Littérature scientifique sur le sujet « Wave Equation Datuming »

Wave-equation datuming (WED) techniques have demonstrated superiority when waves occur on the acquisition surface nonvertically, and traditional static corrections based on the time shift become inaccurate. Meanwhile, as for multicomponent data, those scalar techniques can hardly maintain the vector characteristics for the following multicomponent data processing flows. Considering this, we have developed an elastic-wave datuming approach to handle the static corrections for multicomponent data. Different from those existing scalar WED techniques, the multicomponent data are first decomposed into multicomponent P- and S-wave data. Then, the decomposed data are transformed into the [Formula: see text]-[Formula: see text] domain, and they are extrapolated from the acquisition surface to the datum using the one-way elastic-wave continuation. Finally, the datumed multicomponent data are reconstructed at the output datum by adding up the datumed P- and S-wave data. This elastic WED can guarantee that the same wave modes on different components are equally datumed, and the data remain multicomponent so that they are still applicable to multicomponent-joint processing techniques. Finally, several test examples involved in this paper have proved our method’s effectiveness in multicomponent data datuming application.

Wave‐equation datuming overcomes some of the problems that seismic data recorded on rugged surface topography present in routine image processing. The main problems are that (1) standard, optimized migration and processing algorithms assume data are recorded on a flat surface, and that (2) the static correction applied routinely to compensate for topography is inaccurate for waves that do not propagate vertically. Wave‐based processes such as stacking, dip‐moveout correction, normal‐moveout correction, velocity analysis, and migration after static shift can be severely affected by the nonhyperbolic character of the reflections. To alleviate these problems, I apply wave‐equation datuming early in the processing flow to upward continue the data to a flat datum, above the highest topography. This is what I refer to as “flooding the topography.” This approach does not require detailed a priori knowledge of the near‐surface velocity, and it streamlines subsequent processing because the data are regridded onto a regularly sampled datum. Wave‐equation datuming unravels the distortions caused by rugged topography, and unlike the static shift method, it does not adversely effect subsequent wave‐based processing. The image obtained after wave‐equation datuming exhibits better reflector continuity and more accurately represents the true structural image than the image obtained after static shift.

Seismic land data are commonly plagued by nonhyperbolic distortions induced by a variable near‐surface, low‐velocity layer (LVL). First‐arrival refraction analysis is conventionally employed to estimate the LVL geometry and velocities. Then vertical static time shifts are used to replace the LVL velocities with the more uniform, faster velocities that characterize the underlying refracting layer. This methodology has earned a good reputation as a geophysical data processing tool; however, velocity replacement with static shifts assumes that no ray bending occurred at the LVL base and that waves propagated vertically through the LVL (even though conventional refraction analysis methods, which are used to derive LVL models from seismic data, are less restrictive). These assumptions often are inadequate in thick, complex LVL situations, where resulting errors may considerably hamper a statics‐based velocity replacement procedure. Wave‐equation datuming may be used to perform LVL velocity replacement when statics are inadequate. This method extrapolates the seismic data from the surface to the LVL base with the LVL velocities. Then it extrapolates the data from the LVL base to an arbitrary datum, with the replacement velocity field. The marine analog of such a procedure has been well documented in the geophysical literature, where the object is to remove distortions caused by an irregular water layer. Application of wave‐equation datuming to land data is more difficult because of certain common characteristics of land data (irregular shooting, large data gaps, and crooked line geometry, combined with lower signal/noise) and because the LVL estimation procedure is considerably more difficult. We demonstrate wave‐equation velocity replacement on land data from a western U.S. overthrust belt. The LVL in this region was particularly thick and complicated and ideal for a wave‐theoretical velocity‐replacement procedure. Standard refraction analysis techniques were employed to estimate the LVL, then wave‐equation datuming was used to perform the velocity replacement. Our derived LVL model was not perfect, so some imaging errors were expected because wave‐equation datuming is highly dependent upon the LVL model. Nevertheless, our results show that wave‐equation datuming generally allowed better shallow reflector imaging than could be achieved with conventional statics processing.

Current schemes for removing near-surface effects in seismic data processing use either static corrections or wave-equation datuming (WED). In the presence of rough topography and strong lateral velocity variations in the near surface, the WED scheme is the only option available. However, the traditional procedure of WED downward continues the sources and receivers in different domains. A new wave-equation global-datuming method is based on the double-square-root operator, implementing the wavefield continuation in a single domain following the survey sinking concept. This method has fewer approximations and therefore is more robust and convenient than the traditional WED methods. This method is compared with the traditional methods using a synthetic data example.

Nonflat surface topography introduces a numerical problem for migration algorithms that are based on depth extrapolation. Since the numerically efficient migration schemes start at a flat interface, wave‐equation datuming is required (Berryhill, 1979) prior to the migration. The computationally expensive datuming procedure is often replaced by a simple time shift for the elevation to datum correction. For nonvertically traveling energy this correction is inaccurate. Subsequent migration wrongly positions the reflectors in depth.

Submarine canyons incised into the continental slope interfere with the quality of common‐midpoint (CMP) stacked seismic data obtainable from reflectors beneath the sea floor. The interference problem is caused by rough topography in conjunction with the contrast between the acoustic velocity of sea water and the velocity of the exposed rock layers. Geophysicists have long recognized that part of the solution is to replace the traveltimes of raypaths through the water by their traveltimes through an identical thickness of rock. However, use of wave‐equation datuming to effect velocity replacement yields an additional correction for the change in raypath direction that occurs in crossing from rock to sea water; the wave‐equation datuming implementation of velocity replacement is more comprehensive and complete. The wave‐equation datuming method requires an accurate sea‐floor profile as part of the input, along with values of replacement velocity; it does not require knowledge of geology or velocities at depths much greater than the sea floor. Unstacked common‐source and common‐receiver records are processed to appear as if sources and receivers were moved to the water bottom; the velocity of water is replaced; and the sources and receivers are moved back to the sea surface through the replacement medium. The computational method is well‐suited to the irregular surfaces and laterally variable velocities inherent in the problem of submarine canyons. The advantage of this method is that the corrected seismic records accurately emulate the data that would actually be observed if the acoustic velocity of water could be changed physically. The normal‐moveout (NMO) velocity for optimum CMP stacking becomes the root mean square of the layer velocities, including the velocity substituted for that of water. The spurious lateral variation of stacking velocity in the original data is eliminated. Processing of the corrected data through velocity analysis, stacking, migration, and conversion to depth is therefore more reliable.