8.9

I keep checking the news to see updates on the aftermath of the magnitude 8.9 event off Japan's Honshu Island last night (or afternoon locally, 2011 March 11 05:46:23 UTC). The initial estimates of lost of life were far too low for an event of this magnitude which triggered a tsunami. It's beyond Hawaii now, but tsunami warnings were issued throughout the Pacific, including coastal BC. The problem is the nature of the coastline, as people in Port Alberni, where a 1964 tsunami swept away homes and swamped businesses, know well.



Though I don't have any instruments in the water which could sense the wave, at the moment, people still called me at the crack of dawn. I don't envy friends who work in earthquake seismology or in disaster preparedness on days like these.* I'm glad the epicentre was not further south - closer to larger population bases and known significant gas hydrate deposits. There are fears that earthquake shaking could destabilize gas hydrates, which may both hinder slope stability (another tsunami risk) and release large amounts of methane.

*For the most seriously affected people: Médecins Sans Frontières and the Canadian Red Cross can be found at these links.

(figures from Bunting et al., The Science of Tsunamis, Oilfield Review, 2007)

Chilean earthquake and tsunami hazard

A friend pointed out that there was a magnitude 8.8 thrust-fault earthquake, where the Nazca plate is subducting below the South American plate, offshore Maule, Chile, and the subsequent tsunami watch. Another friend followed up with questions about tsunami hazard, so I thought I would try to shed some light on this.

This news saddened me; I spent one month there and made many friends. Shaking along the coast would have been severe and a long aftershock sequence is to be expected. We were in the twin cities Valparaíso and Viña del Mar, west of Santiago, and sailed south towards Concepcíon (115 km NNE of Maule). We were invited by the Catholic University of Valparaíso to do a controlled-source electromagnetic survey to map gas hydrate deposits. Their interest was tsunami hazard. Chile is prepared for earthquakes and tsunami. They still remember the magnitude 9.5 earthquake of May, 1960 – the largest earthquake worldwide in the last 200 years or more. (It's important to remember that the magnitude scale is logarithmic.) Gas hydrates are stable under certain pressure and temperature conditions. If there is upwards movement or shaking of the seafloor, it is possible that they might become destabilized and lead to submarine slumps and slides (imagine a landslide underwater). This in turn might increase the risk of tsunami. The good news was we found little evidence for gas hydrates in this area.

If you see stories about earthquakes or tsunami in the news and find the facts bewildering (or worse, of dubious veracity) a good place to look for information is the USGS website (or for events in Canada, the GSC website - the USGS has the best worldwide coverage combined with public outreach).

The animated image above is an Tsunami Model for Chilean earthquake produced by the Earthquake Research Institute, Tokyo (another place where I've spent one month of my life). You can see that the maximum amplitude of the the waves induced are on the order of centimeters. Tsunami (Japanese for 'habour wave') are defined by the source (a displacement of a large volume of water, due, for instance to earthquake - especially subduction earthquakes, volcano, land- or submarine slide), but they also have some consistent properties in terms of the physics of the waves produced in the oceans themselves. This includes a small amplitude and very long wavelength (like the gravity waves I use in my experiment). They are not defined by height. Heights of centimeters are most common. In fact, in the ocean basin, if you are not using very specialized, high precision instruments (pressure gauges and accelerometers, like those used in my research), you are unlikely to notice a tsunami. There are two main issues which can lead to death and destruction: harbour geometry and the seemingly relentless nature of the waves. As the seafloor shallows at a coast, the waves shoal - they are compressed, slow down and increase in amplitude. Depending on the geometry of the seafloor, the effect can be severe.

This second animation shows this effect schematically. The second problem is also evident: duration. Even a small increase in relative sealevel (run up) can cause a lot flooding of damage. Though these waves are slow moving in shallow water, they are propagating and some structures will fail with continual force of this nature.

