Following the earthquake of magnitude 7.9 near Tonga in May 2006, the Ministry of Civil Defence was caught flat-footed, failing to confirm or discount tsunami warnings being broadcast by television stations. Ministry officials dithered, creating serious public uncertainty.
Last Wednesday’s earthquake and tsunami near Tonga and Samoa resulted in tsunami warnings for New Zealand coastlines. Some have again accused the Ministry of repeating its Ministry of Continual Dithering act again, but did it?
Saturday 3rd October 2009
The Ministry of Civil Defence and Emergency Management (MCDEM) co-ordinates disaster response within New Zealand territory through a layered structure involving local authorities and government departments. It faces a wide range of potential civil emergencies, both man-made and natural. The natural disasters include floods, earthquakes, tsunami and eruptions encompassing the full spectrum that might be expected in a geologically active country straddling the roaring forties. The emergency situations can be grouped into three categories: immediate (for situations that are already in progress by the time they come to the attention of authorities), imminent (for quickly developing situations that leave little time to respond), and forecast (events that can be monitored as they develop).
Tsunami waves coming ashore on New Zealand coasts can be divided into the three categories ”“ those caused by events near New Zealand coasts, waves generated by events in the South Pacific region, and teletsunami (long-distance waves).
Tsunami waves are generated by disturbances of the ocean floor such as earthquake fault movement associated with an undersea earthquake or underwater landslides. The waves move at great speed in the open ocean, typically 700 km/hr and, like the waves that bring our daily tides, tsunami are very long. Tsunami waves are measured in hundreds of kilomteres whereas the wave that brings a tidal change is more than 2,000 kilometres in length.
This is where the confusion begins. Tsunami waves are not tidal waves. They are also very different from the waves that might occur in a storm swell in, say, Cook Strait where waves of 6 metres or more in height are not uncommon. These waves are very short by comparison (just metres or tens of metres) and hence have a shorter run-up when they strike the coast.
When tsunami enter shallower water, their leading edge slows dramatically, allowing the trailing part of the wave to catch-up. As this occurs the relentless push of the wave causes the leading edge to rise higher as the water catches-up. This can allow a wave of less than a metre height when mid-ocean to rise dramatically in harbours and on coasts where the ocean floor slopes upward.
Of the three categories of tsunami that the MCDEM has to respond to, the regional events are the most problematic. Locally generated tsunami are impossible to address at present. They arrive so quickly after a large earthquake or landslide, that there is little or no time to issue warnings. Common sense must prevail. Anyone feeling a strong earthquake that is thought to be local in origin, should move to higher ground quickly if they are near the coast. Teletsunami have arrival times of many hours, giving plenty of time to assess the experience near the earthquake epicentre and calculate the size of the wave that might strike the New Zealand coastlines. In the case of an earthquake off the coast of South America, any resulting tsunami wave could take 12 hours or more to arrive.
This is not the case for regional events. Last Wednesday’s tsunami was expected to make first contact with New Zealand 3 hours after the earthquake, and a tsunami generated near Raoul Island would take about an hour to arrive.
In such a case, MCDEM has three functions to perform in a short period of time ”“ assess, direct and manage. Assess the risk. Direct a response. Manage the aftermath.
The magnitude 8.0 earthquake occurred at 6:48 a.m., initially reported as magnitude 7.9 by the Pacific Tsunami Warning Centre at 7:04 a.m. and magnitude 7.9 by the U.S. Geological Survey a few minutes earlier. By 7:15 a.m. MCDEM should have been in the assessment phase.
The duty controller should have checked the event against a set of criteria, and the magnitude, shallow depth and undersea location should have flagged a tsunami as possible. An audio conference should have been convened by 7:15 a.m. involving GeoNet’s duty seismologist, the Director of Civil Defence and up to three others from the Ministry.
The convening of the conference should have raised MCDEM’s alert status from 0 to 1 ”“ the activation phase. This should have been communicated via MCDEM’s website to show the public and interested agencies that the Ministry was responding.
By 7:25 the audio conference should have been concluded, having judged that a tsunami risk to New Zealand did exist. The PTWC bulletin issued at 7:04 a.m. issued a warning for New Zealand, indicating a first arrival time of 9:44 a.m. at East Cape. MCDEM’s Alert Status should have been raised to 2 ”“ action being taken. This was actually done, but it wasn’t until about 7:45 that media were able to advise that the Crisis Management Centre had been activated.
Behind the scenes, MCDEM was already alerting regional Civil Defence controllers, the text messages being issued at around 7:30 a.m.
