Marine Micropaleontology 55 (2005) 49 – 62 www.elsevier.com/locate/marmicro

A test of different factors influencing the isotopic signal of planktonic foraminifera in surface sediments from the northern South China Sea Ludvig Lfwemarka,T, Wei-Li Honga,1, Tzen-Fu Yuib,2, Gwo-Wei Hungc,3 a Department of Geosciences, National Taiwan University, Taipei, Taiwan, ROC Institute of Earth Sciences, Academia Sinica, P.O. Box 55-1, 115 Taipei, Taiwan, ROC c Institute of Marine Geology and Chemistry, National Sun Yat-sen University, Kaohsiung, Taiwan 804, ROC b

Received 12 October 2004; received in revised form 4 February 2005; accepted 4 February 2005

Abstract The stable isotope composition of planktonic foraminifera is one of the most important proxies in paleoenvironmental research. In this study, three parameters affecting the stable isotope values of Globigerinoides ruber from surface sediment from the northern South China Sea were tested: different cleaning methods, different morphotypes, and different size fractions. Our results show that in the small size fraction, there is a small but significant effect on y13C by oxidizing the tests prior to measurement. Our data also confirm a small but significant difference between different morphotypes of G. ruber. However, the variability caused by the seasonal effect stable isotope value is larger than the effect caused by different cleaning protocols or different morphotypes. The large spread of the isotope values (up to 2x) have some implications to paleoceanographic reconstructions; when measurements are performed on a small number of foraminiferal tests, the isotope value does not necessarily reflect yearly average or a certain season but is a random value of the seasonal variability in that region. D 2005 Elsevier B.V. All rights reserved. Keywords: Planktonic foraminifera; Stable isotopes; South China Sea; Cleaning protocol; Surface sediment

1. Introduction T Corresponding author. Previously at Institute of Earth Sciences, Academia Sinica, P.O. Box 55-1, 115 Taipei, Taiwan, ROC. Fax: +886 2 23636095. E-mail addresses: [email protected] (L. Lfwemark), [email protected] (W.-L. Hong), [email protected] (T.-F. Yui), [email protected] (G.-W. Hung). 1 Fax: +886 2 23636095. 2 Fax: +886 2 2783 9871. 3 Fax: +886 7 5255149. 0377-8398/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.marmicro.2005.02.004

The stable isotope composition of planktonic foraminifera is one of the most commonly used paleoenvironmental proxies. The isotope ratio of d 18O is routinely used for the chronostratigraphic framework of marine sediment records and also often used for the reconstruction of parameters such as marine water temperatures (e.g., Emiliani, 1955;

50

L. Lo¨wemark et al. / Marine Micropaleontology 55 (2005) 49–62

Shackleton, 1967; Weaver et al., 1997), sea level/ice volume changes (e.g., Hays et al., 1976; Shackleton, 1987; Bard et al., 1989), d 18O of sea water and salinity variations (e.g., Duplessy et al., 1992; Maslin et al., 1995; Wang et al., 1995; Rohling, 2000; Lea et al., 2003). Recent approaches include the use of differences in d 18O between different species or regions to infer changes in upper water structure and monsoon intensity (e.g., Ganssen and Troelstra, 1987; Farrell et al., 1995; Chaisson and Ravelo, 1997; Huang et al., 2003; Wei et al., 2003; Rohling et al., 2004). Large scale ocean circulation changes (e.g., Curry et al., 1988; Duplessy et al., 1988; Sarnthein et al., 1994), variations in the carbon cycle and the strength of the biological pumping of carbon in the ocean (e.g., Ganssen and Sarnthein, 1983; Mortlock et al., 1991) can be deduced from variations in d 13C in planktonic and benthic foraminifera. Because even small changes in the isotope ratios may imply significant changes in the environmental system, it is necessary to measure the isotopic signals with the highest possible precision and to exclude any sources of error. Variations caused by different morphotypes of the same species having different preferences for different habitats (depths) and variations due the ontogenetic development can be assessed by strict morphometric criteria when selecting the tests and by the employment of narrow size fractions. The fact that different species show a slight offset in their fractionation compared to what would be expected if they calcified in equilibrium with sea water, the so called vital effect, has been assessed through sediment trap and cultivation experiments where the offset of the specific species is measured (e.g., Erez and Honjo, 1981; BouvierSoumagnac and Duplessy, 1985; Deuser, 1987). Several studies have casually addressed the influence of different chemical treatments on the stable oxygen and carbon isotope measurements of carbonate materials (Duplessy, 1978, and references therein). For example, Emiliani (1966) noted a lowering of up to 0.8x in the y18O of ground and Helium roasted calcite from Tridacna shells, and a 0.2x change in y18O of the planktonic foraminifera Globigerinoides sacculifer due to oxidization with NaClO. A similar change was observed by Savin and Douglas (1973) in Recent planktonic foraminifera (y18O-0.29x, y13C-0.2x) after exposure to NaClO. Particular attention has also been given to the effect of different cleaning protocols

on the studies of aminoacids in foraminiferal organic matter (Katz and Man, 1979) and Mg/Ca ratios in the calcite shells (e.g., Martin and Lea, 2002). Furthermore, the storage of foraminiferal tests in formalin solution resulted in a significant change in the stable isotope values (Ganssen, 1981), and the comparison of foraminifera ultrasonified in alcohol with untreated ones displayed significantly heavier y18O and y13C values, attributed to the removal of coccolith dust from the tests (Voelker, 1999). However, to our knowledge, no published study has systematically addressed the impact of the most commonly used cleaning methods on the stable isotope composition of foraminiferal shells. The purpose of this study is to test whether the most commonly used cleaning methods have any influence on the stable isotope values of planktonic foraminifera, or if the cleaning is a pointless step, actually increasing the risk of introducing errors. We also compare the stable isotope values of different morphotypes/sizes of Globigerinoides ruber, since recent studies have given somewhat disparate results (S. Steinke, pers. com., Lin et al., 2004) on the two morphotypes distinguished by Wang (2000). A third aim is to assess the isotopic variability of planktonic foraminifera in surface sediments caused by seasonal variations in surface water conditions and different calcifying seasons.

