A grain size and n-alkanes record of Holocene environmental evolution from a groundwater recharge lake in Badain Jaran Desert,Northwestern China

Abstract

Vertical water transport plays an important role in the development and maintenance of lakes in arid zones. However, previous study of this phenomenon has focused mainly on the effects of regional precipitation, rather than effects of groundwater recharge, on arid-zone environmental change. We selected a sedimentary profile from a seasonal lake basin located in the hinterland of the Badain Jaran Desert, northwestern China, to assess the Holocene environmental evolution of this region and its response to climate change. Our results show that peat deposition from 11 to 10 ka indicates humidification after Younger Dryas (YD) event; the lake sediment and lower EM2 and lower principal component analysis (PCA)-1 scores indicate relative more humid environment from 10 to 7.8 ka; the lake sediment and higher EM2 and higher PCA-1 scores indicate most humid environment from 7.8 to 5.8 ka; the transition from lacustrine facies to limnetic facies and increasing salinity indicate drying trend from 5.8 to 2.5 ka; the aeolian sand sediment and lowest EM2 and lowest PCA-1 scores indicate driest environment from 2.5 to 0.8 ka; and lacustrine facies and finest grain size indicate relative humid environment during the ‘Little Ice Age’ period. The environment was relative humid in early Holocene, most Humid in mid-Holocene, and dry in mid-late Holocene. The moisture pattern in the hinterland of the Badain Jaran Desert was synchronous with that in the northeastern of Tibetan Plateau, which implies that the groundwater recharge lake was dominated by the intensity of groundwater recharge from monsoon margin area, especially the northeastern of Tibetan Plateau. As a record from groundwater recharge lake without runoff, our study provides the groundwater output quantity record from northeastern of Tibetan Plateau and would benefit the environment change research and regional water resources allocation around the Tibetan Plateau.

Keywords

Badain Jaran Desert effective humidity groundwater recharge lake Holocene paleotemperature

Introduction

The traditional model of lake evolution is focused on the important role of atmospheric precipitation in the hydrological system (An et al., 2000; Chen et al., 2008). Atmospheric circulation is regarded as a key factor for basin hydrological change based on its influence on regional precipitation (Long et al., 2017). In Xinjiang Province (Zhang and Li, 2005b) and the Hexi Corridor (Zhang and Li, 2005a), in the arid regions of northwestern China, the amount of natural recharge of groundwater in the main inland river basins is at least equal to, if not greater than, the surface water resources. Groundwater recharge plays a crucial role in the maintenance and development of the internal water cycle of these basins. Furthermore, arid zones are more sensitive to temperature changes during global climate change (Huang et al., 2016) and the resulting changes in evaporation (Li et al., 2013). Changes in groundwater recharge and evapotranspiration lead to the expansion and shrinkage of lakes (Wang et al., 2016).

The mode of water transport in these regions is characterized by intensely interspheres vertically water circle and material system interaction, which is different considerably from the traditional mode driven by atmospheric precipitation. As a result, the effective humidity, controlled by interspheres vertically water circle, determines the lake evolution mode in arid regions. Therefore, modeling bi-directional groundwater–climate coupling based on the relationship between land hydrology and regional climate is key to understanding the response mechanisms of the water cycle in groundwater recharge lakes under climate change.

The Badain Jaran Desert, located in a monsoon–westerly transition zone, is sensitive to climate change (Yang et al., 2010, 2011). Because the hinterland of this desert area is extremely arid, there is almost no surface runoff and groundwater recharge accounts for more than 90% of total lake recharge (Dong et al., 2016; Ma et al., 2014; Wang et al., 2016). Therefore, the lakes of this region are an ideal location to study environmental evolution under the water vapor transport model. Previous studies of the Holocene environmental evolution of the Badain Jaran Desert hinterland have mainly focused on geomorphology and lacustrine relics (Wang et al., 2016; Yang et al., 2010), without systematic analysis of lake sedimentation records. Based on the analysis of grain size, salinity, ratio of carbon to nitrogen (C/N), and n-alkanes as proxies, we inferred paleotemperature and hydroclimate changes of groundwater recharge lakes during the Holocene in this extremely arid region.

Geographical setting

The Badain Jaran Desert is located in the west of the Alxa Plateau, northwestern China (Figure 1). With an area of 52,162 km², it is the second largest desert in China after the Taklimakan Desert (Zhu et al., 2010). Groups of high megadunes cover the southeast of the desert and reach 200–400 m in height. The dunes strike 10–30°NE with a dominant wind direction of SW–NE. There are 110 lakes in total in the interdune areas of the megadunes, including perennial lakes, seasonal lakes, and extinct lakes. These lakes vary greatly in area and hydrochemical properties (Lu et al., 2010). The largest lake, named Nuoertu, covers an area of 1.5 km² and has a depth of 15.9 m (Zhang et al., 2013). The monthly average temperature in this area ranges from 7.6°C (January) to 27°C (July), and the annual average temperature is 11.1°C (Figure 1c). Precipitation is concentrated between May and September, and the average annual rainfall is 87 mm. Annual evaporation of a lake with a TDS of 304.5 g/L (Lu et al., 2010) is about 1450 mm (Dong et al., 2016; Wang et al., 2016). Without overland runoff, groundwater recharge is the main source of water for these lakes (Dong et al., 2016; Wang et al., 2016). The surface vegetation coverage is sparse and mainly composed of xerophytes and super-arid shrubs and semi-shrubs. The surface sediments of the desert consist mainly of fine sand (Ning et al., 2013).

Figure 1.

(a) The location of Sedimentary profile ZZH in the Zhunzhahan Jaran Basin. (b) The location of Sedimentary profile ZZH on a map of the climate system of China. (c) Monthly average temperatures and precipitation (2010–2014) from V1 automatic meteorological station in the Badain Jaran Desert, Northwest China.

The Badain Jaran Desert lies within the depression basin of the Alxa Plateau. Since the early Pleistocene, the crust has sunk considerably to form this basin, and 150 m of Quaternary sediments have accumulated therein. Grid-like fracture structures are developed in the underlying layers. The desert has experienced MIS (Marine isotope stages) 13–15, MIS 5, and Holocene wetting periods since 1200 ka (Wang et al., 2015) and permafrost periods during the Last Glacial Maximum and the Late Glacial (Ning et al., 2015).

Materials and methods

Lithology

We obtained a 390-cm-long sediment section (named sedimentary profile ZZH) extracted from a seasonal lake depression (39°57′N, 102°37′E; 1246 m a.s.l.) in the hinterland of Badain Jaran Desert. The samples were collected at 2 cm intervals from 0 to 220 cm depth and at 5 cm intervals from 220 to 390 cm depth, which yielded a total of 138 samples. The section can be divided into seven lithological units based on texture, color, and structure (see Figure 2): (1) below 385 cm, brown loose aeolian sand immersing in the groundwater level; (2) 385–348 cm, a black peat layer mixed with abundant plant residues; (3) 348–175 cm, yellow-green silt-sand. In the under layer (348–248 cm), it shows grayish-green. In the upper layer (248–175 cm), it shows permanently yellowish-brown with grayish-green rust spots. (4) 170 −125 cm, a gray-black silt limnetic layer with more silt; (5) 125–100 cm, a yellow sand layer with rust spots; (6) 100–40 cm, a gray-black limnetic layer; and (7) above 40 cm, grayish-yellow moderately consolidated sand layer.

Figure 2.

Sedimentary log and photo of profile ZZH.

Analytical methods

Eight samples were collected from Sedimentary profile ZZH for accelerator mass spectrometry (AMS) ¹⁴C dating to set up chronostratigraphic sequence. Three of the samples were dated at the AMS Dating Laboratory at Beijing University, China, and the other four were obtained by Beta Analytic, Florida of the USA. All dates were calibrated to calendar years before present (BP, before 1950 AD) with program CALIB 6.0.1 using IntCal09 calibration data set (Reimer et al., 2009). The age–depth model (Figure 3) was established with the age control points using fitting spline functions of Clam 2.2 program in statistical software package R (Blaauw, 2010). At the same time, additional organic matter samples were collected to evaluate the possible effect about ‘carbon storage effect’.

Figure 3.

Lithostratigraphy and chronology for Sedimentary profile ZZH.

Grain size distribution was determined using a Mastersizer 2000 particle analyzer from Malvern, the Netherlands. A total of 138 samples were pretreated by heating to 100°C in 106 mL of 10% H₂O₂ and 10 mL of 10% hydrochloric acid to remove organic matter and carbonate and left at room temperature for 12 h. The upper solution of each sample was then removed and shaken in dispersant (0.05 mol/L sodium hexametaphosphate) for 20 min prior to analysis. Grain sizes were measured in the particle size range from 0.02 to 2000 μm with repetitive error of less than 2%. The end-member analysis was carried by the end-member-modeling algorithm (EMMA; Weltje, 1997; Weltje and Prins, 2003).

Total organic carbon and total nitrogen contents were determined with an elemental analyzer. Freeze-dried samples of known mass were treated with hydrochloric acid for 24 h to remove carbonate and then washed with distilled water to neutral pH. The freeze-dried samples were then measured using a Vario-III elemental analyzer made by Elementar, Germany.

