Paleohydrological changes over the last 4000 years in the middle and lower reaches of the Yangtze River: Evidence from particle size and n -alkanes from Longgan Lake

Abstract

We have reconstructed the history of late-Holocene paleohydrological changes in the middle and lower reaches of the Yangtze River using grain size and n-alkane data from a sediment core retrieved from Longgan Lake. We employ changes in the grain size distribution to reflect the water level in the floodplain lake, with a higher percentage of the finer fraction indicating higher water level and vice versa. The n-alkane molecular distribution, average chain length (ACL), and P_aq ratio (C₂₃+C₂₅)/(C₂₃+C₂₅+C₂₉+C₃₁) are used to reflect mainly vegetation composition that is also sensitive to water depth. Our results reveal that the lake water level was relatively low and gradually increased from 4 to 2.7 ka. The period from 2.7 to 1.2 ka exhibited the highest late-Holocene lake water level in this region. The water level then decreased toward the present. This paleohydrological reconstruction agrees with existing paleoclimate reconstructions of the middle and lower reaches of the Yangtze River, confirming that the intensity of Asian monsoon rains is an important factor in affecting paleohydrological changes in this region.

Keywords

central China grain sizes lake sediment paleohydrological change Yangtze River

Introduction

The middle and lower reaches of the Yangtze Valley are strongly influenced by the Asian summer monsoon rains. Droughts and floods have occurred frequently over the last few decades, bringing severe impacts on the ecosystem and local society. It is therefore imperative to understand the evolution of hydrological conditions in this densely populated region of China to improve planning for future such events. Paleoclimate reconstructions can provide a useful long-term perspective on the regional paleohydrologic history. Previous studies in this region are mainly based on stalagmites (Cai et al., 2010; Hu et al., 2008), peat deposits (Huang et al., 2013; Xie et al., 2013), and historical documents (Central Institute of Meteorology Science (CIMS), 1981). Shallow floodplain lakes are widely distributed in the middle and lower reaches of the Yangtze Valley, including the five biggest freshwater lakes in China (Wang and Dou, 1998). Sediment cores retrieved from these lakes can provide invaluable archives of paleohydrological changes since the late-Holocene. However, high-resolution paleohydrological reconstructions spanning more than 1000 years based on lake sediment cores remain rare in this region (Li et al., 2016; Wang et al., 2008; Yang et al., 2002; Yao et al., 2015).

As a part of the development of paleolimnology, a variety of proxies have been developed for paleoenvironmental reconstructions using lake sediments (Meyers, 2003). Among these proxies, the grain size of lake sediment provides a physical index that reflects the transport dynamics of clastic materials (Colman et al., 2007). In addition, n-alkanes are abundant in lake sediments, with contributions from terrestrial and aquatic macrophytes (Bush and McInerney, 2013; Eglinton and Hamilton, 1967). n-Alkane ratios – including the average chain length (ACL), the carbon preference index (CPI), and the P_aq ratio – are widely used in paleolimnology (Castañeda and Schouten, 2011; Ficken et al., 2000; Meyers, 2003). Because both the grain size and n-alkane distribution are sensitive to the hydrodynamics in a lake catchment, combining these ratios to reconstruct paleohydrological changes from lake sediment cores can be especially informative.

For this study, we collected a sediment core from Longgan Lake, located in the floodplain of the middle reach of the Yangtze River, and analyzed the grain size and n-alkane distributions in segments of this core. The aim of this study is to assess the potential of grain sizes and n-alkane ratios for the reconstruction of paleohydrological changes over the last 4 ka in the middle and lower reaches of the Yangtze River.

Materials and methods

Sampling

Longgan Lake (29°52′~30°05′N, 115°19′~116°17′E; 12 m a.s.l.) is located on the boundary between Suqian County, Anhui Province, and Huangmei County, Hubei Province (Figure 1). This lake has an area of 316 km² and was formed after the collapse of Penglize paleolake (Wang and Dou, 1998). The local climate is dominated by the East Asian Summer Monsoon, with an annual mean rainfall of 1291 mm and an annual mean temperature of 16.6°C (Wang and Dou, 1998). The average lake water level is 3.8 m, but it has strong seasonal variations, with higher levels in summer and lower levels in fall and winter. Precipitation and surface runoff from small rivers such as the Liangting, Erlang, Huangmei, Jingzhu and Meichuan constitute its major water sources (Wang and Dou, 1998). A survey of aquatic vegetation in the lake in 1993 reported macrophytes belonging to 50 species comprised 22 emergent or mud-flat macrophytes, 15 submerged macrophytes, and 13 floating macrophytes (Zhang et al., 1996).

