Civil Engineering News

Learning about loss

By Kevin Wilcox

A researcher at Northwestern University examines the profound changes the 2004 Indian Ocean tsunami had on preparedness, warning systems, and the scientific community and advises engineers to remain informed of the latest research.

In the very early hours of December 26, 2004, a massive earthquake shook the floor of the Indian Ocean along the subduction zone where the India Plate slides under the overriding Burma Plate. The earthquake reached an approximate magnitude of 9.1 and lasted about 10 minutes along 750 mi of the subduction zone. 

The temblor unleashed the deadliest tsunami on record, killing an estimated 250,000 people, primarily in Indonesia, India, Thailand, and Sri Lanka. Experts suggest the tsunami waves carried a total energy equivalent to double that of all of the explosives used in World War II. 

As the earth ruptured and waves that would reach heights of 80 ft raced toward unprepared coastlines, scientific understanding of tsunamis took a step backward. The prevailing theory at the time — that a subduction zone’s earthquake potential was governed by the age of the fault and how rapidly it was moving — suggested that the event unfolding in the Indian Ocean was impossible. 

A researcher who has examined the 16 most scientifically significant tsunamis that have occurred in the 10 years since that tsunami says the scientific community has learned a great deal about early detection, how to prepare residents to escape the coast unharmed, and how to build coastal infrastructure offering greater resiliency. But scientists have yet to replace the paradigm that was destroyed that December morning. 

The scientist who has carried out this research, Emile Okal, Ph.D., a professor in Northwestern University’s earth and planetary sciences department, is a member of the scientific community that has worked for decades to better understand tsunamis. Having majored in physics, Okal became interested in the study of tsunamis as he pursued his doctorate at the California Institute of Technology under Hiroo Kanamori, Ph.D., now an emeritus professor there. Kanamori is responsible for many scientific advances that have shed light on tsunamis. “Kanamori was one of the most enlightening members of an already enlightening department,” Okal recalls. “Being Japanese, he had a particular sensitivity to tsunamis, and he passed it along to the graduate students.” 

The prevailing theory at the time — that a subduction zone’s earthquake potential was governed by the age of the fault and how rapidly it was moving — suggested that the event unfolding in the Indian Ocean was impossible. 

Okal says that the 2004 Indian Ocean event changed the perception of tsunamis among both the general public and the scientific community, virtually overnight. “From the standpoint of a person on the street, ‘tsunami’ became a household word. The event probably had the highest tsunami fatality number in the history of mankind,” says Okal, who points out that any large prehistoric events would have affected much smaller populations. “The magnitude of the disaster was what changed our view from the standpoint of society as a whole.”

For the scientific community, he notes, it was the first well-documented, well-recorded tsunami in the Indian Ocean that exported death and destruction across the oceanic basin. The data it generated helped to transform the prevailing wisdom that held that the age and speed of this subduction zone were such that it could produce earthquakes of magnitude no greater than 8.

“Why did the paradigm not work? Because Mother Earth…functions in a different way,” Okal says. “Unfortunately, we are at the stage where we [must] have a principle of precaution. Since we don’t understand and since we haven’t found a better [paradigm], we must operate under the scenario that all subduction zones may have earthquakes that reach magnitude 9.” 

Published reports indicate that the initial ground motions felt on the shores of the Indonesian island of Sumatra didn’t register as especially powerful to longtime residents, but they were unusually prolonged. This created a difficult decision for residents who had been taught to seek higher ground during strong shaking. The shaking that the residents experienced on the coastline was deceptive because they were experiencing what is known in the field as a tsunami earthquake, featuring anomalous wave patterns. “An earthquake has a release of a great deal of seismic energy, and you can compare that with a symphony, if you wish. Classical earthquake theory would dictate that an earthquake would play a balanced spectrum between trebles and bass. Essentially if you were to filter out all the bass and only hear the trebles—the noise of the high frequencies at the high pitches — you could interpret it and predict what you have in the bass. That is what we call a scaling law. If you look at a little portion of the spectrum, you can extrapolate what is elsewhere in the frequency band,” Okal explains. 

Tsunami earthquakes, however, do not conform to these scaling laws. They have subdued waves at higher frequencies and stronger shaking at the lower frequencies. The waves, which Okal refers to in the analogy as treble frequencies, have a period of between 1 and 10 seconds. These are the seismic waves that residents feel and that inflict the most damage on smaller buildings. 

“It turns out that, for reasons that are difficult to explain, a tsunami is excited by the bass — by the very low frequency part of the spectrum. Now you have an earthquake that is a bad guy — a scofflaw of the scaling laws — and plays a deceptively bass symphony and is very low in the trebles. People will not feel the earthquake to its true potential,” Okal explains. 

