(Table T1) Clay-mineral composition of ODP Site 184-1146 sediments

The preliminary results of a comprehensive study of clay minerals (<2 µm) are presented for the upper 2 m.y. of sediments from Ocean Drilling Program Leg 184 Site 1146 from the northern margin of the South China Sea, close to the Pearl River mouth. More than 500 samples were analyzed, and four main mineral species are present: smectite, illite, chlorite, and kaolinite. On a general basis, illite + chlorite and smectite display anti-correlated behavior in relation to changes in the proportion of primary to secondary minerals in the sediment. Low-frequency and high-frequency changes are observed in the smectite/(illite + chlorite) ratio.

Stable isotopes measured on Miocene benthic foraminifera from ODP sites 202-1237 and 184-1146

One of the most enigmatic features of Cenozoic long-term climate evolution is the long-lasting positive carbon-isotope excursion or "Monterey Excursion", which started during a period of global warmth after 16.9 Ma and ended at not, vert, similar 13.5 Ma, approximately 400 kyr after major expansion of the Antarctic ice-sheet. We present high-resolution (1-9 kyr) astronomically-tuned climate proxy records in two complete sedimentary successions from the northwestern and southeastern Pacific (ODP Sites 1146 and 1237), which shed new light on the middle Miocene carbon-isotope excursion and associated climatic transition over the interval 17.1-12.7 Ma. We recognize three distinct climate phases with different imprints of orbital variations into the climatic signals (1146 and 1237 d18O, d13C; 1237 XRF Fe, fraction > 63 µm): (1) climate optimum prior to 14.7 Ma characterized by minimum ice volume and prominent 100 and 400 kyr variability, (2) long-term cooling from 14.7 to 13.9 Ma, principally driven by obliquity and culminating with rapid cryosphere expansion and global cooling at the onset of the last and most pronounced d13C increase, (3) "Icehouse" mode after 13.9 Ma with distinct 100 kyr variability and improved ventilation of the deep Pacific. The "Monterey" carbon-isotope excursion (16.9-13.5 Ma) consists overall of nine 400 kyr cycles, which show high coherence with the long eccentricity period. Superposed on these low-frequency oscillations are high-frequency variations (100 kyr), which closely track the amplitude modulation of the short eccentricity period. In contrast to d13C, the d18O signal additionally shows significant power in the 41 kyr band, and the 1.2 Myr amplitude modulation of the obliquity cycle is clearly imprinted in the 1146 d18O signal. Our results suggest that eccentricity was a prime pacemaker of middle Miocene climate evolution through the modulation of long-term carbon budgets and that obliquity-paced changes in high-latitude seasonality favored the transition into the "Icehouse" climate.

Planktonic foraminiferal stratigraphy, datum levels and ages of ODP Sites 184-1146 and 184-1148, South China Sea

Over 30 first and last occurrence (FO and LO, respectively) planktonic foraminifer datums were recognized from the Oligocene-Miocene section of Ocean Drilling Program (ODP) Site 1148. Most datum levels occur in similar order as, and are by correlation as probably synchronous with, their open-ocean records. Several datum levels represent local bioevents resulting from dissolution and Site 1148's unique paleoceanographic setting in the northern South China Sea. An age of 9.5-9.8 Ma is estimated for the local LO of Globoquadrina dehiscens (257 meters composite depth [mcd]), whereas the local LO of Globorotalia fohsi s.l. (301 mcd) is projected to be at ~13.0 Ma and the local FO of Globigerinatella insueta (367 mcd) is projected to be at ~18.0 Ma. The combined planktonic foraminifer and nannofossil results indicate that the Oligocene-Miocene section at Site 1148 is not complete. Unconformities up to 2-3 m.y. in duration, occurring at and before the Oligocene/Miocene boundary (OHS1, OHS2, OHS3, and OHS4 = MHS1), are associated with slump deposits between 457 and 495 mcd that signal tectonic instability during the transition from rifting to spreading in the South China Sea. Shorter unconformities of <0.5 m.y. duration that truncate the Miocene section were more likely to have been caused by sea-bottom erosion as well as dissolution. A total of 12 Miocene unconformities, MHS1 through MHS12, are mainly affected by dissolution and an elevated carbonate compensation depth (CCD) during Miocene third-order glaciations recorded in deep-sea positive oxygen isotope Mi glaciation events. Respectively, they fall at ~457 mcd (MHS1 = Mi-1), 407 mcd (MHS2 = Mi-1a), 385 mcd (MHS3 = Mi-1aa), 366 mcd (MHS4 = Mi-1b), 358 mcd (MHS5 = MLi-1), 333 mcd (MHS6 = Mi-2), 318 mcd (MHS7 = MSi-1), 308 mcd (MHS8 = Mi-3), 295 mcd (MHS9 = Mi-4), 288 mcd (MHS10 = Mi-5), 256 mcd (MHS11 = Mi-6), and 250 mcd (MHS12 = Mi-7). The correlation of these unconformities with Mi events indicates that some related driving mechanisms have been operating, causing deepwater circulation changes concomitantly in world oceans and in the marginal South China Sea.

