Zhou et al. (1996) discuss that Hale's 1984 DMO operator via a Fourier transform is computationally expensive because the DMO operator is temporally nonstationary. They use the technique of logarithmic time stretching, first introduced by Bolondi et al. (1982) to present another derivation for the frequency-wavenumber log-stretch DMO operator.
Xu et al. (2001) exploit the idea of computational efficiency of the logarithmic time stretching for the PS-DMO operator. I reformulate the work of Xu et al. (2001) using the PS-DMO smile derived in the previous section and following a procedure similar to Hale (1984) and Zhou et al. (1996). This operation is valid for a constant velocity case.
From equation , and following Hale's 1984 assumption that the DMO operator maps each sample of the NMO section (pn) from time tn to time t0 without changing its midpoint location, x [p0(t0,x,h) = pn(tn,x,h)], the 2-D PS-DMO operator in the f-k domain is
Equation () implies a change of variable from t0 to tn. From equation () we have
which I will represent as A or its Fourier equivalent:
Equation () is the foundation of PS-DMO in the f-k domain. I introduce a time log-stretch transform pair,
where tc is the minimum cutoff time introduced to avoid taking the logarithm of zero. The PS-DMO operator in the f-k log-stretch domain becomes
where is the Fourier representation of the log-stretched time axis . The value of the function for either kh =0 or is obtained as zero by taking the limit of the function for either or and applying the L'Hopital's rule.
The previous expression is equivalent to the one presented by Xu et al. (2001). Note that equation () is based on the assumption that p0(t0,x,h) = pn(tn,x,h). This does not include changes in midpoint position and/or common reflection point position. This leads to a correct kinematic operator but one with a poor amplitude distribution along steeply dipping reflectors.
Zhou et al. (1996) solve this problem for PP-DMO in the f-k log-stretch domain by reformulating the f-k log-stretch PP-DMO operator presented by Black et al. (1993). This operator is based on the assumption that the midpoint changes its location after the PP-DMO operator is applied [p0(t0,x0,h) = pn(tn,xn,h)], which leads to a more accurate distribution of amplitudes. Following the derivation used by Zhou et al. (1996) for PP-DMO, for steeply dipping events I derive a more accurate PS-DMO operator in the frequency-wavenumber log-stretch domain. This new operator differs from the previous one in the function of the filter . The new expression is
The values of the phase-like function at the points kh =0 and are obtained using L'Hopital's rule on the limit of the function , since the function is singular at those points. Figure shows a series of impulse responses for this operator.
Note that for a value of , equivalent to , the filter reduces to the known expression for P-wave data Zhou et al. (1996). We can trust the PS results since the PP impulse response, obtained with the filter in equation and , is the same as that obtained by Zhou et al. (1996). Moreover, the amplitude distribution follows Jaramillo's 1997 result. The 3-D representation for this PS-DMO operator is the starting point for the partial-prestack migration operator presented in Chapter 4.