Tsunamis: "Wave that Shook the World"
Thomas H. Heaton
Professor of Engineering Seismology at the California Institute of Technology
Wednesday, March 30, 2005; 3:00 p.m. ET
Thomas Heaton, professor of engineering seismology at the California Institute of Technology, discussed earthquakes, tsunamis and the PBS NOVA episode "Wave that Shook the World."
Heaton was a Research Geophysicist with the U.S. Geological Survey from 1979 until July 1995, at which time he was the USGS Project Chief of the Southern California Seismic Network. He received a Meritorious Service award from the U.S. Department of the Interior in 1995 and he is a past President of the Seismological Society of America. He has written numerous research papers in the fields of earthquake ground motions, earthquake source physics, earthquake hazards in the Pacific Northwest, earthquake warning systems, and tidal triggering of earthquakes.
Editor's Note: washingtonpost.com moderators retain editorial control
over Live Online discussions and choose the most relevant questions for guests and hosts; guests and hosts can decline to answer questions.
Urbana-Champaign, Ill.:
Dear Dr. Heaton,
I very much enjoyed your NOVA special on tsunamis. I am a professor of Civil and Env. Eng. at the University of Illiois (geotech & geotech earthquake engineering specialty) and have been working closely with Dr. Steve Obermeier (retired USGS geologist) for several years. The recent tragic earthquake and tsunami in the Indian Ocean led us to contemplate tsunami hazard in other regions around the globe. We have conceived a new technique that we believe may allow us to determine the occurrence and flow depth of past tsunamis.
Would you kindly discuss areas of the United States and the world that might experience large tsunamis, but have poorly defined tsunami hazard?
Thank you and best regards,
Scott Olson
Thomas Heaton: To my knowledge, all major oocean-basin-wide tsunamis in the past 2 centuries have been generated along subduction zones (ocean basin trenches). All M > 9 earthquakes have generated giant tsunamis (we know of at least 5 of these in the past 100 years). In addition there was a mysterious event in the Aleutian Islands in 1946 that has been given an earthquake magnitude of 7.5, but the tsunami accompanying it was more like what is generated by a M 9. Of course the 1883 Krakatoa eruption also created a great tsunami. Since we do not yet understand the 1946 event (poor seismic instrumentation in those days), it's hard to say where similar events can occur. However, since all earthquakes with M>9 have generated giant tsunamis, I believe that you are asking where can such giant earthquakes happen? I would include the following locations Alaska/Aleutian subduction zone, Cascadia subduction zone (Washington and Oregon), Nankai trough (southern Japan), Sumatra, Chile, southern Peru, and Colombia. There are other locations that are less obvious, but may include the west coast of Costa Rica, Nicaragua, north east coast of New Zealand, and the east coast of Hokkaido, Japan.
_______________________
Prescott, Ariz.:
What re the most effevtive faults?
Are there many waves in an event?
What part of the wave draws water out to sea exposing the bottom?
Thomas Heaton: question 1. see answer to previous question
question 2. There are ususally many waves in the event. These multiple waves are caused by several effects. In the open ocean, the original wave breaks into many different cresta and troughs. In really large earthquakes, these can be separated by up to an hour. As the wave approaches the coast the wave speed changes dramatically with the water depth. Complex bathymetry can break the wave into several pieces. the first wave isw often not the largest.
3. Why does the water draw out first?
While it is typical for the water to recede first, this is not always the case ... someitmes the first wave is a poisitive wave. I have seen different scientists give different reasons for this phenomenon. My answer is that the uplifted part of the ocean bottom generates the positive wave that does most of the damage. However, in most subduction earthquakes, the region of the uplifted ocean floor is actually surrounded by regions that drop down during the earthquake. Since coastlines are closer to this outside annulus of subsidence, the first arriving wave is usually a negative wave.
_______________________
Richmond, Va.:
Dr. Heaton. Has anyone been able to determine the amount of time between the water beginning to retreat and the arrival of the first destructive wave? How much time is there from the warning sign or retreat of wave for those on shore to get to higher ground? Are we talking seconds or five to 10 minutes or more?
Thomas Heaton: I have not seen the numbers on this earthquake. However, I am guessing that the duration of the drawdown was 10 to 20 minutes.
_______________________
Boston, Mass.:
Mr. Heaton,
Thank you for being online today. How much of the energy of a tsunami is dissipated as it travels across the open ocean or sea?
Thomas Heaton: Tsunami waves travel quite efficiently. They are eventually damped by friction of the water flowing over the ocean bottom, particularly as the ocean bottom becomes shallow. Ocean basins can continue to see reverberations of the waves for several days.
_______________________
Silver Spring, Md.:
Dr. Heaton,
How would you characterize the warning systems that are protecting the US east and west coasts? Should they be better equipped?
