1
行政院國家科學委員會補助專題研究計畫
;
成 果 報 告
□期中進度報告
臺灣高山湖泊與東亞夏季風西界及西風帶湖泊之古氣候記錄
對比(IV)
計畫類別：
;
個別型計畫 □整合型計畫
計畫編號：NSC 91－2611－M－110－010－
執行期間： 91 年 8 月 1 日至 92 年 7 月 31 日
計畫主持人：陳鎮東教授
共同主持人：
計畫參與人員：
郭富雯、許志宏、王冰潔、羅立章、邢麗玉、候偉萍、曾筱
君、蔡長利
成果報告類型(依經費核定清單規定繳交)：
;
精簡報告 □完整報告
本成果報告包括以下應繳交之附件：
□赴國外出差或研習心得報告一份
□赴大陸地區出差或研習心得報告一份
■出席國際學術會議心得報告及發表之論文各一份
□國際合作研究計畫國外研究報告書一份
處理方式：除產學合作研究計畫、提升產業技術及人才培育研究計畫、
列管計畫及下列情形者外，得立即公開查詢
□涉及專利或其他智慧財產權，□一年□二年後可公開查詢
執行單位：
中山大學海洋地質及化學所
中 華 民 國 九十二 年 十 月 十三 日
2
本計畫
成果發表在 Palaeogeography, Palaeoclimatology, Palaeoecology, 193,
181-200, 2003 期刊上, 全文請見下頁。
出席國際學術會議心得報告緊接在後。
另有一篇” Deconvoluting past seawater temperature and transparency records
locked in corals in Nanwan Bay, southern Taiwan 正在準備中。
The Dry Holocene Megathermal in Inner Mongolia
Chen-Tung A. Chen
a;, Hsin-Chi Lan
a, Jiann-Yuh Lou
a;1, Yan-Cheng Chen
ba Institute of Marine Geology and Chemistry, National Sun Yat-sen University, Kaohsiung 80424, Taiwan b Institute of Geological Research and Chemical Mines, Ministry of Chemical Industries, Zhuozhou 072754, PR China
Received 20 March 2002; received in revised form 2 December 2002; accepted 12 December 2002
Abstract
The paleoclimate since 14 kyr BP (14C age) was reconstructed based on a 16.22-m-long sediment core collected
from Lake Yanhaizi, a saline lake located near the northern limit of the East Asian summer monsoon in Inner Mongolia. Coarse sediments were deposited there during a shrinkage phase of the lake when sand dunes reactivated. These sediments have low organic carbon contents but high maturity indices, indicating that they were deposited in an arid environment. By contrast, based on high organic contents and low maturity indices, fine sediments were deposited during periods of high lake stand in a humid environment. It was in general dry between 8.0 and 4.3 kyr BP. The above dry and wet phases are consistent with those recovered from the arid^semiarid transition zone elsewhere, but are unlike the widely perceived humid Holocene Megathermal reported in east China and the newly reconstructed record in the alpine Retreat Lake in Taiwan. The discrepancy may be due to a relative insensitivity to humidity changes in these two areas since they have both been under the total influence of the summer monsoon. On the other hand, much enhanced evaporation over higher monsoon precipitation at Lake Yanhaizi reduces the effective humidity in the warm climate near the northern boundary of the summer monsoon. This also accounts for the fact the high-temperature Holocene Megathermal, as revealed in the Okinawa Trough and the northern South China Sea, is correlated to the dry phases at Lake Yanhaizi. Conversely, the 4^2-kyr BP coldest period in the Holocene corresponds to a wet phase at Lake Yanhaizi.
6 2003 Elsevier Science B.V. All rights reserved.
Keywords: Holocene Megathermal; Inner Mongolia ; paleoclimate; East Asian summer monsoon; e¡ective humidity
1. Introduction
The climate of eastern Asia is largely controlled by the East Asian monsoon system which re-sponds to the strength of high-and low-pressure
cells growing and decaying seasonally over the Asian landmass. The monsoon draws winds and moisture mainly from the tropical Philippine and South China seas at its northern boundary of In-ner Mongolia during summer, and this results in more precipitation over the northwestern interior of China. The present-day landward limit of summer monsoon precipitation lies in the northeast^southwest direction in Inner Mongolia (Gao, 1962). Consequently, distinct distributions of loess, sandy loess and sand dune landforms
0031-0182 / 03 / $ ^ see front matter 6 2003 Elsevier Science B.V. All rights reserved. doi :10.1016/S0031-0182(03)00225-6
1 Present address: Department of Marine Science, Chinese
Naval Academy, P.O. Box 90175, Tsoying, Kaohsiung, Tai-wan.
* Corresponding author. Fax: +886-7-525-5346.
E-mail address:[email protected](C.-T.A. Chen).
PALAEO 3031 7-3-03
Palaeogeography, Palaeoclimatology, Palaeoecology 193 (2003) 181^200
which re£ect the decreasing relative humidity ex-tend from the southeast to the northwest. As a result, the waxing and waning of the summer monsoon intensity control the availability of pre-cipitation in Inner Mongolia and re£ect the hor-izontal migration of the landforms.
During the early to mid-Holocene, the condi-tions that prevailed were, for the most part, wet-ter than those today over monsoonal areas, such as those in the Sahara, the Arabian Peninsula, India and eastern Asia (Ritchie and Haynes, 1987 ; Kutzbach et al., 1996; Enzel et al., 1999). This was in response to increased summer insola-tion (Kutzbach and Otto-Bliesner, 1982 ; Kutz-bach and Street-Perrott, 1985 ; COHMAP Mem-bers, 1988). A wealth of records from the Chinese Loess Plateau has depicted this past scenario of the East Asian summer monsoon-in£uenced areas as warm and humid during high-temperature in-terglacials when the summer monsoon strength-ened (Kukla et al., 1988; An et al., 1990) ; records have also demonstrated that the most humid pe-riod of the Holocene was during the warmest Ho-locene Megathermal in eastern China (Shi et al., 1993). On the other hand, during the last gla-ciation, global cooling enhanced the Mongolian High-Pressure Zone. At that time, the winter monsoon was greatly strengthened and the sum-mer monsoon signi¢cantly weakened (An et al., 1990). However, these records mainly came from the intensively studied Loess Plateau.
Although research on the Mu Us Desert, to the north of the loess, has been carried out in recent years, most of this has been limited to the south-ern margin of the desert and has poor temporal resolution during the Holocene (Sun et al., 1998a,b, 1999 ; Sun, 2000 ; Li et al., 2000). The principal objective of this project is to document and understand decade- to century-scale paleocli-matic variations using detailed sedimentary rec-ords from saline lakes in the deserts of Inner Mongolia. This obviously takes advantage of the fact that closed-basin lakes in the deserts are sen-sitive to the balance between precipitation and evaporation, which is directly linked to atmo-spheric circulation (Street-Perrott and Roberts, 1983). Analysis of the lacustrine records in the deserts could address how temporal and regional
variations with respect to the monsoon respond to longer-term shifts in the monsoonal paleoclimate, thereby providing high-resolution records that cover a su⁄ciently long time scale and fully cap-ture the natural range of variability in the mon-soonal system. Additionally, it will o¡er informa-tion that is essential for understanding near-term future climatic variations in the monsoonal re-gion. The sediments in the studied lake have in-deed preserved a high-resolution record over the last 14 000 yr BP, one of the most complete and detailed sets of Holocene paleoclimatic records obtained within the eastern Asian monsoonal re-gion.
2. Study area
Lake Yanhaizi (108‡25PE^108‡29PE, 40‡06Pc40‡ 10PN, 1180 m a.s.l., Fig. 1a) is one of the many hypersaline lakes located 500 km west of Beijing on the Ordos Plateau, Inner Mongolia. It is 800 km from the nearest ocean, the Bohai Sea, and about 1150 km from the East China Sea. The lake covers a maximum area of 18 km2, has a maxi-mum water depth of about 0.5 m in summer, and a drainage area of about 2000 km2 in the Mu Us Desert (Zhang et al., 1992).
The mean annual temperature is 5.9‡C, with a low mean monthly temperature of 312.2‡C in January and a high mean monthly temperature of 21.6‡C in July. Climatically, the strong season-al in£uence of the East Asian monsoon reaches the area during the summer, resulting in rainfall from July to September (Zhang and Lin, 1992). The mean annual precipitation is only 277 mm, concentrated between July and September. How-ever, the mean annual evaporation is 2604 mm, mostly between April and August. The bedrock of the lake is mainly composed of Lower Cretaceous yellow^greenish sandstones belonging to the Dong Sheng and Yijinghuolou formations.
