Four peat humification-recorded Holocene hydroclimate changes in the southern Altai Mountains of China

Abstract

In this study, Holocene humification records were derived from four ombrotrophic peatlands to investigate the millennial-scale and decadal-centennial-scale variability of peat decomposition and to explore the paleoclimatic and paleohydrological implications. After eliminating the species-specific and site-specific noise and removing the time-dependency effect for each of four peat sequences, the averaged residuals of four peat humification sequences exhibit two-order variations. The millennial-scale variation of averaged residuals of four peat humification is characterized by a bow-shaped curve and is basically moisture-dependent. In detail, the humification in the middle Holocene (~8200–~4000 cal. yr BP) was higher than in the early Holocene (before ~8200 cal. yr BP) and also than in the late Holocene (after ~4000 cal. yr BP). The decadal/centennial-scale variations are superimposed on the bow-shaped curve and have been primarily paced by the sea surface temperature in the North Atlantic whose signals were propagated to Central Asia via the prevailing westerlies, which implies that lower temperature and lower temperature-suppressed evaporation (i.e. elevated moisture level) were most likely responsible for limiting the decomposition activity in the uppermost peat layer. These results indicate the potential for humification records from ombrotrophic peatlands in Central Asia to elucidate paleoclimate variability.

Keywords

Altai Mountains Central Asia Holocene climate ombrotrophic peat peat humification

Introduction

Studies have demonstrated that a higher degree of peat humification indicates a lower moisture condition because plant residues are decayed more intensively by microbial activities in an aerobic environment of the uppermost peat layer during a drier period (Blackford and Chambers, 1993; Castro et al., 2015; Chambers et al., 1997, 2011; Hughes et al., 2012; Payne and Blackford, 2008). In other words, the uppermost peat layer experiences water table fluctuations or soil moisture fluctuations that were in turn determined by the balances between precipitation and evapotranspiration, and organic decay or decomposition proceeds more effectively when the balance between precipitation and evapotranspiration or effective precipitation (i.e. precipitation minus evapotranspiration) is lower under a drier condition (Aaby and Tauber, 2008; Barber, 1981; Blackford and Chambers, 1993; Caseldine et al., 2000; Clarkson et al., 2004; McGlone, 2009; Morgan et al., 2005). Based on this important premise, peat humification has been widely used as a proxy for the degree of peat decomposition in paleoclimatic and paleohydrological reconstructions (Borgmark and Wastegård, 2008; Castro et al., 2015; Huang et al., 2013).

Although this is a sound conceptual basis, questions have been raised about the general applicability of humification as a paleoclimatic or paleohydrological proxy because the humification-based peat-surface wetness reconstructions have been challenged by other proxies-based peat moisture records (Amesbury et al., 2012; Chambers et al., 2011; Hughes et al., 2012; Newnham et al., 2019; Yeloff and Mauquoy, 2006). The inconsistencies between humification and other proxies are most likely attributable to two reasons: (1) site-specific changes in peat botanical composition (Chambers et al., 1997; Hughes et al., 2012; Payne and Blackford, 2008; Zhang et al., 2017) and (2) locality-specific variations in topography and geochemistry of peat (Caseldine et al., 2000; Morgan et al., 2005). Newnham et al. (2019) also challenged the use of humification and found that stronger longer-term trend of humification is attributed to slow anaerobic decay over time because the trend toward wetter summers in the late Holocene cannot be corroborated by independent climate proxies.