Thomas Heaton: This is an important, but difficult question. It is clear that there were problems with the existing tsunami warning system in the December earthquake. In my mind, one of the key issues is being able to quickly determine the appropriate magnitude of giant earthquakes (M>9). This is best done with seismographs and the software that is appropriate for this task. Fortunately, the seismometers alsready exist and are distributed around the globe. When the data is analysed in the appropriate way, then it is easy to tell that an earthquake is a M>9 (and hence a grave tsunami risk). Unfortunately, the data analysis procedures taht are currently used for rapid earthquake analysis are not "tuned" in a way that is needed to determine the magnitude of these giant earthquakes. Remember that 1964 was the date of the last M>9 earthquake (Alaska M 9.3), and most seismologists have had little or no experience with these types of earthquakes.
In addition to seismic data, it is also helpful to measure the actual water motion of the tsunamis. There are several buoys in the Pacific for this task, and the Japanese have pressure sensors on underwater cables off of Japan. This is a more direct measurement, but it is also much more expensive than using the existing seismographic networks.
_______________________
Washington, D.C.:
Is there an explanation as to why this most recent earthquake off of the coast of Sumatra (on Monday) did not cause tsunami's?
Thomas Heaton: The most recent earthquake (M8.7) did indeed cause a tsunami, but it was not nearly as damaging as the December tsunami (M 9.2 earthquake). Why is that? I think the most obvious issue is the size of the earthquake. I have done some calculations that indicate that the kinetic energy of a tsunami should increase by a factor of 100 for every unit increase in magnitude. That means that the tsunami energy of the December event may have been ten times larger than that for the event on Monday. Furthermore, Mondays tsunami hit a region which had already been devastated by a much larger tsunami. People did not stick around the beach this time, and the most vulnerable structures were already gone.
_______________________
Alexandria, Va.:
The world certainly is alive these days, considering the plethora of earthquakes and volcanic eruptions lately. How long do active periods such as this one last? Have seismologists found a regular cycle the earth follows re: seismic activity?
Thomas Heaton: There are several things at play here. The first is an old adage. "Bad things come in three." It may sound stupid, but this is simply a statement of the fact that random events come in clusters. If they were evenly spaced in time, then there must be a clock to tell them when to happen. Random occurrences have no clock and seem to be clustered.
However, as you guessed there is more to it than that. Earthquakes change the stress in the earth and this change typically starts the failure process for more earthquakes. We ususally call these additional events "aftershocks." However, there is nothing especially different about an aftershock than any other earthquake. We are clearly in a cluster of events in Sumatra, and while the chance of another great earthquake is probably less than 1 in 4 in the next year, all of us recognize that this giant fault may still have the potential for other giant earthquakes.
_______________________
Maryland:
Greetings! As you may have heard, Maryland had an earthquake about a month ago. Some people described it as if a "huge truck had drove by their home". This is really odd because as far as I know, there arent any plates in Maryland. How do you explain this phenomenon?
Thomas Heaton: While most of the world's earthquakes occur on plate boundaries, there are other important earthquakes that are away from the plate boundaries. There are still forces in the plate and major earthquakes can still happen, just not so often. I would say that the rate of earthquakes in the eastern US is something like 5% the rate in California. In fact there have been important events in 1886 (Charleston), and three remarkable events in 1811 and 1812 in southern Missouri. If these events were to reoccur today, they may cause tremendous damage (you would wish you were in California, where we have better building codes). It's worth noting that the 1976 Tangshan China (M7.8) earthquake was not on a plate boundary and it killed somewhere between 500,000 and 750,000 people.
_______________________
Munich, Germany:
How is it that the original measurements of the tsunami causing earthquake in December was around 8.5. I would have thought that a couple of measurements around the globe would have been enough to accurately determine the severity of the earthquake within minutes, but instead, the original value was bumped up gradually until it reached its present value of 9.2. How long does it take to get an accurate measure of an earthquake?
Thomas Heaton: This is a critical question for tsuanmi warning. The earthquake itself takes about 6 minutes to occur. If we had sensors above the event we could determine the magnitude within 6 minutes (in principle). However, I am not aware of any data analysis center that could currently do this job. The Japanese have adequate data for this determination, but I am not sure how they are currentloy using the data.
With existing seismic networks, I believe that we could identify the appropriate magnitude for any earthquake in the world within 15 to 20 minutes. This is a matter of using very low-frequency seismic data (less than 0.003 Hz). Unfortunately this is not currently the practice in the existing tsunami warning centers. I suspect that that is about to change, but I am not involved in the design of their methodologies.
_______________________
Thomas Heaton: Dear all,
such good questions. I wish I could type faster! I must go since I need to teach a class.
Tom Heaton
_______________________
Automatically Update Page
Get New Responses
Submit Question
| 