3. Materials and methods
Four cores, YA01, YA02, YA03 and YAS03, were drilled in Lake Yanhaizi (Fig. 1b) during
PALAEO 3031 7-3-03
C.-T.A. Chen et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 193 (2003) 181^200 182
Fig. 1. (a) Map showing the locations of Lake Yanhaizi and sites of other archives referred to in this paper. (b) Coring sites (YA01, Ya02 and Ya03) in Lake Yanhaizi.
PALAEO 3031 7-3-03
two consecutive joint drilling programs. The ¢rst was executed by the Institute of Geology, Swiss Federal Institute of Technology in Zu«rich, Swit-zerland, and the Institute of Geological Research and Chemical Mines, Ministry of Chemical Indus-tries, China, in 1992. The second one was per-formed in 1997 with all the cores archived in the Institute of Marine Geology and Chemistry, Na-tional Sun Yat-sen University (IMGC-NSYSU), Taiwan. The ¢rst expedition has been reported by Bernasconi et al. (1997) and covers Lake He-tongchahannor south of Lake Yanhaizi. For now, most of the analyses have been performed on Core YA01, and these have been supplemented by data from Cores YA02 and YAS03 as well as the lithological variations of Core YA03 and nearby desert sand. Subsampling of the 16.22-m-long Core YA01 was done in the laboratory at 10-cm intervals. The 1-cm-thick samples were then oven-dried at 60‡C and utilized in all further studies.
A LECO CHN-932 elemental analyzer was em-ployed to determine carbon content at 950‡C. After the samples were repeatedly rinsed with 1 N HCl to remove inorganic carbon, total organ-ic carbon (TOC) was determined by di¡erent mixes of NIST (SRM-2704, carbon : 3.348%), LECO-EDTA (carbon : 41.1%, nitrogen : 9.59%) and feldspar powder for calibration (Lou and Chen, 1997). Precision was Q 5% for TOC.
Concentrations of 27 elements (O, Na, Mg, Al, Si, P, S, Cl, K, Ca, Ti, V, Cr, Mn, Fe, Co, Ni, Cu, Zn, Ga, Br, Rb, Sr, Y, Zr, Ba and Pb) were mea-sured with the Rigaku RIX-2000 XRF (Chen et al., 2001). Five grams of ground powder of each sample was palletized under 20 ton cm312 pres-sure for 20 min to form a cake with cellulose as backing. Seven standard samples, namely BCSS-1, MESS-1, PACS-1, MAG-1, NIES-2, SRM-2704 and GBW-07314, were used for calibration. Rep-licate analyses of the same samples gave a preci-sion within 5^10%.
Low-frequency (0.47 kHz) and high-frequency (4.7 kHz) magnetic susceptibility analyses were performed using a Bartington MS-2B sensor which was attached to a MS-2 susceptibility me-ter. Each sample was measured three times. Min-erals were determined with a SIMENS D5000
X-ray powder di¡ractometer at the Institute of Material Science and Engineering, NSYSU. Con-ditions were set at 40 kV, 20 mA, and a CuKK target was chosen with step increases of 0.02‡. Grain size analyses were carried out with a Coulter LS-100 laser particle size analyzer (IMGC-NSYSU) following the procedures de-scribed by Janitzky (1987): samples gently hand-ground in a quartz pestle and mortar were sieved to remove grains larger than 1000 Wm, and then treated with HCl and H2O2 to remove carbonate and organic matter. In the ¢nal stage, sodium hexametaphosphate was added to prevent clay minerals from £occulating, and the samples re-ceived a short period of ultrasonic agitation.
Samples of mostly humin and some organic matter were dated using the AMS at the Rafter Radiocarbon Laboratory, Institute of Geological and Nuclear Sciences, New Zealand, and at the Leibniz-Labor for Radiometric Dating and Isotope Research, Christian-Albrechts-University Kiel, Germany, as well as with the conventional liquid scintillation counting method at the Labo-ratory of Carbon Dating, Department of Geo-sciences, National Taiwan University, Taiwan. One concentrated pollen sample was analyzed in the Leibniz-Labor in order to correct for the res-ervoir e¡ect.
Samples of Core YAS03 were measured for 210Pb with a TENNELEC LB 5100-2800-II count-ing system in the Radio Isotope Laboratory (IMGC-NSYSU) following the procedure by
Hung and Chung (1998). Inorganic stable oxygen and carbon isotope analyses were performed at the Institute of Earth Sciences, Academia Sinica, Taiwan.
4. Results
Core YA01 shows that the upper 3.04 m of the Quaternary sediment is comprised predominantly of mirabilite crystals (Na2SO4W10H2O) containing varying amounts of black quartz sand and silt (Fig. 2a,b). Between 3.04 and 8.29 m, mainly black sand and silty sand are found with the highest abundance between 5.50 and 7.30 m. These become grayish^green between 8.29 and
PALAEO 3031 7-3-03
C.-T.A. Chen et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 193 (2003) 181^200 184
11.70 m, where silt and sand are the major com-ponents ; however, between 9.70 and 12.30 m sand is more abundant. The color becomes black again at 12.70 m, and silt and clay are the major com-ponents between 12.30 and 14.20 m. Greyish^ green silt and sand are found below 14.20 m, but at 16.22 m sandstone is found. Fossil discrim-ination on Core YA01 has not yet been per-formed, but fragments of gastropod of unknown
species are found at depths 12.75, 13.45, 14.74, 17.74, 19.38 and 20.74 m on Core YA02.
14C and calibrated calendar ages using the CALIB 4.3 program (Stuiver et al., 1998) are shown in Table 1; all ages without the extra no-tation of the calendar year (cal. yr BP or cal. kyr BP) quoted in this paper are in 14C years (yr BP or kyr BP). The chronostratigraphy of the core is plotted in Fig. 2a. Results from the nearby Core
Fig. 2. (a) Radiocarbon chronology of Core YA01. Solid squares (F) stand for14C dates of YA01, while solid circles (b) denote 14C dates inferred from Core YA02. Di¡erent segments of the line represent various age models. (b,c) Variations in lithology
with depth for Cores YA01 and YA02, respectively.
PALAEO 3031 7-3-03
YA02 (located more central to the lake ; Fig. 1b) at a similar depth and strata (Fig. 2c;Table 2) are also added with no obvious evidence of systematic di¡erences. The deposition rates of the core are more or less constant except for the section be-tween 12.4 and 14.4 m, where the rate is much slower.
In saline lake environments, the reservoir e¡ect is frequently the most important artifact prohibit-ing accurate carbon datprohibit-ing. Several methods, both indirect and direct, were evaluated to constrain the reservoir e¡ect in Lake Yanhaizi. Extrapolat-ing the 14C ages to the coretop yields an age of 879 yr BP, which is the ¢rst indirect indicator of a possible reservoir e¡ect. Further, taking the sed-imentation rate, 0.112 cm/yr, calculated from ex-cess210Pb (Table 3) and assuming that the age of
the coretop is zero and the sedimentation rate is uniform, an age of 195 yr BP is determined for the sample at 0.27 m depth. The 14C age, how-ever, is 1110 Q 180 yr BP (Table 1), making the age di¡erence of 915 yr the second approximation of the reservoir e¡ect.
Next, direct 14C dating of the lake water yields 887 Q 121 yr BP (Table 2) which agrees with the above two values. Finally, because sporopollen is not in£uenced by the reservoir e¡ect (Regnell, 1992 ; Zhou et al., 1997), the age of direct dating of the sporopollen concentrated in the sediments provides the fourth approximation. The age of the sporopollen at 12.75 m is 8415 Q 50 yr BP (Table 1). On the other hand, the age of the bulk sample at depth 12.75 m interpolated from the two adja-cent dated samples, depths 12.7 and 13.0 m (of
Table 1
Radiocarbon dates for the YA01 sequence
Depth Material N13C 14C age Calibrated age Lab. ref.a
(cm) (x) (14C yr BP) (cal. yr BP) 27 Huminb 324.8 1110 Q 180 1048 NTU2326 390 TOC 326.4 4213 Q 70 4827 NZA7298 749 Humin 325.3 6200 Q 340 7157 NTU2228 900 Humin 324.0 7080 Q 200 7933 NTU2337 950 TOC 325.7 7398 Q 70 8181 NZA7432 1270 Humin 324.3 9020 Q 190 10208 NTU2208 1275 Pollen 325.4 8415 Q 50 9470 KIA7929 1300 Humin 324.4 10026 Q 44 11549 KIA6431 1339 Humin 327.8 12263 Q 75 14265 NZA8334 1449 Humin 330.3 14459 Q 87 17318 NZA8335
a Lab. reference: NTU, Laboratory of Carbon Dating, Department of Geosciences, National Taiwan University, Taiwan;
NZA, Rafter Radiocarbon Laboratory, Institute of Geological and Nuclear Sciences, New Zealand ; KIA, Leibniz-Labor for Ra-diometric Dating and Isotope Research, Christian-Albrechts-University Kiel, Germany.
b Sample where saline groundwater £ows.
