Artificial [Un]intelligence and Disaster Management

 

There is currently intense interest in the possible use of artificial intelligence (AI) in the management of disasters. To what extent is this a real prospect or, alternatively, the fascination of the shiny new toy that soon will be discarded?

To begin with, there are two major arguments against artificial intelligence in its current form. One should note in passing that it is not a new concept but one that has only recently begun to impinge seriously on popular consciousness. The first problem is that it uses vast, and increasing, amounts of energy, which does not help bring climate change under control, or enable the careful stewardship of the planet's resources. Perhaps technological innovation will one day bring this issue under control, but there is no sign of this at present. Secondly, it infringes copyright by misusing people's intellectual property in the training of its algorithms. This may be more of a problem for the arts than for the sciences, in which the fruits of research are supposed to benefit all of us, but at present it is hard to tell.

At the present time, perhaps the greatest potential of AI in disaster management is in its presumed ability to use its algorithms and data banks to provide synthesised information quicker than traditional methods can do so. A report by the Joint Research Centre of the European Commission (Galliano et al. 2024) suggests that in this it is close to disaster management but not quite part of it. Hence, its utility may lie in supporting decision making rather than making the decisions themselves.

A look at the research on AI tends to be more depressing than heartening. First, it is used inductively rather than deductively, which is inefficient, and often grossly so. Secondly, it should not be used as a substitute for thinking, creativity and human interaction. Certain aspects of disaster management are greatly undervalued, and thus poorly researched, but the academic DRR community. One of these is emergency planning, the process of anticipating needs caused by disaster impacts and making arrangements to satisfy them as well as possible with available resources. At the moment it is unclear whether decision making using AI generates risks of incorrect assumptions, distortions, mistaken views of situations or other errors that the technique might magnify. Hence, the safety of AI as a means of emergency management cannot be guaranteed.

Large language models can help chart the progress of public perception of disaster threats and impacts, as manifest in the mass media and social media. However, we now live in a world in which 'manufactured reality' has loomed as large as objective reality as a result of the need to deal with beliefs, opinions and expectations that differ from what science and objectivity would inform and prescribe.

The advent of social media came with a wave of optimism about their utility in reducing disaster risks and impacts (Alexander 2014). Subsequently, the dark side of the media revealed itself: conspiracy theories, subversion, personal attacks, aggression, attempts to destroy reputations, so-called "alternative facts", and so on. Could we be about to experience yet more of this with AI? The challenge with social media is to find an efficient, effective, robust and reliable way of counteracting the effect of misinformation (or disinformation, if you prefer). One of the keys to this is the issue of trust in authority--or its absence. One wonders whether displacing the human element with the computer generated one will increase or reduce trust in the output that results. Scepticism induces me to prefer the latter.

 

Legend has it that, when he was foreign minister of China, Zhou Enlai was asked by a journalist what he thought of the French Revolution and he replied "It's too early to tell." A fantasy, perhaps, but an enjoyable story just the same. It is more genuinely "too early to tell" with AI. What we need is more research on its impact, research that is detached from the process of generating applications for AI and which looks objectively at how well it is working and what problems it either encounters or produces.

Come what may, emergency management is a human activity that requires human input and human reasoning. It is unlikely that this need will ever be satisfied by artificial intelligence. The human mind is too versatile and flexible to be displaced.

More than a quarter of a century ago, Professor Henry Quarantelli, father of the sociology of disaster, published a very perceptive article on the information technology revolution, which was then in its infancy in comparison with what came later. His conclusions are still perfectly valid:-

"...close inspection of technological development reveals that technology leads a double life, one which conforms to the intentions of designers and interests of power and another which contradicts them—proceeding behind the backs of their architects to yield unintended consequences and unanticipated possibilities." (Quarantelli 1997)

We would do well to heed this observation and not embrace artificial intelligence uncritically.

References

Alexander, D.E. 2014. Social media in disaster risk reduction and crisis management. Science and Engineering Ethics 20(3): 717-733.

