HIGH FLYERS THINK TANK

Extreme Natural Hazards

University of Melbourne, Tuesday 30 October 2007

Mr Michael Tarrant
Emergency Management Australia

Michael Tarrant has worked in the disaster management field for the past 20 years and currently manages the Emergency Management Australia research portfolio. His main areas of interest include managing non-routine risk, disaster recovery, business continuity, resilience, simulations and public administration. He has presented and run workshops around the world and has worked at a national level on the development of emergency management publications. Michael has been a member of Standards Australia Risk Management Committee since 1998.

Keynote address: Resilient communities and organisations – pervasive concepts

(Click on image for a larger version)

After some 20 years of working in and observing emergency management practice, resilience appears to be an area of real interest.

So, is it a fad?

First of all I would like to look at a bit of history and at some definitions and concepts in the area. You may be unfamiliar with some of these. I then want to have a quick look at risk, uncertainty and complexity. They are very much the core problems and issues that have to be addressed at the extreme end of the risk spectrum.

(Click on image for a larger version)

To understand the present and the future you have to have an understanding of the past.

War (civil defence)

Emergency management and disaster management in Australia really evolved around the experience of the Second World War. The institute where I work was established in 1956, very much in the mould of civil defence, the Bomb and the cold war: what were we going to do after our cities had been attacked with nuclear weapons?

Response and relief

The thinking in terms of disaster management, or managing these big shocks to our system, was very much a response and relief approach – very simplistic thinking, by and large. There is a wonderful example of the plan the British put together in which, after a section of London had been nuked, they were going to take the buses off line after the morning peak hour, evacuate the 750,000 people that had been affected by the bombing, and try to get most of the buses back on for the afternoon peak. So our thinking has come a long way since the civil defence days.

Natural disasters

As things settled down in the postwar period, natural disasters were starting to get on the agenda. It was a really interesting decade for those interested in shocks. We had polio in 1953, the big floods through New South Wales and Victoria in 1954–55 and then a very significant flu epidemic during 1955. So it was a pretty challenging time.

The 1967 fires in Hobart were a major turning point in terms of natural disasters. Lance Barnard, then a budding politician in Tasmania, observed the problems posed to government by these fires and decided that if he ever got into government he would do something about it. In 1972, when Barnard was elected, the civil defence college was plucked out of the Department of Interior and relocated into the Defence Department.

The Natural Disasters Organisation was born in 1974. That was nicely cemented by Cyclone Tracy and the Brisbane floods, two very significant extreme events. The Americans followed us with the formation of FEMA (Federal Emergency Management Agency) in 1979. US President Jimmy Carter set that up as a result of yet another series of very significant hurricanes.

The idea of how to allocate resources to manage disasters was now broadening. There are not many options when your context is civil defence – you've got bombed cities and people just need relief and support, whereas with natural disasters there are possibilities for action before the event.

PPRR

Australia picked up the concept of PPRR from a Governors' Conference in America in 1976 – where this whole idea of Prevention, Preparedness, Response and Recovery was agreed to as a framework for thinking about how to manage the effects of shocks to society. Thinking tended to focus on natural disasters. Agencies were set up to think about a broader range of ways of managing exposure to risk.

Technological disasters

In the 1970s, starting off with Rachel Carson's The Silent Spring, there was concern about chemicals and concern about the environment – very similar to the wave we are getting now with climate change. There were the sentinels talking about these issues very early on, and slowly momentum built with societies and decision-makers and people start changing their way of thinking about the world.

During the 1970 and 1980s there was a series of high profile technological disasters and we start to see the first serious interest in failures of socio-technical systems: the Flixborough chemical plant explosion, the Herald of Free Enterprise ferry capsize, the Bradford stadium fire, the Hillsborough stadium crush, the Piper Alpha oil rig, the Kings Cross underground fire, the Seveso chemical release in Italy and the Mississauga hazardous chemical train wreck in Canada. These incidents stimulated research interest in complex system failures. Key researchers in the field were people such as Barry Turner, James Reason, Aaron Wildavsky, and Charles Perrow. The other influence on disaster management thinking from this period was the work on risk communication by authors such as Slovic, Fischhoff , Douglas and Kasperson.

Lifelines to critical infrastructure

One very important branch of thinking in the complex systems sphere was the idea of lifelines, linear engineered structures on which our society and economy depend. These include power lines, water and gas mains and telecommunications The whole issue of these linear engineered structures were of interest to the disaster management field.

However, the idea of lifelines very rapidly morphed into critical infrastructure, with the sudden realisation that our economy and our society depend on an interlocking web of networked services. In a slower, less sophisticated economy we were able to tolerate large outages.