Table 2
Radiocarbon dates for the YA02 sequence and saline water of the lake
Depth Material N13C 14C age Calibrated age Lab. ref.a
(cm) (x) (14C yr BP) (cal. yr BP)
0 Saline water 887 Q 121 788 ASA
236 TOC 1973 Q 264 1923 ASA 405 TOC 3658 Q 123 3978 ASA 1044.5 TOC 7859 Q 207 8627 ASA 1105 TOC 8189 Q 200 9232 ASA 1608 TOC 14571 Q 344 17447 ASA 2030 Humin 318.7 16450 Q 460 19609 NTU2357
a Lab. reference: ASA, Institute of Archaeology, Chinese Academy of Social Sciences, China; NTU, Laboratory of Carbon
Dating, Department of Geosciences, National Taiwan University, Taiwan.
PALAEO 3031 7-3-03
C.-T.A. Chen et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 193 (2003) 181^200 186
age 9020 Q 190 and 10 026 Q 44 yr BP, respec-tively), is 9188 yr BP, which is 773 yr older than the age of the sporopollen. Accordingly, 879 yr was taken as the reservoir e¡ect in this study. This is in reasonable agreement with the hard water e¡ect estimated at about 1000^2000 yr for lakes in Inner Mongolia (Ren, 1998).
After subtracting 879 yr, ages were then linearly interpolated for each sample by assuming a uni-form sedimentation rate between each major change in the sedimentation rates accompanied by major lithologic changes. The linear sedimen-tation rates thus obtained vary signi¢cantly within the core with high values of 0.157 cm yr31 be-tween 14.2 and 16.22 m and relatively low values of 0.033 cm yr31 between 12.5 and 14.2 m. The variations in the relative contents of sand, silt and clay with age are shown inFig. 3a.
The TOC values are generally low, between 0.09% and 1.08%. There are three sections with relatively high TOC, between 3100 and 4200, 5800 and 6400, and 8000 and 13 400 yr BP (Fig. 3b). The higher values are considered representative of higher biomass productivity in the drainage (Pedersen and Calvert, 1990; Lou and Chen, 1997 ; Lou et al., 1997). The productivity in an arid area is, in turn, controlled by the e¡ective precipitation (An et al., 1993). Thus, these three segments represent a more humid environment. Desert sand samples from the nearby Mu Us Desert have a TOC content of 0.17%, similar to the values found in the dry segments of the core. Higher TOC/TN ratios have also been taken to represent higher productivity on land (Lou and Chen, 1997 ; Meyers, 1997), but the signals here are mixed (Fig. 3c). The absence of data for the last 2.8 kyr BP has resulted from the nitrogen
analysis which was unreliable due to the pertur-bation caused by the presence of saline minerals. Although the three humid segments in fact do have slightly higher TOC/TN values than the dry ones, i.e. between 4300 and 5900, and 6400 and 8000 yr BP, the oldest dry segment before 13 400 yr BP has the highest TOC/TN. This may re£ect a preferential decomposition of organic ni-trogen. The nearby desert sand samples have a TOC/TN ratio of about 9, similar to that found in the dry segments in the mid-Holocene.
According to Ding et al. (1999), the sand ( s 63Wm) content, an indicator of sand desert has never been absent since S1-1 (marine isotope stage 5a) in the Yulin section situated in the tran-sitional zone between the Mu Us Desert and the Loess Plateau downwind of the winter monsoon from our study area (Fig. 1a). As a result, Lake Yanhaizi must have also been receiving desert sand since S1-1. Desert sand samples near Lake Yanhaizi are characterized by a rounded, well-sorted texture and have a modal grain size of about 300 Wm (Fig. 4a). The sediment samples in the dry segments have a similar texture to that of the dune sand (Fig. 4b), implying that the samples were deposited during sand dune or sand sheet progradation periods ; that is, during a shrinkage phase of the lake. In contrast, sediment samples in the wet segments have a distinct size mode of about 10^30 Wm although some sand is still present (Fig. 4c). We assume that during pe-riods of high lake stand, the periphery of the lake was pushed outward. It is understood that a larg-er watlarg-er body would have trapped more aeolian dust (Fryberger et al., 1983) and provided better preservation of these ¢ne wind-carried dusts in the water. At the same time, a higher groundwater table would have sustained the vegetation, which, in turn, would have prevented the sand dune and sand sheet from entering the lake (Nemoto et al., 1997). High precipitation rates also increase the scavenging of aeolian material by rain (Windom, 1975). All of these mechanisms should have left the lake with ¢ner sediments in the humid seg-ments. This is indeed what has been observed in the case of Lake Yanhaizi.
The maturity index is de¢ned as the ratio of feldspars to the sum of feldspars and quartz
Table 3
Activities of excess210Pb for Core YAS03
Depth 210Pb ex. (cm) (dpm g31) 1 1.60 3 0.90 5 0.43 7 0.25 9 0.18 12 0.00 PALAEO 3031 7-3-03
Fig. 3. Secular variations in proxies in Lake Yanhaizi. (a) Sediment composition, (b) TOC, (c) TOC/TN, (d) maturity index, (e) low-frequency magnetic susceptibil-ity, (f) high-frequency magnetic susceptibilsusceptibil-ity, and (g) frequency-dependent susceptibility.
PALAEO 3031 7-3-03 C.-T.A. Chen et al. /Palaeogeog raphy, Palaeoclima tology, Palaeoecolo gy 193 (2003) 181^200 188
Fig. 4. Volumetric particle size distributions for (a) desert sand sampled nearby, and samples at depth (b) 1100, (c) 1380, and (d) 460 cm in Core YA01. PALAEO 3031 7-3-03 C.-T.A. Chen et al. /Palaeogeog raphy, Palaeoclima tology, Palaeoecolo gy 193 (2003) 181^200 189
(Hsu, 1989). While we did not calibrate the inte-grated area of the 2-theta to intensity plot of feld-spars and quartz to their actual concentrations here, we directly used the integrated area to cal-culate the maturity indices. Only 23 samples were analyzed, but the higher values generally occur in the dry segments (Fig. 3e). As a rule, a higher maturity index indicates that the deposited sedi-ments had undergone low-extent chemical weath-ering in drier environments; on the other hand, a lower index indicates that the deposited sediments had undergone higher-extent chemical weathering in more humid environments (Hsu, 1989). There-fore, the higher maturity indices in the mid-Holo-cene supports the notion of a drier climate recon-structed from other proxies. The interpretation of the maturity index to the humidity of the environ-ment is, of course, true only when there is no variation in the mineral composition of the prov-enance of dust and dune sand. Indeed, Wen (1990) has shown that both the mineralogy and chemical composition of regolith materials con-tributed to the Loess Plateau having very little variation across the whole of northern China.
Fig. 3e,f reveal the coherent variations in low-frequency (Lf) and high-low-frequency (Hf) magnetic susceptibilities. The susceptibilities within the core £uctuate between 4.2U and 33U1038 m3 kg31, with the highest values during the last 500 yr. Qualitatively, these variations seem to correlate with the abundance of sand content (Fig. 3a). Frequency-dependent susceptibility (F.D.S.), de-¢ned as {[(Lf3Hf)/Lf]U100%}, falls below 6% and shows only minor variability over the past 14 kyr BP (Fig. 3g). According toEvans and Ro-kosh (2000), the F.D.S. typically falls within the range of 0^4% for completely unweathered, or only slightly weathered, parental loess, whereas well-developed soils yield values between 8% and 12%. Thus, the low F.D.S. indicates that the sedi-ments of our study area have undergone only in-signi¢cant soil development, which is in accor-dance with the low TOC values throughout the core. Magnetic susceptibility in our study area must have been controlled by di¡erent mecha-nisms than those related to loess and paleosol. However, without further analysis of mineralogy and other magnetic parameters, it can only be
concluded that the samples are only slightly weathered.
