Tsunami Plan Revised

The Ministry of Civil Defence and Emergency Management has revised its Tsunami Advisory and Warning Plan.

Thursday 25th February 2010

Following criticism of its management of recent tsunami scares, the Ministry of Civil Defence and Emergency Management (MCDEM) has revised the way it communicates tsunami information with emergency managers, broadcasters, news agencies, and others with responsibilities during times of emergency.

Early on the 4th of May 2006, a magnitude 7.9 earthquake struck near Tonga’s capital Nuku’Alofa. The Pacific Tsunami Warning Centre issued a tsunami warning for New Zealand complete with estimated arrival times at populated centres. The ministry assessed the situation and decided that a threat did not exist, but failed to report its judgement to news media. Meanwhile, the warning was picked up by international television broadcasters who were amazed at the lack of response in New Zealand. Local news media sought clarification of the situation and were unable to contact ministry officials.

In the absence of official advice, several hundred people in Gisborne decided to evacuate to higher ground. The time of danger passed, and a wave of significant size was not generated.

When two strong earthquakes struck the Puysegur Trench south of New Zealand on the 30th of September 2007, MCDEM’s advisory and warning plan was found to be wanting. The first quake of magnitude 7.3 struck at 6:24 on the Sunday night, followed by a magnitude 6.8 event at 10:48 p.m. For the first event, the ministry issued a statement to some media outlets an hour after the earthquake, and cancelled a tsunami advisory for Otago and Southland five minutes after a 250 mm sea level change was recorded at Dog Island near Bluff. Despite a Memorandum of Understanding arrangement with Radio New Zealand to broadcast tsunami information, the National programme hourly news programme did not mention the earthquake and the tsunami likelihood until 8 p.m. Both Radio New Zealand and MCDEM ignored the second earthquake.

New Zealanders had to rely on two overseas websites, the Pacific Tsunami Warning Centre and the U.S. Geological Survey for timely earthquake and tsunami information. The PTWC site was difficult to access because of the heavy traffic, and its bulletins advised that a tsunami could be locally destructive, but a widespread tsunami threat did not exist. The Dog Island measurement was the largest wave recorded for these events.

When a magnitude 8.0 earthquake and tsunami struck Samoa and Tonga in September 2009, the Ministry of Civil Defence and Emergency Management again seemed to be dithering. With news programmes painting a picture of a steadily worsening tsunami situation in both Samoa and Tonga, the ministry’s response was invisible to news media and the public. It had actually assessed and activated its plans but, while emergency managers around the country knew that something was happening, public statements were not issued. The ministry compounded its error by participating in a news interview during which an appalling lapse occurred. Its representative sounded distant from the issue at hand, and stated that it would issue a tsunami warning, if necessary, “in about an hour” which coincided with the estimated time for the first wave to arrive at East Cape.

The need to dramatically improve its public communication had failed to register with the ministry after three serious events over three years. The ministry maintained a 1960s “we know best” attitude in an environment in which the public is better informed and expects its officials to be responsive. Criticism was widespread in both print media and television.

With the issuing of its Tsunami Advisory and Warning Plan this week, the Ministry of Civil Defence and Emergency Management seems to have almost “got it.”

The plan covers a number of scenarios from locally generated tsunami to distant events that allow hours of preparation. Tsunami waves travel at speeds of 700 km/h in deep water, meaning that locally generated tsunami will come ashore within an hour of being generated, waves from a regional source arrive within 1 to 3 hours, and distant events take more than 3 hours to arrive. Hence local source tsunami allow little or no preparation time, regional source tsunami require prompt action by officials, and distant source waves can be more thoroughly examined before warnings are issued. The flow of information from tsunami experts and agencies is defined, events likely to generate tsunami risk to New Zealand have been identified and, most importantly, the ministry’s communication responsibilities have been clearly laid down.

