Monitoring and imaging water saturation variations using ballistic body- and surface-waves from seismic ambient noise correlations

We develop a new method to monitor and locate seismic velocity changes in the subsurface using seismic noise interferometry. We use the ballistic direct P-, refracted P- and Rayleigh waves computed from 30 days records on a dense nodal array located above the Groningen gas field (the Netherlands). We infer the daily relative phase velocity changes as a function of frequency (for Rayleigh waves) and propagation distance with a cross-wavelet transform processing. Assuming a one-dimensional velocity change within the medium, the induced ballistic wave phase shift exhibits a linear trend as a function of the propagation distance. Measuring this trend for the fundamental mode and the first overtone of the Rayleigh waves for frequencies between 0.5 and 1 Hz enables us to invert for S-wave daily velocity changes in the first 1.5 km of the subsurface. Combined with P-wave velocity variations observations using a similar methodology, we interpret the changes as caused by slight water saturation variations induced by deep fluid flows in the carbonate layer below 800 m depth.