5. Discussion
5.1. Dry Holocene Megathermal in Inner Mongolia
Sections where coarse sediment dominates (Fig. 3a) show similar sedimentary textures to those of the nearby dune sands and have high maturity indices (Fig. 3d) but low organic carbon contents (Fig. 3c) ; thus, they denote arid environments. On the other hand, ¢ner sediments with high organic contents and low maturity indices imply that these sediments were actually deposited in a humid en-vironment. Three humid phases, 13.4^8.0, 6.4^5.8 and 4.3^3.2 kyr BP, were identi¢ed, with the ¢rst phase the wettest and the third the second wettest. The Holocene Megathermal appears to be arid. The ¢rst wet phase, which started at 13.4 kyr BP, denotes the onset of the last deglaciation (The European BYlling^AllerYd interstadial), which is characterized by warmer, wetter conditions result-ing in an increase in vegetation coverresult-ing in Inner Mongolia (Zhou et al., 1996). This 13.4^8.0 kyr BP humid episode roughly corresponds to the monsoon maximum which occurred in countries that encompass the Arabian Sea between about 7850^8850 and about 11 300 yr BP (Sirocko et al., 1993). It may also be supported by the postu-lated low-latitude vegetation development as based on the observation of a period of promi-nent high atmospheric methane concentration in the early Holocene (Blunier et al., 1995).
Hafsten (1970) proposed the term ‘Megather-mal’ to represent the warmest period from 8.2 to 3 cal. kyr BP since the Last Glacial Maximum. The GRIP ice core isotope records also re£ect higher optimum temperatures for the Holocene between ca. 8.6 and 4.3 cal. kyr BP (Johnsen et al., 2001). In eastern Asia, paleotemperatures de-rived from sporopollen assemblages in the eastern part of Hebei Province (Shi et al., 1994) and on the Loess Plateau (Sun and Zhao, 1991) also show that the Holocene Megathermal was the warmest in the Holocene (Fig. 5a,b).
PALAEO 3031 7-3-03
C.-T.A. Chen et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 193 (2003) 181^200 190
In order to clarify whether the paleoenviron-ment record obtained at Lake Yanhaizi is a local signal or a regional record, here we compare our results with other proxy records at the periphery of the summer monsoon-in£uenced areas. From variations in pollen concentration and organic carbon content, Zhou et al. (1996) found that there was a dry phase from 7.5 to 3.5 kyr BP in the Midiwan pro¢le, 250 km south of Lake Yan-haizi (Fig. 1a). A relatively arid climate was also reported from 7 to 5.6 kyr BP in the Tengger Desert (Guo et al., 2000), 300^600 km west of Lake Yanhaizi. Further, after the increase of aeo-lian material at ca. 6.9 kyr BP, arid and deterio-rating conditions were reconstructed and were de-termined to range between 5.6 and 4.5 kyr BP in the Dali Nor area (Haoluku, 116‡45PE, 42‡57PN, 1295 m a.s.l.), and to have been followed by a minor amelioration at 4.5^3.0 kyr BP (Wang et al., 2001). In addition, the water level of Dali Nor
Lake (116‡30PE, 43‡20PN) began to drop as early as ca. 7 kyr BP (Geng and Zhang, 1988). Persis-tent arid climate from 8 to 3 kyr BP was also inferred from Core QH85-14C, drilled in Qinghai Lake (An et al., 2000). A striking abundance of deciduous broad-leaf pollen occurred from 11 to 8 kyr BP, implying late Pleistocene and early Holocene maximum monsoon precipitation. A second humid epoch occurred between 1.6 and 2.5 kyr BP. After 1.5 kyr BP, the percentage of broad-leaf pollen rose to 30%, but this may have been due to human perturbations. Paleoclimate records form Badain Jaran and Tengger deserts further west indicated that the ¢rst humid impulse after a long-term aridity occurred between 12 and 13 kyr BP but it was again generally dry between 32 and 10 kyr BP (Pachur et al., 1995 ; Rhodes et al., 1996 ; Wu«nnemann and Pachur, 1998).
For the Younger Dryas period, our records show that it was wet in Lake Yanhaizi (Fig. 3). -8 -4 0 4 Loess Plateau ∆ T ( oC) 0 20 40 60 TOC (%) 0 2,000 4,000 6,000 8,000 10,000 12,000 14,000 Age (14C yr BP) Retreat Lake -6 -3 0 3
Eastern Part of Hebei ∆
T (
oC)
c b a
Fig. 5. Sporopollen assemblages derived temperature di¡erences vT (T3Taverage) in (a) the eastern part of Hebei Province (Shi et al., 1994) and (b) the Loess Plateau (Sun and Zhao, 1991). (c) Variations in the TOC (%) in Retreat Lake.
PALAEO 3031 7-3-03
This is contrary to the notion that a cold climate has always been accompanied by a dry climate in east China (An, 2000). This may be due to the insensitivity of our study area. As a result, our record does not show the rapid transitions from dry to humid and back to dry climatic conditions recorded in the Midiwan pro¢le (Zhou et al., 1996 ; Zhou et al., 1999) during the Younger Dryas. Alternatively, this may be a result of dif-ferent controlling climate systems at these two places. For example, it was reported to be humid in western Tibet and at the northern and north-eastern margins of the Tibetan Plateau during the Younger Dryas (Lehmkuhl, 1997 ; Lehmkuhl and Haselein, 2000). There is also still no evidence of a glacier advance in High Asia during the Younger Dryas because of an intensi¢ed summer monsoon (Lehmkuhl, 1997).
Synchronous changes in several past records reconstructed for the northern limit of the summer monsoon may re£ect regional changes in e¡ective humidity (precipitation minus evapo-ration). The data here clearly indicate that the Holocene Megathermal was an arid period in the northern limit of the summer monsoon. This is in direct contrast to the humid phenomenon reported elsewhere in China and to the previously presumed notion that the northern limit of the summer monsoon moved landward during this period (Shi et al., 1994).
To illustrate this point, the high TOC content at Retreat Lake in Taiwan, taken as a wet signal (Chen et al., 2003), is almost the mirror image of the TOC signal in Lake Yanhaizi where low TOC is taken as a re£ection of low humidity (Fig. 5c). This is evidence that the Holocene Megathermal started at 8 kyr BP but began to cool at 4 kyr BP.
Sun et al. (1998a) reported that sandy loam soils were widely developed in the eastern part of northern China as a result of increased monsoon-al precipitation between 9.0 and 3.0 kyr BP, broadly coincident with the Holocene Optimum. Similarly, magnetic susceptibility of the Baxie loess pro¢le delineated a humid episode from 9.7 to 5.3 kyr BP (An et al., 1993). A compilation of pollen-based biome reconstruction shows a northwest shift of forest zones (increased annual moisture availability) in China at 6 kyr BP and
this is possibly on account of direct radiative e¡ects which produced a stronger-than-present summer monsoon (Yu et al., 1998, 2000). It is evident that the Holocene Megathermal was warm and generally wet in eastern China (Shi et al., 1993) and Taiwan. However, it was arid in Inner Mongolia.
It has been reported that the fossil extents of the sandy deserts in China during the Holocene Optimum contracted and retreated to the west of Helan Mountain (Sun et al., 1998a); however, based on our reconstruction, the sandy desert in Inner Mongolia actually expanded during the mid-Holocene. Nevertheless, previous studies have made few speci¢c comments on the paradox that the warmest Holocene Megathermal did not result in wet conditions in all of these areas. The nature and timing of the Holocene climate changes, therefore, have long bewildered scientists studying eastern Asia.
5.2. Possible mechanism
In order to interpret an asynchronous Holocene Optimum phenomenon in East Asia, An et al. (2000) proposed a progressive weakening of the summer monsoon as the summer solar radiation anomaly decreased progressively and the East Asian monsoon index declined through the Holo-cene. An et al. (2000) de¢ned the Holocene Opti-mum as an e¡ective moisture maxiOpti-mum, without making reference to temperature. However, some of An et al.’s records do not show an earlier max-imum humidity during 10^8 kyr BP, but instead, their Baxie and Weinan loess pro¢les indicate that the optimum occurred during the mid-Holocene rather than the early Holocene (An et al., 2000).
Wang et al. (1999) reported a Termination I and early Holocene humidity maximum event in the South China Sea based on a paleosalinity recon-struction and the concurrent presence of a higher £uvial input of mud which may point to a humid condition in south China. This also contradicts the notion of an asynchronous Holocene Opti-mum whereby the maxiOpti-mum humidity occurred in 3.0 kyr BP in south China (An et al., 2000).