Peat humification-climate researches have been mostly undertaken in North Europe (Amesbury et al., 2012; Borgmark and Wastegård, 2008; Vorren et al., 2012), East Asia (Hong et al., 2001; Ma et al., 2008; Wang et al., 2004; Xiao et al., 2017), North America (Payne and Blackford, 2008), Australia and New Zealand (Burrows et al., 2014; Jara et al., 2017; Newnham et al., 2019; Viau and Gajewski, 2009; Wilmshurst et al., 2002). In Central Asia, the work of peatland research gained more international recognition in recent decades due to several high-resolution alpine ombrotrophic peatlands (e.g. Kelashazi, Narenxia, Ganhaizi, Yushenkule, and Tuolehaite) in the southern Altai Mountains. A variety of proxies have been used to reconstruct the regional paleoclimatic changes (e.g. Feng et al., 2017; Rao et al., 2019, 2020; Xiao et al., 2017; Yang et al., 2019; Zhang et al., 2018a, 2018b) and the related proxies mainly include pollen assemblages (Feng et al., 2017; Wang and Zhang, 2019; ; Zhang et al., 2020a, 2020b, 2020c), stable isotopes of plant macrofossils (Rao et al., 2019, 2020; Yang et al., 2019; Zhang et al., 2018c) and n-alkane compositions (Wu et al., 2020; Zhang et al., 2016). However, the importance and implication of peat humification in the southern Altai Mountains are not in-depth investigated.

In this study, we present four Holocene humification records with reliable AMS ¹⁴C dates from four ombrotrophic peatlands (Narenxia, Tuolehaite, Ganhaizi, and Kelashazi) in the southern Altai Mountains of China. We test the feasibility of humification records in the southern Altai Mountains for extracting a regional signal at decadal-centennial and millennial scales by eliminating the site-specific and locality-specific noise and removing the time-dependency effect via the widely applied method in other regions (Borgmark and Wastegård, 2008; Newnham et al., 2019). We then investigate the coherent humification patterns among four sites via between-site comparison with other paleoclimate records to determine its climatic implication in the southern Altai Mountains.

Regional settings

The southern Altai Mountains (Figure 1) are bioclimatically the southernmost extension of Siberian taiga forests and are the only taiga-dominated area of China (Chen, 2010). The regional climate is mainly influenced by the westerlies throughout the year with the Siberian High being rather important in winters (Aizen et al., 2001; Meeker and Mayewski, 2002). The westerlies transport water vapor along the Irtysh Valley to the southern Altai Mountains and precipitation forms when the air is forced to rise along the wind-facing slopes. The precipitation is ~200–~300 mm in low elevations, ~300–~500 mm in middle elevations and ~500–~600 mm in high elevations (Wang and Zhang, 2019). The distribution of monthly precipitation during 1970–2015 is characterized by two peaks: one at April–September (50%–68%) and another at November–December (13%–21%) (Figure 1e; Li et al., 2020a). The abundant precipitation, together with topographic depressions, provides suitable settings for peat development. According to a recent field investigation (Nurbayev et al., 2008), the peatlands cover about 108,811.8 hm² in the areal extent, and they distribute in the elevations ranging from 1700 to 2500 m a.s.l. Precipitation and meltwater from snow and glacier are main water resources for growth of peat plants (Zhang et al., 2016). The peat plants are composed of Carex (mainly include Carex altaica, Carex pamirensis and Carex lasiocarpa) and Moss (mainly include Herba Sphagni, Sphagnum nemoreum and Polytrichum commune) (Zhang et al., 2016). Four ombrotrophic peatlands (Narenxia, Tuolehaite, Ganhaizi and Kelashazi) in the southern Altai Mountains are investigated in this study.

Figure 1.

Regional setting of the southern Altai Mountains within China and the studied sites. (a) Narenxia (NRX) Peatland. (b) Tuolehaite (TLHT) Peatland. (c) Ganhaizi (GHZ) Peatland and (d) Kelashazi (KLSZ) Peatland. The diagram (e) is mean monthly temperature and mean monthly precipitation (1958–2017) in the southern Altai Mountains (Li et al., 2020a).

Narenxia (NRX, also called Tielishahan) Peatland (48.80°N, 86.90°E, 1760 m a.s.l.) (site a in Figure 1) is located in an intermontane depression at the Kanas Basin, northwestern portion of the southern Altai Mountains. The modern biomes in the surroundings are dominated by coniferous forests and meadow steppes that distribute in the elevation belt between 1500 and 2600 m, and the coniferous forests are mainly composed of Larix sibirica and Picea obovata (Zhang et al., 2016). The peat-based Holocene vegetation and climate histories were inferred from pollen, isotope and n-alkane data (Feng et al., 2017; Zhang et al., 2016, 2018a, 2018c).

