Temporal and spatial evolution of Holocene vegetation and lake hydrological status,China

Abstract

Studying past climate change has important scientific significance in exploring long-term climatic variability, assessing the nature and human’s contribution to climate warming, and predicting future climate change. Here, we presented a data synthesis of pollen records, organic geochemical proxies, as well as lake-level records, along with numerical climate classification and lake-level simulations to show temporal and spatial evolution of past vegetation and hydrologic change during the Holocene in China. Both geological data and numerical modeling indicated that the evolution of vegetation shows out-of-phase relationships with lake hydrological status in various regions. China as a whole experienced optimal vegetation cover in the mid-Holocene. However, areas dominated by the Asian monsoon and the Westerlies did not uniformly show an early Holocene hydrological optimum. For the Westerlies-controlled regions, optimal conditions prevailed during the mid-Holocene. These apparent contradictions were closely related to various driving factors in different geographical regions. Precipitation has been recognized as a major influence on lake hydrological status of arid and semi-arid in northeast China, whereas in other regions of China, the combined effect of evaporation and precipitation was more influential. Likewise, the limiting factor of vegetation conditions was asynchronous in China, for example, moisture in northwestern China, temperature in the Qinghai-Tibet Plateau, as well as moisture and temperature in the monsoon marginal zones and monsoonal regions. The results provide insights into the underlying climate-forcing mechanisms, and demonstrate a new perspective on Quaternary environment change research.

Keywords

lake hydrological status lake-level records lake sediments organic geochemical proxies pollen records vegetation

Introduction

Many studies have been conducted regarding Chinese vegetation and lake hydrological status during the Holocene in particular regions using numerous proxy data from different sources (An et al., 2000; Chen et al., 2008; Herzschuh, 2006; Li et al., 2009b; Liu et al., 2005, 2013; Long et al., 2010; Parmesan, 2006; Qian et al., 2009; Yu at al., 2000). A recent synthesis of Chinese lakes showed that the highest stands of most lakes in the Tibetan Plateau occurred during the early Holocene, while the arid region lakes in northwestern China reached their maximum lake level in the mid-Holocene, and overall high lake-levels in the monsoonal regions were observed during the mid-Holocene (Shen, 2013). A paleovegetation pattern reconstructed by Yu et al. (2000) indicated that favorable vegetation conditions in China prevailed during the mid-Holocene when the forest biomes in eastern China systematically shifted northward and extended westward. These findings have been supported by other studies. For example, shorelines of a series of lakes on Tibetan Plateau showed a humid climate with a persistent and stable high lake level occurred during the early Holocene (Liu et al., 2013). Paleoclimate records on sedimentology, lake-level, and geochemistry data indicated that maximum moisture conditions with the highest lake levels and densest vegetation cover prevailed on the mid-latitude arid Asian region in the mid-Holocene (Chen et al., 2008). Fossil pollen records from 20 sites in the East Asian monsoon margin region indicated that during the Holocene, the region was dominated by steppe vegetation, except for the mid-Holocene period when forest patches were relatively common (Zhao and Yu, 2012). However, as documented by lake-level records in southwest China, high lake-level conditions existed in the area dominated by the Indian monsoon for the early Holocene (Zhang et al., 2016). Based on the numerical modeling and geological data, Li and Morrill (2010) found that most of the lake levels in monsoonal Asia began to rise in the post-glacial or beginning of the Holocene, and reached their highest level during the early Holocene. Although previous records and simulations as discussed above have paid much attention to millennial-scale variations in vegetation and lake hydrological status, few studies have shown a comprehensive understanding for the out-of-phase relationship between them during the Holocene. In addition, some scientists have proposed the hypothesis that lake evolution was asynchronous among various climate zones (Shen, 2013; Wu and Ma, 2011).

In order to explore the temporal and spatial evolution patterns and forcing factors for Holocene vegetation and lake hydrological status in China as well as investigate the relationship between them, this paper presents an analysis and synthesis of various proxies including organic geochemical data, pollen records, and lake-level records from various Köppen climate zones and different geographical regions of China. Organic geochemical proxies and pollen records, providing direct information on past changes in the regional vegetation and climate, were used to reconstruct Holocene vegetation history (Aravena et al., 1992; Bowen, 1991; Krishnamurthy et al., 1986; Li et al., 2009b; Stuiver et al., 1975; Tian et al., 2008; Xiao et al., 2004; Zhao et al., 2010, 2011). Artemisia (%) Chenopodiaceae (%), Pinus (%), Betula (%), Quercus (%), Cyperaceae (%), trees (%), herbs (%), and A/C were collected for inferring the specific plants types. Lake-level records with rich resources for past hydrologic changes (Li and Morrill, 2010; Street-Perrott and Harrison, 1985; Street-Perrott and Roberts, 1983) were chosen to imply long-term changes of lake hydrological status. Meanwhile, 22 lake-level records were verified by the Community Climate System Model (CCSM 3.0), a lake energy-balance model, and a lake water-balance model. Our results help to provide a deeper understanding for Quaternary environment change research and will provide the basis for the reconstruction of past climate change and prediction of future climate variability.

Data, methods, and simulations

Numerical climate classification

Köppen-Geiger climate classification is one of the most widely used climate classification system, which is developed based on the vegetation distribution connected to annual cycles of precipitation and temperature (Lohmann et al., 1993; Naipal et al., 2015; Peel et al., 2007). This modified classification identifies five main climate regions: tropical (A), arid (B), temperate (C), cold (D), and polar (E) (Table 1). The A-, C-, and D-zones are further subdivided closely related to the seasonality of temperature and precipitation (Baker et al., 2010).

Table 1.

Description of Köppen climate symbols and defining criteria (from Peel et al., 2007).

First	Second	Third	Description	Criteria^a
A			Tropical	T_cold ≥ 18°C
	f		Rainforest	P_dry ≥ 60 mm
	m		Monsoon	P_dry ≤ 60 mm and P_dry ≥ (100–MAP/25) mm
	w		Savannah	P_dry ≤ 60 mm and P_dry ≤ (100–MAP/25) mm
B			Arid	MAP < 10 P_th
	w		Desert	MAP ≥ 5 P_th
	s		Steppe	MAP < 5 P_th
		h	Hot	MAT ≥ 18°C
		k	Cold	MAT < 18°C
C			Temperate	T_hot > 10°C and 0°C < T_cold < 18°C
	s		Dry Summer	P_sdry < 40 mm and P_sdry < P_wwet/3
	w		Dry Winter	P_wdry < P_swet/10
	f		Without dry season	Not Cw or Cf
		a	Hot Summer	T_hot ≥ 22°C
		b	Warm Summer	T_hot < 22°C and count (T_mon > 10°C) ≥ 4
		c	Cold Summer	T_cold ≥ −38°C and count (T_mon > 10°C) < 4
D			Cold	T_hot > 10°C and T_cold ≤ 0°C
	s		Dry Summer	P_wdry ≥ P_swet/10
	w		Dry Winter	P_wdry < P_swet/10
	f		Without dry season	Not DS or Dw
		a	Hot Summer	T_hot ≥ 22°C
		b	Warm Summer	T_hot < 22°C and count (T_mon > 10°C) ≥ 4
		c	Cold Summer	T_cold ≥ −38°C and count (T_mon > 10°C) < 4
		d	Very Cold Summer	T_cold < −38°C
E			Polar	T_hot < 10°C
	T		Tundra	T_hot ≥ 0°C
	F		Frost	T_hot < 0°C

MAP: mean annual precipitation; MAT: mean annual temperature; T_hot: temperature of the hottest month; T_cold: temperature of the coldest month; T_mon10: number of months where the temperature is above 10; P_dry: precipitation of the driest month; P_sdry: precipitation of the driest month in summer; P_wdry: precipitation of the driest month in winter; P_swet: precipitation of the wettest month in summer; P_wwet: precipitation of the wettest month in winter; P_threshold: varies according to the following rules (if 70% of MAP occurs in winter then P_threshold = 2 × MAT, if 70% of MAP occurs in summer then P_threshold = 2 × MAT + 28, otherwise P_threshold = 2 × MAT + 14). Summer (winter) is defined as the warmer (cooler) 6-month period of AMJJAS (ONDJFM).

Based on a series of 0.5°×0.5° monthly gridded precipitation and temperature datasets during 1961–2013 at meteorological stations released by the National Meteorological Information Center, and a 0.5°×0.5° gridded elevation dataset downloaded from http://www.ngdc.noaa.gov/mgg/topo/globe.html, we divided China’s climate into five main climate groups, 12 types, and 28 subtypes according to criteria of Table 1. As depicted in Figure 1, China’s main climate groups are Arid (B), Temperate (C), Cold (D), and Polar (E), and main climate types include Steppe climate (Semi-arid) (Bs), Desert climate (Bw), Mild temperate with dry winter (Cw), Mild temperate, Fully humid (Cf), Snow with dry winter (Dw), and Tundra climate (ET).

Figure 1.

Overview map showing the lake sites selected in this study (numbers indicated refer to the lakes, see Table 2), the dominant circulation systems, and the numerical climate classification.