Unfortunately for MCDEM, this activity was largely invisible to the general public at a time when media were talking to people in Samoa and learning that damaging tsunami had been generated, carrying away people and property. It is alleged that Radio New Zealand tried to contact the Ministry at about 7:45 a.m. and was told to call back after 8 o’clock.
So far, so good. Despite being invisible to the public, the Ministry seems to have done well up to this point, but soon tripped over itself. Rule number one during the direction phase of an alert should be no interviews. The focus should be on issuing factual statements every quarter of an hour to all media. These should be concise, and show progress since the last statement. No interviews. Interviews distract key people from their tasks and, as was the case on Wednesday morning, create confusion. The interviewers usually have little understanding of the science involved and tend to either ask irrelevant questions or try to sensationalise the event by criticising the response. Some of the criticism might be justified, but levelling it at a time when the public needs information does not help the situation.
Engaging in interviews showed an appalling lapse of judgement by Ministry officials, and simply reinforced the view that the Ministry was again dithering, just like in May 2006. The opposite was true, but the alleged comment on television that an announcement on the risk would be made “in about an hour” which coincided with the first arrival at East Cape, was bad judgement indeed. No doubt it was due to being flustered by the badgering of the interviewer, but avoiding interviews would have prevented the situation from occurring.
With radio and television carrying stories of destructive waves at Pago Pago, the Ministry should have notified that it was at Alert Level 3 (warning in place) by 7:55 a.m. This must appear on the MCDEM website and be communicated to the media so that the public can see some decisive action being taken. Silence is seen as dithering. Once again, bulletins should have been issued regularly, showing purposeful action being taken.
Reported activity over the next two hours of Wednesday morning has focussed on regional response ”“ evacuations on the Coromandel Peninsula, groups and businesses moving customers to higher ground in the Bay of Plenty etc. The decentralised Civil Defence response appeared to be working apart from foolhardy interviews given by MCDEM staff.
News of the cancellation of MCDEM’s warning for New Zealand filtered through the media at about 11 a.m. but was badly timed. The tsunami gauges at Raoul Island had started registering sea-level disturbances of up to a metre at about 10 a.m. and these were continuing, with the gauges at East Cape and North Cape showing an abrupt increase in disturbance at 11 o’clock.
The commencement times were about two hours later than forecast first arrival times for Raoul and just over an hour later for East Cape. The Pacific Tsunami Warning Centre does not own chapter and verse on tsunami forecasting, their warnings include estimated times. With the benefit of hindsight, the cancellation was right in principle, but could have encouraged the foolhardy to return to turbulent coastal waters. In any case, better to be judged for a bad decision, rather than no decision at all.
In day-to-day operation, MCDEM has to act like any other government department. In times of emergency, it has to become a very different beast, altogether. It must be authoritative and assess, direct and manage ”“ quickly. On Wednesday morning the media minders should have been locked in a quiet room out the back with a nice sudoku. On the front line should have been decisive communicators, well-versed in Civil Defence operation, the science behind the emergency, empowered to act and direct. This is where MCDEM let itself down.
The Ministry has made improvements since the appalling performance of May 2006, but communication is still its Achilles’ heel.
Closely following, is its inability to grasp the need for a public national warning system. The plethora of regional warning systems may meet the decentralised model that the Ministry adores, but it doesn’t work. There is too much uncertainty. The problem lies with the fact that people are increasingly mobile. We stray outside our home regions for work and holidays, and will only think about warning systems when an emergency presents itself. By then it’s too late to wonder whether our present locality uses sirens, incoherent helicopter messages, radio, subscribed text messaging, pagers, roving police cars, automated diallers, door-knocking or semaphore flags to alert us to danger.
By gaining a backbone and adopting a few national standards, MCDEM could streamline its performance when an emergency looms. Thought should be given to a five stage alert level system, where,
0 = no event
1 = activated
2 = acting
3 = warning
4 = event pending
5 = emergency in progress
and the status of the Ministry and its plans should be available to the public at all times.
It may seem an academic exercise to criticise MCDEM for poor performance when tsunami events are so infrequent. However, if earthquake activity goes through a period of increased activity, then the situation will change. What’s more, looming technology deployments will present the Ministry with more potential crises which fit into the imminent category. Metservice is deploying thunderstorm radar which will dramatically improve its ability to forecast tornado activity. Deployment of deep-sea wave buoys in the South Pacific will present another dilemma ”“ who will act on the information they provide?
[ Back to Wild Land Main Page ]
Well done! Good writing and excellent, plain speaking explanation of what is required. Sure would save a lot of people getting their t..s in a tangle!
I thought they did much better this time – although I did notice on Radio Live they were giving interviews…
Oh well… practice makes perfect and at this rate they will be the best in the world soon enough!
🙂