Thursday, 30 December 2010

An Evaluation of the Recovery Strategy after the 6 April 2009 Earthquake in L'Aquila, Central Italy


In physical terms, the earthquake of 6 April 2009 at L'Aquila was a moderate seismic event. The Irpinia earthquake of 23 November 1980 released 5.6 times more energy, and it thus affected an area ten times as large. However, the very high level of vulnerability of built structures in L'Aquila meant that the effect of the 2009 event was disproportionately serious. It was thus the worst seismic disaster to occur in Italy for 29 years. As a result, it can be regarded as a test of the national civil protection system in its current form and also an opportunity to appraise the evolution of national policies, organisation and techniques for managing the recovery from large disasters.[1] This article evaluates the response to the disaster with particular reference to the medium term and the use of transitional housing.

The Irpinia-Basilicata earthquake of 1980 was not only as a major national disaster, it was also as the dawn of a new era of civil protection. At the time, Law no.996 of 1970, governing emergency response, had not been fully enacted and hence the system was far from complete, even in the rudimentary manner of the time. However, the Friuli earthquakes of 1976 had given valuable experience in how to cope with large catastrophes, including the organisation of hospital response to mass casualty situations, relief columns, prefabricated shelter and the use of a relief commissar to direct operations. Nevertheless, in Irpinia the response was far from efficient. It took another quarter of a century to create a system composed of trained, equipped volunteers and professional workers organised into a coherent network of managers and responders, centrally directed and equipped with modern communications.

The abolition of conscription in the Italian armed forces effectively deprived the civil protection system of a well of autonomous manpower that had to be compensated for by increasing the training and equipment of the country's 3,600 civil protection volunteer forces. The new system, bolstered by the federalism inherent in the 'Bassanini' decree-law of 1998 (DL 112, articles 107 and 108) took responsibility from the prefects and gave it to the regions and provinces, leading to an uneasy compromise between central state and devolved entities. Nonetheless, it worked, not least because the L'Aquila emergency was nowhere near as large as past and potential earthquake disasters further south. Hence, in many respects the real test is yet to come, and the strategy employed at L'Aquila is likely to be a poor guide to how it will be managed.

Immediate response: hours to days

The response to the developing emergency in L'Aquila involved a doctrine akin to that used in the invasion of Baghdad, namely overwhelming force.[2] Civil protection and emergency medical structures in Abruzzo Region were extremely weak and the national response effectively sidelined them and replaced them with resources drawn from the entire nation and dominated by the regions and provinces that are most powerful in terms of civil protection, notably Trento, Roma, Emilia-Romagna and Lombardy. Although the response confirmed the primacy of the National Fire Brigades as the country's principal emergency responder—i.e. the lead agency—the failure of local systems meant that the first and most urgent need, treatment of seriously injured victims, had to be accomplished by military means using techniques developed for the evacuation of wounded soldiers from battlefields, or medevac.[3]

The early response to the disaster established and consolidated a policy that was to have profound effects on the longer-term emergency, namely the supplanting of local authority by the national hierarchy.[4] Although, in legal terms, the mayors of municipalities are the ultimate civil protection authorities, in the province of L'Aquila the Department of Civil Protection, and in places the Italian Armed Forces, commanded. This included the crucial decision to translocate the entire population of L'Aquila and of the centres of other towns, the first time in modern Italian history that a major city has been totally and mandatorily evacuated.

The division of the affected area into seven districts, managed from a national DiComaC (Centre of Direction for Command and Control), established a geographical pattern based on the cascading principle of command centres, in which the Mixed Operations Centres (COMs) in larger settlements act as points of reference for the Municipal Operations Centres (COCs) in the smaller towns.

The initial operation succeeded in that, by saturating the area with rescuers and ensuring that they were well coordinated, the main problems were overcome. These included the need to re-establish medical care, given serious damage to San Salvatore, the region's main hospital (two field hospitals were used), the need to protect the public by interdicting areas of structural collapse, and the need to provide food and shelter to the survivors.

