Figure F8 illustrates planktonic foraminifer and nannoplankton datum levels, as well as the calculated sedimentation rates and unconformities found at Site 1148. Very high sedimentation rates of >100 m/m.y. account for the accumulation during rifting of the exceptionally thick lower Oligocene section, whereas the Miocene sediments were deposited during relatively stable seafloor spreading periods at moderate rates of 15–30 m/m.y. The 457- to 495-mcd interval, including the slump, marks the transition from rifting to stable spreading of the SCS (Briais et al., 1993; Wang, Prell, Blum, et al., 2000; Li et al., 2003). It is not clear, however, how this major tectonic event in the regional West Pacific affected paleoceanography of other regions where a more complete sediment record exists (e.g., Pacific: Kroenke et al., 1993; Atlantic: Flower et al., 1997; Mutti, 2000; Subantarctic: Billups et al., 2002).
At least four major unconformities are present in the upper Oligocene (Fig. F6), respectively, at ~488 mcd (OHS1), ~478 mcd (OHS2), 472 mcd (OHS3), and 460 mcd (OHS4). OHS1 at 488 mcd coincides with the lower/upper Oligocene boundary and represents a break of at least 0.5 m.y., based on the LO of C. cubensis (28.5 Ma) at 487.77 mcd and the LO of Sphenolithus distentus (27.7 Ma) at 485.34 mcd. It is also marked by the most pronounced changes in all physical property records (Wang, Prell, Blum, et al., 2000) (Fig. F9). OHS2 is indicated by the sudden disappearance of P. opima at ~478 mcd (27.1 Ma) and by the FO of P. pseudokugleri (25.9 Ma) from only ~3 m above this level, suggesting a hiatus of ~1 m.y. (between 26.0 and 27.0 Ma). OHS3 and OHS4 lie within the slump between 457 and 472 mcd, although the whole interval can be collectively considered as representing a single unconformity if the slump deposition had occurred in a very short period of time rather than during the entire period of ~23.4–25.4 Ma. Different lithofacies and biofacies distinguish two slumped packages, between 458 and 460 mcd and between 460 and 472 mcd, that correspond to two slumping epochs, each footing on an unconformity surface. Although no microfossil events were observed directly at the 472-mcd level (OHS3), the LO of Sphenolithus ciperoensis (25.5 Ma) at 474 mcd and the LO of Zygrhablithus bijugatus (23.8–24.5 Ma) at 470 mcd suggest a gap at the base of slump would be at least 1 m.y. in duration, between ~24.5 and 25.5 Ma. This is because the 4- to 5-m slumped section between these two datums cannot fully represent a deposition of 1 m.y., even if an error bar of 0.5–1.0 m for the datums was considered. Sediments close to 460 mcd underwent a change in the slumped lithofacies and biofacies, and the presence of two bioevents, the FO of P. kugleri (23.8 Ma) and the LO of Reticulofenestra bisectus (23.9 Ma), indicates a horizon close to the Oligocene/Miocene boundary coinciding with OHS4. Because the average sedimentation rate during the early Miocene at Site 1148 was ~15 m/m.y. (Fig. F7), we may assume that a rate at least twice as fast as 15 m/m.y could account for the gravity flow deposition and the lower slumped package between 460 and 478 mcd could represent a <0.5-m.y. deposition between 24.0 and 24.5 Ma. Consequently, the unconformity across the Oligocene/Miocene boundary would be ~0.5 m.y. in duration between 23.5 and 24.0 Ma and the upper slumped package between 458 and 460 would likely accumulate in the first 0.1 m.y. after the boundary unconformity (i.e., at 23.4 Ma). Because the OHS4 event occurred across the Oligocene/Miocene boundary, it is also assigned to represent the first Miocene unconformity event, MHS1.
Although there are no apparent post-Oligocene slumps, 11 surfaces collectively called unconformities MHS2–MHS12 within the relatively complete Miocene section are associated with either some bioevents clustering together and causing shorter and incomplete biozones or with heavy dissolution (Figs. F7, F8). The FO of Globigerinoides altiapertura (20.5 Ma) at 408 mcd and the LO of P. kugleri (21.5 Ma) at 411 mcd indicate a Miocene unconformity MHS2 of at least 0.5 m.y., which includes a dissolution event at ~407 mcd. MHS3–MHS12 also match the dissolution events at 385, 366, 358, 333, 318, 308, 295, 288, 256, and 250 mcd, respectively (Fig. F8). Among them, MHS4 and MHS9 erased part of the sediment record for Zones N6 and N10 (Fig. F7), whereas others mark sudden changes in various physical property signals that were used for subdividing lithostratigraphic units (Fig. F9). Except for those defined by planktonic foraminifer and nannofossil datums, many of these Miocene dissolution events found at Site 1148 and collectively called unconformities may each represent a missing interval of <0.3 m.y.
The altered planktonic foraminifer assemblage found at Site 1148 was affected by dissolution as a result of lysocline and CCD fluctuations, especially during the Miocene. Few complete tests are preserved in samples affected by strong dissolution, and this pattern occurs repeatedly at certain levels, representing heavy dissolution cycles (Figs. F8, F9). This phenomenon is widespread in the tropical western Pacific where the lysocline is elevated (Kennett et al., 1985; Kroenke et al., 1993).
Major dissolution events occurred in the later part of the middle Miocene (~308 and ~288 mcd), apparently corresponding to the "carbonate crash" widely recorded in tropical oceanic settings (Roth et al., 2000). This has been attributed to the formation or strengthening of the Atlantic Deep Water and Antarctic Deep Water, which caused an elevated CCD and strong deep-sea dissolution worldwide (Woodruff and Savin, 1991; Hodell and Woodruff, 1994; Lyle et al., 1995). In addition to these events, at Site 1148 there is a series of smaller-scale dissolution events at ~457, ~407, ~385, ~366, ~358, ~333, ~318, ~295, ~288, ~256, and ~250 mcd, respectively (Fig. F8). On the basis of correlation, these dissolution events appear to fall at or close to the positive oxygen isotope Mi events of Miller et al. (1991, 1996, 1998) or the MLi–MSi events of Hardenbol et al. (1998) that signal Miocene glaciations. Specifically, the Miocene dissolution events identified at Site 1148 correspond at least in age to isotopic glaciations Mi-1 (457 mcd), Mi-1a (407 mcd), Mi-1aa (385 mcd), Mi-1b (366 mcd), MLi1 (358 mcd), Mi-2 (333 mcd), MSi1 (318 mcd), Mi-3 (308 mcd), Mi-4 (295 mcd), Mi-5 (288 mcd), Mi-6 (256 mcd), and Mi-7 (250 mcd) (Fig. F9). This correlation also implies that all heavy dissolution observed at Site 1148 was related to a corrosive deep water that was strengthened during major glaciation periods.
A positive relationship between the Miocene dissolution events and Mi glaciations indicates a link to deep-sea watermass changes in the SCS (Fig. F10). Some appear to coincide with deep-sea hiatuses recognized elsewhere by Keller and Baron (1983, 1987) and Keller et al. (1987), although their dissolution origin is still debated (Haq, 1991; Kroenke et al., 1993; Spencer-Cervato, 1998). Apart from those associated with the upper Oligocene slump, most Miocene unconformities found at Site 1148 have been more likely related to dissolution than to local tectonics, although some of them (e.g., MHS4 and MHS8) (Fig. F10) could have been affected also from sediment removal by underwater currents.