
A construct of internal multiples from surface data only: the concept of virtual seismic events
Author(s) -
Ikelle Luc T.
Publication year - 2006
Publication title -
geophysical journal international
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.302
H-Index - 168
eISSN - 1365-246X
pISSN - 0956-540X
DOI - 10.1111/j.1365-246x.2006.02857.x
Subject(s) - multiple , construct (python library) , computation , computer science , surface (topology) , context (archaeology) , scattering , range (aeronautics) , geology , algorithm , seismology , mathematics , geometry , optics , physics , engineering , arithmetic , programming language , paleontology , aerospace engineering
SUMMARY We here describe one way of constructing internal multiples from surface seismic data only. The key feature of our construct of internal multiples is the introduction of the concept of virtual seismic events. Virtual events here are events, which are not directly recorded in standard seismic data acquisition, but their existence allows us to construct internal multiples with scattering points at the sea surface; the standard construct of internal multiples does not include any scattering points at the sea surface. The mathematical and computational operations invoked in our construction of virtual events and internal multiples are similar to those encountered in the construction of free‐surface multiples based on the Kirchhoff or Born scattering theory. For instance, our construct operates on one temporal frequency at a time, just like free‐surface demultiple algorithms; other internal multiple constructs tend to require all frequencies for the computation of an internal multiple at a given frequency. It does not require any knowledge of the subsurface nor an explicit knowledge of specific interfaces that are responsible for the generation of internal multiples in seismic data. However, our construct requires that the data be divided into two, three or four windows to avoid generating primaries. This segmentation of the data also allows us to select a range of periods of internal multiples that one wishes to construct because, in the context of the attenuation of internal multiples, it is important to avoid generating short‐period internal multiples that may constructively average to form primaries at the seismic scale.