Developments in management over the past 30 years have often produced very significant gains in productivity and efficiency from organisations in the private and public sectors. One of the very interesting trends is the significant increase in reliability of standard services. That was one of the big results of the move to centralisation: to get away from small utility providers and each town having its own systems: 'What a good idea, we'll connect them up in grids, and then we can swap power around.' It was a very good approach and it has produced significant gains in reliability.

The only problem is that when it does fail, it can fail catastrophically. And then the surprise factor of not having something that has been very reliable, with all the small outages being minimal, suddenly hits and produces significant shocks to the system. This is not to say that networking is intrinsically bad, we just need to develop approaches for what we are going to do when there is a major failure. As a society we have been too willing to consume the entire efficiency dividend and not invest in managing the risk we have transferred into the future. There was a need for a more holistic approach which leads into risk management and whole of government approaches.

Community focus, ERM and whole of government

I would like to make the point that I am not describing a linear transition in the evolution of thinking about disaster management. Organisations and institutions are spread across this framework, depending on roles and cultures. It's very much an onion ring: we have become more and more sophisticated, with more and more thinking about how we approach these issues.

The idea of communities being involved, that the state and our organisations can't do it all, and to effectively manage these sorts of events and to get good outcomes for the community and our society and our economy requires more inclusive approaches to managing the risks of extreme events.

Organisations such as the CFA (Country Fire Authority), in particular, were ground breakers and started off some terrific work in the 1980s and into the '90s with programs such as Community Fireguard and participatory models for communities to be involved in reducing their risk from extreme events.

The other one is emergency risk management. Until well into the 1990s, emergency management was very much a marginal activity. So the push towards using risk management centred on two really important ideas. Firstly, we wanted a conceptual framework and a language that fitted the mainstream and, secondly, we wanted to be able to broaden the range and focus of emergency management.

The definition of risk was a very important starting point. Risk is the chance of something happening which will have an impact upon objectives. It focused the role of emergency or disaster management on objectives. Whose objectives? The community's objectives? People in communities have all sorts of objectives. How do we manage under those circumstances? How do we motivate people? How do we get good outcomes? For example, how do we trade off the interests of the tourist sector against those of the timber industry? All those sorts of issues needed to be addressed. The critical point was it helped the emergency management sector to recognise the complex range of interests that motivated people's decisions and choices. It provided a language and conceptual framework that allowed for the potential to integrate with the mainstream.

Security

Security emerged in 2001, in particular, but I would like to see a much broader sense of the word 'security'. It certainly has driven significant investment by governments.

Core part of governance

In the past 10 years we have seen a growing recognition that the management of major shocks to an organisation is a core part of good governance.

Complexity and turbulence

That brings me to our current environment. There has been steadily growing recognition of the complexity and uncertainties of the environment in which organisations and communities exist. The environment involves complexity of systems, globalisation, interconnectedness, interdependencies and just-in-time. All of these processes and contexts have brought a whole range of criticalities into the way we manage our economy and society. There is a real need to develop approaches that maintain our quality of life but effectively cope with a more complex and turbulent world.

Resilience

At present resilience is just a word – we need to flesh it out, we're building it up as a concept – but it is certainly a way of thinking about the world. And you can see over the last 50 years a steady level of evolution in the way disasters and their management have been conceptualised.

Context

I will make a few quick comments about the issues highlighted on this slide, without going into them in any great detail. There is a wide range of drivers in our environment causing turbulence and providing challenges to our systems. Our context is very much 'How do we deal with non-routine risk?'

Donald Kettl was an early sentinel in this area, he started to think about how disasters or non-routine risks produce particular challenges to our society and its institutions. There are some real problems concerning how the structure of government, or the structure of business units in an organisation, make dealing with cross-cutting issues very difficult. One of the trends in the last 20 years, both in the public and private sector, is driving efficiency: 'How do we optimise? How do we get it more efficient – more and more and more efficient?' Now, that's not a bad thing. Don't get me wrong; there is no way I want to go back to a 1960s public service. We don't want to go back to that sort of level of lock-step, mechanistic development.

However, one of the downsides is that we have consumed the entire efficiency dividend from our reforms. We haven't reinvested a lot of the gains we made in managing the risk we have transferred into the future. We've taken the lot, and I don't think we've invested very much in what we are going to do when things don't go right. And I believe this holds for both the public and private sectors.

Walker and Salt put this quite neatly. Although they are talking very much in terms of resources, I would argue that it applies across the board. Greater efficiency cannot solve all our problems, and is in fact creating more. We are actually building and reinforcing silos. The whole management ethos over the last 20 years has been around key performance indicators and key result areas; the focus has been on continually driving efficiencies, but what we are doing is forgetting about the links between the silos – we are not thinking about the whole.

How governance arrangements in the public and private sectors manage cross-cutting issues such as extreme events need further development. These are the issues we are trying to grapple with.

