Postglacial changes in the Asian summer monsoon system: a pollenrecord from the eastern margin of the Tibetan Plateau
WEIJIAN ZHOU, SHI-YONGYU, GEORGE S. BURR, GEORGE J. KUKLA, A. J. T. JULL, FENG XIAN, JIAYI XIAO, STEVENM.COLMAN, HUAGUI YU, ZHAO LIU AND XIANGHUI KONG
BOREAS Zhou, W., Yu, S.-Y., Burr, G. S., Kukla, G. J., Jull, A. J. T., Xian, F., Xiao, J., Colman, S. M., Yu, H., Liu, Z. &Kong, X. 2010 (July): Postglacial changes in the Asian summer monsoon system: a pollen record from the easternmargin of the Tibetan Plateau. Boreas, Vol. 39, pp. 528–539. 10.1111/j.1502-3885.2010.00150.x. ISSN 0300-9483.
A new pollen record constrained by 32AMS radiocarbon dates from the Hongyuan peatland in the Zoige Basinreveals the long-term dynamics of an alpine wetland ecosystem on the eastern margin of the Tibetan Plateau overthe last 13 500 years. Changes in pollen assemblages and influxes suggest that local vegetation has experiencedthree distinct stages, from alpine coniferous forest–meadow (13 500–11 500 cal. aBP), through alpine coniferousforest (11 500–3000 cal. aBP), back to alpine coniferous forest–meadow (3000 cal. a BP–present). This record re-flects an ecosystem response along a transition zone where the South Asian and East Asian monsoon systems mayhave had different palaeoclimatic influences. A comparison of this record with other pollen records across theTibetan Plateau shows common features with regard to large-scale Holocene climatic changes, but highlights apattern of regional temporal and spatial variability that depends on the topography and position relative to theSouth Asian and East Asian monsoon fronts.
Weijian Zhou (e-mail: [emailprotected]), State Key Laboratory of Loess and Quaternary Geology, Instituteof Earth Environment, Chinese Academy of Sciences, Xi’an 710075, China, and Xi’an AMS Center of IEECAS andXi’an Jiaotong University, Xi’an 710049, China; Shi-Yong Yu (e-mail: [emailprotected]), Feng Xian, Huagui Yu,Zhao Liu and Xianghui Kong, State Key Laboratory of Loess and Quaternary Geology, Institute of Earth Environ-ment, Chinese Academy of Sciences, Xi’an 710075, China; George S. Burr and A. J. T. Jull, NSF-Arizona AMSLaboratory and Department of Physics, University of Arizona, Tucson, AZ 85721, USA; George J. Kukla, Lamont-Doherty Earth Observatory, Columbia University, Palisades, NY 10964, USA; Jiayi Xiao, Department ofGeographical Sciences, Nanjing Normal University, Nanjing 210097, China; Steven M. Colman, Large LakesObservatory and Department of Geological Sciences, University of Minnesota Duluth, Duluth, MN 55812, USA;received 2nd August 2009, accepted 27th January 2010.
The Asian monsoon, including two subsystems, namelythe South Asian monsoon (or the Indian monsoon) andthe East Asian monsoon, is a large-scale atmospheric cir-culation pattern that dominates climatic conditions in avast region ranging from maritime to inland Asia. Theseregions display distinct meteorological characteristics, butare tied to one another by the exchange of energy, moist-ure, and momentum of the atmosphere (Ding & Chan2005). Between the two regions lies a transition zone,stretching approximately along the eastern margin of theTibetan Plateau, where this interaction takes place (Wang& Lin 2002; Ding & Chan 2005). Little is known, how-ever, about how these two subsystems interact in thistransition zone at a longer time scale.
The Tibetan Plateau influences the atmosphere overthe Eurasia landmass in several ways. Essentially, itacts as a topographic barrier to general atmosphericcirculation, as a source of sensible heat, particularlyduring the summer season, and as a frictional surfaceinfluencing local atmospheric circulation (Ye & Gao1979). In winter, cold and dry northeastern winds de-rived from the Mongolian High dominate the weather,whereas warm and wet air masses from the tropicaloceans are dominant during summer. This settingmakes the Tibetan Plateau an ideal site for studying thelong-term dynamics of the Asian summer monsoon
system (Wang et al. 1993; Wei & Gasse 1999; Kanget al. 2000; Prokopenko & Catto 2005).
Previous analyses of sediment cores from a numberof lakes on the Tibetan Plateau have delineated a gen-eral picture of postglacial variations of the Asian sum-mer monsoon (Lister et al. 1991; Morinaga et al. 1993;Sun et al. 1993; Wang et al. 2002; Shen 2003; Morrill2004; Tang et al. 2004; Ji et al. 2005; Shen et al.2005a, b; Herzschuh et al. 2006, 2009; Kramer et al.2010), but the details of temporal and spatial variationsare still unclear. What is clear is that the TibetanPlateau has experienced significant climatic fluctuationsduring the Holocene. The precise timing of these chan-ges is, however, hindered by temporal uncertaintiesassociated with radiocarbon reservoir age errors whendating hard-water lakes (e.g. Van Campo et al. 1996;Henderson 2004; Morrill 2004) or pre-aged bulkorganic matter (cf. Yu et al. 2007).
Recent studies of oxygen isotopes in speleothemshave provided convincing evidence that the Asiansummer monsoon is controlled primarily by variationsin Northern Hemisphere summer insolation over thelong term (Dykoski et al. 2005; Wang et al. 2005, 2008).However, the geographical details of these changes canbe thoroughly understood only by synthesizing well-dated palaeoclimate records with sufficiently high
DOI 10.1111/j.1502-3885.2010.00150.x r 2010 The Authors, Journal compilation r 2010 The Boreas Collegium
mailto:[emailprotected]
mailto:[emailprotected]
mailto:[emailprotected]
mailto:[emailprotected]
mailto:[emailprotected]
mailto:[emailprotected]
mailto:[emailprotected]
mailto:[emailprotected]
mailto:[emailprotected]
mailto:[emailprotected]
temporal resolution (e.g. Morrill et al. 2003; Colmanet al. 2007; Yu et al. 2009). This paper focuses on pollenrecords from the northern extent of the Asian monsoondomain, as both the production and transport of pollenin this area are directly related to the summer monsoonduring the growing season, which is also an importantlimiting factor of the alpine ecosystem on the TibetanPlateau (Yu et al. 2001).
