Pujos (1992) suggested four genera and species as the most useful nannofossils for interpretation of Pliocene-Pleistocene paleoecology: Coccolithus pelagicus, Helicosphaera spp., Rhabdosphaera spp., and Syracosphaera spp. Because they constitute only a small portion of the upper photic-zone flora, stratigraphic variations within them are not easily identified (Table 2, PDF format only, oversize material, this volume). To re-examine their usefulness for paleoceanography, the other dominant components of the flora should be excluded from counting, but such a study is not within the scope of this investigation. Instead, this study examines the usefulness of more abundant components of the flora for paleoceanographic interpretations.
Modern, very small placoliths thrive in tropical upwelling settings (Okada and Honjo, 1973), and they are very abundant components of hemipelagic Quaternary nannoflora (e.g., Biekart, 1989; Okada and Wells, 1997). On the other hand, Molfino and McIntyre (1990a, 1990b) demonstrated that the abundance of Florisphaera profunda, a lower photic-zone dweller, declined when the nutricline rose as a result of intensified trade winds. A recent study confirmed the usefulness of F. profunda abundance as an indicator of primary production (Beaufort et al., 1997). These data suggest a new hypothesis: the relative abundance of very small placoliths and F. profunda move in opposite directions when the nutrient profile changes in the euphotic zone.
Actually, in Hole 994C, the abundance of F. profunda expressed as the ratio against the total upper-photic taxa (Fig. 7B) and the abundance of very small placoliths within the upper photic-zone flora (Fig. 7C) show generally reversed stratigraphic trends except in the upper Pleistocene. The stratigraphic trend for the small placoliths, meanwhile, shows no clear relationship with that of the F. profunda ratio (Fig. 7). These observations demonstrated that an increased abundance of the very small placoliths can be interpreted as an indication of increased nutrient supply into the upper-photic water. Because the location of Site 994 is at the periphery of the subtropical central gyre, such a condition could have occurred when the Gulf Stream shifted its course eastward and brought an increased influence of the gyre margin water, or possibly of the Gulf Stream itself, to the site.
Quantitative study of nannofossils in Pacific surface sediments show that the modern central gyre flora is characterized by F. profunda abundance higher than 50% of the total flora (Tanaka, 1991), equivalent to F. profunda ratio of 1.0. A preliminary study at Deep Sea Drilling Project (DSDP) Site 445 (Philippine Sea) indicated that an F. profunda ratio of >1.0 has prevailed since ~2.5 Ma (Okada, 1983). In Hole 994C, the F. profunda ratio remains below 1.0 for the entire period, and mostly lies between 0.8 and 0.2 (percentage abundance of 45% and 17%, respectively; Fig. 7). These data indicate that the waters overlying Site 994 have fluctuated between oligotrophic gyre waters whose stratification is strong and mesotrophic gyre margin waters in which moderate upwelling occurs. The change in the F. profunda ratio indicates that the site was mostly under the central gyre influence between 4.6 and 2.3 Ma, intermittently during 2.1 and 1.3 Ma, and continuously for the last several tens of thousand years.
Very small placoliths shows four distinctive intervals of increased abundance at, 6.0-4.6 Ma, 2.3-2.1 Ma, 2.0-1.8 Ma, and 1.4-0.9 Ma. Moderately strong upwelling is suspected for these time periods (Fig. 7). The small fluctuations observable in the lower upper and middle upper Pleistocene may indicate intermittent weak upwelling. The short time interval at around 1.15 Ma when the abundance peaked at 55% may be a time of particularly strong upwelling (Table 2, PDF format only, oversize material, this volume). This short interval coincides with the temporary disappearance of medium-sized Gephyrocapsa that occurred as a result of phylogenetic evolution (Matsuoka and Okada, 1990). The high abundance of the very small placoliths, therefore, may partly be the result of the phylogenetic evolution of Gephyrocapsa.
Nannofossil data indicate an almost continuous upwelling in the latest Miocene to earliest Pliocene interval (6.0-4.6 Ma; Fig. 7). This coincides with the high productivity interval suggested by diatom flora at Site 997 (Ikeda et al., Chap. 35, this volume). The early Pleistocene interval (1.4-0.9 Ma), in which a fairly strong upwelling is suggested by the abundance of very small placoliths, also agrees with abundance of diatom.
The climatic interpretation inferred from the study of the Quaternary and latest Pliocene planktonic foraminifers in Hole 997A (Nishi, et al., Chap. 34, this volume) also supports the paleoceanographic interpretation of the nannofossil assemblage data. The foraminifer data indicate an oscillation of cool and warm periods of short durations for the latest Pliocene (2.2-1.7 Ma), and a stable, cool period for the 1.2- to 0.95-Ma time interval when the very small placoliths registered sharp abundance increases (Fig. 7). The cool foraminifer assemblage resulted from increased upwelling associated with the gyre margin waters.
The stratigraphic trend in species diversity expressed as the Shannon-Wiener Function (DH) shows no meaningful relationship with that of the F. profunda ratio (Fig. 7). Therefore, this function is not a useful paleoceanographic proxy in the Blake Ridge area. This result suggests, however, that the abundance of F. profunda has no significant relationship with the assemblage diversity of the upper photic-zone nannoplankton. The generally reversed trend observable between the function and the relative abundance of very small to small placoliths can be easily explained by the fact that: (1) the function is strongly controlled by the relative abundance of dominant taxa, and (2) the upper photic-zone flora is dominated in most intervals by the very small and small placoliths.