Wednesday, March 30, 2011

Japan tsunami animation - wave interference

This NOAA animation of the calculated tsunami is quite impressive:

Listen to the accompanying mini-lecture. The resulting wave patterns due to bathymetry, wave reflection and refraction are visually stunning.

This is just a computer prediction of the expected tsunami.  We'll take a look at tsunami data soon.

Tuesday, March 29, 2011

Tohoku earthquake - the tsunami

Finally, I am getting to the tsunami.

I'll be presenting some of the images that I find most impressive, although not too much in the way of photos or videos.

The figure above shows the maximum tsunami amplitudes forecast (i.e., predicted, not observed) by the NOAA (National Oceanographic and Atmospheric Administration) Center for Tsunami Research using the MOST (method of splitting tsunamis) model.

The scale saturates in Japan (I'll look for a higher resolution image for the near epicenter effects), is about 0.5 meters in Hawaii. 0.25 meters on the west coast of the U.S., and about 0.1 meters on the west coast of South America.

Saturday, March 26, 2011

Seismic spectra and building response

Seismic waves can be analyzed for their characteristic periodicities. Longer period components will have more effect on taller buildings, while shorter periodicities will have greater impact on shorter buildings. This phenomenon was recognized in the 1986 Mexico earthquake. This phenomenon is nicely explained in a web page from MCEER, the Multidisciplinary Center for Earthquake Engineering.

There are some interesting links to Japanese data on the Tohoku earthquake from Tolyo University. One is: 

The comparison of long-period ground motion: The feature of seismic ground motion in city center and influence by architectural structure (By:Dr.Furumura and Dr.Takemura)
A velocity response spectrum is calculated according to waveform recorded at Earthquake Research Institute in Bunkyou-ku district. The figure shows comparison with that of Niigata Chuetsu earthquake in 2004. The Chuetsu earthquake had strong long-period ground motion with period of 7 second, which causes only a minor damage mainly among skyscraper buildings. The 2011 Tohoku earthquake has about the same strength of velocity response occurred in a wide periodic band from 0.5 to 20 second. Not only skyscrapers but also all kind of buildings from wooden building (below 0.5 second) to a low-slung building (around 1second) and skyscraper buildings (few second).

Friday, March 25, 2011

Groundshaking damage to buildings, Japan

So much of the damage done after the Tohoku quake resulted from the tsunami, but there was also damage of buildings due to the groundshaking in high areas that were not damaged by the tsunami. Some reports and images can be found at the Earthquake Engineering Research Institute web site.

A number of field reports on damage done in the epicentral area are given here.

A report by Shunzuke Otani on Building Damage in Sendai says:

"Miki Shoji Co, a commercial business company, reported building damage statistics by external observation along the Sendai main street; out of 175 buildings
          No damage: 95
          Some damage, but could be operational in short time: 66
          Severe damage, need some time before use: 14
          Collapse or tilting: 0
Sendai is the largest city in the affected area.  A church in Mito City, Ibaragi Prefecture, suffered severe damage in the tower. A automobile ramp to the roof parking area collapse and killed three persons in automobiles; the ramp is supported by steel columns and connected to the main structure of a super market building."  Images of these are shown below.

Sendai church damage

Sendai parking garage damage

Tuesday, March 22, 2011

Mercalli intensities for the Tohoku earthquake

RMS is a provider of products, services, and expertise for the quantification and management of catastrophe risk. Below is a Mercalli intensity map produced by RMS for the Tohuku earthquake.

The Mercalli intensity map indicates damages to structures and human reactions to the event. Mercalli intensities tend to be highest near the epicenter, and decrease with distance.

[I mis-read my colors, so this paragraph is not true: In this case, the highest intensities of VI are not at the epicenter. This is presumably because of regional geology; unconsolidated rocks amplify the groundshaking, so the intensity pattern depends in part on the geology of the region.]

The Mercalli scale is described by the USGS. The description for Mercalli intensity VI is:
V. Felt by nearly everyone; many awakened. Some dishes, windows broken. Unstable objects overturned.
VI. Felt by all, many frightened. Some heavy furniture moved; a few instances of fallen plaster. Damage slight.
VII. Damage negligible in buildings of good design and construction; slight to moderate in well-built ordinary structures; considerable damage in poorly built or badly designed structures; some chimneys broken. VIII. Damage slight in specially designed structures; considerable damage in ordinary substantial buildings with partial collapse. Damage great in poorly built structures. Fall of chimneys, factory stacks, columns, monuments, walls. Heavy furniture overturned.
Because of the earthquake-resistant building construction in Japan, the groundshaking of the quake did not do much damage. The tsunami was to blame for most of the destruction.

Monday, March 21, 2011

How much did the rest of Japan move?

This map from JPL and NASA based on gps measurements shows movements up to 5 meters (about 15 feet) on the mainland itself.

Ground acceleration for the Tohoku quake

Tohoku, by the way, is Japanese for northeast, since that is the part of the island of Honshu where the earthquake occurred. It is also being referred to as the Sendai earthquake, Sendai being the largest big city nearby.