Tuolehaite (TLHT) Peatland (48.44°N, 87.54°E, 1700 m a.s.l.) (site b in Figure 1) is a small alpine peatland in the middle part of the southern Altai Mountains. The modern biomes in the surroundings are also coniferous forests and meadow steppes, and the coniferous forests are mainly composed of Larix sibirica and Pinus sibirica. The Holocene vegetation variations and climate changes were inferred from pollen records (Li et al., 2020b; Zhang et al., 2020b).

Ganhaizi (GHZ) Peatland (87.56°E, 48.41°N, 1926 m a.s.l.) (site c in Figure 1) is situated ~5 km away from TLHT Peatland. The vegetation around the GHZ Peatland is basically alpine meadow steppes that are well embedded with large patches of Larix sibirica and Picea obovata forests in higher elevations and also in the north-facing and west-facing slopes. The Holocene climate history was inferred from n-alkane data (Ran et al., unpublished data).

Kelashazi (KLSZ, also called Halasazi) Peatland (48.12°N, 88.37°E, 2450 m a.s.l.) (site d in Figure 1) is located within the alpine meadow zone almost touching the upper tree line and has a large amount of frost mounds (typical domed peat mounds with average height about 3–4 m) (Zhang et al., 2018b). The area around KLSZ valley is naturally landscaped with alpine meadows at 2800–2370 m a.s.l. The Holocene vegetation variations and associated climate changes were inferred from pollen, isotopes and brGDGTs records (Rao et al., 2019, 2020; Wang and Zhang, 2019; Wu et al., 2020; Zhang et al., 2020c).

Material and methods

Fieldwork and chronology

All cores were extracted using a Holland-made peat corer at four sites in the southern Altai Mountains (see Figure 1a–d). The depth of the core is 400 cm in NRX Peatland, 380 cm in TLHT Peatland and 600 cm in GHZ Peatland (Figures 2 –5). Two cores were collected in KLSZ Peatland (Figure 5): KLSZ-1 is 550 cm and KLSZ-2 is 100 cm. NRX core was sliced at 1-cm interval and other cores were sliced at 2-cm interval immediately after the core was retrieved in the field. Total 44 radiocarbon ages (Figures 2 –5) were obtained from these four sites and these ages were already published (Feng et al., 2017; Wang and Zhang, 2019; Zhang et al., 2016, 2020b, 2020c). The related age-depth models were established by invoking the Bayesian age-depth model package Bacon 2.2 (Blaauw and Christen, 2011) using the mathematics software “R” version 3.4.2. All dates were calibrated to calendar years before present (cal. yr BP = years before 1950 AD) using the Intcal13 calibration curve (Reimer et al., 2013). All sliced samples were taken to the laboratory and refrigerated at 4°C to prepare for analysis of humification and loss on ignition (LOI).

Figure 2.

Information of Narenxia peat core: photo, ¹⁴C age, lithology. (a) Peat growth rate (mm yr⁻¹). (b) LOI (%). (c) Original light absorbance (fine line) and LOI-corrected light absorbance (coarse line). (d) Three-point smoothed curve and the linear trend (i.e. time-dependencies). (e) Residuals (differences between the three-point smoothed data and the linearly detrended data).

Figure 3.

Information of Tuolehaite peat core: photo, ¹⁴C age, lithology. (a) Peat growth rate (mm yr⁻¹). (b) LOI (%). (c) Original light absorbance (fine line) and LOI-corrected light absorbance (coarse line). (d) Three-point smoothed curve and the linear trend (i.e. time-dependencies). (e) Residuals (differences between the three-point smoothed data and the linearly detrended data).

Figure 4.

Information of Ganhaizi peat core: photo, ¹⁴C age, lithology. (a) Peat growth rate (mm yr⁻¹). (b) LOI (%). (c) Original light absorbance (fine line) and LOI-corrected light absorbance (coarse line). (d) Three-point smoothed curve and the linear trend (i.e. time-dependencies). (e) Residuals (differences between the 3-point smoothed data and the linearly detrended data).

