Toward establishing a maritime proxy record of
the East Asian summer monsoons for the late Quaternary
Kuo-Yen Wei
a;, Tzu-Chien Chiu
a;b, Yue-Gau Chen
aa Institute of Geosciences, National Taiwan University, P.O.Box 13-318, Taipei, Taiwan b Lamont^Doherty Earth Observatory, Palisades, NY 10964, USA
Accepted 19 June 2003
Abstract
An astronomically tuned late Quaternary planktic foraminiferal N18O record of Site MD972142 (12‡41.33PN,
119‡27.90PE; 1557 m water depth) in the southeastern South China Sea was established. The difference in N18O
between MD972142 and ODP792 of the Sulu Sea is regarded as a maritime proxy of the summer monsoon intensity over the South China Sea and Southeast Asia. The profile of this maritime proxy matches well with the summer monsoon index obtained from the terrestrial record of Louchuan, central Chinese Loess Plateau. The amplified planktic N18O signals of the South China Sea relative to the Sulu Sea record are partly caused by the changing
intensities of the East Asian Monsoons at the glacial^interglacial time-scale throughout the late Quaternary. ; 2003 Elsevier B.V. All rights reserved.
1. Introduction
In studying the Asian summer monsoon as a major component of the global climate system, various indices have been developed to quantify its intensity for modern monsoon climatology, such as monsoon rainfall over India, the vertical shear in the zonal wind between 850 and 200 hPa over the monsoon region (the M1* index of Web-ster and Yang, 1992). In paleoclimatological stud-ies, researchers have also established various proxies to gauge the intensity of paleomonsoons. For instance, the magnetic susceptibility in the paleosol^loess sequences of the Chinese Loess
Plateau has been used as a good proxy of summer monsoon (An et al., 1990, 1991; Verosub et al., 1993 ; Porter et al., 2001), whereas the loess grain size has been considered as an index for the winter monsoon (Ding et al., 1995). For paleoceano-graphic records, the relative abundance of Globi-gerina bulloides, a planktic foraminifer a⁄liated with upwelling induced by the southwesterly summer monsoon in the NW Indian Ocean, has demonstrated to be a good proxy for the marine records of the Indian Ocean (Prell et al., 1990). Nevertheless, establishing a robust proxy of the East Asian summer paleomonsoon intensity from sedimentary records in the South China Sea (SCS) has been posited as a major challenge. Several proxies have been proposed. For in-stance, the deviation of the N18O signal of the
mixed-layer planktic foraminifer Globigerinoides ruber from the main pattern of glacial^interglacial
* Corresponding author. Tel. : +886-2-23691143; Fax: +886-2-23636095.
E-mail address:[email protected](K.-Y. Wei).
Marine Geology 201 (2003) 67^79 R
Available online at www.sciencedirect.com
£uctuation was considered as an index of summer monsoon intensity, as inferred from its covariance with reduced sea-surface salinity and increased input of £uvial clay (P. Wang et al., 1999). In-creases in herbaceous pollen and charcoal from the same marine sedimentary sequence of Core 17940 in the northern SCS during the glacial also has been regarded indicating aridity and hence a sign of weakened summer monsoon (Sun and Li, 1999). More recently, Jian et al. (2001) strategically selected two sites from the summer and winter upwelling areas in the SCS and used four upwelling proxies (benthic forami-niferal £ux, organic carbon £ux, relative abun-dance of Uvigerina peregrina, estimated depth of thermocline) to infer indirectly the strength of summer and winter monsoons.
In this paper, we propose that the di¡erence in N18O of the planktic foraminifer Globigerinoides ruber between the SCS and the Sulu Sea can serve as a proxy of the intensity of summer monsoon over Southeast Asia for the late Quaternary. Such deviation in N18O of the SCS from the Sulu Sea
record is considered to be related to the intensity of precipitation which, in turn, was mainly en-hanced by summer monsoons in the SCS and sur-rounding lands. We will ¢rst present a newly ob-tained astronomically tuned timescale of planktic N18O for the past 870 kyr from Site MD972142, eastern SCS (Fig. 1), and then compare it with a comparable record of ODP Site 769 from the Sulu Sea (Linsley and Dunbar, 1994 ; Linsley, 1996). In interpreting the di¡erence between these two rec-ords, we will present our rationale in using it as a proxy of the summer monsoon intensity over the vast area of the Southeast Asia for the late Qua-ternary. Temporal variations of this summer mon-soon index are then compared with the classical magnetic susceptibility record at Louchuan in the central Chinese Loess Plateau (Lu et al., 1999 ; Heslop et al., 2000). The striking similarity in these two proximal records of summer monsoons appears to support our interpretation.
2. Site MD972142 and its oceanographic setting During the 1997 IMAGES-III-IPHIS Cruise
(Chen and Beaufort, 1998), Core MD972142 (12‡41.33PN, 119‡27.90PE) (Fig. 1) was raised from the continental slope o¡ Palawan Island at a water depth of 1557 m. The sea£oor at this site is much shallower than the calcite lysocline (V2500 m ; Rottman, 1979) of the SCS so that calcareous foraminifera and nannofossils are well preserved. Because of the moderately low sedimentation rates at this site and long core re-covery in length (35.91 m), a record with su⁄-ciently high resolution, covering the whole Brunhes magnetochron is preserved in this core (Lee, 2000).