Figure 1.

(a) Map of the Longgan Lake core in the middle and lower reaches of the Yangtze River with the wind directions of the moist summer Indian and East Asian monsoons and the dry winter Asian monsoon shown. (b) Stateline map of Longgan Lake and surrounding region (downloaded from http://map.qq.com/?center=116.062134%2C29.874378&l=10).

A 3-m sediment core (29°54.756′N, 116°01.118′ E; abbreviated as LGH) was collected using a sampler set (Eijkelkamp, Netherlands) from Longgan Lake on 2 June 2011. At that time, a severe drought existed in the middle and lower reaches of the Yangtze River, and Longgan Lake was almost dry. The coring site was 1 km away from the west lakeshore. Its macrophyte community was populated mainly by Zizania aquatica and minor Trapa incisa. The core was sliced at 2-cm intervals in the field. The upper 14 cm consisted of coarse sediment particles with yellow color. The 14- to 50-cm layer was mainly gray clay. From 50 to 74 cm, the clay color varied between black and red. The 74- to 300-cm layer was gray clay. This study was limited to the upper 190 cm of the core.

Grain size analysis

Grain size analysis was conducted in Taishan College, using a Mastersizer 2000 particle size analyzer (Malvern, England). Air-dried samples of about 0.2 g were first pretreated by adding 30% H₂O₂ and 10% HCl to remove organic matter and carbonate, then mixed with 10% (NaPO₃)₆, and finally ultrasonically dispersed for 10 min before measurement. The analytical error was <1%.

Lipid extraction and analysis

The freeze-dried samples were first passed through a 60 mesh sieve. About 15 g of each sample was extracted with an accelerated solvent extraction system (ASE-100, Dionex) using CH₂Cl₂/methanol (9:1, volume/volume) as the solvent. An internal standard of 5-β(H)-cholane (Chiron, Norway) was added before extraction. The combined extracts were fractionated into aliphatic and polar fractions using silica gel column chromatography with hexane and methanol as eluting solvents, respectively.

The aliphatic fraction containing n-alkanes was quantified using a Shimadzu GC-2010 gas chromatograph (GC) equipped with a flame ionization detector (FID) and a DB-5 column (30 m × 0.25 mm i.d., 0.25 µm film thickness). The sample was injected in splitless mode with the injector temperature at 300°C. The oven temperature was initiated at 70°C and held for 1 min, then ramped to 210°C at a rate of 10°C min⁻¹, and finally ramped to 300°C at a rate of 3°C min⁻¹ and kept isothermal for 25 min. An n-alkane mixture standard containing n-C₁₆ to n-C₃₀ was tested before the sample running to identify the n-alkane distributions of the samples. The absolute abundances were calculated by comparison of peak areas with the internal standard and adjusted with the relevant response factors.

Dating analysis

The chronology of the LGH sediment core is based on ¹⁴C measurements of the bulk organic matter of five core segments conducted in the Accelerator Mass Spectrometry Center of Arizona University (Table 1). The calendar ages of the core segments were calculated using the clam age-depth model (Blaauw, 2010; Figure 2).

Table 1.

Results of ¹⁴C AMS dating from Longgan Lake, central China.

Lab code	Depth (cm)	Materials	¹⁴C date (yr BP)	Calibrated age (cal. yr BP)
AA95214	51	Sediment	898 ± 37	737–912
AA95215	75	Sediment	1514 ± 37	1331–1424
AA96506	99	Sediment	2560 ± 38	2684–2756
AA95216	151	Sediment	3263 ± 41	3396–3574
AA96505	195	Sediment	3688 ± 39	3908–4098

Calibrated age ranges were obtained using the clam age-depth model with 2σ error intervals (Blaauw, 2010).

Figure 2.

Depth-age profile in the LGH sediment core generated using the Clam package (Blaauw, 2010).

