next up previous [pdf]

Next: Conclusions Up: Ayeni: 4D seismic with Previous: Numerical example

Discussion

Data from permanent encoded low-energy seismic sources have the form of passive data recording (Figure 4). However, because the shot locations and timings are known and can be perfectly repeated, a strong limitation of passive data is eliminated. Furthermore, because the seismic array is permanent, data recording can be repeated perfectly. If these data sets are recorded for long enough, direct imaging of these randomly encoded data can give images of comparable quality to conventional data recording and processing (Figures 5 to 7). The poor resolution of these images in Figure 7 is due to the fact that these data were imaged with the original seismic sweeps without any source designature. These results and it's derivatives can be imporved significantly by first deconvolving the source wavelet before migration. Even then, during imaging wavefield correlations, ambient noise and cross-talk artifacts from different seismic sweeps destructively interfere, whereas the true reflections constructively interfere. Whereas, the images from the short-duration data (Figure 6) are not clean enough to generate reliable time-lapse images, those from the long-duration data (Figure 7) are clean enough to generate interpretable time-lapse images.

It is important that the time-lapse images obtained from this type of recording can provide information similar to that provided by conventional recordings. The time-lapse images computed prior to cross-equalization (Figure 8) are similar to those from conventional data recording and processing (not shown). After cross-equalization, time-lapse images from the proposed method and from conventional methods are similar (Figures 10 and 11). Even when all the data recorded using the proposed method are imaged with a strongly smoothed baseline velocity model, interpretable geomechanical information can still be obtained (Figure 12). These apparent displacement components are similar to those from conventional time-lapse seismic data (Figure 13).

With this approach, there can be significant cost-savings in both data acquisition and processing. Encoding eliminates the need for long waiting periods that would be otherwise required for low-energy seismic sources that require data stacking. In addition, encoding reduces the data storage requirements. Furthermore, independent of the number of encoded sources, direct imaging can be posed as a single phase-encoding shot-profile migration.


next up previous [pdf]

Next: Conclusions Up: Ayeni: 4D seismic with Previous: Numerical example

2010-11-26