This study documents a record of Holocene climaticchanges in the transition zone along the eastern marginof the Tibetan Plateau (Fig. 1) that spans the last 13 500years. Our record is based on detailed palynologicalstudies and high-resolution AMS radiocarbon datingof a 4.5-m-long peat and clay sequence from the Hon-gyuan peatland in the Zoige Basin (Fig. 1). The Hon-gyuan peatland is the world’s largest high-altitudemarsh, covering an area of about 500 000 ha with athickness of peat varying from 3 to 5m (Thelaus 1992;Bjork & Thelaus 1996). Previous studies reveal thatsedge (Cyperaceae) peat accumulated continuously inthe Zoige Basin during the Holocene under cold and
wet conditions (Zhou et al. 2002). Records of peat ac-cumulation contain useful information about past cli-matic changes (Wang et al. 1993, 2004; Yan et al. 1999;Hong et al. 2003). The goal of this work is essentially torefine the chronology of climatic changes during theHolocene. This record also complements other palaeo-climatic records from the Tibetan Plateau, which, takentogether, may enhance our knowledge about the dy-namics of the Asian monsoon systems across the Tibe-tan Plateau during the Holocene.
Study area and site description
The Zoige Basin (latitude 321100–341100N, longitude1011450–1031250E) is located on the eastern edge of theTibetan Plateau, �2 km southwest of Hongyuan, thecapital of Hongyuan County (Fig. 1). It is an inter-montane basin controlled by three major fault zonesthat strike WNW, NE and NW. The average elevationof the basin is 3400ma.s.l. Previous drillings reveal that
Fig. 1. Map showing the locations of Holocene pollen records from the Tibetan Plateau. 1=Hongyuan peatland (this study); 2=LakeShayema (Jarvis 1993); 3=Lake Qinghai (Shen J et al. 2005); 4=Lake Hurleg (Zhao et al. 2007); 5=Lake Naleng (Kramer et al. 2010);6=Ren Co (Tang et al. 2004); 7=Lake Haideng (Tang et al. 2004); 8=Lake Zigetang (Herzschuh et al. 2006); 9=Selin Co (Sun et al. 1993);10=Lake Sumxi (Van Campo &Gasse 1993); 11=Bangong Co (Van Campo et al. 1996). The right upper map shows the detail of Hongyuanpeat coring site (solid circle) in the Zoige Basin, east Tibetan Plateau. AAM=Arctic airmass; WJS=westerly jet stream; SAM=South Asianmonsoon; EAM=East Asian monsoon.
BOREAS Postglacial changes in the Asian summer monsoon system, Tibetan Plateau 529
a large freshwater lake existed within the basin betweenabout 900 and 30 kaBP (Wang & Xue 1997; Chen et al.1999), leaving thick sediments (�300m) with an arealextent of about 6300 km2. Peat initiation (Yan et al.1999) suggests that the basin was drained at about30 kaBP owing to the piracy of the Yellow River, whenit cut through the mountain barrier on the east, andflowed out along a zigzag channel (Sun & Zhang 1987).
Temperature and precipitation are positively corre-lated in the Zoige Basin and both exhibit distinctlyseasonal variations, characteristic of a monsoonal cli-mate. Annual mean temperature is �11C, with thelowest monthly mean temperatures of �10.91C inJanuary and the highest monthly mean temperature of111C in July. The annual mean precipitation is�700mm, most of which occurs in summer months,providing favourable climatic conditions for sedgegrowth and thus peat accumulation.
Following a climate gradient, vegetation on theTibetan Plateau exhibits distinct zonality. The ZoigeBasin is located on the ecotones of temperate steppe,high-cold meadow and temperate desert, and localvegetation shows remarkable vertical variations (Wu1980). For example, most of the study area is vegetatedby high-cold meadow, where Carex muliensis andKobresia humilis are the two major species that formpeat. Other species include Polygonum viviparum andChamaesium paradoxum. Alpine shrub-meadow coversthe mountains above 3800ma.s.1., whereas mountainsbetween 3800 and 3000ma.s.l. are covered by alpineconiferous forests composed mainly of Picea purpurea,P. likiangensis and P. asperata as well as of Abies fabriand A. faxoniana. Mixed forests with Tsuga chinensis,Pinus densata, P. armandii, P. tabulaeformis, Betulaplatyphylla, Populus davidiana, Quercus liaotungensis,Q. baronii, Tilia intonsa, Fraxinus chinensis,Acer davidiiand Hippophae rhamnoides occur below 3000ma.s.l.
Methods
An undisturbed 4.5-m-long sequence of monolithic peatand clay complex was recovered from the Hongyuanpeatland in the Zoige Basin (3214604200N, 10213100000E),by digging a trench and pushing a tin box corer into thecleaned exposure. The core was logged in the Xi’an StateKey Laboratory of Loess and Quaternary Geologyusing greyscale measurements. The greyscale values ex-hibit cyclic changes parallel to total organic carbon(TOC) content along the core. Sediment description, de-tailed results of greyscale, and the TOC data have beenpublished elsewhere (Zhou et al. 2002).
Bulk peat samples for radiocarbon dating from se-lected levels were pretreated with 10% HCl and NaOHsolutions, and then rinsed repeatedly with distilled wa-ter through a 60-mm mesh sieve to remove fine sedgeroots. To minimize the ‘hard water effect’, we en-
deavoured to date remains of emerged wetland herbs inthis treeless swampy landscape. To this end, the re-sidual peat cellulose and leaf fragments with sizes lessthan 180 mm were picked out and dried in an electricoven for radiocarbon dating. Details of this method aregiven in Zhou et al. (2002). Dating was conducted in theAMS Laboratory at the University of Arizona. Thed13C value of these samples was also measured si-multaneously to infer the source of these herbaceousremains in terms of C3 versus C4 plants. Radiocarbondates were calibrated using the INTCAL 98 tree-ringdata set using the CALIB 4.2 computer program (Stuiveret al. 1998). The calibrated ages are quoted with twostandard deviation uncertainties. We formulated anage–depth model based on median-probability ages ofthe calibrated dates using the cubic-spline regressionmethod (Heegaard et al. 2005).
The core was subsampled at various intervals forpollen analyses. For example, the sections above 1mand below 4m were sampled at 2–4 cm intervals,whereas the section in between was sampled at 8-cmintervals. Aliquot bulk samples of 5 cm3 were processedfollowing the guidelines of Berglund & Ralska-Jasiewiczowa (1986). Samples were pretreated with a10% HCl solution to remove carbonates, and thentreated with 10% NaOH in a hot water bath to get ridof humic and fulvic acids. Prior to these chemicaltreatments, one tablet of exotic Lycopodium clavatumwith a known number of spores (e.g. 18 583 grains pertablet in this study) was added as marker grains toenable the calculation of fossil pollen concentrationsexpressed as grains cm�3 (Stockmarr 1971), whichwere then converted to pollen influxes by multiplyingaccumulation rates known from age–depth model.
