Processing steps with alternative implementations

Seismic processing is characterized by a sequence of processing steps, that is more or less fixed or standardized; however, for each of these steps there are number of alternative ways for their implementation. The choice of the most appropriate way depends on geophysical considerations; therefore the processing geophysicist should have a thorough understanding of the geophysical principles that underly these processes.
An overview of the processing steps with, for each step a number of alternative implementations, is given below:

  1. Editing

    • spike removal
    • noise burst removal

  2. Statics

    • field statics, elevation corrections
    • first arrival picking and modeling of the near surface
    • reflection signal picking and residual (short wavelength) statics
    • stackpower optimization (e.g. simulated annealing)
    • wave equation redatuming

  3. Velocity analysis

    • CVG, CVS, semblance, the eigenvalue method, coherency inversion
    • moveout correction and stacking
    • normal incidence times and normal incidence wavefront curvature
    • velocity model building
    • relationship of stacking velocity with dmo and time migration velocity

  4. Signal-to-noise enhancement

    • straight stack, weighted stack, diversity stack
    • velocity stack and parabolic Radon transform
    • median filter based methods
    • Wiener filter, matched filter, output energy filter
    • Karhunen Loeve transform
    • f,x-domain prediction filtering

  5. Deconvolution

    • spectrum estimation
    • wavelet estimation and designature
    • dereverberation and deghosting
    • deabsorption
    • estimation of the earth response
    • least-squares filters: spiking deconvolution and gapped deconvolution
    • vibroseis processing
    • data adaptive deconvolution
    • deterministic deconvolution
    • deconvolution methods:
      • least-squares Wiener filters
      • maximum-, minimum entropy or parsimonious deconvolution
      • homomorphic deconvolution
      • ARMA model deconvolution

  6. Multiple elimination

    • predictive deconvolution in (t,x)-domain or linear (tau,p)-domain
    • differential moveout methods:
      • (weighted) stack
      • k,f-domain filtering
      • parabolic Radon transform
    • SRME = free surface related multiple elimination
    • dereverberation with wave equation redatuming of sources and receivers

  7. DMO

    • dmo and velocity analysis
    • dmo and pre-stack imaging (psi)
    • 2D and 3D Kirchhoff implementation

  8. Migration

    • time migration - velocity analysis
    • depth migration - velocity model building
    • migration algorithms:
      • finite difference
      • Fourier domain; e.g. phase-shift, phase-shift plus interpolation
      • Kirchhoff or summation or (weighted)diffraction stack migration
      • reverse time migration (RTM)



HOME