Conclusions

The final results of the project are inconclusive but suggest that high resolution acquisition of seismic data at the frequencies needed to accurately identify relatively thin Morrow sandstones versus shales near the top of channels incised into Mississippian limestones may be difficult to impossible. However, due to the environmental problems faced during this survey, this does not mean that it will be impossible for another survey during a dry period to identify sands versus shales using changes in seismic attributes. Such a survey might test various sources such as deeper wells with a larger Pentalite dynamite source, or a survey which uses a different source such as a high-frequency vibrator prior to acquisition. This survey though, was only able to identify channels, and general channel thicknesses based on amplitude and isochron anomalies. The only possible seismic signatures due to sand versus no sand are found on line 3 (Fig. 27) where a channel anomaly that contains sand has a lower amplitude than one that is slightly thicker and contains no sand, and on Line 1 (Fig. 23) where a channel anomaly associated with a thicker channel with productive sandstone has about the same amplitude as an area which has no sand, but it thinner and was expected to have a lower amplitude anomaly. These characteristics, however do not seem to be consistent throughout the data set, and should not be trusted to predict sandstone thickness. In addition, at the vertical resolution possible with this data set, any change in seismic character due to changes in sandstone thickness, would probably be overwhelmed by the overall changes in thickness of the Morrow to Mississippian clastic section.

If future experiments are done to determine if it is possible to identify the productive Morrow sandstones in this area, they should only be done on a small scale, during dry weather, with a different source, source depth, and/or charge size, with a system that has a large dynamic range, and with the expectation that a lot of 60 and 120 Hz electronic noise will be in the area. Alternatively, since the thickest, most productive sandstones appear to occur above the channels (though productive sandstones do not occur above all the channels), and since increasing amounts of 2-D seismic data point to increasing complexity in the channels, a 3-D seismic survey may go a long way towards mapping out the true position and connectivity of the channels, and identifying their thickness with reflection amplitude. This may aid in finding places where thick sandstones could occur but have been overlooked, and may also aid in determining how the productive sandstones penetrated in the various wells in the fields are connected. However, unless higher frequencies than those acquired in this experiment are recorded at the Morrow-Mississippian clastic interval, identification of sandstone versus shale is remote, and dry holes are still possible.

One final, specific conclusion that can be drawn from this experiment is the modification of the seismic anomaly channel map. Channel margins determined from seismic anomalies visible on this data set have been added to those determined from previous seismic data (Fig. 28). This new map adds to the observation that some of the channels are fairly sinuous, that intrachannel highs with thin Morrow-Mississippian clastic sections do occur, and that the channel margin in sections 8 and 17 may be farther west than previously mapped. Although not indicated on the map in Figure 28, the seismic data collected during this experiment could also be used to map the thickness of the Morrow-Mississippian clastic section using the seismic amplitudes and isochron thickness variations of the peaks above and below the top of the Morrow.

The broader implications of this experiment suggest that extreme care must be take when acquiring high-resolution seismic data over an existing field in an area with a large amount of cultural noise such as power lines and other sources of electronic noise. Moisture in the cable, if only in small amounts, can be a great detriment to acquiring signal above 60 Hz if sources of electrical noise are present in the area. Notch filters, either during acquisition or processing may help reduce or eliminate the noise, but at a cost of phase shifts and frequency holes in the data that can be severe enough to make stratigraphic interpretations extremely difficult. Spectral balancing and deconvolution can help, but will not boost higher frequency signal if it is not there or is of extremely low amplitude compared to the noise. Also, this experiment suggests that loess may not be the best place to set of shots since it appears to soak up a lot of higher frequency energy.