Salinity was calculated based on ions identified via liquid chromatography. Samples were shaken for 3 min with deionized water and centrifuged. Then, the samples and solutions were filtered through a 0.22-μm filter prior to analysis. Ions were identified with an ICS-1500 ion chromatography system produced by Dionex Corporation, America.

The n-alkanes were identified with a 450-GC gas chromatograph produced by Varian Corporation, America. Three grams of each freeze-dried sample was extracted with the extract liquor (CH₂Cl₂:MeOH = 93:7) for 12 h and ultrasonicated at room temperature five times for 20 min each. The resulting extracts were combined and filtered, and the filtrate was concentrated under gentle ultrapure nitrogen flow. The concentrate was passed through a silica gel column and eluted with n-hexane. The eluate was concentrated in a gentle ultrapure nitrogen stream and derivatized with BSTFA. The volume of the eluate was directly measured using the Varian 450-GC device; a flame ionization detector and a set of n-alkanes (nC12 to nC40) of known concentrations were used as internal standards for quantification. The concentration distributions of n-alkanes were used to calculate environmental proxies, such as average chain length (ACL) (Poynter et al., 1989), carbon preference index (CPI) (Allan and Douglas, 1977), odd-even predominance (OEP) (Yunker et al., 1994), nC27/(nC27 + nC31) (Bai et al., 2009; Xie et al., 2003), proportion of aquatic plant n-alkanes (Paq) (Ficken et al., 2000), proportion of land plant n-alkanes (Pwax) (Ficken et al., 2000), and the total concentration of n-alkanes (Con_nak).

To further determine the response patterns of the distributions of n-alkanes and grain size to climate change, we conducted a dimensionality reduction analysis. The mode selection and the operation of the dimension reduction analysis were carried out using the software Canoco v. 5.0. Because the maximum length of the gradient of the detrended correspondence analysis (DCA) is 1.4 times the standard deviation, that is, less than 2, principal component analysis (PCA; linear model) was used (Birks and Birks, 1998).

Results

Core chronology

The freshwater snail shells from Wosigu Lake (this lake has retreated at present), located in the southeast of the Badain Jaran Desert, were dated at 47–273 cal yr BP (laboratory preference: BA111630) and 306–469 cal yr BP (laboratory preference: BA111631), which are consistent with the latest humid period ‘the Little Ice Age’. The modern aquatic plant material from Tian’e Lake in the northwest of the Badain Jaran Desert was dated to be modern (laboratory preference: LUG06-132), indicating that even the aquatic plant material could be directly used to reconstruct chronosequence without correction of ‘Carbon Pool Effect’. Generally, there may be ‘carbon pool effect’ in shells and aquatic plants. As suggested above, the ¹⁴C dating results included in this study are less strongly affected by the carbon pool and are therefore reliable evidence for reconstruction of the Holocene environment.

The detailed sample properties and dating results are listed in Table 1. All the dating results were used to establish age-depth model in Clam (see Figure 3). The deposition rate stands overall stability value of about 0.25 mm·yr⁻¹ except for that above 100 cm with a maximum value of about 5.5 mm·yr⁻¹ and that below 320 cm with a maximum value of about 2.5 mm·yr⁻¹. As the sediment mainly derived from aeolian sand around the lake in the groundwater recharge lake without almost surface runoff input, the deposition rate is higher when the lake had not yet formed or drying and profile lay nearer with aeolian sand around the lake, vice versa. The depositional pattern is very different from that in general lakes of arid area. The resulting age–depth relationship indicates that the peat and the aeolian sand layer has accumulated at a higher constant rate, while the lacustrine strata has accumulated at a lower constant rate, which is consistent with the wind transport capacity during the corresponding periods. As mentioned above, the photo of profile ZZH also shows homogeneous lithology without hiatus or rapid accumulation records. Despite the lack of extra dating data, we regard the chronosequence to be credible.

Table 1.

Dating results for the ZZH section in the hinterland of the Badain Jaran Desert.

Lab-Ref	Sample number	Depth (cm)	Material dated	¹⁴C Age (a BP)	Error (a)	Age (cal yr BP)
Beta-457614	ZZHA78	44	Charcoal	140	30	0–284
Beta-457615	ZZHA62	76	Plant residues	170	30	0–290
Beta-453861	FX1502	149	Plant residues	3160	30	3449–3273
Beta-453860	FX1501	168	Plant residues	3260	30	3403–3565
BA121091	ZZHA-LQ	248	Plant residues	8350	30	9295–9461
BA121090	ZZH-01	348	Plant residues	9245	50	10265–10555
BA121089	ZZH-02	385	Plant residues	9560	45	10722–11098

General changes in grain size, TOC, and salinity

Most of the sediment in the core, 43–67%, is fine sand, with a median grain size range of 130–200 μm (see Figure 4). Grain size varies considerably in the upper layers (0–150 cm, 0–3 cal ka BP). Below this interval, the median grain size falls within a relatively narrow range between 135 and 165 μm. Based on lithology and grain size, we divided the section into seven zones. Thereinto, the relatively finer fraction covers more content in Zone A, Zone B, Zone D, and Zone F, while the relatively coarser fraction covers more content in Zone C, Zone E, and Zone G.

Figure 4.

Sediment grain size record of Sedimentary profile ZZH.

There are large variations in the salinity and C/N, which are illustrated in Figure 4. The C/N ranges from 6.6 to 36 with a mean of 14. There is higher value of C/N in Zone A, Zone F, and Zone H. The salinity varied from 0.14 to 9.80 mg/g with a mean of 2 mg/g. The salinity increased sharply above 200 cm (about 6ka) and the highest salinity value occurred at the Zone G.

Variations of n-alkanes

The n-alkane proxies show considerable variability (see Figure 5). All the proxies wildly fluctuate at Zone H (above 60 cm). Below 60 cm, the majority n-alkane proxies fluctuate weaker. The total concentration of n-alkanes (con_nak) exhibits higher values at Zone D. ACL, OEP, and C31/(C31 + C27) sharply increased at the depth of 200 cm (about 6 ka). Paq sharply decreased and Pwax sharply increased at depth of 150 cm (about 3 ka).

Figure 5.

Variations of n-alkanes proxies in Sedimentary profile ZZH.

Results of PCA

The PCA results based on grain size and n-alkanes proxies in Sedimentary profile ZZH reflect environmental change in groundwater recharge lake in this extremely arid zone. The first two principal components of the analysis account for 33% and 28% of the total variance, respectively. Grain size, TOC, and con-nak are high-load variables of the first principal component; CPI, OEP, and nC27/(nC27 + nC31) are high-load variables of the second principal component. The projection of proxies on a biplot diagram (see Figure 6) shows cluster distribution; these cluster groups correspond to the zones defined based on grain size (Figure 4) and n-alkanes (Figure 5).

Figure 6.

Biplot of PCA result for main climatic proxies data from Sedimentary profile ZZH.

Discussion

Environmental significance of the proxies

The lake from which Sedimentary profile ZZH was extracted is a groundwater recharge lake with almost no surface runoff. The sediment of Sedimentary profile ZZH consists mainly of authigenic lake sediments and aeolian sand of the basin. The source, transport capacity, and distance from the lake’s shore affect lake sediment grain size (Xiao et al., 2009). The sediment source and transport capacity are relatively stable in the Badain Jaran Desert. Therefore, the grain size is most sensitive to lake level; the finer particle sizes indicate farther distance from the lake shore and a higher lake water level (Li et al., 2018; Xiao et al., 2009) when the lake is relatively stable. While if the lake rapidly expands, the coarse aeolian sand around the lake would be transported into lakes and make coarser grain size. On the contrary, if the lake rapidly shrinks, the exposed lacustrine sediment, between migratory dune and lake, and suspended load driven by wind would be main source for lake depositional process; the grain size of source would be finer with the lake shrinking. The area covered by exposed lacustrine sediment would be enough broad to prevent from the aeolian sand which is proved by the remaining exposed lacustrine sediment along the lake margin. This would make the grain size of source particle finer and the sediment further finer. The relative more suspended load prompt more silt in the sediment, which matches with grain size of suspended load.

The organic matter and n-alkanes are mainly authigenic and transported from the basin around lake. The organic matter content and the total concentration of n-alkanes reflect lake biomass (Finlay and Kendall, 2007; Meyers and Lallier-Vergés, 1999). Because organic matter content and the total concentration of n-alkanes are influenced by microorganisms, which are controlled by environmental change, these values reflect effective humidity (Meyers and Lallier-Vergés, 1999). C/N is a common proxy for source of organic matter (Krishnamurthy et al., 1986; Meyers and Lallier-Vergés, 1999) and depositional environment (Wang et al., 2013a). In general, the mean C/N value of benthos is about 3.0, phytoplankton is about 6.0, alga is less than 10, aquatic plant range from 12 and 20, and terrestrial higher plant range from 14 to 23 (Müller and Mathesius, 1999). As for lake without runoff in the desert, organic matter mainly includes endogenous plants in the lake, such as alga and aquatic plant, and exogenous higher plants around the lake. When the lake expands, more phytoplankton and alga were produced and relatively less terrestrial plant was transported to bosom of the lake, showing lower value of C/N. This is different from the lake with runoff.