One of the first such tsunamis in recorded history struck northeastern Japan on June 15, 1896. Residents there reported only mild shaking on the coastline. But the earthquake generated a tsunami with waves nearly 100 ft high in the near field, killing more than 27,000 at a time when the country had far fewer coastal residents than it does today. “We’re not exactly sure why such earthquakes occur or in which kinds of environments they take place,” Okal says. “There are still very many questions that we have about them.” 

The 2004 Indian Ocean earthquake also cast a harsh spotlight on the issue of tsunami preparedness. There was no warning system in place for the Indian Ocean at the time of the earthquake and tsunami. The Pacific Tsunami Warning Center, in Ewa Beach, Hawaii, part of the National Oceanic and Atmospheric Administration’s National Weather Service, issued two tsunami information bulletins for the Pacific basin. However, the Indian Ocean was beyond the center’s mission area in 2004, and effective communication channels had not been established with countries surrounding the Indian Ocean. 

“We knew there was a big event. But nobody was in charge of warning the people at risk in the Indian Ocean,” Okal says. “Over the last 10 years, there have been many warning centers that have been built from scratch in various regions or countries.” New centers have been constructed in Indonesia, India, Turkey, and Puerto Rico, for example. And because the Indian Ocean was included in the mission of the Pacific Tsunami Warning Center after the 2004 earthquake, officials there will pass along information gathered to peers in the new centers. 

“Hopefully there will be people dedicated, 24 hours a day seven days a week, to monitoring the state of affairs in the Indian Ocean,” Okal says. “These people will have, hopefully, designed, built, and tested their communications.” Scientists estimate that had such a warning system been in place in 2004, it might have provided residents in some of the affected areas warnings of between 20 minutes and two hours, perhaps saving thousands of lives. 

The 2004 Indian Ocean earthquake also cast a harsh spotlight on the issue of tsunami preparedness. There was no warning system in place for the Indian Ocean at the time of the earthquake and tsunami.

Okal is the author of the paper “The Quest for Wisdom: Lessons from 17 Tsunamis, 2004–2014,” published last September in the United Kingdom in the Philosophical Transactions of the Royal Society A. (The paper can be accessed at Okal examines 16 of the largest and most significant tsunamis from the approximately 130 that have occurred since the 2004 Indian Ocean tsunami, assigning each a “wisdom index” based on the warnings issued and the public response to those warnings. He found a mixed record of performance, noting technological and procedural advances as well as some missed opportunities. 

Taken together, he finds that there has been significant progress since the 2004 Indian Ocean tsunami in warnings to areas in the far field — that is, places more than 1,000 km from the epicenter but still likely to be affected. There were just three deaths in the far field from the tsunamis he examined, and two of the three who lost their lives ignored evacuation orders. That compares with 50,000 far-field deaths from the Indian Ocean tsunami. However, Okal finds that rapid warnings and evacuations in the near field are still problematic, as is detecting low-frequency tsunami-creating earthquakes. 

“I’m of the opinion that in the very near field — let’s say 100 or 200 kilometers — centralized warnings in order to save populations are not very practical. It is a matter of people at risk taking their fate in their own hands and evacuating immediately by themselves,” Okal says. “The shaking is the warning.”

This approach to near-field warnings becomes a problem when an area experiences a tsunami earthquake with minimal high-frequency shaking. Okal says public education campaigns should now include a caveat that weak shaking that lasts an exceptionally long time also is a warning. This will be a far more challenging concept to effectively convey because people experience the passage of time differently during an emergency, and in hindsight they often estimate the time that passed inaccurately. 

“There is an effort now. We don’t know how to attack the bull by the horns, but there is a lot of thinking about how we are going to educate people to try to be sensitive to the duration aspect of any kind of shaking that they perceive,” Okal says. 

Of the tsunamis that Okal addresses in his paper, the 2011 disaster along the coast of the Japanese region of Tohoku stands as a tragic illustration of the difficulties of near-field warnings and evacuations. The tsunami danger was promptly recognized, and a warning was issued by the Japan Meteorological Agency. However, because the shaking continued for several minutes, initial estimates of the shaking and, thus, of the tsunami risk were significantly lower than the actual 9.0 magnitude. “After four or five seconds, scientifically, you cannot give an accurate estimate of an earthquake if it is greater than magnitude 5,” Okal says. “There were people who were evacuated to the roofs of 3-story buildings and were caught by the waves because the waves were 10 stories high. That’s very tragic. 

“It’s all part of this very tricky business,” he continues. “The community itself was not prepared to evacuate to a sufficiently high altitude.” The experiences in Tohoku, however, do illustrate the value of an educated populace in minimizing the risks posed by tsunamis.