Figure 5.

Information of Kelashazi (KLSZ-1 and KLSZ-2) peat cores: photo, ¹⁴C age, lithology. (a) Peat growth rate (mm yr⁻¹). (b) LOI (%). (c) Original light absorbance (fine line) and LOI-corrected light absorbance (coarse line). (d) Three-point smoothed curve and the linear trend (i.e. time-dependencies). (e) Residuals (differences between the 3-point smoothed data and the linearly detrended data).

Methods

loss on ignition (LOI)

All samples were oven-dried for 12 h. The dried samples were incinerated at 550°C to derive the percentages of loss on ignition (LOI). For each sample, three repeat measurements were taken and averaged. The related results were applied to correct the mineral content and to estimate a cut-off point for inclusion in humification analyses (Gehrels et al., 2006).

humification analyses

Humification measurement was conducted using the commonly used alkali-extraction method (Blackford and Chambers, 1993). First, dried peat sample (0.1 g) was dissolved in 50 ml of 8% NaOH and simmered for 1 h. Second, the simmered sample was filtered to get a solution after cooling and then deionized water was added to the solution to reach a final volume of 100 ml. The light absorbance measurement was done at a light wavelength of 540 nm by an Agilent Technologies Cary 60 UV-Vis spectrophotometer after the 100-ml solution was further mixed for 3 h. To ensure the precision and the consistency of the measurements, the solution was divided into three parts and each of three parts was measured for three times. The measured value was expressed as the percentage of light absorbance, which is proportional to the amount of dissolved humic matter or the degree of humification (Aaby and Tauber, 2008). That is, a lower percentage of light absorbance denotes a lower degree of humification and a higher percentage shows a higher degree of humification.

The relationship between light absorbance and humification can be distorted by minerogenic constituents. It was thus necessary to correct light absorbance values for any distortion caused by minerogenic constituents. A simple linear correction can be made based on the minerogenic content determined by LOI (Chambers et al., 2011; Zhang et al., 2017). To remove aberrant data and “random errors,” the minerogenically-corrected humification data were smoothed via a three-point moving-window smoothing method. Furthermore, another correction also needs attentions, that is, the correction for time-dependency of humification. In other words, humification will definitely increase with time if all other factors are constants (Clymo, 1984). To remove the time-dependency effect, the three-point moving-window smoothed data of each humification sequence were detrended by applying a linear regression line (Borgmark and Wastegård, 2008; Newnham et al., 2019), and the residual values or the differences between the three-point moving-window smoothed data and the linearly detrended data are considered to be indicative of the degree of decomposition (Burrows et al., 2014; Newnham et al., 2019).

In order to depict a regional picture of humification-indicated moisture variations, we averaged the individual humification sequences using a region-averaging method. Specifically, we assigned an ordinal value to each 100-year time slice in a particular humification sequence. The ordinal value that ranges from −4 (the lowest) to 4 (the highest) is the ratio between the average value of 100-year time slice considered and the highest 100-year average of the entire humification sequence. The average of the assigned ordinal values for each 100-year time slices from all four sequences of this study gives the regionally-averaged humification. To show the consistency among the participating humification sequences, mean errors (i.e. error bars) are defined as the sum of differences between the regionally-averaged humification and each individual humification divided by the number of the participating sequences. This region-averaging method does smooth out the details of individual sequences, but it produces a regionally representative picture if all of the participating sequences in a particular region exhibit a more or less geographically coherent pattern (Davis et al., 2003; ; Zhao et al., 2009).