In n-alkanes, higher percentages of long-chain n-alkanes (>C22) indicate more abundant vascular plants (Eglinton and Hamilton, 1967). ACL (ACL = ∑(Ci*[Ci])/∑[Ci], [Ci]) is the concentration of n-alkanes with the carbon number Ci (Poynter and Eglinton, 1990; Poynter et al., 1989). As the temperature rises, the proportion of longer n-alkane chains increases (Eglinton and Hamilton, 1967). ACL values indicate the relative paleotemperature change to some extent (Poynter et al., 1989; Zhou et al., 2005). However, with changes in vegetation associated with significant climate change, ACL may mainly reflect changes in local plant species (Pu et al., 2011).

Phytoplankton and algae are predominantly characterized by short-carbon n-alkanes with a maximum carbon number of C17 (Blumer et al., 1971; Cranwell et al., 1987; Gelpi et al., 1970); submerged plants and phytoplankton are dominated by C21, C23, and C25 (Cranwell, 1984; Viso et al., 1993); emergent plants are dominated by n-alkanes with maximum carbon numbers of C27 and C29 (Cranwell, 1984); and terrestrial plants are dominated by n-alkanes with maximum carbon numbers of C27, C29, and C31 (Eglinton and Hamilton, 1967; Pancost et al., 2002). The CPI is the ratio of even to odd numbers of carbon atoms; CPI = (([C21]–[C31]) + ([C23] – [C33])odd/2 × ([C22] – [C32]) (Allan and Douglas, 1977). The CPI in fresh sediment represents the relative proportion of higher plants to algae and bacteria. A greater CPI value indicates a greater proportion of higher plants and a lower proportion of algae and bacteria (Bray and Evans, 1961; Marzi et al., 1993). The CPI is also affected by microbial degradation and thermal alteration processes, which are controlled by climatic factors during and after sedimentation. Lower CPI values in lake sediments represent higher microbial activity and warmer climatic conditions, and the reverse is also true (Andersson and Meyers, 2012; Jansen and Nierop, 2009; Kuder and Kruge, 1998; Xie et al., 2004; Zheng et al., 2007; Zhou et al., 2010b). The proportion of aquatic plant n-alkanes (Paq), Paq = ([C23] + [C25])/([C23] + [C25] + [C29] + [C31]) (Ficken et al., 2000), reflects the contribution of aquatic macrophytes to sedimentary organic matter. Submerged plants and phytoplankton are dominant when the Paq value is greater than 0.4 (Ficken et al., 2000). The proportion of land plant n-alkanes (Pwax), Pwax = ([C27] + [C29] + [C31])/([C23] + [C25] + [C27] + [C29] + [C31]) (Ficken et al., 2000), reflects the contribution of emergent and terrestrial plants (Pu et al., 2011; Zheng et al., 2007).

Woody plants are dominated by nC27 and n29, and herbs are dominated by nC31 (Cranwell, 1973; Meyers and Ishiwatari, 1993); therefore, changes in nC27/(nC27 + nC31) reflect changes in the types of plant in the area around the lake (Bai et al., 2009; Xie et al., 2003). Because most woody plants in the Badain Jaran Desert area are shrubs, higher nC27/(nC27 + nC31) values represent more abundant shrubs, which grow in drier and warmer conditions (Xie et al., 2003).

The high-load variables of the first principal component are mainly grain size and TOC content. The load of the first principal component is negative to TOC content and the percentage of very fine sand and positive to median grain size. Because TOC and grain size reflect the lake level and effective humidity, the first principal component indicates effective humidity. Higher scores of the first principal component correspond to lower effective humidity. The high-load variables of the second principal component are mainly CPI, OEP, and nC31/(nC27 + nC31). The load of the second principal component is positive to nC27/(nC27 + nC31) and negative to CPI and OEP. The second principal component is related to paleotemperature. A higher score for the second principal component corresponds to higher paleotemperature.

The phase change of paleotemperature

PCA2 high-load variables consist of the CPI, OEP, and nC27/(nC27 + nC31), which together represent temperature conditions. According to the PCA-2 scores, we reconstructed the paleotemperature change during Holocene in a lake of Badain Jaran Desert. As is shown in Figure 7e, the paleotemperature conditions from PCA-2 scores (see Figure 7e) demonstrated lower value before 10 ka, relatively higher value between 10 and 9.2 ka, highest value between 8.1 and 5.8, and lower value after 5.8 ka on a whole, while the value in 8.8–8.2 ka is relatively lower between 10 and 5.8 ka, the value in 4.5–4.0 ka and 0.6–0.1 ka is relatively lower between 5.8 and today, and the value in 2.4–1.6 ka is relatively higher between 5.8 and today.

Figure 7.

Comparison with regional and global records on temperature and effective humidity change: (a) PCA-1 scores for ZZH profile (this study), (b) EM2 of grain size for ZZH profile (this study), (c) redness record from Qinghai Lake, in the northeastern of Tibetan Plateau (Ji et al., 2005), (d) δ¹⁸O record from Dongge Cave (Dykoski et al., 2005; Wang et al., 2005), (e) PCA-2 scores for ZZH profile (this study), (f) δ¹³C record from Zhuyeze Lake, in southeastern of Badain Jaran Desert (Long et al., 2010), (g) woody pollen percentage, in Northern China (Li et al., 2004), (h) temperature record from Qinghai Lake, in the northeastern of Tibetan Plateau (Hou et al., 2016), (i) temperature change in 30–90°N during Holocene (Marcott et al., 2013), (j) Northeast Atlantic overflow recorded in silt-sized particles (10–64 m) for NEAP-15K (Bianchi and McCave, 1999), and (k) July insolation at 37°N latitude (Laskar et al., 2004).

In order to evaluate the geographic representation of our record, we compared with some published records from surrounding areas.

To the south of the desert, warm environment in early to middle Holocene was widely reported. The warmest period is about 8.5–6.7 ka from Zhuyeze δ¹³C records (Long et al., 2010, 2012) and 7.3–5.8 ka from Hongshui River δ¹⁸O records (Zhang et al., 2000); the second warmest period is about 4.5 ka from Hongshui River δ¹⁸O records (Zhang et al., 2000) and 4.4 ka from Zhuyeze δ¹³C records (Long et al., 2010), which is quite similar to our study, implying our paleotemperature records could apply to the study region. Interestingly, the pattern of desert temperature value is very similar to that in Qinghai Lake (see Figure 7h), in the northeastern of Tibetan Plateau (Hou et al., 2016). The highest temperature period occurred at about 5.8–8.1 ka, second highest temperature period occurred at about 10–9.3 ka, and relatively colder period occurred after 5.8 ka. This is consistent with the desert pattern. Even the abrupt period matches with the desert results, such as in about 8.6, 4.7, and 1.7 ka. However, the records in the central and south of Tibetan Plateau show warm in early Holocene and relatively cold in middle and late Holocene (Herzschuh et al., 2006; Thompson, 1997; Thompson et al., 2006), inconsistent with the desert pattern.

To the west of the desert, a warm early and middle Holocene phase was reported in Hexi Corridor and Xinjiang. The warm period implied by pollen record in eastern of Hexi Corridor (Mao et al., 2007) and Xinjiang (Jiang et al., 2013) lasted between 9.2 and 5.8 ka, while there was prominent difference between temperature in early Holocene and middle Holocene, which implied the summer insolation (Laskar et al., 2004) rather than the remaining ice covers in high latitudes area (Ruddiman, 2001; Tarasov et al., 2007) dominated temperature change in Xinjiang.

To the east of the desert, the Daihai Pollen record shows warmest period in 8.5–6.5 ka and second period in 10–8.8 ka, and cold period in late Holocene (Li et al., 2004), which is coincident with the desert.

To the north of the desert, the warmest period occurred during 6.6–4.5 ka in central Mongolia implied by the Lake Ugii Nuur record (Wang et al., 2011) and during 5.5–4 ka in northern Mongolian Plateau implied by pollen in Lake Gun Nuur (Feng et al., 2013), which were later than that in the desert. The delay of warmest period in Mongolian Plateau implied greater impact of ice sheets in higher latitude area.

Compared with global temperature records (Marcott et al., 2013), the desert temperature values exhibited overall similarity and regional characteristics, that is, the desert cooling trend is almost identical with the global temperature records after 4 ka, and desert warm period is similar to global temperature records before 6 ka. The desert temperature was overall synchronous with orbital factors (see Figure 7j) and ice sheet of high latitudes in millennial scale. In 11–10 ka, the remaining ice covers in high-latitude area (Ruddiman, 2001; Tarasov et al., 2007), the remaining local permafrost extent (Dong et al., 1997; Shan, 1996, 1997; Yu et al., 2017), and the lower summer insolation (Laskar et al., 2004) result in a lower temperature. In 10–8.8 ka, the highest summer insolation (Laskar et al., 2004), higher CO₂ concentration (Bush, 2005; Indermühle et al., 1999), and the stronger North Atlantic Deep Water (NADW) flow (Bar-Matthews et al., 1999) produced, preserved, and brought more heat for mid-latitude areas to keep a relatively higher temperature. In 8.8–8.1 ka, lowest CO₂ concentration (Bush, 2005; Indermühle et al., 1999) and weaker NADW flow (Bar-Matthews et al., 1999) contributed to trough period of temperature. In 8.1–5.8 ka, with the increasing CO₂ concentration (Bush, 2005; Indermühle et al., 1999) and final melting of ice covers, although the summer insolation slightly decreased (Laskar et al., 2004), the temperature reached the maximum value during Holocene. After 5.8 ka, with the largely decreasing in summer insolation (COHMAP Members, 1988; Laskar et al., 2004), the temperature gradually decreased. But higher temperature events in about 5–4.5 ka and 2.4–1.6 ka occurred in the desert, Zhuyeze Lake (Long et al., 2010), Daihai Lake (Li et al., 2004), Qinghai Lake (Hou et al., 2016), Erlongwan Maar Lake (Tao and Qi, 2012), and Hongyuan Peat (Hong et al., 2003), which were synchronous with the weaker NADW flow (Bar-Matthews et al., 1999). Hence, the abrupt climate event may be dominated by the Thermohaline Circulation.