2. Materials and method 2.1. Location and hydrography The two box cores M1 (119827.98VE, 21825VN, 2993 m water depth) and F (118835.03VE, 20814.97VN, 2735 m water depth) used in this study were taken from the north-easternmost South China Sea during an Ocean Researcher I cruise in 2004 (Fig. 1). The sediment at these locations consists of hemipelagic muds. The surface water of the South China Sea is characterized by the inflow of saline Western Philippine Water through the Luzon strait that is mixed with fresh river water from the surrounding land areas (Wyrtki, 1961). The surface circulation in the basin is controlled by the strong northeast monsoon driving a cyclonic gyre over the whole basin during winter and

L. Lo¨wemark et al. / Marine Micropaleontology 55 (2005) 49–62

116°

120°

118°

122°

51

124°

24°

22° 200m

20°

1000m

M1 F

2000m

3000m km

0

50 100

4000m

18° Fig. 1. Bathymetric chart showing the locations of the box cores in the north-eastern South China Sea.

the weaker southwest monsoon driving an anticyclonic gyre, primarily in the southern part of the basin, during summer (Wyrtki, 1961; Liang et al., 2000). In the northernmost part of the South China Sea the surface hydrography is especially complicated due to the mixing of several distinctly different water masses. In winter, warm saline Kuroshio-derived water enters through the Bashi Strait, this water is mixed with cold surface water entering from the Taiwan Strait (Wyrtki, 1961). In summer, the hydrography is dominated by warm waters from the southern South China Sea (Shaw and Chao, 1994). Summer conditions are usually oligotrophic and typhoons mixing the upper surface can have a large impact generating short term blooming events (Liu and Liu, 2002). In the northern South China Sea a sea surface temperature difference of 5–6 degrees between winter (~ 23 8C) and summer (~ 29 8C) is present (Levitus and Boyer, 1994). 2.2. Sample preparation The sample material was dispersed in distilled water, and wet sieved with a 63 Am sieve until no

more fine material was released. The sieved material was ultrasonified in tap water for a maximum of 30 s in order to disperse clay aggregates and to loosen adhering clay particles. The sample was wet sieved again to remove the loosened material. The sediment was flushed onto filter paper and dried at 45 8C overnight. The dried sample was dry sieved to separate the size fractions 63–250 Am, 250–350 Am, and N 350 Am. Only the size fraction 250–350 Am was used for this study. The planktonic foraminifera Globigerinoides ruber was picked out under microscope, and separated into three types according to their morphology (Fig. 2). Morphotype I corresponds closely to G. ruber sensu stricto (s.s.) of Wang (2000), whereas morphotype II correspond to G. ruber sensu lato (s.l.) of Wang (2000) and morphotype III correspond to the kummerform of Hecht and Savin (1972) and Hecht (1974). Morphotype II and III are characterized by flattened and minute last chambers, respectively. Each morphotype was subsequently split into two size fractions, 250–297 Am, and 297–350 Am, to minimize the effect of ontogenetic changes in stable isotope composition. Ten repetitive measurements of stable oxygen and carbon isotopes on each

52

L. Lo¨wemark et al. / Marine Micropaleontology 55 (2005) 49–62

Fig. 2. Representative specimens of the three morphotypes of Globigerinoides ruber distinguished in this study. Morphotype I approximately corresponds to the sensu stricto of Wang (2000), whereas morphotype II corresponds to G. ruber sensu lato (s.l.) of Wang (2000) and morphotype III corresponds to the kummerform of Hecht and Savin (1972) and Hecht (1974). Morphotype II is characterized by a flattened and asymmetrical last chambers and morphotype III is characterized by a minute last chamber.

size fraction of the three morphotypes were then performed at the Institute of Earth Sciences, Academia Sinica using a Finnigan MAT252 mass spectrometer with Kiel Device (Thermoquest-Finnigan). The machine precision of the mass spectrometer measurements is F 0.03x for carbon isotopes and F Wet Sieving

0.05x for oxygen isotopes measured on the in-house standard (LB-32). All data are reported in x vs. the PDB-standard. For each stable isotope measurement five foraminiferal tests were used in the size fraction 297–350 Am and 6 tests in the smaller size fraction 250–297 Am.

Dry Sieving

Set A Set B

Foraminifers were separated into five groups

MS

Set C Set D Set E Ul O Cr t ac in a raso Na xidiz lco n k Cl ed ho ic b O w l ath ith

Fig. 3. The sediment was first washed over a sieve with a 63 Am mesh and then dry sieved into two size fractions (250–297 and 297–350 Am) before 5–6 foraminiferal tests were measured on a Finnigan MAT252 mass spectrometer. Additionally, on foraminifera from station F, four different cleaning methods were applied to four subsets from each size fraction of the foraminifera before the isotope measurements.

L. Lo¨wemark et al. / Marine Micropaleontology 55 (2005) 49–62

53

Table 1 Station F Morphotype I

297 –350 Am

Untreated

13

y C

A1 A2 A3 A4 A5 A6 A7 A8 A9 A10 AVG STD

1.48 1.38 1.66 1.32 1.78 1.55 1.43 1.53 1.48 1.31 1.49 0.16

250–297 Am 18

Station F

y O

13

18

y C

y O

-2.84 -3.16 -2.64 -2.48 -2.37 -2.53 -2.69 -3.38 -2.59 -2.94 -2.76 0.34

1.10 1.38 1.39 1.70 1.36 1.14 1.40 1.21 1.42 1.12 1.32 0.18

- 2.81 - 2.90 - 2.45 - 2.20 - 2.92 - 2.13 - 2.62 - 2.13 - 2.41 - 2.89 - 2.54 0.33

Station F

297–350 Am

250– 297 Am

Morphotype II

13

18

y C

y O

y13C

y18O

II1 II2 II3 II4 II5 II6 II7 II8 II9 II10 AVG STD

1.72 1.75 1.71 1.59 1.60 1.49 1.57 1.67 1.52 1.65 1.63 0.06

-1.86 -2.51 -2.01 -2.41 -2.01 -2.04 -2.40 -2.42 -2.59 -2.99 -2.32 0.33

1.27 1.12 1.52 1.24 1.33 1.24 1.27 1.04 1.23 1.17 1.24 0.13

-2.27 -2.41 -2.35 -2.36 -2.34 -2.11 -2.73 -2.58 -2.52 -2.42 -2.41 0.17

Station F

Cleaning method B

13

y C

B1 B2 B3 B4 B5 B6 B7 B8 B9 B10 AVG STD

1.42 1.23 1.19 1.22 1.59 1.62 1.65 1.24 1.19 1.51 1.39 0.21

18

y O

13

18

y C

y O

Morphotype III

y13C

y18O

y13C

y18O

-2.88 -2.68 -2.61 -3.17 -2.92 -3.33 -3.18 -2.85 -2.55 -3.32 -2.95 0.29

1.44 1.29 1.37 1.36 1.18 1.40 1.28 1.56 1.29 1.20 1.34 0.11

- 2.68 - 2.46 - 2.91 - 2.41 - 2.85 - 2.08 - 2.72 - 2.70 - 2.86 - 2.69 - 2.64 0.25

