10,323
edits
Changes
Jump to navigation
Jump to search
Line 53:
Line 53: + + Line 61:
Line 63: + + − + − − [[file:seismic-migration_fig5.png|thumb|{{figure number|5}}(a) CMP stack showing reffections from a salt dome in the Gulf of Mexico. (b) Stack after DMO processing. Steep portions of diffractions and reflections are now preserved. (c) Migration of the DMO-processed data shows the steep flank of the salt dome to be a particularly strong reflector.]] − + − − [[file:seismic-migration_fig6.png|thumb|{{figure number|6}}(a) Unmigrated and (b) migrated stack of DMO-processed data from the Gulf of Mexico.]]
no edit summary
Each of the different approaches has specific advantages such as computational efficiency, accuracy for imaging steep reflectors, and accuracy in the presence of spatial variation of velocity. Likewise, each can produce undesirable processing artifacts related to some limitation in data quality such as poor signal to noise ratio, too coarse a spatial sampling interval, and missing data (e.g., due to seismic source misfires).
Each of the different approaches has specific advantages such as computational efficiency, accuracy for imaging steep reflectors, and accuracy in the presence of spatial variation of velocity. Likewise, each can produce undesirable processing artifacts related to some limitation in data quality such as poor signal to noise ratio, too coarse a spatial sampling interval, and missing data (e.g., due to seismic source misfires).
[[file:seismic-migration_fig5.png|left|thumb|{{figure number|5}}(a) CMP stack showing reffections from a salt dome in the Gulf of Mexico. (b) Stack after DMO processing. Steep portions of diffractions and reflections are now preserved. (c) Migration of the DMO-processed data shows the steep flank of the salt dome to be a particularly strong reflector.]]
==Velocity: the key parameter==
==Velocity: the key parameter==
==Poststack versus prestack migration==
==Poststack versus prestack migration==
[[file:seismic-migration_fig6.png|thumb|{{figure number|6}}(a) Unmigrated and (b) migrated stack of DMO-processed data from the Gulf of Mexico.]]
While migration algorithms are capable of accurately imaging reflections from steep interfaces, shortcomings in CMP stacking lead to destruction of such reflections before conventional ''poststack migration'' is applied. Two alternatives to poststack migration of CMP stacked data preserve reflections from steep interfaces. Migration can be applied to the unstacked data (so-called ''prestack migration'') so that the data need not be reduced to an approximation to zero offset before migration. The improvement in imaging of steep reflectors by this approach, however, is bought at the price of a great increase in the amount of computation required for the migration.
While migration algorithms are capable of accurately imaging reflections from steep interfaces, shortcomings in CMP stacking lead to destruction of such reflections before conventional ''poststack migration'' is applied. Two alternatives to poststack migration of CMP stacked data preserve reflections from steep interfaces. Migration can be applied to the unstacked data (so-called ''prestack migration'') so that the data need not be reduced to an approximation to zero offset before migration. The improvement in imaging of steep reflectors by this approach, however, is bought at the price of a great increase in the amount of computation required for the migration.
A cost-effective and accurate alternative to full prestack migration is to apply poststack migration to data that have had the added step of ''dip moveout'' (DMO) applied after normal moveout (NMO) correction, but before the data are stacked. DMO, a form of ''partial'' prestack migration, completes the process that NMO only imperfectly accomplishes—it converts data recorded with separated sources and receivers to a close approximation to zero-offset data, preserving reflections from both gently dipping and steep reflectors. Figure 5 shows the improvement in imaging of the steep flank of a salt dome achieved by poststack migration when applied to DMO-processed data.
A cost-effective and accurate alternative to full prestack migration is to apply poststack migration to data that have had the added step of ''dip moveout'' (DMO) applied after normal moveout (NMO) correction, but before the data are stacked. DMO, a form of ''partial'' prestack migration, completes the process that NMO only imperfectly accomplishes—it converts data recorded with separated sources and receivers to a close approximation to zero-offset data, preserving reflections from both gently dipping and steep reflectors. [[:file:seismic-migration_fig5.png|Figure 5]] shows the improvement in imaging of the steep flank of a salt dome achieved by poststack migration when applied to DMO-processed data.
The additional accuracy of either DMO or prestack migration over that of conventional poststack migration demands special care in the field acquisition of seismic data. Too coarse a spatial sampling, that is, too large a geophone group interval, may preclude high resolution imaging of steep reflectors by any migration method.
The additional accuracy of either DMO or prestack migration over that of conventional poststack migration demands special care in the field acquisition of seismic data. Too coarse a spatial sampling, that is, too large a geophone group interval, may preclude high resolution imaging of steep reflectors by any migration method.
The example in Figure 6 shows imaging of reflections from steep faults. While migration of CMP-stacked data (not shown here) shows the faulting, reflections from ''the faults themselves'' are absent. Details of the fault reflections seen on the DMO-processed result can be diagnostic of sealing along the faults.
The example in [[:file:seismic-migration_fig6.png|Figure 6]] shows imaging of reflections from steep faults. While migration of CMP-stacked data (not shown here) shows the faulting, reflections from ''the faults themselves'' are absent. Details of the fault reflections seen on the DMO-processed result can be diagnostic of sealing along the faults.
The schematic diagrams shown here have been two-dimensional (2-D) representations, and the illustrations have all involved 2-D migration of 2-D seismic data. Invariably, the earth's subsurface has three-dimensional (3-D) complexity. As a result, the mispositioning of recorded reflections extends in two lateral directions, and migration must be done as a 3-D process (see “[[Three-dimensional seismic method]] for Reservoir Development”).). It suffices here to state that migration is fundamentally incomplete unless it is applied as a 3-D process to 3-D data.
The schematic diagrams shown here have been two-dimensional (2-D) representations, and the illustrations have all involved 2-D migration of 2-D seismic data. Invariably, the earth's subsurface has three-dimensional (3-D) complexity. As a result, the mispositioning of recorded reflections extends in two lateral directions, and migration must be done as a 3-D process (see “[[Three-dimensional seismic method]] for Reservoir Development”).). It suffices here to state that migration is fundamentally incomplete unless it is applied as a 3-D process to 3-D data.