Galliano, D.A., A. Bitussi, I. Caravaggi, L. De Girolamo, D. Destro, A-M. Duta, L. Giustolisi, A. Lentini, M. Mastronunzio, S. Paris, C. Proietti, V. Salvitti, M. Santini and L. Spagnolo 2024. Artificial Intelligence Applied to Disasters and Crises Management: Exploring the Application of Large Language Models and Other Ai Techniques to the European Crisis Management Laboratory Analyses. European Crisis Management Laboratory, Disaster Management Unit JRC E.1, European Commission Joint Research Centre, Ispra, Italy, 46 pp.

Quarantelli, E.L. 1997. Problematical aspects of the information/communication revolution for disaster planning and research: ten non-technical issues and questions. Disaster Prevention and Management 6(2): 94-106.
 

Cities, Cultural Heritage and the Culture of Responding to Floods

 

In the autumn of 1972 I visited a town called Puerto Lumbreras, located in the Province of Murcia in southeastern Spain. It had just been affected by a major flash flood, the effect of a gota fria meteorological phenomenon of the kind that so devastated Valencia in October 2024. The Rambla de Nogalte stream, which runs through the centre of town, was a mass of churned-up mud. In the middle, a lorry was wrapped around the remains of a concrete post, a massive boulder was perched on the first floor of an apartment block that had been gutted by the floodwaters and the bloated corpse of a pig lay amid the detritus washed down by the raging waters. In the surrounding countryside, sediment and water had devastated the orchards and as far as the eye could see the landscape was full of rotting oranges, visible as thousands of bright dots amid the greys and browns of the mud and silt deposits. There had been deaths and there was much destruction. On the approach road a steel girder viaduct had been folded up and swept kilometres downstream by the violence of the water. Powerful floods struck Puerto Lumbreras again in 2012.

Europe is not well protected against flooding. Even in orderly, well-organised Germany its impact can be devastating. In 2021 a colleague who studies natural hazards wrote to me that "our institute is all but destroyed and colleagues have lost their homes". At least 184 people died and devastation was widespread. The response to these floods revealed a lack of crucial connections in civil protection between the federal government and the states.

It is perfectly clear now that climate change is causing episodes of extreme weather to be more common and more violent, so why are we not better prepared? Why are the lessons of these devastating events so easily forgotten or ignored? Each new disaster reveals the shortcomings of hazard mitigation and disaster preparedness. In Valencia in October 2024 warning failed spectacularly and many people died because they did not know how to protect themselves and did not have enough awareness of the flood risk. Shortly before that, on the Noto Peninsula of western Japan severe flooding struck the area that had been devastated by the 1st January 2024 earthquake. Damage to physical infrastructure was so great that there were severe problems with bringing aid and assistance into the area - once again.

Over the last half a century research on disasters has grown to an extent that was unimaginable in 1970. Some 140 academic journals carry scientific and social scientific papers on hazards, risks, disasters and resilience. Why has this not solved the problem? The answer is that there is a yawning gap between what we know and what we do with that knowledge. There are also areas that are not so popular with researchers, and one of these is emergency planning. In far too many places around Europe and the world, the providers of knowledge and the decision makers in public administration are engaged in a dialogue of the deaf. Yet it need not be so, if only we can help the political culture to make decisions on firmer basis of evidence and encourage the providers of that evidence to make it more accessible to non-specialists.

Despite this, the recipe for a safer world is clear and easy enough to describe. First of all, we need a change in culture towards something more inclusive and more serious. What is civil protection? The answer is that all of us are civil protection: it needs to be a collective effort to keep ourselves safe, something we think about individually, as families, as communities and as members of wider society. Culture is hard to transform, requiring much effort and many resources, but it is not impossible if there is a steadfast enough determination to achieve the changes.

