Southern Mexico City Shear-Wave Velocity Estimation Using Ambient Seismic Noise

Abstract We estimate the velocity structure of the subsoil of Ciudad de México, specifically in the sub-basin of the Chalco Valley encompassing the Municipalities of Xochimilco, Coyoacán, Iztapalapa, and Tláhuac. Our model is based on the analysis of ambient seismic noise (ASN) data recorded from 383 triaxial stations, which were organized into 22 arrays, for which each was comprised of approximately 18 stations. These stations recorded ASN data simultaneously over a span of five days. To characterize the diffusive properties of the seismic noise, we calculated energy density functions and spectral density correlation coefficient matrices across all stations and the three components. Through the comprehensive analysis of the processing approaches, we successfully recovered Green’s functions in the radial, vertical, and transverse components. To distinguish the non-diffuse components of the Rayleigh-wave pulses from the radial and vertical components of Green’s functions, we employed a blind signal analysis procedure that independently evaluates each component. Subsequently, using frequency–time analysis techniques, we computed the travel times for wavetrains between various locations. To generate tomographic images, travel times were inverted under the assumption of linear trajectories and using Love- and Rayleigh-group velocity dispersion curves. In addition, we calculated horizontal-to-vertical spectral ratios (HVSRs) for each station. On the basis of results from cluster analysis, we conducted a joint inversion integrating the HVSR spectral ratios with the dispersion curves. In total, we derived 180 1D shear-wave velocity models.