The transportation of salt in the basin was mainly driven by the in£ow of Western Philippine Sea (WPS) waters entering from the Luzon Strait (between Taiwan and Luzon), and meanwhile the waters in the basin are diluted by river runo¡ from the surrounding lands (Wyrtki, 1961). Sum-mer sea-surface salinities range between 32.8 and 33.6x, but the north^south gradient in salinity in the SCS increases during the winter with a branch of warm, saline Kuroshio enters into the SCS through the Luzon Strait (Shaw, 1989, 1991), mixing with colder waters from the north via the Taiwan Strait. These two surface water masses combine and £ow southward along the coast of Indochina Peninsula and result in a counterclock-wise circulation. The salinities during winter in-crease generally, with a wide range from 33.2 to 34.4x (Levitus and Boyer, 1994). Changes in the freshwater budget have had a great impact on the circulation. According toWytrki (1961), the saline waters ( s 34.9x) which enter the SCS as a sub-surface £ow are mainly from the North Subtrop-ical Lower Water while the less saline waters (34.2V34.3x) in the upper part are from the North Paci¢c Intermediate Water. The di¡erence between the Smax and Smin in the vertical pro¢le
decreases as the waters £ow southward into the basin due to vertical mixing. On the other hand, the salinity of the surface water shows also a southward decreasing trend. The gradient along the north^south transect is controlled partly by the dilution e¡ect of freshwater input from the Indochina Peninsula. The SCS water £ows north-wards out through the Luzon Strait along the coast of Luzon and enters to the WPS (Wyrtki,
1961 ; Nitani, 1972 ; Gong et al., 1993 ; Chen and Huang, 1996). The SCS water £ows out mainly at the intermediate depth of 350^1350 m, and joins the Kuroshio, entrained as the left-hand part of the Kuroshio £owing northward (Chen and Huang, 1996). The salinity range of the out£ow water is between 34.45 and 34.6x (Chen and Huang, 1996). It is obvious that precipitation and thus freshwater in£ux to the basin plays an important role in regulating the salt budget of the SCS.
For the late Quaternary, another major factor that a¡ects paleoceanographic conditions is the large-amplitude sea-level £uctuation associated with global ice-volume changes. There is a large extent of continental shelves distributed in the western Paci¢c marginal seas, and therefore, gla-cial-induced sea-level drop shrank the area of marginal seas signi¢cantly (P. Wang, 1999). In the SCS for example, during the glacial periods,
large areas of the continental shelves (such as the Sunda Shelf) were exposed, closing the gateways between the southwestern SCS and Indian Ocean. Shrinkage of basin area and enhanced northeast-ern monsoon would both tend to lower the mois-ture £ux conveyed inland, consequently aridity in mainland China became enhanced during glacial periods (P. Wang, 1999).
3. Materials and methods
The sedimentary sequence at Site MD972142 consists dominantly of ¢ne-grained hemipelagic sediments intercalated with more than a dozen discrete tephra layers. The thickness of the indi-vidual tephra layers ranges from 0.5 to 10 cm ; they are on average thicker than 1 cm (Wei et al., 1998b).
For stable oxygen and carbon isotope analyses,
Fig. 1. Location map of East Asia and Paci¢c showing locations of the three sites examined : MD972142 in the eastern SCS, ODP769 in the Sulu Sea, and Louchuan on the central Chinese Loess Plateau. The ticked line marks the current northwestern limit of the summer monsoon.
samples were taken every 4^12 cm as to obtain a temporal resolution of at least 1 sample per 2000 years. Specimens of planktic foraminifera Globi-gerinoides ruber s.s (white) in the 250V300-Wm size fraction were picked from sieved sediments for isotopic analysis. Con¢ning the size of forami-niferal specimens in such a narrow range results in having picked specimens in the same ontogenetic stage (Brummer et al., 1987; Wei et al., 1992), thus excluding vital e¡ects in stable isotope sig-nals. About eight specimens of Globigerinoides ruber were immersed in CH3OH and subjected
to ultrasonic vibration for 6 s three times and then immersed in NaOCl for 24 h. After being cleaned by deionized water ¢ve times and dried, the specimens were reacted with 100% H3PO4 at
70‡Cto generate CO2 gas in a Kiel Device ; then
the CO2 gas was sent into the Finnigan MAT
Deltaplus mass spectrometer to determine its
N18O and N13Cvalues. All N18O and N13Cvalues were converted to the Pee Dee Belemnite scale.
The NBS-19 was used as calibration standard (N18O = 32.20x, N13C= +1.95x). External
pre-cision of the measurements was better than 0.08x for N18O and 0.04x for N13C, as shown
by routine repeatedly analysis of the internal lab-oratory standard.
In all 250 specimens of Globigerinoides sacculi-fer ( s 300-Wm size fraction) picked from ¢ve depths in the uppermost part of Core MD972142 were sent to Rafter Radiocarbon Laboratory, In-stitute of Geological and Nuclear Sciences, New Zealand, for 14Caccelerator mass spectrometry
(AMS) dating. The last appearance datum (LA) of pink Globigerinoides ruber at about 120 ka has been considered a good biostratigraphic marker in the western Paci¢c (Thompson et al., 1979). Lee et al. (1999)reported recently that the age of this datum is 127 ka at Site MD972151 in the south-west SCS. The last occurrence of pink G. ruber at Site MD972142 was found at a subdepth of 910 cm.
Fig. 2. Oxygen isotope pro¢le of Globigerinoides ruber, Core MD972142. The depth scale is cm in subdepth below the sea£oor. Numbers mark the marine isotope stages correlated to the low-latitude stack ofBassinot et al. (1994). Shown on the top are the magnetostratigraphy fromLee (2000), the last occurrence datum of pink G. ruber, and the position of the Australasian microtek-tite layer (marked as mtk) (Lee and Wei, 2000).