Terrestrial pollen taxa, such as trees, shrubs and up-land herbs, are used to infer regional-scale changes invegetation. Their relative abundance is expressed as apercentage based on the sum of all palynomorphs ex-cluding wetland herbs, algae and ferns, which representlocal vegetation and thus can provide complementaryinformation about past climatic changes in terms ofwater-level fluctuations. For these local taxa, theirabundances are relative to the sum of all palyno-morphs. Given the poor pollen concentrations in thistreeless landscape, several slides were made for eachsample, and at least 600 pollen grains were counted toensure statistical significance for the relative abundanceof pollen taxa. For most of the samples, this numbercould usually be reached after counting three slides.There were only five samples with pollen counts lessthan 200 grains. These samples were included in thecalculation of pollen percentages for the completenessof the record, but were indicated in the pollen percentagediagram with arrows. Zonations of pollen assemblagesare based on terrestrial taxa using stratigraphically con-strained cluster analysis in the CONISS module of TGVIEW
2.0.2 (Grimm 1987).
530 Weijian Zhou et al. BOREAS
Results
Stratigraphy, chronology and sedimentation rates
Our trenching in the Zoige Basin reveals a peat–claydual complex (Fig. 2). We stopped at about 4.5m depthbecause of the waterlogged condition of the trench. Se-diments below 4.0m are grey-green silty clay withabundant remains of leaves. The section above this le-vel is brownish peat, comprising abundant un-decomposed sedge (Cyperaceae) remains. Dark-brownbands occur frequently along the core, revealing re-peated changes in redox conditions, probably asso-ciated with fluctuations in water level (Zhou et al.2002). Our coring reveals that these textural character-istics are similar to those in adjacent bogs (Yan et al.1999; Hong et al. 2003).
A total of 32AMS radiocarbon dates were obtained(Table 1). The average value of d13C of these samples isabout �27%, slightly above the end-member value(�30%) of C3 plants. This difference implies the pre-sence of the radiocarbon ‘hardwater effect’ in this sys-tem, which is estimated to be about 50 years accordingto the core-top age defined by regression analyses. Toobtain a precise chronology, this error was removed
Fig. 2. Biplot of calibrated AMS radiocarbon dates against depth,and calculated sedimentation rates. The heavy line is the age–depthmodel formulated by the cubic-spline fit. The grey-shaded envelopedenotes the 95% confidence level. Arrows indicate the timing of peatinitiation at this coring site.
Table 1. Radiocarbon dates of the Hongyuan peat section, east Tibetan Plateau.
Lab access ID Depth (cm) Materials dated d 13C (% vs. PDB) 14C age (a BP) Error (1s) 2s calibrated age range (a BP)
AA-29614 2 Peat cellulose �27.39 2245 50 2350–2130AA-29612 30 Peat cellulose �25.00 1540 40 1530–1330AA-29611 50 Peat cellulose �27.10 1705 45 1720–1520AA-29610 82 Peat cellulose �27.57 2725 50 2950–2750AA-29608 121 Peat cellulose �27.85 4310 55 5050–4700AA-29609 138 Peat cellulose �28.11 4235 55 4880–4570AA-29605 160 Peat cellulose �27.40 4815 55 5660–5330AA-29604 169 Peat cellulose �27.10 4970 55 5890–5590AA-29603 186 Peat cellulose �26.91 5810 65 6760–6440AA-29602 190 Peat cellulose �26.63 5840 110 6950–6350AA-29607 216 Peat cellulose �27.25 6040 60 7160–6720AA-29606 228 Peat cellulose �27.06 6280 65 7330–7000AA-29601 249 Peat cellulose �27.43 6745 60 7690–7480AA-29600 262 Peat cellulose �27.11 7035 65 7970–7690AA-29599 277 Peat cellulose �27.68 7985 70 9030–8630AA-29598 293 Peat cellulose �27.56 8200 70 9410–9010AA-29597 303 Peat cellulose �26.91 8455 70 9550–9280AA-29596 319 Peat cellulose �27.14 7480 75 8410–8060AA-29594 346 Peat cellulose �27.05 8835 90 10 200–9600AA-29593 349 Peat cellulose �27.02 8775 80 10 150–9550AA-29592 375 Peat cellulose �27.29 8850 70 10 190–9690AA-29591 383 Peat cellulose �26.93 9255 90 10 680–10 220AA-31639 386 Peat cellulose �26.71 9460 70 11 100–10 500AA-31640 390 Peat cellulose �27.27 9630 70 11 180–10 740AA-31641 395 Peat cellulose �27.38 9950 95 11 950–11 150AA-31642 400 Leaf fragments �27.66 10 315 70 12 850–11 650AA-29590 401 Leaf fragments �27.99 10 360 80 12 850–11 750AA-29589 414 Leaf fragments �26.83 10 185 100 12 450–11 250AA-31643 420 Leaf fragments �27.04 10 280 75 12 750–11 550AA-29588 434 Leaf fragments �27.38 11 040 95 13 400–12 650AA-31644 445 Leaf fragments �27.02 11 550 80 13 900–13 150AA-29587 449 Leaf fragments �26.91 11 395 85 13 800–13 000
BOREAS Postglacial changes in the Asian summer monsoon system, Tibetan Plateau 531
systematically from the modelled ages. The calibratedages are, with one exception, stratigraphically con-sistent along the core (Fig. 2), and provide a firmchronological framework for the past 13 500 years. Ac-cording to the age–depth model, peat did not accumu-late until 11 500 cal. a BP (Fig. 2), corresponding toregional climate amelioration following the YoungerDryas climatic event on the eastern margin of theTibetan Plateau (Lister et al. 1991; Gasse & VanCampo 1994; Yan et al. 1999). Rates of peat accumu-lation continued to rise and culminated at about9700 cal. a BP (Fig. 2), indicating a progressive increaseof wetland biomass in response to the gradualstrengthening of the summer monsoon (Jarvis 1993;Sirocko et al. 1993; Wang et al. 1999; Gupta et al. 2003;Morrill et al. 2003; Dykoski et al. 2005; Yu et al. 2009).