Results and discussion

Chronology

The ¹⁴C AMS age data correlate closely with the depth (r² = 0.93; Figure 2), suggesting a near constant sedimentation rate in the upper 1.9 m. In this study, we ignore the reservoir effect for the LGH sediment core, which has also been ignored in other studies of lake sediments in eastern China (Chen et al., 2009; Peng et al., 2005; Wang et al., 2008). A reasonable basis for ignoring the reservoir effect is the predominance of terrestrial input in these lake sediments. As an example, detailed investigation of organic matter in Chaohu Lake, adjacent to Longgan Lake, has established that allochthonous contributions dominate the organic matter in the surface sediments (Wu et al., 2016). According to the chronology, the upper 190 cm covered the last 4 ka BP.

Grain size distribution and its paleohydrological significance

The variations of the contents of six size fractions (<2, 2–5, 5–16, 16–32, 32–63, and >63 µm) are shown in Figure 3. The size fraction of <2 µm varies synchronously with that of 2–5 µm, whereas the fractions of 5–16, 16–32, and 32–63 µm show similar variations (Figure 3). Consequently, we combine the fractions of <2 and 2–5 to represent the fine clay particles, the fractions of 5–63 µm as silt, and the fraction of >63 µm as sand, consistent with previous studies (Conroy et al., 2008; Peng et al., 2005). The clay and silt fractions predominate throughout the entire core, with the sand fraction almost always lower than 5%. From 4 to 2.7 ka, the percentage of clay fraction generally increases, and it reaches maximal values during 2.7–1.2 ka. From 1.2 ka onwards, the clay percentage decreases and is less than 30% from 0.4 ka to the present (Figure 4).

Figure 3.

Down-core variations of six size fractions (<2, 2–5, 5–16, 16–32, 32–63, and >63 µm) in the LGH sediment core.

Figure 4.

Profiles of the clay, silt, and sand percentages, and the medium diameter (Md) in the LGH sediment core. The vertical dashed lines indicate the four climatic stages discussed in the text.

The grain size distribution in lake sediments is a robust proxy for lacustrine paleohydrological conditions owing to its close relation with the transport dynamics and sedimentary environment (Conroy et al., 2008; Liu et al., 2016; Peng et al., 2005; Yan et al., 2011). Surface runoff (Sheng et al., 2015) or wind (Vandenberghe, 2013) carries clastic materials into lakes, their transport being dependent on the flow energy. A comprehensive study by Chen et al. (2003) proposed that the grain size fraction correlated with the shrinkage or expansion of lakes on centennial to millennial timescales, with the coarser fraction preferentially deposited in drier intervals. We argue that this interpretation works on the late-Holocene sediment in the LGH core. Longgan Lake is shallow and relatively flat-bottomed (Wang and Dou, 1998). Both paleolimnological studies and literature records (Shen, 2013) reveal that this lake evolved closely with surrounding lakes and the easily submerged nearby middle and lower Yangtze Plain. Thus, in the following discussions, an increase in the silt particle relative to clay is interpreted to record a decrease in water level and a dry condition. In contrast, an increase in the clay fraction implies a condition with a high water level and wet climate.

n-Alkane distributions and their paleohydrological significance

In the LGH core, the concentrations of individual n-alkanes (C₁₅–C₃₅) range from 0.02 to 1895.5 ng/g dry weight. The homologs with chain length > 20 show a strong odd-over-even predominance (Figure 5). In the upper 30 cm, n-C₂₅ and n-C₂₃ alkanes are the dominant components, supporting a major contribution from aquatic submerged/floating plants (Ficken et al., 2000; Liu et al., 2015). Although Longgan Lake has suffered moderately from human disturbance in the last few decades and its nutrient condition is characterized as meso-eutrophic, the lake still supports tremendous populations of aquatic macrophytes under normal water level conditions (Yang et al., 2006). In the lower part of the sediment core (30–180 cm), the proportions of n-C₂₉ and n-C₃₁ alkanes increase, implying a larger contribution from emersed plants growing in the lakeshore zone and from terrestrial higher plants growing in the catchment.

Figure 5.

Histograms of long chain n-alkanes in selected samples from the LGH sediment core.

Three widely used alkane proxies are utilized here to describe variations in the n-alkane distributions and to evaluate their associated environmental changes (Figure 6). The ACL values vary between 25.3 and 29.4, with a mean value of 28.3. The CPI values fluctuate between 1.5 and 5.7 and average 2.6. The P_aq (a proxy to reflect the proportion of aquatic macrophytes relative to that of the emersed aquatic and terrestrial plants) has values ranging from 0.2 to 0.8 and averaging 0.4:

Figure 6.

