Elsevier

Earth-Science Reviews

Volume 79, Issues 3–4, December 2006, Pages 205-227
Earth-Science Reviews

The nature of noise wavefield and its applications for site effects studies: A literature review

https://doi.org/10.1016/j.earscirev.2006.07.004Get rights and content

Abstract

The aim of this paper is to discuss the existing scientific literature in order to gather all the available information dealing with the origin and the nature of the ambient seismic noise wavefield. This issue is essential as the use of seismic noise is more and more popular for seismic hazard purposes with a growing number of processing techniques based on the assumption that the noise wavefield is predominantly consisting of fundamental mode Rayleigh waves. This survey reveals an overall agreement about the origin of seismic noise and its frequency dependence. At frequencies higher than 1 Hz, seismic noise systematically exhibits daily and weekly variations linked to human activities, whereas at lower frequencies (between 0.005 and 0.3 Hz) the variation of seismic noise is correlated to natural activities (oceanic, meteorological…). Such a surface origin clearly supports the interpretation of seismic noise wavefield consisting primarily of surface waves. However, the further, very common (though hidden) assumption according which almost all the noise energy would be carried by fundamental mode Rayleigh waves is not supported by the few available data: no “average” number can though be given concerning the actual proportion between surface and body waves, Love and Rayleigh waves (horizontal components), fundamental and higher modes (vertical components), since the few available investigations report a significant variability, which might be related with site conditions and noise source properties.

Introduction

Over the last two decades, many cities have grown considerably and demographists expect a similar trend for at least the two next decades. These urban areas are very often built on soft sediments, and a large number is unfortunately located in seismic areas, emphasizing the need for a careful and reliable assessment of site amplification phenomena. This issue has been addressed for a long time by scientists and engineers who developed many techniques to identify the main characteristics of site responses for soft deposits (i.e., resonance frequencies and amplification factor). These techniques may be grouped into three main categories. The first is based on a numerical simulation approach (see Panza et al., 2001 for a review of numerical simulation methods), and is coupling with classical geophysical and geotechnical tools (such as seismic refraction, seismic reflection, boreholes, penetrometers, etc.) in order to provide reliable estimates of the required input parameters (sediment depth, S and P wave velocities, etc.). However, many such classical geophysical tools suffer severe limitations in urbanized areas, mainly because of their cost (seemingly prohibitive especially in developing countries which face many other priorities), and their environmental impact that is less and less accepted by the community (use of explosives, drillings, etc.). The second category of technique consists in directly measuring the site response on the basis of earthquake recordings on specific stations located on carefully chosen sites. Although this technique provides an unbiased experimental estimation of the site transfer amplification factor, its use in areas of low to moderate seismicity is limited by the time required to gather a significant number of recordings with satisfactory signal to noise ratio. Finally, the latest category of methods, based on ambient noise recordings, became more and more popular over the last decades as it offers a convenient, practical and low cost tool to be used in urbanized areas. Two techniques are predominantly used to determine site response parameters: the simple horizontal to vertical Fourier amplitude spectral ratio (HVSR), and the more advanced array technique.

The ability of the HVSR technique to provide a reliable information related to site response has been repeatedly shown in the past (Nakamura, 1989, Lachet and Bard, 1994, Kudo, 1995, Bard, 1998). However, its theoretical basis is still unclear as two opposite explanations have been proposed. Nakamura, 1989, Nakamura, 2000 claims that the horizontal to vertical spectral ratio mainly reflects the S-wave resonance in soft surface layer (removing effects of surface waves), and hence that HVSR curves provide a consistent estimate of the site amplification function. This “body wave” interpretation has been contradicted in several papers highlighting the relationship between the HVSR and the ellipticity of fundamental mode Rayleigh waves (Lachet and Bard, 1994, Kudo, 1995, Bard, 1998), and thus seriously questioning the existence of any simple direct correlation between HVSR peak value and the actual site amplification factor (the HVSR peak would then be associated with the vanishing of the Rayleigh waves vertical component, instead of amplification of S-wave on horizontal components). This brief summary about the two hypothetical origins of the HVSR peak shows the close link between the composition of the seismic noise wavefield (body or surface waves) and the interpretation of the HVSR curve.

To go one step further, when one assumes that the seismic noise wavefield is mainly constituted by surface waves, then new questions arise concerning the type of surface wave (Rayleigh or Love waves) and propagation mode (fundamental or higher modes). Answering these questions is very important for the processing and interpretation of array microtremor recordings, which mainly focus on the derivation of surface waves dispersion curve. For instance, processing of horizontal components should definitively take into account the simultaneous existence of both Rayleigh and Love waves (Arai and Tokimatsu, 1998, Arai and Tokimatsu, 2000, Yamamoto, 2000, Bonnefoy-Claudet, 2004), while improper identification of surface waves type (Rayleigh or Love) or order (fundamental or harmonics) might severely bias the inversion of S-wave velocity profile (Tokimatsu et al., 1992, Tokimatsu, 1997, Beatty et al., 2002, Zhang and Chan, 2003, Bonnefoy-Claudet, 2004, Wathelet, 2005).