4. Oxygen isotope stratigraphy
Results of the oxygen stable isotope measure-ments are plotted in Fig. 2. The pro¢le shows a £uctuation pattern comparable to previously pub-lished records of the late Quaternary in the low-latitude Paci¢c (e.g. V28-238 by Shackleton and Opdyke, 1973 and DSDP 805Cby Berger et al., 1993aon Ontong Java Plateau ; ODP 769 in Sulu Sea byLinsley and Dunbar, 1994). The Brunhes/ Matuyama geomagnetic reversal at 33.70 m (Lee, 2000) assists in positioning the marine isotopic stages 19 and 20. The occurrence of Australasian microtektites at the subdepths 33.85^34.45 m further constrains the age of this interval to be V770^790 ka (Izett and Obradovich, 1992 ;
Tauxe et al., 1996 ; Lee and Wei, 2000). These biostratigraphic, magnetostratigraphic and micro-tektite markers were used as independent controls
in assisting the correlation of the resulted pro¢le with the low-latitude stack (Bassinot et al., 1994) (Fig. 3). The depths and corresponding ages of the controlling points are listed in Table 1. The newly calibrated ages of the various independent markers are listed, too.
Compared to the typical glacial^interglacial dif-ference of V1.4x in N18O observed in
open-ocean records of the Ontong Java Plateau (Shackleton and Opdyke, 1973 ; 1976; Berger et al., 1993a,b), the amplitude of £uctuation as large as V2x for the glacial^interglacial cycles in MD972142 is signi¢cantly larger. This ampli¢ed N18O variation observed in this core is considered to be an ensemble e¡ect of global ice-volume var-iation and local temperature and salinity changes. The semi-isolated con¢guration of the SCS Basin during periods of low sea-level stand together with the changing salt budgets (varying intensities
Fig. 3. Oxygen isotope timescale of Globigerinoides ruber, Core MD972142, calibrated against the low-latitude oxygen isotope stack ofBassinot et al. (1994)(shown as the dotted line). The scale for the MD972142 N18O is shown on the left axis while that for the low-latitude stack is on the right axis. Shown on the top are the magnetic polarity fromLee (2000), the last occurrence datum of pink G. ruber, and the interval of Australasian microtektites (marked as mtk) (Lee and Wei, 2000). The ¢nal ¢xpoints for the timescale are listed inTable 1.
of terrestrial freshwater runo¡ and exchange e⁄-cacy with open-ocean saline waters) should have contributed to amplifying the N18O variation.
To enable a comparison of the oxygen isotope time-series of MD972142 with the isotopic pro¢le yielded from the same species from Site ODP769 at 8‡N latitude in the Sulu Sea (Linsley and Dun-bar, 1994; Linsley, 1996), we calibrated the ODP769 record to the timescale of MD972142 by curve matching. The curve matching was con-ducted using the AnalySeries Program, Version 1.2. (Paillard et al., 1996).
5. Maritime proxyof the East Asia summer monsoons
The calibration of the N18O pro¢le of ODP769
against the timescale of MD972142 allows a close comparison of the two time-series (Fig. 4A). Giv-en the fact that both sites (MD 972142 in the SCS and ODP 769 in the Sulu Sea) are located in the northwest margin of the West Paci¢c Warm Pool (WPWP) with similar latitudes, the oxygen
iso-Table 1
Chronological delineators in MD972142
Depth Age Method
(cm) (ka) 1 1 AMS14C 69 5 AMS14C 145 6 201 12 AMS14C 225 14 AMS14C 241 16 297 21 AMS14C 425 34 601 52 MIS 3.1 675 64 747 80 MIS 5.1 813 94 865 106 MIS 5.4 893 122 MIS 5.5 929 128 MIS 6.0 995 136 1034 148 1056 156 1104 166 1140 176 1172 194 MIS 7.1 1347 214 MIS 7.3 1379 224 MIS 7.4 1419 234 1493 250 1589 266 MIS 8.4 1663 288 MIS 8.5 1705 296 MIS 8.6 1778 316 MIS 9.2 1838 328 MIS 9.3 1896 342 1924 348 MIS 10.3 1974 358 2042 370 MIS 11.1 2090 384 MIS 11.23 2126 392 2221 420 2256 434 MIS 12.2 2284 452 2425 482 MIS 13.11 2441 492 MIS 13.12 2457 500 MIS 13.13 2489 510 MIS 13.2 2529 524 MIS 13.3 2589 536 MIS 14.2 2713 574 MIS 15.1 2777 596 2809 604 MIS15.4 2857 616 MIS 15.5 2881 624 2921 630 2929 644 2961 662 Table 1 (Continued).