The primary hazard intrinsically associated with earthquakes is groundshaking. When the elastic energy stored in the crust is released by the earthquake and motion on the fault, the stored energy causes the ground to vibrate, or accelerate back and forth.  The figure below from the USGS shows the peak acceleration in the vicinity of the quake caused by the shaking earth.

The peak ground acceleration in the graph below was almost three times the Earth's gravitational acceleration (g).

By default, buildings are designed to withstand the vertical acceleration due to gravity, but not necessarily to withstand horizontal ground accelerations of the same magnitude. This is why earthquake-resistant buildings must be designed to deal with. Japan has done a good job with this, so the major damage in the quake was not caused by the groundshaking.

Sunday, March 20, 2011

Fault movement for the Tohoku Japan earthquake

At times after major disasters, there is so much information, but so spread out. Harvard has set up the Japan Sendai Earthquake Portal for data sharing and exchange, with links to other relevant sites. That portal led me to the figure below.

The figure below, from Penn State, shows co-seismic slip - movement on the fault itself. The red dots are actually locations of the main earthquake and aftershocks.

From the figure caption: A preliminary comparison of an estimate of coseismic slip (created in collaboration with Thorne Lay at UC Santa Cruz and Hiroo Kanamori at CalTech) with the aftershocks and bathymetry. The rainbow-filled circles show the slip pattern for the Mw 9.0 earthquake. The deeper red circles are the aftershock locations.

Credit: Rupture model by C. J. Ammon, Penn State, T. Lay, UC Santa Cruz, and H. Kanamori, Caltech. Aftershock data from the US Geological Survey.

Thursday, March 17, 2011

Thrust faulting for the Japan earthquake

The images below are taking from a USGS poster on the Japanese earthquake.

As in the last post, the balloon diagrams are "fault plane solutions." These are based on the P-waves arriving at different seismic stations, and whether the first pulse of energy is a compression or a dilatation (stretching apart). These patterns reveal the orientation of the fault on which the earthquake occurs, and the nature of the faulting. Note that in the top figure, the boundary between the black and white portions of the balloon diagrams are parallel to the major plate boundary between the Pacific plate (to the east) and the Eurasian pate (to the west). This tells us what the strike of the fault is.  Also, the pattern of black (compressions) and white (dilatations) shown corresponds to a thrust fault occurring on a fault plane shallowly dipping to the northwest, as suggested by the pattern of earthquake locations shown on the second diagram. Thus, we can use seismic waves to tell us what kind of faulting of the Earth's crust occurred 25 km below the surface, even though no one is there to directly observe it.

Slip of almost 50 meters for Japan quake

The figure below from the University of California at Santa Barbara shows a maximum slip of almost 50 meters (165 feet) on the fault accompanying the Japan earthquake. The star shows the epicenter of the main shock. Red dots are the aftershocks (M>3). The black line indicates the major plate boundary

Oher interesting resources for this event, some of which I will return to, are referred to from the IRIS web page.

Wednesday, March 16, 2011

Animation of Japan foreshocks-main shock-aftershocks

This animation from YouTube showns the main shock at 1:16, with both preceding foreshocks and succeeding aftershocks. Because the energy release increases by a factor of 27.5 with each step in magnitude, most of the energy is released by the main shock.

Beware if you are in the library: music is included.

 You can get a sense of all that activity by looking at the lists at the USGS/NEIC website.

Japan quake was magnitude 8.9, is now 9.0

"The USGS has updated the magnitude of the March 11, 2011, Tohoku earthquake in northern Honshu, Japan, to 9.0 from the previous estimate of 8.9. Independently, Japanese seismologists have also updated their estimate of the earthquake’s magnitude to 9.0. This magnitude places the earthquake as the fourth largest in the world since 1900 and the largest in Japan since modern instrumental recordings began 130 years ago.

"The USGS often updates an earthquake’s magnitude following the event. Updates occur as more data become available and more time-intensive analysis is performed. There are many methods of calculating the energy release and magnitude of an earthquake. Some methods give approximate values within minutes of the earthquake, and others require more complete data sets and extensive analysis. Due to inherent uncertainties in the modeling of energy and magnitude, the results from different agencies often vary slightly. These magnitude discrepancies arise from the use of different data and techniques. For more information on why magnitudes change, see the Earthquake Hazards Program FAQ website."

Tuesday, March 15, 2011

Largest earthquakes in history

By magnitude (Japan quake comes in as #4)

Event Date/time Lat Long Mag FatalitiesRegion
1. 1960-05-22 19:11 -38.29 -73.05 9.5 1,655 Chile
2. 1964-03-28 03:36 61.02 -147.65 9.2 125 Prince William Sound, Alaska
3. 2004-12-26 00:58 3.295 95.982 9.1 227,898 off the west coast of northern Sumatra
4. 2011-03-11 05:46:23 38.322 142.369 9.0
Near the East Coast of Honshu, Japan
5. 1952-11-04 16:58 52.76 160.06 9.0
Kamchatka, Russia
6. 2010-02-27 06:34:14 -35.846 -72.719 8.8
Offshore Maule, Chile
7. 1906-01-31 15:36 1.0 -81.5 8.8 1,000 Colombia-Ecuador
8. 1965-02-04 05:01 51.21 -178.50 8.7
Rat Islands, Alaska
9. 2005-03-28 16:09 2.074 97.013 8.6 1,313 Northern Sumatra, Indonesia