Lake evolution and environment change based on multiple proxies

The study area is sensitive to climate change, especially change in humidity, because of its location in a typical arid monsoon–westerly transitional zone (Wang et al., 2016). We have assessed the lake evolution and changes in the effective humidity in the study area based on analysis of variations in grain size, TOC, TDS, total concentration of n-alkanes, and PCA-1 score.

During 11–10 ka, the deposition of a peat layer below the lacustrine layer and the relatively large amount of TOC and higher EM2 indicate higher effective humidity after Younger Dryas (YD) event. Higher C/N value (Meyers and Lallier-Vergés, 1999; Wagner, 2000) and C31/(C31 + C27) (Cranwell, 1973; Meyers and Ishiwatari, 1993) indicate more terrestrial herbs and less aquatic plant, which is in agreement with the lithologic character. The abrupt climate event after the YD is widely recorded around the desert, such as Huahai Basin (An et al., 2012), Zhuyeze Lake (Li et al., 2009; Long et al., 2012), Huangqihai Lake (Zhang et al., 2016), Juyanze Lake (Hartmann and Wunnemann, 2009), and Qinghai Lake (Ji et al., 2005), implying it is responding to climatic warming in regional scale.

Between 10 and 7.8 ka, lacustrine deposit indicates relative humid environment than that before 10 ka. However, the grain size, the EM2, con_nak, and PCA-1 scores indicate not exactly the same moisture conditions. The grain size and PCA-1 score indicate relatively more humid in 10–9 ka than that in 9–7.8 ka. However, more silt content, higher Paq, lower Pwax, lower salinity, and decreasing C/N indicate gradually wetting and more aquatic plant between 10 and 8.5 ka in the desert. One possible explanation for this apparent discrepancy is that the sedimentary process is influenced by both lake and swamp as the transition of peat layer and lacustrine facies layer in 10–9 ka. The coarse aeolian sand is blocked by more emerging plant and terrestrial plant and swamp in the lakeside in 10–9 ka which is supported by the higher C/N and lower salinity, respectively. After 8.5 ka, the coarser grain size, lower Paq, higher Pwax, and relatively higher salinity suggested relatively drier and less aquatic plant than before. However, Radix auricularia in the high lake remains of ZZH basin and Planorbids in high lake remains of BRZH basin, which is located in the basin of ZZH profile and 3 km far away from ZZH core, were dated to be 8594 ± 45 cal yr BP and 8438 ± 74 cal yr BP, respectively (Wang et al., 2016). This corroborated the transformation from humid in 10–8.5 ka to relatively dry in 8.5–7.8 ka. The relative dry period in 8.5–7.8 ka was not completely synchronous with the relative cold period in 8.8–8.2 ka, suggesting the local climate change was not the single factor to influence environment change in Sedimentary profile ZZH. The dry event was in phase with the records in Asian Monsoon margin areas, such as Zhuyeze Lake (Long et al., 2010), Qinghai Lake (Ji et al., 2005), Dali Lake (Liu et al., 2015), and Huahai Lake (Li et al., 2016); and Monsoon area, such as Dahu Peat (Zhong et al., 2010) and Erhai Lake (Dearing et al., 2008). But, the records in the western of Tibetan Plateau (Gasse et al., 1991, 1996) implied more humid environment antiphase with the records in the desert, northern, and central Tibetan Plateau. The westerlies area did not show apparent response to the cold event (Chen et al., 2016; Wang and Feng, 2013). The weaker Asian monsoon event (Wang et al., 2005) and low temperature event (Hou and Fang, 2012) triggered less precipitation (Hou et al., 2012; Liu et al., 1998; Wang et al., 2014) and less glacier meltwater (Li et al., 2012; Wischnewski et al., 2011) during 8.5–7.8 ka in northern and central Tibetan Plateau. The humid period of the desert in 10–8.8 ka was inconformity with the dry records in Xinjiang (An et al., 2012; Chen et al., 2016; Wang et al., 2013b), Northern China (Feng et al., 2006; Ran and Feng, 2013; Wen et al., 2010), North Mongolia Plateau (Tarasov et al., 2007; Wang et al., 2011; Wang and Feng, 2013), and conformity with South China (Dearing et al., 2008; Dykoski et al., 2005; Wang et al., 2005; Wei et al., 2010; Yamamoto et al., 2010; Zhao et al., 2011b) and Tibetan Plateau (Gasse et al., 1991, 1996; Herzschuh et al., 2006; Hou et al., 2012; Kramer et al., 2010). The desert lake may be strongly influenced by Asian Monsoon in some way.

Between 7.8 and 5.8 ka, the local maximum of con_nak is evidence for more productivity in situ and most humid environment, in agreement with local maximum EM2, PCA-1 scores, and finer grain size. The most humid period in the desert was in conformity with the records in the Westerlies area (An et al., 2012; Chen et al., 2008, 2016; Huang, 2006; Jiang et al., 2013; Liu et al., 2008; Ran and Feng, 2013; Wang et al., 2013) and northern Mongolia Plateau (Demske et al., 2005; Feng et al., 2013; Tarasov et al., 2007; Wang and Feng, 2013; Wang et al., 2011), a little later than the most humid period in Southern China (Dearing et al., 2008; Ma et al., 2008; Wang et al., 2007; Wei et al., 2010), almost in conformity with the most humid records in northern China (Chen et al., 2015; Jiang et al., 2006), monsoon margin area (Ji et al., 2005; Long et al., 2010; Liu et al., 2015; Xiao et al., 2004; Xu et al., 2010), and northern and Central of Tibetan Plateau (Gasse et al., 1996; Morrill et al., 2006; Wang et al., 2014; Wischnewski et al., 2011). The expansion of desert lake maybe response to strong Asian Monsoon, especially the strong East Asian Monsoon.

Between 5.8 and 5.5 ka, a sharp drop of PCA-1 scores, EM2, and con_nak, in agreement with sharp rise of grain size and salinity, suggests the quickly salinizing and shrinking of lake and more arid environment. Between 5.5 and 3.0 ka, the increasing C/N value, ranging from 8.6 to 19.5, suggests more emergent plant (Finlay and Kendall, 2007; Krishnamurthy et al., 1986; Müller and Voss, 1999) and the higher Paq, ranging from 0.3 to 0.5, also suggests more emergent plant (Cheung et al., 2015; Ficken et al., 2000). Hence, increasing Paq and C/N value suggests more emergent plant and less terrestrial plant or submerged plant, indicating a shallow lake environment between 5.5 and 2.5 ka. It is in agreement with the lithology of limnetic facies. Between 3.0 and 2.5 ka, sharply decreasing Paq and higher C/N value, ranging from 14 to 20, suggests less aquatic plant and more terrestrial plant. And sharp salinization suggested gradually shrinking between 5.8 and 2.5 ka. The particle sediment could only directly derive from emerged paleolake sediment around lake and suspended sediment from farther afield, as the lake gradually shrink, more fine particles from emerged paleolake sediment was transported to remaining lake and more suspended sediment (Lancaster, 1995; Tsoar and Pye, 1987) was directly transported to lake by wind action, which is supported by increasing silt content. The interpretation about grain size needs more study on grain size modern process. A drying trend is identified by the lithology and lower con_nak despite the discrepancy occurred between different proxies in 5.8–2.5 ka. The drying trend in the desert was verified by the absence of high lake level (Wang et al., 2016) and calcareous cement (Yang et al., 2011). The drying trend is inconformity with the warming records in Westerlies area (Chen et al., 2008, 2016; Ran and Feng, 2013) and northern Mongolia Plateau (Wang et al., 2011; Wang and Feng, 2013), conformity with the drying records in northern China, southern China, monsoon margin area, and northern and Central of Tibetan Plateau during 5.8–2.5 ka, that is, the drying trend was in phase with the weakening Monsoon (Dong et al., 2010; Wang et al., 2005, 2010).

Between 2.5 and 0.8 ka, the lithology of aeolian sand and multiproxy, including coarser grain size, minimum EM2, PCA-1 scores, and highest C/N, document the arid environment and less aquatic plant. Between 0.8 and 0.1 ka, the lithology of lake deposit and multiproxy, including finer grain size, maximum EM2, PCA-1 scores, and lower C/N, suggest humid environment. After 0.1 ka, the aeolian sand, higher C/N, lower EM2, and PCA-1 score indicate the desert environment same as today.

