2011 Off the Pacific Coast of Tohoku Earthquake”

  The problem to be solved by the present invention is to detect the occurrence of a tsunami accurately because the needle does not shake out or break down even when a huge tsunami is approaching, and maintenance is easy. .

The present invention is basically an invention for detecting the occurrence of a tsunami by measuring the fluid movement of a large block of seawater that is electrically conductive, that is, the slight fluctuation of a magnetic field generated by an electric current (the following steps) A method for detecting a tsunami including a program, a program for detecting a tsunami, and an apparatus for executing the program).
Specifically, the present invention provides:
A program used in an apparatus including a storage device and an arithmetic processing device to detect a tsunami,
Measuring magnetic field over time at land measurement points to understand magnetic field fluctuations,
A step of obtaining an average value of the magnetic field measured for a predetermined time from the current time, and comparing the average value with the value of the current magnetic field to obtain a difference between the current magnetic field value and the average value, and the difference is An executable instruction for causing the apparatus to execute a step of determining that a tsunami has occurred when a predetermined value recognized as a normal magnetic field fluctuation calculated from a magnetic field fluctuation is continuously exceeded for a predetermined time; The program.
There are a plurality of measurement points. The magnetic field fluctuation is grasped on the basis of the difference signal obtained by obtaining a difference signal by obtaining a magnetic field difference between an arbitrary point in the measurement point and another point. The average value is obtained for the difference signal. Thus, the magnetic field is measured at a plurality of points on the land in order to remove the influence of turbulence of geomagnetism and the like.
An apparatus including a storage device including the program and an arithmetic processing device includes a magnetic field measuring device for measuring a magnetic field, an output device for displaying a result calculated by the arithmetic processing device, and generation of a tsunami. And a display device that displays the occurrence of a tsunami when detected. In addition, this apparatus can further include a system that sounds an alarm upon detection of the occurrence of a tsunami and sends a breaking news to the media.
In addition, the flowchart of the tsunami detection using the above-mentioned program was shown in FIG.

  The magnetic field is obtained based on one component, two components, or three components of the three components of the magnetic field orthogonal to each other.

  The predetermined value is a constant multiple (kσ: k is a constant (for example, 2 to 4)) of a standard deviation (σ) of the difference from a certain past to the present.

  According to the present invention, tsunami can be detected on land, and maintenance of the observation apparatus is simplified. Moreover, since the occurrence of a tsunami can be observed in real time by measuring the magnetic field change, an accurate and accurate tsunami prediction can be made immediately.

It is the schematic which shows the observation point of a magnetic field fluctuation condition. It is a waveform chart figure which shows the observation result in each point. It is an example of magnetic field fluctuation observation (total magnetic force F) at the Hosokura observation point when a tsunami occurs. Specifically, a chart of changes in the total magnetic force (the square root of the sum of the squares of the three components) in Hosokura is shown. The observation date and time is 2 hours from 14:00 to 16:00 on March 11, 2011 when the earthquake (tsunami) occurred. The solid line is the observed waveform, the broken line is the moving average value (M), and the two-dot chain line is M ± 3σ (σ is the standard deviation). It is a chart which shows the change of perpendicular magnetic force (Z component) in Hosokura at the time of tsunami generation. The observation date and time is 2 hours from 14:00 to 16:00 on March 11, 2011 when the earthquake (tsunami) occurred. The solid line is the observed waveform, the broken line is the moving average value (M), and the two-dot chain line is M ± 3σ. It is a chart which shows the difference of the total magnetic force of Hosokura and Memanbetsu at the time of tsunami occurrence. The observation date and time is 2 hours from 14:00 to 16:00 on March 11, 2011 when the earthquake (tsunami) occurred. The solid line is the observed waveform, the broken line is the moving average value (M), and the two-dot chain line is M ± 3σ. It is a chart which shows the difference of the perpendicular magnetic force (Z component) of Hosokura and Memanbetsu at the time of tsunami generation. The observation date and time is 2 hours from 14:00 to 16:00 on March 11, 2011 when the earthquake (tsunami) occurred. The solid line is the observed waveform, the broken line is the moving average value (M), and the two-dot chain line is M ± 3σ. It is a correlation diagram of the tsunami detection and the water level meter of Soma in the total magnetic difference signal change of Hosokura and Memanbetsu at the time of tsunami occurrence. It is a flowchart of the tsunami detection using the above-mentioned program.