Results

Narenxia (NRX) peatland

The NRX core comprises dark brown peat throughout the whole depth with a mean peat growth rate (PGR) of 0.46 mm/yr (Figure 2). The PGR can be classified into five parts: 0.20 mm/yr at 400–325 cm, 0.44 mm/yr at 324–205 cm, 0.61 mm/yr at 204–101 cm, 0.48 mm/yr at 100–20 cm and 0.63 mm/yr in the upper 20 cm (Figure 2a). LOI remains consistent at 65%–90% with a mean 87.56% (Figure 2b). Marked oscillations are observed in the light absorbance values which vary between 10% and 65% with a decreasing trend since Holocene (Figure 2c). After finishing the LOI correction, removing aberrant data via three-point moving method and time-dependency effect via a linear regression curve (Figure 2d-e), the humification residuals (i.e. differences between three-point smoothed data and linearly detrended data) show negative values at 400–323 cm (~9500–~6200 cal. yr BP), at 217–203 cm (~3900–~3600 cal. yr BP), at 167–59 cm (~3000–~1000 cal. yr BP) and at 17–1 cm (past ~200 years). Other intervals (~6200–~3900, ~3600–~3000, ~1000–~200 cal. yr BP) are featured by positive values of humification.

Tuolehaite (TLHT) peatland

The TLHT core is consisted of a light brown peat at 358–0 cm and a grayish lacustrine layer at 380–358 cm (Figure 3). The PGR can be classified into five parts with a mean 0.4 mm yr⁻¹ (Figure 3a). In details, the PGR (0.5 mm yr⁻¹) was relatively high at 358–334 cm (prior to ~9700 cal. yr BP) and was followed by the relatively low values with a mean of 0.2 mm yr⁻¹ at 332–280 cm (~9700–~7600 cal. yr BP). The PGR slightly increased (0.3 mm yr⁻¹) at 278–178 cm (~7600–~4300 cal. yr BP) and then abruptly increased to a high level with a mean of 0.6 mm yr⁻¹ at 176–134 cm (~4300–~3500 cal. yr BP). The PGR decreased rapidly at 134 cm reaching 0.25 mm yr⁻¹ and then switched to a high level (mean 0.7 mm yr⁻¹) in the upper 64 cm (past 1000 years). LOI remains consistent at 50%–95% with a mean 76.99% (Figure 3b). Marked oscillations are evident in the light absorbance values which vary between 10% and 50% away from the bottom lacustrine layer (Figure 3c). After finishing the minerogenical correction and removing the time-dependency-effect (Figure 3d), humification at TLHT Peatland (Figure 3e) experiences one lower-value interval at ~10,200−~8200 cal. yr BP (358–296 cm), one higher-value interval at ~8200–~4000 cal. yr BP (294–164 cm) and then a decreasing trend at ~4000–~1200 cal. yr BP (164–68 cm). There is an increasing trend in the past 1200 years (66–0 cm).

Ganhaizi (GHZ) peatland

The GHZ core can be divided into four units: grayish mud at 600–580 cm, dark brown peat at 580–500 cm, light brown peat at 500–50 cm and again dark brown peat in the upper 50 cm (Figure 4). The PGR (mean 0.65 mm yr⁻¹) experiences an increasing trend with a quick reducing trend in the upper 50 cm (Figure 4a). LOI remains consistent at 80%–95% with a mean 90.44% (Figure 4b). Marked oscillations are evident in the light absorbance values which vary between 20% and 50% away from the bottom lacustrine layer (Figure 4c). After finishing the minerogenical correction and removing the time-dependency-effect (Figure 4d), humification at GHZ Peatland (Figure 4e) had five negative values at 580-480 cm (~10,400−~8400 cal. yr BP), at 454–432 cm (~7800–~7400 cal. yr BP), at 372–298 cm (~6400–~5200 cal. yr BP), at 256–140 cm (~4600–~3100 cal. yr BP) and at 48–0 cm (past ~2000 years). Other four intervals (~8400–~7800, ~7400–~6400, ~5200–~4600, and ~3100–~2000 cal. yr BP) are featured by positive values of humification.