III1 III2 III3 III4 III5 III6 III7 III8 III9 III10 AVG STD

1.53 1.37 1.34 1.34 1.45 1.27 1.49 1.44 1.37 1.47 1.40 0.09

-2.27 -2.92 -2.69 -2.53 -2.84 -2.46 -3.14 -2.62 -2.77 -3.04 -2.69 0.30

0.91 1.22 1.08 1.43 1.29 1.50 1.36 1.41 1.61 1.64 1.35 0.23

-2.77 -2.81 -3.18 -2.57 -2.30 -2.55 -3.26 -3.01 -2.42 -2.67 -2.75 0.32

Station F

Station M1

Cleaning method C

13

y C

C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 AVG STD

18

y O

13

18

y C

y O

Morphotype I

y13C

y18O

y13C

y18O

0.95 1.37 1.32 1.47 1.23 1.59 2.31 0.62 1.16 1.40 1.34 0.51

-2.53 -3.22 -2.90 -2.91 -3.01 -2.90 -2.55 -3.48 -3.11 -3.48 -3.01 0.33

1.24 1.18 1.63 1.29 1.29 1.03 1.58 1.41 1.64 1.47 1.38 0.20

- 2.33 - 3.07 - 3.13 - 2.82 - 2.57 - 2.59 - 2.92 - 3.28 - 2.85 - 2.63 - 2.82 0.29

M1-1 M1-2 M1-3 M1-4 M1-5 M1-6 M1-7 M1-8 M1-9 M1-10 AVG STD

1.52 1.29 1.51 1.66 1.58 1.52 1.56 1.41 1.89 1.62 1.52 0.11

-3.16 -3.10 -3.18 -3.07 -3.29 -3.84 -3.07 -2.94 -3.01 -2.78 -3.24 0.27

1.35 1.54 1.21 1.33 1.48 1.39 1.42 1.60 1.47 1.34 1.39 0.11

-1.93 -2.79 -3.57 -3.05 -3.00 -3.03 -3.28 -2.69 -2.64 -2.95 -2.95 0.51

Cleaning method D

y13C

y18O

y13C

y18O

D1 D2 D3 D4

* * 1.08 1.27

* * -2.91 -2.90

1.07 1.05 1.20 1.30

- 2.40 - 2.70 - 2.22 - 2.91

Station F

(continued on next page)

54

L. Lo¨wemark et al. / Marine Micropaleontology 55 (2005) 49–62

Table 1 (continued) Station F Cleaning method D

y13C

y18O

y13C

y18O

D5 D6 D7 D8 D9 D10 AVG STD

1.50 1.80 1.38 1.20 1.50 1.40 1.39 0.22

- 2.91 - 2.28 - 2.69 - 3.24 - 2.56 - 2.67 - 2.77 0.35

1.08 1.33 1.23 1.13 1.11 0.94 1.14 0.12

-2.60 -4.62 -2.67 -2.94 -2.71 -2.57 -2.83 0.66

Cleaning method E

y13C

y18O

y13C

y18O

E1 E2 E3 E4 E5 E6 E7 E8 E9 E10 AVG STD

1.48 1.48 1.39 1.34 1.02 1.52 1.18 1.20 1.42 1.43 1.35 0.17

- 2.64 - 2.83 - 2.48 - 2.41 - 2.73 - 2.41 - 2.85 - 2.64 - 3.31 - 2.73 - 2.70 0.31

1.02 1.32 1.07 1.37 1.05 1.13 1.26 1.17 1.19 1.04 1.16 0.12

-2.70 -2.79 -2.65 -2.97 -2.53 -2.96 -3.37 -2.51 -2.92 -2.74 -2.81 0.25

Station F

Additionally, on samples from station F, four different cleaning methods were applied upon subsets of the two size fractions of morphotype I. The first cleaning method (set B) consists of cautious ultrasonification in ethanol for 10 s and immediate removal of the ethanol solution in order to remove dispersed material. In the second method (C), the tests were first cracked to expose their inner chambers and then ultrasonified in ethanol. In the third method (D), the tests were oxidized with NaClO for 24 h in order to remove organic material and then ultrasonified in ethanol. In the fourth method (E), the tests were oxidized with NaClO, cracked, and finally ultrasonified in ethanol (Fig. 3). The cleaned foraminifera (10 repetitive measurements of 5 or 6 foraminiferal tests for each set) were then measured on the same mass spectrometer as the untreated tests (Table 1). 2.3. Statistical evaluation We used unpaired Student t-Test to test if there is a significant difference between the untreated and

cleaned foraminifera, between the two size fractions, between the two stations, and between the different morphotypes. The unpaired Student t-Test is suitable in situations where the sample size is small, the two groups of samples are independent, and when the standard deviation of the population is unknown. We used a two-tailed test with the significant level 0.05.

3. Results The distribution of the y18O values in morphotypes I to III is quite large, with values ranging from 1.7x to 3.4x and 2.1x to 3.3x in the larger and smaller size fractions, respectively (Fig. 4a). This results in standard deviations around 0.3x. The averages range from 2.4x to 2.8x. Morphotype 2 has highest y18O values in both size fractions. There is a significant difference (a = 0.05) between morphotype I and II, and between type II and III in size fraction 297–350 Am, and between morphotype II and III in the smaller size fraction. The y13C values of the

L. Lo¨wemark et al. / Marine Micropaleontology 55 (2005) 49–62

55

Different Morphotypes

a -1.5

250-297 µm

297-350 µm

δ18O(permil)

-2.0 II

-2.5

I

II

III

I

III

-3.0

-3.5

b

2.0

297-350 µm

250-297 µm

δ13C(permil)

1.8 1.6

II I

1.4

III

III

I

1.2

II

1.0 0.8

Fig. 4. a) The y18O values of the different morphotypes of G. ruber in site F. The spread of the data points is larger than 1.5x. The averages of morphotype II in both size fractions are heavier than for the other two morphotypes. In size fraction 297–350 Am, morphotype II is significantly different from I and III. In size fraction 250–297 Am, morphotypes II is only significantly different from morphotype III. b) The y13C values of the different morphotypes in site F. The spread is almost 1x, and in size fraction 297–350 Am, there is a significant difference between morphotype II and the two other morphotypes.