Down-core variations of n-alkane average chain length (ACL), the carbon preference index (CPI), and the aquatic macrophyte versus aquatic macrophyte and terrestrial plant ratio (P_aq). The vertical dashed lines indicate the four climatic stages discussed in the text.

Similar to the grain size data, the alkane results support division of the whole core into four stages (Figure 6). The first stage (4–2.7 ka) is characterized by a general increase in CPI and P_aq values and a decrease in ACL values. In the second stage (2.7–1.2 ka), the proportion of short chain n-alkanes increases to near 0.1 from the low values in the first stage. The third stage is featured by an abrupt increase in P_aq values and decrease in ACL values. The CPI values reach their highest in the whole core during the fourth stage, whereas the P_aq ratio reaches a higher value at 0.3 ka and then decreases toward the present.

In shallow lakes like Longgan Lake, vegetation communities in the lake and growing in the catchment are the major factors influencing the n-alkane distribution patterns preserved in lake sediments (Meyers, 2003). Algae and photosynthetic bacteria contribute mainly short chain n-alkanes (Cranwell, 1987), while aquatic macrophytes synthesize a high proportion of medium chain C₂₁, C₂₃, and C₂₅ n-alkanes (Duan et al., 2014; Ficken et al., 2000; Liu et al., 2015). These plants are typically limited to water depths less than 4 m (e.g. Sifeddine et al., 2011). The long chain C₂₇, C₂₉, and C₃₁ n-alkanes in lake sediments originate mainly from terrestrial vascular and some emersed macrophytes (Castañeda and Schouten, 2011; Rielley et al., 1991). From these source generalizations, we interpret that the changes in the vegetation community are the primary factor to control the variations of P_aq and ACL in the LGH sediment core. During stages with large contributions from submerged and/or floating plants, the P_aq values will increase, corresponding with the decrease in ACL values.

In the LGH sediment core, most CPI values are lower than 3.0 (Figure 6). During the 4–2.7 ka time interval, the mean CPI value is even as low as 1.9. Both terrestrial and aquatic macrophytes, the predominant contributors of n-alkanes (>C₂₀) in land settings (Bush and McInerney, 2013; Chikaraishi and Naraoka, 2003; Eglinton and Hamilton, 1967; Ficken et al., 2000), have higher CPI values (normally >5; Bush and McInerney, 2013). It is therefore not reasonable that the low CPI values in the LGH sediment core result from the direct input from fresh plant materials. Instead, lower CPI values may result either from preferential microbial degradation of odd-number homologs relative to even ones during transport of n-alkanes from the catchment to the lake and their deposition in the sediment or from contributions of even-number n-alkanes produced by microbiota (Freeman and Pancost, 2014). Because the concentrations of the even-number n-alkanes are little changed throughout the core whereas those of the odd-number homologs vary substantially (Figure 5), we favor the preferential degradation hypothesis.

Multi-proxy reconstruction of late-Holocene paleohydrological changes in Longgan Lake

It is noteworthy that the grain size fractions correlate closely with the alkane ratios (Table 2). Two examples are the negative correlation between medium diameter of grain size (Md) and ACL (r = −0.62, p < 0.001) and the positive correlation between Md and CPI (r = 0.63, p < 0.001). These correlations imply that the two independent indicators respond similarly to environmental factors. Combining the grain size data and n-alkane distributions, we divide the paleohydrogical changes in Longgan Lake over the last 4 ka into the four stages: 4–2.7, 2.7–1.2, 1.2–0.4, and 0.4–0 ka (Figure 7).

Table 2.

Correlation analysis between grain sizes and n-alkane ratios.

	ACL	CPI	P _aq	Md	Clay %	Silt %
CPI	−0.56**
P _aq	−0.95**	0.56**
Md	−0.62**	0.63**	0.71**
Clay	0.48**	−0.45**	−0.60**	−0.82**
Silt	−0.48**	0.43**	0.60**	0.91**	−0.99**
Sand	−0.01	0.20	0.03	0.24*	−0.18	0.06

ACL: average chain length; CPI: carbon preference index.

p < 0.01; *p < 0.05.

Figure 7.

Comparisons of the P_aq, CPI, clay percentage, and clay/silt ratio in the LGH sediment core with the rainfall amount in the Southwest China (Hu et al., 2008). The vertical dashed lines indicate the four climatic stages discussed in the text.