Data sources and analyses

Holocene lake records in this paper were chosen based on four criteria. (1) The selected lakes must have reliable chronologies and successive sedimentary sequences in the Holocene. (2) The record length should cover most of the Holocene without documented depositional hiatuses. (3) The indicators derived from the records must include pollen records, lake-level records, and at least one type of organic geochemical proxies. (4) The selected lake-level records should be reconstructed by the effective proxies except organic geochemical proxies and pollen records. Considering the above-mentioned criteria, 22 lakes derived from various Köppen climate zones and different geographical regions of China were selected (Figure 1), in which 21 lakes contain total organic carbon (TOC) records, 15 lakes contain C/N records, and another 15 lakes contain δ¹³C records (Table 2). The radiocarbon ages (¹⁴C yr BP) of lakes were calibrated to calendar years (cal. yr BP) by the software of Calib611; if necessary, we removed all possible old-carbon effects using information provided in the original publications. The calibrated ages were used in compiling organic geochemical proxies and total pollen concentration throughout the text (Figures 2 –4), and lake-level records and representative pollen indicators are shown in Tables 2 and 3.

Table 2.

Information on lake records in organic geochemical proxies, pollen, and lake-level datasets from China.

ID	Site	Lat.	Long.	Alt. (m)	Proxies	Köppen climate classification			Physical geographic regionalization	Relationship between pollen records and organic geochemical proxies			Lake level			Vegetation conditions			References
ID	Site	Lat.	Long.	Alt. (m)	Proxies	First	Second	Third	Physical geographic regionalization	TOC	C/N	δ¹³C	8.5–6 ka	8.5 ka-PI	6 ka-PI	8.5-6 ka	8.5 ka-PI	6 ka-PI	References
1	Balikun Lake	43°40′N	92°48′E	1575	TOC, Pollen	Arid	Steppe	Cold	Arid and semi-arid China	−	/	/	Low	Low	High	Low	Low	High	Xue and Zhong (2011); Tao et al. (2010)
2	Bosten Lake	42°05′N	87°03′E	1048	TOC, C/N, Pollen	Arid	Desert	Cold	Arid and semi-arid China	−	−	/	Low	Low	High	Low	Low	High	Zhang et al. (2007, 2010)
3	Wulungu Lake	47°12′N	87°17′E	479	TOC, C/N, δ¹³C, Pollen	Arid	Steppe	Cold	Arid and semi-arid China	−	−	−	Low	Low	High	Low	Low	High	Jiang et al. (2007); Liu et al. (2008a)
4	Bangong Co	33°40′N	79°30′E	4241	TOC, Pollen	Arid	Desert	Cold	Qinghai-Tibet Plateau	*	/	/	High	High	High	Low	High	High	Campo et al. (1996); Fan et al. (1996); Yang (2016)
5	Chen Co	28°56′N	90°35′E	4420	TOC, C/N, Pollen	Polar	Tundra	/	Qinghai-Tibet Plateau	+	+	/	High	High	High	Low	High	High	Zhu et al. (2009)
6	Nam Co	30°42′N	90°40′E	4718	TOC, C/N, Pollen	Polar	Tundra	/	Qinghai-Tibet Plateau	+	+	/	High	High	High	Low	High	High	Chen et al. (2017); Quan et al. (2011); Zhu et al. (2008)
7	Naleng Lake	31°06′N	99°45′E	4200	TOC, C/N, δ¹³C, Pollen	Polar	Tundra	/	Qinghai-Tibet Plateau	+	+	+	High	High	High	Low	High	High	Kramer et al. (2010a, 2010b)
8	Zigetang Co	32°04′N	90°50′E	4560	TOC, δ¹³C, Pollen	Polar	Tundra	/	Qinghai-Tibet Plateau	*	/	*	High	High	Low	Low	High	High	Herzschuh (2006); Li et al. (2009a)
9	Gahai Lake	37°08′N	97°33′E	2850	TOC, C/N, δ¹³C, Pollen	Arid	Steppe	Cold	Qinghai-Tibet Plateau	−	+	−	High	High	Low	Low	High	High	Chen (2010); Guo (2012)
10	Qinghai Lake	36°32′N	99°36′E	3200	TOC, C/N, δ¹³C, Pollen	Cold	Dry Winter	Cold Summer	Monsoon marginal zones	+	+	+	High	High	Similar	Low	High	High	Shen et al. (2004, 2005a)
11	Daihai Lake	40°33′N	112°39′E	1220	TOC, C/N, Pollen	Cold	Dry Winter	Warm Summer	Monsoon marginal zones	+	+	/	High	High	High	Low	High	High	Xiao et al. (2004, 2006); Xu et al. (2010)
12	Hulun Lake	49°0′N	117°25′E	545	TOC, δ¹³C, Pollen	Arid	Steppe	Cold	Monsoon marginal zones	+	/	−	Low	High	High	Low	High	High	Hu et al. (1998); Wen et al. (2010); Wang and Ji (1995)
13	Zhuye Lake	39°03′N	103°40′E	1309	TOC, C/N, δ¹³C, Pollen	Arid	Desert	Cold	Monsoon marginal zones	+	+	−	High	High	Similar	Low	High	High	Li et al. (2009b, 2012)
14	Erhai Lake	25°47′N	100°12′E	1974	TOC, δ¹³C, Pollen	Temperate	Dry Summer	Warm Summer	Monsoon regions	+	/	+	High	High	High	Low	High	High	Shen et al. (2005b, 2006)
15	Erlongwan Maar Lake	42°18′N	126°22′E	722	TOC, C/N, δ¹³C, Pollen	Cold	Dry Winter	Warm Summer	Monsoon regions	+	+	−	High	High	High	Low	High	High	Liu et al. (2008b); You and Liu (2012)
16	Xingkai Lake	44°57′N	132°25′E	69	TOC, C/N, δ¹³C, Pollen	Cold	Dry Winter	Warm Summer	Monsoon regions	+	+		High	High	High	Low	High	High	Wu et al. (2010a, 2010b)
17	Lugu Lake	27°42′N	100°50′E	2690	TOC, Pollen	Temperate	Dry Winter	Warm Summer	Monsoon regions	+	/	/	High	High	High	Low	High	High	Zheng (2014a, 2014b)
18	Dahu Lake	24°41′N	115°00′E	255	TOC, C/N, δ¹³C, Pollen	Temperate	Without dry season	Hot Summer	Monsoon regions	−	−	−	High	High	High	High	High	Low	Zhong et al. (2010); Xiao et al. (2007)
19	Dajiu Lake	31°28′N	110°04′E	1780	TOC, C/N, δ¹³C, Pollen	Temperate	Dry Winter	Hot Summer	Monsoon regions	−	−	−	High	High	High	Low	Low	High	Ma et al. (2008); Zhu et al. (2006)
20	Poyang Lake	29°01′N	116°08′E	1324	δ¹³C, Pollen	Temperate	Without dry season	Hot Summer	Monsoon regions	/	/	+	High	High	High	Low	Low	High	Ma et al. (2004); Jiang and Piperno (1999)
21	Xingyun Lake	24°20′N	102°47′E	1740	TOC, C/N, δ¹³C, Pollen	Temperate	Dry Winter	Warm Summer	Monsoon regions	+	+	+	High	High	High	Low	High	High	Zhang et al. (2014); Cao et al. (2014)
22	Huguangyan Maar Lake	21°09′N	110°17′E	87.6	TOC, C/N, δ¹³C, Pollen	Temperate	Dry Winter	Warm Summer	Monsoon regions	+	+	+	High	High	High	High	High	High	Mingram et al. (2004); Wang et al. (2007)

TOC: total organic carbon; PI: pre-industrial.

‘+’ indicates a similar trend between total pollen concentrations and geochemical proxies; ‘−’ indicates an opposite trend, while ‘*’ indicates that there is no obvious relationship between them; and ‘\’ means that there are no data available in that lake.

Figure 2.

Relationship between total pollen concentration (10³ grains/g or grains/cm³) and total organic carbon (TOC, %). Gray bars suggest pollen concentration and black curves indicate TOC. Roman numerals show the geographical zones of China: the arid and semi-arid area in northwest China (I), the Qinghai-Tibet Plateau (II), the monsoon marginal zones (III), and the monsoonal regions (IV).

Figure 3.

Relationship between total pollen concentration (10³ grains/g or grains/cm³) and carbon nitrogen ratio (C/N). Green bars indicate pollen concentration and black curves show C/N.

Figure 4.

Relationship between total pollen concentration (10³ grains/g or grains/cm³) and organic carbon isotope (δ¹³C, ‰). Blue bars indicate pollen concentration and black curves show δ¹³C.

Table 3.

Information on fossil pollen records in representative pollen indicators from China.

