Conclusions

Starting with nonconclusive velocity logs as a result of difficult-to-log slow formation with extremely high internal velocity contrasts and the harshness of an uncentered logging tool string due to the need to log without the centralizing bow springs, we tried to improve the quality of the data obtained and present measures to evaluated data previously inaccessible with standard processing techniques. Even though the velocity profile produced correlates well with other logs obtained (neutron porosity, bulk density, self-focusing electrical resistivity data, and magnetic susceptibility) (Fig. F12) and offers a reasonable estimate for the location of a prominent shelf unconformity, we should emphasize the limitations of the data. Possible errors introduced may result from

1. A bias in choosing and defining the data categories;
2. The possibility of exclusion of rare good data in a given interval where misleading values represent the majority of the data;
3. An uncertainty of the precise depth location of data from transmitter and receiver spacings that are collecting different regions along the tool string for a given tool position;
4. The mixing of non-depth shifted raw data of logging runs one and two; and
5. A bias in the final correlation and no linear depth shift to the resistivity log using the only general guideline that low resistivity zones are most commonly denser and, therefore, acoustically faster.

Nevertheless, in contrast to seismically derived velocity information (Tinivella et al., Chap. 16, this volume) the velocity information presented here is still more detailed and allows the investigation of the seismic character at least on the scale of the defined logging units (Fig. F12). We hope that the new data will help seismostratigraphers, modelers, and sedimentologists to uncover the beauties and complexities of the Antarctic shelf.