We all remember the 2004 Indian Ocean Earthquake and tsunami, but this does not mean that is what you should imagine every time you here the word tsunami. The 2004 event was a magnitude 9.3 which is barely conceivable. During today's event 4.7±0.9 x 10¹⁶ Joules of energy was released. During the 2004 event, 20 x 10¹⁷ Joules which is forty times more (source: USGS). Wave heights were up to 250 centimeters in the ocean itself - clearly a entirely different situation. Further, there has been an active tsunami-monitoring system in the Pacific for decades - not so in the Indian ocean. The Indian ocean event was a megathrust earthquake. While we can expect megathrust earthquakes in other subduction zones around the Pacific, today's event was not one.

So while this event was large enough to warrant tsunami alerts, the real concern, as is generally the case, is the nature of the housing near the epicenter. The death toll will be considerably less (likely by a factor one thousand) than the much smaller Haitian earthquake six weeks ago, largely because of building standards.

Pacific tsunami

{image from Preliminary Earthquake Report
U.S. Geological Survey, National Earthquake Information Center
World Data Center for Seismology, Denver}

At 17:48:11 UTC (which was 1:48 pm EDT) today, a magnitude 8.0 earthquake hit on the Tonga trench, just off Samoa. A tsunami has been generated. No reports yet on the number of casualties in Samoa. Authorities are monitoring the risk around the Pacific basin and in Hawaii.

When I worked for the GSC, I knew when a large earthquake hit because the seismologist on duty would run down the hallway, or I'd see it on the seismometers on display, or I'd arrive at work and there would be a news crew. It got so that seeing the news crew was like arriving at work to find the flag at half-staff. It made me sad, prior to knowing why. Now I learn these things from the American Geophysical Union twitter feed.

Earthquake prediction remains an elusive goal. Successes are few (and are either the product of dumb luck, or on the most intensively studied fault system in the world, where population density happens to coincide with wealth and the means for investment, technology and expertise on the San Andreas - and even there, while a tremendous achievement, the the prediction was more vague than one would hope). I do not think this will improve any time soon; it in in the nature of earthquakes that probabilities are likely to be the best information earth scientists can provide. Unless there is some measurable, consistent precursor yet to be discovered, or more convincingly documented, there is not much which can be done. The makes one feel helpless and frustrated. It is one reason I did some research on earthquake hazard in Cascadia (British Columbia, Oregon, Washington and northern California), which is not directly tied to the bulk of my research.

Tsunamis on the other hand, are much more predictable. Since they are triggered by things like major earthquakes, or marine slumps and slides, there are measurable precursors. In the Pacific, where industrialized nations have recognized their risk for decades*, there is a well-established tsunami warning system. There are tide gauges and wave rider buoys and satellite altimetry data all available in real-time. It's hard to think about how much of the loss of life after the Sumatran tsunami on Boxing Day, 2004 was preventable. (It was triggered by a staggering 9.2 event which released an order of magnitude more energy than today's event).

Events like these make me wish I could do something. The instrument we deployed recently, in the North Pacific has and will see the effects of this. The energy from an earthquake of that magnitude radiates outward in all directions and makes the earth ring like a bell. The oscillations will already have been recorded by our gravitymeter (which is sensitive to acceleration and hence displacement of the seafloor). This would have happened approximately 12 minutes after the event. The tsunami waves are not traveling anywhere near as fast. Should it hit Hawaii (which is not a given) it will not arrive until 7:18 EDT (23:18 UTC) and it would be hours later when and if it reaches my instrument. If it does, the differential pressure gauge will record it. The waves I use to make measurements obey the same physics and are of similar frequencies as tsunami waves. So what interests me is have we deployed an experiment meant to image what is below the seafloor, which could also provide useful additional information to either the numerical tsunami modellers or to the warning system. Then, at least, I would feel like I was doing something to help.

*or centuries, in the case of Japan. The Haida of BC preserved their memories of the 1700 event in their oral tradition. The earthquake which triggered that tsunami has been estimated to have been a magnitude 9.0.