next up previous print clean
Next: Acknowledgments Up: Haines and Guitton: Electroseismics Previous: Real Electroseismic data

Conclusions and future work

We have shown that non-stationary PEF's provide an effective means for the electroseismic signal/noise separation. We demonstrate that the physics of the problem can be used to develop models for PEF estimation in this case where the signal may be so much weaker than the noise that alterations of the data cannot produce suitable models. The generality of this approach is important in electroseismic signal processing, as it enables us to attenuate the strong coherent noise and enhance previously obscured signal, without any a priori knowledge of the signal. These results are far superior to those obtained with f-k filtering, stationary PEF's Haines and Guitton (2002), or other means. The real data example shown here is encouraging, and indicates that the development of an effective, fully general, algorithm for processing electroseismic data is an attainable goal. The next major step toward a generally applicable algorithm is the development of a better model for noise PEF estimation. One likely option is horizontal geophone data, as it should closely resemble the coseismic noise. The surprisingly poor quality of the horizontal data collected as part of this project makes it unsuitable for such a use, but is likely due to the poor condition of the geophones. We plan to collect additional data using better quality geophones, and hope to find that these data can be used for noise PEF estimation.

Assuming that horizontal geophone data can effectively be used for PEF estimation, we suggest the following processing scheme for electroseismic data:

1.
Pre-processing (60 Hz removal, frequency filtering).
2.
Develop rough velocity model using coseismic and/or geophone data.
3.
Estimate signal PEF on dipole amplitude pattern, using velocity model.
4.
Estimate noise PEF on horizontal geophone data.
5.
Use these non-stationary PEF's in the signal/noise separation outlined here.

next up previous print clean
Next: Acknowledgments Up: Haines and Guitton: Electroseismics Previous: Real Electroseismic data
Stanford Exploration Project
7/8/2003