Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 328 – 340 www.elsevier.com/locate/palaeo
Holocene climate development on the central Tibetan Plateau: A sedimentary record from Cuoe Lake Wu Yanhong a,b,*, Andreas Lu¨cke b, Jin Zhangdong a, Wang Sumin a, Gerhard H. Schleser b, Richard W. Battarbee c, Xia Weilan a a
b
Key Laboratory of Lake Sedimentation and Environment, Nanjing Institute of Limnology and Geography, Chinese Academy of Sciences, Nanjing 210008, China Institute of Chemistry and Dynamics of the Geosphere V: Sedimentary Systems, Research Centre Ju¨lich, D-52425 Ju¨lich, Germany c Environmental Change Research Centre, University College London, 26 Bedford Way, London, WC1H 0AP, UK Received 14 September 2005; received in revised form 22 September 2005; accepted 29 September 2005
Abstract We present a climate record for central Tibet spanning the time between 10500 and 1600 cal years BP from a 5.3-m-long sediment core from Cuoe Lake. The analyses show that high Sr concentration, high TOC content, high C/N ratio and enriched y13C of lake sediments corresponded to stronger chemical weathering, high productivity, high lake level and warmer/wetter climate and vice versa. As revealed by the variations of these proxies, the climatic evolution during this period can be divided into five stages and several substages. These are a cold/dry stage before 10140 cal years BP, a transition stage from 10140 to 8560 cal years BP, an optimum stage (warm/wet) from 8560 to 5750 cal years BP, a cool/dry stage from 5750 to 4000 cal years BP, a transition stage from 4000 to 3000 cal years BP and a drying-up stage after 3000 cal years BP. These changes are presumably reflections of variations in Indian monsoon intensity which, according to that interpretation, should have been enhanced from 8560 cal years BP, slightly weakened after 5750 cal years BP, and gradually diminished since 3000 cal years BP in this region. D 2005 Elsevier B.V. All rights reserved. Keywords: Climate variability; Lake sediment; Cuoe Lake; Tibetan Plateau; Holocene
1. Introduction Despite its importance for the understanding of the Asian monsoon system during the past millennia, only a few continuous palaeoclimatic records are available from the Tibetan Plateau up to now. Most of them are * Corresponding author. Key Laboratory of Lake Sedimentation and Environment, Nanjing Institute of Limnology and Geography, Chinese Academy of Sciences, Nanjing 210008, China. Tel.: +86 25 86882145. E-mail address:
[email protected] (W. Yanhong). 0031-0182/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.palaeo.2005.09.017
from the plateau’s margin and based on tree ring, ice core, lake sediments as well as discontinuously outcropping lake sediments (Wang et al., 1983; Thompson et al., 1989, 1997; Fang, 1991; Gasse et al., 1991, 1996; Lister et al., 1991; van Campo and Gasse, 1993; Fontes et al., 1993; Gu et al., 1993; Sun et al., 1993; Li et al., 1994; Avouac et al., 1996; Shen and Tang, 1996; Wang et al., 1996; Yao et al., 1997a,b; Li et al., 1999; Wei and Gasse, 1999; Lehmkuhl and Haselein, 2000; Li et al., 2001; Tang and Li, 2001). A 13,000year record from Sumxi Co (western Tibet) was presented by Gasse et al. (1991) showing that conditions
W. Yanhong et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 328–340
in the early–middle Holocene were warmer and wetter than at present. There were warm, humid pulses at ~12.5 and ~10.0 ky BP, in phase with the steps of the last deglaciation, separated by a return to cold, dry conditions at ~11.0–10.0 ky BP. After 5.5 ky BP, there was a return to drier conditions in Sumxi Co. Three other continuous lacustrine records, Bangong Co (Gasse et al., 1996), Siling Co (Gu et al., 1993) and Qinghai Lake (Lister et al., 1991), show similar trends in climate change to that of Sumxi Co. The most characteristic features of these records are an unstable transition phase from 13.0 to 10.0 ky BP, the sudden establishment of wet conditions around 10.0 ky BP and maximum aridity around 4–3 ky BP (Gasse et al., 1991). Two ice cores (Dunde ice core and Guliya ice core) from the Tibetan Plateau record the climatic variations in the Holocene with high resolution. The longer ice core, from the Guliya ice field, indicated that after the Younger Dryas cold event, the temperature increased after 11.5 ky BP and dropped suddenly at 7 ky BP. The Hypsithermal started and ended earlier than in other places around the world, and its duration was 4.5 ky. From 7.2 to 5.0 ky BP, a relative cold phase occurred. The temperature increased continuously again after 5.0 ky BP except for a cold phase from 3.5 to 2.5 ky BP. Another ice core, the Dunde ice
329
core, indicated a different climatic variation pattern which is more consistent with global climatic change (Thompson et al., 1989, 1997; Yao et al., 1997a,b). Tree ring records from the south of the Tibetan Plateau and northern margin of the Tibetan Plateau recorded the climatic variation in the past 2.0 ky, which demonstrated several cold phases such as the Little Ice Age and several warmer phases (Wu and Lin, 1981; Wang et al., 1983; Kang et al., 1997; Bra¨uning, 1999). On the contrary, data from the interior of the plateau are relatively sparse, making it difficult to evaluate differences in the climatic development between the central Tibetan Plateau and the more marginal regions. However, this information is necessary in order to unravel the climatic development of the plateau and adjacent regions. The present Chinese climate is characterized by a marked gradient in continentality and aridity from southern and eastern China to western China (Wei and Gasse, 1999; He et al., 2004). On the Tibetan Plateau, the present climate is very diverse and changes from extreme continental climate in steppe, semi-desert and desert regions on the northern and western plateau to humid climate in well-vegetated mountain regions on the southern and eastern plateau (Zhen et al., 1996). These differences are primarily
Fig. 1. Sketch maps of China and the Tibetean Plateau with the sampling location Cuoe Lake near the City of Naqu. The coring site of the core CE-2 is indicated in the inset.
