Reflections on the response to the false tsunami warnings on September 12, 2007

Posted on September 23, 2007  /  4 Comments

Chanuka Wattegama who authored the primer on the use of ICTs in disaster mitigation for the UNDP looks at the responses of littoral nations from South Africa to Thailand to the Bengkulu event.

Nation special

If the 2004 Indian Ocean tsunami was a disaster marked by inaction, what happened on September 12, 2007 was marked by plenty of action, but a dearth of right action. It was certainly not an exemplary implementation of pre-determined and meticulously planned disaster avoidance activities. Did it make the vulnerable communities feel more secure? Or did it merely add to the confusion and chaos? Wasn’t what happened on that crucial evening another good lesson on how not to react to a disaster? Does this mean we still have lot to learn?Risk mitigation through disaster warning is a serious business. It is not as simple as a politician or a government official calling the national TV station and ordering evacuations or worse, the closure of roads. It is an end-to-end process with the hazard monitors at one end and communities at the other. In between are many intermediaries with defined roles. They are expected to play their assigned roles, not exceed their roles and not to play the role of others. If this balance is broken somewhere, as we have seen, it can lead to adverse consequences.

Powered by ScribeFire.


  1. A compilation on LIRNEasia’s disaster work in relation to the above article is at:

  2. The article is excellent and raises important questions and though it has a critical tone at times, which may not always be helpful in a development process scenario, or maybe I imagined it, it is nevertheless very useful.

    My opinion, as someone not heavily involved in this area, is that running through such a large scale warning scenario couldn’t have been all bad as we now get to see what happened in the process, where the congestion occured, what can be done about it, where the panic buttons were pressed, how the disaster management centre communicated with the outside world, the problems they had to deal with and what can be improved upon.

    I can be wrong here, but I believe there has not been a large scale public communication of this nature which was effectively a drill since the boxing day tsunami.

    The experience in Thailand was also interesting because even though they didn’t issue a warning, some people, nevertheless panicked. This raises the very important point that they did not effectively communicate that there was NO warning. Apparently they did so 3 hours later, which was a mistake.

    That there is no warning is as equally important to communicate as it will squash rumors. Rumors are probably the biggest evil in this kind of scenario, because they will make the public take actions they need not take.

    Whether based on the information that was available, and given that tsunami warnings seem to be an inexact science, a warning need not have been issued, I do not know.
    Thats for the expert geologists to debate.

    If indeed it is an inexact science, and one contrarian (Thailand) called the dice right, it might not necessarily mean the others were wrong to issue warnings, especially if its a law of averages kind of situation.

    If our best geologists called it one way, a non-geologists like me is not going to question them too much.

  3. Interesting response.

    Few points to consider:
    In Sri Lanka the designated authority for tsunami hazard information is the Department of Meteorology; not the Geological Bureau. This signifies that there is not much of a role that geologists in Sri Lanka can play in determining whether an earthquake in the Sunda Trench is likely to give rise to a tsunami hitting the Sri Lanka coast. The Pacific Tsunami Warning Center (much strengthened since the 2004 Indian Ocean tsunami) and the Japanese warning center are the best sources of information on the subject. They projected the arrival times for Sri Lanka (after 8 PM in the Southern coast), if a tsunami were to be generated, but did not issue any warnings as such. Clearly, there was no need to issue evacuation orders as early as 6 PM.

    It is one thing to recognize that tsunami prediction is an inexact science; it is quite something else to say it’s akin to tossing a coin. There is science here, but it’s science that has to be applied in a short time frame and factoring in multiple variables. A good example is Charitha Pattiaratchi’s paper at He is working on developing a critical latitude marker for the Sunda Trench, whereby we can be almost certain that earthquakes that occur south of the marker will not pose a danger to Sri Lanka (but will to Western Australia, where he lives). The analysis in the paper that can be accessed from the above post had been done in real time. The travel time calculations for various parts of Asia that were communicated around 5:20 LK time on the 12th of September were generated by models in real time. The science exists, we should not ignore it and put our people in unnecessary danger.