different morphotypes also spread in a large range, which is about 1.8x to slightly less than 1x (Fig. 4b). The averages range from 1.6x to 1.3x, with standard deviations not larger than 0.22x. A significant difference was found between morphotypes I and II, and between II and III in the large size fraction. In the smaller size fraction the differences between the three morphotypes were smaller and not statistically significant. The y18O values of the different cleaning methods also spread in a large range, from about 2.0x to 3.5x, and the averages range from 2.5x to 3x (Fig. 5a). Except for the group D in size fraction 250–297 Am, where one extreme outlier is observed at more than 4.5x, the standard deviations are about 0.3x. In y18O, no significant differences between the untreated foraminifera (set A) and the cleaned samples were detected. The range of y13C values for the different cleaning methods is about 1x (1.8x–0.6x), and the averages range from 1.5x to 1.2x (Fig. 5b). The standard deviations generally are

about 0.2x, except for 297–350 Am, set C, where an outlier causes an unusually large standard deviation. We found significant differences between the untreated and cleaning method E in both size fractions and between untreated (A) and method D in the smaller size fraction. This suggests an influence of the oxidation step on the measured y13C values. Comparing the stable oxygen isotopes of the foraminiferal tests from the two different sampling sites F and M1, there is a significant difference in the bigger size fraction. Although we do not find a significant difference in the smaller size fraction, the t-value was very close to the critical value. The y18O averages of the tests from these two sampling sites are 2.8 F 0.4x (F) and 3.2 F 0.3x (M1), respectively, in the size fraction 297–350 Am, and 2.5 F 0.3x (F) and 2.9F0.5x (M1) in the smaller size fraction (Fig. 6a). Thus, the northern station M1 is about 0.4x lighter than station F. In contrast, there is no significant difference between

56

L. Lo¨wemark et al. / Marine Micropaleontology 55 (2005) 49–62

Different Cleaning Methods

a -1.5

δ18O(permil)

-2.5 -3.0

250-297 µm

297-350 µm

-2.0 A

B

D

E

A

B

C

C

D

E

-3.5 -4.0 -4.5 -5.0

b

2.5

δ13C(permil)

297-350 µm

250-297 µm

2.0

1.5

A

B

C

D

E

A

B

C D

E

1.0

0.5 18

Fig. 5. a) The y O values of the different cleaning methods. The spread is large and the averages of the different cleaning methods deviate from the untreated sample by up to 0.3x. However, these differences are not statistically significant. b) The y13C values of the different cleaning methods. The values spread almost 2x. In the smaller size fraction, methods D and E that include an oxidation step have significantly lower y13C values than the untreated sample.

the two sampling sites in their y13C values. The averages range from 1.5x to 1.3x with standard deviations around 0.2x (Fig. 6b).

4. Discussion Globigerinoides ruber is a subtropical, shallow dwelling spinose planktonic foraminiferal species commonly used in paleoceanographic reconstructions. This species lives in the photic zone of the water column, primarily in the upper 50 m and shows little daily vertical migration (Be´, 1977). In subtropical and transitional waters, G. ruber primarily secretes its shell during the summer months, therefore, the y18O of G. ruber in the sedimentary record often is taken to reflect summer surface water conditions (Deuser et al., 1981; Duplessy et al., 1981; Ganssen, 1983). However, sediment trap data from the South China Sea (Wiesner et al., 1996) and the tropical western Pacific (Kawahata et al., 2002)

suggest that the maximum flux of G. ruber is not necessarily bound to the summer season. Evidence from sediment traps in the Saragasso Sea suggests that the stable oxygen isotopes are subject to a vital effect of about 0.20x relative to sea water (Deuser, 1987). Due to the photosynthetic activity of dinoflagellates living in symbiosis with foraminifera, the calcareous tests of G. ruber are enriched in 13 C relative to sea water (Hemleben et al., 1989; Bemis et al., 2000). The repetitive measurements of stable isotopes on Globigerinoides ruber in surface samples from the northern South China Sea show three interesting features. First, there is an unexpectedly large variability in the stable isotope measurements. Second, there is a significant difference between different morphotypes. Third, there is a statistically significant difference in d 13C between the tests exposed to an oxidizing agent and the untreated tests, whereas the other cleaning methods do not have any significant effect.

L. Lo¨wemark et al. / Marine Micropaleontology 55 (2005) 49–62

a

57

Different Stations -1.5

250-297 µm

297-350 µm

δ18O(permil)

-2.0 -2.5

F F M1

-3.0 M1 -3.5 -4.0

b

2.0

297-350 µm

250-297 µm

δ13C(permil)

1.8 1.6 F

M1

1.4 F

M1

1.2 1.0 18

Fig. 6. a) The difference in y O values between the two sampling sites. The averages of site F are higher than site M1 in both size fractions. There is a significant difference between these two sampling sites in the big size fraction, in the small size fraction the difference is large and close to being significant. b) The difference in y13C values between the two sampling sites. The averages of site F are smaller than site M1, but the differences are not statistically significant.

4.1. Stable isotope variability in the surface sediment 4.1.1. Variability in d18O values The range of almost 1.5x observed in the measured y18O values (Fig. 4) corresponds to a temperature difference of 6–7 8C (e.g., Epstein et al., 1953; O’Neil et al., 1969; Shackleton, 1974; Erez and Luz, 1983). This agrees with the reported winter and summer temperatures that range from around 23– 29 8C (Levitus and Boyer, 1994). Because only 5 tests were used for the analysis, the measured value does not necessarily represent a yearly average. Rather, the value of each measurement is the outcome of randomly mixed tests from different seasons. Although it is unlikely to coincidentally pick 5 out of 5 tests with extreme winter or summer values, there is a realistic chance of choosing 5 tests whose average is much closer to typical winter or summer values than yearly average. We therefore believe that in our test of different cleaning methods, where 50 measurements were made in each grainsize fraction, it is reasonable to assume that the lightest and heaviest

isotopes (the outliers) are close approximations of typical (but not maximal) summer and winter temperatures, respectively. This interpretation is supported by sediment trap data from the same region (Lin et al., 2004). The heaviest winter values are slightly heavier than 2x, compared to 1.8x in our surface samples. Summer sediment trap values generally lie around 3.5x, which is about the same as the lightest points from our surface samples. To avoid this kind of spread in the isotope values, larger numbers of foraminiferal tests should be used. The larger the number of tests used for each individual measurement are, the smaller the chance of coincidentally measuring only extreme tests should be and the narrower the spread would be expected to be (Schiffelbein and Hills, 1984). The isotopic signal recorded in the surface sediment can also be further complicated by seasonal variations in foraminiferal productivity. Sediment trap data from the Saragasso Sea show a strong increase in foraminiferal flux during the winter season (Deuser, 1987), which would bias any isotopic signal toward