Age model of ODP Site 184-1146 (Table 1)

Clay mineral assemblages at ODP Site 1146 in the northern South China Sea are used to investigate sediment source and transport processes and to evaluate the evolution of the East Asian monsoon over the past 2 Myr. Clay minerals consist mainly of illite (22-43%) and smectite (12-48%), with associated chlorite (10-30%), kaolinite (2-18%), and random mixed-layer clays (5-22%). Hydrodynamic and mineralogical studies indicate that illite and chlorite sources include Taiwan and the Yangtze River, that smectite and mixed-layer clays originate predominantly from Luzon and Indonesia, and that kaolinite is primarily derived from the Pearl River. Mineral assemblages indicate strong glacial-interglacial cyclicity, with high illite, chlorite, and kaolinite content during glacials and high smectite and mixed-layer clay content during interglacials. During interglacials, summer enhanced monsoon (southwesterly) currents transport more smectite and mixed-layer clays to Site 1146 whereas during glacials, enhanced winter monsoon (northerly) currents transport more illite and chlorite from Taiwan and the Yangtze River. The ratio (smectite+mixed layers)/(illite+chlorite) was adopted as a proxy for East Asian monsoon variability. Higher ratios indicate strengthened summer-monsoon winds and weakened winter-monsoon winds during interglacials. In contrast, lower ratios indicate a strongly intensified winter monsoon and weakened summer monsoon during glacials. Spectral analysis indicates the mineral ratio was dominantly forced by monsoon variability prior to the development of large-scale glaciation at 1.2 Myr and by both monsoon variability and the effects of changing sea level in the interval 1.2 Myr to present.

(Table 2) Age model of ODP Site 184-1146

Orbital tuning of benthic d18O is a common approach for assigning ages to ocean sediment records. Similar environmental forcing of the northern South China Sea and the southeast Asian cave regions allows for transfer of the speleothem d18O radiometric chronology to the planktonic and benthic d18O records from Ocean Drilling Program Site 1146, yielding a new chronology with 41 radiometrically calibrated datums, spanning the past 350 kyr. This approach also provides for an independent assessment of the accuracy of the orbitally tuned benthic d18O chronology for the last 350 kyr. The largest differences relative to the latest chronology occur in marine isotope stages (MIS) 5.4, 5.5, 6, 7, and 9.3. Prominent suborbital-scale structure believed to be global in nature is identified within MIS 5.4 and MIS 7.2. On the basis of the radiometrically calibrated chronology, the time constant of the ice sheet is found to be 5.4 kyr at the precession band (light d18O lags precession minima by -55.4°) and 10.4 kyr at the obliquity band (light d18O lags obliquity maxima by 57.4°). These values are significantly shorter than the single 17 kyr time constant originally estimated by Imbrie et al. (1984), based primarily on the timing of terminations I and II and the 15 kyr time constant used by Lisiecki and Raymo (2005, doi:10.1029/2004PA001071).