Pollen analysis
Pollen concentrations in this core are generally low, andfor most of the samples only about 600 pollen grainswere identified and counted. Nevertheless, this quantityis sufficiently statistically significant for pollen percen-tages to represent local vegetation (Yu et al. 2001).Furthermore, the local vegetation appears to have a lowspecies diversity. The tree and shrub taxa are dominatedby Abies, Picea, Pinus, Tsuga, Betula and Rosaceae.Artemisia, Asteraceae p.p., Ephedra and Poaceae are themajor upland taxa. The wetland taxa are mainlyCyperaceae, Myriophyllum, Ranunculus and Umbelli-fereae. Relative abundances of these species vary alongthe core (Fig. 3), and their changes in influxes are pre-sented in Fig. 4. A total of three pollen assemblage zonescan be assigned numerically, based on changes in bothpollen percentage and influx, and are referred to here asZ-1, Z-2 and Z-3.
Zone Z-1 (450–400cm; 13500–11500cal. aBP). – Pollenassemblages of this zone are dominated by coniferous treessuch asAbies,Picea andPinus and by upland herbs such asArtemisia and Asteraceae p.p. This zone can be further di-vided into two subzones. Subzone Z-1a (450–415cm;13500–12400cal. aBP) is marked by the dominance ofconiferous trees over upland herbs (Figs 3, 4). The percen-tage values of Abies pollen are high, accounting for60–80% of the terrestrial taxa. Asteraceae p.p. and Poa-ceae dominated the upland herb communities during thisperiod. Subzone Z-1b (415–400cm; 12400–11500ca-l. aBP) is characterized by a dramatic decrease in Abiespollen and a corresponding increase in Artemisia pollen(�80%).Other coniferous species, such asPicea,Pinus andTsuga, vanished, at the same time as the pollen of broad-leaved trees and shrubs disappeared. This sudden change inpollen abundances corresponds to the sharp transitionfrom silty clay to peat and a reversal of radiocarbon dates(Fig. 2), which may indicate a sedimentary hiatus.
Zone Z-2 (400–85 cm; 11 500–3000 cal. a BP). – Thiszone is represented by high values of Abies pollen(�80%). Values of wetland herb pollen, mainly Cyper-aceae, begin to decrease gradually at about 10 000 ca-l. aBP, and then increase after 6500 cal. aBP (Fig. 3).Two further subzones can be identified. Subzone Z-2a(400–190 cm; 11 500–6500 cal. aBP) is marked by a sub-stantial increase in Abies pollen at the lower boundary,along with significant reductions in the values of uplandherbs, such as Artemisia, Asteraceae p.p., Ephedraand Poaceae. The percentage value of Abies pollenthen remains nearly constant throughout this subzone(Fig. 3), whereas its influxes decrease gradually fromc. 9700 cal. aBP (Fig. 4). Broad-leaved tree andshrub pollen, such as Betula and Rosaceae, occurredinitially at about 10 000 cal. aBP, and their influxesreached maximum values at about 9700 cal. aBP.Values of Abies pollen remain nearly the same through-out Subzone Z-2b (190–85 cm; 8500–3000 cal.aBP), whereas the abundances of other coniferousspecies tend to decrease, with a gradual increase in up-land herb pollen. The influxes of Abies and Cyperaceaepollen continue to decrease throughout this subzone,and there are substantial reductions in the influxes ofother species.
Zone Z-3 (85–0 cm; 3000 cal. a BP–present). – Thiszone is characterized by substantial decreases in the in-fluxes of all tree species, and by increases in uplandherbs, such as Artemisia and Asteraceae p.p. In addi-tion, the influxes of wetland herbs, such as Cyperaceae,Ranunculus and Umbellifereae, increase slightly (Fig.4). A striking feature of this zone is the substantialreduction in the value of Abies pollen at the lowerboundary. Local vegetation was again dominated byupland herbs, including Artemisia, Asteraceae p.p.,Ephedra and Poaceae (Fig. 3). Along with this trend is asignificant increase in the values of wetland herb pollen,such as Cyperaceae, Ranunculus, Myriophyllum andUmbellifereae. Abundances of Cyperaceae pollen ac-count for about 80% of the wetland taxa, and this valueremains constant throughout the zone.
Discussion
Holocene climate and vegetation in the Zoige Basin
The well-dated pollen record from the Zoige Basinprovides an opportunity to infer regional-scale vegeta-tion changes on the eastern margin of the Tibetan Pla-teau over the past 13 500 years. The floristic diversity isgenerally low in this area: this is the nature of high-coldvegetation. We strongly argue that climate has driventhese changes, in terms of variations in the Asian sum-mer monsoon system, although human impact cannotbe totally ruled out. This appears to be true particularlyfor the last 3000 years, for which our pollen record
532 Weijian Zhou et al. BOREAS
Fig.3.Pollen
percentagediagram
fortheHongyuanpeatcore
intheZoigeBasin,east
TibetanPlateau.Theabundance
ofrare
speciesisexaggeratedby10times
andindicatedwithhatching.
Arrowsindicate
analysedlevelswithpollen
countslower
than200grains.
BOREAS Postglacial changes in the Asian summer monsoon system, Tibetan Plateau 533
Fig.4.Pollen
influxdiagram
fortheHongyuanpeatcore
intheZoigeBasin,eastTibetanPlateau.
534 Weijian Zhou et al. BOREAS
shows a substantial reduction in tree pollen. Anthro-pogenic deforestation during the late Holocene hasbeen reported in southwestern China (Dearing 2008),and traces of human activity have been found in theLake Qinghai area (Rhode et al. 2007). This areashould not be an exception, although archaeologicalevidence of human agricultural and/or pastoral activ-ities during the late Neolithic Age and early dynasticperiod has not been reported to date. In summary, ourpollen record shows that local vegetation has experi-enced three major stages.
Stage I (13 500–11 500 cal. a BP) – Alpine coniferousforest–meadow landscape. – The meadow communityis dominated by Artemisia and Asteraceae p.p., whichare the two major components of the high-cold mea-dow community that occurs on the western TibetanPlateau today (Yu et al. 2001). Abies, along with otherconifers and shrubs, appears to occur in the entire basinduring this period (Yan et al. 1999). Stratigraphically,this stage is characterized by a lacustrine environmentin this basin, representing cold and wet conditions thatcan be correlated with the Younger Dryas event, asobserved in the Guliya ice-core record from the north-west Tibetan Plateau (Thompson et al. 1989) and inclimate proxy records from the Chinese loess Plateau(Zhou et al. 1996, 1998, 2001).