58

L. Lo¨wemark et al. / Marine Micropaleontology 55 (2005) 49–62

isotopic conditions reflecting that season. In contrast, sediment trap experiments from the western equatorial Pacific do not display such a strong seasonality (Kawahata et al., 2002). However, that study also showed that the connection between phytoplankton productivity and foraminiferal test flux is not as strong as might be expected. Therefore, absolute fluxes of the specific species are needed to accurately assess the influence of variations in foraminiferal flux on the average isotope signals. In the South China Sea the foraminiferal productivity shows a more variable pattern with several peaks during summer as well as winter monsoon regimes (Wiesner et al., 1996). Thus, the average, if sufficient foraminiferal tests are used, could be expected to represent yearly average. There is a significant difference in the average y18O values between the two stations, the northern station M1 being approximately 0.4x lighter than station F, indicating up to two degrees warmer surface waters. Whereas summer sea surface temperatures are largely uniform in the South China Sea, the winter temperatures generally decreases with increasing latitude (Levitus and Boyer, 1994). We attribute the observed discrepancy to the winter time intrusions of warm Kuroshio water (Wyrtki, 1961) that would have a stronger influence on the northern station resulting in lighter y18O-values. 4.1.2. Variability in d13C Values Although the size of the spread observed in y13C from surface sediment Globigerinoides ruber is similar to the one observed in sediment traps from the same region (Lin et al., 2004), there is an offset of about 1x between the two data sets. Whereas most surface sediment values fall between 1 and 2x, the sediment trap data fall between 0 and 1x. The sediment trap data show a clear seasonal variability with high y13C values during the summer months and low values in the winter. Lin et al. (2004) interpreted this as an effect of seasonal variations in surface water productivity. A number of studies have shown a positive correlation between primary productivity and y13C of the phytoplankton (e.g., Deuser, 1970; Sarnthein et al., 1988). The enhanced removal of organic carbon relatively rich in 13C during the winter season results in lighter y13C seawater values and subsequently lighter foraminiferal calcite, and vice versa for the summer months. Shipboard and CZCS-SeaWiFS data confirm high nutrient

levels in the northern South China Sea in winter and low during summer (Liu et al., 2002), corroborating the view that it is the increased primary productivity in the winter season that leads to an isotopically lighter y13Csignal in the planktonic foraminifera. The 1.5x range in y13C observed in the G. ruber from the surface sample therefore is interpreted as the seasonal range of y13C in the seawater, the heaviest values representing summer and the lightest representing winter conditions. The offset of about 1x between our surface sample and the sediment trap data presented by Lin et al. (2004) most likely is due to the Suess effect (Suess, 1965). The combustion of fossil fuels has raised the atmospheric CO2 level from its preindustrial level around 280 ppm and a y13C value of 6.5x (Friedli et al., 1986) to the present 375 ppm with a y13C value around 8x (Whorf and Keeling, 1998; Keeling and Whorf, 2004). This introduction of anthropogenic, isotopically light carbon has been shown to result in an offset of almost 1x between plankton from the water column and core top foraminifera in the Arctic Ocean (Bauch et al., 2000). Variations in seawater carbonate ion concentration (carbonate ion effect) have also been shown to cause changes in the y13C of planktonic foraminifera (Spero et al., 1997; Russell and Spero, 2000). Because the uppermost centimeters of the sediment have been homogenized by bioturbation, the foraminiferal tests sampled from the surface sediment represent a mixture consisting of primarily preindustrial foraminifera and a mixture of foraminifera from different seasons. 4.1.3. Influence of seasonal variations on downcore reconstructions The large variability observed in the y18O and y13C values from the surface samples have some implications to the interpretation of downcore foraminiferal data in paleoceanographic reconstructions when small sample sizes are used. When large sample sizes are used for the stable isotope analysis, the isotope value measured will be close to the true average of the mixed, multi-annual foraminiferal assemblage at that level. Depending on whether planktonic foraminiferal production is even over the year or concentrated to a certain season the measured value will correspond to yearly average or season specific conditions, respectively. When sample sizes decrease, however, the measured value no longer represents the average but a

L. Lo¨wemark et al. / Marine Micropaleontology 55 (2005) 49–62

random value following a normal distribution between two extreme values. Our surface sample data clearly reflect this phenomenon with most of the values relatively close to the average but with a noteworthy portion of the measurements close to the extreme values predicted from sediment trap data. Thus, a large part of the variability observed in downcore records is not due to climatic variability or preparation related errors, but simply reflects the regional seasonal variability. 4.2. Size fractions and ontogenetic effects on d13C and d18O Although the size fraction analyzed was chosen to only include adult stages of Globigerinoides ruber, a significant difference in y13C between the small and large size fractions in morphotypes I and II was observed. The smaller size fractions are about 0.2x and 0.4x lighter than the larger foraminifera in morphotypes I and II, respectively. The photosynthetic activity of the dinoflagellates living in symbiosis with the foraminifera have been shown to cause an enrichment of 13C in the calcite shell relative to sea water y13C (Hemleben et al., 1989; Bemis et al., 2000). This could be interpreted as a difference in vital effect or habitat between morphotypes I/II, and morphotype III. However, no significant difference was observed between the smaller and the larger tests in d 18O values. We therefore speculate that the vital effects of morphotypes I and II are slightly different from morphotype III. In morphotypes I and II the symbiotic activity of the dinoflagellates play a more important role in the larger specimen, i.e. later in the life cycle, resulting in an enrichment in y13C relative to the smaller ones. In morphotype III no such trend was observed. 4.3. Different morphotypes The differences in y18O and y13C between morphotype I and II, and II and III in the larger size fraction seem to corroborate the view of Wang (2000) that there are some differences between the particular morphotypes. The heavier y18O values of morphotype II, corresponding to bsensu latoQ of Wang (2000), also agree with the notion that Globigerinoides ruber s.l. lives at a deeper level than does G. ruber s.s.

