Islands Business - Cover Story/ Tsunami: EXPERTS DIFFER OVER TSUNAMI THREAT

But agree there’s no room for complacency

Samisoni Pareti

Checking out the damage.

A disaster simulation over the South Pacific published by this magazine one month after the Boxing Day Tsunami was exaggerated, says a geophysicist with the Pacific Tsunami Warning Centre in Hawaii.

Dr Gerard Fryer believes Fiji’s geophysical structure like the shallow North Fijian Basin and fringing reefs act as tsunami mitigating factors. But the damage would be widespread still and Dr Fryer concurred with the magazine’s assertion that there was no room for complacency. The tsunami cover story was written by the late magazine publisher Robert Keith-Reid with input from then deputy director of SOPAC, Russell Howorth. Their simulation as explained by Keith-Reid was influenced heavily by the July 17, 1998, Aitape Tsunami. At least 2200 people died, 10,000 fled inland and today PNG authorities are still working to help the traumatised people,” the late Keith-Reid wrote.

“After an earthquake felt at 6.49 pm, a series of three waves from 10 metres to 15 metres high crashed ashore 20 minutes later, surging inland initially at tree top height for 400 metres to 500 metres along a 14-kilometre sector of the coast. Experts were puzzled why it was that such large waves were generated by a moderate Force 7 earthquake and focussed on such a narrow area. The answer; seismic surveys showed the waves appeared to have been pushed up by a large submarine landslide caused close to the shore.”

Below is an extract of our interview with the Haiwaii-based geophysicist.

Is it true to say that the earthquake and tsunami that hit western Solomons confirmed what you experts had been saying all along that if the earthquake hits so close to any of these islands, any tsunami warning will be too late?

“Not entirely. If the earthquake is directly offshore, then the first positive wave will arrive in as little as five minutes. If the water offshore is deep (if you have an extensive shelf offshore, then the tsunami will take several minutes longer to arrive and in shallow waters a tsunami travels much more slowly). Right now, it takes us almost 15 minutes to get a warning out, but once we have more seismometers feeding data into this centre (a matter of increasing our communications bandwidth, which should happen within a few months), we should be able to get the warning time down to about 12 minutes. While that might be too late for people within 100 kilometres of the earthquake, it would certainly benefit those far away. For people within 100 kilometres of the earthquake, this is not as bad as it sounds, since they will have felt a natural tsunami warning: severe shaking of the ground.

If no advanced tsunami warning of say 2 to 3 hours can be issued, there is really nothing one can do, is this so?

“No, there is quite a bit that you can do, but it all has to be done locally. The biggest challenge in tsunami warning (or any warning for that matter) is getting the message over that last mile, from the local government office or police station to the people on the beach. With more and more people carrying mobile phones, pushing out an SMS message would probably be the most effective means of warning people. In July 2006 there was a tragedy on the south coast of Java. A “slow” earthquake (one which doesn’t produce much high-frequency shaking and is therefore not widely felt) threw up a big tsunami which crashed ashore and killed 700 people. We took longer than usual on that one (slow earthquakes are notoriously difficult to measure) and did not send out a warning message until 17 minutes after the earthquake. The first waves hit the shores about five minutes after our message, so people there had little chance. The real tragedy was where most of the deaths occurred, in the Pangandaran area, which the tsunami didn’t reach until almost 40 minutes after the earthquake. If our message had been relayed, and if there had been a notification system set up to get the word to the general public, then most of those poor people would have had 20 minutes in which to escape. The biggest challenge is that last step in the chain-getting the word to the public.”

SOPAC recently got donor agencies and partners for a weeklong meeting in Suva, aimed at encouraging the formulation of national action plans for island countries. Is this the only practical solution, getting coastal dwellers to move inland, and for all coastal buildings including resorts and other developments to abide by stricter building codes?

“It sounds like they’re thinking the right way. Yes, evacuation plans and strong buildings mean your community will be resilient to tsunamis and will be able to bounce back quickly after one occurs. One additional concept should be added: vertical evacuation. If you have a resort building of five or more floors it is going to survive the tsunami (witness the videos from Thailand in 2004). Those buildings would make an excellent refuge for anyone nearby, especially for small children or the infirm, who might not be able to walk fast enough to escape the inundation. Get people into the building and move them up a couple of floors and they’ll be safe.”

The last story ISLANDS BUSINESS published on tsunami contained a projected simulation of a devastating tsunami in the South Pacific. What is your comment on that projected simulation, is it too drastic, unrealistic?
“It’s rather unrealistic. First off, because the North Fiji Basin is fairly shallow, the tsunami would take a full two hours to reach Fiji. Even if the warning system did not evolve between now and then, we’d get a warning out in 15 minutes, and it would be elevated to a Pacific-wide warning by about 20 minutes. Fiji would have at least an hour and a half to prepare. I don’t think the “cry wolf” syndrome will be too bad (i.e., people inured to warnings because there are too many false alarms). The false alarm rate is steadily declining, so that the problem should decline too. It is very unlikely too that we’ll go 20 years without another major tsunami, so understanding of the threat will not be limited just to those who remember 2004. Some of the detailed destruction is overblown. While the Yasawas would be hit hard, the mainland of Viti Levu would not, because the Yasawas and all of the intervening shallow banks and reefs would provide a lot of protection. The northern coasts of Viti Levu and Vanua Levu would only suffer serious damage opposite passes in the reef. While it would be a horrible event, I don’t think it would be as severe as (the late) Mr Keith-Reid suggested. The main point, however, to Mr. Keith-Reid’s piece is that Pacific islands nations cannot afford to be complacent. I agree completely with him there.”

Is it true that unless islands countries have their own earthquake detecting equipment, the Pacific Tsunami Warning Centre (PTWC) would take at least an hour to locate the epicentre of any earthquake before it can issue a tsunami warning?

“It takes us more than an hour to verify that a tsunami has been generated. But we don’t wait for confirmation of a tsunami before issuing a warning; we issue warnings whenever we feel that an earthquake is so big, so shallow, and so close to the ocean that it poses a real tsunami threat. We can almost always work out where and how big an earthquake is, and to get a warning message out, within fifteen minutes. We are striving to reduce that time.”

How many countries that PTWC serves in the Pacific do not have earthquake detecting equipment?

“In almost all countries of the Pacific there are seismometers of the Global Seismographic Network, run by the Incorporated Research Institutions for Seismology. You can find a map of their network at http://www.iris.edu/about/GSN/GSNMaps/Maps/SubMaps/GSN_FDSN.gif. We don’t get all of that data yet, but once our bandwidth is increased we’ll bring in all the GSN data as well as all the data from seismometers of the International Monitoring System. Even more seismometers would mean faster warning, but even with the sparse network of instruments that exists, we can usually figure out the epicentre of an earthquake (though not its magnitude) in seven or eight minutes. (We can do this because seismic waves travel fast: an earthquake in Tonga will be recorded at Honiara only five minutes after it happens).”

Apart the above, what other lessons should we the small islands of the Pacific learn from the Solomon tsunami?

“Lesson number 1: If you feel the ground shake, and if that shaking continues for more than a minute, or if it is so strong that you have difficulty standing, then move away from the ocean as soon as the shaking stops. If you can get to 20 metres elevation, or get half a kilometre in from the ocean, your chances of survival are excellent.”