Stage II (11 500–3000 cal. a BP) – Alpine coniferouslandscape. – Local vegetation was dominated byAbies. The establishment of alpine coniferous forests inthe catchment during this period is consistent withwarm and wet conditions, as expected for the mid-Holocene climate optimum (An et al. 2000). Pollenpercentage data indicate that the summer monsoon be-gan to be enhanced immediately after the YoungerDryas stadial (Figs 3, 4). Note that this inference mightbe unreliable because of the possible existence of asedimentary hiatus across the transition from silty clayto peat. However, pollen influx data indicate thatthe summer monsoon front did not reach this areauntil c. 10 800 cal. a BP, as supported by other studieselsewhere (e.g. Hong et al. 2003; Shen 2003; Shenet al. 2008).
Stage III (3000 cal. a BP–present) – Alpine coniferousforest–meadow landscape. – The gradual decreases inthe value of Abies pollen, along with the expansion ofupland herbs, indicate that alpine coniferous forest–meadow landscape was established in the catchmentonce again. This floristic change reveals a cooling trendduring the late Holocene (Herzschuh et al. 2006), whichcan be correlated with the substantial weakening of thesummer monsoon in this area (Jarvis 1993; Sirockoet al. 1993; Gupta et al. 2003; Morrill et al. 2003;Dykoski et al. 2005).
Other Holocene pollen records from the same basinshow similar characteristics. For example, Yan et al.(1999) described four distinct stages of local vegetationsuccession over the past 14 200 years. Hong et al.(2003) described a similar pattern of climatic changesfrom this area based on d13C measurements of sedgecellulose, including single species. The timing of theirisotopic record broadly corresponds to that of our re-cord. Our record is also consistent with the pollen dataof Shen C.M. et al. (2005), who described a long pollenrecord from the Hongyuan area. Their record docu-ments 18 pollen zones that reflect climatic changesover the past 180 ka. The uppermost section of theircore overlaps with our peat sequence and has abroadly similar pollen zonation, but it does not havesufficient temporal control to enable a close compar-ison with our record. Their record does show that theHongyuan region has experienced repeated and ex-treme floristic variability through time, and that theinferred climatic changes during the Holocene havebeen relatively mild, compared with those of earliertimes.
In a comprehensive study of the relationship be-tween pollen assemblage and climate on the easternTibetan Plateau, Shen et al. (2006) demonstrated thatthe two dominant climatic factors controlling varia-tions in the modern pollen assemblage are annual pre-cipitation and summer (July) temperature. Theyanalysed 227 surface samples from sites where me-teorological data were available. Annual precipitationshows a strong positive correlation with arboreal taxa,which include Abies, Picea, Pinus and Tsuga, and anegative correlation with herbal taxa, such as Com-positae, Artemisia and Gramineae. Summer tempera-ture is positively correlated with all of the arborealtaxa, as well as with Artemisia, and is negatively cor-related with Compositae and Gramineae. Their workjustifies the use of coniferous tree pollen as a reliableclimate proxy in this area. Moreover, the pollen ofconifers, such as Abies and Picea, appears to be de-rived locally rather than from long-distance transport(Yu et al. 2001). Here we use the percentage of Abiespollen as a proxy for the summer monsoon intensity.This record is shown in Fig. 5A. It is generally con-sistent with the reconstructions of the Indian summermonsoon from Dongge Cave (Fig. 5B, Dykoski et al.2005) and Shanbao Cave (Fig. 5C, Shao et al. 2006),but is distinct from the Lake Qinghai record (Fig. 5D,Lister et al. 1991), which represents the variations ofthe East Asian summer monsoon. The similarity ofour Abies pollen record from the Hongyuan peatlandand the d18O records from the Dongge and Shanbaocaves reflects the influence of the South Asian mon-soon on climatic conditions in this area. The primaryfeature of our pollen record is a marked period ofstable and relatively high effective moisture prevailingfrom 11 500 to 3000 cal. a BP. This agrees closely
BOREAS Postglacial changes in the Asian summer monsoon system, Tibetan Plateau 535
with the d18O records of Dongge and Shanbao caves,now generally regarded as a good proxy of the SouthAsian monsoon. The oxygen isotope record from LakeQinghai shows a different trend between 11 500and 8000 cal. a BP, as compared with the other threerecords, with progressively higher d18O values duringthis period. This deviation is interpreted here as re-flecting regional differences in climatic conditions be-tween the various sites. The most obvious cause forsuch variability is the relative contributions made bythe South Asian and East Asian monsoon systems.
Holocene climatic variability across the Tibetan Plateauas a function of topography
Speleothem studies, such as the Dongge cave (Dykoskiet al. 2005; Wang et al. 2005) and the Shanbao cave(Shao et al. 2006; Wang et al. 2008) d18O records, pro-vide a persuasive argument that summer monsoonvariability in China is controlled primarily by theNorthern Hemisphere summer insolation on orbitaltime scales. However, regional climatic fluctuations,such as those seen between Lake Qinghai and theZoige Basin, are probably influenced by geographicdifferences, especially regarding topography andposition relative to the South Asian or East Asianmonsoon fronts. Because the South Asian and East
Asian monsoon systems are not synchronized, therewill be times when they are out of phase (Hong et al.2003), and regional variations may be amplified at thesetimes. When this occurs, different parts of the TibetanPlateau should be influenced, to a greater or lesser ex-tent, by either the South Asian or East Asian monsoonsystem, according to their location. Such an effect couldcontribute to long-term climatic effects such as anasynchronous Holocene climate optimum (An et al.2000; He et al. 2004).
In order to examine this possibility further, we plot-ted inferred relative palaeoclimate information from anumber of pollen records on the Tibetan Plateau, alongwith the Hongyuan pollen data (Fig. 6). The locationsof these studies include sites from western Tibet (VanCampo & Gasse 1993; Van Campo et al. 1996), centralTibet (Sun et al. 1993; Tang et al. 2004; Herzschuh et al.2006), and eastern Tibet (Jarvis 1993; Shen et al. 2005b;Zhao et al. 2007; Kramer et al. 2010), as shown in Fig.1. The compilation plotted in Fig. 6 identifies relativegrowth conditions from each site, as interpreted by theauthors of the studies. Absolute growth conditions varyfrom site to site, but relative optimal conditions are wellcharacterized, with the most favourable growth condi-tions generally corresponding to the mid-Holoceneoptimum. Sustained temporal changes in growth condi-tions for each site may be induced by long-term changesin summer precipitation or temperature. Highly variablegrowth conditions are also shown in Fig. 6 for somesites, according to published interpretations. The sitesare arranged by decreasing elevation.
