Volume 40 Number 2
  • E-ISSN: 1365-2478



Six known methods of seismic phase unwrapping (or phase restoration) are compared. All the methods tested unwrap the phase satisfactorily if the initial function is a simple theoretical wavelet. None of the methods restore the phase of a synthetic trace exactly.

An initial validity test of the phase‐unwrapping method is that the sum of the restored wavelet phase spectrum and the restored pulse‐trace phase spectrum (assuming the convolutional model for the seismic trace) must be equal to the restored phase spectrum of the synthetic trace. Results show that none of the tested methods satisfy this test. Quantitative estimation of the phase‐unwrapping accuracy by correlation analysis of the phase deconvolution results separated these methods, according to their efficiency, into three groups. The first group consists of methods using wavelet information. These methods make the wavelet phase estimation more effective than the minimum‐phase approach, if the wavelet is non‐minimum‐phase. The second group consists of methods using the phase increment Δø(Δω) between two adjacent frequencies. These methods help to decrease the time shift of the initial synthetic trace relative to the model of the medium. At the same time they degrade the trace correlation with the medium model. The third group consists of methods using an integration of the phase derivative. These methods do not lead to any improvement of the initial seismic trace.

The main problem in the phase unwrapping of a seismic trace is the random character of the pulse trace. For this reason methods based on an analysis of the value of Δø(Δω) only, or using an adaptive approach (i.e. as Δω decreases) are not effective. In addition, methods based on integration of the phase derivative are unreliable, due to errors in numerical integration and differentiation.


Article metrics loading...

Loading full text...

Full text loading...


  1. Angeleri, G.P.1983. A statistical approach to the extraction of the seismic propagating wavelet. Geophysical Prospecting31, 726–747.
    [Google Scholar]
  2. Clayton, R.W. and Wiggins, R.A.1976. Source shape estimation and deconvolution of teleseismic body waves. Geophysical Journal of the Royal Astronomical Society47, 151–177.
    [Google Scholar]
  3. Gogonenkov, G.N.1987. Investigation of detailed structure of sedimentary thicknesses by seismic prospecting. Moscow. Nedra. (in Russian).
  4. Knox, K.T. and Thompson, B.T.1974. Recovery of images from atmospherically degraded short‐exposure photographs. Astrophysical Journal193, L45–L48.
    [Google Scholar]
  5. McGowan, R. and Kuc, R.1982. A direct relation between a signal time series and its unwrapped phase. IEEE Transactions on Acoustics, Speech and Signal ProcessingASSP‐30, 719–726.
    [Google Scholar]
  6. Mitrofanov, G.M.1986. Processing of phase spectra of multichannel seismograms. Geology and Geophysics10, 90–108 [in Russian].
    [Google Scholar]
  7. Poggiagliolmi, E., Berkhout, A.J. and Boone, M.M.1982. Phase unwrapping: possibilities and limitations. Geophysical Prospecting30, 281–291.
    [Google Scholar]
  8. Shafer, R.W.1969. Echo removal by discrete generalized linear filtering. Research Laboratory of Electronics. Massachusetts Institute of Technology, Technical Report No. 466.
  9. Stoffa, P.L., Buhl, P. and Bryan, G.M.1974. The application of homomorphic deconvolution to shallow‐water marine seismology ‐ Part 1: Models. Geophysics39, 401–416.
    [Google Scholar]
  10. Timoshin, Y.V. and Shatilo, A.P.1988. Method of seismic signal phase spectrum determination. Reports of the Ukrainian Academy of Sciences B 11, 16–18 [in Russian].
  11. Tribolet, J.M.1977. A new phase unwrapping algorithm. IEEE Transactions on Acoustics, Speech and Signal ProcessingASSP‐25, 170–177.
    [Google Scholar]
  • Article Type: Research Article
This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error