Luminescence Chronology and Radiocarbon Reservoir Age

Journal of Earth Science, Vol. 29, No. 3, p. 695–706, June 2018 ISSN 1674-487X Printed in China https://doi.org/10.1007/s12583-017-0972-9

Luminescence Chronology and Radiocarbon Reservoir Age Determination of Lacustrine Sediments from the Heihai Lake, NE Qinghai-Tibetan Plateau and Its Paleoclimate Implications

Fuyuan An *1, 2, 4, Zhongping Lai5, Xiangjun Liu1, 2, Yixuan Wang1, 3, Qiufang Chang1, Baoliang Lu1, Xiaoyun Yang5, 6 1. Key Laboratory of Comprehensive and Highly Efficient Utilization of Salt Lake Resources, Qinghai Institute of Salt Lakes, Chinese Academy of Sciences, Xining 810008, China 2. Qinghai Provincial Key Laboratory of Geology and Environment of Salt Lakes, Qinghai Institute of Salt Lakes, Chinese Academy of Sciences, Xining 810008, China 3. University of Chinese Academy of Sciences, Beijing 100049, China 4. State Key Laboratory of Cryosphere Sciences, Cold and Arid Regions Environmental and Engineering Research Institute, Chinese Academy of Sciences, Lanzhou 730000, China 5. School of Earth Sciences, China University of Geosciences, Wuhan 430074, China 6. Department of Geology, Baylor University, Waco 76798, USA Fuyuan An: https://orcid.org/0000-0003-3136-8847

ABSTRACT: The accurately determining the lake 14C reservoir age has a crucial significance for climatic reconstruction. In this study, the optically stimulated luminescence (OSL) dating method is em- ployed to date samples from highstand lacustrine sediments, palaeoshoreline, fluvial terrace, and the alluvial fan of the Heihai Lake catchment. Accelerator mass spectrometry (AMS) 14C dating was also used to date fossil plants from highstand lacustrine sediments. Based on the calculations of linear regression with OSL against radiocarbon ages for same layers of two sections, the quantitative 14C reservoir ages were estimated to lie between 3 353 and 3 464 yr during the 1.8 to 2.4 ka, which showed temporal variation. The sources of old carbon are the dissolution of carbonate bedrocks distributed along the Kunlun Mountain. The OSL ages of the different members of the hydatogen sedimentary system at Heihai Lake catchment indicate that a stronger hydrologic condition occurred from 3.0±0.2 to 1.8±0.2 ka, with a maximum lake level of 9 m higher than present. This humid stage was widely recorded in different sediments on the QTP and Chinese Loess Plateau (CLP), indicating its broad synchronicity across the Asian Summer Monsoon region. The enhanced East Asian Summer Monsoon (EASM) and the Indian Summer Monsoon (ISM) resulted in the increase of moisture availability for the Heihai Lake area during this stage. KEY WORDS: luminescence and radiocarbon dating, 14C reservoir age, Heihai Lake, Qinghai-Tibetan Plateau (QTP), palaeoclimate implications.

0 INTRODUCTION 2016; Zeng et al., 2015; An Z S et al., 2012; Chen et al., 2008). The northern Qinghai-Tibetan Plateau (QTP) is a joint The QTP has changed the atmospheric and hydrological circu- area that was influenced by the East Asian Summer Monsoon lations and reshaped the local and global climate (Yang K et al., (EASM), Indian Summer Monsoon (ISM), and the mid-latitude 2014). Accurate and reliable dating of lacustrine deposits is Westerlies (Fig. 1a), and exerts a significant impact on the re- crucial for understanding these climate change patterns and its gional and global climate through a thermal and mechanical forcing mechanisms. Radiocarbon (14C) dating is the most forcing mechanism (An F Y et al., 2018, 2012; Wang et al., commonly applied method for establishing chronologies of lacustrine sediments. However, its application could be re- *Corresponding author: [email protected] stricted due to the ubiquitous hard water reservoir effect on © China University of Geosciences and Springer-Verlag GmbH young Holocene sediments, especially in the QTP (Mischke et Germany, Part of Springer Nature 2018 al., 2013; Hou et al., 2012a). The reservoir effect on lacustrine sediments in the QTP varies significantly, from 650 yr at Manuscript received April 24, 2017. Ahung Co to 6 670 yr at Bangong Co (Hou et al., 2012b). Thus, Manuscript accepted September 1, 2017. the accurate determination of the reservoir effect for 14C dating

An, F. Y., La, Z. P., Liu, X. J., et al., 2018. Luminescence Chronology and Radiocarbon Reservoir Age Determination of Lacustrine Sediments from the Heihai Lake, NE Qinghai-Tibetan Plateau and Its Paleoclimate Implications. Journal of Earth Science, 29(3): 695–706. https://doi.org/10.1007/s12583-017-0972-9. http://en.earth-science.net 696 Fuyuan An, Zhongping Lai, Xiangjun Liu, Yixuan Wang, Qiufang Chang, Baoliang Lu and Xiaoyun Yang

Figure 1. Map showing the location of the study area. (a) MSML represents Modern Summer Monsoon Limit; (d) the red dots represent the sample sites. HL. Heihai highstand lacustrine sediments; HP. Heihai paleoshoreline deposits; HF. Heihai alluvial fan deposits; HT. Heihai fluvial terrace deposits. is critical for establishing reliable age control. generation (SAR) protocol (Murray and Wintle, 2000) has re- Approaches for estimating the reservoir age involves a sulted in extensive applications for dating lacustrine sediments modern calibration approach, linear extrapolation, geochemical (e.g., Fan et al., 2014; Liu and Lai, 2012; Long et al., 2011; Liu models, stratigraphic alignment, and independent age determi- et al., 2010; Shen et al., 2008), and it has been shown that nations (E et al., 2015; Hou et al., 2012b). Each of these meth- silt-sized lacustrine sediments have been adequately bleached ods has its own disadvantages for applications in lakes on the (Thomas et al., 2003). By comparing the OSL dating of silt QTP. The modern calibration and linear extrapolation ap- quartz and the 14C dating of organic matter for lake samples proaches only provide an estimate of the modern 14C reservoir with the same layers, the 14C reservoir age can be obtained age, and different ages can be obtained for samples from dif- (Long et al., 2011; Shen et al., 2008). In the current study, we ferent sites. Geochemical modeling is a theoretical method for try to evaluate the lake reservoir effect of 14C dating for Heihai determining reservoir age, but it is difficult to estimate the ini- Lake by comparing the radiocarbon and OSL ages for samples tial conditions and parameters of the models in the geological from the same lacustrine layers, and further discuss its paleo- time scale. Independent age determination is the most reliable climatic changes combining other hydatogen sedimentary sys- approach for estimating the past variations in the 14C reservoir tem at Heihai Lake catchment. age, thus this approach is widely used in determining the “old carbon” error in lacustrine sediments (Hou et al., 2012a, b; Liu 1 MATERIALS AND METHODS and Lai, 2012; Long et al., 2011; Shen et al., 2008). 1.1 Study Area Improvements in the optically stimulated luminescence The Heihai Lake (36°N, 93°15′E, 4 445 m a.s.l.) is located (OSL) dating method such as the use of the single aliquot re- in the Kunlun Mountains in the northern region of QTP (Figs. Luminescence Chronology and Radiocarbon Reservoir Age Determination of Lacustrine Sediments 697

