Sediments recovered from Site 1081 represent a relatively continuous hemipelagic section spanning the last ~9 m.y. The micropaleontological studies were carried out on core-catcher samples from Hole 1081A. Additional samples from within the cores were examined for calcareous nannofossil- and diatom-based biostratigraphy. Because both siliceous and calcareous microfossil index species are present throughout this site, an integrated biostratigraphic framework can be established (Fig. 20; Table 2). Sedimentation rates are fairly constant within the late Miocene and early Pliocene (average of 4 cm/k.y.). Sedimentation rates within the late Pliocene are the highest recorded for Site 1081A (9–15 cm/k.y.), but they are reduced at the Pliocene/Pleistocene boundary (~7 cm/k.y.). With the exception of planktonic foraminifers, all microfossil groups (calcareous nannofossils, benthic foraminifers, diatoms, and radiolarians) show marked fluctuations in abundance (Fig. 21), which are reflected in the lithology (see "Lithostratigraphy" section, this chapter). Planktonic foraminifers are absent, rare, or replaced by pyrite from Sample 175-1081A-12H-CC to the bottom of the hole.
Calcareous nannofossils were studied in core-catcher samples from Hole 1081A. Additional samples from within the cores were examined close to datum events to improve the stratigraphic resolution. Overall abundance ranges from abundant to very abundant within the upper (Cores 175-1081A-1H through 9H) and lower parts of the section (Cores 175-1081A-39X through base). Samples from Cores 175-1081A-10H through 25X are commonly barren or poor in calcareous nannofossils (Fig. 21), although some nannofossil-rich sediment occasionally is found within narrow intervals (lower part of Core 14H; upper part of Core 18H; upper part of Core 21H). A third unit spanning Cores 175-1081A-26X through 38X contains common to abundant nannofossils. Preservation is good to very good except within the barren and nannofossil-poor interval, which commonly contains heavily etched and dissolved specimens.
Within the sampling resolution, the sedimentation appears continuous throughout the entire section (Fig. 20). Site 1081 terminated in the late Miocene within Zone NN10. Based on the oldest identified calcareous nannofossil datum, the bottom age is estimated at 9 ± 0.2 Ma.
The nannofossil-based biostratigraphy of the Neogene part of Site 1081 is imperfectly constrained because of the scarcity of index species. Most of the biohorizons used to define the Miocene and Plio-cene are based on the first occurrence (FO) or last occurrence (LO) of Discoaster species. These datum events are difficult to find in high latitudes, marginal seas, and eastern boundary current areas. Also, the other marker species belong to genera (i.e., Amaurolithus, Cerato-lithus, and Triquetrorhabdulus) that are more common in low latitudes. Consequently, the Zone NN18/NN17, NN15/NN14, and NN14/NN13 boundaries could not be identified within this section.
This zone covers the last 21.9 mbsf of Hole 1081A. The FO of the Emiliania huxleyi acme, which defines the Zone NN21b/NN21a boundary, is between Samples 175-1081A-2H-1, 100 cm, and 2H-4, 90 cm.
This interval spans 0.2 m.y. from the middle of oxygen-isotope Stage 8 to the middle of isotope Stage 12. The LO of Pseudoemiliania lacunosa, the datum event for the Zone NN20/NN19 boundary, was identified between Samples 175-1081A-4H-CC and 5H-2, 120 cm.
In addition to the zonal boundary events, five biohorizons are identified within this interval. These are the LO of the Small Gephyrocapsa acme (0.6 Ma; Weaver, 1993) at 40.81 mbsf; the LO of Reticulofenestra asanoi (0.83 Ma) at 54 mbsf; the LO of Small Gephyrocapsa acme (0.96; Gartner, 1977) at 63.69 mbsf; the LO of Helicosphaera sellii (1.25 Ma) at 83.46 mbsf; and the LO of Calcidiscus macintyrei (1.67 Ma) at 107.88 mbsf.
The top of Zone NN18 is defined by the disappearance of the last Discoaster species, D. brouweri (between Samples 175-1081A-18H-1, 93 cm, and 18H-4, 80 cm). This species is one of the few star-shaped calcareous nannofossils to be found consistently within the Neogene interval of Hole 1081A. D. pentaradiatus, whose LO event is used to define the Zone NN18/NN17 boundary, is too sparse within the late Pliocene part of the section to allow for a clear definition of this zonal boundary.
The datum event for the Zone NN17/NN16 boundary (LO of D. surculus) was identified between Samples 175-1081A-23X-5, 80 cm, and 23X-CC. The base of this zone is defined by the LO event of one of the few non-Discoaster marker species of the Neogene, Reticulo-fenestra pseudoumbilica. This datum (3.82 Ma) occurs slightly before the late/early Pliocene boundary and can therefore be used to approximate this epoch boundary.