Short term: the first six months

Earthquakes are the archetypical sudden impact disasters, in that they occur virtually without warning and can instantly leave very large numbers of people homeless.[5] In the case of L'Aquila, about 67,000 residents suffered this fate. In most instances, there is a transition from improvised shelter (cars, buses, undamaged public buildings, etc.) to transitional settlement and then to reconstructed permanent housing. The transitional shelter may involve a sequence in its own right, for example using caravans (trailers) and then prefabricated buildings. Commonly, the passage from improvised shelter to the first transitional housing is completed within a few days or weeks.[6]

The decision to accommodate survivors of the L'Aquila earthquake in tents and hotels for six months was unusual in comparison with practice elsewhere in the world. Climatically, the prolonged use of tents was feasible, as the period spanned the spring and summer, although there were problems of very high temperatures in mid-summer and intense rainstorms with localised flooding. The 171 tent camps were mostly small, self-contained entities set up in any available local space within or near to the damaged settlements. In some cases, heavy-handed security measures led residents to complain of being cooped up in lagers, but essentially the policy worked. It was nonetheless hard going to live for half a year in an eight-person tent.

By accident or design, about one third of the survivors found their own accommodation, one third were accommodated in tents and the remainder went into hotels. The problem with this last category was that most of the hotel accommodation was located close to the Adriatic Sea coast at considerable distance from L'Aquila and separated from it by the natural barrier of the Gran Sasso mountain, the highest in Apennine Italy. Social surveys revealed that many adults suffered from a sense of abandonment and disorientation after prolonged absences, but paradoxically children fared better in the hotels than in the tent camps, as social structure and a sense of community appeared to survive better in the former than in the latter.

During the early phase of the aftermath several problems emerged that were to become chronic later on. One was the degradation of the infrastructure and services of the affected area and another was loss of productive employment. It is axiomatic that recovery from disaster needs sources of work and income.[7] The primary source of these is usually the construction industry, although if recovery does not empower other sectors of the economy the result can be a 'boom' followed by a 'bust' when the reconstruction is either finished or stalls.[8]

It is estimated—unofficially, as official statistics are hard to obtain—that between 16,000 and 26,000 jobs were lost as a result of the L'Aquila earthquake. Faced with loss of accommodation and the abandonment of town centres, professional people left the area in large numbers. School enrolments declined. The main employer in L'Aquila is its university, which was left in a precarious position with all major buildings damaged to a greater or lesser extent. Fiscal incentives for employment have been severely limited and suspension of enrolment fees at the university has barely enabled it to hold its own. Moreover, the devolution of taxation so ardently promoted by successive Italian governments since the 1990s has proved advantageous to some provinces of Italy and fiscally regressive to others. L'Aquila is the worst affected example of the latter. Finally, without improvement of the local infrastructure, the area has suffered economic stagnation and decline.

Medium to long term: months to years

The decision to move the Italian G8 summit from La Maddalena, Sardinia, to L'Aquila was, at the time (July 2009), seen as a gesture of solidarity with the victims of the earthquake. However, it had remarkably little impact on their plight.[2] Upgrades to the local infrastructure were limited to an extremely small area. Funds pledged by foreign powers never materialised as the recession began to bight. However, the spotlight remained on L'Aquila and by the autumn transitional housing was available for 24,000 homeless survivors.

The flagship of the programme is the C.A.S.E. (Complessi Antisismici Sostenibili ed Ecocompatibili) project to construct 184 housing units on 19 sites to accommodate 15,500 residents (Figure 1). These two- or three-storey buildings are base isolated against earthquakes. They are constructed of wood with concrete base plates and steel frames. A more modest solution is provided by the M.A.P. (Moduli Abitativi Provvisori) prefabs erected on more than 50 sites, half of them in the municipality of L'Aquila, and housing 8,500 survivors (Figure 2).[9]

Figure 1. C.A.S.E. three-storey units at Assergi, L'Aquila


Figure 2. M.A.P. units and damaged settlement at Villa Sant'Angelo (AQ)

Whereas the price of a basic prefabricated dwelling of 40 sq. m. is about €12,000-15,000, the C.A.S.E. units cost more than 20 times as much, or an average of €280,607 per unit or €3,750 per sq. m., including public spaces.[10] These are remarkably high figures for transitional housing and represent an entirely new policy. The C.A.S.E. policy is a remarkable achievement, the rehousing of more than 15,000 people in new buildings on greenfield sites in only six months, with protection against future earthquake damage. The C.A.S.E. and M.A.P. projects represent the latest evolution and most extravagant form of the prefabricated post-disaster transitional dwelling. There are various unresolved issues with these buildings, listed as follows.

