DISCUSSION

The LPTM log anomaly at Site 999 is abrupt and marks a distinct change in the physical and chemical properties of sediments above and below the LPTM anomaly (Fig. 4). The contrast in physical and chemical properties extends uniformly for 10-15 m above and below the LPTM claystone and indicates significant and long-term differences in the proportion of primary lithologies and their porosity across the LPTM. These observations are consistent with a rapid and extreme environmental change that persists beyond the LPTM anomaly itself. The LPTM log anomaly at Site 1001 does not appear as clear as that shown for Site 999; however, the FMS/SFLU resistivity values also indicate long-term differences in the physical properties of sediments above and below the anomaly. The LPTM interval is also characterized by a thin (short) interval of significantly lower Ca (i.e., carbonate) values, probably reflecting the shallowing of both the carbonate compensation depth (CCD) and the lysocline. Significant environmental changes are also documented by evidence for reduced sea-floor ventilation (e.g., faint lamination, diminishing bioturbation; Sigurdsson, Leckie, Acton, et al., 1997; Bralower et al., 1997).

Downhole and core logging data thus support proposals that the LPTM claystone (log anomaly) is the product of a relatively short paleoceanographic event with an abrupt onset, followed by longer term changes in the ocean circulation that have been documented in contrasting sediment compositions in both low- and high-latitude sites (e.g., Kennett and Stott, 1991; Bralower et al., 1995; Thomas and Shackleton, 1996). At Site 999 there is no indication in the logging data of gradual changes preceding the LPTM that may have eventually crossed a threshold value resulting in the LPTM event. However, the FMS record at Site 1001 shows a 3-m transition from relatively high resistivities below the LPTM down to the local minimum resistivity value marking the LPTM, indicating that there is a gradual change in physical properties preceding the LPTM anomaly at this location (Fig. 7).

At Site 999, the abrupt onset of the FMS anomaly followed by a transition zone to higher values is similar to the pattern displayed by Ca and Fe intensities and magnetic susceptibility data (Fig. 4, Fig. 5); however, these lithologic indices return to their pre-LPTM levels within 1 m above the base of the LPTM, whereas FMS resistivities do not. Ca intensities decrease to a uniform minimum within 3 samples (i.e., 4 cm) indicating an onset time of less than a few thousand years (1.2 k.y. at a sedimentation rate of 32 m/m.y.) and is similar to C isotope patterns displayed by this and other sites worldwide (Bralower et al., 1997; Thomas and Shackleton, 1996).

The LPTM clay includes the negative ¹³C excursion, minimum values of Ca, Fe, and magnetic susceptibility, and is defined as 25 to 37 cm thick at Site 999. This clay is the base of a broader zone (97 cm thick) referred to as the LPTM "interval" defined by low resistivity, and the gradual return of Ca, Fe, magnetic susceptibility, and ¹³C to pre-LPTM levels. At Site 1001 the LPTM clay, including the characteristic FMS patterns observed at Site 999, correlates to index layer "I" and appears to be just 9.2 cm thick. However, the extreme ¹³C anomaly at Site 1001 is ~30 cm thick, which also matches the extent of constant minimum values in Ca, indicating that the LPTM clay is approximately 30 cm thick at this location. The LPTM interval at Site 1001 is 80-cm-thick and includes index layers "H" to "I" (Fig. 6). Thus thickness estimates for the LPTM interval and LPTM clay are comparable at these widely separated sites.

Our estimate of 97 cm for the LPTM interval and 37 cm (maximum) for the LPTM clay at Site 999 corresponds to approximately 30 k.y. and 11 k.y., respectively, assuming sedimentation rates of 32.2 m/m.y. calculated from nannofossil datums (Sigurdsson, Leckie, Acton, et al., 1997). A comparable calculation at Site 1001 indicates a duration of only 21 k.y. for the LPTM interval and 8 k.y. for the LPTM clay (i.e., 80 and 30 cm at 37.4 m/m.y.). These estimates are comparable to those based on species richness parameters and isotopic anomalies, which indicate the duration of the LPTM interval is on the order of 25 to 50 k.y. and that the event itself was in place in less than 10 k.y. (Thomas and Shackleton, 1996; Steineck and Thomas, 1996). The downhole log measurements, primarily at Site 999, indicate that this "short" event is followed by chemical/physical properties changes persisting for millions of years that is not evident in the isotope data. Downhole logging data also provide constraints on the onset time of the LPTM event. The geological high-sensitivity magnetic tool (GHMT) was used at Hole 1001A and a magnetic polarity reversal stratigraphy has been established (V. Louvel and B. Galbrun, unpubl. data). Analysis of GHMT data and conventional paleomagnetic analysis of cored samples indicate that Chron 24r extends from 254.4 to 212.2 mbsf (log depth), which corresponds to 55.904-53.347 Ma (V. Louvel and B. Galbrun, unpubl. data; Cande and Kent, 1995) (Fig. 4). The FMS image indicates the LPTM clay at Site 1001 begins at 240.867 mbsf. Assuming a Late Paleocene sedimentation rate of 37.4 m/m.y., based on nannofossil datums (Sigurdsson, Leckie, Acton, et al., 1997), the LPTM claystone appears abruptly at 55.54 Ma. The occurrence of the LPTM at sometime early in C24r is consistent with other studies (e.g., Aubry et al., 1996; Berggren and Aubry, 1996), and the quantitative estimate of 55.54 Ma is similar to estimates made in recent studies (e.g., LPTM onset at 55.5 Ma; Thomas and Shackleton, 1996). This observation also provides further, independent support for the depth location of the LPTM based on downhole measurements. Unfortunately, GHMT was not used at Site 999, and it was not possible to obtain magnetostratigraphic data from the cores as the drilling overprint could not be removed (B. Galbrun, pers. comm., 1998).

Linear sedimentation rates used in this study are over estimated because they are calculated across distances (time intervals) that include the pronounced decrease in carbonate content which presumably reflects intense dissolution (i.e., an interval of lower sedimentation rate). Therefore our quantitative estimates of onset and duration times for the LPTM must be considered minimum values.