Noise measurements were carried out in the Cologne area (Germany), and the resonance frequency of each site was estimated from the main peak in the spectral ratio between the horizontal and vertical component. For 32 of these sites, the thickness of the sedimentary cover was known from boreholes, and a clear correlation between resonance frequency and sediment thickness was observed. A formula that correlates cover thickness with frequency of the main peak in the horizontal-to-vertical spectral ratio was derived. In addition, a best-fitting shearwave-velocity distribution with depth, v s (z), as well as a relationship between average shear-wave velocity V ¢ s and thickness of the sedimentary cover, was calculated. By using all of the noise measurements and applying the derived relationships, we obtained a subsoil classification for the Cologne area.

[1] Seismic noise contains information on the local S-wave velocity structure, which can be obtained from the phase velocity dispersion curve by means of array measurements. The H/V ratio from single stations also contains information on the average S-wave velocity and the total thickness of the sedimentary cover. A joint inversion of the two data sets therefore might allow constraining the final model well. We propose a scheme that does not require a starting model because of usage of a genetic algorithm. Furthermore, we tested two suitable cost functions for our data set, using apriori and data driven weighting. The latter one was more appropriate in our case. In addition, we consider the influence of higher modes on the data sets and use a suitable forward modeling procedure. Using real data we show that the joint inversion indeed allows for better fitting the observed data than using the dispersion curve only.Citation: Parolai, S., M. Picozzi, S. M. Richwalski, and C. Milkereit (2005), Joint inversion of phase velocity dispersion and H/V ratio curves from seismic noise recordings using a genetic algorithm, considering higher modes, Geophys. Res. Lett., 32, L01303,

No abstract

Noise measurements were carried out at 381 sites in the Cologne area (Germany) using both short period and broad band sensors. The large number of data allowed both assesment of the influence of different sensors in the site response estimation and to compare the widely used H/V technique with the recently proposed Fourier Phase Spectral Method (FPSM). The results show that short period sensors are able to reliably retrieve site effects at frequencies well below their corner frequencies. Moreover, the H/V method should be preferred to the FPSM in determining the fundamental resonance frequency of soils. Finally, a map showing the resonance frequency distribution in the studied area was drawn using the results obtained applying the H/V technique. *

S U M M A R YTo study the applicability of the passive seismic interferometry technique to near-surface geological studies, seismic noise recordings from a small scale 2-D array of seismic stations were performed in the test site of Nauen (Germany). Rayleigh wave Green's functions were estimated for different frequencies. A tomographic inversion of the traveltimes estimated for each frequency from the Green's functions is then performed, allowing the laterally varying 3-D surface wave velocity structure below the array to be retrieved at engineering-geotechnical scales. Furthermore, a 2-D S-wave velocity cross-section is obtained by combining 1-D velocity structures derived from the inversion of the dispersion curves extracted at several points along a profile where other geophysical analyses were performed. It is shown that the crosssection from passive seismic interferometry provides a clear image of the local structural heterogeneities that are in excellent agreement with georadar and geoelectrical results. Such findings indicate that the interferometry analysis of seismic noise is potentially of great interest for deriving the shallow 3-D velocity structure in urban areas.