This interval spans 1.72 m.y. between 5.54 and 3.82 Ma. The horseshoe-shaped calcareous nannofossils of the genera Amaurolithus, Ceratolithus, and Triquetrorhabdulus, which define the various zonal boundaries within this interval, are virtually absent from the analyzed samples. The Zone NN12/NN11 boundary was identified between Samples 175-1081A-34X-CC and 35X-2, 120 cm.
This interval is defined by the range of D. quinqueramus and its synonym species D. berggrenii. The FO of D. quinqueramus (Zone NN11/NN10 boundary) was identified within Core 175-1081A-47X at the mean depth of 430.50 mbsf.
The presence of D. loeblichii and D. neorectus, two species restricted to Zone NN10 within Samples 175-1081A-47X-CC and 48X-CC, confirms that Site 1081 terminated within this interval.
The uppermost assemblage is dominated by Globigerinoides sacculifer and Globigerina bulloides, and Hastigerina siphonifera is abundant. Neogloboquadrina pachyderma (dextral), Orbulina universa, Globorotalia truncatulinoides, Globigerinella calida, Globigerinoides ruber, Globigerina falconensis, Globigerinoides sacculifer, Globigerina quinqueloba, and Globorotalia crassaformis also are present (Table 3). This assemblage is a mixture of warm-water fauna associated with the Angola Current and cool-water fauna. N. pachyderma (dextral) and G. bulloides are common in the fringes of upwelling cells and are associated with the Benguela Coastal Current (Little et al., 1997). There is little downcore variation (Table 3). Tropical species, Globorotalia menardii and Globorotalia tumida, are present in Samples 175-1081A-6H-CC and 7H-CC and may indicate intervals of strong Angola Current flow. Sample 175-1081A-9H-CC is barren, and Samples 175-1081A-10H-CC and 11H-CC contain rare planktonic foraminifers (Table 3).
The majority of the core is affected by dissolution and pyritization. Samples 175-1081A-12H-CC through 24X-CC, 26X-CC, 28X-CC, 29X-CC, and 31X-CC through 35X-CC are barren. Planktonic foraminifers in Samples 175-1081A-37X-CC through 49X-CC are replaced by pyrite, although Samples 37X-CC, 41X-CC, and 43X-CC contain a trace of unreplaced foraminifers. Foraminifers in Sample 49X-CC are completely replaced. Samples 175-1081A-25X-CC and 30X-CC contain only trace numbers of planktonic foraminifers. Assigning the planktonic foraminifers to zones is difficult because of the generally very low abundance and lack of index fossils. Samples 25X-CC through 49X-CC probably are Pliocene to late Miocene in age.
In Sample 175-1081A-25X-CC, planktonic foraminifers are rare. Globorotalia puncticulata is present, but G. inflata is absent. This absence may be caused by dissolution, but it may also be an evolutionary event. The first-appearance datum for G. inflata occurs in the late Pliocene. Sample 175-1081A-30X-CC contains only trace amounts of foraminifers. The age is constrained to the late Miocene to Plio-cene based on the presence of Globigerinoides woodi, which ranges from the late Oligocene to late Pliocene, and on N. pachyderma, which ranges from the late Miocene to the Holocene (Kennett and Srinivasan, 1983).
Samples 175-1081A-37X-CC to the base of the section are of Mio-cene age. The boundary may be higher in the section (e.g., 175-1081A-34X-CC), as suggested by the other microfossil groups, but the interval is barren of planktonic foraminifers. Furthermore, most of the samples below 37X-CC are pyritized. Replacement by pyrite destroys some diagnostic information, such as wall structure and tooth place preservation, and renders the zonation tentative.
The few existing foraminifers in Sample 175-1081A-37X-CC were tentatively identified as Globorotalia conoidea (middle to late Miocene) and D. altispira (early Miocene to late Pliocene). Sample 175-1081A-39X-CC provides some stratigraphic control for the base of the core: the presence of G. conoidea indicates an age of middle to latest Miocene. Sphaeroidinellopsis seminulina is also present. Sample 175-1081A-41X-CC contains N. pachyderma and G. conoidea and is of late Miocene age. Sample 175-1081A-43X-CC has very few specimens, which only constrains the age to the Miocene: Sphaer-oidinellopsis seminulina seminulina, Orbulina universa, and Globigerinella obesa; however, the assemblage in Sample 175-1081A-46X-CC is constrained to the late Miocene by the presence of N. acostaensis, G. conoidea, G. sacculifer, O. universa, and S. seminulina seminulina. Samples 175-1081A-47X-CC through 49X-CC contain Globorotalia conoidea (middle to latest Miocene).
