Green’s function representation for seismic interferometry by deconvolution

Despite the strength of seismic interferometry to retrieve new seismic responses by crosscorrelating observations at different receiver locations, the method relies on a number of assumptions which are not always fulfilled in practice. Some practical circumstances that may hamper interferometry by crosscorrelation are: one-sided illumination, irregular source distribution, varying source spectra, multiples in the illuminating wave field, intrinsic losses, etc. To account for some or more of these effects, several authors have proposed interferometry by deconvolution, using a variety of approaches (Schuster and Zhou, 2006; Snieder et al., 2006; Mehta et al., 2007; Vasconcelos and Snieder, 2008; Wapenaar et al., 2008a, 2008b; van der Neut and Bakulin, 2009; Berkhout and Verschuur, 2009; van Groenestijn and Verschuur, 2010). Here we discuss a Green’s function representation of the convolution type as a unifying framework for seismic interferometry by multi-dimensional deconvolution (MDD).