1b, 1c, and 1d). This lake catchment is dominated by a large to avoid age underestimation (Lai and Brückner, 2008). The area of Triassic, Paleogene, and Carboniferous clastic rocks. purity of the isolated quartz was verified by infrared light Vast piedmont alluvial/fluvial fan sediments fill the catchment stimulation (830 nm) and no obvious IRSL was observed in all (Zhang et al., 2013). samples. The pure quartz samples were then mounted onto the The annual precipitation is 100–200 mm with the majority center (0.5 cm-diameter) of 0.97 cm-diameter stainless steel of the precipitation falling during the summer season in this discs with silicone oil. OSL measurements were carried out on region. The mean annual temperature is 2.0 oC. The mean an automated Risø TL/OSL-DA-20 reader. Stimulation was by January temperature at the Wudaoliang meteorological station blue LEDs at 130 °C for 40 s, and detection was through 7.5 (4 780 m a.s.l.), which is 50 km to the south, is -8.4 °C and the mm-thick U-340 filters. Preheat was performed using 260 °C mean July temperature is 12.1 °C (Zhang et al., 2013). The lake for 10 s for the regenerative dose, and cut-heat was 220 °C for is an open-basin lake. It is fed by two streams from the glacials 10 s for the test dose. The signals of the first 0.64 s stimulation and discharges into the Kunlun River, which finally flows into were integrated for growth curve construction after back- the Qarhan Salt Lake into the enclosed Qaidam Basin (QB). ground subtraction, which used the last 25 channels in the shine down curve. The concentrations of uranium (U), thorium 1.2 Section and Samples (Th) and potassium (K) were measured by neutron activation The HL1 section (35°57′28.95″N, 93°17′33.23″E, 4 446 m analysis in the Chinese Institute of Atomic Energy in Beijing. a.s.l.) is ~1.0 m higher than HL2 (35°57′42.30″N, For the 38–63 μm grains, the alpha efficiency value was taken 93°17′34.99″E, 4 445 m a.s.l.). The two sections are made up as 0.035±0.003 (Lai et al., 2008). The cosmic ray dose rate of six lacustrine sedimentary layers, which include an abun- was estimated for each sample as a function of depth, altitude, dance of interbedded aquatic plant species residues (Pota- and geomagnetic latitude (Prescott and Hutton, 1994). The mogeton sp.). Six plant residue samples for 14C dating and six water content was calculated based on moisture mass/dry mass lacustrine sediment samples for OSL dating were collected with (Aitken, 1985) and, taking into account the uncertainties dur- stratigraphic sequences to determine the chronology and the ing burial, an error of 5% was adopted for the water content to reservoir effect of 14C dating for the lake. dose rate calculations.

In addition, three OSL dating samples were collected from An equivalent dose (De) was determined using a combina- the paleoshoreline (namely the HP section, 35°59′57.22″N, tion of the single-aliquot regenerative dose (SAR) protocol 93°17′54.29″E, 4 452 m a.s.l.), the fluvial terrace of the lake (Murray and Wintle, 2000) and a standardized growth curve outfall (HT, 35°57′11.70″N, 93°19′53.13″E, 4 446 m a.s.l.), and (SGC) method (Lai and Wintle, 2006), i.e., the SAR-SGC the alluvial fan of the southern outfall (HF, 35°57′00.91″N, method. For each sample, six aliquots were measured by SAR 93°15′50.32″E, 4 468 m a.s.l.) in the catchment (Figs. 1d and 2). to construct a SGC growth curve, of which the test dose- In the HP section, the upper layer is an aeolian deposit with a corrected OSL of a natural aliquot could get an equivalent dose. few gravels, whereas the bottom layer is beach gravel and sand. The materials used for radiocarbon dating were submerged One sample was collected from the bottom layers of this sec- plant residue from the lacustrine sediments sections. These tion. The HT section consists of subangular gravels of fluvial samples were dated by the AMS 14C method in the Radiocar- sediments, of which one sample was collected. The HF section bon Dating Laboratory of Peking University. The 14C ages were is in the surface alluvial fan, which is made up of gravel, sand, converted to calendar years before comparing them with the and silt, and has poor roundness and sorting. One sample was OSL ages. All radiocarbon dates were calibrated to calendar collected from this section. ages and adopted the internationally ratified calibration curve IntCal04, which is based on dendrochronologically dated tree 1.3 Measurement Techniques rings, as well as the OxCal v3.10 program (Reimer et al., 2004) The OSL samples were treated using the chemical (Table 1). Grain-size of samples was measured using Malvern pre-treatment procedures according to Lai (2000). It is impor- Matersizer 2000 laser grain-size analyzer in Qinghai Institute of tant to ensure the efficient removed of feldspar contamination Salt Lakes, Chinese Academy of Sciences.

Table 1 Radiocarbon and OSL dating ages for the same layers of HL1 and HL2 section of Heihai Lake, the radiocarbon reservoir ages of Heihai Lake were calculated by the age differences of the two dating methods. Radiocarbon ages calibrated using calibration curve IntCal04 and the 2σ precision.

Sample Depth Dating material 14C age Cal. age (2σ) OSL age 14C Reservoir age ID (m) (yr BP) (Cal yr BP) (ka) (yr)

HL1-1 0.55 PF 6 215±50 5 175±135 1.8±0.2 3 375±68 HL1-2 0.80 PF 6 680±45 5 595±85 2.2±0.2 3 395±115 HL2-1 0.15 PF 6 700±35 5 610±70 2.3±0.2 3 310±130 HL2-2 0.30 PF 6 915±40 5 805±85 2.3±0.2 3 505±115 HL2-3 0.55 PF - - - - 2.4±0.2 - - HL2-4 0.80 PF 7 050±35 5 940±70 2.1±0.3 3 840±230

PF. Potamogeton fossil; - -. indicated the values were not detected. 698 Fuyuan An, Zhongping Lai, Xiangjun Liu, Yixuan Wang, Qiufang Chang, Baoliang Lu and Xiaoyun Yang

Figure 2. (a) Preheat plateau test of Sample HL2-1 showing the dependence of equivalent dose. Four aliquots were used per preheat temperature, and the error o bars represent 1 standard deviation. The dashed line is drawn at the average De of the aliquots measured with preheat temperature between 220 and 300 C. (b) The dose recovery experiment of Sample HL2-1, given dose is 7.9 Ga. (c) The recycling ratio and recuperations of Sample HL2-1 as a function of preheat temperature for the same aliquots as in (a). A dashed line is drawn at 1.0. (d)–(f). Histograms of De distributions of part samples, n represents the number of aliquots.

2 RESULTS De determination. 2.1 Luminescence Characteristics In addition, the dose recovery test is needed to check for

Investigating the influence of preheating on the charge the suitability of SAR procedures for De determination (Murray transfer from light-insensitive traps to light-sensitive ones is and Wintle, 2003). A dose recovery test was performed on rather important (Aitken, 1998). In order to choose appropriate Samples HL2-1 (Fig. 2b). Six natural aliquots were stimulated preheat conditions that minimize thermal transfer for De deter- twice by blue-light stimulation at 130 °C for 40 s, and the ali- mination using the SAR protocol, a preheat temperature plateau quots were irradiated at a laboratory dose of 7.9 Ga, which is test for Sample HL2-1 was conducted (Fig. 2a). Preheat tem- close to the natural De. The measured De was 7.4±0.3 Ga. peratures between 220 and 300 °C, with an interval of 20 °C, Therefore, the ratio of measured to the given De was 0.94±0.05. were tested for 10 s and the cut-heat temperature of 220 °C was Wintle and Murray (2006) suggested that the ratio of measured maintained for 10 s by using the heating rate with 5 °C/s. For to the given De can be acceptable between 0.9 and 1.1. each temperature point, four aliquots were used for De deter- The recycling ratio is also a check method for the sensi- mination. A temperature plateau exists between 240 and 280 °C, tivity change correction (Murray and Wintle, 2000). The recy- thus a preheat temperature of 260 °C was selected for routine cling ratios for all the aliquots are in the range of 0.9–1.1. The Luminescence Chronology and Radiocarbon Reservoir Age Determination of Lacustrine Sediments 699

