A combination of two methods holds promise for faster land seismic acquisition.

An experiment was conducted near the city of Alpine in West Texas to investigate a unique seismic acquisition technique. The potential of the technique involves the union of the High Fidelity Vibratory Seismic (HFVS) acquisition method with either the Slip-Sweep or the Concatenated method of Vibroseis acquisition. Among the benefits provided by HFVS recording is the ability to separate the sources during the processing of the seismic data. Source separation allows for enhanced spatial sampling, which improves the resolution and the bandwidth of the seismic image. Among the benefits of the Slip-Sweep or the Concatenated method of seismic acquisition is an increase in the production rate. An increase in the production rate can be used to collect data faster, or to acquire more seismic data per unit area, which can be used to improve spatial sampling or increase the fold. The new technique capitalizes on the benefits inherent in both methods.

Methodology

The HFVS method requires one vibrator set. There can be one or more vibrators in the vibrator set; however, the minimum field effort in sweeps per vibrator point (VP) is equal to the number of vibrators in the set. Another HFVS requirement is that with each sweep on a VP a vibrator must be "phased," typically 90° or 180°, from the other vibrators. The minimum field effort ensures all the vibrators in the set are "phased" relative to the others with one sweep on each VP. The production rate of the HFVS seismic acquisition is roughly equivalent to a conventional acquisition with similar recording parameters. However, some extra acquisition and processing time can be involved with the handling of the uncorrelated data from each sweep and the measured motion of each individual vibrator required with the HFVS acquisition.

The Concatenated method requires one vibrator set. The multiple sweeps associated with each VP are simply concatenated (piggy-backed) with the listen time at the end. The Concatenated method saves listen and reset times between the multiple sweeps on the vibrator point.

The Slip-Sweep method utilizes multiple vibrator sets. Production rates increase dramatically when two or more vibrator sets sweep simultaneously. Slip-Sweep can also be used in a "safer" application with the sets sweeping only during listen and reset or programmed "dead" time of the other set, with no simultaneous sweeping. The "safer" application makes the Slip-Sweep a cousin to the Concatenated method. The economic advantage of the Slip-Sweep technique can be somewhat offset by the extra expense of additional vibrators necessary to comprise sets, which depends on the crew's capability.
The new technique simply combines the two methods. For example, the HFVS combined with the Concatenated method requires as many concatenated sweeps on a VP as there are vibrators in the set. With each concatenated sweep one of the vibrators is "phased" relative to the others. When the HFVS is combined with the Slip-Sweep method, the minimum number of sweeps per vibrator set must equal the number of vibrators within each set.

Field implementation

The experiment was conducted into a 6.8-mile (11-km) 2-D receiver spread comprised of 164 stations at 220-ft (67-m) intervals. The geophone array consisted of six geophones over 55 ft (17 m) and was deployed inline. Roughly 1 mile (1.6 km) of the receiver line was recorded by the sources. The fold of the datasets is approximately 24. The common data point (CDP) spacing of the datasets varies with the source separation allowed by the particular acquisition technique and ranges from 27.5 ft to 110 ft (8.4 m to 33.5 m). Four Vibroseis datasets were collected. The base Vibroseis sweep is 8 seconds in duration with a 5-second listen time. The vibrators were swept over a frequency range of 8-80 Hz with an emphasis of 3dB/octave. The taper is 200 ms. The first three datasets were acquired as controls for the comparison of the two acquisition techniques with each other and the norm. The fourth dataset was obtained with the new combined technique to be evaluated with the three baselines.

Test #1 represents a conventional production scenario that was collected with four vibrators stacked on each of the 24 VPs. Four sweeps with the base parameters were performed at each source point. The net effect after processing is four vibrators with four sweeps on 220-ft (67-m) surface intervals.

Test #2 was recorded with the HFVS method. Four vibrators were distinctly positioned relative to a 220-ft VP flag on four consecutive 220-ft flags. Four base sweeps were executed with the HFVS phase rotation and each sweep output to tape. All four vibrators moved up 220 ft, which ensured the relative positions and the process repeated for 1 mile of vibrator points. The net effect after source separation is a single vibrator performing 4 sweeps on 55-ft (16.8-m) source intervals.

Test #3 was obtained with the Slip-Sweep technology. It was used in the "safe" mode so the dataset could also be representative of the Concatenated sweep method. In the "safe" mode the two vibrator sets did not sweep simultaneously. Each vibrator set had a sweep sequence comprised of four base sweep segments. After each base segment of vibrator Set One, 8 seconds of dead time was programmed. The four base segments of vibrator Set Two were interleaved into the dead time of vibrator one, plus a 5-second listen time after the last base sweep. The sets were initially deployed 110 ft (33.6 m) apart and moved up 220 ft upon the completion of each sequence. One mile or 24 Slip-Sweep set sequences were recorded by the sets. When the Slip-Sweep method is used, the vibrator sets are realized in processing, but the individual vibrators are not. After the sets were separated in processing, the net effect is two vibrators with four sweeps on 110-ft source intervals.

Test #4 (Figure 1) was recorded with the same sweep sequencing as the Slip-Sweep baseline Test #3, with two primary differences. During the base sweep segments for each sequence of each vibrator set, the vibrators were "phased" according to the HFVS constraint. Because Test #4 was recorded with the HFVS constraint, each individual vibrator that comprised a set can be separated from the composite record. Thus the HFVS portion of the combination allows for the same distinct positioning of the vibrators relative to each 220-ft VP as in Test #2. The net effect of the acquisition technique, as displayed in Figure 1, is one vibrator with four base sweeps on 55-ft (16.8 m) surface spacing.

Data analysis

Each dataset was processed independently of the others. First break, refraction statics and velocities were picked and calculated on the four datasets. Additional steps in the processing flow include predictive deconvolution, spectral whitening and residual statics. The datasets were CDP-stacked and migrated according to the corresponding surface intervals.

The processed datasets are presented in Figure 2a through 2d. Note the similarities of the data outlined by the green ellipse in Figures 2a (base case), 2b (HFVS baseline) and 2d (HFVS/Slip-Sweep). The HFVS/Slip-Sweep data represented by Figure 2d was acquired in roughly 60% of the time as the Conventional Production data displayed in Figure 2a. The traces displayed in Figures 2b and 2d were decimated post migration from 27.5 ft to 110 ft (8.4 m to 33.5 m) CDP intervals for comparison with the Conventional Production case (110-ft, 33.5 m CDPs) shown in Figure 2b. The traces in Figure 2c were decimated post migration from 55 ft to 110 ft (16.8 m to 33.5 m) CDP intervals. Of course, in production the much improved spatial sampling would be preserved to enhance signal-to-noise, especially for 3-D data.

Conclusions

The HFVS is not a faster acquisition method, but the technique allows for the individual sources that comprise the vibrator sets to be separated. The Slip-Sweep method improves acquisition efficiency, and the vibrator sets can be extracted; however, the individual sources cannot. Therefore, an acquisition technique that provides improved spatial resolution as well as acquisition efficiency is welcome.