There are marked similarities between the Pliocene section in Hole 986D and some of the other Pliocene-aged deposits mentioned above, in particular the Geodia sp.Globigerina bulloides Zone (D) in Well 7117/9-1 on the Senja Ridge (Eidvin et al., 1993a) and the Globigerina bulloides-Cassidulina teretis Zone (BB-FB) in Well 7316/5-1 in the Vestbakken Volcanic Province (Eidvin et al., 1998b; Fig. 8). Although the foraminifer assemblages in Wells 7117/9-1 and 7316/5-1 differ somewhat from those of Hole 986D in their generally lower diversity and poorer preservation, many of the recorded species are the same. In all sections, C. teretis and M. zaandamae dominate the calcareous benthic assemblages, whereas G. bulloides dominates the planktonic faunas. In addition, all sections contain many reworked calcareous and agglutinated specimens from Miocene to Eocene deposits. However, the analyzed section in Hole 986D contains more benthic species of deep-water affinity.
A maximum age of 2.7 Ma is assigned to the sections in Wells 7117/9-1 and 7316/5-1, and these sections probably represent sediments deposited after the great increase in the supply of ice-dropped material at the Leg 104 sites at the Vøring Plateau (Jansen and Sjøholm, 1991; Eidvin et al., 1993a, 1998b). A minimum age of 2.4 Ma is based on the fact that G. bulloides is common in Pliocene sediments older than 2.4 Ma at the Vøring Plateau sites (Spiegler and Jansen, 1989). However, G. bulloides is also found in the warmest interglacials of the last 1 Ma (Kellogg, 1977).
The assemblages in Hole 986D also have a strong affinity with the faunas recorded from the section between 151.27 and 502.97 mbsf in Hole 910C (Leg 141). This section also has a benthic assemblage with C. grossus (Osterman, 1996) and common C. teretis, M. zaandamae, and Epistominella associated with rare E. albiumbilicatum. The planktonic assemblage is dominated by N. atlantica (sin.) and G. bulloides. Paleomagnetic and fossil evidence date this section to the late Pliocene. A maximum age of 2.7 Ma is assigned to the base of the section (Spiegler, 1996; Fig. 8).
The interval in between Sample 104-643A-7H-2, 74-78 cm, and 7H-5, 74-78 cm, on the Vøring Plateau also contains common C. teretis, M. zaandamae, N. atlantica (sin.), and G. bulloides. Paleomagnetic records date this section from latest early Pliocene to earliest late Pliocene (Spiegler and Jansen, 1989; Osterman and Qvale, 1989; Fig. 8).
According to Channell et al. (Chap. 10, this volume), the uppermost 150 m at Site 986 appears to record the Brunhes/Matuyama boundary and the Jaramillo Subchron. The base of the drilled section (at ~950 mbsf) is interpreted to lie within the Matuyama Chron (age <2.58 Ma) with the normal polarity interval (interpreted as the Olduvai Subchron) occurring from ~730 to 750 mbsf. The age of the Olduvai Subchron is 1.76-1.98 Ma according to Cande and Kent (1992). This implies that the LO of N. atlantica (sin.) and the last common occurrence of G. bulloides are <1.7 Ma. In other words, this results in an apparent discrepancy of 0.7 m.y. at Site 986 compared to other ODP/Deep Sea Drilling Project Sites in the Norwegian Sea or the North Atlantic. This discrepancy is especially unlikely for the transitional dwelling G. bulloides. Extensive reworking of planktonic foraminifers at Site 986 could explain this problem, but few of the tests show any sign of wear. Less than 200 k.y. is an extremely short time to deposit the analyzed section. However, such extreme accumulation rates, in the 2.7-2.4 Ma time interval, are recorded from the western margin of the Barents Shelf (Eidvin et al., 1993a, 1998b), the Norwegian Sea continental shelf (Eidvin et al., 1998a), and the northern North Sea (Eidvin and Riis, 1992). For the time being, we are unable to find a solution on the mentioned discrepancy between the paleomagnetic and biostratigraphic data.
The porous noncompressed assemblages observed between 907.22 and 955.32 mbsf, between 830.22 and 841.32 mbsf, and at 794.72 mbsf consist exclusively of agglutinated taxa. The absence of calcareous benthic and planktonic foraminifers from these intervals indicates that the deposition of the assemblages has taken place most likely below the local carbonate compensation depth, although diagenetic dissolution of carbonate cannot be totally excluded.
The four silicified compressed assemblages and the seven mixed agglutinated assemblages are associated with a calcareous faunal component consisting of relatively few benthic specimens and more common planktonic taxa. It seems also likely that the calcareous benthic component is at least partly reworked, as suggested by the common occurrences of broken specimens.