The probable forcing mechanisms of Holocene hydroclimate change

In this study, millennial-scale Holocene environmental changes in lake basin from Badain Jaran Desert imply that it was arid to humid in after YD (11–10 ka), relative humid in the early Holocene (10–7.8 ka), most humid in the middle Holocene (7.8–5.8 ka), dry in the mid to late Holocene (5.8–0.8 ka), and a little humid in 0.8–0.1 ka. The moisture change was synchronous with the temperature change in millennial scale before 2 ka, but the phases of higher temperature and higher humidity do not always coincide in century scale, such as the relative humid, proved by higher EM2 and higher lake level (Wang et al., 2016), and cold environment in about 8.3 ka; the relative humid, proved by higher PCA-1 score, and cold environment in about 7.4 ka; dry and warm environment in 1.8–1.4 ka; and humid and cold environment in 0.8–0.1ka.

To clarify the similarities and differences with the lake evolution pattern with runoff, we compare our results with other proxy from Asian Monsoon margin surrounding the desert. The most humid period of the desert was basically conformity with the records from southeastern of Monsoon margin, such as at 7.9–4.5 ka in Daihai Lake (Xiao et al., 2004), 7.7–6 ka in Dali Lake (Liu et al., 2015), 7.4–4.8 ka in Zhuyeze Lake (Li et al., 2009; Long et al., 2010), and 8.1–5.9 ka in Qinghai Lake (Ji et al., 2005), relatively later than the records from northwestern of Monsoon margin, such as 8.8–5.5 ka in Huahai Lake (Li et al., 2016; Wang et al., 2012) and 8.5–7.5 ka in Hongshui River and different with the records in westerlies core area (Chen et al., 2016; Jiang et al., 2013; Wang et al., 2013b) and northwestern of the desert (Hartmann and Wunnemann, 2009, COHMAP Members, 2003). The relative humid period in early Holocene was different with the dry records from eastern of Monsoon margin area, such as Daihai Lake (Xiao et al., 2004), Dali Lake (Liu et al., 2015) and conform with the records from the northeastern of Tibetan Plateau, such as Qinghai Lake (Ji et al., 2005). The environment change pattern was synchronous with the northeastern of Tibetan Plateau.

Despite the synchronism with the Monsoon margin area records, the lake evolution should not be directly dominated by the Monsoon intensity. In fact, the precipitation could barely infiltrate to be groundwater or gather to be runoff to influence the lake evolution in the desert. Groundwater accounts for the main water source of modern lakes in the Badain Jaran Desert; annual precipitation of 87 mm accounts for 10% of the total supply. The reconstructed annual precipitation in the middle Holocene was 300 mm, 37% higher than present, in the Shiyang River Basin in the southeast of the desert (Guo et al., 2000; Zhao et al., 2007) and 340 mm, 10% higher than the present, in the northeast of the Tibetan plateau (Wang et al., 2014). The calcareous root tubes of the Badain Jaran Desert imply that annual precipitation was approximately 200 mm (Li et al., 2015). In the Zhunzhahan Jaran Basin, where Sedimentary profile ZZH was collected, the paleolake inferred from the lake level maintained an area of 1.83 km² and a depth of 26 m. Annual precipitation is hypothesized to have been 200 mm (2.2 times present annual precipitation) with an annual evaporation capacity of 1450 mm, as much as that of a modern salt-water lake. Wang et al. (2016) calculated the annual groundwater recharge to be 2.3 × 10⁶ m³, 6.3 times the annual precipitation recharge. Given the higher temperature and lower salinity in the early to middle Holocene, the evaporation capacity was likely to be greater than the modern annual evaporation capacity of 1450 mm. Therefore, the annual groundwater recharge was likely greater than 2.3 × 10⁶ m³, and the ratio of groundwater recharge to precipitation recharge was more than 6.3, approximately equal to that in the modern. Consequently, there is reason to infer that change in groundwater recharge and evaporation capacity dominated the evolution of lakes in the Badain Jaran Desert during the Holocene. The groundwater dominated the lake evolution during Holocene (Dong et al., 2016; Ma et al., 2014; Wang et al., 2016) and most of groundwater comes from southeastern of the desert, that is, the Northeastern of Tibetan Plateau (Lu et al., 2010; Wu et al., 2014). Hence, the synchronism with the Monsoon margin area records actually implied correlation between the relative humid in the Monsoon margin area and the groundwater recharged amount to the desert.

To further probe into the possibility and mechanism of groundwater from the Northeastern Tibetan Plateau recharging to the desert lake, we compared our records with the amount of precipitation and glacier meltwater, which are key factors for groundwater recharged amount (Cheng and Jin, 2013; Immerzeel et al., 2010; Zhang et al., 2005; Zhou et al., 2010a). The reconstructed precipitation in northeastern of Tibetan Plateau, such as Donggi Cona Lake (Wang et al., 2014), Kuhai Lake (Wischnewski et al., 2011), and Koucha Lake (Herzschuh et al., 2009), showed highest value during 11–9 ka and second highest value during 7.5–5 ka. Although the precipitation covered highest value (Herzschuh et al., 2009; Wang et al., 2014; Wischnewski et al., 2011), the glaciers had retreated significantly after YD certified by a series of moraine deposit and loessic silts capping a latero-frontal moraine (Hu et al., 2016; Owen et al., 2005, 2003a, 2003b). The only plausible explanation for the apparent discrepancy is the huge amount of glacier meltwater under high temperature. The more glacier meltwater and more precipitation in northeastern of Tibetan Plateau made it possible to provide more groundwater to the desert in early Holocene. Another large-scale glacial recession with high precipitation occurred after 8.2 ka event, which is certified by ¹⁴C age (Jiao et al., 2000; Shijie and Keqin, 1990; Zhao et al., 2011a) and cosmogenic radionuclide age (Owen et al., 2003a, 2003b; Rose et al., 1998) of moraine deposit. The higher amount of glacier meltwater and precipitation during 11–9 ka and 7.5–6 ka, as key factors to provide more groundwater recharged amount, mainly match with the humid period in the desert lake. This further determined the importance of precipitation and glacier meltwater in northeastern of Tibetan Plateau for the lake evolution in the hinterland of Badain Jaran Desert.

Conclusion

Based on the grain size, salinity, and n-alkane proxies of Sedimentary profile ZZH from the hinterland of the Badain Jaran Desert, we reconstructed changes in Holocene effective humidity and paleotemperature in a groundwater recharge lake of arid region. The records imply warm and humid environment in early to mid Holocene, and cold and dry in mid to late Holocene. The temperature was broadly synchronous with northeastern of Tibetan Plateau and Northern Hemisphere temperature, driven mainly by orbital factors and Thermohaline Circulation. The effective humidity was consistent with northeastern of Tibetan Plateau and monsoon margin area records. However, the desert lake evolution was mainly dominated by exogenous groundwater rather than local precipitation or runoff. Considering groundwater comes from southeastern of the desert, effective humidity may indicate recharge intensity of groundwater from northeastern of Tibetan Plateau.

Footnotes

Acknowledgements

We are grateful to Shi Xuezhao and Wang Yueguang on experimental work and Prof. Li Yu, anonymous reviewer, for very helpful comments.

Funding

This project was supported by the he National Natural Science Foundation of China (Nos. 41530745 and 41871021), the Fundamental Research Funds for the Central Universities (No. lzujbky-2018-it82) and the Fundamental Research Funds for Chinese Academy of Geological Sciences (No. YYWF201627).

References

Allan

Douglas

(1977) Variations in the content and distribution of n-alkanes in a series of carboniferous vitrinites and sporinites of bituminous rank. Geochimica et Cosmochimica Acta 41: 1223–1230.

Zhao

et al . (2012) A high-resolution record of Holocene environmental and climatic changes from Lake Balikun (Xinjiang, China): Implications for central Asia. The Holocene 22: 43–52.

Porter

Kutzbach

et al . (2000) Asynchronous Holocene optimum of the East Asian monsoon. Quaternary Science Reviews 19: 743–762.

Andersson

Meyers

(2012) Effect of climate change on delivery and degradation of lipid biomarkers in a Holocene peat sequence in the Eastern European Russian Arctic. Organic Geochemistry 53: 63–72.

Bai

Fang

Nie

et al . (2009) A preliminary reconstruction of the paleoecological and paleoclimatic history of the Chinese Loess Plateau from the application of biomarkers. Palaeogeography, Palaeoclimatology, Palaeoecology 271: 161–169.

Bar-Matthews

Ayalon

Kaufman

et al . (1999) The Eastern Mediterranean paleoclimate as a reflection of regional events: Soreq cave, Israel. Earth and Planetary Science Letters 166: 85–95.

Bianchi

McCave

(1999) Holocene periodicity in North Atlantic climate and deep-ocean flow south of Iceland. Nature 397: 515–517.

Birks

HJB

Birks

HJB

(1998) D.G. Frey and E.S. Deevey Review 1: Numerical tools in palaeolimnology – Progress, potentialities, and problems. Journal of Paleolimnology 20: 307–332.

Blaauw

(2010) Methods and code for ‘classical’ age-modelling of radiocarbon sequences. Quaternary Geochronology 5: 512–518.

10.

Blumer

Guillard

RRL

Chase

(1971) Hydrocarbons of marine phytoplankton. Marine Biology 8: 183–189.

11.

Bray

Evans

(1961) Distribution of n-paraffins as a clue to recognition of source beds. Geochimica et Cosmochimica Acta 22: 2–15.

12.

