3-D vertical seismic profiling at LLNL Site 300

Abstract

The initial goal of the 3-D Vertical Seismic Profiling (VSP) work at LLNL was to characterize seismic wave velocities and frequencies below the vadose zone to design the acquisition geometry for a 3-D shallow surface seismic reflection survey. VSPs are also used routinely to provide a link between surface seismic data and well logs. However, a test 2-D seismic line recorded at LLNL in the Spring of 1994 indicated that obtaining high quality reflection images below the vadose zone, yet shallower that 50 m, would require an expensive, very finely sampled survey ({lt} 1 m receiver spacing). Extensive image processing of the LLNL 2-D test line indicated that the only reliable reflection was from the top of the water table. Surprisingly, these results were very different than recent 3-D seismic work recorded at other sites, where high quality, high frequency surface (up to 300 Hz) reflection images were obtained as shallow as 20m. We believe that the differences are primarily due to the comparatively deep vadose zone at LLNL (15 to 30m) as compared to 0-5m at other sites. The thick vadose zone attenuates the reflection signals, particularly at the high frequencies (above 100 @). In addition, the vadose zone at LLNL creates a seismogram in which surface-propagating noise overlaps with the reflection signals for reflections above 50 m. By contrast, when the vadose zone is not thick, high frequencies can propagate and noise will not overlap with reflections as severely. Based on the results from the 2-D seismic line and the encouraging results from a VSP run concurrent with the 2-D seismic experiment, we modified the objectives of the research and expanded the scope of the VSP imaging at LLNL. We conducted two 3-D multi-offset VSP experiments at LLNL in the Summer and Fall of 1994. These VSP experiments were designed to characterize the seismic propagation characteristics at two different locations on the LLNL site: the first was a well with a relatively shallow water table (10m), while the second was a well with a relatively deep water table (about 25m). Other goals of the VSP experiments were to characterize the velocity structure in the vicinity of boreholes, and to attempt to image reflections away from the boreholes. We found that the usable frequency content appeared to vary with water table level.