2518 1 Istituto di Ricerca sul Rischio Sismico - CNR, Via Bassini 15 - 20133 Milano, Italy. Email: mariellacaleffi@hotmail.com 2 Istituto di Ricerca sul Rischio Sismico - CNR, Via Bassini 15 - 20133 Milano, Italy. Email meroni@irrs.mi.cnr.it 3 Istituto di Ricerca sulla Dinamica dei Sistemi Economici Milano, Italy; Email: montini@spbo.unibo.it 4 Istituto di Ricerca sulla Dinamica dei Sistemi Economici Milano, Italy; Email: zoboli@idse.mi.cnr.it 5 Istituto di Ricerca sul Rischio Sismico - CNR, Via Bassini 15 - 20133 Milano, Italy. Email zonno@irrs.mi.cnr.it VARIATION OF TOURIST FLOWS IN THE UMBRIA REGION AFTER THE SEPTEMBER 26, 1997, COLFIORITO (CENTRAL ITALY) EARTHQUAKE Mariella CALEFFI 1 , Fabrizio MERONI 2 , Anna MONTINI 3 , Roberto ZOBOLI 4 And Gaetano ZONNO 5 SUMMARY Vulnerability studies have been extensively developed in Italy starting form residential buildings, in order to preserve its rich architectural patrimony. This paper presents some results of a CNR project aiming at the definition of a model for the evaluation of socio-economic vulnerability to earthquakes. The Umbria-Marche region has been selected, as case study, due to the long sequence of earthquakes that shaken the area starting on September 26, 1997 and because is one of the most important historic areas of Central Italy. The first step was to define the criteria for quantifying the socio-economic vulnerability of the region. Socio-economic vulnerability is evaluated through a selected set of indicators including population density, housing conditions, the structure of local economy, artistic and historical sites, etc. The dataset includes data on 254 variables organised in 9 groups and 48 indicators that are elaborated at the municipality level for the two regions (338 municipalities). The economic evaluation of the impact of the 1997, Colfiorito (Central Italy) earthquake on the tourist flows in the Umbria region is discussed to illustrate one of the various possible uses of the vulnerability indicators. Before the 1997 earthquake, the annual arrivals in Assisi were about 550,000 (1996), compared to a total population of about 25,000 people living in Assisi. After the earthquake, the tourist arrivals in Assisi fell down drastically to 9,205 arrivals in October ‘97, i.e. -83% on the same month of the previous year, followed by a similar trend in the subsequent months. A specific methodology, the Event Study, sometimes used in the studies on financial market shocks, is applied to evaluate if the shock effect due to the earthquake on tourist arrivals can be considered significant at a statistical level and is able to change the pattern and timing of tourist flows in the area. Finally, the role of media, in terms of intensity and quality of the information, is analysed because this factor can change the level of risk perception by people. Evolution of earthquake intensity is therefore analysed using indicators of the frequency and type of the news on the earthquake released by a popular television channel. INTRODUCTION From 26th September 1997, Central Italy has been affected by a long sequence of earthquakes with an intensity that in the first day reached magnitude 5.9 and that caused eleven death, more than 100 people injured and wide damages to the historical and monumental patrimony [Amato et alii 1998]. The area where the earthquake occurred in 1997 has been chosen as a pilot area in a project of the National Research Council (CNR) aimed at assessing the socio-economic vulnerability relatively to the seismic risk. By observing ex ante the infrastructural, social, economic and artistic aspects may be differentiated the areas in terms of vulnerability and of reaction in case of a seismic event. Socio-economic vulnerability can be considered as a multiplication factor of natural hazard for defining the risk associated to disasters and is the object of an increasing attention in Disaster Emergency Management (DEM). When integrated in a risk assessment and management model, the analysis ofsocio-economic vulnerability can contribute to implement prevention and mitigation measures, estimate the potential economic damage, identify the most critical uncertainties about impact and reaction. The analysis of socio-economic vulnerability is however still rough in many respects given the limited experience in socio- economic vulnerability evaluation on both theoretical and practical grounds [see Wisner 1997, Cella et alii 1998, Petrini 1996, Geipel 1979]. In this study, several indicators have been worked out, at a municipal level for the 338 municipalities in Umbria and Marche, relatively to specific aspects (e.g., residential structures, population