330
W. Yanhong et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 328–340
due to the monsoon circulation and the influence of the northern hemispheric westerlies, which had already formed before the Holocene as a result of the uplift of the Tibetan Plateau over the past millions of years (An et al., 2001). The high altitude of the Tibetan Plateau influences the range of monsoon and atmospheric circulation. The Southwest (Indian) and Southeast (Pacific) Monsoons penetrate China from the Indian Ocean and Pacific Ocean, respectively, during the summer months. The Indian monsoon reaches the Siling basin in central Tibet. In western Tibet, some monsoon winds can influence the southern portion (Bangong basin). The southeast monsoon cannot reach the interior of the plateau. The Tibetan Plateau blocks the mid-latitude westerlies, resulting in a split jet stream that moves to the north and south of the plateau. The westerlies affect the northern margin of the plateau but rarely reach the plateau itself (Boehner, 1994; Benn and Owen, 1998; Wei and Gasse, 1999). There are many lakes in the central Tibetan Plateau, most of which are closed basins, supplied only by rainfall over a restricted catchment area and not receiving meltwater from glaciers. Sediment cores from these lakes may reflect climate change and monsoon activity more accurately and more sensitively than open drainage lakes. In this paper, lacustrine sediments from Cuoe Lake, a closed lake in the central Tibetan Plateau, dated to between 10,500 and 1600 cal years BP were used to reconstruct Holocene variations in climate and monsoon activity. 2. Site description Cuoe Lake (31824V–31832VN, 91828V–91833VE) is located on the central Tibetan Plateau about 40 km west of the city of Naqu at 4532 m a.s.l. (Fig. 1). Like many other lakes on the Tibetan Plateau, the basin of Cuoe Lake is of tectonic origin generated by faulting during the uplift of the plateau. The basin is mainly controlled by sets of faults running SSW to NNE. The general setting of the lake and the basic limnological data are given in Table 1. The outcrops in the catchment of the lake are mainly composed of lacustrine, pluvial and alluvial deposits of Quaternary age overlaying amaranth sandstone and clay sandstone of Tertiary age. The vegetation in this area combines the features of alpine grassland and desert1 (Tibet Municipality Geology 1
Geomorphology and Quaternary Geology in Nagqu area, TibetQ compiled by the Tibetan Plateau scientific investigation team, Chinese Academy of Sciences, 1977 (unpublished internal document).
Table 1 General geographical, climatological and limnological data for Cuoe Lake on the central Tibetan Plateau Altitude
4532 m a.s.l.
N–S length E–W width Area Catchment area Annual precipitation Annual evaporation Annual average temperature Maximum water depth Inflows Outflow pH Salinity
~14.8 km ~5.7 km 61.3 km2 1081 km2 300–600mm (1954–1997A.D.) ~1000 mm 2.0 8C 5.0 m 14 streams None 9.8 (measured in July 2000) 12.06 g/l (measured in July 2000)
Data compilation is based on Tibet Municipality Geology and Mineralogy (1993) and bGeomorphology and Quaternary Geology in Nagqu Area, TibetQ compiled by the Tibetan Plateau Scientific Investigation Team, Chinese Academy of Sciences, 1977 (unpublished internal document).
and Mineralogy Bureau, 1993; Wang and Dou, 1998) and mainly resembles a xeric grass steppe. 3. Materials and methods 3.1. Sampling and lithology A 5.3-m-long core (CE-2) was taken in July 2000 using a piston corer from the deepest point of Cuoe Lake (Fig. 1). The upper 1.6 m of the sediment profile was sampled at intervals of 1 cm while below 1.6 m the sampling interval was enlarged to 2 cm. Sampling was done in the laboratory and the material was subsequently freeze-dried for storage and analysis. According to the primary aspect, the lithology of the core CE-2 is mainly composed of silty clay and clayey silt, except the deepest part, and can be divided into seven phases (Fig. 4): 0–35 cm: black mud, silty clay 35–40 cm: silty clay with sapropelic deposition and plant residues 40–371 cm: clayey silt, silty clay with gravel and plant residues interbedded with several layers of peat 371–458 cm: silty clay and clayey silt 458–461cm: sand and gravel 461–483cm: silty clay and clayey silt 483–520 cm: sand and gravel. The dark colour of the sediment in the upper 35 cm is not caused by a high organic content but by the high content of ferrous sulfide (after Jin, unpublished data). The sand and gravel in the deepest part suggest it to be
W. Yanhong et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 328–340
of fluvial origin. Lacustrine sediments occurred at a depth of 483 cm. 3.2. Methods For grain-size analysis, samples of 5.0 g were treated with HCl (5%) to remove carbonates and subsequently with NaOH (5%) to remove organic matter. The residues were dispersed with an ultrasonic oscillator for 15 min after mixing with a dispersant solution ((Na2PO3)6). The grain size was determined using a Malvern Laser Grainsize Meter. For Rb and Sr analysis, 5 g of dry sediment from each sample was ground to a fine powder (b 38 Am) in an agate mortar. The powder samples were then pressed into round discs, and the content of Rb and Sr was determined using a VP-320 XRF spectrometer. The relative standard deviation of measurements is less than 1 10 6. After removal of plant residues using a 100-Am sieve and removal of carbonate by washing with HCl (5%), a sample aliquot was dried again, ground and sieved using a 64-Am sieve and then analyzed on a CE440 elemental analyzer to determine total organic carbon (TOC) and total nitrogen (TN) content. Samples for stable carbon isotope analysis were sieved using a 100-Am sieve to separate abundant coarse plant remains from the fine sediment. Carbonates were removed in the fine fraction by treatment with HCl (5%), which was not necessary for the plant remains fractions. Stable carbon isotope ratios were determined separately for the fine sediment fraction and the plant remains. For analysis of the latter fraction, clean plant remains (up to 3
331
cm length) were hand-picked from the coarse fraction and ground. In 106 samples out of a total of 341, enough material for isotopic analysis could thus be obtained. The isotopic composition of the samples is reported in the conventional y - notation in per mil relative to the VPDB standard: y13 Csample ðxÞ ¼ Rsample Rstandard =Rstandard 1000 13
