Is Japan at Risk for Another Great Earthquake (M>8) in the Near Future?

john's picture

Unfortunately it is very possible, at least according to the calculations on this web site.  We can use the earthquake viewer, found under the "Tools tab", to define a selection polygon in the Japan region (Figure 1 below).  A relatively recent addition to the viewer tool then computes the numerical probability for events M>5, M>6, M>7, and M>8 for time periods 1 month from now, 1 year from now, and 3 years from now, occurring within the defined region.  Figure 1 below shows that the probability for an M>8 earthquake is very high.

In fact: 18.2% for the next month, 92.5% for the next year, and 99.9% for the next 3 years (starting date is today, January 24, 2013).

This is clearly a very disturbing result.  However, it seems to be consistent with the basic NTW forecast method, describe in previous blogs and published as [1].  The NTW method is based on the Gutenberg-Richter magnitude-frequency relation, which states that for every M>8 earthquake in a region, one expects about 1000 M>5 earthquakes.  So if we begin counting M>5 earthquakes just after the most recent M>8 event, then after another 1000 M>5 earthquakes have occurred, another M>8 earthquake should be expected.  

As a simple sanity check, we downloaded the ANSS seismic catalog for the region bounded  by latitudes 23 degrees and 46 degrees north, and longitudes 128 degrees to 147 degrees east.  This region encompasses Japan and in fact more territory, but the great majority of the active seismicity is along the eastern coastline of Japan.

This catalog tells the tale -- indeed, about 1000 M>5 earthquakes have in fact occurred since the M9.1 Tohoku earthquake of March 11, 2011.  So this result is at least consistent with the calculations of the NTW forecast method.  

As another check, we looked at other known seismically active zones around the Pacific Rim including Sumatra, Indonesia, Chile, California, Taiwan, New Hebrides, Central California, and a few other seismic zones.  We found that the probabilities for M>8 are nowhere close to those seen in and around Japan.  

We can also ask:  Is it reasonable that, so soon after the great Tohoku earthquake that another great earthquake should occur in the region?  To find the answer, we don't have to look far.  Recall that the great M9.2 Andaman-Sumatra earthquake occurred on December 26, 2004 just west of Sumatra.  Then 3 months later, on March  28, 2005, the M8.6 Northern Sumatra earthquake ruptured the subduction fault zone immediately to the south of the M9.2 rupture zone.  And last year on April 11, 2012, another M8.6 earthquake struck a fault zone to the west of the M9.2 earthquake.  So indeed, it seems that another M>8 earthquake might well be expected to occur in the Japan region.

Another example of spatial-temporal association of 2 or more great earthquakes was the March 28, 1964 M9.2 Prince William Sound, Alaska earthquake, followed by the February 4, 1965 M8.7 Rat Islands earthquake.  

On the other side of this argument, however, is the May 22, 1960 M9.5 Validivia, Chile earthquake, which was not followed by another event having M>8.  It should also be noted that Shearer and Stark (http://www.pnas.org/content/early/2011/12/12/1118525109) have argued that the recent occurrence of a number of great earthquakes around the world since 2004 is due to random clustering.

We can only hope that, should this event actually occur, the damage and any possible tsunami would not cause the kind of destruction and loss of life that the M9.2 Tohoku event caused almost 2 years ago.  Or possibly worse, as discussed in this nightmarish scenario: http://www.yomiuri.co.jp/dy/national/T120830004931.htm

[1] J.B. Rundle, J.R. Holliday, W.R. Graves,  D.L. Turcotte, K.F. Tiampo and W. Klein, Probabilities for large events in driven threshold systems, Phys. Rev. E, 86, 021106 (2012)

Figure 1:  Defined region and earthquake probabilities.

Figure 2: Spatial probability contours.

Risk Alert