Figure 6 offers a plateau-wide picture of Holoceneclimatic changes. All of the sites show persistent chan-ges during the Holocene, with up to six identifiablepollen zones, or subzones, during that time. A distinctregional variability is observed in the timing, durationand stability of climatic conditions across the plateau.However, no consistent north–south or east–westvariability is discernible from this small data set. Incontrast, a topographic effect is apparent. This is seenin the mid-Holocene climate optimum, which startedearlier and ended later at the lower-elevation site. Si-tuated in a desert setting, Lake Hurleg appears to be anexception, where dry climate prevailed during the mid-dle Holocene. The Hurleg Lake site is different from theother sites in other respects as well, as emphasized byZhao et al. (2007). In this context, the most notabledifference in the Hurleg Lake site is the improvement ingrowth conditions in the late Holocene, in sharp con-trast to the progressively drier climate exhibited at mostother sites. Temperature, as a limiting factor for vege-tation growth, appears to be a function of elevation -the higher the elevation, the lower the temperature.Therefore, these elevation differences should inevitablyresult in a spatial variation in climatic conditions. It isalso true that topographic features directly influencemodern monsoon rainfall (Hoyos & Webster 2007).
Fig. 5. Comparison of palaeoclimate records from the east TibetanPlateau and neighbouring areas. A.Abies pollen percentage data fromthis study. B. Dongge cave speleothem d18O record (Dykoski et al.2005). C. Shanbao cave speleothem d18O record (Shao et al. 2006). D.Lake Qinghai ostracode d18O record (Lister et al. 1991).
536 Weijian Zhou et al. BOREAS
Conclusions
A detailed and well-dated pollen record from the ZoigeBasin sheds new light on the postglacial dynamics of thealpine ecosystem on the eastern Tibetan Plateau. Localvegetation has experienced significant changes, fromalpine coniferous forest–meadow, through alpine con-iferous forests, back to an alpine coniferous forest–meadow landscape during the last 13 500 years, pre-sumably regulated by the rise and fall of the Asiansummer monsoon system. The pollen record from theZoige Basin is broadly consistent with oxygen isotoperecords from Donge and Shanbao caves, and highlightsdifferences between Holocene climate histories at LakeQinghai. The timing, nature and duration of climaticchanges as expressed in pollen records across theTibetan Plateau suggest a relationship with sample ele-vation, and highlight distinct geographic differences.All of the sites examined in this study are strongly in-fluenced by monsoon precipitation, and this is probablythe major controller of Holocene vegetation on theTibetan Plateau.
Acknowledgements. – This project was supported by the NationalScience Foundation of China and the National Basic Research Pro-gram of China. A portion of the work was funded by the USNationalScience Foundation (Grant No. 0622305). We thank Professor Z. S.An and Professor F. B. Wang for their kind support and guidance.We are also grateful to Professor J. N. Haas, Professor J. A. Pio-
trowski, and an anonymous reviewer for their technical and linguisticimprovements to the manuscript.
References
An, Z. S., Porter, S. C., Kutzbach, J. E., Wu, X., Wang, S., Liu, X.,Li, X. & Zhou, W. J. 2000: Asynchronous Holocene optimumof the East Asian monsoon. Quaternary Science Reviews 19,743–762.
Berglund, B. E. & Ralska-Jasiewiczowa, M. R. 1986: Pollen analysisand pollen diagrams. In Berglund, B. E. (ed.): Handbook of Holo-cene Palaeoecology and Palaeohydrology, 455–484. John Wiley &Sons, Chichester.
Bjork, S. & Thelaus, M. 1996: Peatlands in Kachudo, Tibet: appliedwetland research. Ymer 116, 112–131.
Chen, F. H., Liu, G. X., Bloemendal, J. & Zhang, P. Z. 1999: An 800ky proxy record of climate from lake sediments of the Zoige Basin,eastern Tibetan Plateau. Palaeogeography, Palaeoclimatology,Palaeoecology 151, 307–320.
Colman, S. M., Yu, S.-Y., An, Z. S., Shen, J. & Henderson, A. C. G.2007: Late Cenozoic climate changes in China’s western interior: areview of research on Lake Qinghai and comparison with otherrecords. Quaternary Science Reviews 26, 2281–2300.
Dearing, J. A. 2008: Landscape change and resilience theory: a pa-laeoenvironmental assessment from Yunnan, SW China. TheHolocene 18, 117–127.
Ding, Y. & Chan, J. C. 2005: The East Asian summer monsoon: anoverview.Meteorology and Atmospheric Physics 89, 117–142.
Dykoski, C. A., Edwards, R. L., Cheng, H., Yuan, D., Cai, Y.,Zhang, M., Lin, Y., Qing, J., An, Z. S. & Revenaugh, J. 2005: Ahigh-resolution, absolute-dated Holocene and deglacial Asianmonsoon record from Dongge Cave, China. Earth and PlanetaryScience Letters 233, 71–86.
Fig. 6. Summary diagram of Holocene pollenstudies from the Tibetan Plateau. The sites arearranged by elevation. Site-specific relativegrowth conditions are shown for each record.Relative growth conditions were taken fromeach reference (see text). Locations of the sitesare shown in Fig. 1. Triangles denote 14C agecontrol.
BOREAS Postglacial changes in the Asian summer monsoon system, Tibetan Plateau 537
Gasse, F. & Van Campo, E. 1994: Abrupt postglacial climate eventsin West Asia and North Africa monsoon domains. Earth andPlanetary Science Letters 126, 435–456.
Grimm, E. C. 1987: CONISS: a FORTRAN 77 program for strati-graphical constrained cluster analysis by the method of incrementalsum of squares. Computers & Geosciences 13, 13–35.
Gupta, A. K., Anderson, D. M. & Overpeck, J. T. 2003: Abruptchanges in the Asian southwest monsoon during the Holocene andtheir links to the North Atlantic Ocean. Nature 421, 354–357.
He, Y., Theakstone, W. H., Zhang, Z. L., Zhang, D. A., Yao, T. D.,Chen, T., Shen, Y. P. & Pang, H. X. 2004: Asynchronous Holoceneclimatic change across China. Quaternary Research 61, 52–63.
Heegaard, E., Birks, H. J. B. & Telford, R. J. 2005: Relationships be-tween calibrated ages and depth in stratigraphical sequences: anestimation procedure by mixed-effect regression. The Holocene 15,612–618.
Henderson, A. C. G. 2004: Late Holocene environmental change on theNE Tibetan Plateau: a palaeolimnological study of Lake Qinghai andLake Gahai, China, based on stable isotopes. Ph.D. thesis, Uni-versity College London, 222 pp.