Secondly, we need to make emergency planning more rigorous and standardise it on the basis of well-chosen benchmarks. Plans need to be based on detailed but flexible scenarios so that we can anticipate what will need to be done when the next emergency strikes. Foresight is difficult but by no means impossible to exercise and a wide range of methods exists to generate it, including expert advice, focus groups and trend analysis.

Thirdly, we need to invest in civil protection systems that are fully present and integrated at the national, regional and local levels. The last of these is the theatre of operations, as in essence all disasters are local affairs. This is where the resources need to be concentrated. The regional tier of government should coordinate and support local efforts and the national level should weld all of this into a fully harmonised system marked by compatibility and mutual support. Within this, information sharing and warning are paramount. There is also much to learn from the experience of managing disasters in other countries so the transfer of information needs to be international.

Fourthly, relationships of trust and participation need to be built between the authorities and the general public. In Florence, the Italian civil protection service is trialing an app that gives people information on the hazards that affect their own locality, the location of vital services and the procedures to follow in the event of an emergency, as well as offering warnings in times of crisis. Let us hope that it becomes popular.

Florence is an interesting case, as the floods that so severely damaged it in 1966 had world-wide repercussions, especially regarding the city's art and architectural treasures. Since then, the River Arno has been dredged, embankments have been raised, floodwater detention and storage areas have been created, and a major dam has been built that should regulate the flow of one of the tributaries of the Arno. In addition the city's emergency plan has been comprehensively revised.

Although it is axiomatic that prevention is better than emergency response, however much we spend on mitigating disaster, we can never afford to spend less on responding to it. The public does not tolerate parsimony in the aftermath of great destructive emergencies.

So will the next flood be less severe than the one in 1966? It is doubtful. On the positive side, heating oil will not be a problem. In the 1960s many central heating systems in the city ran on oil and the rupture of tanks mixed it with the water and mud of the flood, greatly worsening the impact. Furthermore, emergency planning was rudimentary and much of the response to the disaster was improvised, which will no longer be true.

In 1966 the city centre had many fewer cars in it than it does now. A car will float in less than half a metre of water. Scenes of chaos and devastation in major cities such as Valencia and Genoa give us a sense of the major blockages and mayhem created when large numbers of vehicles are swept away.

In Florence today accommodation is at a premium and the population of ground-floor apartments has swelled enormously compared to what it was in the 1960s. Moreover, the tourist population is at an all-time high and is now a year-round phenomenon. Even if local residents become aware of risks and how to confront them, tourists rarely are, given that they are transient visitors. Finally, many of the priceless works of art that were damaged in 1966 have been restored and put back exactly where they were before. Others, such as frescoes and large statues, cannot be moved. The underlying problem is that the potential for larger, fiercer storms may have gone some way to invalidating structural measures based upon previous levels of flooding.

According to a recent report by the European Commission, Italians feel vulnerable to extreme weather but have low levels of awareness and preparedness compared to the average for the 27 EU member states. Disaster readiness needs to be taught in schools and evening classes, discussed in public forums, promoted by all levels of government and encouraged at the level of families and communities. It needs to be a process and a dialogue. Evidence from around the world shows that it works best when it is a form of participatory democracy. It needs awareness, application, seriousness and solidarity.  If we can arrive at a situation in which disaster preparedness, readiness and response have become a fundamental public service, on a level with water supply, refuse collection and healthcare, then we will be well and truly on the way to winning the battle against natural hazards.

Resilience is an illusion

 

After much pondering of the question, I have come to the conclusion that resilience is an illusion. This is not to denigrate the work of resilience managers, as there is obviously much to be done to reduce the risk and impact of adverse events. However, the concept of resilience is, I think, suspect.

Many researchers who have adopted the concept when trying to interpret disaster risk reduction have followed the work of Crawford Stanley "Buzz" Holling, as expressed in his 1973 paper (Holling 1973). Some have even stated that the 2100-year-old concept of resilience was invented by Holling. Previously (Alexander 2013), I thought that Holling was wrong about resilience. On reflection, I now believe he was right and the concept itself was suspect.