regeneration dose of zero Ga (R0) is to measure recuperation, al. (2011) suggested that luminescence dating is satisfying to which was calculated by comparing its sensitivity-corrected date much younger samples (from 3 710 to 37 yr) for the lacus- OSL signal to the sensitivity-corrected natural signal. Recu- trine sediments from the Qinghai Lake. A modern sample of peration was in all cases less than 7% for all samples (Fig. 2c). aqueous sediments in Heihai region was determined, the De of The zero dose in Fig. 3a gave an OSL signal similar to the this sample is only 1.05±0.07 Ga at the depth of 0.4 m (Chang background, implying negligible recuperation contribution to et al., 2017), indicating the residual doses of samples for lacus- the De. Typical OSL decay curves and a growth curve for the trine sediments in this study are negligible. De histogram of part Sample HL2-1 are shown in Fig. 3. The OSL signals decreased samples were show in Figs. 2d, 2e, and 2f, implying that the very quickly during the first second of the stimulation, which lacustrine samples could be satisfactorily bleached. In the HL1 suggests that the OSL signal is dominated by the fast compo- and HL2 sections, the OSL ages are basically in order with the nent (Fig. 3a). The typical growth curve of the sample can be stratigraphic sequence, again suggesting that bleaching should well-fitted by an exponential plus a linear function. It is evi- not be a problem in these samples. dently found that no saturation was observed in this sample The 14C and OSL ages of the HL1 and HL2 sections are (Fig. 3b). listed in Tables 1 and 2. In the HL1 section, two 14C ages are 5 175±135 Cal yr BP and 5 595±85 Cal yr BP, and two OSL 2.2 Dating Results ages are 1.8±0.2 ka and 2.2±0.2 ka. In the HL2 section, three Lacustrine sediments were sometimes considered to be 14C ages are from 5 610±70 to 5 940±70 Cal yr BP, and the partially bleached prior to burial. However, considerable suc- four OSL ages are from 2.3±0.2 to 2.4±0.2 ka (the HL2-4 cess have been obtained applying OSL dating to a number of adopts the maximum with the error). For determining the lake different lacustrine settings to reconstruct the paleoclimate reservoir (LR), the OSL ages against radiocarbon ages were changes on the northeastern QTP (Zeng and Xiang, 2017; E et plotted, and the two group chronologic results show a remark- al., 2015; Liu et al., 2015, 2011; Fan et al., 2014, 2012). Liu et able positive relationship with a high linear correlation

2 400 1.2 (a) (b) 2 000

16 Gy 1 600

Lx Tx 0.8

1 200 TD

800 0.4 N

Normalized OSL( / )

OSLintensity (c/0.16s) 400 0 Gy

0.0 012340 5 10 15 20 Stimulation time (s) Dose (Gy)

Figure 3. Luminescence characteristics for aliquots of Sample HL2-1. OSL decay curves are shown in (a). Dose response curves are shown in (b).

Table 2 Sample information, grain size, environmental radioactivity, water content, and OSL dating results for the current study sections

Sample Depth Potamogeton Mean K Th U Water Dose rate De OSL age fossil abundance grain size content ID (m) (%) (μm) (%) (ppm) (ppm) (%) (Ga/ka) (Ga) (ka)

HL1-1 0.55 35 5.6 3.38±0.10 12.25±0.32 2.96±0.30 20±5 4.49±0.32 7.90±0.96 1.8±0.2 HL1-2 0.80 15 19.5 1.63±0.06 9.75±0.26 4.70±0.29 22±5 3.26±0.22 7.10±0.21 2.2±0.2 HL2-1 0.15 20 9.8 2.26±0.08 11.52±0.32 2.44±0.19 21±5 3.48±0.28 7.90±0.21 2.3±0.2 HL2-2 0.30 38 6.3 2.92±0.10 12.12±0.33 3.18±0.23 20±5 4.19±0.30 9.70±0.48 2.3±0.2 HL2-3 0.55 25 10.9 2.29±0.08 10.63±0.30 2.58±0.20 21±5 3.41±0.24 8.30±0.28 2.4±0.2 HL2-4 0.80 30 6.6 2.90±0.09 12.63±0.34 6.22±0.31 23±5 4.64±0.31 9.80±0.31 2.1±0.3 HP1-1 0.70 1.01±0.05 4.83±0.17 1.44±0.16 13±5 1.92±0.13 5.80±0.20 3.0±0.2 HT1-1 0.30 1.35±0.05 7.12±0.22 2.25±0.23 15±5 2.65±0.19 5.20±0.62 2.0±0.3 HF1-1 0.40 1.62±0.06 7.71±0.25 2.70±0.39 5±1 3.32±0.26 7.80±0.67 2.3±0.3

The Abundances of Potamogeton fossil were estimated values obtained from field investigation. 700 Fuyuan An, Zhongping Lai, Xiangjun Liu, Yixuan Wang, Qiufang Chang, Baoliang Lu and Xiaoyun Yang

nabe et al., 2010). Thus, modern reservoir corrections cannot be extrapolated “a priori into the past” (Geyh et al., 1974). Lockot et al. (2015) determined the 14C ages of three living plants from the Heihai Lake and argued a spatially consistent modern reservoir effect of 6 465±75 yr, which is higher than our determined age. However, the paleo plant residues in the highstand lacustrine sediments of the southeastern Heihai Lake deposited between 1.8±0.2 and 2.4±0.2 ka. Considering the temporal and spatial variation of the reservoir effect, the reservoir age range from 3 353 to 3 464 yr is reasonable, because climatic condition of this period is rather humid and the Heihai Lake ex- panded approximate one third of the present area according to

the area’s calculation of the exposed highstand lacustrine sedi- Figure 4. Plot of OSL ages against 14C ages for HL1 and HL2 section (dia- ments around the lake. Generally, lake level variations and the mond with error bars). The undeterminate radiocarbon age (HL2-3) was corresponding hydrological changes can influence the reservoir evaluated by the average age of HL2-2 and HL2-4. Black line represented effect (Lockot et al., 2015; Wu et al., 2011). Large lakes have the linear regression. an increased CO2 exchange rate with the atmosphere and dilute the former 14C-depleted lake water using an enhanced runoff (R2=0.924) (Fig. 4). It can be deduced that the LR follows the supply, thereby possessing less reservoir effect than small lakes formula: LR=Y–1 000X, here Y is radiocarbon ages (Cal yr BP) (Geyh et al., 1997). In addition, Heihai Lake is fed by glacier and X is OSL ages (ka). The linear regression equation (Y= meltwater and the sites of HL1 and HL2 both lie at the end of 1 184X+3 022) as shown in Fig. 4 is substituted in the above an alluvial fan that originated from the Kunlun Mountain gla- formula and it is solved for the equation: LR=184X+3 022. ciers, and there are no perennial streams on the piedmont allu- Therefore, when the OSL ages rang from 1.8 to 2.4 ka, corre- vial fan at present. However, the HF, HL1, and HL2 samples spondingly the LR varies from 3 353 to 3 464 yr. The paleo- all validated the presence of a strong hydrologic process in the shoreline section (HP) has an age of 3.0±0.2 ka, thereby sug- period of 2.4–1.8 ka. Therefore, the meltwater that entered the gesting the high lake level of Heihai Lake sustained at least 1.2 lake from the glaciers was most likely well-aerated at that time, ka. Other sections that are related to the hydrologic process in which resulted in an abundance of atmospheric 14C mixed in this catchment showed similar ages such as the top fluvial ter- the water, and further decreased the reservoir effect of the es- race (HT) of the lake outfall, which had an age of 2.0±0.3 ka, tuarine area. and the piedmont alluvial fan (HF) of the southern, which had The temporal variation of reservoir effect could extremely an outfall of 2.3±0.3 ka (Table 2). influence the paleoclimate reconstruction in lakes (Hou et al., 2012a). In the Nam Co Lake, the reservoir ages decreased dra- 3 DISCUSSION matically with depth, from 2 476 to 1 200 yr across the upper 3.1 Temporal Variation of the14C Reservoir Effect for 200 cm sediments (Zhu et al., 2008). The same phenomenon Heihai Lake appeared in the Qinghai Lake (1 650 to 800 yr between 1 000 The accurate radiocarbon dating of lake sediments is to 7 000 Cal yr BP) (Wang et al., 2007) and the Sugan Lake commonly hampered by the existence of the 14C reservoir ef- (2 627 to 4 342 yr during the Late Holocene) (Zhou et al., fect in arid areas such as the QTP due to sparse terrestrial 2009). Lockot et al. (2015) combined four age models for the vegetation in the catchments (Mischke et al., 2013; Hou et al., three core-units of Heihai Lake and finally determined an age 2012a). Fortunately, abundant fossil Potamogeton plants were belt for the reservoir-corrected and calibrated years arises. It preserved in the highstand lacustrine sediments for Heihai Lake, has a mean range of ~2 900 yr and an increasing uncertainty which provide suitable materials to determine its reservoir ef- towards the bottom (reach to ~3 700 yr). This age error would fect and reach the range from 3 353 to 3 464 yr during 1.8 to cause an erroneous paleoclimatic interpretation in the millen- 2.4 ka. Potamogeto, a submerged plant, is widely distributed in nial scale, especially in the Holocene. Our data show the reser- freshwater and saltwater lakes, including lakes on the QTP (Liu voir ages of Heihai Lake vary for 110 yr between 2.4 to 1.8 ka. and Liu, 2016). The aquatic plant residues in the Heihai Lake The temporal changes of reservoir effect should be noted as the belonged to ribbon-leaf Potamogeton plants, which are 5–25 age-depth model with radiocarbon ages for lacustrine se- cm long, 0.5–1 cm wide, and lack stalks. These fossil plant quences in the close basin is determined. residues accumulated compactly and create layered features in the wind-erosive lacustrine sediments on the southeast shore of 3.2 Origin of the 14C Reservoir Effect in Heihai Lake Heihai Lake. Lockot et al. (2015) studied two core sequences located in Previous lake studies show that, in comparing 14C dating the center of Heihai Lake. A positive correlation was observed with other dating techniques, the reservoir effect may vary between the allochthonous input of dolomite and the inverse temporally (e.g., Zhou et al., 2009; Geyh et al., 1974). Paleo- 14C ages, thereby suggesting an origin of the 14C-dead carbon climatic controlled on chemical weathering rates, depositional from a limestone catchment. Liu et al. (2009) reported that the processes, vegetation cover, and soil development vary through radiocarbon age was in excess for the 210Pb and 137 Cs ages of time and result in temporal changes in the reservoir age (Wata- the same layer by 3 400 yr, which indicates a reservoir age of Luminescence Chronology and Radiocarbon Reservoir Age Determination of Lacustrine Sediments 701

