C Al

Chinese Science Bulletin

**** ******* ** ***** *****

Pulleniatina Minimum Event during the last deglacia-

tion in the southern South China Sea

AN Yang & JIAN ZhiMin

State Key Laboratory of Marine Geology, Tongji University, Shanghai 200092, China

The planktonic foraminiferal faunal census of core MD 05-2894 (7 2.25 N, 111 33.11 E, water depth 1982

m), retrieved from the southern South China Sea (SCS) during the Marco Polo cruise in 2005, was

performed to investigate the abundance changes of a subsurface dweller, Pulleniatina obliquiloculata.

The results display that the abundance of P. obliquiloculata nearly declines to zero during 16.0 14.9 ka,

corresponding to the Heinrich 1 (H1) cold interval. The unexpected decrease of P. obliquiloculata occurs

in the adjacent cores, roughly between 17 and 14.8 ka based on the previous studies. Accordingly, the

Pulleniatina Minimum Event in the last deglaciation can serve as a good stratigraphical indicator, at

least in the southern SCS. To further explore the changes of sea surface temperature (SST) and sub-

surface seawater temperature (SSST), we made parallel Mg/Ca measurements on surface dweller Glo-

bigerinoides ruber and subsurface dweller P. obliquiloculata tests. Since the last deglaciation, the SSTs

show a continuous increasing trend towards the late Holocene, while the warming of the subsurface

water is punctuated by a 2 -cooling interval across the deglacial Pulleniatina Minimum Event. Both

increased 18O differences between G. ruber and P. obliquiloculata, and increased temperature differ-

ences between surface and subsurface water suggest a shoaling of the mixed layer during the deglacial

Pulleniatina Minimum Event. Therefore, we consider that the significant changes in the upper ocean

structure are responsible for the Pulleniatina Minimum Event during the last deglaciation in the south-

ern SCS.

Pulleniatina Minimum Event, subsurface seawater temperature, upper ocean structure, the last deglaciation, the southern South

China Sea

Pulleniatina obliquiloculata, a tropical subsurface dwe- obliquiloculata is closely related to subsurface seawater

lling planktonic foraminifer, calcifies most of its test temperatures (SSSTs), especially SSSTs in winter, and

below the mixed layer. In the global ocean, its highest the depth of thermocline (DOT). Studies on this species

abundance in surface sediments occurs in a relatively have become an important issue in paleoceanography of

the western Pacific[5 8].

narrow belt between about 10 N and 10 S, which coin-

cides generally with the equatorial current systems in Paleoceanographers have documented an extremely

the Atlantic, Indian and Pacific Oceans[1]. High concen- low abundance of P. obliquiloculata between 4.5 and

trations (more than 20%) of the species are found in the 3.0 cal. ka in deep sea sediment cores in Okinawa

Trough [5,6,9 11] and the northern SCS[12], which is defined

western equatorial Pacific and the northern South China

Sea (SCS)[1,2]. Previous studies in northwestern Pacific[3] as Pulleniatina Minimum Event (PME). It correlates

and the northern SCS[2] demonstrate that P. obliquilocu- well with the Chinese lake[13] and stalagmite[14] records,

lata is closely related to warm and saline currents like Received November 5, 2008; accepted March 5, 2009, published online May 15, 2009

doi: 10.1007/s11434-009-0290-4

the Kuroshio Current. Sediment traps from the northern

Corresponding author (email: ************@*****.***)

SCS show that P. obliquiloculata occurs during the win- Supported by National Basic Research Program of China (Grant No.2007CB815900),

National Natural Science Foundation of China (Grant No. 40621063) and China Ocean

ter monsoon[4]. So it is believed that the abundance of P. Mineral Resources Research and Development Association (Grant No. DY 115-01-2-3)

Citation: An Y, Jian Z M. Pulleniatina Minimum Event during the last deglaciation in the southern South China Sea. Chinese Sci Bull, 2009, 54: 4514 4519, doi:

10.1007/s11434-009-0290-4

ARTICLES

consisting of mixed tests of G. ruber and G. sacculifer

which show the weakening of East Asian summer mon-

(800 1200 specimens, >154 m) were subjected to

soon during this period. Previous studies ascribed the

AMS 14C dating at Peking University. The results were

Holocene PME to the track and intensity changes of

converted to calendar years based on Fairbanks et al. s[16]

Kuroshio or the decline of winter sea surface temperatu-

res (SSTs)[5,6,9,11]. During the last deglacial warming per- method and CALIB 5.0 software, respectively. Few dif-

iod, there is another low abundance of P. obliquiloculata ferences occurred between 12.41 ka BP and the present.

in the southern SCS similar to the late Holocene PME. From 11.9 to 14 ka, the dataset used by CALIB 5.0 in-

In this study, we ll try to understand when, where and cludes predominantly laminated sediments from the Ca-

why this event happens so as to get a better understand- riaco Basin, along with coral data from numerous inves-

ing of the paleoceanographic changes in the southern tigators of varying sample quality, in contrast, Fairbanks

SCS during the last deglaciation. et al. s[16] dataset uses the 1382-ring floating tree ring

data. We use the latter method to construct our age mo-

1 Materials and methods del because it is more accurate and widely cited.

MD 05-2894 (lat. 7 2.25 N, long. 111 33.11 E, Figure 1, 14

Table 1 AMS C datings of MD 05-2894

1982 m water-depth; MD 94 henceforth), retrieved from AMS 14C years (a BP ( 1 ))

Depth (cm) Cal. years (a BP)

the southern SCS during the Marco Polo cruise in 62.5 501*-**-****-**

2005, is a 10.85 m long CASQ core free of turbidites. 104.5 676*-**-****-**

140.5 108**-**-***** 105

230 samples were sampled in the upper 467 cm with an

188.5 131**-**-***** 101

interval of 2 cm for oxygen, carbon isotopic and Mg/Ca

214.5 132**-**-***** 103

measurements on surface dweller Globigerinoides ruber 284.5 137**-**-***** 121

and subsurface dweller P. obliquiloculata. All laboratory 368.5 148**-**-***** 177

methods can be referred to ref. [15] and all experiments

were completed at State Key Lab of Marine Geology,

Tongji University. The >154 m fraction of each sample

2 Results and discussion

(with the sampling interval of 2 4 cm) was split into an

The precise age model is established based on the cal-

aliquot containing 200 to 300 specimens. Main species,

endar years converted from the AMS 14C dating results.

such as G. ruber, Globigerinoides sacculifer, P. obliqui

The studied interval covers about 18.9 cal. ka (All ages

loculata and Neogloboquadrina dutertrei, were counted.

To construct the precise age model, 7 samples (Table 1) discussed henceforth are calendar ages), with an average

OCEANOLOGY

time resolution of 82 a. The sedimentation rate (SR)

varies much among different time intervals. The upper

134.5 cm Holocene samples have a low average SR of

11.7 cm/ka; while the lower part has a relatively high S-

R of 46.2 cm/ka in average, with its maximum of 252.4

cm/ka (Figure 2(c)). The 8 a/sample-resolution then is

the highest temporal resolution till now for the late Qu-

aternary paleoceanographic study in the southern SCS.

The oxygen isotope values of G. ruber show an ob-

vious glacial/interglacial cycle, varying between 0.89

4 and 3.492 (Figure 2(a)). During the early stage

of the deglaciation (15 18.9 ka), the 18O of G. ruber

fluctuated around 1.7, and then became more nega-

tive to 3.0 in the Holocene. 18O of P. obliquilocu-

lata vary in a similar trend to that of G. ruber (Figure

2(b)), but shows heavier values of 0.1 and 1.3 in

the early deglaciation and the Holocene, respectively.

Figure 1 Locations of cores in the southern SCS related to this

18O of G. ruber have an average difference of 1.3

study.

An Y et al. Chinese Science Bulletin December 2009 vol. 54 no. 23 4515

between the early deglaciation and the Holocene, which event, so we define it as H1-PME. Comparing the result

with that of the adjacent cores such as 17962[18] and M-

is a little larger than that (1.2 ) of P. obliquiloculata.