ID	Site	Artemisia			Chenopodiaceae			Pinus			Betula			Cyperaceae			Quercus			A/C			Trees			Herbs
ID	Site	8.5–6 ka	8.5 ka-PI	6 ka-PI	8.5–6 ka	8.5 ka-PI	6 ka-PI	8.5–6 ka	8.5 ka-PI	6 ka-PI	8.5–6 ka	8.5 ka-PI	6 ka-PI	8.5–6 ka	8.5 ka-PI	6 ka-PI	8.5–6 ka	8.5 ka-PI	6 ka-PI	8.5–6 ka	8.5 ka-PI	6 ka-PI	8.5–6 ka	8.5 ka-PI	6 ka-PI	8.5–6 ka	8.5 ka-PI	6 ka-PI
1	Balikun Lake	Low	Low	Low	High	High	Low	Low	Low	High	Low	High	High	Low	Low	Low	–	–	–	Low	Low	High	–	–	–	Low	High	High
2	Bosten Lake	High	High	High	High	High	Low	–	–	–	–	–	–	High	High	High	–	–	–	Low	Low	High	–	–	–	Low	High	High
3	Wulungu Lake	Low	Low	High	Low	Low	Low	–	–	–	Low	Low	High	–	–	–	–	–	–	High	High	High	Low	Low	High	–	–	–
4	Bangong Co	High	High	High	High	High	Low	–	–	–	–	–	–	Low	High	High	–	–	–	Low	Low	High	–	–	–	–	–	–
5	Chen Co	High	Low	Low	–	–	–	High	High	High	Low	Low	High	Low	High	High	–	–	–	Low	Low	Low	–	–	–	–	–	–
6	Nam Co	High	High	High	High	Low	Low	Low	Low	Low	–	–	–	Low	Low	Low	–	–	–	–	–	–	Low	Low	High	–	–	–
7	Naleng Lake	–	–	–	High	Low	Low	Low	Low	Low	Low	High	High	High	High	Low	–	–	–	–	–	–	Low	High	High	–	–	–
8	Zigetang Co	High	High	High	High	Low	Low	High	Low	Low	Low	Low	High	Similar	Low	Low	–	–	–	–	–	–	–	–	–	–	–	–
9	Gahai Lake	High	High	High	High	High	Low	High	High	High	High	High	High	Low	High	High	–	–	–	High	High	Low	–	–	–	–	–	–
10	Qinghai Lake	High	High	Low	Low	Low	Low	Low	Low	High	High	High	High	High	High	Low	High	High	High	High	Low	Low	Low	High	High	–	–	–
11	Daihai Lake	High	Low	Low	Low	Low	Low	Low	High	High	High	High	High	–	–	–	Low	High	High	High	High	High	Low	High	High	High	Low	Low
12	Hulun Lake	High	High	Low	Low	Low	Low	Low	Low	Low	Low	High	High	Low	Low	Low	–	–	–	High	High	High	Low	Low	High	High	High	Low
13	Zhuye Lake	Low	Low	High	Low	Low	High	High	High	Low	–	–	–	–	–	–	Low	High	High	High	High	High	High	High	High	Low	Low	High
14	Erhai Lake	Low	Low	Low	–	–	–	Low	High	High	High	High	High	–	–	–	Low	Low	Low	Low	High	High	Low	High	High	Low	High	High
15	Erlongwan Maar Lake	High	High	Low	High	High	Low	High	Low	Low	High	High	Low	High	High	Low	Low	Low	High	High	High	Low	–	–	–	High	High	Low
16	Xingkai Lake	High	High	Low	Low	High	High	Low	Low	Low	Low	High	High	High	High	High	Low	Low	Low	High	High	High	–	–	–	–	–	–
17	Lugu Lake	High	High	Low	–	–	–	Low	High	High	High	High	Low	Low	Low	Low	Low	Low	High	Low	High	High	–	–	–	–	–	–
18	Dahu Lake	–	–	–	–	–	–	High	Low	Low	–	–	–	Low	Low	High	High	High	High	–	–	–	High	High	Low	High	Low	Low
19	Dajiu Lake	Low	Low	Low	High	High	High	Low	Low	High	Low	Low	High	–	–	–	Low	Low	High	Low	Low	Low	Low	Low	High	Low	High	High
20	Poyang Lake	Low	High	High	–	–	–	Low	Low	High	Low	Low	High	Low	Low	High	Low	Low	High	–	–	–	–	–	–	–	–	–
21	Xingyun Lake	Low	Low	Low	–	–	–	Low	Low	High	Low	Low	High	Low	High	High	Low	Low	High	–	–	–	–	–	–	Low	Low	Low
22	Huguangyan Maar Lake	High	Low	Low	High	High	Low	–	–	–	–	–	–	High	Low	Low	Low	Low	High	High	Low	Low	High	High	High	Low	Low	Low

PI: pre-industrial.

Significance of the environmental proxies

Pollen grains get washed or blown into lakes and then accumulate in sediments, proving a record of past vegetation. Different types of pollen in lake sediments reflect vegetation around the lake and climate conditions that are favorable for the vegetation (Heuer and Overpeck, 2003). In this paper, Artemisia (%), Chenopodiaceae (%), Pinus (%), Betula (%), Quercus (%), Cyperaceae (%), trees (%), herbs (%), and A/C were selected as representative pollen indicators. They were chosen to be included here not only because they contained topical pollen of the four geographical regions but also because their paleoclimate significance was relatively clear. The pollen spectra dominated by tree species such as Pinus, Betula, and Quercus indicates large-scale forests are developed, while the relatively higher percentage of herbaceous species such as Artemisia, Chenopodiaceae, and Cyperaceae means grassland vegetation cover. Pinus is a less significant environmental indicator owing to its ease of transport over long distances, and Betula grows in the transition zone from brushland to steppes (Li et al., 1985). They both require certain water conditions but are relatively tolerant of lower temperature as inferred from their altitudinal ranges. The genus Quercus is one of the most important tree genera in temperate and subtropical areas of the Northern Hemisphere, indicating a warm and humid environment (Dowsett et al., 1994). Artemisia and Chenopodiaceae are both representative taxa of altitudinal desert zones, but Artemisia generally lives in a more humid environment and Chenopodiaceae is widespread during dry periods. Moreover, in the Artemisia and Chenopodiaceae dominated regions (at least more than half of the pollen sum), A/C can indicate vegetation ecology that suggests desert vegetation when it is <0.5, desert-steppe vegetation when it is 0.5–1.0, and steppe vegetation when it is >1.0 (Jiang et al., 2007; Sun et al., 1994). The family Cyperaceae is meadow vegetation belonging to herbaceous species and mainly grows in a humid environment (Gignac et al., 2004).

Organic matter in soils and sediments, containing a large variety of organic materials including organic geochemical proxies, is widely distributed over the earth’s surface occurring in almost all terrestrial and aquatic environments (Schnitzer, 1978). The TOC content of lake sediments that originally derived from terrestrial and lake sources can indicate primary productivity around (Balascio et al., 2013; Plewa et al., 2006). The C/N ratio can imply the relative amounts of autochthonous aquatic plants versus the allochthonous terrestrial organic matter during the sedimentary period (Wagner et al., 2000). C/N less than 10 indicates that organic matter mainly derives from lacustrine algae, whereas C/N between 14 and 23 indicates terrestrial organic matter sources (Meyers, 1994, 1997; Meyers and Ishiwatari, 2008; Meyers and Vergés, 1999). The origin of organic matter can be inferred relatively from stable carbon isotopic composition of lake sediments. In terms of the difference in photosynthetic reactions, the land plants are divided into C₃, C₄, and Crassulacean Acid Metabolism (CAM) plants with the ranges of δ¹³C value between −34‰ to −23‰, −22‰ to −6‰, and −20‰ to −10‰, respectively (O’Leary, 1988; Tieszen and Boutton, 1989; Zhang et al., 2004). Likewise, the ranges of δ¹³C value in lake aquatic plants are various: −12‰ to −20‰ in submerged plant, −24.00‰ to −30.00‰ in emergent aquatic plant, and δ¹³C value can even reach −35.50‰ in floating plant (Bowen, 1991).

The lake-level fluctuation, resulting from fossil shoreline deposits, sedimentary facies, fossils, and other features indicative of past lake levels, is a highly sensitive indicator for hydrological climate (Herzschuh, 2006; Kleinherenbrink et al., 2014; Li and Morrill, 2010). This paper considered multi-proxies from different profiles for an individual lake in original publications or extracted directly from publications as one record. For instance, a high lake-level period would be assigned to an interval with fine-grained sediments, low carbonate content, high magnetic susceptibility, or abundant planktonic diatoms. Lake-level changes in China at three periods, the early (12–8 ka), the mid- (8–4 ka), and the late-Holocene (4–0 ka), were reconstructed and compared (Table 2) in this paper. The justification for temporal and spatial evolution of lake level was that each individual lake might be different from other lakes on account of different forcing factors, geographic locations, and sedimentary indicators (Chen et al., 2008).

Lake-level simulations

A series of models, the CCSM 3.0, a lake energy-balance, and a lake water-balance model, was used to simulate millennial-scale variations in lake level and examine the reasons for lake-level changes in China between the early (8.5 ka), mid- (6.0 ka), and late-Holocene (ca. 1800 AD) (Li and Morrill, 2010).

