A high-resolution history of vegetation and climate ...

Science in China Series D: Earth Sciences © 2007

Science in China Press Springer-Verlag

A high-resolution history of vegetation and climate history on Sunda Shelf since the last glaciation WANG XiaoMei1,2†, SUN XiangJun3,1, WANG PinXian1 & Karl STATTEGGER4 1

Key Laboratory of Marine Geology, Tongji University, Shanghai 200092, China; Research Institute of Petroleum Exploration and Development, Beijing 100083, China; 3 Institute of Botany, Chinese Academy of Sciences, Beijing 100093, China; 4 Institute of Geosciences, University of Kiel, Olshausenstrasse 40, Kiel, 24118, Germany 2

This paper presents 16500-year-long high-resolution pollen and spore records from sediments of core 18287 on the continental slope of the southern South China Sea. In the period of 16.5―13.9 ka B.P., the low-mountain rainforest dominated the continental slope of the South China Sea. And in the period of 13.9―10.2 ka B.P., the lowland rainforest and ferns expanded greatly, while the low-mountain rainforest shrank, which indicated a warming at the last deglaciation. Also during this period, the pollen sedimentation rates reduced sufficiently. This might imply a rise of the sea level and therefore the submergence of the shelf, resulting in the broadening of the distance between the source area and the slope. After 10.2 ka B.P, decreasing of the fern indicates the early Holocene (10.2―7 ka B.P.) is a cold period, while the increasing of fern marks the rising temperature (7―3.6 ka B.P.). Sunda Slope, pollen and spores, vegetation, last deglaciation, Holocene.

During the last deglaciation, 15―10 ka B.P., enormous changes occurred on the global surface. The ice caps of the Northern Hemisphere melt and sea surface level rose about 120 m, resulting in the deluge of wide range of continental shelf[1]. China has wide continental shelf, whose changes in the last deglaciation has enormous compact. Sunda Shelf, the second largest continental shelf in the world, 1.8×106 km2, emerged during the glacial times owing to low sea surface level. SONNE cruise 115 (December 1996―January 1997) got a lot of high resolution sedimental records[2 7] along a transect across the Sunda Shelf covering the last 50 000 years. But whether the exposed shelf was covered with vegetation or with what vegetation is still an unanswered question, which may be important to our understanding of the past climate changes in the tropics. Li and Sun[8,9] assigned southern continental shelves of the SCS to lowland rainforested land which reflects a humid climate. While the special position of Sunda Shelf was also poin―

www.scichina.com

www.springerlink.com

ted as a dry climate[10]. So the main aim of this paper is to reveal vegetation changes in the Sunda Shelf and the surrounding highland based on a high- resolution pollen record from the hemipelagic sediments, and to reconstruct the history of the sea surface and climate in this region.

1

Climate and environmental setting

Sunda Shelf, also called “Great Asian Bank”, is bordered by the Malay Peninsula, Sumatra, Java and Borneo (Figure 1). It is an important channel connecting the South China Sea and the Indian Ocean. During the Last Glacial Maximum, Borneo (up to 4101 m), Java (up to 3676 m) and Sumatra (up to 4101 m) were connected together to become the “Sunda Land”[11,12]. The Sunda Received April 6, 2006; accepted August 31, 2006 doi: 10.1007/s11430-007-2067-4 † Corresponding author (email: [email protected]) Suppored by the National Key Basic Development Research Program (Grant No. G2000078500)

Sci China Ser D-Earth Sci | January 2007 | vol. 50 | no. 1 | 75-80

μm mesh. Pollen influx was calculated based on the number of the exotic pollen grains. More than 150 terrestrial pollen grains were counted for each sample (i.e. excluding fern spores and pollen of aquatics) . The percentages of each taxa were calculated based on the total terrestrial pollen grains. The chronology of the core was based on six 14C dates measured in Kiel University, and was corrected by an oceanic reservoir age of −400 years[14] and has been calibrated using CALIB 4.1.2[15], that is 10 cm 3570±50 a; 140 cm 8280±50 a; 175 cm 9200±50 a; 288 cm 11680±50 a; 410 cm 14090±50 a; 512 cm 15680±50 a. The average time resolution at Holocene is 135 years and Last Glaciation is 52 years.

