Seismic activity changes in capital region since 3.11


Last updated: 28th February 2012

Seismic activity changes in capital region since 3.11.

  • The estimation below was announced in September 2011, at DANWA-KAI(monthly meeting in ERI and is open to public). At that time, it was already covered by the Medias. Since then, there’s no change in phenomenon nor did we calculate something new.
  • As been announced since before, written on this website does not mean that it is an official opinion of the Earthquake Research Institute. Everything on this website is just a result and achievement of the analysis gained by the individual researchers.
  • What measures can we take for disaster mitigation

(Research content and figures by: Associate Prof. Shinichi Sakai)

Detailed description is made according to the following: “Introduction”, “Method”, “Seismic activity in capital region, before and after 3.11.”. “The margin of error in provisional calculation”,“The Reason why the number is different from what the government is announcing “70% in next 30 years””, “What measures can we take for disaster mitigation?”, FAQ

This website was created for the demand of detailed description by public and media (created on 23rd January 22:00). This website had been used from before to meet such demand.



Small earthquakes occur more frequent than large ones. Smaller the earthquake is, the more likely to occur. This is called the “Gutenberg-Richter’s law”.  For example, in Japan, it is known that magnitude (M) 3 earthquake occurs 10,000 times per year (i.e. once in an hour), M4 earthquake, 1000 times per year (i.e. 3times in a day), M5, 100 times per year (i.e. once in 3days), M6, 10 times per year (i.e. once a month).

On the other hand, many aftershocks will occur when large earthquake occurs. Longer the time after the main shock, smaller the number of aftershocks become. Numerical expression for this, is the “modified Omori formula”. Earthquake Research Committee made an “Evaluation Method of the Aftershock Probability” combining the “Gutenberg-Richter’s law” and the “modified Omori formula”.

The range of application of this method is stated clearly, as “narrowly-defined aftershock”, but Associate professor Sakai et al. considered that the seismic activity in Tokyo region triggered by Tohoku earthquake is also aftershock in a broad sense, and thought that this method is applicable. Therefore, in September, they calculated the probability of M7 triggered earthquake using this method.


Seismic activity in capital region, before and after 3.11.

The left figure below is the earthquake distribution above M3, before 3.11(11th September to 10th March 2011), and the right figure is after the 3.11 (11th March to 10th September 2011). (JMA centralized database)

Left: Distribution of M3 and above seismic activity from 11th September 2010 to 10th March 2011. Right: Distribution of M3 and above seismic activity from 11th March to 11th September 2011. The number is increasing from 47 to 343.


In this “seismic-activity after 3.11″, both usual seismic-activity before 3.11 and seismic-activity triggered by 3.11 are included. because the number of seismic-activity triggered by 3.11 is far more than the usual seismic-activity, it was analysed as triggered seismic-activity. First, the b value of the “Gutenberg-Richter’s law” and the p value of the “modified Omori formula” is estimated. Calculating the probability of M7 (in accuracy M6.7-7.2) by substituting  these estimation in to the  formula, the probability was 98% in the next 30 years. In other word, 70% in the next 4years, by finding out how long it takes to reach upto 70%.


The margin of error in provisional calculation

The margin of error in the above provisional calculation is as follows:

  1. Margin of error caused by implementing the triggered seismic activity to ‘Evaluation Method of the aftershock probability’ which should be for narrowly- defined aftershock.
  2. The margin of error observed when estimating the parameter b value and p value

The precise margin of error in no.1 is not clear, but it is considered to be quite large, for it is used out of the applicable range specified in the prediction method. Also, as shown below, the margin of error for no.2 is also large:

In the provisional calculation indicated above, the b value was estimated 0.67. However, by including the data up to December 2011, the b value changed drastically to 0.78. And in the same way, the estimated value for p also became large. ‘Evaluation Method of the aftershock probability’ includes improved Omori formula. Therefore, the occurrence probability shouldn’t change as long as you don’t change the initial date of reckoning. But because of the changes in such parameter value, the provisional calculation result for probability in 30 years has changed from 98% to 83%.

