Panou, Areti

In the framework of the SESAME project one of the taskswas the compilation of all available ambient noise measurements within urban environments affected by historical or/and recent strong earthquakes in Europe. The aim of such a task was to give an answer to the question; “How does horizontal-to-vertical ambient noise spectral ratio compare with damage in modern cities?”. For this purpose five European urban areas, namely, Angra do Heroismo (Portugal), Fabriano and Palermo (Italy), Thessaloniki and Kalamata (Greece) were selected for which spatial damage information was available either in terms of modified Mercalli intensity or in EMS98 damage grades. The geological setting of the examined sites as well as the causative earthquakes are satisfactorily known. Ambient noise recordings compiled for all examined sites have been homogeneously processed by a technique developed and agreed upon SESAME project. Using a standard multivariate statistical analysis, namely, factor analysis and canonical correlation, the horizontal-to-vertical ambient noise spectral ratio (HVNSR) is correlated with damage pattern observed within examined urban areas. Results show that, in some cases (Thessaloniki, Palermo), the HVNSR seems to be able to differentiate between areas previously shown to be associated with higher damage. In other cases however (Angra do Heroismo, Fabriano, Kalamata), the correlation is not statistically significant indicating thus the complex character of the parameters involved, implying that currently there is no a straightforward way that a value of HVNSR can correctly predict the extent to which a given region will be associated with increased damage.

In order to empirically evaluate the horizontal-to-vertical (H/V) spectral ratio technique, ambient noise measurements performed in about two hundred sites mainly in Europe where weak or/and strong motion data was recorded. Standard Information Sheets (SIS) and earthquake information data were included in the SESAME [Site EffectS assessment using AMbient Excitations] project database, specially designed to facilitate data selection. All noise recordings were processed with JSESAME software to calculate (H/V) spectral ratio, whereas weak and strong motion earthquake recordings were processed with a similarly standardized procedure. For the latter, (H/V) receiver function for all sites were calculated. Experimental site transfer functions obtained from earthquake recordings were compared with the (H/V) spectral ratios from noise recordings in terms of fundamental frequency, amplification bandwidth and amplification level. Similarities and differences between (H/V) spectral ratio of noise and earthquake recordings are presented and discussed. In addition, a dense grid of noise measurements were performed within urban environment of cities affected by strong earthquake (Greece: Thessaloniki, Kalamata, Italy: Palermo). It seems that the (H/V) spectral ratio may satisfactorily indicate areas favorable to the occurrence of higher damage in urban environment. However, quantitative correlation between (H/V) spectral ratio properties and damage distribution (macroseismic intensity, damage grades) in some cases, is difficult to be established given the complexity of parameters involved.

In this Report comparisons of seismic damage distribution with the (H/V) spectral ratio results is attempted, under the framework of the SESAME Project (Site Effects Assessment Using Ambient Excitations, EC-RGD, Project No. EVG1-CT-2000-00026 SESAME), Task A (H/V technique), Work Package 04 (WP04– H/V Technique: Empirical Evaluation).

In this Report the homogeneous data set of earthquake and noise recordings at many sites that is generated under the framework of the SESAME Project (Site Effects Assessment Using Ambient Excitations, EC- RGD, Project No. EVG1-CT-2000-00026 SESAME), Task A (H/V technique), Work Package 04 (WP04– H/V Technique: Empirical Evaluation), is presented. Three main topics are included in the Report: (a) The Standard Information Sheet (SIS) structure, (b) The SESAME Ascii Format (SAF) files, and (c) The SESAME SIS-database. The SIS compiled in this sub-task and SAF format adopted for all data, constitute a homogeneous data set to be processed for further scientific results in the framework of the WP04. In addition, the SIS-database is going to facilitate data selection and management.