Migration Velocity Analysis Using Seismic Multiples

Abstract

Migration velocity analysis (MVA) is a powerful tool for determining the interval velocity model in regions of complex geology. It is based on the fact that the migration process is sensitive to changes in the velocity model. Only when the migration velocity is equal to the interval velocity of a layer will that event be horizontally aligned in a migrated common image gather (CIG) (a CIG is a gather containing traces from the same surface location but from different migrated shot gathers). This is equivalent to saying that the image of the subsurface should be in the same location regardless of the source position. Conventional analysis is concerned with utilising primary events and these alone for the determination of interval velocity. However, it has been found that a greater accuracy can be obtained by utilising multiple events which are normally considered as noise and removed prior to MVA. The new technique is based on the fact that any initial error in velocity between the migration and interval velocities will be increased as the distance travelled through the Earth increases. Hence in migration, multiples will be more sensitive than primaries to changes in the velocity model. Two synthetic examples are presented to demonstrate this. Both show that on a coherency plot, multiples will stack in a more localised region. This should result in increased accuracy in determining interval velocity and subsequently interval thickness. It should also reduce the number of migration iterations needed to converge on an accurate solution. This has implications for reducing the time and cost of the MVA process.