59

Although the heavy values in the smaller size fraction are not statistically significant, they still indicate a deeper habitat. Therefore, morphotype II probably calcifies in sub-surface waters of 30–50 m depth as suggested by Wang (2000). CTD studies from the northern South China Sea show a 2–3 8C decrease and a 0.20–0.25 psu salinity increase, resulting in about 0.4x heavier y18O values (Wang, 2000), in agreement with the ca 0.5x difference observed in our data. The larger size fraction of morphotype II also differs from morphotypes I and III in y13C, possibly indicating a different habitat. We cannot exclude, however, that the differences observed are due to differences in vital effect or in preferential calcification season. The lack of a statistically significant difference in the stable isotopes of the small size fraction might indicate that the observed differences in the larger foraminifera are related to the ontogeny of the foraminifera and that differences are smaller in earlier stages of the life cycle. Wang (2000) used the somewhat larger size fraction 315–400 Am for his study, thus no comparative information about the smaller size fraction is available. 4.4. The effect of different cleaning methods on stable isotope values The first two cleaning methods, cleaning in ultrasonic bath, and cracking combined with cleaning in ultrasonic bath were not statistically significant different from the untreated foraminifera. This suggests that if the sieving procedure was performed adequately and the foraminiferal tests look clean under optical microscope, i.e., no material visible in the apertures, then a second cleaning through cracking and/or ultrasonification in alcohol is not necessary. Presumably, if no contaminating material is visible in the apertures or adhering to the foraminiferal shell, then the potential level of contamination is small and the risk of losing material or introducing an error during the different cleaning steps probably outweighs the presumed benefits of the cleaning. Especially the cracking step, where primarily the outer chambers are cracked open, may result in the loss of carbonate material that is sucked out together with any dispersed coccolith material. Because the early chambers secreted during juvenile and neanic stages were produced under different

60

L. Lo¨wemark et al. / Marine Micropaleontology 55 (2005) 49–62

conditions (diet and probably also vital effect changed during the ontogeny of the foraminifera (Hemleben et al., 1989)), the loss of calcite from the outer chambers may lead to an unwanted shift in stable isotope values toward values representative of earlier stages of the ontogeny. In contrast, the two cleaning methods involving oxidation of the foraminiferal tests with sodium hypochlorite show an offset in y13C of about 0.4x to the untreated foraminifera. This offset is statistically significant (95% confidence) in the smaller size fraction and in set E (oxidation, cracking, and ultrasonification) of the larger size fraction, but not significant in set D (297–350 Am, oxidation and ultrasonification). No significant difference in y18O was observed between the untreated and the oxidized foraminifera. The negative offset is a surprise. Because the oxidation step is introduced in order to remove any remaining organic material and marine organic matter generally has y13Corg values around 20x, the removal of any remaining organic matter would be expected to cause a positive shift in the y13C. In order to better determine the effect of the different cleaning methods additional experiments performed on planktonic foraminifera from a region with minimal seasonal difference in surface water conditions are needed. This would reduce the variability caused by the mixing of foraminifera that calcified under different seasons.

5. Conclusions The large spread of both y18O and y13C in the measured surface samples show that when small samples are used, the obtained values do not necessarily represent yearly average or a certain season. Rather, the values will randomly fall somewhere between seasonal maxima and minima. The measured values will approximately follow a normal distribution with most values close to the average but with a not neglectable portion of the measurement falling close to the extremes. The smaller the number of foraminiferal tests used, the larger the chance of values significantly deviating from the average. The measured differences between different morphotypes corroborate the view of Wang (2000) that Globigerinoides ruber sensu lato (our morphotype II)

calcifies at a larger depth during its adult stage than does G. ruber sensu stricto (our morphotype I). Finally, our experiment with applying different cleaning methods prior to stable isotope analysis shows that for clean samples, the commonly applied ultrasonification in alcohol (sometimes combined with a cracking of the outer chambers) does not have any effect on neither y18O nor y13C values. In contrast, the steps involving oxidation of the calcite shells with sodium hypochlorite resulted in a statistically significant lowering of the y13 C values. Unless the foraminiferal tests are visibly contaminated by coccoliths or adhering clay, cracking and/or ultrasonification are unnecessary steps. However, a more detailed study ought to be performed on surface samples from a region characterized by minimal seasonal variations in order to allow a better quantification of the potential effects of different cleaning methods. Acknowledgements Special thanks to Yoshiyuki Iizuka (IESAS) for the SEM pictures and Rosa Cheng (IESAS) for operating the mass spectrometer. We thank Stephan Steinke (Bremen University) for valuable discussions and for his constructive help with an earlier version of this paper. Mark Maslin (University College London) and Gerald Ganssen (Vrije Universiteit Amsterdam) are cordially thanked for their constructive reviews of a previous version of this paper. We thankfully appreciate economic support by the APEC-program and Academia Sinica, Taiwan. References Bard, E., Fairbanks, R., Arnold, M., Maurice, P., Duprat, J., Moyes, J., Duplessy, J.-C., 1989. Sea-level estimates during the last deglaciation based on d 18O and accelerator mass spectrometry 14 C ages measured in Globigerina bulloides. Quaternary Research, 31, 381 – 391. Bauch, D., Carstens, J., Wefer, G., Thiede, J., 2000. The imprint of anthropogenic CO2 in the Arctic Ocean: evidence from planktic y13C data from watercolumn and sediment surfaces. Deep-Sea Research. Part 2. Topical Studies in Oceanography, 47, 1791 – 1808. Be´, A.W.H., 1977. An Ecological, Zoogeographic and Taxonomic Review of Recent Planktonic Foraminifera. In: Raysay, A.T.S. (Ed.), Oceanic Micropalaeontology, vol. 1. Academic Press, London, pp. 1 – 100.

