Estimation of static corrections to improve the stacking of Ps reflection data from weathered layer converted phases

Tuesday, 15 December 2015
Poster Hall (Moscone South)
Ailiyasi Ainiwaer1, Harold Gurrola2, Richard Kelvin3, Peter Vail3 and Kenneth DeWitt Rogers1, (1)Texas Tech University, Lubbock, TX, United States, (2)Texas Tech University, Geosciences, Lubbock, TX, United States, (3)SEITEL Inc., Houston, TX, United States
In typical reflection processing, PdP phases (P wave reflected from a depth d) for several shot- receiver pairs, with the same midpoint, are stacked after applying moveout corrections (adjustments that move the PdP phases to the correctzero offset delay time). Moveout corrections are dependent on the average velocity to the given depth. The correct average velocity is considered that which produces largest stacked amplitude for the given phase. For three component reflection data, reflected P-to-S converted (PdS) phases are also stacked. Stacking the PdS phases requires estimates of the Vp/Vs ratio to a given depth; again by determining the value that results in the largest stacked amplitude. This method finds the average velocities over a large areas but fails to model small near surface velocity anomalies. These variations in velocity are replaced with shot and receiver static corrections, which are time delays at the shot and receiver to correct for slow velocities in a variable “weathered layer”.Methods to estimate the static corrections due to topography and for P-phases are well developed, but methods to compute static corrections for PdS phases can be problematic. We have adapted the receiver function (RF) method typically used to estimate Vp/Vs ratio and depth to the Moho from Moho Pds phases. As with RFs, we are using deconvolution of the radial component of the reflected seismogram by the vertical component seismogram (that contains mostly PdP phases) to isolate weathered layer converted phases and reverberations (PdPwS, PdPSwS, and PdPPwS phases, where w is the depth of the weathered layer). Similarly, we deconvolve the vertical component seismograms by the radial to isolate weathered layer phases from the PdS phase ending in P (PdSwP, PdSSwP and PdSPwP). These phases are weak in individual seismograms, but we can stack all the RFs computed for a common receiver gather (data from all shots recorded at one receiver) to enhance these phases.