Herzschuh, U., Kramer, A., Mischke, S. & Zhang, C. 2009: Quanti-tative climate and vegetation trends since the late glacial on thenortheastern Tibetan Plateau deduced from Koucha Lake pollenspectra. Quaternary Research 71, 162–171.
Herzschuh, U., Winter, K., Wunnemann, B. & Li, S. 2006: A generalcooling trend on the central Tibetan Plateau throughout the Holo-cene recorded by the Lake Zigetang pollen spectra. QuaternaryInternational 154/155, 113–121.
Hong, Y. T., Shibata, Y., Hirota, M., Uchida, M., Leng, X. T., Jiang,H. B., Xu, H., Wang, H., Yi, L., Hong, B., Lin, Q. H. & Zhu, Y. X.2003: Correlation between Indian Ocean summer monsoon andNorth Atlantic climate during the Holocene. Earth and PlanetaryScience Letters 211, 371–380.
Hoyos, C. D. & Webster, P. J. 2007: The role of intraseasonal varia-bility in the nature of Asian monsoon precipitation. Journal ofClimate 20, 4402–4424.
Jarvis, D. I. 1993: Pollen evidence of changing Holocene monsoonclimate in Sichuan Province, China. Quaternary Research 39,325–337.
Ji, J., Chen, J., Liu, L., Liu, X., Shen, J. & Balsam, W. 2005: Asianmonsoon oscillations in the northeastern Qinghai–Tibet Plateausince the late glacial as interpreted from visible reflectance ofQinghai Lake sediments. Earth and Planetary Science Letters 233,61–70.
Kang, S. C., Wake, C. P., Qin, D. H., Mayewski, P. A. & Yao, T. D.2000: Monsoon and dust signals recorded in Dasuopu glacier,Tibetan Plateau. Journal of Glaciology 46, 222–226.
Kramer, A., Herzschuh, U., Mischke, S. & Zhang, C. 2010: Lateglacial vegetation and climate oscillations on the southeasternTibetan Plateau infererred from the Lake Naleng pollen profile.Quaternary Research 73, 324–335.
Lister, G. S., Kelts, K. R., Chen, K. Z., Yu, J. Q. & Niessen, F. 1991:Lake Qinghai, China: Closed-basin lake levels and the oxygen iso-tope record for ostracoda since the latest Pleistocene. Palaeogeo-graphy, Palaeoclimatology, Palaeoecology 84, 141–162.
Morinaga, H., Itota, C., Isezaki, N., Goto, H., Yaskawa, K., Kusa-kabe, M., Liu, J., Gu, Z., Yuan, B. & Cong, S. 1993: O-18 and C-13records for the last 14,000 years from lacustrine carbonates of SilingCo (Lake) in the Qinghai–Tibetan Plateau. Geophysical ResearchLetters 20, 2909–2912.
Morrill, C. 2004: The influence of Asian summer monsoon variabilityon the water balance of a Tibetan lake. Journal of Paleolimnology32, 273–286.
Morrill, C., Overpeck, J. T. & Cole, J. E. 2003: A synthesis of abruptchanges in the Asian summer monsoon since the last deglaciation.The Holocene 13, 465–476.
Prokopenko, A. & Catto, N. 2005: Paleolimnological records fromAsian lacustrine systems: archives of large-scale seasonal atmo-spheric circulation changes in the Northern Hemisphere. Qua-ternary International 136, 1–4.
Rhode, D., Zhang, H., Madsen, D. B., Gao, X., Brantingham, J., Ma,H. & Olsen, J. W. 2007: Epipaleolithic/early Neolithic settlements
at Qinghai Lake, western China. Journal of Archaeological Science34, 600–612.
Shao, X. H., Wang, Y. J., Cheng, H., Kong, X. G., Wu, J. Y. &Edwards, R. L. 2006: Long-term trend and abrupt events of theHolocene Asian monsoon inferred from a stalagmite d18O recordfrom Shennongjia in Central China. Chinese Science Bulletin 51,221–228 (in Chinese, with English abstract).
Shen, C. M. 2003: Millennial-scale variations and centennial-scaleevents in the Southwest Asian monsoon: Pollen evidence from Tibet.Ph.D. thesis, Louisiana State University, 286 pp.
Shen, C. M., Liu, K.-B., Tang, L. & Overpeck, J. T. 2006: Quantita-tive relationships between modern pollen rain and climate in theTibetan Plateau. Review of Palaeobotany and Palynology 140,61–77.
Shen, C. M., Liu, K.-B., Tang, L. & Overpeck, J. T. 2008: Numericalanalysis of modern and fossil pollen data from the TibetanPlateau. Annals of the Association of American Geographers 98,755–772.
Shen, C. M., Tang, L., Wang, S., Li, C. & Liu, K.-B. 2005a: Pollenrecords and time scale for the RM core of the Zoige Basin, north-eastern Qinghai–Tibetan Plateau. Chinese Science Bulletin 50,553–562.
Shen, J., Liu, X. Q., Wang, S. M. & Matsumoto, R. 2005b: Palaeo-climatic changes in the Qinghai Lake area during the last 18,000years. Quaternary International 136, 131–140.
Sirocko, F., Sarnthein, M., Erlenkeuser, H., Lange, H., Arnold, M. &Duplessy, J. C. 1993: Century-scale events in monsoonal climateover the past 24,000 years. Nature 364, 322–324.
Stockmarr, J. 1971: Tablets with spores used in absolute pollen ana-lysis. Pollen et Spores 3, 615–621.
Stuiver, M., Reimer, P. J., Bard, E., Beck, J. W., Burr, G. S., Hughen,K. A., Kromer, B., McCormac, G., van der Plicht, J. & Spurk, M.1998: INTCAL98 radiocarbon age calibration, 24,000–0 cal BP.Radiocarbon 40, 1041–1083.
Sun, G. Y. & Zhang, W. F. 1987: Old channels of the HuangheRiver on Ruoergai Plateau and its palaeogeographic significance.Scientia Geographica Sinica 7, 266–272 (in Chinese, with Englishabstract).
Sun, X., Du, N., Chen, Y., Gu, Z., Liu, J. & Yuan, B. 1993: Holocenepalynological records in Lake Selincuo, Northern Xizang. ActaBotanica Sinica 35, 943–950 (in Chinese, with English abstract).
Tang, L., Shen, C. M., Liu, K.-B., Yu, S.-Y. & Li, C. H. 2004: Cli-matic changes in southeastern Tibet since the Last Glacial Max-imum – pollen records from the southeastern Tibetan Plateau.Science in China, Series D 34, 436–442 (in Chinese).
