Influence of instruments on the H/V spectral ratios of ambient vibrations

Abstract

International audienceFor an optimal analysis of the H/V curve, it appears necessary to check the instrument signal to noise ratio in the studied frequency band, to ensure that the signal from the ground noise is well above the internal noise. We assess the reliability and accuracy of various digitizers, sensors and/or digitizer-sensor couples. Although this study is of general interest for any kind of seismological study, we emphasize the influence of equipment on H/V analysis results. To display the impact of the instrumental part on the H/V behavior, some series of tests have been carried out following a step-by-step procedure: first, the digitizers have been tested in the lab (sensitivity, internal noise...), then the three components sensors, still in the lab, and finally the usual user digitizers-sensors couple in lab and outdoors. In general, the digitizer characteristics, verified during this test, correspond well to the manufacturer specifications, however, depending on the digitizer, the quality of the digitized waveform can be very good to very poor, with variation from a channel to another channel (gain, time difference etc.). It appears very clearly that digitizers need a warming up time before the recording to avoid problems in the low-frequency range. Regarding the sensors, we recommend strongly to avoid the use of “classical” accelerometers (i.e., usual force balance technology). The majority of tested seismometers (broadband and short period, even 4.5 Hz) can be used without problems from 0.4 to 25 Hz. In all cases, the instrumentation should be checked first to verify that it works well for the defined study aim, but also to define its limit of use (frequency, sensitivity...)

Evaluation of the influence of experimental conditions on H/V results from ambient noise recordings

“…The results from this study show that (1) basically any equipment that has been tested can be used except accelerometers as sensors (for a more detailed discussion see Guillier et al 2005), (2) recorder parameterization does not matter significantly as long as signal saturation is avoided, and (3) the sensor can be installed without too much precision as long as logical conditions are kept in mind (e.g., no over-tilting nor weight on the sensor) and the sensor is fully in contact with the medium on which it is set up. It is not recommended to use interface to set up the sensor, unless really necessary (e.g., to make it easier to level the sensor), in which case it is recommended to use a container filled with sand.…”

“…These experimental field tests complement the laboratory instrumental tests performed in Bergen in 2002 (Guillier et al 2002a(Guillier et al ,b, 2005, where the manufacturers specifications and equipment reliability were checked. The goal of this series of tests is to check if the various recording settings (gain, sample frequency, .…”

“…The results and conclusions of the Bergen series of tests are presented by Guillier et al (2005) and SESAME WP02 team (2002) and are not further developed in this paper. …”

Méthode H/V en géotechnique. Application à un modèle bicouche

“…Plusieurs chaînes d'acquisition ont été testées lors du projet SESAME afin de contrôler leur capacité à fournir des signaux de qualité suffisante pour que la méthode H/V puisse être appliquée. Les vélocimètres ayant une fré quence de coupure à 0,2 Hz (Guillier et al, 2008), voire même 1 Hz (Strollo et al, 2008) sont considérés comme adaptés à l'acquisition de signaux dans la plage de fré quence allant de 0,1 à 20 Hz.…”

Love wave contribution to the ambient vibration H/V amplitude peak observed with array measurements

“…This method is well suited for rapid site characterisation because it facilitates 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 2 the estimation of resonance frequencies, involves only noninvasive measurements with basically a single station, can be conducted quickly (common measurement times at a single location are on the order of ten minutes, e.g. Fäh et al, 2001;Scherbaum et al, 2003;Souriau et al, 2007) and provides reliable results for a wide variety of instrumentation types (Guillier et al, 2008). Thus, H/V measurements are extensively used in microzonation to map the variations in resonance frequencies in densely populated, earthquake-prone areas (e.g.…”