We generated a synthetic seismogram from Site 808 data using the LWD density and ISONIC velocity curves and a source wavelet extracted from the seismic reflection data as described in "Core-Log Seismic Correlation and Seismic Resolution" in the "Explanatory Notes" chapter (Fig. F30). LWD densities from Hole 808I were used from 150 mbsf to TD. As the LWD densities from 0 to 150 mbsf were unreliable due to the interference with casing, we used an exponential curve that changed sharply in the near surface from 1.1 g/cm3 at 0 mbsf to the first measured LWD density value at 150 mbsf. For velocities we used a similar curve for 0-150 mbsf to that used for the densities. The seafloor velocity was set at 1500 m/s, and values were interpolated to the first ISONIC LWD velocity measurement at 150 mbsf.
Figure F30 shows the fit of the synthetic seismogram with 10 traces on either side of Hole 808I. Although there is a good match at the seafloor in both waveform and amplitude, the synthetic seismogram exhibits a reflectivity series generally inconsistent with the reflectivity in the seismic data. There are several zones where a reflector or sequence of reflectors in the synthetic correlate with the seismic data in depth but not in amplitude and waveform: ~170-270, ~400-450, ~525-640, ~700-750, and ~900-925 mbsf (Fig. F30). Although there are mismatches in amplitude and waveform, some of these zones can be correlated to specific log units or to boundaries between them: (1) the base of the ~170- to 270-mbsf sequence of reflections correlates with the log Unit 1/2 boundary, (2) the ~525- to 640-mbsf sequence of reflections correlates with log Unit 3, and (3) the remainder of the synthetic seismogram correlates with log Unit 4. The frontal thrust zone reflection is observed in the synthetic trace at ~400 mbsf, but the amplitude is a factor of 2-3 times higher than the seismic traces nearest the borehole. The décollement zone at ~900-925 mbsf produces a reflection, but it matches poorly with the real data. A sequence of high-amplitude reflections in the synthetic seismogram beneath the décollement zone is not observed in the seismic data.
The match in amplitude and waveform between the synthetic and collected seismic data is notably poor between ~170 and 270 mbsf, between ~750 and 850, and below ~925 mbsf (Fig. F30). The interval from ~170 to 270 mbsf exhibits large variations in both velocity and density that produce reflections not seen in the seismic data. The interval between ~750 and 850 mbsf exhibits density values that are unreasonably low and correlates with (1) reflectivity in the synthetic seismogram that is not observed in the seismic reflection data and (2) poor hole conditions as suggested by differential caliper values >1 in. The high-amplitude reflections in the synthetic seismogram below ~925 mbsf are probably caused by inaccurate data acquired as hole conditions quickly deteriorated after penetrating the décollement zone. Some mismatch between the synthetic seismogram and the 3-D data may be caused by the poor quality of the ISONIC log; however, matched reflections at 725 mbsf and below the décollement zone, for example, are produced by the density log.
In summary, beneath the depth of the casing (~150 mbsf) correlations between the synthetic seismogram and seismic reflection data broadly reflect the defined lithologic boundaries or units. However, the details of amplitude and waveform throughout most of the section do not match the seismic data. Amplitudes in the intervals ~200-400 mbsf, ~750-850 mbsf, and below ~925 mbsf are significantly higher in the synthetic section than in the seismic data and are generated by numerous suspiciously low velocity and density values. Production of a more realistic synthetic seismogram awaits careful correction of the density and velocity curves, which now include spurious values.