Thelaus, M. 1992: Some characteristics of the mire development inHongyuan County, eastern Tibetan Plateau. Proceedings of the 9thInternational Peat Congress 1, 334–351, International Peat Society,Uppsala.
Thompson, L. G., Mosley-Thompson, E., Davis, M. E., Bolzan, J. F.,Dai, J., Yao, T., Gundestrup, N., Wu, X., Klein, L. & Xie, Z. 1989:Holocene–Late Pleistocene climatic ice core records from Qin-ghai–Tibetan Plateau. Science 246, 474–477.
Van Campo, E. & Gasse, F. 1993: Pollen- and diatom-inferred cli-matic and hydrological changes in Sumxi Co Basin (Western Tibet)since 13,000 yr BP. Quaternary Research 39, 300–313.
Van Campo, E., Cour, P. & Sixuan, H. 1996: Holocene environmentalchanges in Bangong Co basin (Western Tibet). Part 2: The pollenrecord. Palaeogeography, Palaeoclimatology, Palaeoecology 120,49–63.
Wang, B. & Lin, H. 2002: Rainy season of the Asian–Pacific summermonsoon. Journal of Climate 15, 386–398.
Wang, F. B., Yan, G. & Lin, B. H. 1993: Organic-carbon stable iso-topic composition from peat in Ruoergai Plateau, eastern part ofQinghai-Xizang Plateau. Chinese Science Bulletin 38, 1114–1118.
Wang, H., Hong, Y. T., Zhu, Y. X., Hong, B., Lin, Q. H., Hai, X.,Leng, X. T. & Mao, X. M. 2004: Humification degrees of peatin Qinghai–Xizang Plateau and palaeoclimate change. ChineseScience Bulletin 49, 514–519.
Wang, L. J., Sarnthein, M., Erlenkeuser, H., Grootes, P. M., Grimalt,J. O., Pelejero, C. & Linck, G. 1999: Holocene variations in Asianmonsoon moisture: a bidecadal sediment record from the SouthChina Sea. Geophysical Research Letters 26, 2889–2892.
538 Weijian Zhou et al. BOREAS
Wang, R. L., Scarpitta, S. C., Zhang, S. C. & Zheng, M. P. 2002:Later Pleistocene/Holocene climate conditions of Qinghai-XizhangPlateau (Tibet) based on carbon and oxygen stable isotopes ofZabuye Lake sediments. Earth and Planetary Science Letters 203,461–477.
Wang, S. M. & Xue, B. 1997: Environmental evolution of Zoige Basinsince 900 ka B.P. and comparison study with Loess Plateau. Sciencein China, Series D 40, 329–336.
Wang, Y. J., Cheng, H., Edwards, R. L., He, Y. Q., Kong, X. G., An,Z. S., Wu, J. Y., Kelly, M. J., Dykoski, C. A. & Li, X. D. 2005: TheHolocene Asian monsoon: Links to solar changes and NorthAtlantic climate. Science 308, 854–857.
Wang, Y. J., Cheng, H., Edwards, R. L., Kong, X. G., Shao, X. H.,Chen, S. T., Wu, J. Y., Jiang, X. Y., Wang, X. F. & An, Z. S. 2008:Millennial- and orbital-scale changes in the East Asian monsoonover the past 224,000 years. Nature 451, 1090–1093.
Wei, K. &Gasse, F. 1999: Oxygen isotopes in lacustrine carbonates ofWest China revisited: implications for post glacial changes in sum-mer monsoon circulation. Quaternary Science Reviews 18,1315–1334.
Wu, C.-Y. 1980: Vegetation of China. 1382 pp. Science Press, Beijing(in Chinese).
Yan, G., Wang, F. B., Shi, G. R. & Li, S. F. 1999: Palynological andstable isotopic study of palaeoenvironmental changes on thenortheastern Tibetan plateau in the last 30,000 years. Palaeogeo-graphy, Palaeoclimatology, Palaeoecology 153, 147–159.
Ye, D. & Gao, Y. 1979: The Meteorology of the Qinghai–Xizang Pla-teau. 278 pp. Science Press, Beijing (in Chinese).
Yu, G., Tang, L., Yang, X. D., Ke, X. K. & Harrison, S. P. 2001:Modern pollen samples from alpine vegetation on the Tibetan Pla-teau. Global Ecology and Biogeography 10, 503–520.
Yu, S.-Y., Ricketts, R. D. & Colman, S. M. 2009: Determining thespatial and temporal patterns of climate changes in China’s westerninterior during the last 15 ka from lacustrine oxygen isotope re-cords. Journal of Quaternary Science 24, 237–247.
Yu, S.-Y., Shen, J. & Colman, S. M. 2007: Modeling the radiocarbonreservoir effect in lacustrine systems. Radiocarbon 49, 1241–1254.
Zhao, Y., Yu, Z. C., Chen, F., Ito, E. & Zhao, C. 2007: Holocene vege-tation and climate history at Hurleg Lake in the Qaidam Basin, north-west China. Review of Palaeobotany and Palynology 145, 275–288.
Zhou, W. J., An, Z. S., Jull, A. J. T., Donahue, D. J. & Head, M. J.1998: Reappraisal of Chinese Loess Plateau stratigraphic sequencesover the last 30,000 years: Precursors of an important Holocenemonsoon climatic event. Radiocarbon 40, 905–913.
Zhou, W. J., Donahue, D. J., Porter, S. C., Jull, A. J. T., Li, X. Q.,Stuiver, M., An, Z. S., Matsumoto, E. & Dong, G. R. 1996: Varia-bility of monsoon climate in East Asia at the end of the last glacia-tion. Quaternary Research 46, 219–229.
Zhou, W. J., Head, M. J., An, Z. S., De Deckker, P., Liu, Z. Y., Liu,X. D., Lu, X. F., Donahue, D., Jull, A. J. T. & Beck, J. W. 2001:Terrestrial evidence for a spatial structure of tropical–polar inter-connections during the Younger Dryas episode. Earth and Plane-tary Science Letters 191, 231–239.
Zhou, W. J., Lu, X. F., Wu, Z. K., Deng, L., Jull, A. J. T., Donahue,D. & Beck, J. W. 2002: Peat record reflecting Holocene climaticchange in the Zoige Plateau and AMS radiocarbon dating. ChineseScience Bulletin 47, 66–70.
BOREAS Postglacial changes in the Asian summer monsoon system, Tibetan Plateau 539