CCSM 3.0 is a state-of-the-art coupled climate model consisted by four component models of the ocean, sea ice, atmosphere, and land that are coupled without flux corrections. The horizontal resolution of the ocean and sea ice models is nominally 1°C, while that of the atmosphere and land models is 2.8°C. Furthermore, there are 40 vertical levels in the ocean model, 26 vertical levels in the atmosphere model, as well as 10 subsurface soil layers in the land model. Three simulations were completed, for 8.5 ka, 6.0 ka, and a pre-industrial (PI) conditions (Figure 6). The 8.5 ka was chose as an early Holocene timeslice because it could represent environmental conditions that were not influenced by the 8.2 ka cooling event. The 6.0 ka timeslice was widely used as a representative period of the mid-Holocene, and the PI was considered as the typical period of the late-Holocene because of relatively less human interference. Table 4 shows the boundary conditions of orbital forcing (Berger, 1978), atmospheric trace gases (Flückiger et al., 1999, 2002; Monnin et al., 2004), continental ice sheets, and vegetation (Peltier, 2004) for each simulation (Li and Morrill, 2010).

Table 4.

Boundary conditions used for CCSM 3 model experiments at 8.5 ka, 6 ka, and PI.

	8.5 ka	6 ka	Pre-industrial
Eccentricity	0.019199	0.018682	0.016724
Obliquity (°)	24.22	24.1	23.45
Long. of perihelion (°)	319.5	0.87	102.04
CO₂ (ppm)	260	280	280
CH₄ (ppb)	660	650	760
N₂O (ppb)	260	270	270
Ice sheet	Peltier (2004), 8.5 ka	Peltier (2004), 0 ka	Peltier (2004), 0 ka
Vegetation	Present-day	Present-day	Present-day

CCSM: Community Climate System Model; PI: pre-industrial.

We conducted lake evaporation calculations using a lake energy-balance model that is based on the model of Hostetler and Bartlein (1990). In this model, the depth of the lake was assumed as 1 m, since the experiments increasing lake depth to 5 and 10 m and increasing lake salinity to 10 ppt show only small changes in lake evaporation. Meanwhile, the vertical level of lake water is not mixed and the surface energy balance for the lake is controlled by water surface evaporation (Li and Morrill, 2010). The equation is as follows:

c_{w} ρ_{w} z \frac{\partial T}{\partial t} = ϕ_{s} + ϕ_{l d} - ϕ_{l u} \pm Q_{e} \pm Q_{h}

(1)

where c_w, ρ_w, z, and T represent specific heat of water (J kg⁻¹ K⁻¹), water density (kg m⁻³), lake depth (m), and lake surface temperature (K), respectively. T is the time, φ_s and φ_ld are shortwave and longwave radiation absorbed by water, φ_lu is longwave radiation emitted by lake, Q_e represents latent heat fluxes, and Q_h represents sensible heat fluxes.

This paper assumed that each grid cell of the CCSM 3.0 model is a single individual that is in hydrological equilibrium with steady-state conditions of climate, and the lake water-balance equation:

V = A_{B} R + A_{L} (P_{L} - E_{L})

(2)

where V is discharge from the lake (m³/yr), A_B and A_L are watershed area (m²) and lake area (m²), R is watershed runoff (m/yr), P_L is on-lake precipitation (m/yr), and E_L is lake evaporation (m/yr). However, A_B and A_L are difficult to obtain for a hypothetical lake, so Eq. 2 is simplified as follows:

\frac{A_{L}}{A_{B}} = \frac{R}{(E_{L} - P_{L})}

(3)

Results

Organic geochemical proxies and pollen records

Since sediment accumulation rate, physical and geographical conditions, and paleoclimate significance of the proxies vary according to the location, numerical relationships between organic geochemical proxies and total pollen concentration have been difficult to achieve. However, we can depict their correlation based on the changing trends of the proxies. The ‘+’ indicates a similar trend between total pollen concentration and geochemical proxies; the ‘−’ indicates an opposite trend, and the ‘*’ indicates that there is no obvious correlation between them, which can be clearly observed in the Figures 2 –4 (Table 2). According to the relationship between organic geochemical proxies and total pollen concentration from 22 lakes, the study region can be divided into arid and semi-arid area of northwest China (I), the Qinghai-Tibet Plateau (II), the monsoon marginal zones (III), and the monsoonal regions (IV) (Figure 5). In arid and semi-arid of northwest China, Bosten Lake, Wulungu Lake, and Balikun Lake have similar characteristics that organic geochemical proxies and pollen concentration show opposite trends. Lakes in this region are typically discharged by glacial-melt water; thus, the strong basin-wide hydrodynamic conditions usually correspond to the abundant exogenous fluxes. Inversely, the pollen is relatively vulnerable under these conditions, so that total pollen concentration shows negative correlations with organic geochemical proxies in arid and semi-arid of northwest China. The Qinghai-Tibet plateau can be divided into a northwest and a southeast part. Bangong Co, Zigetang Co, and Gahai Lake in the northwest plateau show totally different variation trends of organic geochemical proxies and total pollen concentration owing to the low biomass compared with other parts of the plateau. In this case, TOC and C/N in lake sediments cannot accurately respond to Holocene millennial-scale climate changes, and pollen records are more effective in showing vegetation variations. However, there are positive correlations between organic geochemical proxies and pollen concentration in lake sediments of Nam Co, Chen Co, and Naleng Lake from the southeast plateau. Qinghai Lake, Daihai Lake, Hulun Lake, and Zhuye Lake belong to the monsoon marginal zones, where vegetation is dominated by drought-tolerant plants (C₃ plants with heavy δ¹³C values) (Herzschuh, 2006; Kleinherenbrink et al., 2014; Li and Morrill, 2010). Thus, high primary productivity in those regions is usually linked to the relatively high TOC, C/N values, high total pollen concentration, and low δ¹³C values. As a result, lacustrine TOC, C/N, and total pollen concentration indicate similar changing trends during the Holocene, and δ¹³C shows an opposite trend with total pollen concentration in the monsoon marginal zones. Exceptionally, Qinghai Lake with an attitude of 3200 m shows a positive correlation between δ¹³C and pollen concentration, which can be attributed to altitude effects of the plant δ¹³C (Li et al., 1999). In the monsoonal regions, pyrophilous C₄ plants with light δ¹³C values show a high proportion in warm-humid climates, whereas during arid periods, C₄ plants almost disappear and are replaced by frigostable C₃ plants. Therefore, Erhai Lake, Lugu Lake, Poyang Lake, Xingyun Lake, and Huguangyan Maar Lake in those regions are characterized by similar changing trends of organic geochemical proxies and total pollen concentration. Nevertheless, there are some exceptions presenting inverse correlations between δ¹³C and pollen concentration. Located in the relatively high latitude, the negative relationship in Erlongwan Maar Lake and Xingkai Lake is mainly related with various forcing factors of plants δ¹³C in different latitude (Zhou et al., 2013). As for Dahu Lake and Dajiu Lake situated in the humid region of China, this is closely associated with high water stands and diluted water.

Figure 5.

Spatial distribution of relationship between total pollen concentration and organic geochemical proxies: (a) TOC, (b) C/N, and (c) δ¹³C.

Lake-level records and simulations

Figure 6 indicates the millennial-scale variations in lake-level based on the CCSM 3.0, a lake energy-balance as well as a lake water-balance model, and the simulated results show that lake-level varies by regions and through time. In monsoonal regions and the monsoon marginal area, lake levels begin to rise from the early Holocene, then decrease around the mid-Holocene ca. 6 cal. ka BP, and in the late-Holocene, lake level reaches its lowest level (Figure 6a). This general trend in the evolution of lake level is in accordance with most lake records from the two regions. In arid and semi-arid of northwest China, the models show low lake level at 8.5 ka relative to 6.0 ka and PI, and high lake level at 6.0 ka relative to PI (Figure 6b), which is also in line with the three lake-level records in the area. Lakes in the Qinghai-Tibet Plateau reach their highest lake level during the early Holocene, and fall to their lowest level in the late-Holocene, agreeing with most of lake-level records, but Zigetang Co and Gahai Lake show different patterns with high lake-level at PI relative to 6 ka.

Figure 6.

Simulated lake-level changes for (a) 8.5 ka minus 6.0 ka, (b) a 8.5 ka minus PI, and (c)6.0 ka minus PI based on the CCSM 3.0, the lake energy-balance, and the lake water-balance model.

The lake simulations also show the reasons for lake-level changes in China between the early (8.5 ka), mid- (6.0 ka), and late-Holocene (ca. 1800 AD). In the monsoonal regions, the monsoon marginal zones, as well as the southeast Qinghai-Tibet Plateau, the decrease in lake level during the Holocene is generally related to the combined effects of decreasing precipitation and increasing lake evaporation. In the northwest plateau, the main factor driving lake-level rise from 8.5 to 6 ka is the increased precipitation at 6 ka; then precipitation decrease and lake evaporation increase cause the high lake level at 6 ka relative to PI. Lake-level changes in arid and semi-arid of northwest China through the Holocene are only associated with precipitation, the increase in precipitation from early to mid-Holocene, and the decrease in precipitation from mid- to late-Holocene give rise to the high lake-level at the mid-Holocene.