Depth Age Method
(cm) (ka) 3033 690 3073 702 3121 710 3153 718 MIS 18.2 3193 730 MIS 18.3 3265 754 MIS 18.4 3329 766 MIS 19.1 3385 780 3417 792 MIS 20.2 3449 796 3465 808 3473 820 MIS 21.1 3489 828 MIS 21.2 3505 838 MIS 21.3 3537 848 MIS 21.4 3561 858 MIS 21.5 *910 125 LA of pink Globigerinoides ruber *3370 776 Brunhes/Matuyama boundary *3425 793 Australasian microtektite peak
topic values obtained from the same species of planktic foraminiferal tests are expected to be the same, or at least, similar. Yet, MD972142 tends to show more negative N18O values than
ODP769, especially during the interglacial stages (the odd-numbered stages in Fig. 4). The di¡er-ence in N18O between the two sites decreases
dur-ing the glacial periods but increases durdur-ing
in-terglacial intervals (Fig. 4B). Larger gradients (N18O
MD142^N18OODP769) up to 0.5^1.5x exist
for most parts of the interglacial stages but with small di¡erences of 0^0.5x during the glacial stages (Fig. 4B). This result is somewhat coun-ter-intuitive in that the two basins should be more isolated from each other during glacial peri-ods when sea level dropped, therefore showing
Fig. 4. Comparison of oxygen isotope pro¢le of MD972142 with that of ODP769 of the Sulu Sea. Marine oxygen isotopic stages down to stage 21 are marked. Glacial stages are represented by gray stripes. (A) Time-series of N18O of both sites with a time in-terval of 2k (vt = 2k) of both sites are interpolated from the original data pro¢les and shown here. MD972142 shows lower N18O values during most of the last 800 kyr. (B) The N18O di¡erence between the interpolated time-series of MD972142 and ODP769 calculated by subtracting the values of ODP769 from that of MD972142. The depletion of18O is more distinctive during the in-terglacial periods.
di¡erent values. On the other hand, there should be more water exchange between these two basins during interglacial periods when the sea levels were higher.
Over the late Quaternary, such systematic dis-crepancies must result from a signi¢cant di¡er-ence in temperature or salinity, or a combination of both, of the surface waters of these two basins. Furthermore, the amplitude of the £uctuation in MD972142 time-series also is larger, attesting an ampli¢ed e¡ect of the SCS. At present, the sea-surface temperatures (SSTs) of these two basins are quite similar (Levitus and Boyer, 1994). The di¡erence in temperature during the past could be due to the shifting of the western extent of the WPWP (Martinez et al., 1997). On the other hand, the salinity of the Sulu Sea surface water at present is slightly higher than that in the SCS by 0.5x (Levitus et al., 1994).
Although the foraminifera used by both studies were well preserved, and the inter-laboratory dif-ference in oxygen isotopic measurements was evaluated to be small (pers. commun. Linsley, 1998), we could not absolutely rule out the possi-bility that there exists a systematic di¡erence in N18O values between these two time-series intro-duced by experimental practice. To further check if the observed di¡erence is not merely an artifact, we normalized each time-series to its own average value, and calculated the gradient between these two normalized time-series again (Fig. 5B). The result yields an almost identical pattern with
Fig. 4B.
Such an ever-lasting di¡erence in N18O between
the SCS and Sulu Sea must result from a signi¢-cant di¡erence in hydrographic condition between the two basins during the past 800 kyr. The rela-tively more negative N18O values in MD972142
can only be accounted for by higher temperature or lower salinity, or a combination of both. How-ever, the former factor ^ higher SSTs in the SCS ^ is less likely because the Sulu Sea is located closer to the core of the WPWP and is warmer. The glacial^interglacial SST di¡erence in the core WPWP was about 1^3‡C(CLIMAP, 1981; Oh-kouchi et al., 1994; Lea et al., 2000). In contrast, SSTs in the SCS dropped by as much as 3^6‡C (L. Wang and Wang, 1990; Miao et al., 1994;
Thu-nell et al., 1995; Chen and Huang, 1998; Wei et al., 1998a ; Pelejero et al., 1999 ; L. Wang et al., 1999 ; Kienast et al., 2001) (equivalent to V0.75^ 1.5x increase in N18O signal) from the
intergla-cial to glaintergla-cial intervals. The SSTs seldom ex-ceeded the modern temperature in the past. For instance, the increase in SST (reconstructed by alkenone insaturation index) during the warmest period, substage 5.5 (V125 ka) in the last inter-glacial, was less than 1‡Cthan the Holocene (P. Wang et al., 1999). Therefore, temperature variation (mostly becoming colder) alone in the SCS would likely have caused the N18O values to
be more positive relative to the Sulu Sea, opposite to the observed negative tendency. One possible explanation for the light oxygen isotope composi-tion of MD972142 is that over most part of the past 800 kyr, the N18O values of the surface waters
of the SCS were lower than in the Sulu Sea, es-pecially during the interglacial periods. Today, rivers in Indochina, Borneo and Sumatra serve as major sources of freshwater to the SCS, it is conceivable to assume that during the interglacial periods, enhanced precipitation and increased runo¡ would result in lower N18O values for
sur-face waters particularly for the southern part of the SCS. And meanwhile, the Sulu Sea was more or less immune from the in£uence of the increased freshwater runo¡. In fact, long-term observation of land and ocean precipitation of the studied area (the CAMS^OPI monthly precipitation cli-matology available from http ://ingrid.ldgo.colum-bia.edu/) shows that starting from June to the end of October, when southwesterly monsoons are prevailing, the Indochina Peninsula and the SCS are under heavy precipitation. Particularly, the ‘bull-eyes’ of peak precipitation larger than 400 mm/month are located over the southern Indochi-na Peninsula and the eastern part of the SCS. In comparison, the precipitation over Sulu Sea is less than 300 mm/month.
Modern climate data suggest that the strongest convective precipitation in the SCS and over the Indochina Peninsula is associated with the Inter-Tropical Convergence Zone (ITCZ) (Lau and Li, 1984 ; Ho¡man and Heimann, 1997) ; the position of the ITCZ, in turn, is controlled by the north^ south thermal gradient through the troposphere.