Definitions

Regarding a definition of crisis, people ask me whether heatwaves count as a disaster. Well, it depends. But I think the work by Paul 't Hart is really important – the idea of crisis as being the turning point where a community says, 'That's not fair. That's not right.' We lose old people in every heatwave, but when it gets to a certain threshold it pushes people's tolerance and they say, 'Hang on, that's not right.'

In the heatwave that hit France a few years back they lost about 35,000 elderly people who died in their apartments. So that really defined itself as a crisis, and I am sure that if it happened here in Australia, people would start saying, 'Hang on, that's not right.'

When a crisis occurs it usually requires resources to be shifted around and the ability to run in 'non-routine' mode. I will develop that idea in a little more detail.

Disasters and emergencies are not the same thing. An emergency is a short-term event requiring immediate and predetermined actions by trained individuals. A classic example is the recent fire and crash in the tunnel in Melbourne. That emergency was well managed.

A disaster is very different. It is very much about having to run your system in non-routine mode, where you have to be able to get departments, groups and communities to operate very differently on Thursday from the way they operated on Wednesday. And how do we support people through that? How do we support our organisation through that? How do we support our staff through that? That's the challenge. That's why people develop and write disaster plans – because it's a roadmap showing how to run your organisation or whatever you are responsible for in non-routine mode. And that's basically all that an emergency or disaster plan is. It is a set of arrangements for being able to run in non-routine mode.

Let's have a look at what I mean by that.

(Click on image for a larger version)

If you have a look at a risk spectrum from major to minor risks – and it is a spectrum – you see that on the left-hand side minor risks are people tripping over a fault in a carpet, or a bit of foliage hanging across a stop sign. They can be handled through the normal service delivery of an organisation.

However, on the right-hand side, major risks are those which have the potential to generate disasters, the low-probability, high-consequence situations where organisations will be tripped into non-routine mode.

That is where those terms sit in terms of conceptual framework. Incidents and emergencies sit very much on the left-hand side. Okay, an emergency is a hard day at the office for the emergency services, but they are still all sitting within their role. In a disaster, it is very different indeed for significant parts of organisations and communities.

And, for those that come from a corporate world, on the right-hand side are the risks that would require your organisation to use approaches such as business continuity; on the left-hand side it is just business as usual. Your normal arrangements handle and deal with those sorts of risks.

(Click on image for a larger version)

Non-routine risks are a subset of the risk spectrum. We are talking about low-probability, high-consequence situations. These risks typically have high levels of uncertainty, complexity and ambiguity. This whole field is permeated by these characteristics. We are not dealing with areas where you can collect data, analyse it and argue a good, logical case for investment of resources. Those sorts of approaches don't work within our area.

Prevention and mitigation are very expensive. For example, separating people and floodwaters. Virtually every settlement in Australia has been established on a watercourse, because in the 19th century the risks of flooding were far less than the risks of not having water or transport. Once effective water reticulation systems were installed the need to live so close to water no longer applied. We have this residual problem from last century, of having most of our settlements exposed to potential, significant flooding. How might this exposure alter with climate change?

Benefits of mitigation as losses forgone are issues we have to deal with.

There is a long gap between investment and benefits. That is one of the real problems: we can't show politicians and decision-makers a good return on investment quickly.

(Click on image for a larger version)

Wildavsky sees two polar opposites – this is a very important idea – between anticipation and resilience, this set of trade-offs.

Anticipation is very good for static, well-understood systems, but poor for high uncertainties and complexities. The minute you start allocating resources to deal with particular sorts of risks, you are drawing away from your pool of generic resources to deal with the unexpected. And that is a continual tension.

I am putting anticipation and resilience up as two polar extremes. Obviously, every solution is going to be somewhere in between those two; you are never going to have solutions at the two poles..

People tend to like the anticipation approach. It has a good feel about it: 'We've got things under control. We've got the plan'. Just like they had in New Orleans: 'We know how we are going to shift 100,000 people out of the city.'

So anticipation is one approach. It works very well for static, well-understood systems.

Resilience again presents the need for the development and allocation of generalised resources – being able to flexibly adapt your entity, your organisation, your company, your business, your community to deal with these issues.

And it involves toleration of variability or local solutions. It is politically very brave to be able to say, 'No, we've got to let communities make up their own minds and take responsibility.' Managing risk is a social and political process. How do we manage inappropriate use of land? Who bears the costs and who benefits? Perhaps the most efficient approach may not be the most resilient. For example, with power generation, should our society invest in a technology such as nuclear which is very expensive with very long lead times, or do we invest in a diversified range of technologies around wind, waves and solar which may not be as efficient but is diversified and very resilient.

So I hope that gives you a bit of a feel for how we are thinking, and some of the issues we have to deal with.