Estimating the sediment thickness and the geometry of the bedrock is a key component of many hydrogeological and seismic studies and can be of major importance to understand how earthquake waves are transferred or attenuated through the upper crust. Among other aims, the horizontal- to vertical (H/V) spectral ratio ambient noise is a non-invasive passive seismic method that has been used to estimate the depth of the bedrock below sedimentary layers. The H/V method (Nakamura, 1989) uses a single, broad-band three-component seismometer to record ambient vibrations produced by local surface sources such as traffic and other human activities or from far-distance sources such as oceanic waves and wind for lower noise levels. The seismic noise recorded are signals of low-amplitude motions of the ground. Periods from noise associated to wind and human activities are predominantly below 0.1 Hz, whereas noise generated by near-shore oceanic waves and current is at higher periods (SESAME, 2005). For the measurement of the ambient noise I use a 24-bit Cityshark II acquisition system (LEAS) connected to a 3-component LE-3D/5s Lennartz seismometers. The seismometer allows recording ambient noise measurements between 0.2 Hz and 50 Hz and adjustment of the sampling rate. The spectral analysis is performed by using the open source GEOPSY software (Wathelet, 2005) and is mainly used to compute a sit's resonance frequency. The resonance frequency can be interpreted using regression equations to estimate depth to the bedrock and eventually also sediment thickness.
Currently, I'm working on H/V data gathered in Brussels to improve the bedrock depth map below the city. This in collaboration with the Geological Survey of Belgium.
In the past, I collaborated with T. Lecocq to develop a new workflow that includes the automated H/V calculation of continuous seismic data of a stationary station installed near a microzonation site. Continuous analysis benefits from nightly measurements that allow controlling the variation in HVSR results obtained by mobile stations during the day. This research is still in progress but offers great possibilities to improve bedrock depth estimations.