Natural Disaster Risk Communication – No More Chatterbox

Natural Disaster Risk Communication – No More Chatterbox

When natural processes impact the human environment we call them natural hazards. When natural hazards impact the human environment, we call them natural disasters. When making predictions about what might happen in future, we call this the assessment of natural disaster risk.

Natural disaster risk is typically defined as a combination of the likelihood of an event occurring and the consequences should the event happen. The spectra of risk then typically becomes a range of relative risk - Very High, High, Moderate, Low, and Very Low. These risk levels are then considered to be variously either acceptable, tolerable, intolerable, or unacceptable.

However, the current language around natural hazards and natural disaster risk is not only unhelpful, it is also misleading. Significant consequences can all too easily be “risked away” by changing the assumptions made about the likelihood of an event to reduce the chances of a natural hazard occurring. However, it is the lack of clear communication of the possible consequences of natural hazards that have led to the greatest level of misunderstanding about natural disaster risk.

In December 2020, Radio New Zealand reported that the Alert Level of Ruapehu had been raised to "2". When asked what this meant, the GNS duty volcanologist Mike Rosenberg said "People should be aware, as I'm sure they are, that Ruapehu is an active volcano and things can change very quickly. People should have a slightly higher level of awareness and pay attention to the information that's provided when they go to the mountain."

A year earlier, in December 2019, another active volcano had had its Alert Level raised to 2. It was reported that GNS Science and the National Geohazards Monitoring Centre continued to closely monitor Whakaari/White Island for further signs of activity.

Although this meant that an eruption could occur at any time, it did not mean an eruption was likely to occur. That implied an eruption was not likely to occur.

Should the public stay away from the volcano? Was Alert Level 2 sufficient communication? On a bright sunny day on 9 December 2019, Whakaari/White island erupted. 22 people on the island at the time have since died and more were seriously injured. What was the risk and was it acceptable?


"Likelihood", once a binary term (either likely or unlikely) has now been diced into an array of relative probabilities, including qualitative relativities such as very likely, moderately likely, unlikely, very unlikely, and extremely unlikely, or quantitative relativities such as Almost Certain (1 in 10 chance), Likely (1 in 100), Possible (1 in 1000), Unlikely (1 in 10000), Rare (1 in 100,000) or Barely Credible (1 in 1,000,000).

Most people (scientists, and engineers included), do not understand probabilities and uncertainty, particularly when applied over a time span of more than 1 year. The term “return period” has unfortunately contributed to a false sense of security that low probability events present. An earthquake or a volcanic eruption that has an average recurrence interval of 1000 years, has a 1-in-10 chance of occurring in the next 100 years. That descriptor is Almost Certain. But the chances of it occurring tomorrow, or even this year, is possible but unlikely. These are the terms normally used to communicate the chances of a natural hazard occurring.

On 31 March 2009, six Italian scientists from the National Commission for the Forecasting and Prevention of Major Risks expressed the view that a major earthquake in L’Aquila was unlikely to occur, but the possibility could not be excluded. Swarms of small earthquakes had been occurring since October 2008, and many had occurred in March 2009.

On 6 April 2009, a 6.3 magnitude earthquake occurred in L’Aquila, killing 309 people.

The six scientists were initially sentenced and jailed on the basis that their risk assessment was incomplete, inaccurate, and contradictory. In short, it was considered that the scientists had given the local community a false sense of security. US scientists conveyed alarm at the idea of subjecting earthquake risk assessment to the criminal justice system. In their successful appeal, Professor Seth Stein stated that “our ability to predict earthquake hazards is, frankly, lousy”.


"Consequence" also has its own plethora of terms. What does Minor, Moderate, or Major actually mean when describing consequence? Or Catastrophic, Major, Medium, Minor and Insignificant?

The former Minister of Earthquake Recovery, Gerry Brownlee, tells the story of buying a property in Christchurch prior to 2010. When he went to purchase the property his lawyer said, “do you realise that the property is susceptible to liquefaction”?

Where do I sign”, Mr Brownlee responded.

Thousands of properties in Christchurch were “susceptible to liquefaction”. Indeed, most of the properties on the Christchurch flat land had the liquefaction hazard on the property Land Information Memoranda (LIMs).

But no one communicated what the consequential risk of liquefaction actually meant. Also, prior to 2010, there was no experience of a major liquefaction event in New Zealand.

In hindsight, what Gerry Brownlee’s lawyer should have said was this:

“Do you realise that this property is susceptible to liquefaction? What that means is that it is quite foreseeable that you will go to sleep at night in a park-like setting and wake up badly shaken and think you are living on a beach. Your house will be cracked and out of plumb and you will have sand dunes in your living room. Outside there will be water and sand as far as the eye can see, and you will not have reticulated sewerage for at least 12 months”.

Signing up to the consequence would not have been so easy.