Guo et al. (2000)found a prolonged, somewhat drier interval during the mid-Holocene in
north-PALAEO 3031 7-3-03
C.-T.A. Chen et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 193 (2003) 181^200 192
ern Africa and China. Further work points to a drier 6 kyr BP in south China (Zhengtang Guo, private communication, 2000). What these au-thors adopted to justify the mid-Holocene dryness is the notion of a mechanism which involved a general weakening of the monsoons, with the dri-er climate in the monsoonal area extending in a broad geographic realm from northern Africa to eastern China (Guo et al., 2000). But, there are indeed records showing a humid Megathermal, such as in the loess pro¢les denoted by higher magnetic susceptibility (An et al., 1993, 2000) and in the TOC record of Retreat Lake (Fig. 5c). Also, other than the oceanic patterns that
Guo et al. (2000) cited, there are warmer sea-sur-face temperatures reconstructed from foram abundances of high-resolution cores retrieved in the Okinawa Trough and South China Sea (Jian et al., 1996).
The dry Holocene Megathermal in Inner Mon-golia may also be elucidated by monsoon dynam-ics. These require that increased atmospheric con-vergence and rising motion (and hence increased monsoon precipitation) are balanced by increased upper-level divergence and subsidence (and hence decreased precipitation). In relatively simple mon-soon systems, such as the American monmon-soon, the geographic relationship between the zones of con-vergence and of subsidence are relatively straight-forward and give rise to observed bi-polar (wetter/ drier) patterns in precipitation anomalies. The sit-uation in Asia is more complex, however, because of the interplay between the Paci¢c, India and the East African monsoon systems and because of the strong in£uence of the so-called winter monsoon. Nevertheless, it is clear that increases in the extent of the monsoon during the Holocene were accom-panied by decreased precipitation in zones around the core region of the expanded monsoon (Sandy P. Harrison, private communication, 2001).
It is argued here, however, that the discrepancy was more likely to have been due to the greatly enhanced rate of evaporation than to the higher monsoon precipitation in Inner Mongolia. This would have reduced the e¡ective humidity in the warm, arid climate near the northern boundary of the summer monsoon. This explanation would account for the fact that the highest temperatures
in the Holocene Megathermal, as revealed in the eastern part of Hebei Province (Shi et al., 1994;
Fig. 5a), the Loess Plateau (Sun and Zhao, 1991;
Fig. 5b), the Okinawa Trough and the northern South China Sea (Fig. 6a^c), correlate well with the dry phases in Inner Mongolia. Conversely, the 4.3^3.2-kyr BP coldest period in the Holocene corresponds with a wet phase in Inner Mongo-lia when evaporation decreased as temperature dropped. This also contradicts previous sugges-tions that increased evaporation due to the tem-perature maximum at 6000 yr BP was unable to overcompensate for the intensi¢ed precipitation, hence making it become wetter (Sarnthein, 1978).
Fig. 7 shows the monthly precipitation (P) and evaporation (E) at weather stations in Hunjinchi and Daihai. Data for 1971^1980 for Hunjinchi, located near Lake Yanhaizi, are plotted (data were provided by the station). Indeed, the higher the temperature, the higher the precipitation. Al-though this was accompanied by a greater in-crease in evaporation, the rate of inin-crease in evap-oration was much higher, resulting in a higher E3P at higher temperatures. The resulting higher E3P at higher temperatures may account for the fact that the e¡ective humidity was lower during the Holocene Megathermal in Inner Mongolia.
Faced with the increasing trend in temperature due to global warming (Mann et al., 1998 ; Shi et al., 1999), there is another way, based on lake-level £uctuations, to assess the hypothesis that in the future the northern border of monsoon-in-£uenced areas will become drier as it gets warmer. Although no long-term lake-level observations are available for Lake Yanhaizi, the observed lake-level £uctuations in Qinghai Lake, 800 km to the southwest (Fig. 1a), have shown a decreasing trend since the early 1900’s (Qin and Huang, 1998). Lake water balance studies for the 1958^ 1990 period demonstrate that evaporation ex-ceeded water input (precipitation plus runo¡), resulting in lower lake levels (Qin and Huang, 1998). A similar trend with respect to lower lake levels since 1960 has been monitored in Lake Dai-hai (Wang and Feng, 1992), and the data from a nearby weather station, Liancheng, has the same P, E as well as E3P vs. temperature relationships (Fig. 7c,d) as that of the weather station near
PALAEO 3031 7-3-03
Lake Yanhaizi. With global warming, the climate may indeed bring about greater evaporation and an overall drier environment in Inner Mon-golia.
5.3. Correlation with marine records
Pulleniatina obliquiloculata is a tropical plank-tonic foraminifer living primarily in a narrow belt between about 10‡N and 10‡S, and is very sen-sitive to winter temperature (Li et al., 1997).
Variations in P. obliquiloculata in the Okinawa Trough (Core 255, 123‡07PE, 25‡12PN, w.d. 157 5 m ; Core 170, 125‡48PE, 26‡38PN, w.d. 1470 m;
Jian et al., 1996; Li et al., 1997 ; Ujiie¤ and Ujiie¤, 1999) and the South China Sea (Core 17940, 117‡25PE, 20‡07PN, w.d. 1727 m; Jian et al., 1996) indicate that this warm-water species was low in abundance around 10 kyr BP but that it abruptly increased in abundance about 7 kyr BP just prior to showing a drastic decrease from 4360 to 3260 yr BP (Fig. 6a^c). This 4.3^3.2 kyr BP decline, known as the Pulleniatina Minimum Event (Jian et al., 1996 ; Li et al., 1997 ; Ujiie¤ and Ujiie¤, 1999 ; Wang et al., 1999), may be in-dicative of cooling winter sea-surface tempera-tures (Jian et al., 1996 ; Li et al., 1997). Analysis of coccolith (Florisphaera profunda) records have con¢rmed this (Cheng and Wang, 1998). Wei et al. (1997)also found that nannofossil preservation improved during these periods, probably re£ect-ing a local coolre£ect-ing event. They further reported that the relative content of F. profunda had the same distribution as P. obliquiloculata in the South China Sea.
The Pulleniatina Minimum Event around 4 kyr BP along with the paucity of the species around 10 kyr BP corresponded to the two wet phases around 4 and 10 kyr BP found at Lake Yanhaizi (Fig. 6). It was also found in Retreat Lake that there was a hiatus possibly due to the total drying out of the lake between 3.9 and 2.1 cal. kyr BP (Chen et al., 2003; Fig. 5c). The cooling event has also been deduced from coral Sr/Ca ratios and lake sediments at other localities in Taiwan (Liew and Hsieh, 2000), and possibly correlate to the late Holocene neoglaciation registered as a glacier advance in the Tibetan Plateau ( Lehm-kuhl, 1997). On the other hand, the high abun-dance ‘P. obliquiloculata maximum’ corresponded to the dry phase between 4.2 and 8 kyr BP at Lake Yanhaizi (Fig. 6) and the humid phase be-tween 8.0 and 4.2 kyr BP in Retreat Lake (Fig. 5c), signaling two contrasting scenarios in Inner Mongolia and eastern China through the Holo-cene climate variation. The mid-HoloHolo-cene Mega-thermal also correlates with the disappearance of Globorotalia truncatulinoides from the cores of the South China Sea and Okinawa Trough between 8
Pulleniatina obliquiloculata
Fig. 6. Variations in Pulleniatina obliquiloculata (%) in cores taken from (a) the South China Sea (Core 17940-2; Jian et al., 1996) and (b,c) the Okinawa Trough (Cores 255 and 170; Li et al., 1997) and (d) their correlation with TOC (%) records in Lake Yanhaizi. The vertical lines connecting the panels indicate places where the 14C age is 10 kyr BP, and
the numbers at the top of the ¢rst panel denote the marine isotope stages. The hatched areas denote the Pulleniatina Minimum Event.
PALAEO 3031 7-3-03
C.-T.A. Chen et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 193 (2003) 181^200 194
and 4 cal. kyr BP, which indicates a mid-Holo-cene reduction in the thickness of the North Pa-ci¢c subtropical mode water thermostad in the western North Paci¢c (Jian et al., 2000).
Kim and Kennett (1998)have shown the Holo-cene marine transgression in the Yellow Sea oc-curred between 11.3 and 7 kyr BP based on
benthic foraminiferal and stable isotopic data, while on the west coast of Korea, the marine transgression started at 13 kyr BP (Park et al., 1994). This transgression correlates not only with the ¢rst humid phase in Lake Yanhaizi, but also with the humidity maximum during 15^9 cal. kyr BP in the South China Sea (Core 17940,
Fig. 7. (a) Average monthly precipitation (P) and evaporation (E). (b) E3P vs. temperature between 1971 and 1980 at the Hang-jienqi weather station. (c) P and E. (d) E3P vs. temperature between 1959 and 1988 at the Liancheng weather station located near Lake Daihai (Wang and Feng, 1992).
PALAEO 3031 7-3-03
117‡25PE, 20‡07PN, w.d. 1727 m; Wang et al., 1999). It is suggested that such good correlations in climate changes over such a wide geographical distribution both on land and in the ocean are highly indicative that land and the ocean inter-acted.