The first stage (4–2.7 ka) is characterized by a gradual increase in the lake water level, indicated by a monotonic increase in the clay percentage and quite low CPI values. In this stage, silt fraction content is higher than 50% and a little higher than the percentage of clay fraction. The clay percentage in the LGH sediment core is distinct before and after 3.2 ka in the interval of 4–2.7 ka. The period of 4–3.2 ka has a mean clay percentage of 38.7%, while the mean clay percentage increases to 48.3% from 3.2 to 2.7 ka (Figure 7). This difference in clay percentage suggests the water level was higher in the period of 3.2–2.7 ka than during 4–3.2 ka.

During the second stage (2.7–1.2 ka), the lake water level probably reached its highest elevation in the late-Holocene, as supported by the highest clay percentage and the relatively higher CPI values. In this second stage, the content of clay can be nearly double that of silt. From 1.2 ka onwards, the lake water level decreases in two steps: 1.2–0.4 and 0.4–0 ka. From 0.4 to 0 ka, the lake shrank dramatically, as recorded by the accumulation of the coarse fraction. Meanwhile, the shallower lake condition favored the establishment of aquatic plants, as indicated by the relatively higher P_aq values from 1.4 to 0 ka. These plants grew close to the coring location, and the shorter transport time of their long chain n-alkanes resulted in less degradation and relatively higher CPI values.

The East Asian summer monsoon circulation controls the hydrological condition in the middle and lower reaches of the Yangtze River (Zhang et al., 2007). In this study, the paleohydrological conditions reconstructed from the LGH sediment core are consistent with previous paleoclimate records in the middle and lower reaches of the Yangtze River. Changes in the aerobic bacteria derived hopanoid flux imply that the drought that prevailed from 4 to 3 ka in the middle reaches of the Yangtze River was followed by relatively wet conditions from 3 to 1 ka (Xie et al., 2013). Rainfall amount reconstructed by the difference of oxygen isotope compositions between two stalagmites from the same water vapor transport route (Hu et al., 2008) showed a drought from 3.6 to 3.2 ka, an interval with relatively higher precipitation from 3.2 to 1.4 ka, and then a relatively dry interval from 1.4 to 0 ka (Figure 7). Similarly, a lacustrine sediment sequence retrieved from the Jianghan Plain recorded a change from lacustrine conditions to a swamp at 3.7 ka, a dry interval from 2.3 to 2.1 ka, and a wet interval between 2.1 and 1.2 ka, with the 1.2 ka to recent period recorded as drought (Li et al., 2014). Also, a pollen record from Chao Lake indicates a drought period from 4.8 to 2.2 ka, relatively wet conditions from 2.2 to 1.0 ka, and dry conditions from 1.0 to 0.2 ka (Wang et al., 2008). In addition, leaf wax δD data from nearby Poyang Lake imply a dry interval from 1.4 to 0.8 ka and 0.5 to 0.2 ka during the past 1400 years (Yao et al., 2015).

We note that the LGH record has some inconsistencies with previous records. The greatest one is the interval of 4–2.7 ka (Figure 7). Other studies concluded that this interval could be separated into two sub-stages: a drier period from 4 to 3.2 ka and a subsequent wet interval (Hu et al., 2008; Xie et al., 2013). In fact, the clay percentage in the LGH sediment core is different before and after 3.2 ka in the interval of 4–2.7 ka (Figure 7). It is odd that the alkane ratios are very constant from 4 to 2.7 ka, mismatching the shifts of grain sizes. This inconsistency between alkane ratios and grain size fractions may result from heavy degradation of organic matter, as suggested by the fairly low CPI values, or from dilution of the sediment by allochthonous mineral matter.

Human influences, especially in the last 1 ka, could have complicated the sediment record. Pollen results from Longgan Lake imply enhanced human activity since 0.8 ka (Tong et al., 1997). In the LGH sediment core, the Md reaches the highest values in the period from 0.4 to 0 ka, with the sand fraction > 10% (Figure 4). Farming activities and forest clearance in the catchment can accelerate soil erosion, and thus higher amounts of coarse material will be transported to the lake. The alkane ratios can also be affected by human activity such as deforestation. Along the whole sequence, the P_aq values show a general increasing trend, but they have relatively high values only at 1.2–0 ka, a period characterized by low lake water level (Figure 7). This trend probably also resulted from deforestation. Before 1.2 ka, forests in the catchment were likely less disturbed, and consequently land plant derived n-alkanes could be imported to lake sediments in large amounts. However, after 1.2 ka, forest clearance eliminated this source and made vegetation in and around the lake to become the dominant n-alkane contributors. This scenario can also help explain the higher CPI values from 0.4 to 0 ka; the shorter transport distance would to some extent reduce microbial degradation and leave relatively higher CPI values (Figure 7).