L. Lo¨wemark et al. / Marine Micropaleontology 55 (2005) 49–62 Bemis, B.E., Spero, H.J., Lea, D.W., Bijma, J., 2000. Temperature influence on the carbon isotopic composition of Globigerina bulloides and Orbulina universa (planktonic foraminifera). Marine Micropaleontology, 38, 213 – 228. Bouvier-Soumagnac, Y., Duplessy, J.-C., 1985. Carbon and oxygen isotopic composition of planktonic Foraminifera from laboratory culture, plankton tows and recent sediment: implications for the reconstruction of paleoclimatic conditions and of the global carbon cycle. Journal of Foraminiferal Research, 15 (4), 302 – 320. Chaisson, W.P., Ravelo, A.C., 1997. Changes in upper watercolumn structure at site 925, late Miocene–Pleistocene: planktonic foraminifer assemblage and isotopic evidence. In: Shackleton, N.J., Curry, W.B., Richter, C., Bralower, T.J. (Eds.), Proceedings of the Ocean Drilling Program, Scientific Results, vol. 154, pp. 255 – 268. Curry, W.B., Duplessy, J.C., Labeyrie, L.D., Shackleton, N.J., 1988. Changes in the distribution of 13C of deep water ACO2 between the last glaciation and the Holocene. Paleoceanography, 3, 327 – 337. Deuser, W.G., 1970. Isotopic evidence for diminishing supply of available carbon during diatom bloom in the Black Sea. Nature, 225, 1069 – 1071. Deuser, W.G., 1987. Seasonal variations in isotopic composition and deep-water fluxes of the tests of perennially abundant planktonic Foraminifera of the Saragasso Sea: results from sediment-trap collections and their paleoceanographic significance. Journal of Foraminiferal Research, 17 (1), 14 – 27. Deuser, W.G., Hemleben, C., Spindler, M., 1981. Seasonal changes in species composition, numbers, mass, size, and isotopic composition of planktonic foraminifera settling into the deep Saragasso Sea. Palaeogeography, Palaeoclimatology, Palaeoecology, 33, 103 – 128. Duplessy, J.C., 1978. Isotope studies. In: Gribbin, J. (Ed.), Climatic Change. Cambridge University Press, London, pp. 44 – 67. Duplessy, J.C., Be´, A.W.H., Blanc, P.L., 1981. Oxygen and carbon isotopic composition and biogeographic distribution of planktonic foraminifera in the Indian Ocean. Palaeogeography, Palaeoclimatology, Palaeoecology, 33, 9 – 46. Duplessy, J.C., Shackelton, N.J., Fairbanks, R.G., Labeyrie, L., Oppo, D., Kallel, N., 1988. Deepwater source variations during the last climatic cycle and their impact on the global deepwater circulation. Paleoceanography, 3 (3), 343 – 360. Duplessy, J.C., Labeyrie, L., Arnold, M., Paterne, M., Duprat, J., van Weering, T.C.E., 1992. Changes in surface water salinity of the North Atlantic Ocean during the last deglaciation. Nature, 358, 485 – 488. Emiliani, C., 1955. Pleistocene temperatures. Journal of Geology, 63, 538 – 578. Emiliani, C., 1966. Paleotemperature analysis of Caribbean cores P6304-8 and P6304-9 and a generalized temperature curve for the past 425,000 years. Journal of Geology, 74 (2), 109 – 126. Epstein, S., Buchsbaum, R., Lowenstam, H., Urey, H., 1953. Revised carbonate-water isotope temperature scale. Geological Society of America Bulletin, 64, 1315 – 1325. Erez, J., Honjo, S., 1981. Comparison of isotopic composition of planktonic foraminifera in plankton tows, sediment traps and

61

sediments. Palaeogeography, Palaeoclimatology, Palaeoecology, 33, 129 – 156. Erez, J., Luz, B., 1983. Experimental paleotemperature equation for planktonic foraminifera. Geochimica et Cosmochimica Acta, 47, 1025 – 1031. Farrell, J.W., Murray, D.W., McKenna, V.S., Ravelo, A.C., 1995. Upper ocean temperature and nutrient contrasts inferred from Pleistocene planktonic foraminifer y18O and y13C in the eastern Equatorial Pacific. In: Pisias, N.G., Mayer, L.A., Janecek, T.R., Palmer-Julson, A., van Andel T.H. (Eds.), Proceedings of the Ocean Drilling Program, Scientific Results, 138, pp. 289 – 311. ¨ schger, H., Siegenthaler, U., Stauffer, B., Friedli, H., Lftscher, H., O 1986. Ice core record of the 13C/12C ratio of atmospheric CO2 in the past two centuries. Nature, 324, 237 – 238. Ganssen, G., 1981. Isotopic analysis of foraminifera shells: interference from chemical treatment. Palaeogeography, Palaeoclimatology, Palaeoecology, 33, 271 – 276. Ganssen, G., 1983. Dokumentation von kqstennahem Auftrieb anhand stabiler Isotope in rezenten Foraminiferen vor Nordwest-Afrika. Meteor-Forschungsergebnisse. Reihe C, Geologie und Geophysik, 37, 1 – 46. Ganssen, G., Sarnthein, M., 1983. Stable isotope composition of foraminifera: the surface and bottom waters record of coastal upwelling. In: Suess, A.E., Tiede, J. (Eds.), Coastal Upwelling, Its Sediment Record, Part A. Plenum, New York, pp. 99 – 121. Ganssen, G., Troelstra, S.R., 1987. Palaeoenvironmental changes from stable isotopes in planktonic foraminifera from Eastern Mediterranean sapropels. Marine Geology, 75, 221 – 230. Hays, J.D., Imbrie, J., Shackelton, N.J., 1976. Variations in the earth’s orbit: pacemaker of the ice ages. Science, 194 (4270), 1121 – 1132. Hecht, A.D., 1974. Intraspecific variation in recent populations of Globigerinoides ruber and Globigerinoides trilobus and their application to paleoenvironmental analysis. Journal of Paleontology, 48 (6), 1217 – 1234. Hecht, A.D., Savin, S.M., 1972. Phenotypic variation and oxygen isotope ratios in recent planktonic foraminifera. Journal of Foraminiferal Research, 2 (2), 55 – 67. Hemleben, C., Spindler, M., Anderson, O.R., 1989. Modern Planktonic Foraminifera. Springer-Verlag. 335 pp. Huang, B., Cheng, X., Jian, Z., Wang, P., 2003. Response of upper ocean structure to the initiation of the North Hemisphere glaciation in the South China Sea. Palaeogeography, Palaeoclimatology, Palaeoecology, 196, 305 – 318. Katz, B.J., Man, E.H., 1979. Effects of ultrasonic cleaning on the amino acid geochemistry of foraminifera tests. Geochimica et Cosmochimica Acta, 43 (9), 1567 – 1570. Kawahata, H., Nishimura, A., Gagan, M.K., 2002. Seasonal change in foraminiferal production in the western equatorial Pacifc warm pool: evidence from sediment trap experiments. DeepSea Research. Part 2. Topical Studies in Oceanography, 49, 2783 – 2800. Keeling, C.D., Whorf, T.P., 2004. Atmospheric CO2 Records from Sites in the SIO Air Sampling Network, Trends: A Compendium of Data on Global Change. Carbon Dioxide Information Analysis Center. Oak Ridge National Laboratory, U.S. Department of Energy, Oak Ridge, Tenn., USA.

