Full text loading...
-
Site effects and H/V method: comparison of different approaches
- Publisher: European Association of Geoscientists & Engineers
- Source: Conference Proceedings, 4th EEGS Meeting, Sep 1998, cp-43-00130
- ISBN: 978-94-6282-127-9
Abstract
It has long been noticed that the effects of geology on ground shaking is an important aspect of earthquake engineering. In general site response estimate from instrumental recordings is accomplished by removing the source and the path effects from recorded seismograms using information relevant to bedrock ground motion. When ground motion recordings at the base of the soil layers (e.g. down-hole array) are not available, simplified tecniques that only requieres recordings at the top of the layers could be used. Basically these methods may be divided in to two main categories (Bard, 1994): the 'Reference Site' and the 'Non Reference Site' techniques. In this study we adopted a non reference site technique which requires to take the speetral ratio between the horizonthal and the vertical components (WV) of the signal recorded at surface. Many authors agree that this method, when applied to earthquake data (Receiver function method (Langston, 1977)) provide a reliable estimation of the site response. Nevertheless it seems to underestimate the absolute level of amplification with respect to those inferred by the reference site or down hole recording methods. Differently, other authors pointed out that WV ratio applied to microtremors (Nakamura method, (Nakamura, 1989)) allows a reliable evaluation of the fundamental resonance frequencies of soft deposit, but it fails in providing higher harmonies, Moreover the obtained amplitude value is generally lower with repsect to the amplification measured with other techniques. In this study we utilize short period recordings as weIl as microtremors to examine frequency dependent site effects. Data were recorded by a temporary network layed out during the earthquake sequence occurred in Central Italy (Umbria Marche) in september 1997. The emphasis of this preliminary analysis is both on a comparison of the results obtained by Nakamura and Receiver function methods and on assessing the influence of different deconvolutive approaches to estimate the speetral ratio.