5.4. Aeolian dust records
Due to its dryness and proximity to the Gobi Desert ^ thought to be the major source of dust in the Greenland ice cores (Biscaye et al., 1997) ^ Inner Mongolian lakes provide an excellent set-ting with which to examine paleoenvironmental changes, especially those involving the long-range transport of aeolian dusts. Liu (1985) stated that the paucity or absence of silt in the desert sand dunes is not unexpected in the deserts northwest of the Loess Plateau, implying that silts ^ as an abrasion product of sand dune terrain ^ are read-ily lost by de£ation, leaving only sands in the source area. Recent ¢eld observations indicate that the silt-rich alluvial fan systems in the Hexi Corridor region (between the Tengger Desert and Loess Plateau) have supplied a considerable amount of material to the loess columns of that region (Derbyshire et al., 1998). Variations of aeolian silt £ux in downwind sink areas can thus serve as a gauge of source area conditions.
Periods of higher aeolian £ux in the GISP2 ice core (O’Brien et al., 1995; Fig. 8a) are found to correlate with lesser amounts of sediments ¢ner than 50 Wm (Fig. 8b), which are the basic fraction of wind-blown dusts (Liu, 1985) in our study area (clay accounts for a negligible amount as com-pared to silt and sand) and corresponds to dry periods in Lake Yanhaizi (Fig. 8c). This is possi-bly evidence of a teleconnection in climate sys-tems between East Asia and polar, high-latitude areas.
In most studies, an increase in dust accumula-tion has been regarded as an indicator of in-creased aridity in the interior of Asia (Nilson and Lehmkuhl, 2001 ; Rea, 1994; Rea and Leinen, 1988). The atmospheric circulation modes in east-ern Asia are long-distance dust transport via upper-level westerly winds along with strong ver-tical air motions which incorporate dust into the
Westerly Jet over Inner Asia during interglacial times, and low-level Eastern Asian winter mon-soon, which prevailed during Glacial times (Pye and Zhou, 1989). But it has long been problem-atic that maximum dust £ux to the North Paci¢c has been recorded in the mid-Holocene (Rea and Leinen, 1988; Wang et al., 1998), while Holocene rates of loess accumulation on the Loess Plateau have been low (Pye and Zhou, 1989).
Because no evidence of arid mid-Holocene con-ditions has been reported for the early 1990s,Pye and Zhou (1989) had to invoke di¡erent patterns of atmospheric circulation to explain the di¡eren-ces in the timing of the dust £ux maximum be-tween the continent and the North Paci¢c where dust originates mainly from Inner Asia. It is, how-ever, now clear that the arid Holocene Megather-mal in Inner Mongolia may easily account for the greatly increased dust £ux found in the North Paci¢c. Silt content was indeed lower or even ab-sent during this period in Lake Yanhaizi (Fig. 3a). Relatively higher dust deposition rates in the cen-tral Loess Plateau were also found during 10^5 kyr BP in the Holocene epoch (An, 2000). Be-sides, in the wake of a higher £uvial input of mud during Termination I and the early Holo-cene, there was indeed a period characterized by higher contents of silt in the two South China Sea cores, 17940-1/2 and 17939-2, (Wang et al., 1999). To summarize, the Asian winter monsoon trans-ports dusts from their source area in Inner Mon-golia to Lake Yanhaizi and the Loess Plateau and, ¢nally, to the South China Sea and the North Paci¢c. A dry mid-Holocene in Lake Yan-haizi corresponds to a higher dust deposition rate in the downwind regions.
6. Conclusions
Preliminary evidence from a sediment core col-lected from Lake Yanhaizi in Inner Mongolia in-dicates that three humid phases, 13.4^8.0, 6.4^5.8 and 4.3^3.2 kyr BP, have occurred since 14 kyr BP. Notwithstanding this fact, in general, the Ho-locene Megathermal was marked by relative dry-ness. This is in complete agreement with ¢ndings noted in the arid^semiarid zones elsewhere in
In-PALAEO 3031 7-3-03
C.-T.A. Chen et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 193 (2003) 181^200 196
ner Mongolia, but it is contrary to the humid Holocene Megathermal found elsewhere in China and Taiwan. The reason for this may be that much enhanced evaporation over higher monsoon precipitation at Lake Yanhaizi reduced e¡ective humidity. The dry (warm) Holocene Megathermal at Lake Yanhaizi also corresponds to the high temperatures found in the Okinawa Trough and the South China Sea, and provides direct evidence for the missing source of aeolian dust in the North Paci¢c. Such correlations may very well be indicative of the in£uence of the Asian mon-soon which is associated with the global atmo-spheric circulation system.
Acknowledgements
We are indebted to Prof. K.J. Hsu for his en-couragement, and Ms. H.I. Huang and Mr. I. Hsu for their assistance with the ¢eld work. We also wish to thank Prof. Yu-Chia Chung, Dr. Chung-Ho Wang, Dr. Peter B. Yuan, Mrs. H.Y. Li and Mrs. K.H. Chen for their assistance with the analyses. Drs. P. De Deckker, J. Bloemendal, F. Lehmkuhl, J. Dobson, Z.T. Guo and S. Harri-son helped to strengthen an earlier version of the manuscript. This work was ¢nancially supported by grant NSC90-2611-M-110-006 from the Na-tional Science Council of the ROC.
Fig. 8. (a) First and dominant empirical orthogonal function (EOF1), quantifying common behavior among Holocene glacio-chemical species in the GISP2 ice core (O’Brien et al., 1995). Arrows pointing to the major positive EOF1 deviation periods denote a higher in£ux of marine sea salt and terrestrial dusts. (b) Contents of grain size less than 50 Wm. (c) Environmental conditions represented by the TOC of Lake Yanhaizi.
PALAEO 3031 7-3-03
References
An, Z.S., 2000. The history and variability of the East Asian palaeomonsoon climate. Quat. Sci. Rev. 19, 171^187. An, Z.S., Liu, T.S., Lu, Y.C., Porter, S.C., Kukla, G.J., Wu,
X.H., Hua, Y.M., 1990. The long-term paleomonsoon var-iation recorded by the loess-paleosol sequence in central China. Quat. Int. 7/8, 91^95.
An, Z.S., Porter, S.C., Zhou, W.J., Lu, Y.C., Donahue, D.J., Head, M.J., Wu, X.H., Ren, J.Z., Zheng, H.B., 1993. Epi-sode of strengthened summer monsoon climate of Younger Dryas age on the Loess Plateau of central China. Quat. Res. 39, 45^54.
An, Z.S., Porter, S.C., Kutzbach, J.E., Wu, X.H., Wang, S.M., Liu, X.D., Li, X.Q., Zhou, W.J., 2000. Asynchronous Ho-locene optimum of the East Asian monsoon. Quat. Sci. Rev. 19, 743^762.
Bernasconi, S.M., Dobson, J., Mckenzie, J.A., Ariztegui, D., Niessen, F., Hsu, K.J., Chen, Y., Yang, Q., 1997. Prelimi-nary isotopic and paleomagnetic evidence for Younger Dryas and Holocene climate evolution in NE Asia. Terra Nova 9, 246^250.
Biscaye, P.E., Grousset, F.E., Revel, M., Van der Gaast, S., Zielinski, G.A., Vaars, A., Kukla, G., 1997. Asian prove-nance of glacial dust (stage 2) in the GISP2 ice core, Summit, Greenland. J. Geophys. Res. 102 (C12), 26675^ 26781.
Blunier, T., Chappellaz, J., Schwander, J., Stau¡er, B., Ray-naud, D., 1995. Variations in atmospheric methane concen-tration during the Holocene epoch. Nature 374, 46^49. Chen, C.T.A., Wann, J.K., Lou, J.Y., 2001. Aeolian £ux of
metals in Taiwan in the past 2600 years. Chemosphere 43, 287^294.
Chen, C.T.A., Wann, J.K., Lou, J.Y., 2003. High resolution paleoclimatic records in Taiwan, submitted.
Cheng, X.R., Wang, P.X., 1998. Variations in Late Quaternary upper ocean structure of Okinawa Trough ^ a nannofossil approach. Sci. China (Ser. D) 41, 290^296.
COHMAP Members, 1988. Climatic changes of the last 18,000 years: observations and model simulations. Science 241, 1043^1052.
