On the Identification of Industrial Components in the Horizontal-to-Vertical Spectral Ratio (HVSR) from Microtremors

摘要

In seismic hazard studies, the Horizontal-to-Vertical Spectral Ratio (HVSR) is nowadays routinely considered as a quick way to assess possible amplification effects. However, because of several issues that can affect the data, the HVSR cannot be considered as valid per se, and a careful data evaluation is necessary. In this study, a series of HVSR curves are evaluated in order to highlight industrial components that can significantly alter the natural HVSR. First, a controlled-source experiment is carried out in order to define the characteristics of spurious industrial signals. Data analysis shows that the coherence functions and the mildly smoothed amplitude spectra plotted with linear scales can help significantly in identifying industrial components that can be otherwise difficult to highlight. Coherence functions appear particularly interesting because of their ability to reveal the presence of industrial components independently of their amplitude. Field data from three sites are then analyzed on the basis of the evidence obtained through the controlled-source experiment. For the third site, data recorded on two different days are considered. While in the first data set no significant industrial component is present, in the second and third data sets, a series of remarkable industrial signals that severely alter the natural HVSR are identified. The assessment of the coherence functions and mildly smoothed amplitude spectra is therefore suggested as a valuable support to avoid pitfalls in the interpretation of the experimental HVSR. Finally, two quick and dirty procedures aimed at reducing the effect of industrial components are also presented.