D97-2151[19], we will see that the H1-PME extensi- vely

Furthermore, both species have clear responds to the

occurs in the southern SCS, roughly between 17 and

Younger Dryas event (13.2 11.5 ka), with reduced ox-

14.8 ka. But no such event exists in the northern SCS

ygen isotopes by ~1.0 and ~0.8, respectively (Fig-

(17940[17], Figure 3(a)), so the H1-PME may be a uni-

ure 2). Such changes clearly indicate the existence of

que event and can be a good stratigraphical indicator in

millennial-scale climate fluctuations like the Younger

the southern SCS.

Dryas event in the southern SCS.

Figure 2 Oxygen isotopic records of planktonic foraminifera and

the sedimentation rate in core MD 05-2894. a, G. ruber. b, P. obliquil-

Figure 3 The abundance of P. obliquiloculata of various cores in

oculata; c, sedimentation rate. PME is the Holocene Pulleniatina [17]

the southern SCS and 17940 in the northern SCS. a, 17940 ; b, MD

Minimum Event, and YD is Younger Dryas event. H1-PME is the de- [18] [19]

05-2894, this study; c, 17962 ; d, MD97-2151 . PME is the Holo-

glacial Pulleniatina Minimum Event correlating to H1 event. Average

cene Pulleniatina Minimum Event; H1-PME is the deglacial Pulleni-

values of various periods are represented by the dotted lines. Ar-

atina Minimum Event correlating to H1 event.

14

rows show the AMS C radiocarbon datings.

Paleoceanographers have thoroughly investigated the

It is indicated by the faunal census data that the abun-

Holocene PME. They first ascribed this event to winter

dance of P. obliquiloculata varies from 0 to 21.6% from

SST declines[5,12], further studies concluded that track

the last glacial maximum (LGM) to present, with an

and intensity changes of Kuroshio Current might also

average of 5.4%. The abundance of this species shows a

cause the event[5,6,9,11], and also considered ENSO fact-

decreasing trend since the last deglaciation. Its highest

or[20]. Lin et al.[10] has performed multi-species isotopic

abundance is documented in the early deglaciation (16.2

analysis and G. ruber Mg/Ca SST reconstructions. Bas-

ka; Figure 3(b)), while its lowest concentrations are re-

ed on the data in the western Pacific marginal seas (ma-

corded during two time-intervals, 3.7 5.2 ka and 14.9

inly in Okinawa Trough and SCS), she insisted on that

16.0 ka. P. obliquiloculata has a very low abundance of

the Holocene PME be independent of any above-menti-

1.9% in average, with its lowest value of 0.4% during

oned paleoceanographic changes. Till now we indeed

3.7 5.2 ka. This section correlates well with the Holo-

can not attribute the Holocene PME in the southern SCS

cene PME which is extensively documented in marginal

(Figure 3) to variations of Kuroshio Current; we can not

seas of the western Pacific[5,6,9 11]. But after the PME,

expect decreases of Mg/Ca-derived SSTs in Okinawa T-

the abundance is still lower than the early Holocene in

rough[10,21] or the southern SCS[22,23], either. According

the southern SCS, which is different from that in the

to the numerical simulation study[20], El Ni o strength-

northern SCS (Figure 3(a)) and Okinawa Trough[12].

ened after the Holocene PME, which is contradictory to

What s more, P. obliquiloculata shows another dramati-

the ENSO hypothesis. Observed only in the southern

cally low abundance between 14.9 and 16.0 ka. The av-

SCS, H1-PME may have a reason completely different

erage abundance of 1.1% is obviously lower than that of

from that occurred in the Holocene, which means Ku-

the upper and lower parts of this core (usually greater

than 10%). This event lasts for 1.1 ka from 16.0 to 14.9 roshio can not be the triggering factor. No obvious SST

ka (Figure 3(b)) and correlates well with the Heinrich 1 declines are documented in the southern SCS during H1-

4516 www.scichina.com csb.scichina.com www.springer.com/scp www.springerlink.com

ARTICLES

PME[20], so SST changes are not the cause. Then we

must divert our attention to other influencing factors su-

ch as subsurface water temperatures and nutrient supply,

etc.