As shown in FIG. 1, the present inventor has provided an observation point in Hosokura (HSK) in Miyagi Prefecture and continuously observes the geomagnetic field.
At the Hosokura observation point, three-component geomagnetic observation has been conducted since March 2004 using a pair of fluxgate magnetometers at the Hosokura Mine in Kurihara City, Miyagi Prefecture. One is installed in a tunnel 70m underground and the other is installed 1m below the surface. The underground magnetometer records 0.5 seconds and the ground magnetometer records 1 second. Both observation data are time synchronized by GPS radio waves.

  In addition, each recording system uses photovoltaic power generation or a battery as a power supply source, and is independent of the commercial power system, so there is almost no influence such as a power failure accompanying an earthquake.

  The 2011 Tohoku-Pacific Ocean Earthquake (Moment Magnitude (Mw) 9.0) occurred on March 11, 2011 at 14:46:18 (Japan time). The distance from the Hosokura observation point to the epicenter is about 100 km.

FIG. 2 shows the geomagnetic field signal at the time of the earthquake at each observation point. The geomagnetic data of Kamaoka (KAK) in Ishioka City, Ibaraki Prefecture and Memanbetsu (MMB), Hokkaido are provided by the Japan Meteorological Agency Geomagnetic Observatory, and Okutama (OKM) in Tokyo is provided by the Tokai University Earthquake Prediction Research Center. It is. Earthquake time is indicated by T ₀ of FIG. A few minutes after the occurrence of the earthquake, the earthquake should have arrived at each observation point, but the magnetic field has not changed significantly. However, referring to the magnetic field signal at the Hosokura observation station, remarkable magnetic field changes were observed from the earthquake to the next 10 to 18 minutes due to the huge earthquake and the subsequent tsunami.

  For the measurement of geomagnetism, three-component observation is performed. Accordingly, attention may be paid to only one of the three components (for example, the perpendicular magnetic force component: Z component) or two components (for example, one of the vertical component Z and the horizontal component), or all three components. You may pay attention to.

  Anomalies in magnetic field fluctuations can be observed just by observing the observed data, but the average value (M) of the magnetic field from a given time before the observation time is obtained, and the average value and the current magnetic field are calculated. By comparing, the tsunami can be detected more objectively.

  FIG. 3 is data of the total magnetic force (the square root of the sum of the squares of the three components) at the Hosokura observation point on the date of the occurrence of the huge earthquake. In the figure, the solid line is the observed waveform of the total magnetic force, the broken line is the moving average value (represented by M) for 20 minutes, and the two-dot chain line is M ± 3σ. FIG. 3 shows that 3σ is too large and should be about 2σ as a threshold for determining a tsunami in a signal from only one station.

FIG. 4 is data of perpendicular magnetic strength (Z component) at the Hosokura observation point on the date of the earthquake.
Also in this case, similarly to the total magnetic force, 3σ is too large as a threshold for determining a tsunami in a signal from only one observation station, and should be about 2σ.

  As described above, the occurrence of a tsunami can be detected from data from only one station, but there is a better method. In other words, to eliminate noise caused by causes other than the tsunami, such as geomagnetic disturbances, common to multiple observation points, the difference signal of the magnetic field change observed at the observation point close to the tsunami and the observation point far from the tsunami is used. By doing so, it became possible to detect the magnetic field change based on the tsunami with high sensitivity.

  FIG. 5 is a graph of the difference signal of the total magnetic force between the observation data of Hosokura and Memanbetsu. A solid line is an observed waveform, a broken line is a moving average value for 30 minutes, and a two-dot chain line is M ± 3σ. The difference signal usually shifts almost without deviating from the average value, but it was confirmed that it deviated greatly when a huge tsunami occurred. By detecting this, it is possible to detect a tsunami using a difference signal of magnetic field changes at multiple observation points.

  FIG. 6 is a graph of the difference signal of the vertical magnetic field Z of the observation data for Hosokura and Memanbetsu. A solid line is an observed waveform, a broken line is a moving average value for 30 minutes, and a two-dot chain line is M ± 3σ. As in FIG. 5, the difference signal usually shifts with little deviation from the average value, but it was confirmed that the difference signal greatly deviated when a huge tsunami occurred. In this way, by detecting the Z component, which is one of the three components, tsunami detection using the difference signal of the magnetic field change at the multi-observation point is possible.