where R is the C/12C ratio of the sample and the standard, respectively. 4. Chronology Initially, 10 samples of bulk sedimentary material (sieved to 100 m) were collected for radiocarbon age determinations and analyzed by AMS (Tokyo University, Japan) and liquid scintillation counting (Key Laboratory of Lake Sedimentation and Environment, CAS, China). Owing to the low organic matter content, the radiocarbon ages of three samples could not be reliably determined. Seven 14C ages were calibrated with OxCal 3.5 and median values of the 2r error bands are used to establish an age–depth relation (Table 2). The uppermost sample CE-1-9 (47–50 cm) revealed a calibrated age of 5135 years BP, which would have made the sediment surface much too old (~3500 years). Furthermore, it contradicted 137Cs and 210Pb dating of the upper layer of another core from Cuoe Lake (CE-1) which gave recent ages (Wu et al., 2001). Obviously, the radiocarbon ages are influenced either
Table 2 Radiocarbon ages of bulk organic matter (OMbulk) and plant remains (Plant r.) from sediment samples of Cuoe Lake (core CE-2) as derived from liquid scintillation counting (LSC) and accelerator mass spectrometry (AMS) Sample
Depth (m)
Material Method Laboratory
14 C age (years BP)
Calibrated age (2r cal years BP)
Median (cal years BP)
Average (cal years BP)
Error estimate (F years)
CE-1B CE-35B CE-35D CE-39B CE-39D Ce-1-9 Ce-1-10 Ce-1-2 Ce-1-3 Ce-1-4 Ce-1-6 CE-286D Ce-1-8
0.01 0.35 0.35 0.39 0.39 0.47–0.50 1.70 1.93–1.98 2.37–2.40 3.09–3.14 3.76–3.82 4.12 4.74–4.84
OMbulk OMbulk Plant r. OMbulk Plant r. OMbulk OMbulk OMbulk OMbulk OMbulk OMbulk Plant r. OMbulk
3260 F 30 4660 F 40 4410 F 35 4480 F 35 4340 F 40 4510 F 130 6660 F 80 6426 F 45 8030 F 95 8690 F 90 9126 F 330 7650 F 50 11,570 F 90
3567–3396 5573–5305 5255–4866 5298–4972 5031–4833 5580–4833 7671–7425 7427–7253 9254–8599 10,145–9502 11,197–9490 8539–8368 13,851–13,176
3469 5412 5020 5116 4866 5135 7530 7366 9000 9627 10,238 8411 13,480
3482 5439 5061 5135 4932 5207 7548 7340 8926 9824 10,344 8454 13,513
85 134 195 163 99 373 123 87 328 321 853 85 338
AMS AMS AMS AMS AMS AMS AMS LSC LSC LSC LSC AMS AMS
Poland Poland Poland Poland Poland Japan Japan China China China China Poland Japan
Calibration was performed with OxCal 3.5 (Bronk Ramsey 2000) using the INTCAL 98 calibration data set. Given calibrated age ranges refer to 2r results of calibration. Median ages were used for the establishment of the age–depth relation of core CE-2.
332
W. Yanhong et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 328–340
by a strong reservoir effect or an input of old carbon from the catchment area. Such problems of radiocarbon dating when using bulk organic matter are well known from other investigations and were detected in earlier studies of lacustrine sediments from the Tibetan Plateau to a similar or even larger extent (Sun et al., 1993; Gu et al., 1993; Fontes et al., 1996; Wang et al., 2002; He et al., 2004). One factor might be that in the Cuoe Lake catchment calcareous silt and limestones occur. The alkaline waters of the tributaries and the lake itself can dissolve carbonate and soil organic matter. This may introduce dissolved inorganic carbon (DIC) with an old radiocarbon signature to aquatic and terrestrial plants where it is assimilated and afterwards deposited in the lake sediments. Therefore, another six samples were selected for 14C AMS dating at the Poznan˜ Radiocarbon Laboratory, Poland. These include three samples of bulk organic matter from the surface and from depths of 35 cm and 39 cm, and samples of macroscopic plant remains from depths of 35 cm and 39 cm as well as one from a depth of 412 cm (Table 2). The results were calibrated following the same method as for the former seven samples. The samples were selected to confirm the existence of an bold carbonQ effect, to estimate the respective reservoir age and to confirm a hiatus expected at a depth of 37 cm. The macro-remains were not previously used for dating because identification was not possible. However, the material proved to be very homogeneous all through the core and showed characteristics similar to plants growing in swamps or in the littoral zone of lakes (F. Bittmann, personal communication). The calibrated age of the uppermost sediment sample CE-1B (1 cm) appeared to be 3469 years BP and proves a respective reservoir age of the bulk organic matter samples. The ages of plant remains at depths of 35 cm and 39 cm appeared to be 392 years and 250 years younger than the age of bulk organic matter at the same depths. This indicates that not only the authigenic organic matter but also the terrestrial organic matter or aquatic macrophytes were influenced by bold carbon.Q Therefore, it was decided to establish an age–depth model based on linear interpolation between the ages of the bulk organic matter samples corrected for the reservoir age of 3470 years (Fig. 2). However, the assumption of a constant reservoir age has to be seen as an approximation. The radiocarbon ages of sample CE-35B and CE-1-10 were excluded from the calculation of the age–depth relation because of an age inversion. According to this relation, the age of the surface would be 1100 years BP but not recent. This time lag
Fig. 2. Relationship between depth and age of CE-2 core. Thirteen 14 C dating results were calibrated with OxCal 3.5 (Bronk Ramsey 2000) using the INTCAL 98 calibration data set and exhibit the reservoir effect. Corrected ages are obtained by subtracting reservoir age (3470 years) from 14C ages. A hiatus emerged around 37 cm in the CE-2 core and is further supported by lithology. The relationship between depth and age was established using seven corrected ages together with the calibrated 14C age of plant remains at a depth of 412 cm, which is presumably not influenced by the reservoir effect.
could not be attributed to a reservoir effect but has to be interpreted as a sedimentary hiatus at a core depth between 40 and 35 cm. Supporting evidence for this includes (1) the development of the swampy deposition of sapropel layer with a lot of plant remains, (2) anoxic sedimentary conditions were recorded by Th/U (b1) and the only evaporate mineral, gypsum, was detected (Jin, unpublished data), and (3) a previously recognized climate change around 2000 years BP, which probably also influenced Cuoe Lake. Around 2 ky BP, there occurred an abrupt cold and dry event in the central part of the Tibetan Plateau demonstrated by ice core and lacustrine sediments (Gasse et al., 1991, 1996; Sun et al., 1993; Tang and Li, 2001), which resulted in widespread drop in lake level and lake shrinkage. Interestingly, the age of the plant remains from a depth of 412 cm (CE-286D) differs remarkably from that of the uncorrected bulk samples. Presumably, at that time, these plants used atmospheric CO2 as an inorganic carbon source for photosynthesis only. This would be the case for land plants or for macrophytes with most of the plant parts emerged. Within the uncertainties of the material and the age model, the uncorrected age of the
W. Yanhong et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 328–340
Fig. 3. Triangle graph of grain-size composition shows that the sediment of CE-2 core was mainly composed of silt and clay.
plant sample CE-286D fits well with the corrected ages of the other bulk samples.