In order to explore the reasons of H1-PME, we ran

parallel oxygen, carbon isotopic and Mg/Ca ratio meas-

urements on G. ruber and P. obliquiloculata tests. Mg/

Ca SSTs are derived by the equations of Deckens et

al.[25], Lea et al.[26] and Huang et al.[27], respectively. SS-

Ts of coretop samples derived by the equation of Huang

et al.[27] (28.0, averaged by spring-summer and fall-

winter SSTs) are closer to modern annual SST[28] at this Figure 4 SSTs and SSSTs of MD 05-2894 and SST comparisons

between MD 05-2894 and MD97-2151. a, Mg/Ca-derived SSST, 3-

site, therefore we adopted Huang et al. s[27] equation-de-

point smoothed values are represented by dashed line; b, Mg/Ca-

rived SSTs in this study. Furthermore, we use Anand et derived SST, averages of various periods are indicated by dotted

al. s[26] and Huang et al. s[27] equation to obtain P. obli- 37 [30]

lines; c, UK -derived SST of MD97-2151 . PME is the Holocene

Pulleniatina Minimum Event, and YD is Younger Dryas event. H1-

quiloculata Mg/Ca SSSTs, respectively. The difference

PME is the deglacial Pulleniatina Minimum Event correlating to H1

between the two estimates is about 0.29, which may event.

be caused by the differences between Sargasso Sea and

STs continuously decrease after reaching its maximum

SCS. We use Huang et al. s[27] estimates because the eq-

in early to mid-Holocene, which displays a similar trend

uation come from SCS. Here we specially addressed the

to that of Timor Sea[8]. The warming process of subsur-

two PMEs in the Holocene and the last deglaciation; de-

face water was interrupted in the Holocene PME, H1-

tailed Mg/Ca-derived paleotemperatures will be discu-

PME and YD event. SSSTs decline as much as ~3.1,

ssed elsewhere.

4.8 and 3.1 in the three periods, respectively, the

Mg/Ca-derived SSTs vary from 23.7 to 29.8,

most obvious among which is the ~2 decrease in aver-

with a similar trend to that of 18O (Figure 4(b)). They

age during H1-PME.

rise from an average of 25.5 in early deglaciation to

Previous studies demonstrated that the 18O differ-

28.8 in the Holocene. The~3.3 difference is close to

ences ( 18O) of planktonic subsurface dweller such as

those of the adjacent cores such as MD97-2151 (2.9 [30]) P. obliquiloculata and surface dweller such as G. ruber

OCEANOLOGY

and MD01-2390 (3 [22]). Mg/Ca-derived SSTs show a can be used to indicate the relative depth of thermocline.

gradual increase since H1-PME, with no punctuations Large 18O indicate a shallow thermocline, and vice

during the two PMEs (Figure 4(b)), but SSTs derived by versa[7,32]. So do the temperature gradients ( T) between

transfer functions have a remarkable decrease during the SSTs and SSSTs. Large T indicate a shallow mixed layer,

two periods. For instance, winter SSTs of 17962 drop by and vice versa. During the deglacial PME, SSTs show no

as much as about 6 [31] between 17 and 15 ka. Such obvious declines while SSSTs have an average decrease

differences may be due to the No-analog [15,23] prob- of ~2, so T becomes larger, which means a shallow

lem. That is to say, the low abundance of P. obliquilo- thermocline then. The shallow thermocline is also indi-

culata makes the application of transfer functions in the cated by larger 18O values during the PMEs. Besides,

southern SCS problematic. SSTs from nearby cores both 18O and T show a deep thermocline in the early

(MD97-2151[30] and MD01-2390[22]) demonstrate no deglaciation and a shallow thermocline in the Holocene.

obvious changes between 17 and 14.8 ka, and then it can H1-PME occurred exactly during the transition of the two

be concluded that SST declines are not the reason for stages. The SSSTs decrease and upper ocean structure

the PM-Es. changes (Figure 5) influence the habitat of P. obliquilo-

As shown in Figure 4(a)), SSSTs have a fluctuation culata and cause the low abundance of this species.