Furthermore, in convectively active regions such as the tropical islands and monsoon prevailing areas, the amount e¡ect on N18O is signi¢cant
(Dansgarrd, 1964). For instance, the N18O values
drop from 0x to 37x when precipitation in-creases from 0 to 350 mm/month (Ho¡man and Heimann, 1997). To demonstrate the link between the negative N18O values and the summer
mon-Fig. 5. Comparison of the terrestrial proxy of summer monsoons at Louchuan, Chinese Loess Plateau (A) with the maritime summer monsoon proxy (B). (A) Magnetic susceptibility of Louchuan Section in the central Chinese Loess Plateau, adopted fromHeslop et al. (2000). Paleosol units (S1^S8) of the Louchuan Section are marked by light gray. (B) Maritime summer mon-soon proxy of the SCS. The proxy is the di¡erence in N18O between the average-normalized time-series of MD972142 and ODP769. Marine oxygen isotope stages down to stage 20 are marked. Magnetostratigraphy shown on top of each panel and the Australasian microtektite (mTK) layers at the three sites provided with independent chronological markers that support the corre-lation.
soons, we examined the precipitation data col-lected by IAEA at Bangkok for Year 1968^ 1998. The data show clearly that precipitation over these 30 years has been concentrated in the summer^early autumn and the N18O values dropped to as low as 312x when the monthly precipitation exceeded 500 mm. The amount e¡ect was estimated as 31.4x per 100 mm/month in-crease in precipitation (Fig. 6).
Based upon the above reasoning, the negative excursion in N18O of Site MD972142 from that of
ODP769 (Fig. 5B) can be regarded as an indica-tion of excessive input of freshwater/precipitaindica-tion to the SCS, and thus an indicator of summer monsoon intensity. The increased excursion (v18O in Fig. 5B) during the interglacial periods
is consistent with most previous inferences about the East Asian monsoon variation in the SCS, that the summer monsoon was stronger during the interglacial periods with a concomitant weak-ening of the winter monsoon (L. Wang and Wang, 1990 ; Huang et al., 1997 ; L. Wang et al., 1999 ; Jian et al., 2001). The variation pattern of this summer monsoon index (Fig. 5B) shows several important features : (1) precipitation (summer monsoon) was strengthened generally during the interglacial stages, and (2) the oxygen
isotopic stage 14 is peculiar in showing heavy pre-cipitation and strong summer monsoon, which is at odds with other glacial stages.
More strikingly, this ‘maritime summer mon-soon’ index agrees very well with the magnetic susceptibility record of Louchuan, central Chinese Loess Plateau (Heslop et al., 2000) (Fig. 5A). The magnetic susceptibility of loess^paleosol sequen-ces has long been regarded a proxy of summer monsoons over the Chinese Loess Plateau (An et al., 1990, 1991; Porter et al., 2001). The paleo-sol units are usually developed during the inter-glacial stages and characterized by high magnetic susceptibility. Here we adopted the astronomi-cally tuned time scale of Heslop et al. (2000) to make a comparison between the maritime and terrestrial proxies of the summer monsoons. The widespread Australasian microtektites caused by an asteroid impact on the Indochina Peninsula slightly before the Matuyama/Brunhes geomag-netic boundary serves as an independent age marker to line up the three records under scru-tiny: the microtektites (symbolized as mTK in
Fig. 5) were found within the oxygen isotope stage 20 in Core MD972142 (Lee and Wei, 2000) and ODP769 (Schnieder et al., 1992 ; Kent and Schneider, 1995), and within the loess unit L8 at the Louchuan section (Fig. 5) (Li et al., 1993, ¢de Zhou and Shackleton, 1999).
Fig. 7. Power spectra of maritime summer monsoon proxy of the South China Sea for the last 800 kyr. The vertical bar represents half of the 80% con¢dence interval whereas the horizontal bar is the bandwidth of the spectral analysis. Note that the time-series is dominated by the periodicity of eccentricity of 100 kyr.
Fig. 6. Regression of N18O of precipitation against the monthly precipitation amount at the Bangkok Station over the period of 1968^1998. The linear relationship suggests the existence of an amount e¡ect of precipitation on the N18O values of precipitation. The amount e¡ect is estimated to be 31.4x per 100 mm precipitation. The data were obtained from the website of the Global Network for Isotopes in Pre-cipitation, managed by the Isotope Hydrology Information System (http://isohis.iaea.org).
Generally speaking, high values of magnetic susceptibility shown by the Louchuan Section cor-respond to negative excursions shown by the mar-itime proxy (Fig. 5). For a long time it has been perplexing that the thickest paleosol unit (S5) spans across a prolonged period of marine oxygen isotope stages 15^13 (MIS 15^13). Here the v18O
shows large negative values over most of MIS 15^ 13, suggesting that the intensi¢ed summer mon-soons might not only have caused high precipita-tion over Southeast Asia and the SCS, but also induced warm and wet conditions over the Chi-nese Loess Plateau in this particular period (Fig. 5). Similarly, the paleosol unit S3 also extends into the ¢rst half of MIS 8 where the summer monsoons were intensi¢ed (Fig. 5). Paleosol unit S1 also ¢nds its counterpart during MIS 5 in the maritime index. The only interval where the mar-itime index does not correspond well is in the second half of MIS 7 where the low v18O gradient
does not match well with the high SI values of the loess. But, on the other hand, the peak SI values in the early part, namely those during 650^850 ka, match quite well with high v18O values in their
counterparts, respectively (Fig. 5). A spectral analysis of the maritime summer monsoon index reveals that the time-series is dominated by the 100-kyr periodicity of eccentricity (Fig. 7). The explanation power of this newly established mar-itime proxy appears to merit it a faithful indicator of the Asian summer monsoon intensity at least at the glacial^interglacial time scale. The MD972142 oxygen isotope pro¢le indicates that the western margin of the WPWP over the southeastern SCS has been highly variable, partially contributed by the changing intensities of summer monsoons at the glacial^interglacial time-scale.