62

L. Lo¨wemark et al. / Marine Micropaleontology 55 (2005) 49–62

Lea, D.W., Pak, D.K., Spero, H.J., 2003. Sea surface temperatures in the western equatorial Pacific during marine isotope stage 11. In: Droxler, A., Poore, R., Burckle, L. (Eds.), Earth’s Climate and Orbital Eccentricity: The Marine Isotope Stage 11 Question, Geophysical Monograph Series, vol. 137. AGU, Washington, DC, pp. 147 – 156. Levitus, S., Boyer, T., 1994. World Ocean Atlas Volume 4: temperature. NOAA Atlas NESDISUS Government Department of Commerce, Printing Office, Washington, DC. 117 pp. Liang, W.-D., Jan, J.-C., Tang, T.-Y., 2000. Climatological wind and upper ocean heat content in the South China Sea. Acta Oceanographica Taiwanica, 38, 91 – 114. Lin, H.-L., Wang, W.-C., Hung, G.-W., 2004. Seasonal variation of planktonic foraminiferal isotopic composition from sediment traps in the South China Sea. Marine Micropaleontology, 53, 447 – 460. Liu, K.-K., Liu, C.-S., 2002. National Center for Ocean Research, Progress Report 2002, Taipei. Liu, K.-K., Chao, S.-Y., Shaw, P.-T., Gong, G.-C., Chen, C.-C., Tang, T.Y., 2002. Monsoon-forced chlorophyll distribution and primary production in the South China Sea: observations and a numerical study. Deep-Sea Research. Part 1. Oceanographic Research Papers, 49, 1387 – 1412. Martin, P.A., Lea, D.W., 2002. A simple evaluation of cleaning procedures on fossil benthic foraminiferal Mg/Ca. Geochemistry, Geophysics, Geosystems, 3 (8401). Maslin, M.A., Shackleton, N.J., Pflaumann, U., 1995. Surface water temperature, salinity and density changes in the Northeast Atlantic during the last 450,000 years: heinrich events, deep water formation and climatic rebound. Paleoceanography, 10 (3), 527 – 544. Mortlock, R.A., Charles, C.D., Froelich, P.N., Zibello, M.A., Saltzman, J., Hays, J.D., Burckle, L.H., 1991. Evidence for lower productivity in the Antarctic Ocean during the last glaciation. Nature, 351, 220 – 223. O’Neil, J., Clayton, R., Mayeda, T., 1969. Oxygen isotope fractionation in divalent metal carbonates. Journal of Chemical Physics, 51 (12), 5547 – 5558. Rohling, E.J., 2000. Paleosalinity: confidence limits and future applications. Marine Geology, 163, 1 – 11. Rohling, E.J., Sprovieri, M., Cane, T., Casford, J.S.L., Cooke, S., Bouloubassi, I., Emeis, K.C., Schiebel, R., Rogerson, M., Hayes i, A., Jorissen, F.J., Kroon, D., 2004. Reconstructing past planktic foraminiferal habitats using stable isotope data: a case history for Mediterranean sapropel S5. Marine Micropaleontology, 50, 89 – 123. Russell, A.D., Spero, H.J., 2000. Field examination of the oceanic carbonate ion effect on stable isotopes in planktonic foraminifera. Paleoceanography, 15 (1), 43 – 52. Sarnthein, M., Winn, K., Duplessy, J.-C., Fontugne, M.R., 1988. Global variations of surface ocean productivity in low and mid latitudes: influence on CO2 reservoirs of the deep ocean and atmosphere during the last 21,000 years. Paleoceanography, 3 (3), 361 – 399. Sarnthein, M., Winn, K., Jung, S.J.A., Duplessy, J.-C., Labeyrie, L., Erlenkeuser, H., Ganssen, G., 1994. Changes in east Atlantic

deepwater circulation over the last 30,000 years: eight time slice reconstructions. Paleoceanography, 9 (2), 209 – 267. Savin, S.M., Douglas, R.G., 1973. Stable isotope and magnesium geochemistry of recent planktonic foraminifera from the South Pacific. Geological Society of America Bulletin, 84, 2327 – 2342. Schiffelbein, P., Hills, S., 1984. Direct assessment of stable isotope variability in planktonic foraminifer populations. Palaeogeography, Palaeoclimatology, Palaeoecology, 48, 197 – 213. Shackleton, N.J., 1967. Oxygen isotope analyses and Pleistocene temperatures re-assessed. Nature, 215, 15 – 17. Shackleton, N.J., 1974. Attainment of isotopic equilibrium between ocean water and the benthic foraminifera Genus Uvigerina: isotope changes in the ocean during the last glacial. Les Me´thodes quantitatives d’e´tude des variations due climat au cours du Pleistocene, Colloques Internationaux de Central National de la Recherce Scientifique Paris, 219 pp. Shackleton, N.J., 1987. Oxygen isotopes, ice volume and sea level. Quaternary Science Reviews, 6 (3-4), 183 – 190. Shaw, P.-T., Chao, S.-Y., 1994. Surface circulation in the South China Sea. Deep-Sea Research, 41 (11/12), 1663 – 1683. Spero, H.J., Bijma, J., Lea, D.W., Bemis, B.E., 1997. Effect of seawater carbonate concentration on foraminiferal carbon and oxygen isotopes. Nature, 390 (6659), 497 – 500. Suess, H.E., 1965. Secular variations of the cosmic-ray produced carbon-14 in the atmosphere and their interpretations. Journal of Geophysical Research, 70, 5937 – 5952. Voelker, A.H.L., 1999. Zur deutung der dansgaard-oeschger ereignisse in ultra-hochauflfsenden Sedimentprofilen aus dem europ7ischen nordmeer. Berichte-Reports, Institut fqr Geowissenschaften, Christian-Albrechts-Universit7t zu Kiel, 9, 278. Wang, L., 2000. Isotopic signals in two morphotypes of Globigerinoides ruber (white) from the South China Sea: implications for monsoon climate change during the last glacial cycle. Palaeogeography, Palaeoclimatology, Palaeoecology, 161, 381 – 394. Wang, P., Wang, L., Bian, Y., Jian, Z., 1995. Late Quaternary paleoceanography of the South China Sea: surface circulation and carbonate cycles. Marine Geology, 127, 145 – 165. Weaver, P.P.E., Neil, H., Carter, L., 1997. Sea surface temperature estimates from the Southwest Pacific based on planktonic foraminifera and oxygen isotopes. Palaeogeography, Palaeoclimatology, Palaeoecology, 131 (3-4), 241 – 256. Wei, K.-Y., Chiu, T.-C., Chen, Y.-G., 2003. Toward establishing a maritime proxy record of the East Asian summer monsoons for the late Quaternary. Marine Geology, 201, 67 – 79. Whorf, T.P., Keeling, C.D., 1998. Rising carbon. New Scientist, 157 (2124), 54. Wiesner, M.G., Zheng, L., Wong, H.K., Wang, Y., Chen, W., 1996. Fluxes of particulate matter in the South China Sea. In: Ittekot, V., Sch7fer, P., Honjo, S., Depetris, P.J. (Eds.), Particle Fluxes in the Ocean. Wiley, New York, pp. 293 – 312. Wyrtki, K., 1961. Physical oceanography of the south-east Asian waters. NAGA Report Vol. 2, Scientific Results of Marine Investigations of the South China Sea and the Gulf of Thailand. Scripps Institution of Oceanography, La Jolla, CA. 195 pp.