from 16.0 to 23.2, slightly larger than SSTs. Both of The Holocene PME is widely distributed in marginal

them show a similar trend to 18O, that is, an increasing seas of the western Pacific, but H1-PME is only docu-

mented in the southern SCS, so some unique paleocean-

trend towards the Holocene. Different from SSTs, SS-

An Y et al. Chinese Science Bulletin December 2009 vol. 54 no. 23 4517

surface water from the northern SCS mixed layer is dra-

ined into the upper seasonal thermocline; moves south-

ward and becomes subsurface water of the southern SC-

S[35]. Then the southern SCS SSSTs show an obvious de-

cline; the thermocline shoals; and P. obliquiloculata de-

crease. In order to understand the reasons why SSSTs

decline during H1-PME, further studies are still needed

to discuss the links between various paleoceanographic

changes.

Figure 5 The temperature gradients ( T) between SSTs and

18 18

SSSTs, and the O differences ( O) between P. obliquiloculata

and G. ruber. a, The temperature gradients ( T) between SSTs and

3 Conclusions

18 18

SSSTs; b, the O differences ( O) between P. obliquiloculata

and G. ruber. PME is the Holocene Pulleniatina Minimum Event,

Based on a careful planktonic foraminiferal faunal cen-

and YD is Younger Dryas event. H1-PME is the deglacial Pulleni-

atina Minimum Event correlating to H1 event.

sus count, we discovered that P. obliquiloculata nearly

disappears between 16.0 and 14.9 ka in the southern

ographic changes may exist in the southern SCS during

SCS. This interval correlates well with the Heinrich 1

the last deglacial warming period. Koutavas et al.[33] co-

cooling event. The H1-PME is extensively documented

mpared SSTs of V21 30 in the eastern Pacific with

in the southern SCS, thus can be a good stratigraphical

18287 in the southern SCS and implied a more El

indicator in this region. Mg/Ca-derived SSTs gradually

Nino-like condition between 17.0 and 14.8 ka, which is

increase since the last deglaciation, with little punctua-

contemporary with H1-PME in the southern SCS. But

tion during the Holocene PME, H1-PME or Younger

there is no such event in the western equatorial Pacific[8],

Dryas event. But SSSTs demonstrate significant decr-

so H1-PME cannot be ascribed to the tropical factors

eases during these periods. Both 18O and temperature

like ENSO. Here we present the following hypothesis

differences become larger during the two PMEs, indi-

about the reasons of H1-PME. One is related to that

cating the shoaling of the thermocline. The restructured

North Atlantic Deep Water (NADW) weakens while

upper ocean may be responsible for the SSSTs decline,

North Pacific Deep Water (NPDW) strengthens during

which results in the remarkable decrease of P. obliqui-

H1[34]. Cold NPDW upwells to be subsurface water

loculata during the deglacial Pulleniatina Minimum Ev-

when blocked in the southern SCS; then SSSTs decrease;

ent. Such restructuring may be caused by the upwelling

the thermocline shoals and the abundance of P. obliqui-

of NPDW, gradual opening seaways, or better thermo-

loculata drop. Another hypothesis is that seaways con-

cline ventilation during the last deglaciation. Further st-

necting the southern SCS with Indian and Pacific Ocean

udies are still needed to know the exact reasons for these

gradually open during the last deglaciation. The restruc-

changes.

tured upper ocean makes subsurface dweller P. obliqui-

loculata absent in this region. Furthermore, forced by

Samples in this study were provided by Marco Polo Cruise in 2005

winter monsoon, the mixed layer in the upper SCS will which was sponsored by IMAGES. We sincerely thank all the onboard

deepen. When the meridional seawater temperature gra- scientists and the crew. We should also thank Cheng Xinrong and Qiao

Peijun for their assistance in isotopic and Mg/Ca measurements.

dient increases, the thermocline is best ventilated. More

1 B A W H. An ecological, zoogeographical and taxonomic review of semblages. Palaeogeogr Paleoclimatol Palaeoecol, 1981, 35:

recent planktonic foraminifera. In: Ramsay A T S, ed. Oceanic 241 279

Micropaleontology. London: Academic Press, 1977. 1 100 4 Chen R H, Jian Z M, Zheng Y L, et al. Seasonal variations of the