Bush

(2005) CO₂/H₂O and orbitally driven climate variability over central Asia through the Holocene. Quaternary International 136: 15–23.

13.

Chen

Jia

Chen

et al . (2016) A persistent Holocene wetting trend in arid central Asia, with wettest conditions in the late Holocene, revealed by multi-proxy analyses of loess-paleosol sequences in Xinjiang, China. Quaternary Science Reviews 146: 134–146.

14.

Chen

et al . (2015) East Asian summer monsoon precipitation variability since the last deglaciation. Scientific Reports 5: 111186.

15.

Chen

Yang

et al . (2008) Holocene moisture evolution in arid central Asia and its out-of-phase relationship with Asian monsoon history. Quaternary Science Reviews 27: 351–364.

16.

Cheng

Jin

(2013) Groundwater in the permafrost regions on the Qinghai-Tibet Plateau and it changes. Hydrogeology and Engineering Geology 40: 1–11 (in Chinese with English abstract).

17.

Cheung

M-C

Zong

Wang

et al . (2015) δ13Corg and n-alkane evidence for changing wetland conditions during a stable mid-late Holocene climate in the central Tibetan Plateau. Palaeogeography, Palaeoclimatology, Palaeoecology 438: 203–212.

18.

COHMAP Members (1988) Climatic changes of the last 18,000 years: Observations and model simulations. Science 241: 1043–1052.

19.

Cranwell

(1973) Branched-chain and cyclopropanoid acids in a recent sediment. Chemical Geology 11: 307–313.

20.

Cranwell

(1984) Alkyl esters, mid chain ketones and fatty acids in late glacial and postglacial lacustrine sediments. Organic Geochemistry 6: 115–124.

21.

Cranwell

Eglinton

Robinson

(1987) Lipids of aquatic organisms as potential contributors to lacustrine sediments – II. Organic Geochemistry 11: 513–527.

22.

Dearing

Jones

Shen

et al . (2008) Using multiple archives to understand past and present climate–human–environment interactions: The lake Erhai catchment, Yunnan Province, China. Journal of Paleolimnology 40: 3–31.

23.

Demske

Heumann

Granoszewski

et al . (2005) Late glacial and Holocene vegetation and regional climate variability evidenced in high-resolution pollen records from Lake Baikal. Global and Planetary Change 46: 255–279.

24.

Dong

Naiang

Chen

et al . (2016) New observational and experimental evidence for the recharge mechanism of the lake group in the Alxa Desert, north-central China. Journal of Arid Environments 124: 48–61.

25.

Dong

Chen

Wang

et al . (1997) The evolution of deserts with climatic changes in China since 150 ka B.P. Science in China(Series D:Earth Sciences) 40: 370–382.

26.

Dong

Wang

Cheng

et al . (2010) A high-resolution stalagmite record of the Holocene East Asian monsoon from Mt Shennongjia, central China. The Holocene 20: 257–264.

27.

Dykoski

Edwards

Cheng

et al . (2005) A high-resolution, absolute-dated Holocene and deglacial Asian monsoon record from Dongge Cave, China. Earth and Planetary Science Letters 233: 71–86.

28.

Eglinton

Hamilton

(1967) Leaf epicuticular waxes. Science 156: 1322–1335.

29.

Fan

Zhang

et al . (2015) History and mechanisms for the expansion of the Badain Jaran Desert, northern China, since 20 ka: Geological and luminescence chronological evidence. The Holocene 26: 532–548.

30.

Feng

Zhang

et al . (2013) Holocene climate variations retrieved from Gun Nuur lake-sediment core in the northern Mongolian Plateau. The Holocene 23: 1721–1730.

31.

Feng

Tang

Wang

et al . (2006) Holocene vegetation variations and the associated environmental changes in the western part of the Chinese Loess Plateau. Palaeogeography, Palaeoclimatology, Palaeoecology 241: 440–456.

32.

Ficken

Swain

et al . (2000) An n-alkane proxy for the sedimentary input of submerged/floating freshwater aquatic macrophytes. Organic Geochemistry 31: 745–749.

33.

Finlay

Kendall

(2007) Stable isotope tracing of temporal and spatial variability in organic matter sources to freshwater ecosystems. In: Michener

Lajtha

(eds) Stable Isotopes in Ecology and Environmental Science. London: Springer, pp. 283–333.

34.

Gasse

Arnold

Fontes

et al . (1991) A 13,000-year climate record from western Tibet. Nature 353: 742–745.

35.

Gasse

Fontes

Van Campo

et al . (1996) Holocene environmental changes in Bangong Co basin (Western Tibet). Part 4: Discussion and conclusions. Palaeogeography, Palaeoclimatology, Palaeoecology 120: 79–92.

36.

Gelpi

Schneider

Mann

et al . (1970) Hydrocarbons of geochemical significance in microscopic algae. Phytochemistry 9: 603–612.

37.

Guo

Chen

Shi

(2000) The application of GIS and water and energy budget to the study on the water rebuilding of paleo-lake – A case in Shiyang River drainage. Scientia Geographica Sinica 5: 422–426 (in Chinese with English abstract).

38.

Hartmann

Wunnemann

(2009) Hydrological changes and Holocene climate variations in NW China, inferred from lake sediments of Juyanze palaeolake by factor analyses. Quaternary International 194: 28–44.

39.

Herzschuh

Kramer

Mischke

et al . (2009) Quantitative climate and vegetation trends since the late glacial on the northeastern Tibetan Plateau deduced from Koucha Lake pollen spectra. Quaternary Research 71: 162–171.

40.

Herzschuh

Winter

Wünnemann

et al . (2006) A general cooling trend on the central Tibetan Plateau throughout the Holocene recorded by the Lake Zigetang pollen spectra. Quaternary International 154–155: 113–121.

41.

Hong

Lin

et al . (2003) Correlation between Indian Ocean summer monsoon and North Atlantic climate during the holocene. Earth and Planetary Science Letters 215: 317–318.

42.

Hou

Fang

(2012) Characteristics analysis and synthetical reconstruction of regional temperature series of the Holocene in China. Journal of Palaeogeography 14: 243–252 (in Chinese with English abstract).

43.

Hou

Chongyi

Xiao

(2012) Synthetical reconstruction of the precipitation series of the Qinghai-Tibet Plateau during the Holocene. Progress in Geography 31: 1117–1123 (in Chinese with English abstract).

44.

Hou

Huang

Zhao

et al . (2016) Large Holocene summer temperature oscillations and impact on the peopling of the northeastern Tibetan Plateau. Geophysical Research Letters 43: 1323–1330.

45.

Zhang

et al . (2016) Late Quaternary glacial advances in the eastern Qilianshan, north-eastern Tibet, as inferred from luminescence dating of fluvioglacial sediments. Journal of Quaternary Science 31: 587–597.

46.

Huang

Guan

et al . (2016) Accelerated dryland expansion under climate change. Nature Climate Change 6: 166–171.

47.

Huang

(2006) Holocene Climate Variability of Arid Central Asia Documented by Bosten Lake, Xinjiang, China. Lanzhou, China: Lanzhou University (in Chinese with English Abstract).

48.

Immerzeel

Van Beek

Bierkens

(2010) Climate change will affect the Asian water towers. Science 328: 1382–1385.

49.

Indermühle

Stocker

Joos

et al . (1999) Holocene carbon-cycle dynamics based on CO₂ trapped in ice at Taylor Dome, Antarctica. Nature 398: 121–126.

50.

Jansen

Nierop

KGJ

(2009) Methyl ketones in high altitude Ecuadorian Andosols confirm excellent conservation of plant-specific n-alkane patterns. Organic Geochemistry 40: 61–69.

51.

Shen

Balsam

et al . (2005) Asian monsoon oscillations in the northeastern Qinghai–Tibet Plateau since the late glacial as interpreted from visible reflectance of Qinghai Lake sediments. Earth and Planetary Science Letters 233: 61–70.

52.

Jiang

et al . (2013) Holocene vegetational and climatic variation in westerly-dominated areas of Central Asia inferred from the Sayram Lake in northern Xinjiang, China. Science China Earth Sciences 56: 339–353.

53.

Jiang

Guo

Sun

et al . (2006) Reconstruction of climate and vegetation changes of Lake Bayanchagan (Inner Mongolia): Holocene variability of the East Asian monsoon. Quaternary Research 65: 411–420.

54.

Jiao

Yao

Shi-Jie

(2000) Evolution of Glaciers and Environment in the West Kunlun Mountains During the Past 32 ka. Journal of Glaciolgy and Geocryology 22: 250–256.

55.

Kramer

Herzschuh

Mischke

et al . (2010) Holocene treeline shifts and monsoon variability in the Hengduan Mountains (southeastern Tibetan Plateau), implications from palynological investigations. Palaeogeography, Palaeoclimatology, Palaeoecology 286: 23–41.

56.

Krishnamurthy

Bhattacharya

Kusumgar

(1986) Palaeoclimatic changes deduced from 13C/12C and C/N ratios of Karewa lake sediments, India. Nature 323: 150–152.

57.

Kuder

Kruge

(1998) Preservation of biomolecules in sub-fossil plants from raised peat bogs – A potential paleoenvironmental proxy. Organic Geochemistry 29: 1355–1368.

58.

Lancaster

(1995) The Geomorphology of Desert Dunes. New York: Routledge.