Despite the lack of theoretical agreement about the nature of the ambient seismic noise wavefield, the number of site-specific studies based on noise recordings has increased dramatically in recent years. Time has come thus for a clear assessment of the HVSR and array methods, in order to better identify their actual possibilities and limitations. This was one of the main goals of the SESAME European project (Site EffectS using AMbient Excitations, EESD project n° EVG1-CT-2000-00026). Within this framework, a wide set of numerical and experimental researches work has been conducted over the last 4 years to investigate the nature of the ambient noise wavefield in the frequency range of interest for site effects estimation purposes (i.e., from 0.2 to 10 Hz). The first step of this SESAME project has been to update a survey of the scientific literature dealing with seismic noise, in order to establish a state of the art about the knowledge of the nature of the ambient noise wavefield. The present paper is devoted to a presentation of the main outcomes of this comprehensive survey. After reviewing the evolution of concerns about seismic noise over the year (Section 2) we will focus on its origin (Section 3) and the composition of the noise wavefield (Section 4), with special attention to the respective proportion between a) surface and body waves, b) Rayleigh and Love waves, c) fundamental and higher modes.

Section snippets

The use of seismic noise: main historical periods

Seismic noise is observed very early on, from the very beginning of instrumental seismology in the nineteenth century. A pendulum led Bertelli (1872), in Italy, to observe that it was continuously moving for hours or days with regional weather conditions. He noticed a correlation between the long period noise and disturbed air pressure. Since this date many studies about seismic noise have been carried out. We can distinguish three predominant time periods in subsequent studies of ambient

Origin of the seismic noise

Noise is the generic term used to denote ambient vibrations of the ground caused by sources such as tide, water waves striking the coast, turbulent wind, effects of wind on trees or buildings, industrial machinery, cars and trains, or human footsteps, etc. Clearly classifying all noise sources is not an easy task. Gutenberg (1958) established a list of the different types of sources according to their frequency. Asten (1978) and Asten and Henstridge (1984) reached the same conclusions in a

Composition of the seismic noise wavefield

Ideally the goal would be to split up the noise wavefield into body waves (P, SV, SH) and surface waves (Rayleigh and Love waves), and quantify the proportion of each type of waves. The literature survey does prove that achieving that goal is not an easy task, and that many issues are still wide open. This section will thus summarize the relatively few results that have been obtained concerning 1) the relative proportion of surface waves and body waves, 2) the relative proportion of Rayleigh

Conclusions

This review outlines that, while an overall agreement is reached concerning the origin and gross characteristics of seismic noise, this is not the case for the composition of seismic noise wavefield, mainly because of the lack or scarcity of data.

It seems well established now that the spatial and temporal characteristics of seismic noise are closely related with its natural or cultural origin (low frequency microseisms or higher frequency microtremors, respectively). The amplitude variations of

Acknowledgements

This work was supported by the European Union through the research program Energy, Environment and Sustainable Development (EC-contract No: EVG1-CT-2000-00026), the EUROSEISRISK project (contract No: EVG1-CT-2001-00040) and the RDT-SismoDT for the seismic noise measurement in Grenoble. We thank P. Bodin and M. Asten for their helpful comments on this study. Their suggestions have strongly improved the present manuscript.

References (176)

  • Z. Al Yuncha et al.

    The use of ambient vibration seismic noise measurements for the estimation of surface soil effects: the Motril city case (southern Spain)

    Pure and Applied Geophysics

    (2004)
  • A. Alfaro et al.

    Measurements of microtremors in Barcelona: a tool for seismic microzonation

  • A. Alfaro et al.

    Preliminary map of soil's predominant periods in Barcelona using microtremors

    Pure and Applied Geophysics

    (2001)
  • A. Ansal et al.

    Microtremor measurements for the microzonation of Dinar

    Pure and Applied Geophysics

    (2001)
  • M.A. Ansary et al.

    Use of microtremors for the estimation of ground vibration characteristics

  • H. Arai et al.

    Evaluation of local site effects based on microtremor H/V spectra

  • H. Arai et al.

    Effects of Rayleigh and love waves on microtremor H/V spectra

  • H. Arai et al.

    S-wave velocity profiling by inversion of microtremor H/V spectrum

    Bulletin of the Seismological Society of America

    (2004)
  • H. Arai et al.

    S-wave velocity profiling by joint inversion of microtremor dispersion curve and horizontal-to-vertical (H/V) spectrum

    Bulletin of the Seismological Society of America

    (2005)
  • H. Arai et al.