2 Pflaumann U, Jian Z M. Modern distribution patterns of planktonic planktonic foraminiferal flux in the central South China Sea (in

foraminifera in the South China Sea and western Pacific: a new Chinese with English abstract). J Tongji Univ, 2000, 28: 73 77

transfer technique to estimate regional sea-surface temperatures. 5 Li B H, Jian Z M, Wang P X. Pulleniatina obliquiloculata as a pa-

Mar Geol, 1999, 156: 41 83 leoceanographic indicator in the southern Okinawa Trough during

3 Thompson P R. Planktonic foraminifera in the western north Pacific the last 20000 years. Mar Micropaleontol, 1997, 32: 59 69

during the past 150000 years: comparison of modern and fossil as- 6 Jian Z M, Wang P X, Saito Y, et al. Holocene variability of the Ku-

4518 www.scichina.com csb.scichina.com www.springer.com/scp www.springerlink.com

ARTICLES

roshio Current in the Okinawa Trough, northwestern Pacific Ocean. cropaleontol, 2003, 49: 335 364

Earth Planet Sci Lett, 2000, 184: 305 319 21 Sun Y B, Oppo D W, Xiang R, et al. Last deglaciation in the Oki-

7 Tian J, Wang P X, Chen R H, et al. Quaternary upper ocean thermal nawa Trough: subtropical northwest Pacific link to Northern Hemi-

gradient variations in the South China Sea: implications for East sphere and tropical climate. Paleoceanography, 2005, 20: PA4005,

Asian monsoon climate. Paleoceanography, 2005, 20: PA4007, doi: doi:10.1029/2004PA001061

10.1029/2004PA001115 22 Steinke S, Chiu H, Yu P, et al. On the influence of sea level and

8 Xu J, Holbourn A, Kuhnt W, et al. Changes in the thermocline monsoon climate on the southern South China Sea fresh water

structure of the Indonesian Outflow during Terminations I and II. budget over the last 22,000 years. Quat Sci Rev, 2006, 1477 1488

Earth Planet Sci Lett, 2008, 273: 152 162 23 Steinke S, Yu P, Kucera M, et al. No-analog planktonic foraminiferal

9 Ujii Y, Ujii H. Late Quaternary course changes of the Kuroshio faunas in the glacial southern South China Sea: implications for the

Current in the Ryukau Arc region, northwestern Pacific Ocean. Mar magnitude of glacial cooling in the western Pacific warm pool. Mar

Micropaleontol, 1999, 37: 23 40 Micropaleontol, 2008, 66: 71 90

10 Lin Y S, Wei K Y, Lin I T, et al. The Holocene Pulleniatina Mini- 24 Moy C M, Seltzer G O, Rodbell D T, et al. Variability of El

mum Event revisited: geochemical and faunal evidence from the Ni o/Southern Oscillation activity at millennial timescales during

Okinawa Trough and upper reaches of the Kuroshio Current. Mar the Holocene epoch. Nature, 2002, 420: 162 165

Micropaleontol, 2006, 59: 153 170 25 Dekens P, Lea D, Pak D, et al. Core top calibration of Mg/Ca in

11 Li T G, Liu Z X, Hall M A, et al. Heinrich event imprints in the tropical foraminifera: refining paleotemperature estimation. Geochem

Okinawa Trough: evidence from oxygen isotope and planktonic fo- Geophys Geosyst, 2002, 3: 1022, doi:10.1029/2001GC00 0200

raminifera. Palaeogeogr Paleoclimatol Palaeoecol, 2001, 176: 26 Lea D W, Pak D, Spero H. Climate impact of Late Quaternary Equa-

133 146 torial Pacific sea surface temperature variations. Science, 2000, 289:

12 Jian Z M, Li B H, Pflaumann U, et al. Late Holocene cooling event in 1719 1724

the western Pacific. Sci China Ser D-Earth Sci, 1996, 39: 543 550

27 Huang K F, You C F, Lin H L, et al. In situ calibration of Mg/Ca ratio

13 Peng Y, Xiao J, Nakamura T, et al. Holocene East Asian monsoonal

in planktonic foraminiferal shell using time series sediment trap: a

precipitation pattern revealed by grain-size distribution of core

case study of intense dissolution artifact in the South China Sea. Geo-

sediments of Daihai Lake in Inner Mongolia of north-central China.

