No abstract

Measurements of the wave shape of the seismic transient generated by dropping a weight indicate that the weight drop has a much higher efficiency than shooting dynamite. The surface wave which is generated by the weight drop is very large, but it is minimized by the use of large seismometer groups, many drops per trace, and large offset distances. For experimental studies, a special recording instrument adaptable for either shooting or weight dropping is used. The weight truck is hydraulically operated. Weight dropping versus shooting studies indicate that the quality of weight drop records is limited by the large surface wave generated by the weight impact on the ground.

No abstract

FIG. 6. Cross-borehole depth section. processing scheme, we apply an FK filter to the commonsource-point gathers to remove the low velocity receiver ho!e tube waves. We then sort the filtered data to commonreceiver-point gathers and apply an FK filter to remove the source-hole tube waves. Finally, we resort the data back to common source-point-gathers for further processing.Unlike conventional surface seismology. a source-receiver pair in the cross-hole setting generates reflections from both above and below. Thus, in our third processing step, we select upgoing (or downgoing) events by FK filtering, as is commonly done in VSP.In the fourth and key step, which is illustrated schematically in Figure 5, we convert each source-receiver time trace to a depth trace. We first use the arrival times of the direct waves to estimate the velocity structure for the region between the two boreholes. Using these velocity estimates, we can compute, for a particular source-receiver pair. the arrival time from a reflector at any depth within the region between the well bores. (In the present work, we simply assumed straightline raypaths and flat reflectors; the method can certainly be refined by employing more elaborate modeling.) Then, since we know the reflection arrival time as a function of depth, we can invert this relation to convert each event on the time trace to depth.The fifth and final step is to stack the data generated by multiple source-receiver pairs. The result of this processing for the field data is displayed in Figure 6. The top half of the section results from imaging with sources and receivers below the reflector. while the bottom half of the section results from imaging with sources and receivers above the reflector. The hourglass appearance of the depth section is caused by the reflection point geometry imposed by the source-and-receiver locations in the two boreholes. Observe that we can see clearly the interface between the glacial till and a 5 ft thick limestone at a depth of I IO ft. The roof of the coal seam at 265 ft is also prominent. ConclusionOur initial work demonstrates that high resolution crossborehole imaging using the full waveform signal is possible provided that the tube wave noise can be suppressed. These low velocity tube waves reverberate with little attenuation within both source-and-receiver well bores, reducing the signal-to-noise ratio of the reflections. Further work needs to be done refining our processing stream, e.g., improving fhe velocity analysis, introducing a deconvolution step, and considering alternative methods of spatial filtering. Perhaps more importantly, we need sources and receivers specifically designed for cross-borehole use.

A study was undertaken to evaluate the effectiveness of the high‐resolution seismic technique for the mapping of stratigraphic and structural controls in the Gas Hills uranium district, Wyoming. The test area is one in which uranium deposits are in Tertiary sediments which unconformably overlie a Mesozoic Paleozoic section. Paleochannels on the unconformity appear to control the localization of the uranium. Drilling in the area allows an evaluation of the effectiveness of the study. Using both sonic and density logs, we computed synthetic seismograms to evaluate the feasibility of predicting the success of the seismic reflection technique and to test this prediction using surface seismic methods. The field study was undertaken utilizing primarily two energy sources—a high‐frequency vibrator (40–350 Hz), and one‐pound dynamite charges shot in 10-ft holes. A limited amount of data was also acquired using detonating cord on the surface. Some three‐dimensional (3-D) data were also acquired, and a later study acquired passive seismic data. The seismic reflection data were successful not only in delineating the unconformable surface and in mapping paleodrainages on the unconformity, but also in defining channel deposits within the Tertiary section. Correlation with the logs shows the success of the study. Several areas were delineated where one would undertake tight drilling patterns, and other areas were delineated in which one might minimize or eliminate exploratory drilling. The synthetic seismograms also could have predicted the success of the seismic work.