333
layers emerging at the deepest part of the core and some clayey sand layers interbedded in the upper part (Fig. 4). Before 10140 cal years BP (below 484 cm), the main grain-size component was sand and gravel with the content of the N64-Am fraction being more than 40%, and reaching a maximum of 66%. The major components from 10140 to 9560 cal years BP (484–466 cm) and from 8840 to 8580 cal years BP (444–436 cm) were silt (4–64 Am, content N 50%) and clay (b 4 Am, content N30%), without sand or gravel. From 9560 to 8840 cal years BP (466–444 cm), the major components were still silt and clay, but with a small amount (N10%) of sand and pebbles. A peak in the N 64-Am fraction occurred at 9300 cal years BP (458 cm). From 8580 to 1650 cal years BP (436–40 cm), the sediment was a clay–silt with some sand and gravel. The abrupt changes of grain-size composition, especially the sudden increase of the coarser fractions, for example, at 5700 cal years BP (260 cm), 4350–3970 cal years BP (210–200 m) and 2500 cal years BP (around 100 cm), must be related to environmental events (Fig. 4).
5. Results 5.2. Rb, Sr, and Rb/Sr 5.1. Grain size The sediment of core CE-2 is mainly composed of silty clay and/or clayey silt (Fig. 3), except for the sand
The Rb/Sr ratio of the CE-2 core shows a clear pattern. The highest period of the Rb/Sr ratio was before 8500 cal years BP (434 cm). In this high Rb/
Fig. 4. Proxy variation in CE-2 core. Sand content represents the content of N64-Am fraction, which reflects the hydraulic energy variation in the lacustrine phase. Rb/Sr indicates an environment change occurring around 430 cm (8600 cal years BP). y13C of plant remains was generally higher than that of bulk organic matter at the same depth, indicating the aquatic source of organic matter as the supplement to terrestrial plants/soils.
334
W. Yanhong et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 328–340
Sr ratio period, the Sr concentration increased continuously except for a slight decrease from 9100 to 8500 cal years BP, while the Rb concentration fluctuated with high values (Fig. 4). After 8500 cal years BP, the Sr and Rb concentrations fluctuated in the opposite pattern and the Sr concentration generally showed high values while the Rb concentration displayed low values. The sharp drop of Sr and increase of Rb caused several small peaks of the Rb/Sr ratio around 4500 cal years BP and after 1700 cal years BP. 5.3. TOC and C/N The TOC content varied from 0.26% to 6.55% (Fig. 4). Before 8520 cal years BP (below 434 cm), TOC was, in general, lower. A rapid increase in TOC values occurred at 8520 cal years BP (434 cm). After a rapid increase, the high TOC phase lasted from 8520 to 5760 cal years BP with a period of low values from 7380 to 6730 cal years BP. After 5760 cal years BP, the TOC was lower but increased again around 3800 cal years BP. After 3800 cal years BP, TOC values exhibited a declining trend. The TN content and C/N ratio have a similar pattern of variation as compared to TOC content (Fig. 4). 5.4. Stable carbon isotope The y13C of bulk organic matter in the CE-2 core changed from 28.1x to 17.6x (Fig. 4). Generally, the y13C was higher before 5760 cal years BP, except for a short period with depleted values around 6600 cal years BP. After 5760 cal years BP, the y13C declined except for some small peaks , the lowest y13C ( 28.1x) emerging from 1660 to 1650 cal years BP. The y13C of plant remains is 3.6–11.5x higher than that of bulk organic matter and varied from 12.0 to 21.1x. This indicates that the source of organic matter in the CE-2 core was not only terrestrial plants and soil organic matter but also aquatic macrophytes. Most probably, due to the very high salinity, algae did not make an appreciable contribution to the sedimentary organic matter.
6. Discussion The grain-size composition of sediments is generally used to describe changes of the sedimentary environment (Lambiase, 1980) related to hydrologic energy variations. In the lacustrine environment, grain size fluctuations through time indicate changes in lake
water level. The finer grain-size spectrum represents higher lake levels while the coarser grain-size spectrum represents relatively low lake levels reflecting changes between wet and dry periods, respectively (Wang and Feng, 1991; Chen and Wan, 1999). Although hydrologic energy is the main factor controlling the grain-size distribution, the relationship between grain-size composition and climate humidity is not linear. Increasing runoff can still transport coarser material over longer distances within the lake, even to the centre of the lake, when the energy is high enough, e.g., at high precipitation rates. In core CE-2, the grain size reveals that below 444 cm (before 8840 cal years BP), the sedimentary environment was in the fluvial phase, while the lacustrine phase developed afterwards. During the fluvial phase, the sedimentary environment can be divided into three sub-phases, namely, a river channel sub-phase before 10,140 cal years BP, an over-bank lake sub-phase from 10,140 to 9560 cal years BP and a river channel subphase from 9560 to 8840 cal years BP. A stable lacustrine phase developed from 8840 to 1650 cal years BP. During the lacustrine phase, the sand and gravel content in the sediment, therefore, can be seen as the reflection of the variations of runoff. Under wet climatic condition, increasing runoff enhances the hydrologic transport energy, resulting in coarser particles in the central lake sediment. Fractionation of Rb and Sr occurs in surface geochemical processes due to the difference in their geochemical character (Leinen and Heath, 1981; White et al., 1994). Previous studies have shown that the Rb/Sr ratio in weathering residue increases as weathering intensity increases due to a higher Sr activity and a more inert nature of Rb (Parrington et al., 1983; Chen et al., 1998, 1999). In lake sediments, the relationship between Rb/Sr ratio and weathering intensity is the opposite of that of the weathering residue (Jin et al., 2001), as it is more dependent on the transportation mechanism of Rb and Sr. In Cuoe Lake, a sequential extraction experiment shows that almost all Rb is sequestered in the residual (silicate) phase, whereas 30– 50% of Sr is bound to carbonate and/or to Fe–Mn oxides (Jin, unpublished data). We suggest that the more inert Rb was carried into the basin together with the silicate input resulting from physical weathering, especially during the cold periods, but that the Sr bound to carbonate and/or Fe–Mn oxides was leached into solution during chemical weathering and was transported as dissolved Sr2+. Accordingly, stronger chemical weathering leads to more dissolved Sr2+ being transported into the depositional basin, resulting in a
W. Yanhong et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 328–340
335
Fig. 5. Sr concentration and TOC varied with almost the same pattern, indicating that the high Sr concentration as well as high TOC content represents increased macrophyte input and stronger chemical weathering when the climate was warmer and wetter. Curves represent a five-point running mean.