3

Figure 1

The position of Core 18287-3.

Land was tectonically stable during the late Quaternary. Today, the slope of the shelf is small from the coast to the outer shelf, and a slight sea level change could cause a large shift of the shore line. A large quantity of terriclastics accumulates in this region, both today and during the Last Glacial Maximum. The precipitation is high and evenly distributed, and the temperature is warm throughout a year (≥3000 mm/a).

2

Materials and methods

A gravity core (GC 18287-3, located at 5°39′N, 110°39′E) was recovered during R/V SONNE cruise 115 during December 1996 to January 1997[2] from the upper continental slope of the Sunda Shelf in the southern SCS (Figure 1). GC 18287-3 is located near the mouth of the glacial Molengraaff or North Sunda River[2,13]. The water depth was 598 m and the core length was 5.66 m. According to PARASOUND sub-bottom profiles[2] and sedimentological analyses, the core consists of undisturbed hemipelagic silty clay. A total of 112 pollen samples, each 10 mL in volume, were collected at internals of 5 cm from this core. All samples were prepared in the pollen lab at Tongji University using hydrochloric and hydrofluoric acids method. To concentrate the pollen, the material remainings were washed through a 7 76

Palaeoecological results

More than 100 pollen types were identified. They could be classified into the following groups: Tropical montane rainforest: Podocarpus, Dacrycarpus, Dacrydium, Phyllocladus and Ericaceae; Low montane rainforest: mainly of Castanopsis, Quercus, Elaeocarpaceae, Myrtaceae, and Theaceae; Tropical lowland rainforest: Trema, Rutaceae, Sapotaceae, Anacardiaceae, Araliaceae, Ilex, Mallotus/Macaranga, Euphorbiaceae, Palmae, Calamus, Meliaceae, Actinidiaceae, Dipterocarpaceae, etc., occurred in a large number of taxa, but with a few pollen grains in each taxon; Mangroves: mainly Phizophora and Sonneratia. The pollen diagram can be divided into 3 zones, based on major changes in the relative abundance of taxa in the diagram (Figures 2 and 3). Zone 1: (5.66―4 m, 16500―13900 a B.P.): Tree pollen is dominant (±75%). Herb pollen, mainly Cyperaceae and Poaceae, is in considerable percentages (±25%). Ferns spores are 25% of total pollen of land seed plants, the lowest quantity in the whole profile. In the tree pollen, the lower montane rainforest has the highest percentages (35%), the lowland rainforest and mangroves are 20% and 15% separately, the upper montane rainforest is in very low proportions (2%). The pollen influx is high (1700 · cm−2 · a−1 on average) and very stable in values (Figure 3) before 14600 a B.P., and decline after 14600 a B.P. The high quantity of the low mountain forest and low quantity of fern indicate a low and humid temperature. Zone 2: (4―2.25 m, 13900―10200 a B.P.): The main feature of this zone is the general increase

WANG XiaoMei et al. Sci China Ser D-Earth Sci | January 2007 | vol. 50 | no. 1 | 75-80

Figure 2 Pollen percentages diagram of Core 18287 from the southern South China Sea.