Holling studied ecological systems at a time when there was much interest in small islands. These are attractive because they are closed systems for mass and open systems for energy. Thus they are 'nice and tidy' for study. No matter how big the shocks that they endure, ecological systems tend towards equilibrium. This is homeostasis, a key feature of resilience, according to Holling.

When, in the early 2000s, resilience began to be embraced by students of disaster, problems were soon encountered with homeostasis, colloquially known as 'bounce-back'. The late Bernard Manyena suggested we concentrate on 'bounce forward', the concept of recovery as a transition to a better reality than the pre-disaster one. However, this does not dispense with homeostasis.

Put bluntly, in disaster risk reduction, these days the goalposts are moving faster than the players.  To start with, the pattern of disaster impacts is highly irregular around the world. Secondly, and more importantly, vulnerability, risk, impact and their controlling factors are all trending. As we move towards the end of the Second Age of Enlightenment (cf. Whatmore 2023), we face a situation in which change is absolutely inherent or ingrained. Strong trends are complemented by instabilities that could tip regional and global systems into an entirely different phase. Hence, public discourse and concerns could conceivably change overnight (as they did in the USA after "Nine-Eleven" in 2001). Migration, conflict, climate extremes, proliferating technological failure and associated consequences all pose this kind of threat.

In conclusion, resilience is an idea, or an approach, that we can usefully promote under certain well-defined circumstances. However, in broad terms it is an illusion. It can only be attained by constant adaptation, which is a case of pursuing an ever receding goal. What can we do instead? I recommend going back to vulnerability and endeavouring to identify, understand and reduce it.

References

Alexander, D.E. 2013. Resilience and disaster risk reduction: an etymological journey. Natural Hazards and Earth System Sciences 13(11): 2707-2716.

Holling, C.S 1973. Resilience and stability of ecological systems. Annual Reviews of Ecological Systems 4(1): 1-23.

Manyena, B. 2016. After Sendai: is Africa bouncing back or bouncing forward from disasters? International Journal of Disaster Risk Science 7(1): 41-53.

Whatmore, R. 2023. The End of Enlightenment: Empire, Commerce, Crisis. Allen Lane, Harmondsworth, 496 pp.

A Proposed Strategy to Advocate for Improved Civil Protection in the United Kingdom

 

National elections in the United Kingdom are likely to bring a change in the political complexion of the government. This short essay puts the case for coming together to put pressure on the new leadership to improve British civil protection. The lessons of the Covid-19 pandemic, alas largely negative, show that a good civilian system designed to protect the public against major hazards and threats can save thousands of lives and billions in losses and wasted expenditure. In a world characterised by rising levels of hazard and threat, improvement of the civil protection system is a necessity. It is something that needs to be dragged onto the political agenda as a matter of extreme urgency, a matter to be treated, at last, with the seriousness it demands.

In the future the UK faces a dramatic escalation in threats and hazards, but is it sufficiently prepared to deal with them? Climate change is increasing the magnitude and frequency of floods, storms, droughts, cold snaps and heatwaves. Hence there is potential for floods that equal or exceed those generated in 2008, storms that may be more disruptive than Arwen was in 2021, heatwaves greater than that of July 2022, and so on. Increasing dependency on critical infrastructure makes the country ever more vulnerable to proliferating technological failure, whether it is caused by cyber attack, sabotage or natural forces. Unplanned mass migration is now a common phenomenon throughout the world and the combination of climate change and political and military instability can only lead to greater spontaneous movements of populations. Non-seasonal influenza retains the potential to cause a pandemic on the level of that of 1918-1920. Influenza pandemics occur with a recurrence interval of about 40 years. In the meantime the world inches towards proliferating conflict and rising authoritarianism.

Safety measures at Europe's largest nuclear plant, Zaporizhzhia in eastern Ukraine, hang by a thread. In the case of radioactive emission, it has been estimated that the top event could be 20 times greater than that of Chornobyl in 1986, which covered the entire continent with radioactivity. At a lesser scale, sabotage, targeted assassination, cyber attacks and political interference through social media are increasingly capable of causing runaway chains of adverse consequences.