When this is expressed by a mathematical expression, if the error signal at time t is D (t), the average value M of error signals from a predetermined time to the present time is
M = {D (t−nT) + D (t− (n−1) T) +... + D (t−T)} / n
(Tamping interval is T). The standard deviation σ is
σ ² = {(D (t−nT) −M) ² +... + (D (t−T) −M) ² } / n
It is requested from.
Therefore, the present invention
Absolute value (D (t) -M)> 3σ
Or absolute value (D (t) −M)> kσ (k is an arbitrary constant (for example, 2 to 4))
It is determined that a tsunami has occurred at (t).
The above n is a discrete time from 10 minutes to 30 minutes. In other words, if the sampling interval T is 1 second, n = 600 to 1800, and if T is 2 seconds, n = 300 to 900.

  The sampling interval T may be about 0.5 second to 2 seconds. Since the standard deviation at the normal time is about 1 nT, specifically, for example, it may be 2 nT, but it is preferable that the size to be compared is determined based on the standard deviation. In the above example, it is 3σ.

  Further, when detecting a tsunami, it is preferable that the reference value (threshold value) determined as kσ is continuously exceeded for a predetermined time (for example, about 30 seconds). This is because when a geomagnetic observation is performed, a pulse having a very sharp peak is often observed. It is a well-known story in geomagnetic observation, and the cause is not clear, but it is said to be a car, a human being, or a bird.

  FIG. 7 shows a graph of the difference signal of the total magnetic force of the observation data of Hosokura and Memanbetsu and a graph of the water level meter in Soma in time synchronization. Soma is relatively close to the epicenter and the water level has reached 10m. As can be seen from FIG. 7, the tsunami occurred around 15:00 and reached Soma before 16:00. Therefore, it is possible to issue a tsunami warning 50 minutes before the arrival of the tsunami.

As is clear from the above, it is possible to detect a tsunami using a difference signal of magnetic field changes at multiple observation points. Moreover, it can be seen that a tsunami occurred near an area where a signal greater than a predetermined difference between observation points was generated. As described above, the present invention enables real-time observation in that the magnetic field difference signal is used.

  The present inventor succeeded in observing magnetic field fluctuations associated with a giant tsunami in the Tohoku-Pacific Ocean Earthquake (East Japan Earthquake) that occurred on March 11, 2011 at an observation point created in the Hosokura Mine. This is a breakthrough for seismic electromagnetic observation, and based on this result, for example, an emergency warning for a huge tsunami becomes possible.

MMB Memanbetsu (Hokkaido)
HSK Hosokura (Miyagi Prefecture)
KAK Kashioka (Ibaraki Prefecture)
OKM Okutama (Tokyo)
ESS Esashi (Iwate Prefecture)
Aftershock Area Aftershock Area
Tohoku EQ Tohoku Earthquake
Tsunami

Claims (3)

A method for detecting a tsunami before the tsunami reaches the land,
Measuring magnetic field over time at land measurement points to understand magnetic field fluctuations,
A step of obtaining an average value of the magnetic field measured for a predetermined time from the current time, and comparing the average value with the value of the current magnetic field to obtain a difference between the current magnetic field value and the average value, and the difference is A step of determining that a tsunami has occurred when a predetermined value recognized as a normal magnetic field fluctuation calculated from the magnetic field fluctuation is continuously exceeded for a predetermined time,
The measurement points are a plurality of locations,
The magnetic field variation is obtained based on a difference signal obtained by obtaining a difference signal by obtaining a magnetic field difference between an arbitrary point in the measurement point and another point,
The average value is determined for the difference signal;
Said method. A program used in an apparatus including a storage device and an arithmetic processing device to detect a tsunami,
Measuring magnetic field over time at land measurement points to understand magnetic field fluctuations,
A step of obtaining an average value of the magnetic field measured for a predetermined time from the current time, and comparing the average value with the value of the current magnetic field to obtain a difference between the current magnetic field value and the average value, and the difference is An executable instruction for causing the apparatus to execute a step of determining that a tsunami has occurred when a predetermined value recognized as a normal magnetic field fluctuation calculated from the magnetic field fluctuation is continuously exceeded for a predetermined time;
The measurement points are a plurality of locations,
The magnetic field variation is obtained based on a difference signal obtained by obtaining a difference signal by obtaining a magnetic field difference between an arbitrary point in the measurement point and another point,
The average value is determined for the difference signal;
The program. An apparatus for executing the program according to claim 2, comprising a storage device and an arithmetic processing device,
A magnetic field measuring instrument for measuring the magnetic field;
An output device for displaying a result calculated by the arithmetic processing device;
An apparatus for detecting a tsunami, comprising a display device that displays the occurrence of a tsunami when the occurrence of a tsunami is detected.