View larger version (52K):
[in a new window]
|
Figure 1. Progress in ultra-shallow seismic reflection methods over the past two decades is illustrated by comparing the figures above. The figure on the left shows seismic reflection data collected in 1986 at Great Bend, Kansas, USA, in which a water-table reflection is clearly visible at a time of 24 ms. The figures on the right show seismic reflection data collected in 1998 at the same location. The right half of each seismogram is synthetic, generated by finite-difference wave equation modeling; the left half of each one is real data. Again, the water-table reflection is prominently shown, but shallower (at 19 ms) due to aquifer recharge that occurred in the weeks before the data were collected. The 1998 data also show reflections at 8 ms and 14 ms that were not visible in the 1986 data. The depth to the water table was 2.6 m in 1986 and 2.1 m in 1998. The shallower reflections on the right come from depths of 63 cm and 146 cm respectively, and were verified by digging a grave-sized pit by hand with a shovel. The reflection from 63 cm represents a buried soil layer (paleosol), and the reflection from 146 cm represents an abrupt change from fine sand to gravel. The principal improvement that led to the ability to see the shallower two reflections was increased dynamic range of the seismograph. The 1986 data set was collected with a seismograph that had 12-bit analog-to-digital conversion; the 1998 data employed 24 bits. The increased dynamic range allowed the use of a smaller seismic source (.22 versus 30.06 rifle) that generated the high frequencies that are necessary to see ultra-shallow reflections. The increased number of channels in the later data also allowed the use of smaller geophone intervals, which improved the coherency of the reflections.
|