The absolute abundance of benthic foraminifers is high in Samples 175-1081A-1H-CC through 8H-CC (Fig. 21). This interval is followed by an interval comprising Samples 175-1081A-9H-CC through 17X-CC that are barren or have very low abundance (Table 4). Before this, there is a succession of barren samples or samples with low abundances alternating with samples showing high abundance of benthic foraminifers. In general, preservation follows the absolute abundance pattern, suggesting that the absolute abundance reflects the degree of dissolution.
Only a few benthic foraminifer species are present throughout the entire Hole 1081A. Uvigerina auberiana is present in high relative abundance (between 4% and 89%, average 41%) in the pre-Pleisto-cene interval (Samples 175-1081A-14H-CC through 49X-CC) disregarding barren samples, samples with lithified sediment, and samples with too few specimens on which to base reliable percentage estimates. This species is also present in the Pleistocene part of Hole 1081A (Samples 175-1081A-1H-CC through 10H-CC), but in lower relative abundance (between 1% and 21%, average 5%). Other species present throughout Hole 1081A are Globocassidulina subglobosa and Oridorsalis umbonatus, although they are present in much lower relative abundance compared to Uvigerina auberiana (generally <5% each).
Species restricted to, or with their highest abundance in, the Pleistocene are Bolivina sp.1, Bolivina sp. 2, Bulimina aculeata, Bulimina marginata, and Cassidulina laevigata, (Table 4; Fig. 22). The late Pliocene is dominated by Bulimina exilis and Uvigerina auberiana (between 25% and 90%, average 67%; Fig. 22). The early Pliocene is characterized by high abundance of Cibicidoides pachyderma, stilo-stomellas (often broken), and Uvigerina auberiana. The upper part has high abundance of the arenaceous species Siphotextularia con-cava, whereas the lower part has high abundance of Trifarina sp. 1. Siphotextularia concava seems to be restricted to an interval covering the early/late Pliocene boundary. The Miocene section of Hole 1081A is dominated by Cibicidoides pachyderma and Uvigerina auberiana, with contributions from several species (e.g., Cibicidoides bradyi, Globocassidulina subglobosa, Oridorsalis umbonatus, and stilo-stomellas). The species Cibicidoides brady is not found in post-Mio-cene sediments at Hole 1081A, whereas Cibicidoides pachyderma and Sigmoilinopsis schlumbergeri seem to be restricted to the Mio-cene and early Pliocene.
Radiolarians are present in most of the core-catcher samples of Hole 1081A (Table 5). In the upper sequence, radiolarians are generally abundant and well preserved. In the lower sequence, radiolarians are rare and show signs of dissolution, although concentrations by the sample preparation often produce high abundances of radiolarians in the assemblage slides. Radiolarian fauna indicates a Quaternary to late Miocene age for Hole 1081A. No apparent reworking has been identified.
The radiolarian zones used for this hole are those of Moore (1995) and Motoyama (1996). There are some difficulties in applying the established tropical and Antarctic zonations (Sanfilippo et al., 1985; Lazarus, 1992) because of the absence of index species such as Pterocanium prismatium, Spongaster pentas, Diartus hughesi, Helotholus vema, Prunopyle titan, and Amphymenium challengerae, and because of the extended occurrences up to the uppermost Pleistocene of Eucyrtidium calvertense and Pterocanium trilobum, which became extinct in the Pliocene or earlier Pleistocene in the Antarctic Ocean.
The absence of Axoprunum angelinum indicates that the uppermost cores (175-1081A-1H-CC and 2H-CC) are within either the Pleistocene Collosphaera tuberosa Zone or the Pleistocene to Holo-cene Buccinosphaera invaginata Zone of Moore (1995). A finer zonal resolution could not be achieved because of the absence of B. invaginata.
Although the diagnostic species Anthocyrtidium angulare is absent throughout the core, Samples 175-1081A-5H-CC and 6H-CC are approximately assigned to the Pleistocene A. angelinum Zone or Amphirhopalum ypsilon Zone of Moore (1995) based on the presence of A. angelinum and the absence of Lamprocyrtis neoheteroporos. The diagnostic species C. tuberosa, used to recognize the A. angelinum and A. ypsilon Zones, is absent throughout the core.
The LO of Cycladophora pliocenica, which became extinct at 1.78 Ma in the Antarctic Ocean (Caulet, 1991), is placed in Sample 175-1081A-14H-CC, approximating the Pliocene/Pleistocene boundary between Samples 175-1081A-13H-CC and 14H-CC. Thus, the lower boundary of the A. angulare Zone, originally defined by the LO of P. prismarium, should be placed at about Sample 14H-CC.
The FO of Cycladophora davisiana indicates an age of 2.7 Ma within the interval between Samples 175-1081A-19X-CC and 20X-CC. This event is well correlated to the LO of Stichocorys peregrina (2.69 Ma) in the tropical Pacific (Moore, 1995); thus, it approximately places the lower boundary of P. prismatium at the same interval.