Kelashazi (KLSZ) peatland

Two cores (KLSZ-1 and KLSZ-2) were taken at KLSZ Peatland (Figure 5). KLSZ-1 core is composed of a brown peat (100–60 cm) and a dark brown peat (60–0 cm). KLSZ-2 core is composed of a light brown peat (550–420 cm) and a yellow peat (420–0 cm). Two points should be noted: (1) proxy sequences (i.e. PGR, LOI and light absorbance) in KLSZ Peatland are expressed in ages not depths; (2) data are averaged from KLSZ-1 and KLSZ-2 core for the overlapping period (i.e. ~4100–~3200 cal. yr BP). As shown in Figure 5a, PGR experiences an increasing trend from 0.4 to 0.8 mm/yr at ~11,000–~7000 cal. yr BP and then gradually reduced to 0.18 mm/yr until ~500 cal. yr BP. In the past 500 years, PGR quickly increased up to 1.20 mm/yr. LOI remains consistent at 70%–90% with a decreasing trend in the past 4000 years (Figure 5b). Marked oscillations are evident in the light absorbance values which vary between 20% and 70% with a mean 43.18% (Figure 5c). After finishing the minerogenical correction and removing the time-dependency-effect (Figure 5d), negative values against the detrending linear curve appear for the following intervals: ~11,000–~7600, ~6500–~5300, ~4100–~2900 cal. yr BP and past 2000 years; and positive values for the following intervals: ~7600–~6500, ~5300–~4100, and ~2900–~2000 cal. yr BP (Figure 5e).

Comparisons

As shown in Figures 2-5, PGR has a pattern: GHZ (0.65 mm/yr) > KLSZ (0.54 mm/yr) > NRX (0.46 mm/yr) > TLHT (0.40 mm/yr). The similar pattern is found in LOI: GHZ (90.44%) > KLSZ (87.64%) > NRX (87.46%) > TLHT (76.99%). Marked oscillations are recorded in LOI-corrected light absorbance values that vary between 10% and 60% among four peat cores. The result of mean LOI-corrected light absorbance values is KLSZ (38.25%) > NRX (31.88%) > GHZ (28.39%) > TLHT (25.34%). All four smoothed curves present more or less a similar general trend: a relatively high light absorbance before ~4000 cal. yr BP and a decreasing light absorbance trend since ~4000 cal. yr BP.

For each comparison, we assigned an ordinal value to each 100-year time slice in each humification sequence. The humification residuals of four peat cores are showed in Figure 6. Three observations can be made from these four curves (Figure 6a–d). First, two humification sequences (i.e. GHZ and KLSZ Peatland) are nearly completely consistent in showing those higher and lower values (Figure 6c and d). Second, the humificaiton sequence from TLHT Peatland (Figure 6b) reasonably well corroborates those from GHZ Peatland and KLSZ Peatland. Third, although the humificaiton sequence from NRX Peatland (Figure 6a) does not corroborate the GHZ and KLSZ Peatland sequences (Figure 6c and d) in showing those higher and lower values, it does show a high-degree consistency with the TLHT sequence (Figure 6b). That is, both sequences (Figure 6a and b) exhibit a relatively high-humification period lasting from ~8000–~7000 to ~3500–~3000 cal. yr BP.

Figure 6.

Light absorbance residuals for NRX (a), TLHT (b), GHZ (c), and KLSZ (d) Peatlands. Curve (e): the averaged light absorbance residuals of four peat cores (gray area is the standard deviations).

To depict a regional picture of humification-indicated moisture variations, we averaged the individual humification sequences using the region-averaging method. The averaged humification curve (Figure 6e) is featured by a bow shape, that is, the residual values are generally higher in the middle Holocene (~8200−~4000 cal. yr BP) than in the early Holocene (before ~8200 cal. yr BP) and also than in the late Holocene (after ~4000 cal. yr BP). Furthermore, superimposed on the bow-shaped curve are several major troughs including ones occurred at ~8600–~8400, ~7800–~7400, ~6500–~5300, ~3800–~3200, and ~1900–~1200 cal. yr BP.