Discussion

As widely used paleoenvironment indicators, organic geochemical proxies and pollen records indicate different aspects of vegetation changes according to conventional interpretation (Bowen, 1991; Krishnamurthy et al., 1986; Li et al., 2009b; Stuiver et al., 1975; Tian et al., 2008; Xiao et al., 2004). Organic geochemical and proxies generally imply the overall conditions of the paleoecosystem, whereas pollen records are used for the detailed description of paleovegetation changes (Bowen et al., 1991; Krishnamurthy et al., 1986; Li et al., 2009; Stuiver et al., 1975; Tian et al., 2008; Xiao et al., 2004). Accordingly, we chose the two indicators to reconstruct the millennial-scale variations in vegetation. Our synthesis showed that vegetation and lake hydrology of China did not show synchronous and coherent changes during the Holocene in each climate zone, but both of them showed relatively strong temporal and spatial variability in various geographical regions.

Arid and semi-arid of northwest China

Based on our analysis and synthesis of the organic geochemical indicators and pollen records of Balikun Lake, Bosten Lake, and Wulungu Lake, we reconstructed the Holocene vegetation history in arid and semi-arid of northwest China. The records suggested that during the early Holocene, sediments were characterized by low TOC, C/N, and δ¹³C; low total pollen concentration; and low A/C value, indicating the climate was arid. Meanwhile, pollen assemblages dominated by Chenopodiaceae showed that the lake basins were covered by dry steppe. Sediments that deposited during the middle Holocene were represented by rapid increases of organic geochemical proxies and total pollen concentration. Pollen assemblages also changed significantly in this period, and the increase in Artemisia and A/C ratios implied the environment became slightly more humid and the regional vegetation was desert-steppe (Figure 7). In the late-Holocene, organic geochemical proxies and total pollen concentration were still high but the A/C ratios descended, suggesting a decreasing effective moisture trend but the regional vegetation was still desert-steppe. Millennial-scale variations in lake-level can directly reflect the long-term changes in lake hydrological status according to the previous studies (Chen et al., 2008; Herzschuh, 2006; Kleinherenbrink et al., 2014). Therefore, the optimal lake hydrological conditions in arid and semi-arid northwest China occurred in the mid-Holocene, and lake hydrological status in the late-Holocene was better than the early Holocene (Figure 8).

Figure 7.

Vegetation evolution in China during the Holocene. Green points indicate that optimal vegetation cover occurred in the early Holocene. Red points reflect that large-scale vegetation cover was presented during the mid-Holocene.

Figure 8.

Millennial-scale variations in lake hydrological status from China during the Holocene. Green points indicate that lakes experienced optimal lake hydrological status in the early Holocene. Red points show that favorable water status was prevailed in the middle Holocene.

In arid and semi-arid of northwest China, vegetation and lake hydrological status are strongly influenced by precipitation. Poor hydrological status and low vegetation coverage during the early Holocene in this region were closely related to the weakened mid-latitude Westerlies at the winter-spring season and the reduction in westward propagation of water vapor from the Mediterranean, Black Sea, and North Atlantic (Chen et al., 2008; Jin et al., 2011; Li and Morrill, 2010; Liu and Li, 2016; Shen, 2013). At the same time, large-scale ice sheets at high latitude and meltwater discharges modulated and reduced air and sea-surface temperature also causing decrease in vapor transport of the Westerlies (Chen et al., 2008). Another explanation might be the regional difference in the uplift and descent of air masses (Chen et al., 2008; Herzschuh, 2006; Shen, 2013). Strong insolation in summer at low latitudes caused a strong uplift motion of air mass forming a low pressure cell, which led to a large-scale convergence of the Asia summer monsoon in the lower troposphere, resulting in the intensified subsidence of air masses to the Tibetan Plateau and inevitably dry climate in arid and semi-arid areas (Broccoli and Manabe, 1992; Chen et al., 2008; Herzschuh, 2006; Rodwell and Hoskins, 1996; Ye and Gao, 1979). From the mid- to late-Holocene, the increase in high-altitude air and the sea-surface temperatures caused by the reduced ice sheets might induce more vapor from the boreal winter sea to be transported to the Chinese northwest region (Chen et al., 2008; Numaguti, 1999). Furthermore, high air temperatures in the mid-Holocene would be expected to increase cyclonic activity and synoptic disturbances along the Westerlies, bringing more convective precipitation (Chen et al., 2008). Consequently, vegetation and lake hydrological status experienced synchronous changes during the Holocene in arid and semi-arid of northwest China, indicating a dry early Holocene, a warm and humid mid-Holocene, and a moderately wet and cold late-Holocene. This Holocene climate history appears to be similar to the pattern of the previous environment reconstructions (Boomer et al., 2000; Chen et al., 2008; Jin et al., 2011; Li and Morrill, 2010; Liu and Li, 2016; Shen, 2013; Zhang et al., 2016).

The Qinghai-Tibet Plateau

The Qinghai-Tibet Plateau, with an average elevation of more than 4000 m a.s.l., plays an important role in determining regional weather and climate patterns in East and South Asia (Herzschuh et al., 2006). In the southeast plateau, organic geochemical proxies and total pollen concentration had similar changing trends during the Holocene. Low TOC, C/N, δ¹³C, and total pollen concentration in the early Holocene implied lower production of terrestrial plants and aquatic plankton under cold conditions, and all types of pollen indicated the presence of alpine meadows in this period. As suggested by high TOC and total pollen concentration, the environment in the mid-Holocene turned warm and wet; meanwhile, the vegetation was transformed to the coniferous mixed forest (Figure 7). During the late-Holocene, lower organic geochemical proxies indicated that the temperature decrease caused the re-expansion of alpine meadows. Vegetation reconstructions in the northwest plateau suggested a dominance of desert-steppe vegetation during the early and late-Holocene, and alpine steppes with desert elements tended to dominate the mid-Holocene. On the other hand, lake-level records suggested that most lakes on the Qinghai-Tibet Plateau experienced favorable hydrological status in the early Holocene, suboptimal hydrological conditions at the mid-Holocene, and poor hydrological status during the late-Holocene (Figure 8).

In the Qinghai-Tibet Plateau, temperature is the main limiting factor for vegetation growth, while lake water status relates to precipitation and evaporation. According to sedimentary records from the plateau, temperature variations have undergone three stages: a temperature increase in the early Holocene, peak temperature in the mid-Holocene, and a temperature decrease in the late-Holocene (Shen, 2013; Zhou and Shen, 2007). Therefore, most lakes of the Plateau were characterized by flourishing vegetation in the mid-Holocene (Herzschuh et al., 2010). Paleoclimate studies of the late-Quaternary monsoon history derived from the compiled records revealed that the Indian summer monsoon and East Asia summer monsoon were both highly sensitive to millennial-scale variations in insolation (Agnihotri et al., 2002; Herzschuh, 2006; Leuschner and Sirocko, 2003). Strong summer insolation in the Northern Hemisphere during the early Holocene (Berger and Loutre, 1991) strengthened the Asian monsoon and brought considerable precipitation (Krause and Whitlock, 2013). Meanwhile, enhanced cloud cover blocked a greater fraction of insolation, reducing surface shortwave radiation and surface temperatures and giving rise to the favorable hydrological condition in the early Holocene (Li and Morrill, 2010). On the other hand, rapid glacial melting at the onset of Holocene also contributed to the abrupt increases in lake levels during the early Holocene. In the mid-Holocene, despite the higher air temperature, the increased rainfall could not compensate for the decreased glacial-melt-waters owing to glacial retreat, so lake level began to descend since the mid-Holocene (Li and Morrill, 2010; Shen, 2013). However, because Zigetang Co and Gahai Lake were situated in the northwest plateau where the east Asian summer monsoon had significantly retreated southeastward since the mid-Holocene and then the climate there was mainly dominated by mid-latitude westerly winds, the two lakes showed better hydrological status in the late-Holocene relative to the mid-Holocene (Agnihotri et al., 2002; Leuschner and Sirocko, 2003; Shen, 2013).

Monsoon marginal zones

The Holocene vegetation and hydrologic changes of the monsoon marginal zones were reconstructed based on the analysis of organic geochemical proxies and pollen records derived from Qinghai Lake, Daihai Lake, Hulun Lake, and Zhuye Lake. In the early Holocene, pollen assemblages showed that arid steppe dominated the regions in company with patches of pine forests indicating a relatively arid environment. During the mid-Holocene, the optimal environment prevailed there characterized by the notably increased TOC, C/N, and pollen concentration; heavy δ¹³C; as well as intensive mixed pine and broadleaved forests (Figure 7). In the late-Holocene, the forests disappeared and the vegetation density decreased. On the contrary, lake hydrological status reflected by lake-level records indicated optimal water status occurred in the early Holocene, and lake hydrological status in the mid-Holocene was better than in the late-Holocene (Figure 8).