decrease in the Rb/Sr ratio. Therefore, the Rb/Sr ratio in lake sediments can be used as an indicator of the weathering intensity in catchments. Chemical weathering is sensitive to changes in temperature and humidity. Under cold and dry climate conditions, dissolved Sr content is mainly controlled by temperature and decreases with decreasing temperature. Under warm and wet conditions, precipitation is more important and increasing precipitation introduces more dissolved Sr into the lake sediment (Chen et al., 1999; Jin et al., 2001). In the CE-2 core, Sr concentration varied almost with the same pattern as the TOC variations (Figs 4 and 5). Normally, the TOC content in sediments from unproductive lakes is derived from the plants that have lived in the littoral zone and on the land surrounding the lake (Meyers and Ishiwatari, 1993). The TOC content of lake sediment is therefore a useful proxy for paleoenvironmental reconstruction, reflecting climatic conditions and lake environment. Analyses of Qarhan lake sediments show that higher organic carbon content coincides with warmer rather than cooler periods (Huang et al., 1990). Results from the RH core in the northeastern Tibetan Plateau also show that high TOC values are related to a warm and wet climate (Wang et al., 1994). We therefore suggest that high Sr concentration and high TOC represent periods of higher input of littoral/shore plants when the climate was warmer and wetter than during periods of low Sr and TOC content. The C/N ratio is generally used to identify the source of organic matter in the sediment (Meyers and Ishiwatari, 1993). The C/N ratio in the CE-2 core suggests that the organic matter in sediments was a mixture of soil and littoral organic matter input. The C/N value of soils in more arid areas can be much lower than the C/N value of the original plant and may even reach values as low as 10 (Mayr et al., 2005). The low value in the
bottom layer is probably caused by decomposition of the sediments during the fluvial phase since the oxidation of organic matter causes preferential loss of carbon from the sediment (Sarazin et al., 1992). The drop of the C/N ratio after 3000 cal years BP is probably also caused by enhanced decomposition of organic matter due to a lake level drop and enhanced oxidation of organic matter. In the CE-2 core, TOC content is positively related to the C/N ratio with an R 2 of 0.45, while the relationship between TN content and C/N ratio was worse with an R 2 of 0.05 (Fig. 6). The relationships between TOC content, TN content and C/N ratio indicate that the major factor influencing C/N ratio is TOC content, but not TN content. The C/N ratio therefore partially reflects the primary productivity variation, although the degradation of organic matter may cause an additional decrease of the C/N ratio (Sarazin et al., 1992). The stable carbon isotope composition (y13C) of organic matter in lake sediment has been used as a proxy indicator for climate, mainly based on the relative composition of C3, C4 and CAM plants, which are the source of organic matter in lake sediment, and their different pathways of photosynthetic CO2 fixation (Stuiver, 1975; Krishnamurthy et al., 1986; Bowen, 1991; Wang et al., 1993; Zhang and Wang, 1995; Wu and Wang, 1996). C4 plants are mainly distributed in both middle to low altitudes and their occurrence decreases with increases both in altitude and latitude (Sage et al., 1999). Surveys in the northern and central Tibetan Plateau show that the vegetation is dominated by C3 plants (forest and alpine meadow) at an altitude above 3500 m a.s.l., and the contribution of organic carbon from C4 plants to soil samples is negligible (Lu et al., 2001). Previous works also show that the y13C of C3 plants is sensitive to variations in humidity or precipitation, and that the y13C value is influenced by
336
W. Yanhong et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 328–340
Fig. 6. Relationships between TOC content, C/N ratio and y13C. A strong positive relation between C/N ratio and TOC content and a poor relation between C/N ratio and TN indicate that the C/N ratio is mainly dependent on the TOC content variation. Both TOC content and C/N ratio are positively related with y13C suggesting that all the TOC content, C/N ratio and y13C are controlled by climate variations.
water content in the soil or precipitation, especially where the altitude is above 3500 m a.s.l. (Francey and Farquahar, 1982; Sternberg et al., 1984; Schleser, 1995; Lu et al., 2001).The altitude of the Cuoe Lake catchment area was higher than 3500 m before 12,000 years BP.2 The y13C of sedimentary organic matter at Cuoe lake, therefore, presumably reflects the precipitation fluctuation. As a conclusion, during the lacustrine phase of Cuoe Lake, the high sand content, Sr content, TOC value, C/N ratio and y13C corresponded to warm and wet climatic conditions, stronger chemical weathering and higher primary (input) productivity, and vice versa. 7. Holocene climatic variability in the central Tibetan Plateau Lithology, TOC content, C/N ratio, grain-size composition, y13C composition, Rb and Sr concentrations 2 Wu Yanhong, 2002. Preliminary study on Quaternary environmental change in Central Tibetan Plateau, PhD Dissertation , Graduate Institute, Chinese Academy of Sciences.
and the Rb/Sr ratio (Fig. 7) of the CE-2 core describe the lacustrine and palaeoclimatic development during the period from before 11,000 to 1650 cal years BP in the central Tibetan Plateau. Before 10,140 cal years BP (stage I), the climate was cold and dry and a fluvial sedimentary environment had developed. TOC and Sr concentration were at their lowest value, indicating low chemical weathering and low primary production. Stage II lasted from 10,140 to 8450 cal years BP and can be described as a transition phase. This stage was established in three steps during which the climate presumably reached its optimum and Cuoe Lake developed. The first step (from 10,140 to 9750 cal years BP) corresponded to an over-bank lake sub-phase where Sr concentration, TOC and C/N ratio were low, but Rb concentration was at a maximum. These features of proxies indicate that primary productivity was low, and chemical weathering was weak while physical weathering was strong. y13C was intermediate at this substage and the organic matter originated partly from soil. The climate at this substage can therefore be described as cold and wet. In the second step (from 9750 to
W. Yanhong et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 328–340
337
Fig. 7. Interpretation of proxy variations in CE-2 core. All curves represent five-point running means. For details, refer to the text.