“The death toll was about 18,500, which is a catastrophic number. But I have some colleagues in social sciences who have estimated the number of people who were present in the sweep of land that was completely devastated and razed to the ground by the tsunami on that day (at) about 200,000,” Okal says, “which means about 90 percent of them escaped.”

The value of an educated and experienced populace is further illustrated by the differing experiences of Swedish, German, and Japanese tourists during the Indian Ocean tsunami. At least 539 German and 543 Swedish tourists died in the tsunami. “At the same resorts, in the same hotels, the Japanese tourists, when they saw the sea retreat, climbed to the 26th floor of the hotels and survived,” Okal says. “The Japanese were educated to the dangers of tsunamis. Education plays an incredible role.”

This education can assume many forms, he explains, from the informal oral histories that people in Peruvian fishing villages share about their ancestors running from the shore at the first signs of an earthquake to the numerous signs lining the roads of American Samoa warning of tsunami risks. It also includes the formal drills being staged in northern California and the Pacific Northwest, areas facing risks from the Cascadia Subduction Zone. 


Okal describes a tsunami that was generated by a large earthquake on the Cascadia Subduction Zone in 1700 as an important milestone in understanding the phenomenon. Because this event predates historical records in the area, researchers had to piece together evidence to develop an understanding of it. At first, many in the seismic community believed that the Cascadia Subduction Zone was unlocked, facilitating slow movement between the plates. Researchers, however, later found evidence that the subduction zone is, in fact, locked and building pressure. 

The discovery in the mid-1980s of buried tidal basins was the first clue to a significant past tsunami in the area. Carbon dating later narrowed the possibilities of that tsunami to the 1700s. Researchers examining historical records from Japan found a so-called orphan tsunami that sent waves crashing into the Japanese coast in early 1700. 

As the danger of a future earthquake and tsunami on the Cascadia Subduction Zone has come into sharper focus in recent decades, government agencies have developed computer simulations and prepared inundation maps for a series of different seismic and tsunami scenarios. These efforts to gain a better understanding of tsunami dangers have been coupled with public education campaigns. In March 2010 Okal attended a tsunami evacuation drill in Crescent City, California. Although the tsunami was modeled on an event generated by the Cascadia Subduction Zone, the drill was held to mark the anniversary of the devastating tsunami generated by the so-called Good Friday earthquake, which occurred in the Prince William Sound region of Alaska in 1964. A 20 ft wave crashed onto the coast, killing 12 people and destroying 289 buildings and businesses. 

The evacuation drill was, Okal says, “a very interesting experience. All of the decision makers, the stakeholders, were superb. The fire department, the office of the mayor, the sheriff, the schoolteachers, the Red Cross — all of these participants were superb,” he says. “It was very moving and promising to see the teachers walking little kids to rehearse the evacuation of the school to high ground.” 

The experience, however, was tempered somewhat by a conversation with the proprietor of a coastal motel in the area who advised Okal to ignore the drill as a needless government intrusion. “I think in this respect there is still some education to perform,” Okal says. “But I think, by and large, the seashore communities have been very responsible and very responsive to the threat.”

Okal notes that, in some of the more populated areas along the West Coast of the United States, the major activity centers are not directly facing harm’s way. “The Port of Seattle is in Puget Sound, and the Port of Portland is on the Columbia River,” Okal notes. “There is a risk of course, but it is not as bad as it would be if Seattle — with the size of its harbor — were straight on the open ocean.” 

As the scientific community expands its understanding of tsunami risks, Okal advises engineers to keep abreast of research findings. He notes that he often speaks with engineers seeking an understanding of the tsunami risks to their coastal projects, and the solution is often a balancing act between open, stilted designs that withstand tsunamis and the more traditional structural elements that accommodate earthquake motion.

As the scientific community expands its understanding of tsunami risks, Okal advises engineers to keep abreast of research findings.

Engineering a project for tsunami resistance is further complicated by the paucity of scientific records. Historical accounts of large tsunamis, primarily in Japan, stretch back more than 1,300 years. But in most areas scientific records go back just 100 years. “A fundamental principle of geology and geophysics is that the historical record (to which) we have access is completely dwarfed by the duration over which the dynamics of the earth are going to change,” Okal says. “These dynamic processes in the earth are going to take hundreds of millions of years. 

“But we can say, if an earthquake occurred in Japan in the year 869 AD, it will recur. Perhaps not within a lifetime…but it will recur.”

This article first appeared in the March 2016 issue of Civil Engineering.


- Advertisement -


Please enter your comment!
Please enter your name here

- Advertisement -