Figure 3 Pollen influx diagram of Core 18287 from the southern South China Sea.

in tropical lowland rainforest (from 20% to 30%) and decrease in lower montane rainforest (from 40% to 15%). The quantities of mangroves changed from 20% to 15%, even 5% from 13900 a B.P. to 12900 a B.P., then raised to 20% at 11900 a B.P. The quantities of the herbs decrease to 15% from 30% at 12900 a B.P. The

second feature of this zone is the great increase of fern, which rose to 40%―50% of total pollen of land seed plants. Pollen influx values decreased abruptly to 1/10 of Zone 1, at the top of the zone only 200 cm−2 · a−1 on average, which is the second feature of this zone. The increase of tropical lowland rainforest and fern,


77

and the decrease of lower montane rainforest indicate that the temperature is raised abruptly at the melting time. The declining of pollen influx values, mangrove and herb shows that the rising of sea level led to the submergence of Sunda shelf. Therefore, the distance between the studied site and river mouth was longer and little sediment and pollen could have reached our site during this time. This explains why the pollen influx values of the last glaciation were dramatically reduced. Zone 3: (2.25―0 m, 10200―3570 a B.P.): This zone is quite similar to Zone 2 in percentages. The differences are the slightly higher percentages of tropical lowland rainforest and lower percentages of lower montane rainforest. The most striking feature is that the content of fern was low at 10200―7000 a B.P., while after 7000 a B.P., the content rises greatly. Pollen influx values consistently decreased, only 100 cm−2 · a− 1, reached the minimum of the profile. The comparatively low content of fern at 10200― 7000 a B.P. should be in response to the climatic cooling, while the high content at 10200—7000 a B.P. indicates a warmer climate.

4

Discussions

4.1 The vegetation and climate of the last deglaciation It is important to understand the mechanism of pollen dispersal and transportation when we interpret the pollen data from marine sediments because all of the pollen grains and fern spores in the marine sediments come from surrounding lands. According to the study of the modern distribution pattern of the pollen in the surface sediments in the South China Sea, we know that the pollen is mainly carried by river from the south islands ― such as Borneo[16 18]. Because of the long distance, pollen deposited in the modern South China Sea is very little. While at the last glacial maximum, the sea level dropped to 115―120 m below the present level, and Sunda shelf subaerially exposed.The South China Sea received large mounts of pollen if there was vegetation on the subaerial land. In this profile, the pollen concentration or pollen influx was higher because the source area was closer to the studied area. More pollen was from the subaerial land. Just after the last glacial maximum (16500―13900 a B.P.), pollen influx was high. While at 13900 a B.P., pollen influx reduced to only 78

1/10 of that at 16500 aB.P., which was an indication of rising of sea level and submergence of the Sunda land. At the beginning of the ice melting, pollen in the studied areas mostly came from the Sunda land, and Sunda land was covered by tropical lowland rainforest and lower montane rainforest. The high content of lower montane rainforest and low content of fern indicates the low temperature just after the LGM. But there was no evidence of dryness. This phenomenon also occurs in the core 17962 and 17964[8,16]. Temperature falling has also been registered in West Java[19], north Sumatran[20] and other tropical areas[8,16,21]. But from the pollen data in Indonesia archipelago, north Australia, South America ― and Africa[22 28], we know that there was grassland in the last glacial maximum, indicating the dropping of temperature and rainfall. Li et al.[8] considered the strong winter monsoon absorbed much vapor when it passed the sea, so brought much rainfall to the southern South China Sea, which explains the different rainfall in low latitude. 4.2 The rapid climate shift event at 14 ka At 13900 a B.P., the lowland rainforest and fern expanded abruptly, and the lower mountain forest increased dramatically, which reflected the rising temperature. However, δ18O of G. ruber from the same core decreased from −1.2‰ to −2.9‰ at 14400―14500 a B.P., which suggests sea surface temperature increase[29]. Radiocarbon and gray-scale data of foraminifer from the Cariaco basin of Southern America also showed that sea surface temperature rose at 14400 a B.P.[30]. Accordingly, we think that the change of terrestrial vegetation lagged behind the abrupt climatic change by 500―600 years. This abrupt warming event should correspond to the Bølling-Allerød warming, the most prominent warming phase at the end of the last glaciations, While there was no obvious evidence of the near-glacial conditions, the so-called Younger Dryas, based on the pollen and spore record, although this temperature falling happened in the Tillla Lake in Nigeria, Africa[31,32] and Abiyata Lake of East Africa[33]. 4.3 Climatic shifts events in early and middle Holocene From the great decrease content of fern at 10200―7000 a B.P., we presume that somewhat low temperature climate occurred. At this time the obvious decreasing of Cyperaceae and increasing of Poaceae should indicate