59.

Laskar

Robutel

Joutel

et al . (2004) A long-term numerical solution for the insolation quantities of the Earth. Astronomy & Astrophysics 428: 261–285.

60.

Chen

Jiang

et al . (2012) Geological records for Holocene climatic and environmental changes derived from glacial, periglacial and lake sediments on Qinghai-Tibetan Plateau. Quaternary Science 32: 151–157.

61.

Jie

et al . (2004) Vegetation history and climatic variations during the last 14 ka BP inferred from a pollen record at Daihai Lake, north-central China. Review of Palaeobotany and Palynology 132: 195–205.

62.

Naiang

Cheng

et al . (2009) Holocene environmental change in the marginal area of the Asian monsoon: A record from Zhuye Lake, NW China. Boreas 38: 349–361.

63.

Chen

Shen

et al . (2013) Analysis of changing pan evaporation in the arid region of Northwest China. Water Resources Research 49: 2205–2212.

64.

Naiang

Cheng

et al . (2015) Formation and environmental significance of late Quaternary calcareous root tubes in the deserts of the Alashan Plateau, northwest China. Quaternary International 372: 167–174.

65.

Naiang

Cheng

et al . (2016) Early–middle Holocene hydroclimate changes in the Asian monsoon margin of northwest China inferred from Huahai terminal lake records. Journal of Paleolimnology 55: 289–302.

66.

Wei

Dong

et al . (2018) The paleoenvironmental significance of spatial distributions of grain size in groundwater-recharged lakes: A case study in the hinterland of the Badain Jaran Desert, northwest China. Earth Surface Processes & Landforms 43: 363–372.

67.

Liu

K-B

Yao

Thompson

(1998) A pollen record of Holocene climatic changes from the Dunde ice cap, Qinghai-Tibetan Plateau. Geology 26: 135–138.

68.

Liu

Deng

Xiao

et al . (2015) Insolation driven biomagnetic response to the Holocene Warm Period in semi-arid East Asia. Scientific Reports 5: 8001.

69.

Liu

Herzschuh

Shen

et al . (2008) Holocene environmental and climatic changes inferred from Wulungu Lake in northern Xinjiang, China. Quaternary Research 70: 412–425.

70.

Long

Lai

Fuchs

et al . (2012) Timing of Late Quaternary palaeolake evolution in Tengger Desert of northern China and its possible forcing mechanisms. Global and Planetary Change 92–93: 119–129.

71.

Long

Lai

Wang

et al . (2010) Holocene climate variations from Zhuyeze terminal lake records in East Asian monsoon margin in arid northern China. Quaternary Research 74: 46–56.

72.

Long

Shen

Chen

et al . (2017) Holocene moisture variations over the arid central Asia revealed by a comprehensive sand-dune record from the central Tian Shan, NW China. Quaternary Science Reviews 174: 13–32.

73.

Naiang

et al . (2010) Spatial distribution of lakes hydro-chemical types in Badain Jaran Desert. Journal of Lake Sciences 22: 774–782 (in Chinese with English abstract).

74.

Zhu

Zheng

et al . (2008) High-resolution geochemistry records of climate changes since late-glacial from Dajiuhu peat in Shennongjia Mountains, Central China. Chinese Science Bulletin 53: 28–41.

75.

Wang

Zhao

et al . (2014) Observation of mega-dune evaporation after various rain events in the hinterland of Badain Jaran Desert, China. Chinese Science Bulletin 59: 162–170.

76.

Mao

Zhao

et al . (2007) Pollen assemblages and environment evolution in Shule River Alluvial Fan Oasis of Gansu. Acta Geoscientica Sinica 28: 528–534 (in Chinese with English abstract).

77.

Marcott

Shakun

Clark

et al . (2013) A reconstruction of regional and global temperature for the past 11,300 years. Science 339: 1198–1201.

78.

Marzi

Torkelson

Olson

(1993) A revised carbon preference index. Organic Geochemistry 20: 1303–1306.

79.

Mason

Zhou

et al . (2009) Dune mobility and aridity at the desert margin of northern China at a time of peak monsoon strength. Geology 37: 947–950.

80.

Meyers

Ishiwatari

(1993) Lacustrine organic geochemistry – An overview of indicators of organic matter sources and diagenesis in lake sediments. Organic Geochemistry 20: 867–900.

81.

Meyers

Lallier-Vergés

(1999) Lacustrine sedimentary organic matter records of late quaternary paleoclimates. Journal of Paleolimnology 21: 345–372.

82.

Mischke

(2003) Hydrologic and climatic implications of a multi-disciplinary study of the mid to late Holocene Lake Eastern Juyanze. Chinese Science Bulletin 48: 1411.

83.

Morrill

Overpeck

Cole

et al . (2006) Holocene variations in the Asian monsoon inferred from the geochemistry of lake sediments in central Tibet. Quaternary Research 65: 232–243.

84.

Müller

Mathesius

(1999) The palaeoenvironments of coastal lagoons in the southern Baltic Sea, I. The application of sedimentary C org /N ratios as source indicators of organic matter. Palaeogeography Palaeoclimatology Palaeoecology 145: 1–16.

85.

Müller

Voss

(1999) The palaeoenvironments of coastal lagoons in the southern Baltic Sea, II. δ 13 C and δ 15 N ratios of organic matter – Sources and sediments. Palaeogeography Palaeoclimatology Palaeoecology 145: 17–32.

86.

Ning

Wang

et al . (2013) Spatial characteristics of grain size and its environmental implication in the Badain Jaran Desert. Journal of Desert Research 3: 7.

87.

Ning

Wang

et al . (2015) Characteristics of seasonal frozen ground in Badain Jaran desert. Journal of Glacilogy and Geocryology 5: 1209–1216 (in Chinese with English abstract).

88.

Owen

Finkel

Barnard

et al . (2005) Climatic and topographic controls on the style and timing of Late Quaternary glaciation throughout Tibet and the Himalaya defined by 10Be cosmogenic radionuclide surface exposure dating. Quaternary Science Reviews 24: 1391–1411.

89.

Owen

Finkel

Haizhou

et al . (2003a) Timing and style of Late Quaternary glaciation in northeastern Tibet. Geological Society of America Bulletin 115: 1356.

90.

Owen

Spencer

et al . (2003b) Timing of Late Quaternary glaciation along the southwestern slopes of the Qilian Shan, Tibet. Boreas 32: 281–291.

91.

Pancost

Baas

van Geel

et al . (2002) Biomarkers as proxies for plant inputs to peats: An example from a sub-boreal ombrotrophic bog. Organic Geochemistry 33: 675–690.

92.

Poynter

Eglinton

(1990) Molecular composition of three sediments from hole 717C: The Bengal fan. Proceedings of the Ocean Drilling Program: Scientific Results 116: 155–161.

93.

Poynter

Farrimond

Robinson

et al . (1989) Aeolian-derived higher plant lipids in the marine sedimentary record: Links with palaeoclimate. In: Leinen

Sarnthein

(eds) Paleoclimatology and Paleometeorology: Modern and Past Patterns of Global Atmospheric Transport. London: Springer, pp. 435–462.

94.

Zhang

Wang

et al . (2011) Climatic and environmental implications from n-alkanes in glacially eroded lake sediments in Tibetan Plateau: An example from Ximen Co. Chinese Science Bulletin 56: 1503–1510.

95.

Ran

Feng

(2013) Holocene moisture variations across China and driving mechanisms: Asynthesis of climatic records. Quaternary International 313–314: 179–193.

96.

Reimer

Bard

Bayliss

et al . (2009) IntCal13 and Marine13 Radiocarbon Age Calibration Curves 0-50,000 Years cal. BP. Radiocarbon 55: 1869–1887.

97.

Rose

Derbyshire

Guo

et al . (1998) Glaciation of the eastern Qilian Shan, northwest China. Journal of Quaternary Science 13: 143–152.

98.

Ruddiman

(2001) Earth’s Climate: Past and Future. London: Macmillan.

99.

Shan

(1996) The first discovery and significance of glacial ice wedge in the northeast margin of Tengger Desert. Chinese Science Bulletin 41: 160–163 (in Chinese with English abstract).

100.

Shan

(1997) The process and environmental changes of the ice margin in the purple mud lake area in the west of Helan Mountain. Geological Review 43: 428–433 (in Chinese with abstract).

101.

Shijie

Keqin

(1990) Glacier variations on the south slope of West Kunlun Mountains since 30000 years. Journal of Glaciology and Geocryology 12: 311–318.

102.

Tao

(2012) High-resolution climate evolution derived from the sediment records of Erlongwan Maar Lake since 14 ka BP. Chinese Science Bulletin 57: 3610–3616.

103.

Tarasov

Bezrukova

Karabanov

et al . (2007) Vegetation and climate dynamics during the Holocene and Eemian interglacials derived from Lake Baikal pollen records. Palaeogeography, Palaeoclimatology, Palaeoecology 252: 440–457.

104.

Thompson

(1997) Tropical climate instability: The last glacial cycle from a Qinghai-Tibetan ice core. Science 276: 1821–1825.

105.

Thompson

Yao

Davis

et al . (2006) Holocene climate variability archived in the Puruogangri ice cap on the central Tibetan Plateau. Annals of Glaciology 43: 61–69.

