Skip to main content Skip to footer content

Polarization analysis on three-component seismic data

P. Capizzi, L. De Luca and M. Vitale

Abstract: 

The acquisition of three-component high-density seismic data in WARR surveys, aimed at the two or three-dimensional modelling of buried structures, makes it possible to apply statistical-analysis techniques to estimate the propagation parameters of waves. The availability of the three components allows the use of the information about the state of the seismic wave polarization in order to filter the signal, reducing uncorrelated noise and separating kinematically equivalent seismic phases. In order to study polarization properties of seismic waves in WARR data, a correction of the polarization angle estimate (obtained by the SVD of the covariance matrix of three-component data) was performed. Therefore, it is possible to remove the bias produced by the superposition of a non-polarized component of the wavefield to a possible linearly polarized one, and reduce the estimate variance. An interesting target in seismic profile analysis consists in the simultaneous determination of kinematic and polarization properties of the signal. This procedure is addressed to the estimation of two fundamental attributes of seismic waves: apparent slowness and polarization angle. This method provides good results only when applied to data with a high S/N ratio. Therefore, it should be improved to extend it to low S/N data.