Durability and maintenance. Buildings containing a high proportion of wood require a continuous cycle of maintenance, yet the L'Aquila area does not have experience or a tradition of this. Some uncorrected signs of decay were already apparent after a few months, and the local climate is one of the most extreme in peninsular Italy. One effect of the L'Aquila earthquake has been to cause an abrupt change from stone and concrete construction to building in wood (Figure 3). It remains to be seen what the fire risk will turn out to be, given that traditionally urban areas in peninsular Italy have been of low flammability and hence have not required or obtained large firefighting resources.

Figure 3. Wooden church under construction at Fossa (AQ).

Longevity and future uses. In the plans for the C.A.S.E. and M.A.P. units there is no indication of the intended lifespan of the buildings or of any future use of the units or their sites, other than vague references to 'student housing', which would be inappropriate on such a scale. Such huge investments imply that the transitional housing will span decades rather than years. This is borne out by the fact that remnants of temporary dwellings are still to be found in Messina (1908 earthquake) and Avezzano (1915), as well as in the Belice Valley of western Sicily (1968).

Lack of services and public transport. The C.A.S.E. and M.A.P. accomodate up to 2,500 people at each site, but in almost all cases there are no basic services and there is only very limited public transport. Social cohesion and functional living are not helped by a situation in which all that has been provided is housing and some landscaping.

Isolated sites. At the time of assignment of the units, by far the most popular site was the only one that is located in L'Aquila city itself. Arischia and Assergi are two C.A.S.E. sites that are respectively 15 and 16 km from L'Aquila. Neither are near significant commercial, medical or administrative centres.

Questionable ecological values. The sourcing of materials and use of solar panels may qualify the C.A.S.E. units to be regarded as 'ecocompatible', but the dispersion and isolation of the sites, and the poor quality of public transport have induced a massive dependency on the private car, despite the lack of improvement of the local infrastructure. Moreover, some of the sites have no wastewater treatment facilities. Finally, several are built on conservation land and others are on prime farmland.

Questionable urban values. Much money and effort was expended on landscaping and urbanising the transitional housing sites, with road networks, retaining walls, footpaths, greenery and communal park areas. While this has made for a pleasant environment, although one without any particular local character, it represents a sort of forced modernisation, which entirely breaks with tradition. Given the closure and—one hopes temporary—abandonment of the historical centres, there has been a precipitous loss of the genius loci of the area. It is not clear how much of this can be recovered. At its worst the closure might also represent a form of forced migration. Such a phenomenon is not unknown after modern disasters and was encountered in the southern USA in 2005 after Hurricane Katrina.[11]

Social fragmentation. In the mechanism for assigning the transitional housing units little attention was given to the preservation of the social fabric. The result has been to enhance residents' sense of isolation, abandonment and powerlessness. Social surveys have revealed high levels of post-traumatic stress and depression, especially among women, unemployed people and the elderly. One consequence of the social fragmentation, as observed in social surveys, is an increase in xenophobia and the perception that foreigners have been given privileges in the assignment of housing.

Role of governance.[12] The social fragmentation forms part of a policy that appears to be characterised by divide and rule. Although there are emergent groups of citizens who fight for their rights and for a better future, participatory democracy has not been enhanced by the disaster. Instead, it has been replaced by government paternalism and central direction without significant devolution to local communities.

The fate of the damaged historical centres. A year and a half after the disaster the centre of L'Aquila remained off limits to ordinary citizens. Although there were problems with the removal of 4-5 million tonnes of rubble, it contained some of the most elaborate buttressing ever applied to earthquake-damaged buildings (Figure 4). The use of electro-soldering implies that the buttressing is designed to last for a very long time. Unfortunately, it cannot prevent the decay of the damaged urban fabric, only hold it in place. The result is a completely dysfunctional urban area, with services dispersed or absent, and points of reference left to decay behind the cordons.