It is common to find lags and inertia in civil protection. It is a field that is consistently neglected and that seldom attracts the level of investment required in order to protect society adequately against hazards, threats and disasters. Mami Mizotori, the Head of the United Nations Office for Disaster Risk Reduction (UNDRR) stated in the mid-term report of the Sendai Framework that "progress [in implementing the SFDRR] has stalled and, in some cases, reversed". One of the UK's senior and most accomplished emergency planners recently remarked that "“our societal resilience is the lowest I have ever perceived”. Looking back at the inquiries and reports on large adverse events in the past, it could fairly be argued that no disaster has been managed well in the UK in the last quarter of a century. A consistent trait has been the failure to learn and implement lessons from past events, or indeed from good practice in other European countries.

The United Kingdom does not lack talent and expertise in civil protection. In addition, more than 60 universities teach and research on topics (hazards, risks, disasters, safety, security, etc.) that are pertinent to the field. However, the problem lies in the configuration of the system. In my opinion, the following issues need to be addressed.

Centrism versus devolution. The UK civil protection system is complex, unwieldy and top-heavy. A good system starts at the local level, as this is where expertise and capability are located. No matter how large a disaster is, it is a local affair because that is where the theatre of operations invariably lies. In the Covid inquiry Mr Hugo Keith KC described the system as "a bowl of spaghetti". The linkages between the component parts are unduly complex, but they also leave quite large areas bereft of clear indications as to who is responsible for them, especially when one is dealing with the relationship between central government and the devolved administrations of Wales, Scotland and Northern Ireland (and to a lesser extent the mayoralities of the metropolitan areas).

COBRA, the national 'nerve centre' for directing major emergency actions in the UK, is a relatively small entity that does not match up to the specifications of a well-endowed national emergency operations centre.

The Civil Contingencies Act of 2004 is of debatable value, as evidenced by the fact that it was in effect abandoned during the Covid-19 pandemic emergency. The risk here is that any amendment to or substitution of the Act will merely tinker with the works, rather than effecting the fundamental changes in the system that are needed.

The long-standing tendency to circulate civil servants from one post to another risks losing expertise in the practical administration of civil protection functions. Over the last 15-20 years the profession of emergency manager has tended to lose prestige and has failed to gain an adequate career structure such that expertise can better be built up in terms of human resources.

The regional tier. The devolved administrations and metropolitan mayoralities are the nearest Britain comes to having a regional tier of public administration. These entities lack the coherence of, for example, the 20 Italian regions or 25 Netherlands safety regions. In these countries responsibility for coordinating local emergency actions is delegated to the regions. This makes sense as the problems encountered at the local level are likely to vary from one part of the country to another. Wales has developed three regional emergency response coordination centres, but the rest of the UK is not following suit. Abundant experience from the rest of Europe shows that regional coordination of civil protection resources makes sense.

The local level. For years, local authorities have been starved of funds and resources.  As a result, many of them have done relatively little to develop emergency response capability beyond existing 'blue light' capabilities. With respect to these, the Fire and Rescue Service has also suffered contraction due to lack of funds. Good emergency response is nevertheless dependent on the calibre of local responses in terms of personnel, equipment, expertise and coordination.

The voluntary sector. In some countries, notably Germany, Italy and Australia, the voluntary sector is the backbone of national, regional and local civil protection. In Britain there has been some discussion of the possibility of founding a national civil protection corps, composed of volunteers. The reaction of the voluntary sector has been to reject this idea in favour of better integrating the existing voluntary organisations into the official structure of emergency response. The age of spontaneous voluntarism is well and truly over. Organised voluntarism must be fully incorporated into the civil protection system.

