Calibration of Vp/Vs ratio in isotropic layered model using passive seismic data

A realistic velocity model is the basis for good interpretation of seismic data. Usually the main challenge is deriving the S-wave velocity model because P-wave velocities can be constrained from active seismic (e.g., check shots). Without reliable information about propagation of P- and S-waves we are not able to correctly locate hypocentres of earthquakes even with very well-picked arrival times. In the oil and gas industry, we usually have much better knowledge about P-waves than about S-waves. Information about P-wave velocities is often derived from an active seismic survey and sonic logs. Information about only P-wave velocities is sufficient when we want to use a dense network of receivers (usually 100 and more) for earthquake location (e.g., Duncan and Eisner, 2010). However, it is much more expensive than monitoring with a sparse network. Hence, there is a need to acquire information about S-wave velocity distribution in the subsurface.