Figure 4. Buttressing of a damaged building in L'Aquila city centre.

The context of the disaster

In the L'Aquila earthquake standard patterns of damage to buildings were reproduced an infinitum (Figure 5), suggesting that seismic vulnerability was not only widespread but absolutely endemic. Given the prevalence of earthquakes in the central Apennines, this represents a historical failure to promote and enforce adequate building codes: in fact, until recent revisions of the codes, L'Aquila was placed in the 'moderate seismicity' capacity despite having had, in 1703, an earthquake of estimated magnitude 6.7 that killed at least 6,000 residents and severely damaged most of the city's public buildings. Hence, there was no scientific justification for such laxity, which was doubtless the result of pressure by speculative builders.

Figure 5. Typical form of damage in reinforced concrete apartment building, L'Aquila.

One striking feature of the resp
onse to the L'Aquila earthquake is the lack of separation of civil protection from the business of recovery and reconstruction, a situation which is uncommon in other countries. For example, in Pakistan after the 2005 earthquake a special agency was set up to manage the recovery process.[13] In Italy, the national civil protection service seems to go through a cycle of scandals that repeats itself roughly once every ten years. The scandal of 2009-10 concerned the use of ordinances to accomplish public works, thus using emergency measures simply to bypass the stringent and cumbersome bureaucratic anti-corruption controls on the tendering process. Some €10.6 billion had been disbursed in this manner in eight years, some of it with very debatable justification.[4] Involvement of civil protection in the recovery process in L'Aquila thus tainted both sides rather than creating a workable symbiosis.

Disaster risk reduction: the report card

At the world scale, disaster risk reduction and resilience have become fashionable goals, in part thanks to the United Nations International Strategy for Disaster Reduction (UNISDR) and its Hyogo Framework for Action.[14] This has five priorities for action and it is worth considering how Italy, and L'Aquila, have responded to them.

1. Ensure that disaster risk reduction is a national and a local priority with a strong institutional basis for implementation. An OECD evaluation shows that the institutional basis is strong in Italy at the national level but fragmentary and inadequate below that.1 This is particularly true of Abruzzo Region, where resources either do not exist or have not been sufficiently devoted to the problem. Only a handful of Italy's 8,104 municipalities have comprehensive disaster reduction plans, despite the fact that no other European country is as severely afflicted by calamity. In contrast to many other European countries, neither nationally nor locally is disaster risk reduction and the creation of resilience a significant priority.

2. Identify, assess and monitor disaster risks and enhance early warning. In fundamental terms, risk is composed of two components: hazard and vulnerability. Hazard recognition and monitoring are highly developed in Italy, and that applies to the sector of the central Apennines in which L'Aquila is located. Vulnerability assessment is much less well developed and the is usually only applied to physical vulnerabilities (i.e. the probabilities of structural collapse), not socio-economic ones. At the international level, many experts regard vulnerability as the dominant component of disaster risk.[15]

3. Use knowledge, innovation and education to build a culture of safety and resilience at all levels. Although civil protection has made considerable progress in recent years, and through voluntarism this has involved a proportion of the general public, there is little sign that a culture of safety and resilience is being created. Although there are initiatives to sensitise school children against risks, civil protection and disaster risk reduction are not part of the standard curriculum. Protection structures were particularly weak in Abruzzo at the time of the earthquake and remain so.

4. Reduce the underlying risk factors. This would require considerable investment in structural and non-structural measures. Although there has been some progress in structural protection, much remains to be done. Very little progress has been made in the implementation of non-structural measures. Business continuity management (BCM), for example, is almost completely lacking in large parts of Italy. The effects of this in L'Aquila meant that employment which could have been saved was lost.

5. Strengthen disaster preparedness for effective response at all levels. Some progress is being made, notably in the more progressive regions, such as Friuli Venezia-Giulia, Emilia-Romagna, Lombardy and Sicily. However, much remains to be done.

One can conclude from this brief evaluation that the disaster risk reduction situation in Italy is not promising. It is particularly bleak in Abruzzo. Above all there is a pervasive lack of planning and a corresponding inability to set strategic priorities for resilience.

