PALEOCEANOGRAPHY OF THE SOUTH CHINA SEA

Despite the importance and interconnections of the East Asian and Indian monsoons, few marine-based studies have compared the past monsoonal variations of the two subsystems. Previous ODP studies have focused on the Arabian Sea monsoon (Leg 117: Prell, Niitsuma, et al., 1991; Prell et al., 1992, and references within) and on the Sulu Sea (Leg 124: Rangin, Silver, von Breymann, et al., 1990) to the south of the South China Sea. Many of the East Asian and South China Sea paleomonsoon studies have used traditional and long piston cores to focus on the late Quaternary climate changes. During the last glacial maximum, a lower sea level greatly altered the configuration and area of the western Pacific marginal seas. The three major shelf areas that emerged during the LGM (East China Sea Shelf, Sunda Shelf [the Great Asian Bank], and Sahul Shelf [the Great Australian Bank]) exposed 3,900,000 km² ("maritime continent"), which is comparable in size to the Indian subcontinent. The SCS lost half of its surface area (>52%) as a result of shelf exposure, which changed its configuration into a semi-isolated basin (Wang et al., 1997). Moreover, the most extensive shelf area of the SCS is located in the modern Western Pacific Warm Pool bounded by the 28influenced the thermodynamic role played by the Global Warm Pool.

The central portion of the SCS experienced a considerable decline in the sea-surface temperature (SST) during the LGM. Large decreases in the winter SST in the western Pacific marginal seas and especially in the SCS (Wang and Wang, 1990; Miao et al., 1994; Wang et al., 1995; Chen and Huang, 1998; Chen et al., 1999; Pflaumann and Jian, 1999) are interpreted to indicate that the winter monsoon strengthened, the polar front shifted southward, and the Kuroshio Current migrated eastward. Together with the negligible changes in the summer SST, the South China Sea experienced a much higher seasonality SST during the LGM (Wang et al., 1999). An important consequence of the glacial conditions in the SCS region is the intensified aridity in China. The summer monsoon is the main source of water vapor for rainfall in East China (Chen et al., 1991), and changes in shelf emergence, SST decline, and land-sea heating patterns must have led to a reduction of vapor transport to southern Asia. A rough calculation suggests that the reduction in evaporation from the SCS during the LGM could correspond to one-eighth to one-fourth of the annual precipitation in all of China (Wang et al., 1997). The glacial reduction in water vapor transport helps to explain the intensification of aridity in the China hinterland as evidenced by the extensive distribution of loess deposits. Moreover, the glacial increase of seasonality in the marginal seas may help resolve the tropical paleoclimate enigma in the Pacific; that is, the discrepancy between marine and terrestrial indicators of paleotemperature during the LGM (Stuijts et al., 1988; Anderson and Webb, 1994).

Studies of the late Quaternary have demonstrated the great potential of the SCS's hemipelagic sediments to provide high-resolution paleoenvironment records. A core from the northern SCS (SONNE95-17940, 20&018607´N, 117&018623´E, 1727 m) reveals a highly detailed transition from glacial to Holocene conditions (Fig. F7; Wang et al., 1999). The LGM and isotope Stage 3 are characterized by low fluvial clay content (50%-60%) and high modal grain size (10-25 marked by high clay content (>70%) and low modal grain size (<6.3 and weak summer monsoon precipitation during the glacial regime and a strong summer monsoon and weakened winter monsoon during the Holocene regime. However, with lowered sea level during glacials, a large subareal sediment source is exposed on the shelf of the East China Sea. Deflation and transport of these sediments to the South China Sea during glacials is another possible explanation for the coarser particle sizes. The ¹⁸O data from the mixed-layer planktonic foraminifer Globigerinoides ruber reveal numerous short-term light ¹⁸O events superimposed on the main pattern of glacial-postglacial change (Fig. F7). These events appear to reflect increases in summer monsoon intensity (i.e., reduced sea-surface salinity together with increased input of fluvial clay and decreased modal grain size). The increases in summer monsoon intensity can be correlated with Dansgaard-Oeschger Events 1-10 in the GISP2 ice core (Fig. F7). Also observed in this SCS core are four periods of relatively heavy ¹⁸O associated with low fluvial clay content and larger grain size (i.e., reduced summer monsoon rainfall and increased winter monsoon wind, which correlate with the Heinrich Events 1-4 (Fig. F7). The early Holocene/Preboreal summer monsoon maximum revealed by a broad ¹⁸O minimum and fluvial clay maximum has also been reported from the Arabian Sea (Prell, 1984b; Sirocko et al., 1993). The 8.2-ka cooling event recorded in the GISP2 ice core appears to coincide with a large increase in ¹⁸O and, hence, a decrease in summer monsoon precipitation in the SCS. Similar rapid events in the Bay of Bengal and Andaman Sea have been related to North Atlantic climate change (Colin et al., 1998). The Leg 184 cores, along with the recent cores from the joint German-Chinese Monitor Monsoon expedition (Sarnthein et al., 1994; Sarnthein and Wang, 1999), and the 1997 IMAGES III cruise for the first time provide systematic and high-quality material for studying the long-term evolution and variability of the monsoonal South China Sea.