the dry environment. So the climate can be interpreted as cold and dry. The phenomenon also occurred in other low latitude areas, for example, the temperature fell 1― 2℃ from 8700 a B.P. to 6500 a B.P. in the Titicaca Lake, ― west of South America[34 38], also the climate became ― drier at 8000―4000 a B.P.[39,35 37]. At 7000―3600 a B.P., fern spore rose dramatically, which should mean a warm phase. This cold and warm event may be influenced by the transition of Kuroshio, when it was strong, the sea surface temperature of Okinawa Tough is high, and on the contrary, the sea surface temperature is low. So influenced by weak intensity at 9400―6400 a B.P., the temperature of Sunda shelf fell. When Kuroshio came back to Okinawa Tough[40], the temperature rise again. This rising temperature at middle Holocene was also recorded in the Titicaca Lake[39], ― north Bolivia[41] and Peru[42 44].

Core 18287 from 16500 a B.P. to 3000 a B.P., we can conclude: (1) At the last maximum, the Sunda land was covered by the lowland rainforest and lower mountain rainforest, mangrove grew along the coast. The low content of fern and high content of the lowland rainforest indicate low temperature. (2) The content of the lowland rainforest and fern increased abruptly, while the low mountain forest decreased at 13900 a B.P., which indicates that a warm temperature occurred. The greatly decreasing pollen influx at that time means that Sunda Shelf was flooded, so the source area became far. (3) The climate was cold from 10200 a B.P. to 7000 a

5

We thank Institute of Geosciences, University of Kiel for providing samples and Wu Guoxuan for his help in handling samples, also thank Weng Chengyu for his helpful comments on the manuscript.

Conclusions

From the high-resolution of palynological records in 1

took place.

Lambeck K, Chappell J. Sea level change through the last glacial cycle. Science, 2001, 292: 679―686[DOI]

2

B.P., at 7000—3600 a B.P. a warm climatic conditions

from Station 17962. Acta Bot Sin, 2002, 44(6): 746―752 10

Stattegger K, Kuhnt W, Wong H K, et al. Cruise ReportSONNE115

Indo-Pacific Warm Pool and adjacent land at the Last Glacial

SUNDAFLUT. Sequence Stratigraphy, Late Pleistocene-Holocene Sea Level Fluctuations and High Resolution Record of the Post-

Maximum. Glob Planetary Change, 2002, 35: 25―35[DOI] 11

Flenley J R. Problems of the Quaternary on mountains of the

12

Tjia H D. The Sunda Shellf, Southeast Asia. Z Geomorph N E, 1980,

13

Molengraaff G A F. Modern deep-sea research in the East Indian

Sunda-Sahul region. Quat Sci Rev, 1996, 15: 549―555[DOI]

Pleistocene Transgression on the Sunda Shelf. Berichte-Reports. Geol.- Pal¨aont. Inst. Univ. Kiel, 1997, 86: 211 3

Hanebuth T, Stattegger K, Grootes PM. Rapid flooding of the Sunda Shelf—a late-glacial sea-level record. Science, 2000, 288: 1033―

24 (4): 405―427 Archipelago. The Geogr J, 1921, 57: 95―121[DOI]

1035[DOI] 4

Hanebuth T J J, Saito Y, Stattegger K. The stratigraphic architecture

14

Paleoceanography, 1988, 3: 635―645

surveying. Geo-Mar Letters, 2002, 22: 86―94[DOI] Hanebuth T J J, Stattegger K. The stratigraphic evolution of the

15 16

tion distribution. Mar Geol, 1999, 156: 221―226

Special Publications, 2003, 76: 189―200 Hanebuth T J J, Stattegger K, Schimanski A, et al. Late Pleistocene

