Purpose and Summary

In his case study, Mustafa Erdik showed that many buildings located near the surface fault were torn apart by the fault rupture and collapsed. In addition, sites constituted by soft and liquefiable silts and sands were also vulnerable. This has led to a call for improved assessments of local geological conditions and land-use controls. Most importantly, the codes and regulations must be better enforced. This prescription was underscored by the fact that new buildings appeared more vulnerable than older ones, which may also be explained by a change in building typologies. Large losses to the economy were due to the fact that the epicentral area is the "home" of Turkey's major heavy industry.

Fortunately, the Northridge earthquake resulted in relatively few fatalities, but the direct and indirect losses were substantial, estimated at around USD 42 billion and USD 8 billion, respectively. The greatest direct losses were suffered by the residential and commercial sectors, although public-sector losses were also substantial. William Petak and Shirin Elahi estimated in their case study that approximately half of the direct losses suffered by private and public victims were reimbursed by private insurance and federal/state assistance. After Northridge, the high level of losses and seismologists' predictions of further earthquakes caused most insurers to leave the market. The ensuing crisis in the availability of insurance led to the creation of the California Earthquake Authority (CEA), which is a state-run entity funded by the private sector set up to create a viable earthquake insurance market in the state.

In addition to revisiting the event and subsequent losses, Hiroyuki Kameda discussed the lessons learned from this tragic case. These lessons included recognition of the importance of performance-based design criteria for infrastructures and a multi-disciplinary approach for the management of complex urban disasters. This approach should not be limited investigating the physical phenomena, but should also include social factors. New research directions, which are being supported by research institutes such as the Disaster Prevention Research Institute at the Kyoto University, were discussed further by Norio Okada, who recommended the incorporation of disaster risk management into long-range urban planning and management.

In his discussion of the Chi Chi earthquake, Daigee Shaw pointed out a remarkable aspect of this earthquake - the well-organized disaster response that rescued many victims. Economic loss estimates are around USD 10 billion, of which less than 5% were covered by insurance. In addition, in comparison with other disasters there was very little government financial assistance to the victims.

Petros Dimitriou underlined the good performance of the existing Greek seismic code (when correctly applied) in avoiding losses, even, in this case, if the peak acceleration recorded during the event exceeds the design criteria. He also explained the way in which deviations from the code and construction practices resulted in the large losses encountered.

Donat Faeh provided an overview of the Swiss project on modeling and mapping the ground motion on a regional scale for selected areas in Switzerland, and particularly for the city of Basel. This project included the classification of the vulnerability of buildings to earthquake ground motion and the specification of damage scenarios. Other papers exemplified risk assessments for the Cologne area in Germany (Alexander Allmann), Romania (Friedemann Wenzel), Cyprus (Stella Kythreoti), as well as methodologies (e.g. Patricia Grossi) and tools for risk information management (e.g. Valeri Gitis).

Risk assessments should take into account the triggering effects that earthquakes may have on other types of disasters. For example, William Murphy described the landslides triggered by the Chi Chi earthquake and concluded that it may be necessary to revise current methods of slope stability analysis during earthquakes.

Ian Nunez drew attention to the bad performance of building in the 1999 Izmit earthquake (which was also confirmed by inspections, presented and analyzed by Dominik Lang and Mathias Raschke). He, and others, concluded that a quantum change needs to take place in the culture surrounding building construction. Specifically, he suggests that the quality of concrete be improved as well as the quality of building design by using structural shear walls. Other critical areas for improvements are in technical education and the regulatory environment. Finally, mandatory insurance might involve insurers more directly in mitigation.

Concettina Nunziata et al. modeled the ground motion from possible earthquakes in areas of Naples at the sites of main ancient cultural buildings. These basic data should be helpful in designing measures for retrofitting buildings in a way that avoids damage induced by the uncontrolled use of new materials, as was the case for the cupola of the Assisi Basilica. Michel Geradin demonstrated how these data were used in the Ispra Reaction Wall Laboratory for choosing the most adequate techniques for retrofitting ancient monuments in Lisbon and Palermo. With a network of laboratories, this lab is also validating seismic Eurocodes at full or large scales, which, however, make use of shaking tables and therefore operate at smaller scales.

Alistair Moat concluded in his analysis of 1999 earthquakes that industrial facilities within seismically active zones are at risk from severe disruption and that in most cases the risks are underestimated. Insurance companies and risk managers should identify proactive risk reduction measures. Loss mitigation can be achieved by improved emergency management, and to this end information technology can help incorporating lessons learnt for better training and decision making, as Satoshi Tanaka exemplified for Kobe.

Stephanie Chang showed that damage to transportation systems figures importantly in long-term losses to a national economy mainly because of the lengthy restoration times. The most important example of this was the destruction of the Kobe and regional ports. Hirokazu Tatano and other speakers showed how a reversible physical loss (the Kobe port was repaired after some months) could be accompanied by a non-reversible operational loss. Other ports both in Japan and in South Korea have maintained an activity increase after having been provisionally used after the emergency. The losses can also reverberate through the global economy as discussed by Ioannis Papadakis, who presented an analysis of the Dell Computer Corporation in the US. Dell suffered sizable losses after the 1999 Taiwan earthquake due to disruptions in the global supply of memory chips. Experience with reconstruction problems and funding were presented by Christopher Barham (for developing countries) and Laurie Johnson (for Kobe and California).

A number of papers (e.g. Kiyoshi Kobayashi et al., Paul Freeman et al., Yuri Ermoliev et al.) examined the long-term effects of catastrophes on economic growth. This seems to be a recent and challenging research issue as indicated by the relatively large number of papers devoted to it (as well as to decision theory) by young researchers.

In a survey of eight recent flood and earthquake disasters, Joanne Linnerooth-Bayer and Simon Quijano showed how the losses are transferred from the victims to insurance companies, the government and international aid givers. In their data base, which included countries in Asia, Europe and North America, they show that the sum of insurance claims, public compensation and voluntary aid ranges from 40 to 60 % of the direct losses from these events. However, the distribution is uneven, with insurance playing the major role in some countries and public assistance in others. Voluntary aid appears to play only a minor role.

Since many governments would like to reduce their role in ex post disaster assistance, Charles Scawthorn and other presenters discussed existing institutional arrangements for a national insurance system. One important suggestion is that disaster insurance be mandatory for households and businesses in high-risk areas. However, this raises the issue of how low-income persons will cover the costs. Paul Kleindorfer showed in his presentation that arrangements for bundling natural disaster insurance with fire or property insurance might be an interesting alternative to mandatory insurance. (In this direction, Aniello Amendola et al. presented a case study on possible policy options for mitigation and insurance for an Italian region). It appears that the demand for "bundled" natcat policies is far less elastic with respect to the price of these policies than "unbundled" policies.

Finally, a number of speakers (e.g. Roger Jones, Hiroyuki Tamura et al., Peter Gordon et al.) presented important outcomes and research inputs derived from theoretical papers on decision analysis for resource allocation and mitigation options.