3 400 yr in Kusai Lake. Mischke et al. (2013, 2010) detected a centennial-scale. Thus it substantially produced grain size sort- shell collected from the inflow at the northernmost shore of the ing during deposited processes for the lacustrine sediments, this Donggi Cona Lake and found a LR effect of 3 300 yr. However, could result in the fluctuations of K, U, and Th element con- other sits around the lake showed great variations that ranged tents for the lacustrine sections. The fine particles can absorb from 900 to 20 000 yr, which vary much more than the more concentrations of K, U, and Th element than coarse parti- core-top sediments of the lake and suggests a spatial distribu- cles (Kenworthy et al., 2014). The grain-size distributions of tion for the LR effect and the local differences in the catchment lacustrine sediments for the two sections basically show nega- basement rocks and hydrology. Heihai Lake, Kusai Lake, and tive relationship with the radioactive element concentrations Donggi Cona Lake are all located in the same orogenic belts of (Table 2). In addition, in the two sections, the grayish-yellow the Kunlun Mountain. Carbonate bedrocks are present in the fine silt layers (containing 15%–25% aquatic plant residues) Bayan Har Shan Group of the Triassic Period, which distrib- showed higher K, U, and Th element concentrations, but the uted from the west to east along the Kunlun Mountain (Bureau yellowish-brown coarse silt layers (containing 30%–38% plant of Geology and Mineral Resources of Qinghai Province, 1991). residues) were opposite. This phenomenon indicated the rapid Therefore, similar reservoir ages are found in the three lakes, changes of hydrodynamic force and lake level during the which suggests the same 14C-depleted carbon sources that are high-rate accumulation process could lead to the variations of related to the carbonate bedrock. However, the exogenous external factors (e.g., different grain size particles input) and inputs of the 14C reservoir effect for a lake are mainly com- internal factors (e.g., abundance of aquatic plant residues, the prised of the input of peat and dead terrestrial plants, the dis- oxidation-deoxidizing circumstance in the sediments), which tribution springs and runoffs, crustal 14C-depleted carbon from synthetically affected on the distributions of K, U, and Th ele- tectonic events, and the dissolved inorganic carbon eroded ment concentrations, thus this phenomenon can be considered from carbonate-rich bedrock. These processes are rather com- as a peculiar characteristic during the highstand lacustrine plex and synthetic for the geologic period. Therefore, the 14C sediment fast deposited processes for Heihai Lake. reservoir effect of each lake in this region showed similarities In current study, we assumed that the tectonic activities of due to the consistent 14C-dead carbon origin, but further indi- the Kunlun Mountain involved in the lake level changes, such cated the temporal and spatial variations, as well their peculi- as the vertical movements, were negligible. In actuality, the arity. fluvial terrace formations in this region since the Holocene period were almost attributed to changes of the paleoclimate 3.3 Environment Significance for the OSL Ages (Chang et al., 2017; Wang et al., 2009). Among the five sections, as shown in Table 2, the paleo- The OSL ages of the highstand lacustrine sediments, pa- shoreline section (HP) has an age of 3.0±0.2 ka and is 9.0 m leoshorelines, fluvial terrace and alluvial fan deposits range (4 452 m a.s.l.) higher above the modern lake level of 4 443 m from 3.0±0.2 to 1.8±0.2 ka, thereby suggesting a rather humid in 2015 AD, which indicates the highest lake level for the past climate and hydrological condition as well as a high lake level 3.0 ka. The two highstand lacustrine sediments sections (HL1 during the period (Fig. 5). Previous studies provided evidence and HL2) have ages of 2.4±0.2 to 1.8±0.2 ka with elevations of of the widespread wet stage between ~3 and 2 ka across the 4 445 and 4 446 m. There were no laminas in these sediments, northeastern QTP. Fan et al. (2014, 2012) dated the paleoshore- indicating it was not a deep water deposit. The lake level was lines of Hurleg Lake, Toson Lake, and Gahai Lake of the assumed to be ~4 447–4 448 m a.s.l. with a maximum lake Delingha subbasin on the northeastern edge of the QB, and level of 4–5 m higher than present (the highest elevation of reported the existence of a high lake level in these lakes for the lacustrine sediment margin is ~4 448 m). The fluvial terrace age of 2.2 ka. Liu et al. (2009) reported a high TOC proxy (HT) of the lake outfall was formed at 2.0±0.3 ka with an ele- value in the bulk samples for the lacustrine sediments of Kusai vation of 4 446 m, which is similar to that of highstand lacus- Lake between the range of 2.6 and 1.8 ka. Yu and Lai (2014, trine sediments and suggests that reaching the high lake level could result in overflowing. The terminal age of the top sediments of the piedmont alluvial fan (HF) for the southern lake outfall is also 2.3±0.3 ka, and it resulted in the temporary damming of the outflow of the lake (Stauch et al., 2017). The previous study suggested the lacustrine sediments in Heihai Lake mainly originated from the Kunlun Mountain with the glacio-fluvial and precipitation induced runoff (Ramisch et al., 2016). Therefore it can be deduced that development of the alluvia fan could produce a relative high-rate accumulation for the highstand lacustrine sediments due to its upstream location and input of abundant clastic materials. The OSL data of the two lacustrine sections likewise indicated the lake sediments Figure 5. The fluctuation history of the high lake level for Heihai Lake accumulated during several hundred years, the deposition rate during 3.0–1.8 ka based on OSL dating. The empty square represents the reached to the range of 0.625–65 mm/yr. the extensive fluctua- paleoshoreline deposits, the empty diamond represents the alluvial fan tions of the deposition rate implied the drastic variations of deposits, the empty circles represent highstand lacustrine sediments, and the runoff yield from the alluvial fan within the time of empty triangle represents the fluvial terrace deposits. 702 Fuyuan An, Zhongping Lai, Xiangjun Liu, Yixuan Wang, Qiufang Chang, Baoliang Lu and Xiaoyun Yang