The Ministry for Civil Defence and Emergency Management will issue one of three notifications in response to a tsunami event. They are: a National Advisory – either “no threat to New Zealand” or “potential threat to New Zealand” or a National Warning – “threat to New Zealand.” At the completion of each incident, a cancellation message will be issued for either the National Advisory or National Warning. Where public broadcasts are necessary, it will issue a “request for broadcast of an emergency announcement” to media organisations that have signed a Memorandum of Understanding with the ministry. The announcement request will specify whether the message is to be repeated at 15 minute, 1 hour or similar intervals.

Very little can be done to alert the public to a tsunami generated by a large earthquake close to New Zealand. This was dramatically illustrated on the 26th of March 1947 when a magnitude 7.1 undersea earthquake struck off the North Island’s East Coast. The first wave swept ashore seven minutes later trapping people, with the largest wave arriving shortly after. Run-ups of up to ten metres were recorded as a result of this earthquake.

Other local events can cause sudden changes in sea, river and lake levels, and are discussed in the plan. These include large local earthquakes that don’t generate tsunami by moving the seafloor but cause a seiche in lakes, rivers and harbours as the energy released causes the water to slop back and forth. This happened in Wellington harbour after the magnitude 8 earthquake in 1855, and in rivers as far north as Wanganui.

Undersea landslides can occur with or without an associated earthquake, and can trigger tsunami waves. The plan notes that the giant Ruatoria debris avalanche of about 170,000 years ago generated a wave thought to be a staggering 125 to 700 metres in height, when the 3000 cubic kilometres of debris moved at high speed.

Fortunately such large landslides happen very rarely, but the studying of undersea landslide risk has resulted in a new, very sound piece of advice about long duration, weak earthquakes being added to the conditions that should trigger an immediate response from the public.

The plan states:
“Persons in coastal areas who:
• experience strong earthquakes (hard to stand up);
• experience weak earthquakes lasting for a minute or more;
• observe strange sea behaviour such as the sea level suddenly rising and
falling, or hear the sea making loud and unusual noises or roaring like a
jet engine;
should not wait for an official warning. Instead, let the natural signs be
the warning. They must take immediate action to evacuate predetermined
evacuation zones, or in the absence of predetermined evacuation zones, go
to high ground or go inland.”

Tsunami arising in the south-west Pacific (referred to as regional source tsunami) provide some opportunity to issue warnings, depending on where they were generated. The threshold for announcing a warning of a tsunami threat from this area for New Zealand is set at a magnitude 8 or higher earthquake at a depth of less than 100 km or a warning for New Zealand being issued by the PTWC. Earthquakes with magnitudes between 7.5 and 7.9 at depths less than 100 km or a PTWC watch for New Zealand would trigger an advisory of a potential threat. The key difference here is that the warning is more important than an advisory.

For tsunami generated at a distance, the plan identifies six areas where earthquakes could generate waves that threaten New Zealand. They are: South America, Mexico – Central America, Cascadia (North-western United States of America and Vancouver Island, Canada), Aleutians – Rat Islands, Kamchatka – Kuril Islands, and Japan. The thresholds for earthquakes that would cause an advisory of a potential threat to New Zealand to be announced vary between the zones, from magnitude 8 or above and less than 100 km deep, to undefined. In the latter case PTWC bulletins would provide guidance. The greater transit time will allow the advisory to be upgraded to a warning if necessary, as wave measurements and experiences are reported by communities closer to the source.

Having identified the three types of events – local, regional or distant, the plan lays down the procedure that will be followed when a PTWC bulletin or earthquake notification is received. The available information will be examined against the documented criteria and, if necessary, a panel of tsunami experts will convene by audio conference.

Time to react is tight for regional source tsunami. Despite automation, there is still a lag between an earthquake occurring and information being posted on the US Geological Survey or GeoNet websites. GeoNet is not manned 24 hours a day, but has duty officers on a 20 minute response time, 24 hours a day, 7 days a week. Thus, half an hour could elapse before an earthquake is analysed and information becomes available to officials.

The action guide allows MCDEM 15 to 30 minutes from receipt of either a GeoNet earthquake report or a PTWC bulletin to issue an appropriate response in the form of either an advisory or warning message to local and regional emergency managers. Another 30 minutes has been allowed to issue a public announcement via nominated media organisations.