Lessons learned?

Many publications include "lessons learned" in their titles.[16] However, to be truly learned a lesson must be both practically useful and incorporated in order to create better practice. That is too seldom the case.

The resettlement policy in L'Aquila has led to the replication of a number of problems that were encountered in other countries, notably Turkey, in the 1970s and at the time represented lessons genuinely to be learned.[17] One was the importance of planning in an integrated manner for all stages of the 'disaster cycle': mitigation (risk reduction), preparedness (including prediction and warning), emergency response, recovery and reconstruction. In Italy there is a tendency not to plan and not to understand the purpose of planning, which should enable rather than restrict by coordinating the rational use of resources. Another problem concerns the functionality of transitional settlements. At L'Aquila these are lacking in socio-economic resilience and planned according to criteria that are far too restrictive, especially regarding access to employment and services. Nor is there any indication that the longer term is being planned. Hence, if return to 'normality' involves reoccupation of the damaged urban fabrics, no one can tell when this will take place. The current strategy has apparently been designed under the assumption that it will not occur for decades.

In many places the occurrence of disaster opens a 'window of opportunity' for risk reduction. For example, in Iran the Manjil earthquake of 1990 and the Bam disaster of 2003 both led to step-like improvements in the country's disaster response.[18] It is striking that the L'Aquila earthquake did not lead to similar adjustments. Other parts of the country that have a similar earthquake risk, for example the Garfagnaga and Mugello in Tuscany, appear not to have benefited from a renewed interest in disaster reduction.

The L'Aquila earthquake can only be analysed satisfactorily in political terms (Figure 6), especially with regard to short-term voting behaviour resulting from government largesse and paternalism.[4] Governance and the mechanisms of social participation have been casualties of this emphasis. The main product has been a specious urbanisation, based on greenfield sites and as dependent on the private car as much of the American suburban sprawl of the mid-20th century. As current thinking is that disaster risk reduction must be integrated with sustainable resource usage (and it must be sustainable in its own right), the policies employed in Abruzzo may be regarded as storing up problems for the future, rather than solving them.

Figure 6. The political process of resettlement in L'Aquila.

Vulnerability is at the root of disaster risk. Most attempts to characterise it have concentrated on measuring it in a sectoral manner, for example in the social, economic, physical and cultural spheres. In a model I formulated some years ago[12,19], I identified six components, which can be evaluated with respect to the case of L'Aquila:-

Total vulnerability - life is generally precarious. This is not the case, as it applies mainly to places where poverty is an absolute quantity.

Economic vulnerability - people lack adequate occupation. This was the case before the 2009 earthquake and is doubly so in the aftermath when economic stagnation has set in and jobs have been lost en masse.

Technological/technocratic vulnerability - caused by the riskiness of technology. There is little sign that technology is an important source of vulnerability in L'Aquila.

Residual vulnerability - caused by lack of modernisation. This is a fundamental source of risk in the L'Aquila area, where lack of seismic retrofitting and lack of social participation and preparation are endemic.

Delinquent vulnerability - caused by corruption, negligence, etc. There is significant evidence, although none with scientific validity, that the L'Aquila earthquake was an opportunity for corruption and speculation. Although this aspect is by its very nature hard to evaluate, it seems that, rather than creating the conditions for improvement of safety, disaster in Italy opens a Pandora's box of negative outcomes by creating conditions that organised crime and corrupt people can exploit. At the same time, disaster tends to weaken structures designed to control speculation and crime, especially as the imperatives to get things done lead to the abbreviation of the relevant procedures.

Newly generated vulnerability - caused by changes in circumstances. As noted above, although the transitional settlement policy may have abated seismic risk, it may end up increasing social risk by creating fragmentation and isolation in the social fabric.

A change in attitudes and culture is urgently needed in order to reverse the trends described herein. The concept of resilience has succeeded in being accepted in places where it seemed far more foreign than it does in L'Aquila. A more locally based, socially inclusive policy of recovery would help it to be accepted there as well.


