Reliable Fundamental Frequencies of Soils and Buildings Down to 0.1 Hz Obtained from Ambient Vibration Recordings with a 4.5-Hz Sensor

One of the most widely used methods for site‐effect studies is based on ambient vibration recordings and the frequency peak obtained from the horizontal‐to‐vertical spectral ratio (HVSR; Nogoshi and Igarashi, 1970; Nakamura, 1989) (e.g., Fah et al. , 1997; Gueguen et al. , 1998, 2000; Alfaro et al. , 2001; Duval et al. , 2001; Navarro et al. , 2001; Cara et al. , 2003; Panou et al. , 2005; Chatelain, Guillier, and Parvez, 2008; Bensalem et al. , 2010; Hellel et al. , 2010, 2012). Ambient vibration techniques are also used for the determination of dynamic parameters of buildings (e.g., Dunand et al. , 2004; Farsi et al. , 2009; Gueguen et al. , 2010; Michel, Gueguen, Arem, et al. , 2010; Michel, Gueguen, and Lestuzzi, et al. , 2010). Possible bias associated with experimental conditions and the related precautions to be taken to obtain reliable data from such recordings have been extensively studied (e.g., Withers et al. , 1996; Mucciarelli, 1998; Mucciarelli and Monachesi, 1998; Bour et al. , 1998; Mucciarelli and Gallipoli, 2001; Mucciarelli et al. , 2003; Cara et al. , 2003; Chatelain, Guillier, Cara, et al. , 2008). However, there is still much debate about the type of seismometer that should be used. The main concern relates to the natural frequency of the seismometer, which is directly linked to the frequency range that can be surveyed. The problem arises when dealing with soils or buildings that have fundamental frequencies lower than that of the seismometer. For example, the SESAME (2004) guidelines do not recommend the use of sensors with a natural frequency above the frequency of interest. Thus, the most commonly used sensors for recording ambient vibrations have a 0.2‐ or 1‐Hz natural frequency. An ongoing controversy concerns the use …