106.

Tsoar

Pye

(1987) Dust transport and the question of desert loess formation. Sedimentology 34: 139–153.

107.

Viso

Pesando

Bernard

et al . (1993) Lipid components of the mediterranean seagrass Posidonia Oceanica. Phytochemistry 34: 381–387.

108.

Wagner

Melles

Hahne

et al . (2000) Holocene climate history of Geographical Society O, East Greenland - evidence from lake sediments. Palaeogeography Palaeoclimatology Palaeoecology 160: 45–68.

109.

Wang

Sun

Chen

et al . (2015) Formation and evolution of the Badain Jaran Desert, North China, as revealed by a drill core from the desert centre and by geological survey. Palaeogeography, Palaeoclimatology, Palaeoecology 426: 139–158.

110.

Wang

et al . (2012) Younger Dryas event recorded by the mirabilite deposition in Huahai lake, Hexi Corridor, NW China. Quaternary International 250: 93–99.

111.

Wang

et al . (2013a) Millennial-scale environmental changes in the Asian monsoon margin during the Holocene, implicated by the lake evolution of Huahai Lake in the Hexi Corridor of northwest China. Quaternary International 313: 100–109.

112.

Wang

Ning

et al . (2016) Holocene high lake-levels and pan-lake period on Badain Jaran Desert. Science China Earth Sciences 59: 1633–1641.

113.

Wang

Liu

et al . (2007) The early Holocene optimum inferred from a high resolution pollen record of Huguangyan Maar Lake in southern China. Chinese Science Bulletin 20: 2829–2836 (in Chinese with English abstract).

114.

Wang

Feng

(2013) Holocene moisture evolution across the Mongolian Plateau and its surrounding areas: A synthesis of climatic records. Earth-Science Reviews 122: 38–57.

115.

Wang

Feng

Ran

et al . (2013b) Holocene climate and vegetation changes inferred from pollen records of Lake Aibi, northern Xinjiang, China: A potential contribution to understanding of Holocene climate pattern in East-central Asia. Quaternary International 311: 54–62.

116.

Wang

Feng

et al . (2011) A prolonged dry mid-Holocene climate revealed by pollen and diatom records from Lake Ugii Nuur in central Mongolia. Quaternary International 229: 74–83.

117.

Wang

Cheng

Edwards

et al . (2005) The Holocene Asian monsoon: Links to solar changes and North Atlantic climate. Science 308: 854–857.

118.

Wang

Herzschuh

Shumilovskikh

et al . (2014) Quantitative reconstruction of precipitation changes on the NE Tibetan Plateau since the last glacial maximum – Extending the concept of pollen source area to pollen-based climate reconstructions from large lakes. Climate of the Past 10: 21–39.

119.

Wang

Liu

Herzschuh

(2010) Asynchronous evolution of the Indian and East Asian Summer Monsoon indicated by Holocene moisture patterns in monsoonal central Asia. Earth-Science Reviews 103: 135–153.

120.

Wei

Xue

Cao

et al . (2010) Bulk organic carbon isotopic record of lacustrine sediments in Dahu Swamp, eastern Nanling Mountains in South China: Implication for catchment environmental and climatic changes in the last 16,000years. Journal of Asian Earth Sciences 38: 162–169.

121.

Weltje

(1997) End-member modeling of compositional data: Numerical-statistical algorithms for solving the explicit mixing problem. Mathematical Geology 29: 503–549.

122.

Weltje

Prins

(2003) Muddled or mixed? Inferring palaeoclimate from size distributions of deep-sea clastics. Sedimentary Geology 162: 39–62.

123.

Wen

Xiao

Chang

et al . (2010) Holocene precipitation and temperature variations in the East Asian monsoonal margin from pollen data from Hulun Lake in northeastern Inner Mongolia, China. Boreas 39: 262–272.

124.

Wischnewski

Mischke

Wang

et al . (2011) Reconstructing climate variability on the northeastern Tibetan Plateau since the last Lateglacial – A multi-proxy, dual-site approach comparing terrestrial and aquatic signals. Quaternary Science Reviews 30: 82–97.

125.

Wang

Zhao

et al . (2014) Hydrochemical characteristics and recharge sources of Lake Nuoertu in the Badain Jaran Desert. Chinese Science Bulletin 59: 886–895.

126.

Xiao

Chang

et al . (2009) Partitioning of the grain-size components of Dali Lake core sediments: Evidence for lake-level changes during the Holocene. Journal of Paleolimnology 42: 249–260.

127.

Xiao

Nakamura

et al . (2004) Holocene vegetation variation in the Daihai Lake region of north-central China: A direct indication of the Asian monsoon climatic history. Quaternary Science Reviews 23: 1669–1679.

128.

Xie

Chen

Wang

et al . (2003) Lipid distributions in loess-paleosol sequences from northwest China. Organic Geochemistry 34: 1071–1079.

129.

Xie

Nott

Avsejs

et al . (2004) Molecular and isotopic stratigraphy in an ombrotrophic mire for paleoclimate reconstruction. Geochimica et Cosmochimica Acta 68: 2849–2862.

130.

Xiao

et al . (2010) Pollen-based quantitative reconstruction of Holocene climate changes in the Daihai Lake area, Inner Mongolia, China. Journal of Climate 23: 2856–2868.

131.

Yamamoto

Kawamura

Seki

et al . (2010) Paleoenvironmental significance of compound-specific δ13C variations in n-alkanes in the Hongyuan peat sequence from southwest China over the last 13 ka. Organic Geochemistry 41: 491–497.

132.

Yang

Dong

et al . (2010) Recharge to the inter-dune lakes and Holocene climatic changes in the Badain Jaran Desert, western China. Quaternary Research 73: 10–19.

133.

Yang

Scuderi

Paillou

et al . (2011) Quaternary environmental changes in the drylands of China – A critical review. Quaternary Science Reviews 30: 3219–3233.

134.

Wang

Long

et al . (2017) The age of sand wedges and its environmental significance at the southeast edge of Badain Jaran Desert (in Chinese). Chinese Science Bulletin 62: 9.

135.

Yunker

Macdonald

Whitehouse

(1994) Phase associations and lipid distributions in the seasonally ice-covered Arctic estuary of the Mackenzie Shelf. Organic Geochemistry 22: 651–669.

136.

Zhang

Nie

Liu

et al . (2005) Characteristic and variation of groundwater recharge resources in the middle reaches of Heihe River basin. Advances in Water Science 16: 673–678.

137.

Zhang

Wunnemann

et al . (2000) A Holocene climatic record from arid northwestern China. Palaeogeography, Palaeoclimatology, Palaeoecology 162: 389–401.

138.

Zhang

Tsukamoto

Jia

et al . (2016) Lake level reconstruction of Huangqihai Lake in northern China since MIS 3 based on pulsed optically stimulated luminescence dating. Journal of Quaternary Science 31: 225–238.

139.

Zhang

(2005a) Groundwater Resource of China (Gansu Vol). Peking, China: SinoMaps Press.

140.

Zhang

(2005b) Groundwater Resource of China (Xinjiang Vol). Peking, China: SinoMaps Press.

141.

Zhang

Wang

et al . (2013) Remote sensing on spatial changes of lake area in Badain Jaran Desert hinterland during 1973–2010. Journal of Lake Sciences 4: 514–520 (in Chinese with English abstract).

142.

Zhao

Shi

Wang

(2011a) Comparison between quaternary glaciations in China and the Marine Oxygen Isotope Stage (MIS): An improved schema. Acta Geographica Sinica 66: 867–884 (in Chinese with English abstract).

143.

Zhao

Wang

(2007) Water balance of Shiyang River drainage during 6700–5800 yr BP. Journal of Arid Land Resources and Environment 6: 84–91 (in Chinese with English abstract).

144.

Zhao

(2011b) Holocene vegetation and climate histories in the eastern Tibetan Plateau: Controls by insolation-driven temperature or monsoon-derived precipitation changes? Quaternary Science Reviews 30: 1173–1184.

145.

Zheng

Zhou

Meyers

et al . (2007) Lipid biomarkers in the Zoige-Hongyuan peat deposit: Indicators of Holocene climate changes in West China. Organic Geochemistry 38: 1927–1940.

146.

Zhong

Xue

Zheng

et al . (2010) Climatic changes since the last deglaciation inferred from a lacustrine sedimentary sequence in the eastern Nanling Mountains, south China. Journal of Quaternary Science 25: 975–984.

147.

Zhou

Kang

Gao

et al . (2010a) Response of Zhadang Glacier runoff in Nam Co Basin, Tibet, to changes in air temperature and precipitation form. Chinese Science Bulletin 55: 2103–2110.

148.

Zhou

Zheng

Meyers

et al . (2010b) Postglacial climate-change record in biomarker lipid compositions of the Hani peat sequence, Northeastern China. Earth and Planetary Science Letters 294: 37–46.

149.

Zhou

Xie

Meyers

et al . (2005) Reconstruction of late glacial and Holocene climate evolution in southern China from geolipids and pollen in the Dingnan peat sequence. Organic Geochemistry 36: 1272–1284.

150.

Zhu

Wang

Chen

et al . (2010) Study on the boundary and the area of Badain Jaran Desert based on remote sensing imagery. Progress in Geography 29: 1087–1094 (in Chinese with English abstract).