Multiwave Gaussian beam prestack depth migration of exploration-scale seismic data with complex near-surface effects

2016 ◽  
Vol 14 (4) ◽  
pp. 307-313 ◽  
Author(s):  
Jianguang Han ◽  
Yun Wang ◽  
Changqing Yu
Geophysics ◽  
2009 ◽  
Vol 74 (6) ◽  
pp. WCA35-WCA45 ◽  
Author(s):  
Chaoshun Hu ◽  
Paul L. Stoffa

Subsurface images based on low-fold seismic reflection data or data with geometry acquisition limitations, such as obtained from ocean-bottom seismography (OBS), are often corrupted by migration swing artifacts. Incorporating prestack instantaneous slowness information into the imaging condition can significantly reduce these migration swing artifacts and improve image quality, especially for areas with poor illumination. We combine the horizontal surface slowness information of observed seismic data with Gaussian-beam depth migration to implement a new slowness-driven Gaussian-beam prestack depth migration whereby Fresnel weighting is combined naturally with beam summation. The prestack instantaneous slowness information is extracted from the original OBS or shot gathers using local slant stacks and is combined with a local semblance analysis. During migration, we propagate the seismic energy downward, knowing its instantaneous slowness information. At each image location, the beam summation is localized in a resolution-dependent Fresnel zone; the instantaneous slowness information controls the beam summation. Synthetic and real data examples confirm that slowness-driven Gaussian-beam migration can suppress most noise from inadequate stacking and give a clearer migration result.


Geophysics ◽  
2003 ◽  
Vol 68 (6) ◽  
pp. 1782-1791 ◽  
Author(s):  
M. Graziella Kirtland Grech ◽  
Don C. Lawton ◽  
Scott Cheadle

We have developed an anisotropic prestack depth migration code that can migrate either vertical seismic profile (VSP) or surface seismic data. We use this migration code in a new method for integrated VSP and surface seismic depth imaging. Instead of splicing the VSP image into the section derived from surface seismic data, we use the same migration algorithm and a single velocity model to migrate both data sets to a common output grid. We then scale and sum the two images to yield one integrated depth‐migrated section. After testing this method on synthetic surface seismic and VSP data, we applied it to field data from a 2D surface seismic line and a multioffset VSP from the Rocky Mountain Foothills of southern Alberta, Canada. Our results show that the resulting integrated image exhibits significant improvement over that obtained from (a) the migration of either data set alone or (b) the conventional splicing approach. The integrated image uses the broader frequency bandwidth of the VSP data to provide higher vertical resolution than the migration of the surface seismic data. The integrated image also shows enhanced structural detail, since no part of the surface seismic section is eliminated, and good event continuity through the use of a single migration–velocity model, obtained by an integrated interpretation of borehole and surface seismic data. This enhanced migrated image enabled us to perform a more robust interpretation with good well ties.


2006 ◽  
Author(s):  
Carlos A. S. Ferreira ◽  
João C. R. Cruz

Geophysics ◽  
1998 ◽  
Vol 63 (1) ◽  
pp. 25-38 ◽  
Author(s):  
Xianhuai Zhu ◽  
Burke G. Angstman ◽  
David P. Sixta

Through the use of iterative turning‐ray tomography followed by wave‐equation datuming (or tomo‐datuming) and prestack depth migration, we generate accurate prestack images of seismic data in overthrust areas containing both highly variable near‐surface velocities and rough topography. In tomo‐datuming, we downward continue shot records from the topography to a horizontal datum using velocities estimated from tomography. Turning‐ray tomography often provides a more accurate near‐surface velocity model than that from refraction statics. The main advantage of tomo‐datuming over tomo‐statics (tomography plus static corrections) or refraction statics is that instead of applying a vertical time‐shift to the data, tomo‐datuming propagates the recorded wavefield to the new datum. We find that tomo‐datuming better reconstructs diffractions and reflections, subsequently providing better images after migration. In the datuming process, we use a recursive finite‐difference (FD) scheme to extrapolate wavefield without applying the imaging condition, such that lateral velocity variations can be handled properly and approximations in traveltime calculations associated with the raypath distortions near the surface for migration are avoided. We follow the downward continuation step with a conventional Kirchhoff prestack depth migration. This results in better images than those migrated from the topography using the conventional Kirchhoff method with traveltime calculation in the complicated near surface. Since FD datuming is only applied to the shallow part of the section, its cost is much less than the whole volume FD migration. This is attractive because (1) prestack depth migration usually is used iteratively to build a velocity model, so both efficiency and accuracy are important factors to be considered; and (2) tomo‐datuming can improve the signal‐to‐noise (S/N) ratio of prestack gathers, leading to more accurate migration velocity analysis and better images after depth migration. Case studies with synthetic and field data examples show that tomo‐datuming is especially helpful when strong lateral velocity variations are present below the topography.


Sign in / Sign up

Export Citation Format

Share Document