8840 cal years BP), Sr concentration, TOC content, the C/N ratio and y13C were relatively high, implying a warmer and wetter climate. In the last step (from 8840 to 8560 cal years BP), all the proxies dropped rapidly thus indicating an intermittent cold and dry phase. After the rapid change at the beginning, the climate reached its optimum period during the Mid-Holocene (stage III, from 8560 to 5750 cal years BP). During this period, the lake was enlarged and the lake level raised. The sedimentation environment therefore changed from a riverine system to a lacustrine system. A partly coarser composition of the sediments represents the variation of the hydrologic transport energy caused by precipitation variations. The dramatic drop of y13C, TOC content and the C/N ratio from 7280 to 6750 cal years BP suggests that a severe climatic event probably occurred. The small peak of Rb concentration and Rb/Sr ratio around 6800 cal years BP (380 cm in CE-2 core) indicate weak chemical weathering and relatively strong physical weathering, most likely as a result of cold and wet climate, so that we describe this climatic event as a wet and cold pulse. Except for this cold pulse, y13C, TOC content and the C/N ratio all remained high during this optimum period. Stage IV lasted from 5750 to 3000 cal years BP, starting with a decreasing trend of the Sr concentration,
y13C, TOC content and the C/N ratio and then fluctuated at a lower level. The coarse grain composition (N 250 Am) also disappeared at the beginning of this phase. The characteristics of the proxy variations revealed that the climate became a little cooler and drier than before. A notably higher Rb concentration and Rb/Sr ratio, as well as a low value of y13C of plant remains, suggested a cold and dry event occurring from 4600 to 4200 cal years BP. The second substage starting from ca. 4000 cal years BP was an unstable stage of climate against the background of a cooler and drier climate, indicated by higher fluctuations of almost all proxies. After 3000 cal years BP, TOC content, the C/N ratio and the y13C exhibited a negative trend, where coarse composition (N 250 Am) occasionally emerged in the sediment, reflecting the drying up trends. 8. Conclusion According to the above interpretation, several climate fluctuations have occurred in the central Tibetan Plateau during the Holocene. After an initial transition phase, the climatic optimum should have occurred between 8560 and 5750 cal years BP. A short-term cool and dry shift emerged in the optimum period from 7280 to 6750 cal years BP. The period from
338
W. Yanhong et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 328–340
5750 to 4000 cal years BP was relatively cool and dry. From 4000 to 3000 cal years BP, the climate frequently alternated between warm and cool and dry and wet conditions. After 3000 cal years BP, the climate became more arid, which might be attributed to the effects of the regional monsoon and/or global climatic aridity (Kelts et al., 1989; Lister et al., 1991; Gasse et al., 1991, 1996; Fontes et al., 1993; Gu et al., 1993; Shi et al., 1998). Cuoe Lake is located at the leading edge of the Indian monsoon, so the regional climate can be expected to be sensitive to variations in the monsoon intensity. Consequently, the palaeoclimate record deduced from the various proxies used for the analysis of the CE-2 core can be used to draw the first picture of the activity of the Indian monsoon during the Holocene in the Central Tibetan Plateau. In the early to middle Holocene hydrological optimum, the primary productivity in the lake catchment increased although increasing summer radiation enhanced evaporation in response to abundant monsoonal precipitation at this time. Overall, the monsoon in this region was enhanced after 8560 cal years BP, weakened slightly after 5750 cal years BP, and gradually diminished after 3000 cal years BP. This pattern of monsoon evolution is quite comparable with the result of COHMAP (1988) and the results deduced from the vegetation evolution (Shen and Tang, 1996; Tang and Li, 2001). According to this pattern of climatic variation, the optimum period of the Holocene on the central Tibetan Plateau started earlier and also ended a little earlier than in the marginal areas such as in SumxiLongmu Co (Gasse et al., 1991) and Bangong Co (Gasse et al., 1996). With the continuous aridity trend after 3000 cal years BP, the maximum aridity on the central Tibetan Plateau seems to have appeared after 2000 years BP, leading to an extensive shrinkage of lake volume and a disruption in the sediment accumulation. This is different from the development in the marginal areas where the maximum aridity was found from 3900 to 3200 years BP (Gasse et al., 1996). Acknowledgements We thank W. Laumer for assistance in stable isotope analysis and F. Bittman for his work on identifying the plant remains. We are grateful for financial support from the Chinese b973 projectQ (Grant 2004CB720205), the German Federal Ministry of Education and Research through DEKLIM (Grant