2012) proposed an effective moisture index for the aeolian dominant moisture source to the ASM during the monsoon sea- sediments of eastern QB and indicated the relative increase of son. However, Fan et al. (2014) dated the high paleoshoreline the effective moisture at ~3.4–2.2 ka. The Qinghai Lake also sediments and reconstructed the lake level history of Hurleg Lake, experienced the humid paleoclimatic stage and a high lake level and illuminated that the Westerlies and local topography, rather during the period of ~3 to 2 ka (Liu et al., 2015, 2011; An Z S than the Asian Summer Monsoon, dominate the moisture avail- et al., 2012). This result was found in the Nuomuhong Culture ability in this region during the Holocene. Zhao et al. (2007) site in the eastern QB, which permanently settled in this region considered the effective moisture in the Hurleg Lake region be- during 3 400–2 450 Cal yr BP with a humid climatic condition ing influenced by the Asian Monsoon, Westerlies, and the topog- stage (Dong et al., 2016). The tree rings from the mountains of raphy. Thus, the difference in climate interpretation of the QB the northeastern QTP were also proven to be relatively moist in suggests complexity of the Holocene climate change near the this period (Yang B et al., 2014). northwestern limit of the summer monsoon influence, especially in the Late Holocene Period. 3.4 Moisture Source of the Humid Stage of 3.0 to 1.8 ka In current study, we synthesized the paleoclimatic records The Holocene climate changes in the joint area of Westerly, of the three atmospheric systems zones (EASM, ISM, and EASM, and ISM for the QTP are still an important scientific Westerlies) (Fig. 6). In Fig. 6, the effective moisture index of the issue, especially in the Late Holocene, because of its extremely eastern QB and the summer monsoon index of Qinghai Lake unstable hydrological conditions (Yan and Wünnemann, 2014; show a relative increase between the periods of 3 to 2 ka (Figs. Herzschuh, 2006). Generally, the EASM brings warm and hu- 6a, 6b), thereby suggesting a stronger EASM condition (Yu and mid air from the Pacific Ocean to the southeastern half of Lai, 2014; An Z S et al., 2012). The Seling Co Lake, Bangong China, resulting in a northwestward decreasing trend of the Co Lake, and Hongyuan Peat are located in southwestern China mean annual precipitation. The ISM carries warm and wet air and are dominated by the ISM from the Indian Ocean. Their mass from the Indian Ocean to the southwestern part of China. humidity proxies all indicate a better hydrological condition in The Westerlies dominate the climates in Xinjiang region, even this period, which also suggested the enhancement of the ISM in northwestern part of the QTP during the last glaciation (Ran (Mügler et al., 2009; Hong et al., 2003; Gasse et al., 1996; Gu et and Feng, 2013). However, in the interactional sites of the three al., 1993) (Figs. 6d, 6e, and 6f). The widespread existence of a atmospheric circulation systems such as Heihai Lake (Fig. 1) high lake level in many lakes on the QTP for the period docu- the moisture sources of the Late Holocene are probably more mented the broad synchronicity of the Late Holocene climate complicated. across the monsoon region (Fan et al., 2014, 2012; An Z S et al., Chen et al. (1991) studied the Jiuzhoutai loess profiles of 2012; Liu X J et al., 2011; Mügler et al., 2009; Liu X Q et al., Lanzhou, which is west of the Chinese Loess Plateau, and 2009; Hong et al., 2003; Gasse et al., 1996). Zhang et al. (2011) found paleosol layer forming between 2.7 to 2.0 ka, which studied the δ18O records in the ISM region and the δ13C records corresponds to the Qin-Han warm period in Chinese history. An in both the ISM and EASM regions and suggested that the ISM et al. (2000) considered this humid stage as a relative en- and the EASM were broadly in phase at the millennial time- hancement of the EASM on the CLP. The precipitation of scales during the Holocene period. In addition, the summer Qinghai Lake was dominated by the ASM system and showed a monsoon index and the temperature curve of Qinghai Lake high summer monsoon index during the 3.0 to 2.0 ka, thereby show a positive correlation (Figs. 6b, 6c), likewise indicating a suggesting a relative increase in ASM (An Z S et al., 2012). variation trend of the heat and water as both were in phase dur- The study of aeolian sediments from eastern QB deposited ing the Late Holocene in the northeastern QTP. However, the since Holocene illustrated that the moisture change of the area average moisture index in arid Central Asia (Chen et al., 2008), was controlled by the Asian Summer Monsoon (ASM), but it which is regarded as a valid proxy for the Westerlies to show a was difficult to identify whether the regional precipitation was relatively dry stage between 3 and 1.8 ka (Fig. 6g), indicates the mainly controlled by the EASM or the ISM (Yu and Lai, 2014). Westerlies could not influence the precipitation of this region. The high TOC content and high humidity index in the Kusai Wang et al. (2013) synthesized the moisture index of the Xinji- Lake lacustrine sediments, which is somewhat west of the Hei- ang area and indicated that a gradual wetting climate dominated hai Lake, indicated a relative humid stage between 2.5 and 1.9 the Mid-to-Late Holocene, but showed a relatively dry stage ka, also suggesting a stronger summer monsoonal condition during 3 to 2 ka in this Westerlies-dominant region. Ramisch et (Liu et al., 2009). al. (2016) suggested that the orography of the Kunlun barrier Ran and Feng (2013) synthesized 33 Holocene regional forced moisture-bearing air masses that migrated on the ISM moisture records for the three above-mentioned atmospheric trajectories to suddenly ascend, thereby leading to intensified system zones in China, and found that it had parallel trends be- orographic precipitation such that the barrier blocked northward tween the moisture level in the QTP and the ISM strength re- migration for nearly 90% of summer precipitation. Through a trieved from Arabian Sea, which suggests the QTP had been comparison of the moisture proxies, it is evident that both the under the influence of the ISM throughout the Holocene. Zhang EASM and ISM dominated the high lake level of Heihai Lake et al. (2011) suggested that the moisture is mainly derived from during 3.0–1.8 ka. the ISM to the west of 110°E in southern China and extends fur- Lehmkuhl (1997) suggested the paleosol and peat devel- ther to the east in northern China. Tian et al. (2007) proposed that oped on the terminal moraines of the Nianbaoyuze Mountains the ISM influence extends to the middle part of QTP, and Clem- and adjacent areas (eastern QTP) during 3 to 2 ka, thereby ens et al. (2010) further confirmed that the Indian Ocean is the suggesting a warm period after a glacial advance in the Late Luminescence Chronology and Radiocarbon Reservoir Age Determination of Lacustrine Sediments 703

0 12 3 4 5 (ka) records of Heihai Lake follow the Holocene latitudinal insola- 3 tion gradient between the daily summer insolation rates of the (a) Himalayan barrier (30°) and the latitude of mean daily peak insolation (44°), thereby suggesting the precipitation intensities 2 in the region were forced by differentiated solar heating of the Northern Hemisphere and indicates dependency on the energy 0.6 (b) gradients (Ramisch et al., 2016). 0.5

Effective moisture index of eastern Qaidam Basin 1 0.4 4 CONCLUSIONS 24 14 0.3 Through OSL and C dating of the highstand lacustrine (c) sediments, paleoshoreline, alluvial fan, and the fluvial terrace

of the Qinghai Lake 20 Summer monsoon index deposits of Heihai Lake catchment, two points are conclude as follow. (1) According to the calculations of linear regression of 16 the radiocarbon and OSL dating ages for samples from the

Qinghai Lake Qinghai

temperature (ºC) temperature same depth of highstand lacustrine sediments, the quantitative 14C reservoir ages were determined to lie between 3 353 and 3 12 (d) -6

-5 ‰ 464 yr during the 1.8 to 2.4 ka. We consider the temporal varia- 0.4

-4 O ( ) tion of reservoir effect as an important factor for the paleocli-

(e) -3 δ mate reconstruction in lakes. The radiocarbon reservoir effect

0.3 Lake Co Seling of the Heihai Lake is similar to that of the adjacent Kusai Lake and Donggi Cona Lake. We suggest the origin of the 14C reservoir effect is probably the dissolution of the carbonate

Hongyuan 0.2 bed-rocks that were distributed along the Kunlun Mountain. (2) The OSL ages of highstand lacustrine sediments, pa-

humification index humification (f) leoshoreline, fluvial terraces, and alluvial fans deposits ranged