Conclusion

Paleohydrological changes in Longgan Lake over the last 4 ka were reconstructed using grain size and n-alkane data. Our results reveal that the lake water level was initially relatively low and gradually increased from 4 ka toward 2.7 ka. The period from 2.7 to 1.2 ka was characterized by the highest lake water level in the late-Holocene in this region. The water level then decreased toward the present. This paleohydrological record is in line with those from previous paleoclimate studies of the middle and lower reaches of the Yangtze River, suggesting that the Asian monsoon intensity is an important factor in affecting paleohydrological changes in this region.

Footnotes

Acknowledgements

Jinxiang Zhang, Canfa Wang and Fengfeng Zheng are thanked for helping in the sample collection. Comments from an anonymous reviewer greatly helped us to improve the quality of the manuscript.

Funding

This work was supported by NSFC (41472308), the fundamental research funds for the central universities (CUG150618), and Distinguished Young Scholars of Hubei Province (2016CFA051).

References

Blaauw

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

Bush

McInerney

(2013) Leaf wax n-alkane distributions in and across modern plants: Implications for paleoecology and chemotaxonomy. Geochimica et Cosmochimica Acta 117: 161–179.

Cai

Tan

Cheng

, et al. (2010) The variation of summer monsoon precipitation in central China since the last deglaciation. Earth and Planetary Science Letters 291: 21–31.

Castañeda

Schouten

(2011) A review of molecular organic proxies for examining modern and ancient lacustrine environments. Quaternary Science Reviews 30: 2851–2891.

Central Institute of Meteorology Science (CIMS) (1981) Yearly Charts of Dryness/Wetness in China for the Last 500-Year Period. Beijing: Atlas Press of China (in Chinese).

Chen

Wang

Dai

(2009) Holocene vegetation history with implications of human impact in the Lake Chaohu area, Anhui Province, East China. Vegetation History and Archaeobotany 18: 137–146.

Chen

Wan

Zhang

, et al. (2003) The record of lake sediments in different scales-taking grain size for example. Science in China Series D: Earth Sciences 33: 563–568.

Chikaraishi

Naraoka

(2003) Compound-specific δD-δ¹³C analyses of n-alkanes extracted from terrestrial and aquatic plants. Phytochemistry 63: 361–371.

Colman

, et al. (2007) Late Cenozoic climate changes in China’s western interior: A review of research on Lake Qinghai and comparison with other records. Quaternary Science Reviews 26: 2281–2300.

10.

Conroy

Overpeck

Cole

, et al. (2008) Holocene changes in eastern tropical Pacific climate inferred from a Galápagos lake sediment record. Quaternary Science Reviews 27: 1166–1180.

11.

Cranwell

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

12.

Duan

Cao

, et al. (2014) Hydrogen isotope ratios of individual n-alkanes in plants from Gannan Gahai Lake (China) and surrounding area. Organic Geochemistry 77: 96–105.

13.

Eglinton

Hamilton

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

14.

Ficken

Swain

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

15.

Freeman

Pancost

(2014) Biomarkers for terrestrial plants and climate. In: Turekian

Holland

(eds) Treatise on Geochemistry. 2nd Edition. Oxford: Elsevier, pp. 395–416.

16.

Henderson

Huang

, et al. (2008) Quantification of Holocene Asian monsoon rainfall from spatially separated cave records. Earth and Planetary Science Letters 266: 221–232.

17.

Huang

Meyers

Jia

, et al. (2013) Paleotemperature variability in central China during the last 13 ka recorded by a novel microbial lipid proxy in the Dajiuhu peat deposit. The Holocene 23: 1123–1129.

18.

Zhu

, et al. (2014) Environmental humidity changes inferred from multi-indicators in the Jianghan Plain, Central China during the last 12700 years. Quaternary International 349: 68–78.

19.

