The 14 th World Conference on Earthquake Engineering October 12-17, 2008, Beijing, China EARTHQUAKE LOSS ESTIMATION AND MITIGATION IN EUROPE: A REVIEW AND COMPARISON OF ALTERNATIVE APPROACHES Robin Spence 1 , Emily So 1 , Giovanna Cultrera 2 , Atilla Ansal 3 , Kyriazis Pitilakis 4 , Alfredo Campos Costa 5 , Gökçe Tönük 3 , Sotiris Argyroudis 4 , Kalliopi Kakderi 4 , Maria Luisa Sousa 5 . 1 Department of Architecture, Cambridge University, eMail: rjs2@cam.ac.uk 2 INGV, Rome, 3 KOERI, Bogazici University, Istanbul, 4 Aristotle University, Thessaloniki, Greece, 5 LNEC, Lisbon, Portugal ABSTRACT : The paper contributes to an assessment of the uncertainties involved in the use of current loss modeling methodologies when applied to the estimation of building damage and casualty generation in urban areas. The work derives from studies conducted within the EU-funded LESSLOSS project with the aim of providing a basis for urban planning authorities methods to assess alternative mitigation strategies. Research teams in Istanbul, Thessaloniki and Lisbon developed methods applicable to their own city and building stock. A benchmarking study was then carried out to compare the results of the three approaches when applied to a standardized “urban block”. The paper, presents the results of the benchmarking study, and reviews the differences between the loss estimation approaches used, There are significant differences in surface ground motion, and even greater differences in predicted damage and casualties resulting from the ground motions using the different approaches. The paper discusses possible reasons for these differences and the implications for the estimation of uncertainty in urban loss estimation. KEYWORDS: Earthquake risk, loss estimation, ground motion, building damage, casualties 1. INTRODUCTION: AIMS AND METHODS Earthquake loss estimation is of growing importance both for the planning of appropriate and cost effective earthquake mitigation measures and for insurance purposes, and an understanding of the uncertainty in such estimates is vital for informed decision-making on alternative mitigation strategies. In a recently completed collaborative research project (LESSLOSS SP10) new loss estimation studies were carried out for Istanbul, Lisbon and Thessaloniki, based on new approximately 50-year and 500-year earthquake scenarios, and using methods, software and GIS mapping tools which were developed by separate research teams for each of these cities; these scenarios were used to investigate alternative mitigation strategies. The individual city studies have been reported elsewhere (Spence et al., 2007). The approaches used for the three cities had many common elements, but there were also important differences, and in the final phase of the project a benchmarking study was conducted to compare the results obtained by different methods (KoeriLossV2, AUTHloss and LNECloss) used by the separate research. The study consisted of comparing the results obtained when the three alternative approaches were used for a standardised set of conditions. It involved defining a “standard urban block” in terms of the soil profile, the inventory of building classes, and occupancy; and applying to this urban block several bedrock ground motions of sufficient amplitude to cause significant damage. The ground motions were defined by means of bedrock acceleration time histories. Each team then estimated, using their own chosen methods: (a) the surface ground motion; (b) the level of building damage caused; (c) the number of casualties. These results were then compared. 2. THE STANDARDISED INPUT DATA Two time histories of bedrock ground motion were adopted to describe different hazard levels: 1. A time history for Gebze (GBZ) in the 1999 Kocaeli, Turkey earthquake (max PGA=0.24g) 2. A synthetic time history (S53) developed by INGV for Istanbul (max PGA=0.63g) Further details of these time histories, are given in Table 1.