0.1 Wet from 3.0±0.2 to 1.8±0.2 ka, thereby suggesting the existence of

4 Lake a humid period at that time with a high lake level (maximum 9

index

Bangong Co Bangong

(g) Dry m higher than present) in Heihai Lake. This humid event was 3 wildly recorded in different sediments of the Qinghai-Tibetan Plateau as well as in the loess profile of the Chinese Loess Pla- 2 teau, which includes the East Asian Summer Monsoon and Indian Summer Monsoon dominant zone. Therefore, we argue 1

Average moisture of the arid Central Asia that the increased East Asian Summer Monsoon and Indian 0 12 3 4 5 (ka) Summer Monsoon both influenced the increase in humidity at Figure 6. Part of the proxy records of different atmospheric circulation system the Heihai Lake region in the wet stage. since the Late Holocene: (a) Effective moisture index of the eastern Qaidam Basin (Yu and Lai, 2014); (b) and (c) Summer Monsoon index and temperature ACKNOWLEDGMENTS records of Qinghai Lake (Hou et al., 2016; An Z S et al., 2012); (d) δ18O record This study was supported by the National Natural Science of Seling Co Lake (Gu et al., 1993); (e) humification index of Hongyuan Peat Foundation of China (No. 41401008), West Light Foundation (Hou et al., 2016; Hong et al., 2003); (f) the wet-dry variations of Bangong Co of the Chinese Academy of Sciences (No. Y412021005), Natu- Lake (Mügler et al., 2009; Gasse et al., 1996); (g) average moisture index for ral Science Foundation of Qinghai Province (No. arid Central Asia (Chen et al., 2008). The grey bar shows the wet and warm 2016-ZJ-926Q), and the instrument function development pro- paleoclimatic period in Asian Summer Monsoon zone. gram of the Chinese Academy of Sciences (No. Y410041013). We thank Lupeng Yu, Tianyuan Chen for their help and dis- Holocene. This period corresponded with temperature varia- cussions. The final publication is available at Springer via tions from the last 3 000 yr of the Holocene on the QTP and https://doi.org/10.1007/s12583-017-0972-9. north China based on their temperature curves comparisons. The surface temperature reconstructions for the past 11 300 yr REFERENCES CITED from globally distributed records indicated a relative warm Aitken, M. J., 1985. Thermoluminescence Dating. Academic Press, London period during 3 to 2 ka within the Northern Hemisphere range Aitken, M. J., 1998. An introduction to Optical Dating. Oxford University (29 records, 90°N to 30°N), and showed strong similarities Press, Oxford between high-resolution precipitation records and the hemi- An, F. Y., Lai, Z. P., Liu, X. J., et al., 2018. Abnormal Rb/Sr Ratio in Lacus- spheric temperature reconstruction curves, which consistently trine Sediments of Qaidam Basin, NE Qinghai-Tibetan Plateau: A Sig- associated the relative warmth with relative wetness, and it was nificant Role of Aeolian Dust Input. Quaternary International, 469: related to the interactions for the summer insolation and radia- 44–57. https://doi.org/10.1016/j.quaint.2016.12.050 tive forcing by greenhouse gases during the mid-to-late Holo- An, F. Y., Ma, H. Z., Wei, H. C., et al., 2012. Distinguishing Aeolian Signa- cene in the region (Marcott et al., 2013). The paleoclimatic ture from Lacustrine Sediments of the Qaidam Basin in Northeastern 704 Fuyuan An, Zhongping Lai, Xiangjun Liu, Yixuan Wang, Qiufang Chang, Baoliang Lu and Xiaoyun Yang