This means, if both values are added, the impact should be so large, since the margin of error in result of provisional calculation in no.2 was already quite large. Therefore, please consider that the number such as 98% in 30yeras or 70% in 4 year, has no meaning in itself. However, as being announced by government and researchers, the fact that the Tokyo region earthquake is impending, is true. Now is the time for you to be prepared.


The Reason why the number is different from what the government is announcing “70% in the next 30 years”.

As you might have heard, The Headquarters for Earthquake research Promotion, in the Ministry of Education, Culture, Sports, Science and Technology, had been announcing that the probability of M7 earthquake in southern-Kanto, is “70% in the next 30yeras”. This is different from the provisional calculations: “98% in the next 30years (or 70% in the next 4years)”, derived in this research. This is because of the difference in the coverage of provisional calculation and evaluation.
According to the calculation by the government, they counted the number of M6.7-7.2 seismic activity in the past 150 years, and calculated the probability from its frequency.

This means that they are calculating the probability by counting only the earthquakes which are Tokyo Region earthquake among the steady seismic-activity before there were a seismic-activity triggered by 3.11 in Tokyo Region.On the other hand, this research is covering the seismic activity in Tokyo Region triggered by 3.11.

However, the relevance between triggered seismic-activity and steady seismic activity of the 3.11 is not clear yet. Therefore, it is not appropriate to simply compare the two numbers.


What measures can we take for disaster mitigation?

Not just in Tokyo region, but anywhere in Japan, there’s possibility of magnitude 7-class earthquake occurrence . The Japanese island is formed by earthquakes. It is mandatory to get ready for earthquakes, and in other word, we are here by surviving many of these earthquakes.

Japan is a rare country where taking measures for M7 earthquake is technically possible. Below are the references of what you can do beforehand and what you should do during earthquake:

In occasion of earthquake, put yourself in safe places where “nothing falls on you”, “nothing collapse on you” and “nothing moves toward you”.

  • Check if all the gas and fire wares are turned off, after the shake. Do not attempt to do it while it is still shaking.
  • Careful with your feet. The broken glasses might be scattered on the floor.
  • Open the doors and windows to keep the way out.
  • Do not stay near brick walls and gate. They are easy  to collapse.
  • Stabilize your furniture. It is your own responsibility to keep them stable.
  • Get checked of the earthquake resistance. Any buildings built before 1st June 1981 are built by the previous building law, and Is not safe. Get checked as quick as possible. Many of the local government will help you. (make a call to the disaster-prevention department in your local government)
  • Put antiseismic reinforcement, if needed according to the result of test.
  • Any communication devises could be unavailable by earthquake. Discuss with your family, how to get in touch in case when all devises are not working.

For epicentral earthquake, such as Tokyo region earthquake, it is warned that 80% of the dead will be because of the building collapse and falling of the furniture. And in reality, it was like that by the Great Hanshin-Awaji Earthquake. In contrary, if you have antiseismic reinforcement on your building, and stable the furniture, 80% of the damage could be reduced. (antiseismic reinforcement on school building is an urgent need).

Fires could be prevented if collapse of building is prevented. Fire extinguishing will be easier if the brick walls don’t fall. The damages could be lighter. It is possible to keep the damage as small as possible, by each individuals getting prepared for the possible earthquakes. Please take measures NOW!


Q: There was a front page report on Yomiuri News Paper on 23rd January 2012 about this issue. Were there any differences in seismic activity of the capital region?
A: No. nothing new was found. This research was presented at ERI Danwa-kai in September 2011. And is the same provisional calculation indicating the seismic activity after 3.11 that were announced at the ERI Danwa-kai. The article on the newspaper is a provisional calculation derived by a research group of Associate Prof. Shinichi Sakai. And this page is created according to the context announced by Associate Prof. Shinichi Sakai.

Q: Is this an official opinion of the Earthquake Research Institute?
A: No. This page was created because we received many inquiries after the article on paper. Being announced on this page or other pages in this website does not mean that the fact is admitted by the Earthquake Research Institute or the community of researchers.

Q: Is both “98% in the next 30years” and “70% in the next 4 years”, the same information?
A: Yes, it is the same.

Q: Why is it different from the “70% in the next 30years” announced by the government?
A: Read the above description on: The reason why the number is different from what the government is announcing “70% in next 30years”.