How resilient is the SAR sector?

The resilience definition most applicable to New Zealand’s SAR system comes from the United Nations Office for Disaster Risk Reduction (updated on 2 February 2017): 

“The ability of a system, community or society exposed to hazards to resist, absorb, accommodate, adapt to, transform and recover from the effects of a hazard in a timely and efficient manner, including through the preservation and restoration of its essential basic structures and functions through risk management.”

Most change is gradual, some is exponential and then there are shocks. Shocks in particular can be very destabilising.  

Furthermore, when things go wrong, they can go very wrong on multiple levels, very quickly. So it’s important that we understand how well New Zealand’s search and rescue ‘service model’ will respond under different kinds of scenarios.

We can begin to identify how resilient the New Zealand SAR sector is by identifying how quickly a normal SAR incident could escalate to a full-blown nationally significant one, and exploring how well the system responds under these different conditions. We do this by identifying two scenarios:

1. “Normal SAR event” – a normal instance where the SAR sector receives an alert seeking help for someone in distress
2. “Significant SAR event” – a situation where a normal SAR event escalates to a point where it risks overwhelming the current system capability.

To do this, we first look for a set of elements within the system that, when combined, can lead to a rapid escalation in severity.

As part of this assessment, there are two specific system risk combinations to consider:

• a ‘cascade failure’ - where one part of the system (or a ‘sub-system) fails and leads to a significant knock on effect for other parts within the same system;
• a ‘common mode failure’ - where the failure of one subsystem elements knocks out several adjacent subsystems. For example, common mode failure often originates in a technology system and spreads to associated human decision making systems.

Both events can occur in the same system.

A cascade or common mode failure could occur, for example, if there is a reduction of the number of volunteers below the level required to meet ongoing service demand. As we have discussed elsewhere in this environmental scan, changes in demography, funding and natural renewal of the existing number of volunteers could all lead to a situation where there are not enough volunteers available to respond in situations of abnormally high demand.

Cascade or common mode failure might also occur when messages are relayed between the alerting and communication subsystems, and associated allocation of resource and responsibility. This basically means that a message received at the ‘alert’ stage gets garbled, leading to the wrong or inadequate deployment of people or assets.

The very models used to identify how to respond to different incidents can embed cascade or common mode failures. For example, if the way threats are modelled do not usefully or appropriately show how inputs (e.g. data about the threat) should translate into action (the response), they can lead to failures on multiple levels.

However, we know that the risks associated with SAR incidents do not escalate in a linear or steady way – instead they tend to escalate rapidly (more like a logarithmic or exponential curve). 

As outlined above, we have also identified multiple potentially volatile changes in the environment that will significantly affect both the demand and supply of search and rescue services in the next 5-10 years. 

These include both external factors (e.g. shifts in weather, demography, tourism levels and politics) as well as internal factors (e.g. changes in numbers of volunteers and funding for search and rescue work).

When you combine the potential for SAR events to escalate quickly with major system-wide changes due to the environment, a highly dynamic medium-term picture emerges. This leads us to the tentative conclusion that the system could come under significant stress within the next 5 years, and more probably within 10 years.

A system failure could either be local, or could even ultimately require a National Security System response.

For a public with a very high expectation of delivery quality, all it takes is for one significant combination of events to overload SAR sector capacity.

To illustrate, even if we hold the population, volunteer numbers and operational area relative to assets stable, we can explore three climate change projections that would impact both category 1 and 2 operations:

1. Higher temperatures in the summer holiday period, with heat stress increasing mortality risk.
2. More intense rainfall, leading to an increase of flooding events.
3. Higher winds with an increase in serious storm events, heavier swells and more significant storm surges in areas of prevailing westerlies. This will impact exposed and elevated areas, as well as Cook and Foveaux Straits.

Assuming that the present system can account for a climate event like intense rainfall, the question to ask is at what point does a local failure become un-manageable. This could occur, for example, if any of the following kinds of events happen at the same time as another one:

• the 1991 collapse at Aoraki/Mount Cook
• the 1995 Cave Creek or the 2010 Outdoor Pursuits Centre tragedies
• the Ansett Flight 703 crash in the foothills of the Tararuas
• the LandSAR incidents as experienced in July 2017
• the widespread 2017 flooding event in the South Island.

This is the kind of risk that emerged recently for the Ministry of Civil Defence and Emergency Management when a single earthquake based in Kaikoura had a wide range of impacts at multiple points across New Zealand, including deaths, severe infrastructure damage, liquefaction, land slips and so on.

This could be one of the risks outlined above which impact directly on demand or supply of search and rescue services. But there could also be indirect risks that emerge. 

For example, at a November 2017 workshop with the NZSAR sector, one participant saw a risk of investing in assets that could prove to be redundant if were to be a downturn in tourism because of a future economic crash. Given history is littered with regular economic crashes of one sort or another, this situation is plausible.

When asked how prepared they thought the SAR sector was for such an event, one SAR sector member said the sector was “quite well prepared. Commenting on a draft version of this scan, they said

“...the focus on major environmental events as a SAR risk is questioned. This is more to do with Civil Defence. Not to say, though, that we should not do more to encourage contributing in a civil defence emergency…The promotion of the “One SAR” concept including all community emergency response organisations may be worth exploring.”

LandSAR also appeared to be relatively relaxed about culminating major event risks ahead, while observing that the sector, as currently run and resourced, may never be able to fully meet the public’s expectations, saying:

 “The sector is generally well prepared. Rare, large SAR events would likely expose some shortcomings. Standards may fail to fully meet changing societal expectations (ie. people increasingly expect the highest quality service where human life is at stake, and volunteer SAR models cannot always deliver this). 

The sector is prepared to a certain degree, but there are always limits that could be exceeded.  I think it very unlikely that terrestrial SAR services will be overwhelmed by any alignment of purely SAR responses. Some of the climate based scenarios fall more in the realm of Civil Defence responses and the SAR sector’s ability to assist with these could be tested, in large events of this nature.”[i]

---

[i] LandSAR (2017).  Feedback on the draft NZSAR environmental scan, received via email, 23 November 2017, 8.26AM.