Yang

Yao

, et al. (2016) Precipitation changes records in the sedimentary total organic carbon isotopes from Lake Poyang in the Middle and lower Yangtze River, southern China over the last 1600 years. Quaternary International 425: 292–300.

20.

Liu

Yang

Wang

, et al. (2015) Carbon isotope composition of long chain leaf wax n-alkanes in lake sediments: A dual indicator of paleoenvironment in the Qinghai-Tibet Plateau. Organic Geochemistry 83–84: 190–201.

21.

Liu

Vandenberghe

, et al. (2016) Grain size of Lake Qinghai sediments: Implications for riverine input and Holocene monsoon variability. Palaeogeography, Palaeoclimatology, Palaeoecology 449: 41–51.

22.

Meyers

(2003) Applications of organic geochemistry to paleolimnological reconstructions: A summary of examples from the Laurentian Great Lakes. Organic Geochemistry 34: 261–289.

23.

Peng

Xiao

Nakamura

, et al. (2005) Holocene East Asian monsoonal precipitation pattern revealed by grain-size distribution of core sediments of Daihai Lake in Inner Mongolia of north-central China. Earth and Planetary Science Letters 233: 467–479.

24.

Rielley

Collier

Jones

, et al. (1991) The biogeochemistry of Ellesmere Lake, U.K. – I: Source correlation of leaf wax inputs to the sedimentary lipid record. Organic Geochemistry 17: 901–912.

25.

Shen

(2013) Spatiotemporal variations of Chinese lakes and their driving mechanisms since the Last Glacial Maximum: A review and synthesis of lacustrine sediment archives. Chinese Science Bulletin 58: 17–31.

26.

Sheng

, et al. (2015) Late-Holocene Indian summer monsoon precipitation history at Lake Lugu, northwestern Yunnan Province, southwestern China. Palaeogeography, Palaeoclimatology, Palaeoecology 438: 24–33.

27.

Sifeddine

Meyers

Campello Cordeiro

, et al. (2011) Delivery and deposition of organic matter in surface sediments of Lagoa do Caco (Brazil). Journal of Paleolimnology 45: 385–396.

28.

Tong

Shi

, et al. (1997) Vegetation and climatic quantitative reconstruction of Longgan Lake since the past 3000 years. Marine Geology and Quaternary Geology 53–61 (in Chinese with English abstract).

29.

Vandenberghe

(2013) Grain size of fine-grained windblown sediment: A powerful proxy for process identification. Earth-Science Reviews 121: 18–30.

30.

Wang

Dou

(1998) Lakes in China. Beijing: Science Press (in Chinese).

31.

Wang

Zhang

, et al. (2008) Environmental changes during early-middle Holocene from the sediment record of the Chao Lake, Anhui Province. Chinese Science Bulletin 53: 153–160 (in Chinese with English abstract).

32.

Gao

Chen

, et al. (2016) Response of organic carbon burial to trophic level changes in a shallow eutrophic lake in SE China. Journal of Environmental Sciences 46: 220–228.

33.

Xie

Evershed

Huang

, et al. (2013) Concordant monsoon-driven postglacial hydrological changes in peat and stalagmite records and their impacts on prehistoric cultures in central China. Geology 41: 827–830.

34.

Yan

Sun

Oppo

, et al. (2011) South China Sea hydrological changes and Pacific Walker Circulation variations over the last millennium. Nature Communications 2: 293.

35.

Yang

Shen

Dong

, et al. (2006) Historical trophic evolutions and their ecological responses from shallow lakes in the middle and lower reaches of the Yangtze River: Case studies on Longgan Lake and Taibai Lake. Science in China Series D: Earth Sciences 49: 51–61.

36.

Yang

Zhu

, et al. (2002) Environmental changes revealed by lacustrine sediment from Longgan Lake since Last Glacial Maximum. Journal of Lake Sciences 14(2): 106–110 (in Chinese with English abstract).

37.

Yao

Yang

Liu

, et al. (2015) Hydrological changes of the past 1400 years recorded in δD of sedimentary n-alkanes from Poyang Lake, southeastern China. The Holocene 25: 1068–1075.

38.

Zhang

Jiang

, et al. (2007) Possible influence of ENSO on annual maximum streamflow of the Yangtze River, China. Journal of Hydrology 333: 265–274.

39.

Zhang

Dou

Jiang

(1996) Aquatic vegetation in Longgan Lake. Journal of Lake Sciences 161–168 (in Chinese with English abstract).