Monsoon marginal zones are located in the interaction between the Asian summer monsoon and the westerly winds, where moisture and temperature conditions are both the vital limiting factors for plant growth, and lake-level changes in those regions are strongly affected by precipitation and evaporation. In the early Holocene, increased summer insolation strengthened the Asian monsoon and brought considerable precipitation (Agnihotri et al., 2002; Chen et al., 2008; Herzschuh, 2006; Leuschner and Sirocko, 2003; Shen, 2013). Simultaneously, enhanced cloud cover blocked reduced surface shortwave radiation and surface temperatures (Li and Morrill, 2010), resulting in the optimal hydrological status prevailing at this stage. During the mid-Holocene thermal optimum, the intensity of precipitation was still strong, so flourishing vegetation in the monsoon marginal zones was indicated. In addition, the increasing trends of lake hydrological status in Qinghai Lake and Zhuye Lake during the late-Holocene might associate with the enhancing westerly flow (Agnihotri et al., 2002; Leuschner and Sirocko, 2003; Qian et al., 2009; Shen, 2013). Poor lake hydrological conditions in Hulun Lake during the early Holocene were largely caused by remnant large ice sheets at high latitudes (Chen et al., 2008; Ruddiman, 2006) and simultaneously prevailing dry conditions in the area adjacent to the north (Wang and Ji, 1995).

Monsoonal regions

The organic geochemical indicators and pollen records from lakes spanning the Holocene revealed a detailed history of vegetation over the monsoonal regions. In the early Holocene, pine forest dominated the regions in company with patches of herbs and shrubs, suggesting a wet and cold climatic condition. During the mid-Holocene, evergreen broadleaved forest appeared to have extended over much of the lower latitude, and the higher latitude was dominated by the deciduous broadleaved forest. The late-Holocene was characterized by a decline in broadleaved and then pine forest expanding again. However, on account of increased frequency and intensity of El Nino events (Riedinger et al., 2002; Wang et al., 2007; Xiao et al., 2007) and northward movement of the Inter-Tropical Convergence Zone (Fleitmann et al., 2003; Hastenrath and Greischar, 1993; Wang et al., 2007) in the mid-Holocene, Dahu Lake and Huguangyan Maar Lake showed different vegetation evolution patterns with sparse vegetation coverage. The lake hydrological status of the monsoonal regions showed similar pattern with the monsoon marginal zones on millennial scale, indicating optimal water status in the early Holocene (Figure 8).

The monsoonal regions of eastern China are directly affected by the Asian monsoon. Vegetation conditions are largely related to precipitation and temperature (An et al., 2000), whereas lake hydrological status is mainly controlled by precipitation and evaporation. Paleoclimate studies on lake sediments indicated a strong intensification of both the Indian and the East Asian summer monsoon at the early Holocene (Agnihotri et al., 2002; Chen et al., 2008; Herzschuh, 2006; Leuschner and Sirocko, 2003; Shen, 2013), resulting in considerable precipitation and less lake surface evaporation (lake evaporation). Therefore, favorable hydrological status occurred at the early Holocene in the monsoonal regions. During the mid-Holocene, with the increase in temperature, the intensity of precipitation was still strong as a result of the relatively weak East Asian summer monsoon in lower latitude and a significant enhancement of East winter Asian monsoon in the higher latitude; thus, optimal vegetation cover was occurred at this stage (Li et al., 2017). However, some proxy data from lake sediments in previous studies showed different patterns in lake hydrological status and vegetation changes with our reconstructions during the Holocene. In their review, An et al. (2000) pointed out that the Holocene optimum of the East Asian monsoon was different among various monsoonal regions. Herzschuh (2006) proposed that optimal climate conditions were asynchronous between the Indian summer monsoon and South-East Asian monsoon regions. However, recent observational analysis and modeling studies do not appear to support those hypotheses because of the coarse-resolution (Chen et al., 2008; Feng et al., 2006; Li and Morrill, 2010).

Recent observational analysis suggested monsoon rainfall increased dramatically in southern China, but drought conditions prevailed in northern China over the past several decades (Li and Xu, 2016; Li et al., 2012). Indeed, extreme weather events such as floods, droughts, precipitation, and large storms are likely to become more frequent and more intense (Boukabara et al., 2007; Hansen et al., 2014; Kreyling et al., 2008). Meanwhile, ecological damage is another serious environmental problem (Gimpelhinteregger, 2008). The evolution of Holocene vegetation and lake hydrological status across China suggests that the mid-Holocene thermal optimal is related to the low-latitude summer insolation changes, which is different from current global warming. Therefore, our research regarding the millennial-scale vegetation and hydrological status changes has important scientific significance.

Conclusion

This paper provides a comprehensive review of temporal and spatial evolution of the Holocene vegetation and lake hydrological status from China based on the numerical climate classification, lake-level simulations, pollen records, organic geochemical proxies, as well as paleoenvironmental indicators from various Köppen climate zones and different geographical regions of China. The results showed that vegetation and lake hydrology of China did not show synchronous and coherent changes during the Holocene in each climate zone, while both of them showed relatively strong temporal and spatial variability in various geographical regions. Overall large-scale vegetation cover in China was observed during the mid-Holocene, whereas the Holocene hydrological optimum was asynchronous in China, for example, the mid-Holocene in northwestern China, and the early Holocene in the Qinghai-Tibet Plateau, the monsoon marginal zones, and the monsoonal regions. The out-of-phase pattern of Holocene vegetation and lake hydrological status were closely related to different forcing factors in various regions. Lake hydrological status in arid and semi-arid of northeast China was strongly influenced by precipitation, whereas in other regions of China, there was a combined influence of evaporation and precipitation. Likewise, the limiting factor of vegetation growth was various, for example, moisture in northwestern China, temperature in the Qinghai-Tibet Plateau, as well as moisture and temperature in the monsoon marginal zones and monsoonal regions.

Footnotes

Funding

This work was supported by the National Natural Science Foundation of China (Grant No. 41571178 and 41371009) and the Fundamental Research Funds for the Central Universities (Grant No. lzujbky-2015-143).

References

Agnihotri

Dutta

Bhushan

et al . (2002) Evidence for solar forcing on the Indian monsoon during the last millennium. Earth and Planetary Science Letters 198: 521–527.

Poter

Kutzbach

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

Aravena

Warner

MacDonald

(1992) Carbon isotope composition of lake sediments in relation to lake productivity and radiocarbon dating. Quaternary Research 37: 333–345.

Baker

Diaz

Hargrove

et al . (2010) Use of the Köppen-trewartha climate classification to evaluate climatic refugia in statistically derived ecoregions for the People’s Republic of China. Climatic Change 98: 113–131.

Balascio

D’Andrea

Bradley

et al . (2013) Biogeochemical evidence for hydrologic changes during the Holocene in a lake sediment record from southeast Greenland. The Holocene 23: 1428–1439.

Berger

(1978) Long-term variations of caloric insolation resulting from the Earth’s orbital elements. Quaternary Research 9: 139–167.

Berger

Loutre

(1991) Insolation values for the climate of the last 10 million years. Quaternary Science Reviews 10: 297–317.

Boomer

Aladin

Plotnikov

et al . (2000) The palaeolimnology of the Aral Sea: A review. Quaternary Science Reviews 19: 1259–1278.

Boukabara

Weng

Liu

(2007) Passive microwave remote sensing of extreme weather events using NOAA-18 AMSUA and MHS. IEEE Transactions on Geoscience & Remote Sensing 45: 2228–2246.

10.

Bowen

(1991) Isotopes and Climates. London: Elsevier.

11.

Broccoli

Manabe

(1992) The effects of orography on midlatitude northern hemisphere dry climates. Journal of Climate 5: 1181–1201.

12.

Campo

Cour

Hang

(1996) Holocene environmental changes in Bangong Co basin (Western Tibet). Part 2: The pollen record. Palaeogeography, Palaeoclimatology, Palaeoecology 120: 49–63.

13.

Cao

Wang

Zhang

et al . (2014) Holocene vegetation history, precipitation changes and Indian summer monsoon evolution documented from sediments of Xingyun lake, south-west China. Journal of Quaternary Science 29: 661–674.

14.

Chen

(2010) Paleoclimate study based on high-resolution sediment records since Holocene in Gahai Lake. Doctoral Dissertation, Guangzhou University (in Chinese).

15.

Chen

Feng

et al . (2017) Potential forcing mechanisms of Holocene lake-level changes at Nam Co, Tibetan Plateau: Inferred from the stable isotopic composition of shells of the gastropod Radix. The Holocene 27: 594–604.

16.

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.

17.

Dowsett

Thompson

Barron

et al . (1994) Joint investigations of the Middle Pliocene climate I: PRISM paleoenvironmental reconstructions. Global & Planetary Change 9: 169–195.

18.

Ehleringer

Björkman

(1977) Quantum yields for CO2 uptake in C3 and C4 plants dependence on temperature, CO2, and O2 concentration. Plant Physiology 59: 86–90.

19.

Ehleringer

Cerling

Helliker

(1997) C4 photosynthesis, atmospheric CO2, and climate. Oecologia 112: 285–299.

20.

