← Back to Gallery

Perturbations of Free-Field Seismic Recordings Caused by Soil-Structure Interaction, From the Effects of Buildings to the Impact of Coupling Slabs: Preliminary Results From Empirical Studies Carried Out in Greece.

A proper and reliable measurement of ground motion is essential for an accurate estimate of seismic hazard. Current ground motion databases used to develop Ground Motion Models (GMM) rarely provide information about the station housing and sensor installation conditions. Therefore, users often assume these sensors are installed in free-field condition, i.e. at the natural surface of the ground without disturbances related to nearby buildings or structures. Nevertheless, analysis of information collected from seismological networks worldwide has highlighted a large variety of sensor installation modalities, including the housing typology and seismometers coupling. Recent studies have shown that different installation conditions can have a significant impact at high frequencies on the recorded motion, in comparison with true free-field measurements. This study focuses on the impact of soil-structure interaction at different scales. First, the effect of buildings is addressed by analysing ground motion recorded by sets of sensors installed inside and very close to structures hosting several Greek key accelerometer stations, with respect to free-field measurements. Moreover, those buildings are founded on different types of soil. We observe that the effect of the soil-structure interaction is more significant on soft sites, than on stiff to hard rock ones. Second, and at a smaller scale, we investigate the impact of soil-sensor coupling. To this aim, within the framework of the “ArgoSlab” experiment led in Argostoli on Kefalonia Island, Greece, about thirty seismometers have been installed (i) on concrete slabs of different size and shape, (ii) on seismic pillars, (iii) inside manholes, (iv) outdoor or sheltered, and then compared to free-field stations' recordings. To provide a reliable interpretation, our process has been based on both earthquake recordings and ambient noise in each case. Our results show amplifications (e.g. pillars installations) or deamplifications (e.g. stations in manholes) that can be significantly greater than a factor of 3 at certain frequencies, depending on the type of installation.

Session: From Earthquake Recordings to Empirical Ground-Motion Modelling - I

Type: Oral

Room: Tubughnenq’ 4

Date: 5/2/2024

Presentation Time: 08:45 AM (local time)

Presenting Author: Pauline Rischette

Student Presenter: Yes

Additional Authors