The drill pipe provides a fixed alignment which served as the acoustic equivalent of an optical bench. This mechanical alignment allowed looking in detail at the changes in the seafloor and subseafloor reflections as a function of source distance above the seafloor. Unlike conventional 3.5-kHz profiling, these data are not changing because of horizontal movement of the source but solely because of the height of the source. Because the difference in spreading losses of the different reflections with elevation above the seafloor are miniscule, the amplitudes and duration of the reflections can be directly compared.
The aperture of the Fresnel zone varies with source depth. Table T2 shows the change in Fresnel area of the 4-kHz source with depth. At 5 m above the seafloor, the reflecting area is ~4 m in diameter, compared to 120 m when sounded from the sea surface. For a given transducer beam width, the area insonified at approximately normal incidence changes as the square of the transducer elevation.
Roughness of the seafloor and subseafloor reflections exaggerates this effect, as there are in-phase interfaces normal to the wavefront beyond the Fresnel zone. These combine to lengthen the reflection wavelet.
Figure F7 shows significant differences in the amplitudes of the subseafloor reflections at five heights examined. Note that the amplitudes, durations, and travel times of the peak amplitude of the reflections differ between the traces (Table T3). Each trace is normalized to the amplitude of the seafloor reflection. The amplitude of the portion of the trace before (above) the seafloor reflection indicates the noise background. Data were taken on different days during which the background noise of the receiver on the ship varied. The wash of the ship's forward port thrusters across the 3.5-kHz transducer pod occluded recording when it was thrusting to port. The heavy weather during initial lowering of the vibration-isolated television (VIT) frame to view the proposed spud-in location prevented gathering horizontally changing data. The number of "clean" traces available for stacking differed in each data set. The arrival 4 ms after the seafloor reflection on the fourth trace is from the reentry cone. The drillers measurement indicated that it settled 1.7 m between deployment and setting the casing 9 days later. The acoustic data indicate it settled about 2.8 m.
Two attributes of the subseafloor reflecting interfaces that affect their reflections are as follows:
Applying these criteria, the six reflecting interfaces are characterized in Table T1.
The 0.25-s digitizing window triggered on the direct arrival did not allow a continuous recording of the reflection sequence as the source was lowered and raised. Only those intervals during which the direct water wave (which triggered the acquisition cycle) and seafloor reflections were within the quarter-second were recorded. For the 1-s repetition of the source, this is each 750 m above the seafloor. If there is opportunity to do this experiment again, we would refine the data digitizing scheme to digitize the seafloor reflection continuously. Varying the source repetition rate would be one means for getting additional recording windows capturing both the direct vertical and seabed reflections. This would allow a more detailed examination of the geometric character of the seafloor and deeper reflecting interfaces as a function of source elevation. Because the VIT frame slides on the drill string, there was no azimuthal control. Mounting the source on a remotely operated vehicle or autonomous underwater vehicle would allow freedom of both vertical and horizontal movement.