THE RESULTS OF PROCESSING

For deconvolution I used a simple algorithm (Dwork, 1950):

$$D(f)={{S_0^*(f)}\over{S(f)S^*(f)+N(f)N^*(f)}}.$$

Here D is the deconvolution filter, S₀ is the estimated wavelet, S is the signal in real data, and N is noise. Asterisk shows complex-conjugate values.

Figures , , and show the the comparison of the data before and after deconvolution. Obviously, the P- and SV-wave have been sharpened but the result for the channel where we hoped to see the SH-wave is not encouraging. Probably our next step will be applying our approach to the full dataset where we can use stacking to enhance low-amplitude signal.