Culture and inclusiveness. There is a tendency for the UK civil protection system to be dominated by a para-military ethos that is far from inclusive. Civil emergency management is often considered to be an appropriate second career for former military and police officers. They tend to be middle-aged white males. Some of them have brilliantly made the transition, bringing with them very valuable skills, but others have perpetuated the military and para-military ethos, which is decidedly not appropriate to modern civil protection. Command and control are being replaced by coordination and collaboration. If civil protection is to work, it needs to be an artefact of participatory democracy, and it must be fully inclusive. References to women and girls, people with disabilities and ethnic and cultural minorities are totally absent from the current version of the National Resilience Framework and almost completely missing from the UK Risk Register. Worldwide, there is a vast amount of evidence that this is not conducive to good practice in civil protection. An example is the excoriating report produced by three eminent Japanese women after the 2011 earthquake, tsunami and nuclear release, which highlights the deficiencies of a response system in which decision making is in the hands of elderly males.[1] The best solution to this problem is to promote inclusiveness in emergency preparedness at all levels.

Terminology. The word 'disaster' is seldom used in official UK civil protection parlance, yet how could one possibly call the Covid-19 pandemic tragedy a 'major incident'? Fire, flood, extremes of heat and cold, tempests, tornadoes, bombings, shootings, structural collapses, chronic contamination, infrastructure failure and transportation crashes are all features of recent British history and, where such events have had a profound impact, they deserve to be called disasters, and treated as such.

A strategy

Unfortunately, there is a very high likelihood that the UK will have to face major emergencies that are even more challenging than those of the post-War period so far. Within the limits of what can be achieved, governments have a duty to keep their citizens safe. In Britain there has been talk of innovation, for example in instituting a 'whole of society' approach and in making better use of academic expertise. However, in these respects there is little sign of change. Reports by the House of Lords, the National Preparedness Commission and other bodies, as well as various public inquiries, have drawn attention to the deficiencies of the civil protection system. Indeed, one might not classify it as a system at all, but rather a mere set of fragments of a system.

With a general election pending there is an opportunity to produce some strong advocacy for the development of a better civil protection system. The following are some suggestions for the improvement of the system:-

• A national civil protection agency with a large, well-equipped coordination centre at the heart of government designed to act as the hub of a capillary system that reaches all parts of the country.
  

•  A basic law (in place of the current Civil Contingencies Act) that defines the system at all levels and apportions the fundamental responsibilities at all levels. As clarity is essential, it should be written in plain English, not legalese.

• A regional tier to coordinate local civil protection activities, with strategically placed regional emergency operations centres.

• Empowerment of local authorities to have their own emergency managers and emergency operations centres, fully connected to the system in a capillary manner.

•  Standardised,"all hazards" emergency planning methodology applied at all levels.

•  Full incorporation of recognised voluntary organisations into the civil protection system.

•  Qualified emergency managers, not the police, to coordinate emergency responses.

•  Revision of the National Risk Register and UK Resilience Framework to remedy existing deficiencies (see above) and reflect the revised form of the system.

•  Closer collaboration with European civil protection forces at the national and EU levels.

•  An emphasis on understanding and reducing vulnerability rather than creating resilience, which is a less robust concept.

•  To achieve a great improvement in the role of universities in educating and training emergency managers both as part of continuing education and for new entrants to the emergency planning and management profession.

•  Ensure that the people who produce the plans are also the managers--i.e., the coordinators of emergency operations.

 A suggested strategy is as follows. In the light of a coming general election, we should:-

•  convene interested parties

•  organise a forum in order to debate the best strategy for advocacy

•  endeavour to reach and record a consensus on the best strategy for reforming British civil protection

•  produce two documents: a two-page open letter to the incoming Prime Minister and Home Office Minister and a white paper on what needs to be done, which provides more of the details.

---

[1] Domoto, A., M. Ohara and H. Hara 2011. Disaster risk reduction: a Japanese women's perspective on 3/11. Japan Women's Network for Disaster Risk Reduction, Tokyo, 17 pp. https://www.preventionweb.net/files/32983_32983insidetoprint1.pdf