Discussion

Premises of peat humification record interpretations

A higher degree of peat humification indicates a lower moisture condition because the fact that plant residues are decayed more intensively by microbial activities in an aerobic environment of the uppermost peat layer during a drier period. In other words, organic decay or decomposition proceeds more effectively when the balance between precipitation and evapotranspiration is lower under a drier condition. In the peatlands of the Altai Mountains, the uppermost peat layer is frozen and the land surface is covered by deep snow (50–250 cm) in cold season, thus inhibiting aerobic decay (Rao et al., 2020). It means that the peat decay primarily occurs in warm season (May–September). Under this kind of conditions (i.e. warm-season decay), the uppermost peat layer is subjected to the changes in moisture-related or/and temperature-modulated aeration conditions during the warm season (Barber, 1981; Charman, 2007; Newnham et al., 2019).

Millennial-scale humification-indicated moisture variations

The Holocene that began at ~11,700 cal. yr BP was divided into three stages by the international stratigraphic chronology committee: warming Greenlandian stage (~11,700–~8200 cal. yr BP), warm Northgrippian stage (~8200–~4200 cal. yr BP) and cooling Meghalayan stage (~4200–0 cal. yr BP) (Ran and Chen, 2019; Walker et al., 2012; Zanchetta et al., 2016). The three-stage division seems reasonably corroborated by the averaged light absorbance residuals of four peat cores, although the division lines at ~8200 cal. yr BP and at ~4200 cal. yr BP are not as sharp as those in the sequences used by the international stratigraphic chronology committee. Specifically, the averaged residuals of four peat cores (Figure 5e) are generally higher in the middle Holocene (~8200–~4000 cal. yr BP) than in the early Holocene (before ~8200 cal. yr BP) and than in the late Holocene (after ~4000 cal. yr BP).

Climatic implication of humification in the southern Altai Mountains were investigated through comparing with regional paleoclimate reconstruction in the same and nearby sites. The first site is TLHT Peatland where the A/C (i.e. Artemisia/Chenopodiaceae) ratio was justified to be indicative of basin-wide moisture (Zhang et al., 2020b). As shown in Figure 7b, the A/C ratio-indicated moisture level was generally lower in the middle Holocene (~8500–~4000 cal. yr BP) than in the early Holocene (before ~8500 cal. yr BP) and also than in the late Holocene (after ~4000 cal. yr BP). The second site is GHZ Peatland where the n-alkane-based P_aq (the proportion of aquatic components) was justified to be indicative of peat-surface moisture (Figure 7b, Ran et al., unpublished data). As shown in Figure 7c, the P_aq-indicated moisture level was generally lower in the middle Holocene (~9000–~4500 cal. yr BP) than in the early Holocene (before ~9000 cal. yr BP) and also than in the late Holocene (after ~4500 cal. yr BP). The third site is Kanas Lake (48.72°N, 87.02°E, 1365 m a.s.l.) where the A/C-indicated moisture variations (Figure 7d; Huang et al., 2018) are in a relatively good agreement with the three-stage variations of averaged residuals (Figure 7a). Finally, three-stage variations in averaged residuals in the southern Altai Mountains (Figure 7a) are reasonably corroborated by the synthesized aridity index from four pollen sequences including Achit Lake, Bayan Nuur, Uggi Lake and Telmen Lake in the adjacent western Mongolia (Figure 7f; Zhang and Feng, 2018). The synthesized aridity index curve shows that the moisture level was definitely lower in the middle Holocene (~8500−~4000 cal. yr BP) than in the early Holocene (before ~8500 cal. yr BP) and also than in the late Holocene (after ~4000 cal. yr BP). Therefore, the light absorbance residuals in the southern Altai Mountains (Figure 7a) are well correspondent with well-dated Holocene moisture sequences from same/nearby sites, which demonstrates peat humification in the southern Altai Mountains can be considered as a reliable proxy of moisture.

Figure 7.

Comparison between averaged residuals and Holocene climate records in the southern Altai Mountains and the surrounding areas. (a) Averaged light absorbance residuals of four peat cores (NRX, TLHT, GHZ and KLSZ; this study). (b) A/C-indicated moisture in TLHT Peatland (Zhang et al., 2020b). (c) P_aq-indicated moisture in GHZ Peatland (Ran et al., unpublished data). (d) A/C-indicated moisture in Kanas Lake (Huang et al., 2018). (e) Synthesized aridity index in western Mongolia (Zhang and Feng, 2018). (f) ACL (average chain length)-indicated temperature in GHZ Peatland (Ran et al., unpublished data). (g) Temperature anomaly in the Northern Hemisphere (Marcott et al., 2013).