6. Summary
The late Quaternary planktic foraminiferal N18O record obtained from Core MD972142 (12‡41.33PN, 119‡27.90PE; 1557 m water depth) of eastern South China was correlated with the astronomically tuned low-latitude stack of Bassi-not et al. (1994). The N18O values at Site
MD972142 are always lighter than those recorded
at Site ODP 769 of the neighboring Sulu Sea, suggesting the existence of lower salinities in the SCS. The £uctuation amplitudes in N18O between
the glacial and interglacial periods are also much larger than those of the Sulu Sea (ODP 769) by showing more negative values during the intergla-cial periods. We interpret this phenomenon as caused by enhanced precipitation over Indochina, Sumatra and Borneo during interglacial periods. The di¡erence in N18O between the
average-nor-malized N18O time-series of the SCS (MD972142)
and the Sulu Sea (ODP769) is considered to be a maritime proxy of the East Asia summer mon-soons. This maritime proxy shows good correla-tion with the summer monsoon index obtained from the Louchuan Section of central Chinese Loess Plateau. Notably, high values of maritime summer monsoon index during MIS 9^8.5 (340^ 280 ka) and MIS 15^13 (620^480 ka) can account for the prolonged development of thick paleosols of S3 and S5, respectively. The ampli¢ed planktic N18O signals of the SCS relative to the Sulu Sea record are partly caused by the changing inten-sities of the East Asian monsoons at the glacial^ interglacial time-scale throughout the late Quater-nary.
Acknowledgements
This study is a contribution to the Taiwan IM-AGES Program and the APECProject. Special thanks are extended to Drs. Min-Te Chen and Bernhard Diekmann for their instructions on con-ducting the time-series analyses presented. We thank the scienti¢c party and crew of IMAGES-III-IPHIS-Leg II Cruise for a successful coring in the SCS. The support o¡ered by French MENRT, TAFF, CNRS/INSU and IFRTP to the operation of the cruise vessel Marion Dufrense and the IMAGES Program is highly appreciated. We thank Dr. Brad Linsley for providing us his ODP 769 data and his laboratory standards. Dr. Heslop kindly provided the calibrated magnetic susceptibility data of the Louchuan Section. This study has been supported by Grants NSC-88-2116-002-012 and NSC-90-2166-M002-019 to K.-Y.W. and Y.-G.C. For curation of cores and
sampling service, we acknowledge the Core Lab-oratory located at the National Taiwan Ocean University, funded by the Center for Ocean Re-search, National Science Council, Taiwan.
References
An, Z.S., Kukla, G.J., Porter, S.C., Xiao, J.L., 1991. Magnetic susceptibility evidence of monsoon variation of the Loess Plateau of central China during the last 130,000 years. Quat. Res. 36, 29^36.
An, Z.S., Liu, T.S., Lu, Y.C., Porter, S.C., Kukla, G.J., Wu, X.H., Hua, Y.M., 1990. The long-term paleomonsoon var-iation recorded by the loess-paleosol sequence in central China. Quat. Int. 7/8, 91^95.
Bassinot, F.C., Labeyrie, L.D., Vincent, E., Quidelleur, X., Shackleton, N.J., Lancelot, Y., 1994. The astronomical theory of climate and the age of the Brunhes^Matuyama magnetic reversal. Earth Planet. Sci. Lett. 126, 91^108. Berger, W.H., Bickert, T., Schmidt, H., Wefer, G., Yasuda,
M., 1993a. Quaternary oxygen isotope record of pelagic for-aminifers : Site 805, Ontong Java Plateau. In: Berger, W.H., Kroenke, L.W., Mayer, L.A., et al. (Eds.), Proc. ODP, Sci. Results 130, Ocean Drilling Program, College Station, TX, pp. 363^379.
Berger, W.H., Bickert, T., Schmidt, H., Wefer, G., 1993b. Quaternary oxygen isotope record of pelagic foraminifers: Site 806, Ontong Java Plateau. In: Berger, W.H., Kroenke, L.W., Mayer, L.A., et al. (Eds.), Proc. ODP, Sci. Results 130, Ocean Drilling Program, College Station, TX, pp. 381^ 395.
Brummer, G.-J.A., Hemleben, C., Spindler, M., 1987. Ontog-eny of extant spinose planktonic foraminifera (Globigerini-dae): A concept exempli¢ed by Globigerinoides sacculifer (Brady) and G. ruber (D’Orbigny). Mar. Micropaleontol. 12, 357^381.
Chen, C.-T.A., Huang, M.-H., 1996. A mid-depth front sepa-rating the South China Sea Water and the Philippine Sea Water. J. Oceanogr. 52, 17^25.
Chen, M-T., Beaufort, L., 1998. Exploring Quaternary vari-ability of the East Asian monsoon. Kuroshio Current, and the Western Paci¢c warm pool systems: high-resolution in-vestigations of paleoceanography from the IMAGES III/ MD106 ^ IPHIS cruise TAO. Terr. Atmos. Ocean. Sci. 9, 129^142.
Chen, M.-T., Huang, C.-Y., 1998. Ice-volume forcing of winter monsoon climate in the South China Sea. Paleoceanography 13, 622^633.
CLIMAP, 1981. Seasonal reconstructions of the Earth’s sur-face at the Last Glacial Maximum. Map and Chart Series MC-36, Geological Society of America.