Qinghai-Tibetan Plateau and Its Palaeoclimatic Implications. Aeolian 163–178. https://doi.org/10.1016/j.quascirev.2005.02.006 Research, 4(6): 17–30. https://doi.org/10.1016/j.aeolia.2011.12.004 Hong, Y. T., Hong, B., Lin, Q. H., et al., 2003. Correlation between Indian An, Z. S., Colman, S. M., Zhou, W. J., et al., 2012. Interplay between the Ocean Summer Monsoon and North Atlantic Climate during the Holo- Westerlies and Asian Monsoon Recorded in Lake Qinghai Sediments since 32 cene. Earth and Planetary Science Letters, 211(3/4): 371–380. Ka. Scientific Reports, 2(8): 619–625. https://doi.org/10.1038/srep00619 https://doi.org/10.1016/s0012-821x(03)00207-3 An, Z. S., Porter, S. C., Kutzbach, J. E., et al., 2000. Asynchronous Holocene Hou, J. Z., D’Andrea, W. J., Liu, Z. H., 2012a. The Influence of 14C Reser- Optimum of the East Asian Monsoon. Quaternary Science Reviews, 19(8): voir Age on Interpretation of Paleolimnological Records from the Ti- 743–762. https://doi.org/10.1016/S0277-3791(99)00031-1 betan Plateau. Quaternary Science Reviews, 48: 67–79. Bureau of Geology and Mineral Resources of Qinghai Province, 1991. https://doi.org/10.1016/j.quascirev.2012.06.008 Regional Geology of Qinghai Province. Geological Publishing House, Hou, J. Z., D’Andrea, W. J., Liu, Z. H., 2012b. Geochronological Limitations Beijing (in Chinese with English Abstract) for Interpreting the Paleoclimatic History of the Tibetan Plateau. Qua- Chang, Q. F., Lai, Z. P., An, F. Y., et al., 2017. Chronology for Terraces of the ternary Sciences, 32 (3): 441–452. https://doi.org/10.3969/j.issn.1001- Nalinggele River in the North Qinghai-Tibet Plateau and Implications for Salt 7410.2012.03.10 (in Chinese with English Abstract) Lake Resource Formation in the Qaidam Basin. Quaternary International, Hou, J. Z., Huang, Y. S., Zhao, J., et al., 2016. Large Holocene Summer 430: 12–20. https://doi.org/10.1016/j.quaint.2016.02.022 Temperature Oscillations and Impact on the Peopling of the Northeast- Chen, F. H., Li, J. J., Zhang, W. X., 1991. Loess Stratigraphy of the Lanzhou ern Tibetan Plateau. Geophysical Research Letters, 43(3): 1323–1330. Profile and Its Comparison with Deep-Sea Sediment and Ice Core Record. https://doi.org/10.1002/2015gl067317 GeoJournal, 24(2): 201–209. https://doi.org/10.1007/bf00186017 Kenworthy, M. K., Rittenour, T. M., Pierce, J. L., et al., 2014. Lumines- Chen, F. H., Yu, Z. C., Yang, M. L., et al., 2008. Holocene Moisture Evolu- cence Dating without Sand Lenses: An Application of OSL to tion in Arid Central Asia and Its Out-Of-Phase Relationship with Asian Coarse-Grained Alluvial Fan Deposits of the Lost River Range, Idaho, Monsoon History. Quaternary Science Reviews, 27(3/4): 351–364. USA. Quaternary Geochronology, 23(10): 9–25. https://doi.org/10.1016/j.quascirev.2007.10.017 https://doi.org/10.1016/j.quageo.2014.03.004 Clemens, S. C., Prell, W. L., Sun, Y. B., 2010. Orbital-Scale Timing and Lai, Z. P., 2010. Chronology and the Upper Dating Limit for Loess Samples Mechanisms Driving Late Pleistocene Indo-Asian Summer Monsoons: from Luochuan Section in the Chinese Loess Plateau Using Quartz Reinterpreting Cave Speleothem δ18O. Paleoceanography, 25(4): OSL SAR Protocol. Journal of Asian Earth Sciences, 37(2): 176–185. 545–558. https://doi.org/10.1029/2010pa001926 https://doi.org/10.1016/j.jseaes.2009.08.003 Dong, G. H., Ren, L. L., Jia, X., et al., 2016. Chronology and Subsistence Lai, Z. P., Brückner, H., 2008. Effects of Feldspar Contamination on Strategy of Nuomuhong Culture in the Tibetan Plateau. Quaternary Equivalent Dose and the Shape of Growth Curve for OSL of Silt-Sized International, 426: 42–49. https://doi.org/10.1016/j.quaint.2016.02.031 Quartz Extracted from Chinese Loess. Geochronometria, 30(1): 49–53. E, C. Y., Lai, Z. P., Hou, G. L., et al., 2015. Age Determination for a Neo- https://doi.org/10.2478/v10003-008-0010-0 lithic Site in Northeastern Qinghai-Tibetan Plateau Using a Combined Lai, Z. P., Wintle, A. G., 2006. Locating the Boundary between the Pleistocene Luminescence and Radiocarbon Dating. Quaternary Geochronology, and the Holocene in Chinese Loess Using Luminescence. The Holocene, 30: 411–415. https://doi.org/10.1016/j.quageo.2015.01.007 16(6): 893–899. https://doi.org/10.1191/0959683606hol980rr Fan, Q. S., Ma, H. Z., Cao, G. C., et al., 2012. Geomorphic and Chrono- Lai, Z. P., Zöller, L., Fuchs, M., et al., 2008. Alpha Efficiency Determination for metric Evidence for High Lake Level History in Gahai Lake and Toson OSL of Quartz Extracted from Chinese Loess. Radiation Measurements, Lake of North-Eastern Qaidam Basin, North-Eastern Qinghai-Tibetan 43(2/3/4/5/6): 767–770. https://doi.org/10.1016/j.radmeas.2008.01.022 Plateau. Journal of Quaternary Science, 27(8): 819–827. Lehmkuhl, F., 1997. Late Pleistocene, Late-Glacial and Holocene Glacier https://doi.org/10.1002/jqs.2572 Advances on the Tibetan Plateau. Quaternary International, 38/39: Fan, Q. S., Ma, H. Z., Wei, H. C., et al., 2014. Holocene Lake-Level 77–83. https://doi.org/10.1016/s1040-6182(96)00025-0 Changes of Hurleg Lake on Northeastern Qinghai-Tibetan Plateau and Liu, H., Liu, W. G., 2016. N-Alkane Distributions and Concentrations in Possible Forcing Mechanism. The Holocene, 24(3): 274–283. Algae, Submerged Plants and Terrestrial Plants from the Qing- https://doi.org/10.1177/0959683613517399 hai-Tibetan Plateau. Organic Geochemistry, 99: 10–22. Gasse, F., Fontes, J. C., van Campo, E., et al., 1996. Holocene Environmental https://doi.org/10.1016/j.orggeochem.2016.06.003 Changes in Bangong Co Basin (Western Tibet). Part 4: Discussion and Con- Liu, K., Lai, Z. P., 2012. Chronology of Holocene Sediments from the Ar- clusions. Palaeogeography, Palaeoclimatology, Palaeoecology, 120(1/2): chaeological Salawusu Site in the Mu Us Desert in China and Its Pa- 79–92. https://doi.org/10.1016/0031-0182(95)00035-6 laeoenvironmental Implications. Journal of Asian Earth Sciences, Geyh, M. A., Krumbein, W. E., Kudrass, H. -R., 1974. Unreliable 14C Dating of 45(4): 247–255. https://doi.org/10.1016/j.jseaes.2011.11.002 Long-Stored Deep-Sea Sediments Due to Bacterial Activity. Marine Geology, Liu, X. J., Lai, Z. P., Fan, Q. S., et al., 2010. Timing for High Lake Levels of 17(1): M45–M50. https://doi.org/10.1016/0025-3227(74)90043-7 Qinghai Lake in the Qinghai-Tibetan Plateau since the Last Geyh, M. A., Schotterer, U., Grosjean, M., 1997. Temporal Changes of the Interglaciation Based on Quartz OSL Dating. Quaternary Geochronology, 14C Reservoir Effect in Lakes. Radiocarbon, 40(2): 921–931. 5(2/3): 218–222. https://doi.org/10.1016/j.quageo.2009.03.010 https://doi.org/10.1017/s0033822200018890 Liu, X. J., Lai, Z. P., Madsen, D., et al., 2011. Lake Level Variations of Gu, Z. Y., Liu, J. Q., Yuan, B. Y., et al., 1993. Monsoon Variations of the Qinghai Lake in Northeastern Qinghai-Tibetan Plateau since 3.7 ka Qinghai-Xizang Plateau during the Last 12,000 Years—Geochemical Based on OSL Dating. Quaternary International, 236(1/2): 57–64. Evidence from the Sediments in the Siling Lake. Chinese Science Bul- https://doi.org/10.1016/j.quaint.2010.08.009 letin, 38(7): 577–581 (in Chinese with English Abstract) Liu, X. J., Lai, Z. P., Madsen, D., et al., 2015. Last Deglacial and Holocene Herzschuh, U., 2006. Palaeo-Moisture Evolution in Monsoonal Central Asia Lake Level Variations of Qinghai Lake, North-Eastern Qinghai-Tibetan during the Last 50 000 Years. Quaternary Science Reviews, 25(1/2): Plateau. Journal of Quaternary Science, 30(3): 245–257. Luminescence Chronology and Radiocarbon Reservoir Age Determination of Lacustrine Sediments 705