Spongurus pylomaticus first occurs at Sample 175-1081A-33X-CC, giving an age of 5.2 Ma and a Miocene/Pliocene boundary for the horizon just below this sample. Samples 175-1081A-20X-CC through 33X-CC between the FO of C. davisiana and the FO of S. pylommaticus can be correlated with a zonal sequence from the lower part of the Cycladophora sakaii Zone to the S. pylomaticus Zone in the North Pacific (Motoyama, 1996).
In the lower sequence below Sample 175-1081A-33X-CC, the fauna consists mainly of Spumellaria, and age-diagnostic forms are sparse. Therefore, no zones are defined for Samples 175-1081A-34X-CC through 49X-CC. The rare occurrences of tropical zonal marker species Didymocyrtis antepenultima within Samples 42X-CC, 46X-CC, 47X-CC, and 48X-CC possibly indicate that these samples belong to the middle part of the late Miocene D. antepenultima Zone. The presence of Stichocorys peregrina probably indicates that the deepest sample (49X-CC) is no older than ~ 9 Ma, because its FO approximates that age in the high-latitude Southern Ocean (Lazarus, 1992), although the FO of this species is at ~7 Ma in the tropical region.
S. peregrina seems to prefer tropical to temperate oceanographic conditions (Motoyama, 1996). Its last (common) occurrence is highly variable in age in different places; it disappeared in the late Miocene (~6.8 Ma) in the high-latitude North Pacific, whereas it became extinct at 2.6 Ma in the tropical region. At Hole 1081A, S. peregrina disappeared in Sample 175-1081A-24X-CC, far below the FO of C. davisiana (2.7 Ma). The presence of S. peregrina through the lower sequence indicates temperate oceanographic conditions, and its terminal event indicates a retirement of temperate conditions from the Walvis Basin region. Simultaneously, an Antarctic species, C. pliocenica, first appeared in Sample 24X-CC; it ranges up to Sample 14H-CC, indicating invasion of cooler waters into the study region from a southern higher latitude, which agrees with diatom observations (see below).
Diatom counts and identification were carried out on smear slides and on acid-treated, sieved (63 µm) material from core-catcher samples from Hole 1081A. Additional samples from within the cores were examined close to datum events to improve the stratigraphic resolution and to refine floral changes (Table 6). Diatom preservation is moderate throughout Hole 1081A. The record of diatom abundance points to a substantial increase in deposition during the late Pliocene and early Pleistocene (from ~227 to 50 mbsf), reaching a maximum in the late Pliocene, followed by a decrease within the Pleistocene at about 1 Ma (~77 mbsf; see Table 6; Fig. 21). Overall abundance levels remain low (trace–frequent) in the late Miocene and early Pliocene. This pattern resembles that of DSDP Site 532 (see summary in Hay and Brock, 1992; Berger and Wefer, 1996). We may assume that the diatom content at Hole 1081A reflects a varying nutrient supply that could be related to the upwelling of nutrient-rich deeper waters and high biological productivity over the Walvis Ridge, especially in the late Pliocene.
The diatom biostratigraphic zones used for this hole are those summarized in Barron (1985). Zones could not be applied to Samples 175-1081A-26X-CC through the end of the hole because of the scarcity of diatom valves and concomitant lack of biostratigraphic markers (Table 6). The LO (1.55 Ma) of Rhizosolenia praebergonii var. robusta was recognized between Samples 175-1081A-11H-CC and 12H-1, 90 cm; the LO (0.8 Ma) of the silicoflagellate Bachmanno-cena quadrangula was recognized between Samples 175-1081A-5H-CC and 6H-CC (Fig. 20).
The diatom assemblage consists of a mixture of upwelling-indicator (Chaetoceros resting spores and Thalassionema nitzschioides var. nitzschioides) and oceanic species (e.g., Azpeitia spp., Hemidiscus cuneiformis, and Thalassiothrix spp.). Within Subunit IB, upwelling-related species dominate the diatom assemblage during highest abundance times in the late Pliocene (see "Lithostratigraphy" section, this chapter). They are not common in early Pliocene nor Miocene sediments when oceanic species tend to dominate (Table 6). In addition, the record of two middle- to high-latitude species, Proboscia (=Simon-seniella) curvirostris in Sample 175-1081A-7H-1, 70 cm (at approximately the Brunhes/Matuyama boundary), and P. barboi in Samples 175-1081A-13H-1, 70 cm, to 23X-5, 85 cm (~1.8–2.8 Ma), is interesting to note. P. barboi is a known ancestor of the Pleistocene cold-water diatom P. curvirostris, and it appears to have had similar ecological preference (Fenner, 1991; Barron, 1995). The occurrence of these species may indicate periods of intensified Benguela Current transport, an assumption that is also supported by the co-occurrence of the Antarctic radiolarian C. pliocenica (see above).