Dansgarrd, W., 1964. Stable isotopes in precipitation. Tellus 16, 436^468.
Ding, Z.L., Liu, T.S., Rutter, N.W., Yu, Z.W., Guo, Z.T.,
Zhu, R.X., 1995. Ice-volume forcing of East Asian winter monsoon variation in the past 800,000 years. Quat. Res. 44, 140^159.
Gong, G.-C., Liu, K.-K., Liu, C.-T., Pai, S.-C., 1993. The chemical hydrography of the South China Sea west of Lu-zon and a comparison with the West Philippine Sea. Terr. Atmos. Ocean. Sci. 3, 587^602.
Heslop, D., Langereis, C.G., Dekkers, M.J., 2000. A new as-tronomical timescale for the loess deposits of Northern Chi-na. Earth Planet. Sci. Lett. 184, 125^139.
Ho¡man, G., Heimann, M., 1997. Water isotope modeling in the Asian monsoon region. Quat. Int. 37, 115^128. Huang, C.-Y., Liew, P.-M., Zhao, M., Chang, T.-C., Kuo,
C.-M., Chen, M.-T., Wang, C.-H., Zheng, L.-F., 1997. Deep sea and lake records of the southeast Asian paleomon-soons for the last 35 thousand years. Earth Planet. Sci. Lett. 146, 59^72.
Izett, G.A., Obradovich, J.D., 1992. Laser-fusion 40Ar/39Ar ages of Australasian tektites (abstract). Lunar Planet. Sci. 23, 593^594.
Jian, Z., Huang, B., Kuhnt, W., Lin, H.-L., 2001. Late Qua-ternary upwelling intensity and East Asian monsoon forcing in the South China Sea. Quat. Res. 55, 363^370.
Kent, D.V., Schneider, D.A., 1995. Correlation of paleo-intensity variation records in the Brunhes/Matuyama polar-ity transition interval. Earth Planet. Sci. Lett. 129, 135^ 144.
Kienast, M., Steinke, S., Stattegger, K., Calvert, S.E., 2001. Synchronous tropical South China Sea SST change and Greenland warming during deglaciation. Science 291, 2132^2134.
Lau, K.-M., Li, T., 1984. The monsoon of East Asia and its global associations ^ a survey. Bull. Am. Meteorol. Soc. 65, 114^125.
Lea, D.W., Pak, D.K., Spero, H., 2000. Climate impact of late Quaternary equatorial Paci¢c sea surface temperature varia-tion. Science 289, 1719^1724.
Lee, M.-Y., Wei, K.-Y., 2000. Australasian microtektites in the South China Sea and the West Philippine Sea: Implications for age, size and location of the impact crater. Meteorit. Planet. Sci. 35, 1151^1155.
Lee, M.-Y., Wei, K.-Y., Chen, Y.-G., 1999. High resolution oxygen isotope stratigraphy for the last 150,000 years in the southern South China Sea: core MD972151. Terr. Atmos. Ocean. Sci. 10, 239^254.
Lee, T.-Q., 2000. Geomagnetic secular variation of the last one million years recorded on core MD972142 from the south-east South China Sea. J. Geol. Soc. China 43, 423^433. Levitus, S., Boyer, T.P., 1994. World Ocean Atlas, Vol. 4,
Temperature. NOAA, U.S. Dept. Commerce, Washington, DC.
Levitus, S., Burgett, R., Boyer, T.P., 1994. World Ocean Atlas, Vol. 3, Salinity. NOAA, U.S. Dept. Commerce, Washing-ton, DC.
Li, C.L., Ouyang, Z.Y., Liu, T.S., An, Z.S., 1993. Microtek-tites and glassy microspherules in loess ^ their discoveries and implication. Sci. China B36, 1141^1152.
Linsley, B.K., 1996. Oxygen-isotope record of sea level and climatic variation in the Sulu Sea over the past 150,000 years. Nature 380, 234^237.
Linsley, B.K., Dunbar, R.B., 1994. The late Pleistocene history of surface water N13Cin the Sulu Sea: Possible relationship to Paci¢c deepwater N13Cchanges. Paleoceanography 9, 317^340.
Lu, H., Liu, X., Zhang, F., An, Z., Dodson, J., 1999. Astro-nomical calibration of loess^paleosol deposits at Louchuan, central Chinese Loess Plateau. Palaeogeogr. Palaeoclimatol. Palaeoecol. 154, 237^246.
Martinez, J.I., DeDeckker, P.D., Chivas, A.R., 1997. New estimates for salinity changes in the Western Paci¢c Warm Pool during the Last Glacial Maximum: Oxygen-isotope evidence. Mar. Micropaleontol. 32, 311^340.
Miao, Q., Thunell, R.C., Anderson, D.M., 1994. Glacial^Ho-locene carbonate dissolution and sea surface temperature in the South China and Sulu Seas. Paleoceanography 9, 269^ 290.
Nitani, H., 1972. Beginning of the Kuroshio. In: Stommel, H., Yoshida, K. (Eds.), Kuroshio. University of Washington Press, pp. 129^163.
Ohkouchi, N., Kawamura, T., Nakamura, T., Taira, A., 1994. Small changes in the sea surface temperature during the last 20,000 years: Molecular evidence from the western tropical Paci¢c. Geophys. Res. Lett. 21, 2207^2210.