Derbyshire, E., Meng, X.M., Kemp, R.A., 1998. Provenance, transport and characteristics of modern aeolian dust in west-ern Gansu Province, China, and interpretation of the Qua-ternary loess record. J. Arid Environ. 39, 497^516. Ding, Z.L., Sun, J.M., Rutter, N.W., Rokosh, D., Liu, T.S.,
1999. Changes in sand content of loess deposits along a north-south transect of the Chinese Loess Plateau and the implications for desert variations. Quat. Res. 52, 56^62. Enzel, Y., Ely, L.L., Mishra, S., Ramesh, R., Amit, R., Lazar,
B., Rajaguru, S.N., Baker, V.R., Sandler, A., 1999. High-resolution Holocene environmental changes in the Thar Desert, Northwestern India. Science 284, 125^128. Evans, M.E., Rokosh, C.D., 2000. The last interglacial in the
Chinese Loess Plateau: a petromagnetic investigation of samples from a north-south transect. Quat. Int. 68-71, 77^ 82.
Fryberger, S.G., Al-Sari, A.M., Clisham, T.J., 1983. Eolian dune, interdune, sand sheet, and siliciclastic sabkha sedi-ments of an o¡shore prograding sand sea, Dhahran Area, Saudi Arabia. AAPG Bull. 67, 280^312.
Gao, Y.X., 1962. On some problems of Asian monsoon. In: Gao, Y.X. (Eds.), Some Questions About the East Asian Monsoon. Science Press, Beijing, pp. 1^49.
Geng, K., Zhang, Z., 1988. Geomorphologic features and evo-lution of the Holocene lakes in Dali Nor Area, the Inner Mongolia (in Chinese). J. Beijing Normal Univ. (Nat. Sci.) 4, 94^100.
Guo, Z.T., Petit-Maire, N., Kro«pelin, S., 2000. Holocene non-orbital climatic events in present-day arid areas of northern Africa and China. Glob. Planet. Change 26, 97^103. Hafsten, U., 1970. A sub-division of the Late Pleistocene
peri-od on a synchronous basis, intended for global and universal usage. Palaeogeogr. Palaeoclimatol. Palaeoecol. 7, 279^296. Hsu, K.J., 1989. Physical Principles of Sedimentology: A Readable Textbook for Beginners and Experts. Springer, Berlin.
Hung, G.W., Chung, Y.C., 1998. Particulate £uxes,210Pb and 210Po measured from sediment trap samples in a canyon o¡
northeastern Taiwan. Cont. Shelf Res. 18, 1475^1491. Janitzky, P., 1987. Particle size analysis. In: Singer M. and P.
Janitzky (Eds.), Field and Laboratory Procedures Used in Soil Chrono-sequence Study. USGS Bull. 1648, 11^16. Jian, Z.M., Li, B.H., P£aumann, U., Wang, P.X., 1996. Late
Holocene cooling event in the western Paci¢c. Sci. China (Ser. D) 39, 543^550.
Jian, Z.M., Li, B.H., Huang, B.Q., Wang, J.L., 2000. Globo-rotalia truncatulinoides as indicator of upper-ocean thermal structure during the Quaternary : evidence from the South China Sea and Okinawa Trough. Palaeogeogr. Palaeoclima-tol. Palaeoecol. 162, 287^298.
Johnsen, S.J., Dahl-Jensen, D., Gundestrup, N., Ste¡ensen, J.P., Clausen, H.B., Miller, H., Masson-Delmotte, V., Sveinbjo«rnsdottir, A.E., White, J., 2001. Oxygen isotope and paleotemperature records from six Greenland ice-core stations: Camp Century, Dye-3, GRIP, GISP2, Renland and NorthGRIP. J. Quat. Sci. 16, 299^307.
Kim, J.M., Kennett, J.P., 1998. Paleoenvironmental changes associated with the Holocene marine transgression, Yellow Sea (Hwanghae). Mar. Micropaleontol. 34, 71^89. Kukla, G., Heller, F., Liu, X.M., Xu, T.C., Liu, T.S., An, Z.S.,
1988. Pleistocene climates in China dated by magnetic sus-ceptibility. Geology 16, 811^814.
Kutzbach, J., Otto-Bliesner, B.L., 1982. The sensitivity of the African-Asian monsoonal climate to orbital parameter changes for 9, 000 yr BP in a low-resolution general circu-lation model. J. Atmos. Sci. 39, 1177^1188.
Kutzbach, J.E., Street-Perrott, F.A., 1985. Milankovitch forc-ing of £uctuations in the level of tropical lakes from 18 to 0 kyr BP. Nature 317, 130^134.
Kutzbach, J., Bonan, G., Foley, J., Harrison, S.P., 1996. Veg-etation and soil feedbacks on the response of the African monsoon to orbital forcing in the early to middle Holocene. Nature 384, 623^626.
PALAEO 3031 7-3-03
C.-T.A. Chen et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 193 (2003) 181^200 198
Lehmkuhl, F., 1997. Late Pleistocene, Late-Glacial and Holo-cene glacier advances on the Tibetan Plateau. Quat. Int. 38/ 39, 77^83.
Lehmkuhl, F., Haselein, F., 2000. Quaternary paleoenviron-mental change on the Tibetan Plateau and adjacent areas (West China and Mongolia). Quat. Int. 65/66, 121^145. Li, B.H., Jian, Z.M., Wang, P.X., 1997. Pulleniatina
obliqui-loculata as a paleoceanographic indicator in the southern Okinawa Trough during the last 20,000 years. Mar. Micro-paleontol. 32, 59^69.
Li, B.S., Zhang, D.D., Jin, H.L., Wu, Z., Yan, M.C., Sun, W., Zhu, Y.Z., Sun, D.H., 2000. Paleo-monsoon activities of Mu Us Desert, China since 150 ka B.P. ^ a study of the strati-graphic sequences of the Milanggouwan Section, Salawusu River area. Palaeogeogr. Palaeoclimatol. Palaeoecol. 162, 1^ 16.
Liew, P.M., Hsieh, M.L., 2000. Late Holocene (2 ka) sea level, river discharge and climate interrelationship in the Taiwan region. J. Asian Earth Sci. 18, 499^505.
Liu, T.S., 1985. Loess and the Environment. China Ocean Press, Beijing.
Lou, J.Y., Chen, C.T.A., 1997. Paleoclimatological and paleo-environmental records since 4000 a BP. in sediments of al-pine lakes in Taiwan. Sci. China (Ser. D) 40, 424^431. Lou, J.Y., Chen, C.T.A., Wann, J.K., 1997.
Paleoclimatolog-ical records of the Great Ghost Lake in Taiwan. Sci. China (Ser. D) 40, 284^292.
Mann, M.E., Bradley, R.S., Hughes, M.K., 1998. Global-scale temperature patterns and climate forcing over the past six centuries. Nature 392, 779^787.
Meyers, P.A., 1997. Organic geochemical proxies of paleocea-nographic, paleolimnologic, and paleoclimatic processes. Org. Geochem. 27, 213^250.
Nemoto, M., Ohkuro, T., Xu, B., 1997. The role of weed invasion in controlling sand dune reactivation in abandoned ¢elds in semi-arid Inner Mongolia. China Ecol. Res. 12, 325^336.
Nilson, E., Lehmkuhl, F., 2001. Interpreting temporal patterns in the late Quaternary dust £ux from Asia to the North Paci¢c. Quat. Int. 76/77, 67^76.
O’Brien, S.R., Mayewski, P.A., Meeker, L.D., Meese, D.A., Twickler, M.S., Whitlow, S.I., 1995. Complexity of Holo-cene Climate as reconstructed from a Greenland ice core. Science 270, 1962^1964.
Pachur, H.J., Wu«nnemann, B., Zhang, H., 1995. Lake evolu-tion in the Tengger Desert, northwestern China, during the last 40,000 years. Quat. Res. 44, 171^180.
Park, Y.A., Khim, B.K., Zhao, S., 1994. Sea level £uctuation in the Yellow Sea Basin (in Korean with English abstract). J. Korean Soc. Oceanogr. 29, 42^49.
Pedersen, T.F., Calvert, S.E., 1990. Anoxia vs. productivity: what controls the formation of organic-carbon-rich sedi-ments and sedimentary rocks? AAPG Bull. 74, 454^466. Pye, K., Zhou, L.P., 1989. Late Pleistocene and Holocene
Aeolian dust deposition in north China and the northwest Paci¢c Ocean. Palaeogeogr. Palaeoclimatol. Palaeoecol. 73, 11^23.
Qin, B.Q., Huang, Q., 1998. Evaluation of the climatic change impacts on the inland lake ^ a case study of lake Qinghai, China. Clim. Change 39, 695^714.
Rea, D.K., 1994. The paleoclimatic record provided by eolian deposition in the deep sea: the geologic history of wind. Rev. Geophys. 32, 159^195.