01LD0001 to AL) and a fellowship from the MaxPlanck-Gesellschaft for WY. References An, Z., Kutzbach, J.E., Warren, L.P., Porter, S.C., 2001. Evolution of Asian monsoon and phased uplift of Himalaya–Tibetan Plateau since late Miocene times. Nature 411, 62 – 66. Avouac, J.P., Dobremez, J.F., Bourjot, L., 1996. Paleoclimatic interpretation of a topographic profile across middle Holocene regressive shorelines of Longmu Co (Western Tibet). Palaeogeography, Palaeoclimatology, Palaeoecology 120, 93 – 104. Benn, D.I., Owen, L.A., 1998. The role of Indian summer monsoon and the mid-latitude westerlies in Himalayan glaciation: review and speculative discussions. Journal of the Geological Society (London) 155, 353 – 363. Boehner, J., 1994. Circulation and representativeness of precipitation and air temperature in the southeast of the Qinghai-Xizang Plateau. Geological Journal 34, 55 – 66. Bowen, R., 1991. Isotopes and Climates. Elsevier Applied Science, London, pp. 128 – 131. Bra¨uning, A., 1999. Dendroclimatological potential of drought-sensitive tree stands in Southern Tibet for the reconstruction of the monsoonal activity. IAWA Journal 20 (3), 325 – 338. Chen, Jingan, Wan, Guojiang, 1999. Sediment particle size distribution and its environmental significance in lake Erhai, Yunnan province. Acta Mineralogica Sinica 19 (2), 175 – 182. Chen, Jun, An, Zhisheng, Wang, Yongjin, 1998. Rb/Sr distribution in Luochuan loess profile and paleomonsoon variation since 800 ka. Science in China (D) 28 (6), 498 – 504. Chen, J., An, Z., Head, J., 1999. Variation of Rb/Sr ratios in the loess– paleosoil sequences of Central China during the last 130,000 years and their implications for monsoon paleoclimatology. Quaternary Research 51, 215 – 219. COHMAP Members, 1988. Climatic changes of the last 18 000 years: observations and simulations. Science 241, 1043 – 1052. Fang, Jin-Qi, 1991. Lake evolution during the past 30,000 years in China, and its implication for the environmental change. Quaternary Research 36, 1 – 24. Fontes, J.Ch., Melieres, F., Gibert, E., Liu, Qing, Gasse, F., 1993. Stable isotope and radiocarbon balances of two Tibetan lakes (Sumxi Co, Longmu Co) from 13,000 yr BP. Quaternary Science Reviews 12, 875 – 887. Fontes, J.-C., Gasse, F., Gibert, E., 1996. Holocene environmental changes in Lake Bangong basin (West Tibet): Part 1. Chronology and stable isotope of carbonates of a Holocene lacustrine core. Palaeogeography, Palaeoclimatology, Palaeoecology 120, 25 – 47. Francey, R.J., Farquahar, G.D., 1982. An explanation of 13C/12C variations in tree rings. Nature 297, 28 – 31. Gasse, F., Arnold, M., Fontes, J.C., Fort, M., Gibert, E., Huc, A., Li, Bingyan, Li, Yuanfang, Liu, Qing, Me´lie`res, F., Van Campo, E., Wang, Fubao, Zhang, Qingsong, 1991. A 13,000-year climate record from Western Tibet. Nature 353, 742 – 745. Gasse, F., Fontes, J.Ch., van Campo, E., Wei, K., 1996. Holocene environmental changes in Bangong Co basin (Western Tibet): Part 4. Discussion and conclusions. Palaeogeography, Palaeoclimatology, Palaeoecology 120, 279 – 292. Gu, Zhaoyan, Liu, Jiaqi, Yuan, Baoyin, Liu, Dongsheng, Liu, Rongmo, Liu, Yu, Zhang, Gaungyu, 1993. The changes in monsoon influence in the Qinghai–Tibetan Plateau during the past 12 000 years—geo-
W. Yanhong et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 328–340 chemical evidence from L. Silling Co sediments. Chinese Science Bulletin 38 (1), 61 – 64. He, Y., Theastone, Wilfred H., Zhang, Zhonglin, et al., 2004. Asynchronous Holocene climatic change across China. Quaternary Research 61, 52 – 63. Huang, Qi, Meng, Zhaoqiang, Liu, Hailing, 1990. Preliminary study on paleoclimatic variation in saline lakes of Qaidam Basin. Science in China (B) 20 (6), 652 – 663. Jin, Zhangdong, Wang, Sumin, Shen, Ji, Zhang, Enlou, Li, Fuchun, Ji, Junfeng, Lu, Xinwei, 2001. Chemical weathering since the little ice age recorded in lake sediments: a high-resolution proxy of past climate. Earth Surface Processes and Landforms 26 (7), 775 – 782. Kang, Xingcheng, Grumlich, L.J., Sheppard, P., 1997. The last 1835 years climate changes inferred from tree ring records in Dulan region, Qinghai, China. Quaternary Sciences 1, 70 – 75. (in Chinese). Kelts, K., Chen, K.Z., Lister, G., Qing, Y.J., Hong, G.Z., Niessen, F., Bonani, G., 1989. Geological fingerprints of climate history: a cooperative study of Qinghai Lake, China. Ecological Geol. Helv., vol. 82 (1), pp. 167 – 182. Krishnamurthy, R.V., Bhattacharya, S.K., Kusumgar, S., 1986. Paleoclimatic changes deduced from 13C/12C and C/N ratios of Karewa lake sediments, India. Nature 323, 150 – 152. Lambiase, J.J., 1980. Hydraulic control of grain size distributions in a macrotidal estuary. Sedimentology 27, 433 – 446. Lehmkuhl, F., Haselein, F., 2000. Quaternary paleoenvironmental change on the Tibetan Plateau and adjacent areas (western China and west Mongolia). Quaternary International 65/66, 121 – 145. Leinen, M., Heath, G.R., 1981. Sedimentary indicators of atmospheric circulation in the northern hemisphere during the Cenozoic. Palaeogeography, Palaeoclimatology, Palaeoecology 36, 1 – 21. Li, Yuanfang, Zhang, Qingsong, Li, Bingyuan, Gasse, F., 1994. Ostracod fauna and environmental changes during the past 17000 years in western Tibet. Acta Geographica Sinica 49 (1), 46 – 53 (in Chinese). Li, Shenfeng, Wang, Fubao, Zhang, Jie, Shen, Xuhui, Zeng, Zhaoqi, 1999. Holocene diatom record and environment change in Angren Co, Tibet. Chinese Science Bulletin 44 (3), 302 – 323. Li, Bingyuan, Wang, Sumin, Zhu, Liping, Li, Yuanfang, 2001. 12 ka BP lake environment on the Tibetan Plateau. Science in China (D) 44, 324 – 331 (Supp.). Lister, G.S., Kelts, K., Chen, Ke Zao, Jun-Qing, Yu, Niessen, Frank, 1991. Lake Qinghai, China: closed basin lake levels and the oxygen isotope record for Ostracoda since the latest Pleistocene. Palaeogeography, Palaeoclimatology, Palaeoecology 84, 141 – 162. Lu, Houyuan, Gu, Zhaoyan, Wu, Naiqin, Wang, Luo, Guo, Zhengtang, Wu, Haibin, 2001. Effect of altitude on the organic carbonisotope composition of modern surface soils from Qinghai– Xizang Plateau. Quaternary Sciences 21 (5), 399 – 406. Mayr, C., Fey, M., Haberzettl, T., Janssen, S., Lu¨cke, A., Maidana, N.I., Ohlendorf, C., Scha¨bitz, F., Schleser, G.H., Struck, U., Wille, M., Zolitschka, B., 2005. Palaeoenvironmental changes in southern Patagonia during the last millennium recorded in lake sediments from Laguna Azul (Argentina). Palaeogeography, Palaeoclimatology, Palaeoecology 228, 203 – 227. Meyers, P.A., Ishiwatari, R., 1993. Lacustrine organic geochemistry— an overview of indicators of organic matter sources and diagenesis in lake sediments. Organic Geochemistry 20, 867 – 900. Parrington, J.R., Zoller, W.H., Aras, N.K., 1983. Asian dust: seasonal transport to the Hawaiian Islands. Science 220, 195 – 197. Sage, R.F., Wedin, D.A., Li, Meirong, 1999. The biogeography of C4 photosynthesis: patterns and controlling factors. In: Sage, R.F.,