Fan

Gasse

Huc

et al . (1996) Holocene environmental changes in Bangong Co basin (Western Tibet). Part 3: Biogenic remains. Palaeogeography, Palaeoclimatology, Palaeoecology 120: 65–78.

21.

Feng

Wang

(2006) Holocene climatic and environmental changes in the arid and semi-arid areas of China: A review. The Holocene 16: 119–130.

22.

Fleitmann

Burns

Mudelsee

et al . (2003) Holocene forcing of the Indian Monsoon recorded in a stalagmite from Southern Oman. Science 300: 1737–1739.

23.

Flückiger

Dallenbach

Blunier

et al . (1999) Variations in atmospheric N2O concentration during abrupt climatic changes. Science 285: 227–230.

24.

Flückiger

Monnin

Stauffer

et al . (2002) High-resolution Holocene N2O ice core record and its relationship with CH4 and CO2. Global Biogeochemical Cycles 16: 10-1–10-8.

25.

Gignac

Gauthier

Rochefort

et al . (2004) Distribution and habitat niches of 37 peatland Cyperaceae species across a broad geographic range in Canada. Revue Canadienne De Botanique 82: 1292–1313.

26.

Gimpelhinteregger

(2008) Environmental liability and ecological damage in European law. Journal of Environmental Law 21: 377–380.

27.

Guo

(2012) Holocene climate change documented by Lake sediments from Lake Gahai in the Monsoonal Margin Region, Northwest China. Doctoral Dissertation, Lanzhou University (in Chinese).

28.

Hansen

Isaksen

Benestad

et al . (2014) Warmer and wetter winters: Characteristics and implications of an extreme weather event in the high arctic. Environmental Research Letters 9: 114021–114030.

29.

Hastenrath

Greischar

(1993) Further work on the prediction of northeast Brazil rainfall anomalies. Journal of Climate 6: 743–758.

30.

Herzschuh

(2006) Palaeo-moisture evolution in monsoonal Central Asia during the last 50000 years. Quaternary Science Reviews 25: 163–178.

31.

Herzschuh

Birks

HJB

Jian

et al . (2010) Holocene land-cover changes on the Tibetan plateau. The Holocene 20: 91–104.

32.

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.

33.

Heuer

Overpeck

(2003) Drought: A paleoperspective-paleoclimatology and drought. Available at: https://www.ncdc.noaa.gov/paleo/drought/drghthome.html (accessed 12 November 2003).

34.

Hostetler

Bartlein

(1990) Simulation of lake evaporation with application to modeling lake level variations of Harney-Malheur Lake, Oregon. Water Resources Research 26: 2603–2612.

35.

Wang

Appel

et al . (1998) The environmental magnetic mechanism of magnetic susceptibility changes in Hulun lake sediment. Science China-Earth Sciences 4: 334–339.

36.

Jiang

Liu

et al . (2007) A high-resolution climatic change since Holocene inferred from multi-proxy of lake sediment in westerly area of China. Science Bulletin 52: 1970–1979.

37.

Jiang

Piperno

(1999) Environmental and archaeological implications of a late Quaternary palynological sequence, Poyang Lake, southern China. Quaternary Research 52: 250–258.

38.

Jin

Chen

Morrill

et al . (2011) Causes of early Holocene desertification in arid central Asia. Climate Dynamics 38: 1–15.

39.

Kleinherenbrink

Ditmar

Lindenbergh

(2014) Retracking cryosat data in the sarin mode and robust lake level extraction. Remote Sensing of Environment 152: 38–50.

40.

Kramer

Herzschuh

Mischke

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

41.

Kramer

Herzschuh

Mischke

et al . (2010b) Late Quaternary environmental history of the south-eastern Tibetan Plateau inferred from the Lake Naleng non-pollen palynomorph record. Vegetation History and Archaeobotany 19: 453–468.

42.

Krause

Whitlock

(2013) Climate and vegetation change during the late-Glacial/early-Holocene transition inferred from multiple proxy records from Blacktail pond, Yellowstone national park, USA. Quaternary Research 79: 391–402.

43.

Kreyling

Beierkuhnlein

Elmer

et al . (2008) Soil biotic processes remain remarkably stable after 100-year extreme weather events in experimental grassland and heath. Plant and Soil 308: 175–188.

44.

Krishnamurthy

Bhattacharya

Kusumgar

(1986) Palaeoclimatic changes deduced from 13C/12C and C/N ratios of Karewa lake sediments, India. Nature 48: 150–152.

45.

Leuschner

Sirocko

(2003) Orbital insolation forcing of the Indian Monsoon – a motor for global climate changes? Palaeogeography, Palaeoclimatology, Palaeoecology 197: 83–95.

46.

Chambers

Yang

et al . (2017) Records of east Asian monsoon activities in northeastern China since 15.6 ka, based on grain size analysis of peaty sediments in the Changbai mountains. Quaternary International 447: 158–169.

47.

Bernd

Xia

et al . (2009a) A preliminary study of the Holocene lake level changes and their causes derived from the sediment record of Zigetang Lake, Tibetan Plateau. Earth Science Frontiers 16: 162–167.

48.

Han

Zhen

et al . (1985) Forest in Tibet. Beijing: Science Press, pp. 126–131 (in Chinese).

49.

Chen

Zhang

et al . (1999) The characteristics of carbon isotope composition of modern plants over Qinghai-Tibet Plateau (NE) and its climatic information. Acta Sedimentologica Sinica 2: 325–329 (in Chinese).

50.

Morrill

(2010) Multiple factors causing Holocene lake-level change in monsoonal and arid central Asia as identified by model experiments. Climate Dynamics 35: 1119–1132.

51.

(2016) Asynchronous Holocene Asian monsoon vapor transport and precipitation. Palaeogeography, Palaeoclimatology, Palaeoecology 461: 195–200.

52.

Wang

Cheng

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

53.

Wang

et al . (2012) Holocene climate cycles in north-west margin of Asian monsoon. Chinese Geographical Science 22: 450–461.

54.

Zhou

et al . (2013) Relationship between pollen assemblages and organic geochemical proxies in the Zhuyeze palaeo-lacustrine basin sediments and its response to climate change. Journal of Desert Research 33: 87–93 (in Chinese).

55.

Liu

et al . (2005) Bulk organic carbon isotopic record of Huguangyan Maar Lake, Southeastern China and its paleoclimatic and paleoenvironmental significance since 62 ka BP. Marine Geology & Quaternary Geology 25: 115–126 (in Chinese).

56.

Liu

Lai

Zeng

et al . (2013) Holocene lake level variations on the Qinghai-Tibetan Plateau. International Journal of Earth Sciences 102: 2007–2016.

57.

Liu

Herzschuh

Shen

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

58.

Liu

(2016) The response of lake records to the circulation system and climate zones in china since the last glacial maximum. Acta Geographica Sinica 71: 1898–1910 (in Chinese).

59.

Liu

Zhang

Liu

et al . (2008b) Vegetation and environment history of Erlongwan Maar Lake during the late pleistocene on pollen record. Acta Micropalaeontologica Sinica 25: 274–280 (in Chinese).

60.

Lohmann

Sausen

Bengtsson

et al . (1993) The Köppen climate classification as a diagnostic tool for general circulation models. Climate Research 3: 177–193.

61.

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.

62.

Zhu

Zheng

et al . (2008) High-resolution geochemistry records of climate changes since late-glacial from Dajiuhu peat in Shennongjia Mountains, Central China. Science Bulletin 53: 28–39 (in Chinese).

63.

Huang

Wei

et al . (2004) Organic carbon isotope records of the Poyang Lake sediments and their implications for the paleoclimate during the last 8 ka. Geochimica 33: 279–285 (in Chinese).

64.

Meyers

(1994) Preservation of source identification of sedimentary organic matter during and after deposition. Chemical Geology 144: 289–302.

65.

Meyers

(1997) Organic geochemical proxies of paleoceanographic, paleolimnologic, and paleoclimatic processes. Organic Geochemistry 27: 213–250.

66.

Meyers

Ishiwatari

(2008) Organic matter accumulation records in lake sediments physics and chemistry of lakes. In: Lerman

Imboden

Gat

(eds) Physics and Chemistry of Lakes. 2nd Edition. Berlin: Springer Verlag, pp. 279–328.

67.

Meyers

Vergés

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

68.

Monnin

Steig

Siegenthaler

et al . (2004) Evidence for substantial accumulation rate variability in Antarctica during the Holocene, through synchronization of CO2, in the Taylor Dome, Dome C and DML ice cores. Earth & Planetary Science Letters 224: 45–54.

69.

Naipal

Reick

Pongratz

et al . (2015) Improving the global applicability of the rusle model-adjustment of the topographical and rainfall erosivity factors. Geoscientific Model Development Discussions 8: 2991–3035.

70.

Numaguti

(1999) Origin and recycling processes of precipitating water over the Eurasian continent: Experiments using an atmospheric general circulation model. Journal of Geophysical Research 104: 1957–1972.

71.

O’Leary

(1988) Carbon isotopes in photosynthesis. Bioscience 38: 328–336.

