Thanks for your time and help!

I am working with Apollo project passive seismic experiment (PSE) data, and I have a large set of seismic records (on digital counts) and the corresponding file of poles and zeros.

I need to deconvolve these registers to obtain calibrated seismograms on SI units.

I have read that this can be accomplished with a water-level deconvolution of 10% but I am not sure of how can make this.

I ask for some orientation and if possible bibliographical references

I am working in MatLab

  • $\begingroup$ Hi Oscar and welcome to DSP.SE! I may help give further details if you link what you have read, assuming it is available online. From what you wrote it sounds like there is a correlation of seismic data to water levels and you are trying to normalize that out perhaps-- not clear to me but perhaps others with experience with seismic data analysis will recognize what you are trying to do and why. $\endgroup$ Jun 20, 2020 at 12:47
  • $\begingroup$ Hello @DanBoschen, Thanks for your interest. The Water-Level refers to a type of regularized convolution to solve stability problems introduced for noise on the signal after deconvolution. The document I have is The 1979 Ph.D. Thesis of Peter Horvath link, the chapter 2 is dedicated to the deconvolution with Water-Level Method $\endgroup$ Jun 20, 2020 at 20:03
  • $\begingroup$ Interesting- is it referring to actual water level data or is that an analogy? $\endgroup$ Jun 20, 2020 at 20:12
  • $\begingroup$ @DanBoschen There is no water involved. Please takes a look at link. There the authors expose Water-Level deconvolution but with an algorithm different from Horvath. My main difficulty is finding the correct frequency bins! Thanks for your help. $\endgroup$ Jun 20, 2020 at 20:19
  • $\begingroup$ I had no idea this was called water level deconvolution, but it appears to be what Wertheim (at the old Bell Labs) was doing in 1975: G.K. Wertheim, Deconvolution and Smoothing: Applications in ESCA, J. Electron Spectroscopy and Related Phenomena, 6 (1975) 239-251. I do not have an online link to this paper, just an old scanned copy. Wertheim seems to be using a 10% "water level". Best of success with your project/research! $\endgroup$
    – Ed V
    Jun 20, 2020 at 21:41


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.