Non-conventional seismic differencing in time-lapse II

Vanja Vracar and Robert J. Ferguson


Conventional seismic differencing relies on a number of assumptions which may not always represent reality. Error associated with the use of conventional imaging algorithms and error due to source/receiver coupling variations are assumed to be small relative to the seismic response of fluid transport in a reservoir; source/receiver positioning must also be repeated between surveys in time-lapse. The result is that conventional differencing involves simple match filtering followed by subtraction where the interpretable product is an image of the change in fluid location superimposed upon some background noise level. In reality, however, errors are often very large. We observe that though errors might be large, and with the exception of source/receiver location repeatability, coupling variation and system errors result in differences in seismic amplitude and not necessarily seismic phase. With this observation we develop four non-conventional seismic differencing algorithms: 1) Cross-correlation differencing (CCD), 2) Pseudo cross-correlation differencing (PCCD), 3) Conventional imaging condition differencing (CICD) and 4) Imaging condition differencing (ICD). The CCD and PCCD algorithm uses cross-correlation and Gaussian function to create filtering operator later multiplied by conventional difference and migrated in time and frequency domain, respectively. The CCD proves to be computationally costly, whereas PCCD improves computational cost and resolution. As both algorithms are dependent on the user to move from filtering operator creation to non-conventional differencing to migration, we develop CICD algorithm as a pilot algorithm to ICD. CICD is based on the pore-stack depth migration and conventional differencing. It performs wavefield extrapolation and conventional differencing at the imaging condition. As CICD proves to be robust, we develop ICD that is also based on pre-stack depth migration but performs nonconventional differencing (PCCD) at imaging condition. ICD algorithm fully eliminates dependence on the user, improves resolution and computational cost

View full article as PDF (4.82 Mb)