17

Palaeogeogr Palaerclimat Palaeoecol, 2000, 160: 301―316[DOI]

Geol, 2003, 199: 139―157[DOI] Hanebuth T J J, Stattegger K. Depositional sequences on a late

18

Pleistocene–Holocene tropical siliciclastic shelf (Sunda Shelf, Li X, Sun X J. Palynological records since Last Glacial Maximum

2003, 201: 97―118[DOI] 19

from a deep- sea core in the southern South China Sea. Quat Sci (in Sun X J, Li X, Luo Y L. Vegetation and climate on the Sunda Shelf of the South China Sea during the Last Glaciation pollen results

Van der Kaars S, Dam M A C. A 135000-year record of vegetational and climatic change from the Bandung area, West Java, Indonesia.

Chinese), 1999, 6: 526―535 9

Sun X J, Luo Y L, Huang F, et al. Deep-sea pollen from the South China Sea: Pleistocene indicators of East Asian monsoon. Mar Geol,

southeast Asia). J Asian Earth Sci, 2004, 23 : 113―126[DOI] 8

Sun X J, Li X, Luo Y L, et al. The vegetation and climate at the last glaciation on the emerged continental shelf of the South China Sea.

forced-regressive deposits on the Sunda Shelf (Southeast Asia). Mar 7

Sun X J, Li X, Beug H. Pollen distribution in hemipelagic surface sediments of the South China Sea and its relation to modern vegeta-

Asia and Vicinity-Sedimentol, Stratigra and Petroleum Geol. SEPM 6

Stuiver M, Reimer P J. Extended 14C database and revised CALIB radiocarbon calibration program. Radiocarbon, 1998, 35: 215―230

Sunda Shelf during the past fifty thousand years. In: Sidi F H, Nummedal D, Posamentier H W, et al., eds. Deltas of Southeast

Bard E. Correction of accelerator mass spectrometry 14C ages measured in planktonic foraminifera: Paleoceanographic implications.

of the central Sunda Shelf (SE Asia) recorded by shallow-seismic 5

De Deckker P, Tapper N J, van der Kaars S. The status of the

Palaeogeogr Palaerclimat Palaeoecol, 1995, 117: 55―72[DOI] 20

Newsome J C. Late Quaternary vegetation history of the Central Highland of Sumatra. I. Present vegetation and modern pollen rain.


79

J Biogeogr, 1988, 15: 363―386[DOI] 21

Van der Kaars S. Palynology of eastern Indonesian martine pistoncore: a late Quaternary vegetational and climatic record for Austral-

83[DOI] 33

asia. Palaeogeogr Palaerclimat Palaeoecol, 1991, 85: 239 ―

Abiyata (Ethiopia). Palaeogeogr Palaeoclimat Palaeoecol, 2002, 187:

302[DOI] 22

Van der Kaars S. Marine and terrestrial pollen records of the last glacial cycle from the Indonesian region: Bandung Basin and Banda

259―283[DOI] 34

Dam R A C, Fluin J, Suparan P, et al. Palaeoenvironmental developments in the Lake Tondano area (N. Sulawesi, Indonesia) since

climat Palaeoecol, 2003, 194: 259―279 35

24

Van der Kaars S, Penny D, Tibby J, et al. Late Quaternary pa-

640―643[DOI] 36

and chemical model of Lake Titicaca, Bolivia and Peru. Quat Res,

swamp: Rawa Danau, West Java, Indonesia. Palaeogeogr Palaeo25

Colinvaux P A, De Oliveire P E, Moreno J E, et al. A long pollen

2001, 56: 1―9[DOI] 37

26

Caratini C, Tissot C. Paleogeographical evolution of the Mahakam

aridity in the trop-ical Andes. Geology, 1998, 26: 167―170[DOI] 38

27

Hope G S, Tulip J. A long vegetation history from lowland Iran-

nological Knowledge Boston: Kluwer Academic, 1992, 49―62 39

28

Martinez J, De Deckker P, Barrows T T. Palaeoceanography of the

2003, 3066: 1―21 40

147: 73―99[DOI] 29

(16500―3000 yr B.P.), Quat Res, 2001, 5: 352―362[DOI] 30

1959[DOI]