By casualties (Japan will not make this somewhat subjective list):
Rank Death toll Event Location Year
1. &0000000000830000000000830,000 1556 Shaanxi earthquake China 01556-01-21556
2. &0000000000242419000000242,419–779,000 1976 Tangshan earthquake China 01976-07-281976
3. &0000000000316000000000316,000+ 2010 Haiti earthquake Haiti 02010-01-122010
4. &0000000000250000000000250,000 526 Antioch earthquake Antioch, Byzantine Empire (now Turkey) 00526-526
5. &0000000000235502000000235,502 1920 Haiyuan earthquake China 01920-12-11920
6. &0000000000230210000000230,210 2004 Indonesian earthquake Indonesia 02004-12-262004
7. &0000000000230000000000230,000 1138 Aleppo earthquake Syria 01138-1138
8. &0000000000200000000000200,000 856 Damghan earthquake Iran 00856-12-2          856
9. &0000000000150000000000150,000 893 Ardabil earthquake Iran 00893-03-2893
10. &0000000000142000000000142,000 1923 Great Kanto earthquake Japan 01923-09-1923
11. &0000000000137000000000137,000 1730 Hokkaido earthquake Japan 0171730
12. &0000000000123000000000123,000 1908 Messina earthquake Italy 01908-12-21908
13. &0000000000110000000000110,000 1948 Ashgabat earthquake Turkmen SSR, Soviet Union (now Turkmenistan) 01948-10-01948
14. &0000000000100000000000100,000 1290 Chihli earthquake China 01290-09-21290
15. &0000000000100000000000100,000 1755 Lisbon earthquake Portugal 01755-11-011755
16. &0000000000100000000000100,000 1667 Shamakhi earthquake Azerbaijan 01667-11 1667

Or see the USGS list.

Japan quake, some of the early news

Here were some of the early news stories that impressed my geophysics students Peter Rippberger, Hannah Rosenberg, and Matt Slowinksi:
Up to date coverage of the diasster in Japan from BBC news report.
One of the worst effects of earthquakes are the tsunamis caused by the quake, check out the link to learn more.
Look at the link to see pictures from the devastating event.
Look at this live footage of how quickly the flood spreads.
Here is a video from one of the residents of Japan while the Earthquake was occuring.

Monday, March 14, 2011

Japan earthquake - seismograms

My geophysics students Eamon Hall, Conor Neal, and Stacey Sosenko tracked down this seismogram from the Australian seismograph network:

Seismograms from the western Australia network of seismograms shows the initial pulse a little more than 10 minutes after the main quake, as shown in the figure below.
The Franklin & Marshall seismograph is located about 93 degrees away from the focus and felt the quake about 13 minutes after it struck, as expected.  The seismogram also shows up two  aftershocks later in the day.

Or, check out some more seismograms using the Rapid Earthquake Viewer.

Japan 8.9 quake - the source

On Friday, the morning after the devastating earthquake in Japan, I had my geophysics students compile some information about the quake.

Elvis Andino, Stephanie Douglas, and Andrew Tsang looked at the source parameter. They found:

The epicenter occurred at 38.322°N, 142.369°E on the Japan Trench subduction zone, where the Pacific plate is subducting beneath the Eurasian plate.  The epicenter was 373 km NE of Tokyo. The earthquake focus was about 24 km below the sea floor.

As shown below in the fault plane solution, the earthquake exhibited thrust faulting, consistent with its location along a subduction zone. The NE-SW strike of the fault plane is parallel with the orientation of the plate boundary shown above.

It had a magnitude of 8.9, putting it 5th on the USGS's list of largest earthquakes since 1900, right before the 2010 Chile earthquake,

Large foreshocks preceded the main shock over the previous two days, including a M7.2 quake about 40 km from the epicenter of the big earthquake.

Because Japan is partially surrounded by fault zones, it has had many earthquakes in its history. These have mostly been shallow quakes. The last major earthquake on the same fault was in 2005, a M7.0

Monday, March 7, 2011

Arkansas 4.7 quake, 2/28/11

A magnitude 4.7 earthquake happened in Arkansas on Feb. 28.  It's interesting that the Arkansas Geological Survey has some decent web pages on earthquakes in the New Madrid Seismic Zone, but I couldn't find anything yet on this particular event.

According to the USGS:
This figure [below] shows the distribution of earthquakes in north-central Arkansas that have occurred since June 2010. Earthquakes on the figure are color-coded, with red circles denoting earthquakes that occurred from June 4th, 2010 to February 14, 2011, and orange circles denoting earthquakes that occurred from February 15, 2011 to March 1, 2011. The space-time pattern shows a systematic shift in seismicity to the southwest starting on February 15, 2011. White triangles show the location of real-time seismic stations in the vicinity of the earthquake swarm that are used by the University of Memphis and USGS in their complimentary monitoring activities.

More in following posts.