In a worst case scenario, an hour-and-a-half would elapse before radio stations were able to issue a tsunami warning. In practice, the response time could well be better, but this is a reality in a nation where emergency management involves a number of organisations, devolved responsibilities for warning communities, and after-hours management on a call-out basis. Hence tsunami generated in the Kermadec Islands or Auckland Islands or areas in-between may come ashore without a public warning being issued.

This means that concerns raised over the lack of MCDEM communication during the Puysegur Trench event of September 2007, and the offshore magnitude 6.8 Gisborne earthquake of December 2007 have not been addressed by this plan.

However, the plan should result in improved management of warnings for events such as the Samoa-Tonga tsunami of September 2009 and the Tongan earthquake scare of May 2006.

Having identified the situations which would trigger the announcement of a potential or actual threat of tsunami for New Zealand or the detection of an actual tsunami wave, the plan is less clear on when the ministry will issue a public statement when it considers that there is no threat to New Zealand. This is a condition that has caused controversy in the past, and seen the ministry not meeting public expectations.

The plan describes the process for issuing a public announcement to the effect that experts consider there is no threat to New Zealand coastlines. However, the criteria which would trigger such an announcement are only partially defined. The trigger points specify earthquakes by magnitude (the amount of energy released) but not by their felt effects i.e. the amount of shaking that New Zealanders could feel when a strong earthquake occurs close to New Zealand.

This is a persistent blindness in the ministry’s planning, and clearly at odds with the expectations of many New Zealanders. When we again experience earthquake shaking of the intensity of the Puysegur Trench events of September 2007, there will be public uncertainty over whether to evacuate low-lying areas and when it is safe to return. The decision to move to safer ground is addressed by the ministry’s recommended response to having difficulty standing up during strong shaking or experiencing sustained tremor for longer than a minute. However. When is it safe to return? Is there a risk that successive waves will increase in height?

MCDEM has a responsibility to manage expert analysis of such events and provide authoritative advice on when the period of danger has passed.

Another glaring omission is the failure to specify a standard for a public warning system of sirens, horns or other technology for deployment in high-risk areas. The plan specifically devolves this responsibility to local authorities. It states, “Matters such as local public alerting systems, possible areas of inundation and evacuation arrangements must be incorporated into local plans.”

We tend to be more mobile these days, and many people move between the regions as they pursue work and their careers. This means that we identify less strongly with our communities and are less likely to know about local conventions than perhaps we were in the 1970s. Only when an incident occurs will we wonder whether the sirens we can hear are calling volunteer firefighters to the station or warning of a tsunami alert.

Compare this with the situation in the 1970s: From the Gisborne and District Telephone Directory, 1978, “The Civil Defence warning signal for all emergencies is a series of intermittent siren or signal blasts of two or more minutes duration. This means tune into your nearest operating radio station and follow instructions.” Simple, universal, unambiguous.

The 60-page Tsunami Advisory and Warning Plan is worth reading if you are interested in understanding how the Ministry of Civil Defence and Emergency Management expects to manage tsunami events in the future. It contains templates to be used by ministry staff, and is necessarily detailed and plodding in places, but provides good insight into emergency management in New Zealand.

For those with an interest in the science of tsunami it contains a fascinating plain english annex on tsunami categories and threat. The various causes of tsunami including earthquake, submarine landslide, meteor or asteroid impact, and undersea volcanic eruption or collapse are described. Such activity in certain locations on the planet presents greater or lesser risk to New Zealand, and the known wave guide and deflector zones which help or hinder waves reaching our shores are mentioned. Some of the rare scenarios are a little scary, so it’s not bed-time reading for the nervously inclined!

The report addresses some but not all of the concerns raised during recent tsunami incidents.

[ Back to Wild Land MAIN PAGE ]

2 Responses to “Tsunami Plan Revised”

  1. Flying Deldas says:

    Excellent writing Kim. Of great interest.

  2. Lizzie from Gizzie says:

    This was well timed huh

Leave a Reply