In the preparation of this article I gratefully acknowledge the help of the MICRODIS-L'Aquila research team: Caterina Antinori, Francesco Barbano, Anna Carbonelli, Vincenza Cofini, Christian Iasio, Michele Magni, Fausto Marincioni and Roberto Miniati.


[1] OECD 2010. Italy 2010: Review of the Italian National Civil Protection System (OECD Reviews of Risk Management Policies. Organisation for Economic Co-operation and Development, Paris, 173 pp.

[2] Alexander, D.E. 2010. The L'Aquila earthquake of 6 April 2009 and Italian Government policy on disaster response. Journal of Natural Resources Policy Research 2(4): 325-342.

[3] Gerhardt, R.T., J.S. McGhee, C. Cloonan, J.A. Pfaff and R.A. De Lorenzo 2001. U.S. Army MEDEVAC in the new millennium: a medical perspective. Aviation, Space and Environmental Medicine 2(7): 659-664.

[4] Alexander, D.E. 2011. Civil protection amid disasters and scandals. In E. Pasotti and E. Gualmini (eds) Politica in Italia 2011 (Bologna), Italian Politics 2011 (San Francisco).

[5] Becker, N. 2009. Raising preparedness by risk analysis of post-disaster homelessness and improvement of emergency shelters. Disaster Prevention and Management 18(1): 49-54.

[6] Alexander, D.E., 1984. Housing crisis after natural disaster: the aftermath of the November 1980 southern Italian earthquake. Geoforum 15(4): 489-516.

[7] El-Anwar, O., K. El-Rayes and A. Elnashai 2010. Minimization of socioeconomic disruption for displaced populations following disasters. Disasters 34(3): 865-883.

[8] Haas, J.E., R.W. Kates and M.J. Bowden (eds) 1977. Reconstruction Following Disaster. M.I.T. Press, Cambridge, Massachusetts, 331 pp.

[9] Stucchi, M., C. Meletti, G. Manfredi, M. Dolce (eds) 2009. L'Aquila, April 6th 2009, 3:32am. Progettazione Sismica 03: 1-256.

[10] Calvi, G.M. and V. Spaziante 2009. Reconstruction between temporary and definitive: the CASE project. Progettazione Sismica 03: 221-250

[11] Button, G.V.2006. Voices from the Astrodome and beyond: counternarrative accounts of disaster. Learning from Catastrophe: Quick Response Research in the Wake of Hurricane Katrina. Special Publication no. 40, Natural Hazards Centre, University of Colorado, Boulder, Colorado: 429-442.

[12] Özerdem, A. and T. Jacoby 2006. Disaster Management and Civil Society: Earthquake Relief in Japan, Turkey and India. International Library of Postwar Reconstruction and Development no. 1.I.B. Tauris, London 142 pp.

[13] Cosgrave, J. and M. Herson 2008. Perceptions of crisis and response: a synthesis of evaluations of the response to the 2005 Pakistan earthquake. ALNAP Seventh Review of Humanitarian Action. Active Learning Network for Accountability and Performance in Humanitarian Action, Overseas Development Institute, London, Chapter 4, p. 208.

[14] UNISDR 2005. Hyogo Framework for Action 2005-2015: Building the Resilience of Nations and Communities. United Nations International Strategy for Disaster Reduction, Geneva, 22 pp.

[15] Birkmann, J. (ed.) 2006. Measuring Vulnerability to Natural Hazards: Towards Disaster Resilient Societies. United Nations University Press, Tokyo, 524 pp.

[16] e.g. Fallahi, A. 2007. Lessons learned from the housing reconstruction following the Bam earthquake in Iran. Australian Journal of Emergency Management 22(1): 26-35.

[17] Davis, I. 1978. Shelter After Disaster. Oxford Polytechnic Press, Oxford, 127 pp.

[18] Amini Hosseini, K., M. Kazem Jafari, M. Hosseini, B. Mansouri and S. Hosseinioon 2009. Development of urban planning guidelines for improving emergency response capacities in seismic areas of Iran. Disasters 33(4): 645-664.

[19] Alexander, D.E. 1997. The study of natural disasters, 1977-97: some reflections on a changing field of knowledge. Disasters 21(4): 284-305.