    Comparison of local amplifications estimated from microtremor F–K spectrum analysis with earthquake records

  • M.W. Asten

    Geological control of the three-component spectra of Rayleigh-wave microseisms

    Bulletin of the Seismological Society of America

    (1978)
  • M.W. Asten

    Comment on “Microtremor observations of deep sediment resonance in metropolitan Memphis, Tennessee” by Paul Bodin, Kevin Smith, Steve Horton and Howard Hwang

    Engineering Geology

    (2004)
  • M.W. Asten et al.

    Arrays estimators and the use of microseisms for reconnaissance of sedimentary basins

    Geophysics

    (1984)
  • M.W. Asten et al.

    Enhanced interpretation of microtremor spectral ratios using multimode Rayleigh-wave particle-motion computations

  • T.C. Bache et al.

    High-frequency seismic noise characteristics at the four United Kingdom-type arrays

    Bulletin of the Seismological Society of America

    (1986)
  • S.K. Banerji

    Microseisms associated with the incidence of the south-west monsoon

    Nature

    (1924)
  • S.K. Banerji

    Microseisms and the Indian monsoon

    Nature

    (1925)
  • P.-Y. Bard

    Microtremor measurements: a tool for site effect estimation?

  • K. Beatty et al.

    Simulated annealing inversion of multimode Rayleigh wave dispersion curves for geological structure

    Geophysical Journal International

    (2002)
  • J. Berger et al.

    Ambient earth noise: a survey of the global seismographic network

    Journal of Geophysical Research

    (2004)
  • P. Bernard

    Etude sur l'agitation microséismique

    (1941)
  • P. Bernard

    Etude sur l'agitation microséismique et ses variations

    Annales de l'Institut de Physique du Globe de Paris

    (1941)
  • T. Bertelli

    Osservazioni sui piccolo movimenti dei pendoli in relazione ad alcuni fenomeni meteorologiche

  • B. Bettig et al.

    Analysis of dense array noise measurements using the modified spatial auto-correlation method (SPAC). Application to the Grenoble area

    Bollettino Di Geofisica Teorica ed Applicata

    (2001)
  • D. Bindi et al.

    Microtremor H/V spectral ratio in two sediment-filled valleys in western Liguria (Italy)

    Bollettino Di Geofisica Teorica ed Applicata

    (2001)
  • Bonnefoy-Claudet, S., 2004. Nature du bruit de fond sismique: Implications pour les études des effets de site. Ph.D....
  • Bonnefoy-Claudet, S., Cornou, C., Bard, P.-Y. and Cotton, F., 2004. Nature of noise wavefield. SESAME report, D13.08...
  • Bonnefoy-Claudet, S., Cornou, C., Bard, P.-Y., Cotton, F., Moczo, P., Kristek, J. and Fäh, D., in press. H/V ratio: a...
  • S.P. Bowen et al.

    Microseism and infrasound generation by cyclones

    Journal of the Acoustical Society of America

    (2003)
  • P.D. Bromirski

    Vibrations from the “perfect storm”

    Geochemistry Geophysics Geosystems

    (2001)
  • P.D. Bromirski et al.

    The near-coastal microseism spectrum: spatial and temporal wave climate relationships

    Journal of Geophysical Research

    (2002)
  • J. Capon

    High-resolution frequency–wavenumber spectrum analysis

    IEEE

    (1969)
  • J. Capon et al.

    Multidimensional maximum-likehood processing of a large-aperture seismic array

    IEEE

    (1967)
  • F. Cara et al.

    A study on seismic noise variations at Colfiorito, central Italy: implications for the use of H/V spectral ratios

    Geophysical Research Letters

    (2003)
  • F.J. Chavez-Garcia et al.

    An alternative approach to the SPAC analysis of microtremors: exploiting stationarity of noise

    Bulletin of the Seismological Society of America

    (2005)
  • B. Chouet et al.

    Shallow velocity of Stromboli volcano, Italy, derived from small-aperture array measurements of Strombolian tremor

    Bulletin of the Seismological Society of America

    (1998)
  • Cornou, C., 2002. Traitement d'antenne et imagerie sismique dans l'agglomération grenobloise (Alpes françaises):...
  • C. Cornou et al.

    Contribution of dense array analysis to identification and quantification of basin-edge induced waves. Part I: methodology

    Bulletin of the Seismological Society of America

    (2003)
  • C. Cornou et al.

    Contribution of dense array analysis to identification and quantification of basin-edge induced waves. Part II: application to Grenoble basin (french Alps)

    Bulletin of the Seismological Society of America

    (2003)
  • C. Cornou et al.

    Simulation of seismic ambient vibrations-II H/V and array techniques for real sites

  • Cited by (559)

    View all citing articles on Scopus
    View full text