339
Monson, R.K. (Eds.), C4 Plant Biology. Academic Press, San Diego, CA, pp. 3 – 16. Sarazin, G., Michard, G., Al Gharib, I., Bernat, M., 1992. Sedimentation rate and early diagenesis of particulate organic and carbon in Aydat Lake (Puy de Doˆme, France). Chemical Geology 98, 307 – 316. Schleser, G.H., 1995. Parameters determining carbon isotope ratios in plants. In: Frenzel, B., Stauffer, B., Weiss, M.M. (Eds.), Problems of Stable Isotopes in Tree-Rings, Lake Sediments and Peat-Bogs as Climatic Evidence for the Holocene, ESF Special Issue, vol. 17. Gustav Fischer Verlag, Stuttgart, pp. 71 – 96. Shen, Caiming, Tang, Linyu, 1996. Pollen evidence for monsoon variation in Tibetan Plateau. Acta Micropalaeontologica Sinica 13 (4), 433 – 436. Shi, Yafeng, Li, Jijun, Li, Bingyuan (Eds.), Uplift and Environmental Changes of Qinghai–Xizang (Tibetan) Plateau in the Late Cenozoic. Guangdong Science and Technology Press, Guangzhou China, pp. 3 – 15. Sternberg, L.O., Deniro, M.J., Johnson, H.B., 1984. Isotope ratios of cellulose from plants having different photosynthetic pathways. Plant Physiology 74, 557 – 561. Stuiver, M., 1975. Climate versus change in 13C content of the organic component of lake sediments during the late quaternary. Quaternary Research 5, 251 – 262. Sun, Xiangjun, Du, Naiqiu, Chen, Yinshuo, Gu, Zhaoyan, Liu, Jaqi, Yuan, Baoyin, 1993. Analysis on pollen of lake sediment in Silling Co, Tibet. Acta Botanica Sinica 35 (12), 943 – 950. Tang, Lingyu, Li, Chunmei, 2001. Temporal–spatial distribution of the Holocene vegetation in the Tibetan Plateau. Journal of Glaciology and Geocryology 23 (4), 367 – 374. Thompson, L.G., Tomphson, Mosley, Davis, E., M.E., 1989. Holocene–Late Pleistocene climatic ice core records from Qinghai– Tibetan Plateau. Science 246, 474 – 477. Thompson, L.G., Yao, T., Davis, M.E., et al., 1997. Tropical climatic instability: the last glacial cycle from a Qinghai–Tibetan ice core. Science 276, 1821 – 1825. Tibet Municipality Geology and Mineralogy Bureau, 1993. Local Geological Records of Tibet Municipality. Geology Press, Beijing, p. 132. van Campo, E., Gasse, F., 1993. Pollen- and diatom-inferred climatic and hydrological changes in Sumxi Co Basin (western Tibet) since 13,000 yr B.P. Quaternary Research 39, 300 – 313. Wang, Sumin, Dou, Hongshen (Eds.), 1998. Chinese Lake Records. Science Press, Beijing, p. 327. Wang, Sumin, Feng, Ming, 1991. Lake environment change in Daihai, Inner Mongolia, and its relationship with south-east monsoon. Science in China (B) 21 (7), 759 – 768. Wang, Y., Guangyuan, L., Zhang, X., 1983. The relationship between the tree rings of Qilianshan juniper and climatic change and glacial activity during the past 1000 years in China. Kexue Tongbao 28, 1647 – 1652. Wang, Fubao, Yan, Ge, Lin, Benhai, 1993. Preliminary study on y13C of peat in Zoige Plateau, Tibet. Chinese Science Bulletin 38 (1), 65 – 68. Wang, Sumin, Shi, Yafeng, Shen, Ji, 1994. Preliminary study on paleoclimate and paleoenvironment change of the east of Qinghai–Xizang Plateau since 800 ka BP. Study on Plateau Forming and Evolution, Environment Evolution and Ecosystem of Tibetan Plateau. Science Press, pp. 236 – 248. Wang, Fubao, Han, Huiyou, Yan, Ge, 1996. Paleovegetation and paleoclimate evolution sequence since 30 ka in north-eastern Tibetan Plateau. Science in China (series D) 26 (2), 111 – 117.
340
W. Yanhong et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 234 (2006) 328–340
Wang, R.L., Scarpitta, S.C., Zhang, S.C., et al., 2002. Later Pleistocene/Holocene climate conditions of Qinghai–Xizang plateau (Tibet) based on carbon and oxygen stable isotopes of Zabuye lake sediments. Earth and Planetary Letters 203, 461 – 477. Wei, K., Gasse, F., 1999. Oxygen isotopes in lacustrine carbonates of west China revisited: implications for post glacial changes in summer monsoon circulation. Quaternary Science Reviews 18, 1315 – 1334. White, J.W.C., Cials, P., Figgo, R.A., Keany, R., Markgraf, V., 1994. A high-resolution record of atmospheric CO2 content from carbon isotopes in peat. Nature 365, 153 – 156. Wu, Xiangding, Lin, Zhenyao, 1981. Climatic change during the last 2000 years in Tibet. In: Central Weather Bureau (Ed.), Proceedings of Symposium on Climatic Change. Science Press, Beijing, pp. 18 – 25. Wu, Jinglu, Wang, Sumin, 1996. Carbon isotope in lake sediment and its paleoclimatic significance. Marin Geology and Quaternary Geology 16 (2), 103 – 109.
Wu, Yanhong, Wang, Sumin, Xia, Weilan, Zhu, Yuxin, Yin, Yu, 2001. Environmental variation in the central Tibetan Plateau in the last 200 years. Science in China (D) 44, 322 – 327 (suppl.). Yao, T., Shi, Yafeng, Thompson L.G., 1997a. High resolution record of paleoclimate since the little ice age from the Tibetan ice cores. Quaternary International 37, 19 – 23. Yao, T., Thompson, L., Shi, Yafeng, 1997b. Climate change since last post glacial recorded in Guliya ice core. Science in China (series D) 27 (3), 247 – 252. Zhang, Pingzhong, Wang, Xianbin, 1995. Hydrogen index and carbon isotope composition of lacustrine organic matter—Proxies to indicate lake level variation. Chinese Science Bulletin 40 (18), 1682 – 1685. Zhen, Du, Yang, Qinye, Liu, Yanhua, 1996. Natural environment and its distribution on Tibetan Plateau. In: Sun, Honglie (Ed.), The Formation and Evolution of the Tibetan Plateau. Shanghai Science and Technology Press, Shanghai, pp. 262 – 323.
