Magnetic polarity and microfossil zonations (calcareous nannofossils, foraminifers, and dinoflagellates) used for Leg 149 sites employed the chronostratigraphical time scales of Harland et al. (1990), Gradstein et al. (1994), and Shackleton et al. (in press).
A combined paleomagnetic and biostratigraphic study of the Cenozoic sedimentary sequence was conducted by Zhao et al. (this volume). Continuous and undisturbed Cenozoic sections, of sufficient length to allow the identification of geologically useful magnetic polarity sequences, were obtained from Sites 897, 898, and 900, but not from Sites 899 and 901 (Table 2). Therefore, Sites 899 and 901 were not considered any further by Zhao et al. (this volume).
Sedimentary sections of late Neogene age from Holes 897C, 898A, and 900A recorded a pattern of magnetic polarity reversals that correlates well with the known magnetic polarity time scale for the past 5 Ma, and allows the determination of accurate sediment accumulation rates.
The polarity pattern from the Pliocene-Pleistocene turbidite sequence shows that a reliable magnetostratigraphy can be established from the early Pliocene to the Holocene, including the Gilbert/Gauss boundary (3.58 Ma) through the Matuyama/Brunhes boundary (0.78 Ma). Below the middle Miocene angular unconformity that separates the Pliocene-Pleistocene turbidites from the Miocene and earlier strata, a reliable magnetostratigraphy could not be established because of the extremely weak magnetization of the sediments (Fig. 9, Fig. 10, Fig. 11). This sudden downhole decrease in magnetic signal corresponds in general to a large decrease in iron content and an increase in sulfate concentration (Sawyer, Whitmarsh, Klaus et al., 1994), suggesting magnetic mineral dissolution. This magnetic boundary may be caused by the termination in supply of terrigenous material at the end of the early Miocene and reflects the occurrence of the middle Miocene tectonic folding event.
The biostratigraphic data for Sites 897, 898, 899, and 900 indicate that the stratigraphic column is incomplete between the early Eocene and the late Pleistocene (de Kaenel and Bergen, this volume; Liu, this volume; de Kaenel and Villa, this volume; Liu et al., this volume; Gervais, this volume).
An incomplete foraminiferal biostratigraphy was identified at Sites 897, 898, 899, and 900 (Gervais, this volume) from Upper Cretaceous to Pleistocene sediment. Dissolution and changes in water-mass temperatures affect the quality of biozonation in the Pliocene, Miocene (particularly the latest middle to early late Miocene), and the early Eocene (Gervais, this volume). Several hiatuses occur throughout the sedimentary column.
Benthic foraminifers faunas were investigated from the sediments of Cretaceous to Quaternary age at Sites 897, 898, 899, and 900 (Kuhnt and Collins, this volume; Collins et al, this volume). Knowledge about the abyssal benthic foraminifers in the northeastern Atlantic is limited and sediments were examined to improve biostratigraphic information for sediments deposited below the CCD and to obtain paleoecologic information on the depositional environment of these sites. They provided helpful stratigraphic boundary markers for dating lithologic Unit III (Kuhnt and Collins, this volume).
Calcareous nannofossils are generally common to abundant; preservation is generally good in the upper part of the section. From the upper Aptian to the lower Eocene, major parts of the sediment column are barren of nannofossils, because of deposition below the CCD. A maximum of 38 nannofossils datums (Site 897) could be identified from the Pleistocene to the Lower Cretaceous.
A similar sequence of sedimentation events is seen at Sites 897 to 900, as shown in Figure 12, Figure 13, Figure 14, Figure 15, with some differences in the sedimentation rates between sites:
Figure 16 presents the accumulation rates for Cenozoic sediments at Sites 897, 898, and 900 in further detail. The average accumulation rate over the whole Pliocene-Pleistocene turbidite sequence varies between the three sites, increasing with increasing water depth. The rate averages 99 m/m.y. at Site 897 (Fig. 16A). The overall average is 92 m/m.y. at Site 898, but single rates vary from 64.8 m/m.y. (1.95-1.46 Ma) to a rapid 167 m/m.y. (1.46-1 Ma) and then back to 71 m/m.y. (1 Ma to the present) (Fig. 16B). This interval of higher accumulation rate corresponds in part to thicker and more sandy turbidites, which occur in Cores 149-898-8H to 10H. The sedimentation rate at Site 900 (25.8 m/m.y.) is fairly constant through the Pleistocene (Fig. 16C) and declines to 10.6 m/m.y. for the late Pliocene. This change is consistent with accumulation within the lysocline but above the CCD, where a proportion of the CaCO3 deposited on the seabed is dissolved before burial. In Site 897, the pelagic clay sequence immediately beneath the Pliocene/Pleistocene turbidites has an accumulation rate of 4.2 m/m.y. (Milkert et al., this volume). This is consistent with an elevated CCD prior to the onset of Northern Hemisphere glaciation at 2.6 Ma.