Paillard, D., Labeyrie, L., Yiou, P., 1996. Macintosh program performs time-series analysis. EOS Trans. AGU 77, 379. Pelejero, C., Grimalt, J.O., Sarnthein, M., Wang, L., Flores,
J.-A., 1999. Molecular biomarker record of sea surface tem-perature and climatic change in the South China Sea during the last 140,000 years. Mar. Geol. 156, 109^121.
Porter, S.C., Hallet, B., Wu, X., An, Z., 2001. Dependence of near-surface magnetic susceptibility on dust accumulation rate and precipitation on the Chinese Loess Plateau. Quat. Res. 55, 271^283.
Prell, W.L., Marvil, R.E., Luther, M.E., 1990. Variability in upwelling ¢elds in the Northwestern Indian Ocean ^ 2. Data-model comparison at 9000 years B.P.. Paleoceanogra-phy 5, 447^457.
Rottman, M.L., 1979. Dissolution of planktonic Foraminifera and pteropods in the South China Sea sediments. J. Fora-minif. Res. 9, 41^49.
Schneider, D.A., Kent, D.V., Mello, G.A., 1992. A detailed chronology of the Australasian impact event, the Brunhes^ Matuyama geomagnetic polarity reversal, and global climate change. Earth Planet. Sci. Lett. 111, 395^405.
Shackleton, N.J., Opdyke, N.D., 1973. Oxygen isotope and paleomagnetic stratigraphy of equatorial Paci¢c Core V28-238: Oxygen isotope temperatures and ice volumes on a 105 year and 106year scale. Quat. Res. 3, 39^55.
Shackleton, N.J., Opdyke, N.D., 1976. Oxygen isotope and paleomagnetic stratigraphy of equatorial Paci¢c Core V28-239: Late Pliocene to latest Pleistocene. In: Cline, R., Hays, J.D. (Eds.), Investigation of Late Quaternary Paleoceanog-raphy and Paleoclimatology. Mem. Geol. Soc. Am. 145, pp. 449^464.
Shaw, P.-T., 1989. The intrusion of water masses into the sea southwest of Taiwan. J. Geophys. Res. 94, 18213^18226. Shaw, P.-T., 1991. The seasonal variation of the intrusion of
the Philippine Sea water into the South China Sea. J. Geo-phys. Res. 96, 821^827.
Sun, X., Li, X., 1999. A pollen record of the last 37 ka in deep sea core 17940 from the northern slope of the South China Sea. Mar. Geol. 156, 227^242.
Tauxe, L., Herbert, T., Shackleton, N.J., Kok, Y.-S., 1996. Astronomical calibration of the Matuyama^Brunhes bound-ary: Consequences for magnetic remanence acquisition in marine carbonates and the Asian loess sequences. Earth Planet. Sci. Lett. 140, 133^146.
Thompson, P.R., Be¤, A.W.H., Duplessy, J.-C., Shackleton, N.J., 1979. Disappearance of pink-pigmented Globigeri-noides ruber at 120,000 yr BP in the Indian and Paci¢c Oceans. Nature 280, 554^558.
Thunell, R.C., Anderson, D., Gellar, D., Miao, Q.M., 1995. Sea-surface temperature estimates for the tropical western Paci¢c during the last glaciation and their implications for the Paci¢c warm pool. Quat. Res. 41, 255^264.
Verosub, K.L., Fine, P., Singer, M.J., TenPas, J., 1993. Pedo-genesis and paleoclimate : Interpretation of the magnetic sus-ceptibility record of Chinese loess^paleosol sequences. Geol-ogy 21, 1011^1014.
Wang, L., Wang, P., 1990. Late Quaternary paleoceanography of the South China Sea: Glacial^interglacial contrasts in as enclosed basin. Paleoceanography 5, 77^90.
Wang, L., Sarnthein, M., Erlenkeuser, H., Grimalt, J., Groote, P., Heiling, S., Ivanova, E., Kienast, M., Pelejero, C., P£au-mann, U., 1999. East Asian monsoon climate during the late Pleistocene : High-resolution sediment records from the South China Sea. Mar. Geol. 156, 245^284.
Wang, P., 1999. Response of western Paci¢c marginal seas to glacial cycles: Paleoceanographic and sedimentological fea-tures. Mar. Geol. 156, 5^39.
Webster, P.J., Yang, S., 1992. Monsoon and ENSO: Selec-tively interactive system. Quat. J. R. Met. Soc. 118, 877^926. Wei, K.-Y., Lee, M.-Y., Duan, W., Chen, C., Wang, C.-H., 1998a. Paleoceanographic change in the northeastern South China Sea during the last 15,000 years. J. Quat. Sci. 13, 55^64. Wei, K.-Y., Lee, T.-Q., and Shipboard Scienti¢c Party of IM-AGES III/MD106-IPHIS Cruise (Leg III), 1998b. Late Pleistocene volcanic ash layers in Core MD972142, o¡shore of Northwestern Palawan, South China Sea: A preliminary report. Terr. Atmos. Ocean. Sci. 9, 143^152.
Wei, K.-Y., Zhang, Z.-W., Wray, C., 1992. Shell ontogeny of Globorotalia in£ata (I): Growth dynamics and ontogenetic stages. J. Foraminif. Res. 22, 318^327.
Wytrki, K., 1961. Physical oceanography of the Southeast Asian Waters. In: NAGA Report, Vol. 2, Scripps Inst. of Oceanography, La Jolla, CA, 195 pp.
Zhou, L.P., Shackleton, N.J., 1999. Misleading positions of geomagnetic reversal boundaries in Eurasian loess and im-plications for correlation between continental and marine sedimentary sequences. Earth Planet. Sci. Lett. 168, 117^ 130.