https://doi.org/10.1002/jqs.2777 derestimation Using Different OSL Signals from Lacustrine Quartz and Liu, X. Q., Dong, H. L., Yang, X. D., et al., 2009. Late Holocene Forcing of Polymineral Fine Grains. Quaternary Science Reviews, 22(10/11/12/13): the Asian Winter and Summer Monsoon as Evidenced by Proxy Re- 1139–1143. https://doi.org/10.1016/s0277-3791(03)00045-3 cords from the Northern Qinghai-Tibetan Plateau. Earth and Planetary Tian, L., Yao, T., MacClune, K., et al., 2007. Stable Isotopic Variations in Science Letters, 280(1/2/3/4): 276–284. West China: A Consideration of Moisture Sources. Journal of https://doi.org/10.1016/j.epsl.2009.01.041 Geophysical Research: Atmospheres, 112(D10): 185–194. Lockot, G., Ramisch, A., Wünnemann, B., et al., 2015. A Process- and https://doi.org/10.1029/2006jd007718 Provenance-Based Attempt to Unravel Inconsistent Radiocarbon Wang, A., Smith, J. A., Wang, G. C., et al., 2009. Late Quaternary River Chronologies in Lake Sediments: An Example from Lake Heihai, Terrace Sequences in the Eastern Kunlun Range, Northern Tibet: A North Tibetan Plateau (China). Radiocarbon, 57(5): 1003–1019. Combined Record of Climatic Change and Surface Uplift. Journal of https://doi.org/10.2458/azu_rc.57.18221 Asian Earth Sciences, 34(4): 532–543. Long, H., Lai, Z. P., Wang, N. A., et al., 2011. A Combined Luminescence https://doi.org/10.1016/j.jseaes.2008.09.003 and Radiocarbon Dating Study of Holocene Lacustrine Sediments from Wang, W., Feng, Z., Ran, M., et al., 2013. Holocene Climate and Vegetation Arid Northern China. Quaternary Geochronology, 6(1): 1–9. Changes Inferred from Pollen Records of Lake Aibi, Northern Xinjiang, https://doi.org/10.1016/j.quageo.2010.06.001 China: A Potential Contribution to Understanding of Holocene Climate Marcott, S. A., Shakun, J. D., Clark, P. U., et al., 2013. A Reconstruction of Pattern in East-Central Asia. Quaternary International, 311: 54–62. Regional and Global Temperature for the Past 11,300 Years. Science, https://doi.org/10.1016/j.quaint.2013.07.034 339(6124): 1198–1201. https://doi.org/10.1126/science.1228026 Wang, X., Zhou, A., Sun, Z., 2016. Spatial and Temporal Dynamics of Mischke, S., Aichner, B., Diekmann, B., et al., 2010. Ostracods and Stable Lakes in Nam Co Basin, 1991–2011. Journal of Earth Science, 27(1): Isotopes of a Late Glacial and Holocene Lake Record from the NE 130–138. https://doi.org/10.1007/s12583-016-0634-3 Tibetan Plateau. Chemical Geology, 276(1/2): 95–103. Wang, Y., Shen, J., Wu, J. L., et al., 2007. Hard-Water Effect Correction of https://doi.org/10.1016/j.chemgeo.2010.06.003 Lacustrine Sediment Ages Using the Relationship between 14C Levels Mischke, S., Weynell, M., Zhang, C. J., et al., 2013. Spatial Variability of in Lake Waters and in the Atomosphere: The Case of Lake Qinghai. 14C Reservoir Effects in Tibetan Plateau Lakes. Quaternary International, Journal of Lake Sciences, 19(5): 504–508. 313/314: 147–155. https://doi.org/10.1016/j.quaint.2013.01.030 https://doi.org/10.18307/2007.0502 (in Chinese with English Abstract) Mügler, I., Gleixner, G., Günther, F., et al., 2009. A Multi-Proxy Approach to Watanabe, T., Matsunaka, T., Nakamura, T., et al., 2010. Changes of Or- Reconstruct Hydrological Changes and Holocene Climate Develop- ganic Matter Sources in Sediment Cores from a High-Altitude Lake ment of Nam Co, Central Tibet. Journal of Paleolimnology, 43(4): (Pumoyum Co, Southeastern Tibetan Plateau) over the Last 19 000 625–648. https://doi.org/10.1007/s10933-009-9357-0 Years. Nuclear Instruments and Methods in Physics Research Section Murray, A. S., Wintle, A. G., 2000. Luminescence Dating of Quartz Using an B: Beam Interactions with Materials and Atoms, 268(7/8): 1070–1072. Improved Single-Aliquot Regenerative-Dose Protocol. Radiation https://doi.org/10.1016/j.nimb.2009.10.100 Measurements, 32(1): 57–73. Wintle, A. G., Murray, A. S., 2006. A Review of Quartz Optically Stimu- https://doi.org/10.1016/s1350-4487(99)00253-x lated Luminescence Characteristics and Their Relevance in Sin- Murray, A. S., Wintle, A. G., 2003. The Single Aliquot Regenerative Dose gle-Aliquot Regeneration Dating Protocols. Radiation Measurements, Protocol: Potential for Improvements in Reliability. Radiation Measure- 41(4): 369–391. https://doi.org/10.1016/j.radmeas.2005.11.001 ments, 37(4/5): 377–381. https://doi.org/10.1016/s1350-4487(03)00053-2 Wu, Y. H., Wang, S. M., Zhou, L. P., 2011. Possible Factors Causing Older Prescott, J. R., Hutton, J. T., 1994. Cosmic Ray Contributions to Dose Rates Radiocarbon Age for Bulk Organic Matter in Sediment from Daihai for Luminescence and ESR Dating: Large Depths and Long-Term Time Lake, North China. Radiocarbon, 53(2): 359–366. Variations. Radiation Measurements, 23(2/3): 497–500. https://doi.org/10.1017/s0033822200056617 https://doi.org/10.1016/1350-4487(94)90086-8 Yan, D. D., Wünnemann, B., 2014. Late Quaternary Water Depth Changes Ramisch, A., Lockot, G., Haberzettl, T., et al., 2016. A Persistent Northern in Hala Lake, Northeastern Tibetan Plateau, Derived from Ostracod Boundary of Indian Summer Monsoon Precipitation over Central Asia Assemblages and Sediment Properties in Multiple Sediment Records. during the Holocene. Scientific Reports, 6(1): 25791. Quaternary Science Reviews, 95: 95–114. https://doi.org/10.1038/srep25791 https://doi.org/10.1016/j.quascirev.2014.04.030 Ran, M., Feng, Z. D., 2013. Holocene Moisture Variations across China and Yang, B., Qin, C., Wang, J., et al., 2014. A 3 500-Year Tree-Ring Record of Driving Mechanisms: A Synthesis of Climatic Records. Quaternary Inter- Annual Precipitation on the Northeastern Tibetan Plateau. Proceedings national, 313/314: 179–193. https://doi.org/10.1016/j.quaint.2013.09.034 of the National Academy of Sciences, 111(8): 2903–2908. Reimer, P. J., Baillie, M. G. L., Bard, E., et al., 2004. IntCal04 Terrestrial https://doi.org/10.1073/pnas.1319238111 Radiocarbon Age Calibration, 0–26 Cal Kyr BP. Radiocarbon, 46(3): Yang, K., Wu, H., Qin, J., et al., 2014. Recent Climate Changes over the 1029–1058. https://doi.org/10.1017/s0033822200032999 Tibetan Plateau and Their Impacts on Energy and Water Cycle: A Re- Shen, Z. X., Bloemendal, J., Mauz, B., et al., 2008. Holocene Environmental view. Global and Planetary Change, 112: 79–91. Reconstruction of Sediment-Source Linkages at Crummock Water, https://doi.org/10.1016/j.gloplacha.2013.12.001 English Lake District, Based on Magnetic Measurements. The Yu, L. P., Lai, Z. P., 2012. OSL Chronology and Palaeoclimatic Implica- Holocene, 18(1): 129–140. https://doi.org/10.1177/0959683607085604 tions of Aeolian Sediments in the Eastern Qaidam Basin of the Stauch, G., Schulte, P., Ramisch, A., et al., 2017. Landscape and Climate on Northeastern Qinghai-Tibetan Plateau. Palaeogeography, Palaeocli- the Northern Tibetan Plateau during the Late Quaternary. Geomor- matology, Palaeoecology, 337/338: 120–129. phology, 286: 78–92. https://doi.org/10.1016/j.geomorph.2017.03.008 https://doi.org/10.1016/j.palaeo.2012.04.004 Thomas, P. J., Murray, A. S., Sandgren, P., 2003. Age Limit and Age Un- Yu, L. P., Lai, Z. P., 2014. Holocene Climate Change Inferred from Strati- 706 Fuyuan An, Zhongping Lai, Xiangjun Liu, Yixuan Wang, Qiufang Chang, Baoliang Lu and Xiaoyun Yang

graphy and OSL Chronology of Aeolian Sediments in the Qaidam Ba- and Habitat Relationships in the Kunlun Mountains, Northern Tibetan sin, Northeastern Qinghai–Tibetan Plateau. Quaternary Research, Plateau. Quaternary International, 313/314: 38–46. 81(3): 488–499. https://doi.org/10.1016/j.yqres.2013.09.006 https://doi.org/10.1016/j.quaint.2013.06.020 Zeng, F. M., Liang, M. Y., Peng, S. Z., et al., 2015. Sr-Nd-Pb Isotopic Zhao, Y., Yu, Z. C., Chen, F. H., et al., 2007. Holocene Vegetation and Cli- Compositions of the Neogene Eolian Deposits in the Xining Basin and mate History at Hurleg Lake in the Qaidam Basin, Northwest China. Implications for Their Dust Sources. Journal of Earth Science, 26(5): Review of Palaeobotany and Palynology, 145(3/4): 275–288. 669–676. https://doi.org/10.1007/s12583-015-0575-2 https://doi.org/10.1016/j.revpalbo.2006.12.002 Zeng, F. M., Xiang, S. Y., 2017. Geochronology and Mineral Composition Zhou, A. F., Chen, F. H., Wang, Z. L., et al., 2009. Temporal Change of of the Pleistocene Sediments in Xitaijinair Salt Lake Region, Qaidam Radiocarbon Reservoir Effect in Sugan Lake, Northwest China during Basin: Preliminary Results. Journal of Earth Science, 28(4): 622–627. the Late Holocene. Radiocarbon, 51(2): 529–535. https://doi.org/10.1007/s12583-016-0712-6 https://doi.org/10.1017/s0033822200055909 Zhang, J., Chen, F., Holmes, J. A., et al., 2011. Holocene Monsoon Climate Zhu, L. P., Wu, Y. H., Wang, J. B., et al., 2008. Environmental Changes Documented by Oxygen and Carbon Isotopes from Lake Sediments and since 8.4 Ka Reflected in the Lacustrine Core Sediments from Nam Co, Peat Bogs in China: A Review and Synthesis. Quaternary Science Reviews, Central Tibetan Plateau, China. The Holocene, 18(5): 831–839. 30(15/16): 1973–1987. https://doi.org/10.1016/j.quascirev.2011.04.023 https://doi.org/10.1177/0959683608091801 Zhang, W. Y., Mischke, S., Zhang, C. J., et al., 2013. Ostracod Distribution