72.

Parmesan

(2006) Ecological and evolutionary responses to recent climate change. Ecology, Evolution, and Systematics 37: 275–290.

73.

Peel

Finlayson

Mcmahon

(2007) Updated world map of the Köppen-Geiger climate classification. Hydrology & Earth System Sciences 4: 439–473.

74.

Peltier

(2004) Global glacial isostasy and the surface of the ice-age Earth: The ICE-5G (VM2) model and Grace. Annual Review of Earth and Planetary Sciences 32: 111–149.

75.

Plewa

Meggers

Kasten

(2006) Barium in sediments off northwest Africa: A tracer for paleoproductivity or meltwater events? Paleoceanography 21: 285–293.

76.

Qian

Ding

et al . (2009) An overview of dry-wet climate variability among monsoon-westerly regions and the monsoon northernmost marginal active zone in China. Advances in Atmospheric Sciences 26: 630–641.

77.

Quan

Zhu

et al . (2011) Pollen-inferred climate changes and vertical shifts of alpine vegetation belts on the northern slope of the Nyainqentanglha Mountains (central Tibetan Plateau) since 8.4 kyr BP. The Holocene 21: 939–950.

78.

Riedinger

Steinitzkannan

Last

et al . (2002) A 6100 ^14c yr record of EI Nino activity from the Galapagos Islands. Journal of Paleolimnology 27: 1–7.

79.

Rodwell

Hoskins

(1996) Monsoons and the dynamics of deserts. Quarterly Journal of the Royal Meteorological Society 122: 1385–1404.

80.

Ruddiman

(2006) What is the timing of orbital-scale monsoon changes? Quaternary Science Reviews 25: 657–658.

81.

Schnitzer

(1978) Humic substances: Chemistry and reactions. In: Schnitzer

Khan

(eds) Soil Organic Matter. New York: Elsevier, pp. l–64.

82.

Shen

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

83.

Shen

Jones

Yang

et al . (2006) The Holocene vegetation history of Lake Erhai, Yunnan province southwestern China the role of climate and human forcings. The Holocene 16: 265–276.

84.

Shen

Liu

Ryo

et al . (2004) A high-resolution climatic change since the Late Glacial Age inferred from multi-proxy of sediments in Qinghai Lake. Science China-earth Sciences 34: 582–589.

85.

Shen

Liu

Wang

et al . (2005a) Palaeoclimatic changes in the Qinghai Lake area during the last 18,000 years. Quaternary International 136: 131–140.

86.

Shen

Yang

et al . (2005b) Lake sediment records on climate change and human activities since the Holocene in Erhai catchment, Yunnan Province, China. Science China-earth Sciences 48: 353–363.

87.

Street-Perrott

Harrison

(1985) Lake levels and climate reconstruction. In: Hecht

(ed.) Paleoclimate Analysis and Modeling. New York: Wiley, pp. 291–340.

88.

Street-Perrott

Roberts

(1983) Fluctuations in closed lakes as an indicator of past atmospheric circulation patterns. In: Street-Perrott

Beran

Ratcliffe

RAS

(eds) Variations in the Global Water Budget. Reidel: Dordrecht, pp. 331–345.

89.

Stuiver

Yang

Denton

(1975) Climate versus changes in 13C content of the organic component of lake sediments during the late Quaternary. Quaternary Research 5: 251–262.

90.

Sun

Wen

(1994) Paleovegetation and paleoenvironment of Manas Lake, Xinjiang, N.W. China during the last 14000 years. Quaternary Science 1994: 239–248 (in Chinese).

91.

Tao

Chen

et al . (2010) Holocene vegetation changes interpreted from pollen records in Balikun Lake, Xinjiang, China. Acta Palaeontologica Sinica 48: 194–199 (in Chinese).

92.

Tian

et al . (2008) Pollen assemblage characteristics of lakes in the monsoon fringe area of China. Chinese Science Bulletin 21: 3354–3363.

93.

Tieszen

Boutton

(1989) Stable carbon isotopes in terrestrial ecosystem research. In: Rundel

Ehleringer

Nagy

(eds) Stable Isotopes in Ecological Research. New York: Springer Verlag, pp. 167–195.

94.

Wagner

Melles

Hahne

et al . (2000) Holocene climate history of geographical society Ø, east Greenland – Evidence from lake sediments. Palaeogeography, Palaeoclimatology, Palaeoecology 160: 45–68.

95.

Wang

(1995) Sedimentology of late Quaternary lacustrine deposits and history of lake level fluctuation in Hulun Lake. Journal of Lake Sciences 7: 334–339.

96.

Wang

Liu

et al . (2007) The early Holocene optimum inferred from a high-resolution pollen record of Huguangyan Maar Lake in southern China. Science Bulletin 52: 1285–1291 (in Chinese).

97.

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.

98.

Shen

(2010a) Paleoclimate evolution since 277 kaBP reflected by grain size variation of a sediment core from Lake Xingkai, northeastern Asia. Journal of Lake Sciences 22: 110–118 (in Chinese).

99.

Shen

(2010b) Paleoenviromental and paleoclimatic changes in Lake Xingkai inferred from stable carbon and nitrogen isotopes of bulk organic matter since 28 ka BP. Acta Sedimentologica Sinica 28: 365–372 (in Chinese).

100.

(2011) Lake evolution and climatic and hydrological changes in Arid zone of Xingjiang. Marine Geology & Quaternary Geology 31: 135–143.

101.

Xiao

et al . (2006) Holocene climate changes in the monsoon/arid transition reflected by carbon concentration in Daihai Lake of Inner Mongolia. The Holocene 16: 551–560.

102.

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.

103.

Xiao

Zhou

et al . (2007) Sporopollen vegetation and environmental evolution since last glacial maximum in Dahu swamp, Jiangxi. Science China-Earth Sciences 37: 789–797 (in Chinese).

104.

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.

105.

Xue

Zhong

(2011) Holocene climate variation denoted by Barkol lake sediments in northeastern Xinjiang and its possible linkage to the high and low latitude climates. Science China-Earth Sciences 54: 603–614.

106.

Yang

(2016) Climate and environment changes recorded by sediments from Bangong Co in Tibet since last deglaciation. Doctoral Dissertation, Lanzhou University (in Chinese).

107.

Gao

(1979) Meteorology of the Qinghai-Xizang Plateau. Beijing: Science Press.

108.

You

Liu

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

109.

Zhang

Feng

Yang

et al . (2010) Holocene environmental variations recorded by organic-related and carbonate-related proxies of the lacustrine sediments from Bosten lake, northwestern China. The Holocene 2: 363–373.

110.

Zhang

Zheng

Prokopenko

et al . (2007) The palaeoenvironmental variation from the high-resolution record of the Holocene sediment carbonate and isotopic composition in Bosten Lake and responding to glacial activity. Acta Geologica Sinica 81: 1658–1671 (in Chinese).

111.

Zhang

Xue

(2016) The lake status change of China since the late Quaternary and its significance for palaeoenvironment. Quaternary Sciences 36: 598–611 (in Chinese).

112.

Zhang

Ming

Shi

et al . (2014) Lake sediment records on climate change and human activities in the Xingyun lake catchment, SW China. PLoS ONE 9: e102167.

113.

Zhao

(2012) Vegetation response to Holocene climate change in east Asian monsoon-margin region. Earth-Science Reviews 113: 1–10.

114.

Zhao

(2011) 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.

115.

Zhao

Chen

Zhou

et al . (2010) Vegetation history, climate change and human activities over the last 6200 years on the Liupan Mountains in the southwestern Loess Plateau in central China. Palaeogeography, Palaeoclimatology, Palaeoecology 293: 197–205.

116.

Zheng

Hucai

Qingzhong

et al . (2014a) Vegetational and environmental changes since 15 ka BP recorded by Lake Lugu in the southwest monsoon domain region. Doctoral Dissertation, Yunnan Normal University (in Chinese).

117.

Zheng

Zhang

Ming

et al . (2014b) Vegetational and environmental changes since 15 ka BP recorded by Lake Lugu in the southwest monsoon domain. Quaternary Sciences 34: 1314–1326 (in Chinese).

118.

Zhong

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.

119.

Zhou

Shen

(2007) Eco-Environment Evolution in West China. Beijing: Science Press (in Chinese).

120.

Zhou

Zhang

et al . (2013) The response of organic geochemical proxies in Holocene lake sediments to millennial-scale climate change. Journal of Salt Lake Research 21: 1–9 (in Chinese).

121.

Zhu

Zhang

et al . (2006) Pollen record from Dajiuhu Basin of shennongjia and environmental changes since 15.753 ka BP. Quaternary Sciences 26: 814–826 (in Chinese).

122.

Zhu

Wang

et al . (2008) Environmental changes since 8.4 ka reflected in the lacustrine core sediments from Nam Co, central Tibetan Plateau, China. The Holocene 18: 831–839.

123.

Zhu

Zhen

Wang

et al . (2009) A ~30,000-year record of environmental changes inferred from lake Chen Co, southern Tibet. Journal of Paleolimnology 42: 343–358.