Palynology, 1998, 100: 39―72[DOI] 32

Salzmann U. Late Quaternary Climate and Vegetation of the Sudanian Zone of Northeast Nigeria. Quat Res, 2002, 58: 73 ―

80

368[DOI] 42

Thompson L G, Mosley-Thompson E, Davis M E, et al. Late glacial stage and Holocene tropical ice core re-cords from Huascaran, Peru. Science, 1995, 269: 46―50[DOI]

43

Thompson L G, Davis M E, Mosley-Thompson E, et al. A 25,000-year tropical climate history from Bolivian ice cores. Sci-

Salzmann U, Waller M. The Holocene vegetational history of the Nigerian Sahel based on multiple pollen profiles. Rev Palaeobot

Graf K. Palynological investigations of two post-glacial peat bogs near the boundary of Bolivia and Peru. J Biogeogr, 1981, 8: 353―

Hughen K A, Eglinton T I, Xu L, et al. Abrupt tropical vegetation response to rapid climate changes. Science, 2004, 304: 1955―

31

532―536 41

Steinke S, Kienast M, Pflaumann U, et al. A High-Resolution Sea-Surface Temperature Record from the Tropical South China Sea

Jian Z M, Yoshiki S, Wang P X, et al. Shifts of the Kuroshio axis over the last 20 000 years. Chin Sci Bull (in Chinese), 1998, 43(5):

last glacial maximum in the eastern Indian Ocean: planktonic foraminiferal evidence. Paleogeogr Paleoclimatol Paleoecolo, 1999,

Paduano G M, A vegetation and fire history of Lake Titicaca since the Last Glacial Maximum. Palaeogeogr Palaeoclimat Palaeoecol,

Jaya,Indonesia. Palaeogeogr Palaeoclimat Palaeoecol, 1994, 109: 385―398[DOI]

Ybert J P. Ancient lake environments as deduced from pollen analysis. In: DeJoux C., ltis A. (Eds.) Lake Titicaca. A Synthesis of Lim-

delta in Kalimantan, Indonesia during the Quaternary and late Pliocene. Rev Palaeobot Palynology, 1988, 55: 217―228[DOI]

Seltzer G O, Cross S, Baker P. High-resolution seismic reflection profiles from Lake Titicaca, Peru/Bolivia. Evidence for Holocene

record from lowland Amazonia: Forest and cooling in glacial times. Science, 1996, 274: 85―88[DOI]

Cross S L, Baker P A, Seltzer G O, et al. Late Quaternary climate and hydrology of tropical South America inferred from an isotopic

laeoecology, palynology and palaeolimnology of a tropical lowland climat Palaeoecol, 2001, 171: 185―212[DOI]

Baker P A, Seltzer G O, Fritz S C. The history of South American tropical precipitation for the past 25,000 years. Science, 2001, 291:

33,000 B.P. Palaeogeogr Palaeoclimat Palaeoecol, 2001, 171: 147― 183[DOI]

Baker P A, Bush M, Fritz S C. et al. A vegetation and fire history of Lake Titicaca since the Last Glacial Maximum. Palaeogeogr Palaeo-

Sea. Palaeoclimates, 1998, 3: 209―219 23

Chalie F, Gasse F. Late Glacial-Holocene diatom record of water chemistry lake level change from the tropical East African Rift Lake

ence, 1998, 282: 1858―1864[DOI] 44

Hansen B C S, Seltzer G O, Wright H E. Late-Quaternary vegetation change in the central Peruvian Andes. Palaeogeogr Palaeoclimat Palaeoecol, 1994, 109: 263―285[DOI]
