ASSESSMENT OF EARTHQUAKE-INDUCED DAMAGE LEVEL ON BUILDINGS: ANALYSIS OF TWO DIFFERENT SURVEY METHODS FOR A CASE STUDY

Abstract. Purpose of this paper is to compare different methods for the assessment of earthquake-induced damage on buildings based on survey data, referring to the case study of Castelluccio di Norcia. The seismic events that occurred in Central Italy in 2016 threatened the future of many villages located along the Apennine ridge straddling the Regions of Umbria, Marche, Abruzzo and Lazio: in Castelluccio di Norcia, a minor historical center in the Municipality of Norcia (Umbria), the earthquake occurred on August the 24^th caused some damage on localized buildings, but the strongest seismic event – the one occurred on October the 30^th and with magnitude M_w 6.5, provoked numerous collapses and widespread failure on several buildings of the village, razing to the ground almost the 60 % of the built heritage. After the two earthquakes respectively, the Fire Brigade organized a reconnaissance campaign and flew over the village via UAVs (Unmanned Aerial Vehicles). By acquiring the videos recorded in this framework, that are currently available on the net and originally not recorded for specific survey purposes, the authors produced 3D models of the village allowing to compare the configuration of Castelluccio in the phases pre- and post- the destructive event of 30^th October: in fact, since the level of damage after the earthquake of August the 24^th was very low, the model resulting from the video recorded after this earthquake could be used as a model showing the layout of the village before the main struck of October. The result of this study is the Damage Degree Evaluation (DDE) and the following definition of a map showing for each building its class of damage, according to the distinction provided by European Macro-seismic Scale EMS98. On the other hand, another damage level map was studied: the Civil Protection requested, immediately after the earthquake and for the management of disaster response activities, the activation of the COPERNICUS project, providing for the detection of most damaged buildings of the village for an early census of the non-safe areas. The map of the damage level produced within this project is of course more accurate and precise, since it was collected through different acquisition systems: UAV, close-range photogrammetry, LiDAR (Light Detection And Ranging) and SLAM (Simultaneous Location and Mapping)-based mapping. This paper proposes a comparison between the two different DDEs, in order to define whether the first method, even if based on data downloaded via the web and therefore at lower resolution, and even if acquired with a more rapid evaluation procedure not providing for ground-based surveys, can lead to the construction of damage level maps that are plausible and realistic. The question is if the first method of DDE, even if less accurate, can allow to obtain results that are satisfactory and useful in the process of management and monitoring of natural hazards, providing support for the several implied institutions, in terms of information on catastrophes and first disaster rescue management.