At Whakaari/White Island a common theme from the “survivors” was that the risk had not been adequately explained to them.

The risk was expressed as being low and the fact that they were even permitted to be on the island indicated that the risk was considered by authorities to be acceptable. That was all they knew.

Like Gerry Brownlee they signed up for it without a second thought. Had they been informed that it was entirely foreseeable that they could be burnt alive or asphyxiated by fumes and ash, they could have at least been able to make more informed decisions.

The statistic most often cited when talking about natural disaster risk is that you are more likely to die on the roads. That is true, but that is not a benchmark to set the natural hazard bar against.

It is entirely foreseeable that when we travel on roads there is a risk of an accident. The term road toll implies that there is a trade-off between the costs (including deaths) of road transport and the benefits. This is somehow the price of mobility convenience, and it is clearly “acceptable”.

However, Waka Kotahi does not find any road deaths acceptable. One-star safety rated vehicles are killing our most vulnerable, namely, the young growing up in areas of social deprivation. The evidence is overwhelming. You are twice as likely to die in a one-star safety rated car in a collision than a car with a higher star rating, and 67% of our young who die on the roads do so in one-star safety rated vehicles.

Waka Kotahi could have just stated the statistics and set out the probabilities of dying in a one-star safety rated vehicle. This would have been the conventional risk management approach. However, Waka Kotahi proceeded with an educational video to demonstrate the consequences of having an accident in a one-star rated vehicle - "the safer the car, the safer they are".

This car has a one-star safety rating. What this means is that:

“In a head-on crash, this engine is going to come all the way up to meet you. You will get turned inside-out. Basically, if you get hit on the side you'll end up unrecognisable, and your daughter will be the same. Bang. Snap. Your neck - just like that. And then you will stop talking. Really quiet. No more chatterbox”.

With natural disaster risk, we, therefore, need to be much better at explaining what the potential consequences look like. We also need to talk about uncertainty.

There is a lot of uncertainty about the effects from natural hazard events. This is not just about the likelihood of an event occurring and when it might occur, but also about the cascade of effects and consequences that may or may not play out and who/what may be impacted.

We know that decision making in the wake of natural disasters is sub-optimal. Had the decision been made to withdraw from the most vulnerable land in the eastern suburbs of Christchurch before the Canterbury earthquakes it might have been possible to achieve an optimal outcome. The financial cost of relocation could well have been similar, but the social cost would have been substantially less.

Although this realisation came too late for Christchurch, there is another community where natural disaster awaits. In the township of Franz Josef, multiple natural hazards are likely to occur in the next 50 years with major consequences and which present a significant risk to both the residents and tourists alike. From an extensive local and central government-sponsored community engagement process, the community eventually settled on three options:

  1. Avoiding the hazards by moving the town
  2. Defending against nature in perpetuity or
  3. Living with nature and removing some stopbanks.

Like the Christchurch residential red zone, Government funding would be required. However, despite a lot of talk and some promised funding, nothing has actually been done to date to lessen the risk.

It is quite possible that nothing will have been done by the time the town is inundated with water and gravel from the aggrading Waiho River, or buried by landslide debris from the hillslopes above, or torn apart by a rupture of the Alpine fault which bisects the town.

We seem to be incapable of accepting the warnings that natural hazards present and acting before natural disasters occur. The current approach to risk is part of the problem.

Risk is more than just a function of likelihood and consequences. The international risk standard (ISO 31000) defines risk as “the effect of uncertainty on objectives”. ISO 31000 was revised in 2018 to help manage the uncertainty around decision-making and states that yesterday’s risk management practices are considered to be no longer adequate to deal with today’s threats and they need to evolve. These considerations were at the heart of the 2018 revision to help manage uncertainty.

Accordingly, we need to change our approach to how we assess risk. All living involves risk-taking, and from a natural hazard perspective, New Zealand is a relatively dangerous place to live, with almost every part of the country at risk from some natural hazard or another. Nobody expects absolute safety in the natural environment, any more than they expect absolute safety on the roads. What people can reasonably expect however is that the hazards are identified, in terms that they can understand. People can then make rational decisions based on a clear understanding of the possible consequences.

Although uncertain in time, many natural hazards can be reasonably well-defined spatially. Particular areas have greater or lesser uncertainty effects from natural hazards on what people want to achieve, whether this be where they choose to farm or where they choose to live. This new thinking around risk, being the effect of uncertainty on objectives, is also much more aligned with desired community outcomes, such as sustainability and resilience. This new approach is, therefore, more helpful for risk management, resilience, and adaptation planning by communities as well as by local and central government.

Yesterday’s risk management assessment approaches need to be abandoned and we need to adopt a new approach to looking at how the uncertainty around natural disaster risk affects what we want to achieve.

We need to do this now.

No more chatterbox.