chem Geophys Geosyst, 2008, 9: doi:10.1029/2007 GC001660

Earth Planet Sci Lett, 2005, 233: 467 479

28 Levitus S, Boyer T P. World Ocean Atlas 1994: Temperature. NO-

14 Wang Y J, Cheng H, Edwards L R, et al. The Holocene Asian mon-

AA Atlas NESDIS, Vol. 4. Washington D C U.S. Department of C-

soon: links to solar changes and north Atlantic climate. Science,

ommerce, 1994

2005, 308: 854 857

29 Anand P, Elderfiled H, Conte M. Calibration of Mg/Ca thermometry

15 Jian Z M, Wang B S, Qiao P J. Late Quaternary changes of sea sur-

in planktonic foraminifera from a sediment trap time series. Paleo-

face temperature in the southern South China Sea and their com-

ceanography, 2003, 18: 1050, doi:10.1029/2002PA 00846

parison with the paleoclimatic records of polar ice cores (in Chinese

Zhao M X, Huang C Y, Wang C C, et al. A millennial-scale Uk37

30

with English abstract). Quat Sci, 2008, 3: 391 398

sea-surface temperature record from the southern South China Sea

OCEANOLOGY

16 Fairbanks R G, Mortlock R A, Chiu T C, et al. Radiocarbon calibra-

(8 N) over the last 150kar: monsoon and sea-level influence. Pa-

tion curve spanning 0 to 50,000 years B.P. based on paired

laeogeogr Paleoclimatol Palaeoecol, 2006, 236: 39 55

230-***-***-**

Th/ U/ U and C dates on pristine corals. Quat Sci Rev, 2005,

31 Li B H, Jian Z M, Huang B Q, et al. Sea surface temperature estimate

24: 1781 1796

and subsurface water evolution in the southern South China Sea (in

17 Wang L J, Sarnthein M, Erlenkeuser H, et al. East Asian monsoon

Chinese with English abstract). J Micropaleontol, 2004, 21: 37 43

climate during the Late Pleistocene: high-resolution sediment re-

32 Jian Z, Huang B, Lin H, et al. Late Quaternary upwelling intensity

cords from the South China Sea. Mar Geol, 1999, 156: 245 284

and East Asian monsoon forcing in the South China Sea. Quat Res,

18 Fang D Y, Cheng X R, Wu G X, et al. Paleoceanographic record of

2001, 55: 363 370

the past 30 ka in the southern Nansha Area, South China Sea (in

33 Koutavas A, Stieglitz J, Marchitto T, et al. El-Ni o-Like Pattern in

Chinese with English abstract). Mar Geol Quat Geol, 2000, 20:

Ice Age Tropical Pacific Sea Surface Temperature. Science, 2002,

81 86

297: 226 230

19 Huang C Y, Chen M T, Lee M Y, et al. Planktic foraminifer faunal

34 Sarnthein M, Thorsten K, Grootes P, et al. Warmings in the far nor-

sea surface temperature records of the past two glacial terminations

thwestern Pacific premoted pre- Clovis immigration to America dur-

in the South China Sea near Wanan Shallow (IMAGES core

ing Heinrich Event 1. Geology, 2006, 34: 141 144, doi:10.1130/G

MD972151). Western Pacific Earth Sci, 2002, 2: 1 14

22200.1

20 Ujii Y, Ujii H, Taira A, et al. Spatial and temporal variability of

35 Wang D X, Du Y, Shi P. Evidence for thermocline ventilation in the

surface water in the Kuroshio source region, Pacific Ocean, over the

South China Sea in winter. Chinese Sci Bull, 2001, 46: 774 778

past 21000 years: evidence from planktonic foraminifera. Mar Mi-

An Y et al. Chinese Science Bulletin December 2009 vol. 54 no. 23 4519

Contact this candidate