Rea, D.K., Leinen, M., 1988. Asian aridity and the zonal west-erlies: Late Pleistocene and Holocene record of eolian depo-sition in the northwest Paci¢c Ocean. Palaeogeogr. Palaeo-climatol. Palaeoecol. 66, 1^8.
Regnell, J., 1992. Preparing pollen concentrations for AMS dating ^ a methodological study from a hard-water lake in southern Sweden. Boreas 21, 373^377.
Ren, G.Y., 1998. A ¢nding of the in£uence of ‘hard water’ on radiocarbon dating for lake sediments in Inner Mongolia, China. J. Lake Sci. 10, 80^82.
Ritchie, J.C., Haynes, C.V., 1987. Holocene vegetation zona-tion in the eastern Sahara. Nature 330, 645^647.
Rhodes, T.E., Gasse, F., Ruifen, L., Fontes, J.Ch., Wei, K., Bertrand, P., Gibert, E., Me¤lie'res, F., Tucholka, P., Wang, Z., Chen, Z., 1996. A Late Pleistocene-Holocene lacustrine record from Lake Manas, Zhunggar (northern Xinjiang, western China). Palaeogeogr. Palaeoclimatol. Palaeoecol. 120, 105^121.
Sarnthein, M., 1978. Sand deserts during glacial maximum and climatic optimum. Nature 272, 43^46.
Shi, Y.F., Kong, Z.Z., Wang, S.M., Tang, L.Y., Wang, F.B., Yao, T.D., Zhao, X.T., Zhang, P.Y., Shi, S.H., 1993. Mid-Holocene climates and environments in China. Glob. Planet. Change 7, 219^233.
Shi, Y.F., Kong, Z.Z., Wang, S.M., Tang, L.Y., Wang, F.B., Yao, T.D., Zhao, X.T., Zhang, P.Y., Shi, S.H., 1994. The climate £uctuation and important events of Holocene Mega-thermal in China. Sci. China (Ser. B) 37, 353^365. Shi, Y.F., Yao, T.D., Yang, B., 1999. Decadal climatic
varia-tions recorded in Guliya ice core and comparison with the historical documentary data from East China during the last 2000 years. Sci. China (Ser. D) 42, 91^100.
Sirocko, F., Sarnthein, M., Erlenkeuser, H., Lange, H., Ar-nold, M., Duplessy, J.C., 1993. Century-scale events in mon-soonal climate over the past 24,000 years. Nature 364, 322^ 324.
Street-Perrott, F.A., Roberts, N., 1983. Fluctuations in closed-basin lakes as an indicator of past atmospheric circulation patterns. In: Street-Perrott, F.A., Beran, M.A., Ratcli¡e, R.A.S. (Eds.), Variations in the Global Water Budget. Reidel, Hingham, MA, pp. 331^345.
Stuiver, M., Reimer, P.J., Bard, E., Beck, J.W., Burr, G.S., Hughen, K.A., Kromer, B., McCormac, G., v.d. Plicht, J., Spurk, M., 1998. INTCAL98 Radiocarbon Age Calibration, 24,000-0 cal BP. Radiocarbon 40, 1041^1083.
Sun, J.M., 2000. Origin of eolian sand mobilizaton during the past 2300 years in the Mu Su Desert. China Quat. Res. 53, 78^88.
Sun, J.M., Ding, Z.L., Liu, T.S., 1998a. Desert distributions during the glacial maximum and climatic optimum: Exam-ple of China. Episodes 21, 28^31.
PALAEO 3031 7-3-03
Sun, J.M., Yin, G.M., Ding, Z.L., Liu, T.S., Chen, J., 1998b. Thermoluminescence chronology of sand pro¢les in the Mu Us Desert, China. Palaeogeogr. Palaeoclimatol. Palaeoecol. 144, 225^233.
Sun, J.M., Ding, Z.L., Liu, T.S., Rokosh, D., Rutter, N., 1999. 580,000-year environmental reconstruction from Aeolian de-posits at the Mu Us Desert margin, China. Quat. Sci. Rev. 18, 1351^1364.
Sun, J.Z., Zhao, J.B., 1991. Quaternary of Loess Plateau in China. Science Press, Beijing.
Ujiie¤, H., Ujiie¤, Y., 1999. Late Quaternary course changes of the Kuroshio Current in the Ryukyu Arc region, northwest-ern Paci¢c Ocean. Mar. Micropaleontol. 37, 23^40. Wang, H., Zhao, Q., Yu, L., Zhang, W., 1998. Late
Quater-nary eolian records in the northwest Paci¢c with Glacial cycles: a comparison. Sci. China (Ser. D) 41, 28^34. Wang, H.Y., Liu, H.Y., Cui, H.T., Abrahamsen, N., 2001.
Terminal Pleistocene/Holocene paleoenvironmental changes revealed by mineral-magnetism measurements of lake sedi-ments for Dali Nor area, southeastern Inner Mongolia Pla-teau, China. Palaeogeogr. Palaeoclimatol. Palaeoecol. 170, 115^132.
Wang, L., Sarnthein, M., Erlenkeuser, H., Grimalt, J., Grootes, P., Heilig, S., Ivanova, E., Kienast, M., Pelejero, C., P£aumann, U., 1999. East Asian monsoon climate dur-ing the Late Pleistocene : high-resolution sediment records from the South China Sea. Mar. Geol. 156, 245^284. Wang, S.M., Feng, M., 1992. The relationship between
envi-ronmental change and strength of southeast monsoon in Daihai Lake, Inner Mongolia. Sci. China (Ser. B) 35, 722^ 734.
Wei, K.Y., Yang, T.N., Huang, C.Y., 1997. Glacial-Holocene calcareous nannofossils and paleoceanography in the north-ern South China Sea. Mar. Micropaleontol. 32, 95^114. Wen, Q.Z., 1990. Geochemistry of Loess in China. Science
Press, Beijing, in Chinese.
Windom, H.L., 1975. Eolian contributions to marine sedi-ments. J. Sediment. Petrol. 45, 520^529.
Wu«nnemann, B., Pachur, H.J., 1998. Climatic and environ-ment changes in the deserts of Inner Mongolia, China, since the Late Pleistocene. In : Alsharan, A.S., Glennie, K.W. and Whittle, G.L. (Eds.), Quaternary Deserts and Climate Changes. Balkema, Rotterdam, pp. 381^394.
Yu, G., Prentice, I.C., Harrison, S.P., Sun, X.J., 1998. Pollen-based biome reconstructions for China at 0 and 6000 years. J. Biogeogr. 25, 1055^1069.
Yu, G., Chen, X., Ni, J., Cheddadi, R., Guiot, J., Han, H., Harrison, S.P., Huang, C., Ke, M., Kong, Z., Li, W., Liew, P., Liu, G., Liu, J., Liu, Q., Liu, K.B., Prentice, I.C., Qui, W., Ren, G., Song, C., Sugita, S., Sun, X., Tang, L., Van Campo, E., Xia, Y., Xu, Q., Yan, S., Yang, X., Zhao, J., Zheng, Z., 2000. Palaeovegetation of China: a pollen data-based synthesis for the mid-Holocene and last glacial max-imum. J. Biogeogr. 27, 635^664.
Zhang, J., Lin, Z., 1992. Climate of China. Wiley, Shanghai. Zheng, X.Y., Zhang, M.G., Han, Z.M., Dong, Z.H., Zhang,
B.Z., Gao, Z.H., Sun, D.P., Xue, C., Wang, K.J., 1992. Salt Lakes in Inner Mongolia. Science Press, Beijing, 195 pp., in Chinese.
Zhou, W.J., Donahue, D.J., Porter, S.C., Jull, T.A., Li, X.Q., Stuiver, M., An, Z.S., Matsumoto, E., Dong, G.R., 1996. Variability of monsoon climate in East Asia at the end of the last Glaciation. Quat. Res. 46, 219^229.
Zhou, W.J., Donahue, D.J., Jull, A.J.T., 1997. Radiocarbon AMS dating of pollen concentrated from eolian sediments: Implications for monsoon climate change since the Late Quaternary. Radiocarbon 39, 19^26.
Zhou, W.J., Head, M.J., Lu, X.F., An, Z.S., Jull, A.J.T., Do-nahue, D., 1999. Teleconnection of climatic events between East Asia and polar, high latitude areas during the last de-glaciation. Palaeogeogr. Palaeoclimatol. Palaeoecol. 152, 163^172.
PALAEO 3031 7-3-03
C.-T.A. Chen et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 193 (2003) 181^200 200
Ct1\D:\NSC\91s\成果報告\全球海洋通量聯合研究 9205.doc 表 Y04