The ACL-recorded temperature in GHZ Peatland displays a straight warming trend before ~8500 cal. yr BP and a general cooling trend after ~8500 cal. yr BP (Figure 7f; Ran et al., unpublished data) and the ACL-recorded temperature variations are basically consistent with the synthesized temperature anomaly for the Northern Hemisphere (Figure 7g; Marcott et al., 2013). According to the three-stage division of the averaged light absorbance residuals and the collaborative moisture proxy data from nearby sites (Figure 7a–e), it can be summarized that the early Holocene (before ~8200 cal. yr BP) was warm and wet, the middle Holocene (between ~8200 and ~4000 cal. yr BP) warm and dry and the late Holocene (after ~4000 cal. yr BP) cool and wet. It means that the warm-dry condition in the middle Holocene (~8200–~4000 cal. yr BP) was most likely responsible for the elevated humification through more effective decomposition in an aerobic environment. It also means that wet conditions in the early Holocene (warm-wet) and in the late Holocene (cool-wet) prohibit peat decay in an anaerobic environment. We therefore conclude that this long-term trend signalling decreasing humification in peat sediments can be attributed to regional moisture variability, being different with the trend of humification in New Zealand which is more indicative of long-term decay of peat (Newnham et al., 2019).

Decadal-centennial humification-indicated humification variations

We turned to the light absorbance residuals and found numerous decadal- centennial-scale phases in all four records (Figure 6). The obvious differences of peat humification records are existed in four hydrologically separate sites, such as, the lasting 800-year (~1000–~200 cal. yr BP) higher-humification interval is observed in NRX Peatland, whereas no obvious increasing humification is found in other three cores. The decadal-centennial-scale phases represent shorter-term variability in summer precipitation-evaporation balance along with other and local site factors. Two factors are most likely responsible for these different shorter-term variabilities: (1) vegetation composition in the core over time and (2) specific characteristics of these peats. Marked changes in vegetation composition over time have been reported previously from plant macrofossil analyses at NRX core (Sun, 2012) and KLSZ core (Rao et al., 2019). The peat surfaces exhibit patterns of moist swales and intervening drier hummocks (Newnham et al., 2019) and it has been suggested that these features may migrate across the peat surface over time as part of a natural process of growth dynamics and hence independently of climate variability (McGlone, 2009). This process would likely cause variation in the degree of evapotranspiration and hence peat surface moisture experienced between hummocks and swales which would change as these topographic features migrated across the core sites.

Nevertheless, several consistent patterns are recorded among four different humification records which points to regional climate signal at times outweigh these local factors. Several pronounced decadal-centennial-scale low-humification intervals are superimposed on the bow-shaped curve (Figure 8a), which means strong decadal/centennial scale variability in peat surface moisture was a prevalent feature of Holocene climate. The pronounced low-humification intervals definitely occurred in following intervals: ~8600–~8400, ~7800–~7400, ~6500–~5300, ~3800–~3200, and ~1900–~1200 cal. yr BP (Figure 8a). It is quite noticeable that those low-humification intervals in the southern Altai Mountains are chronologically correspondent with those high-percentage hematite-stained grains in the North Atlantic (Figure 7b; Bond et al., 1997). It means that the peat decay of the southern Altai Mountains was weaker during lower sea surface temperature (SST) intervals as suggested by higher percentages of the hematite-stained grains in the North Atlantic Ocean. It implies that the decadal-centennial-scale humification variations in the southern Altai Mountains have been primarily paced by SST in the North Atlantic Ocean whose signals were propagated to the Central Asia including the southern Altai Mountains via the prevailing westerlies. In other words, the decadal-centennial-scale chronological correspondence between SST and lower humification implies that both lower temperature and lower temperature-suppressed evaporation (i.e. elevated moisture level) were most likely responsible for inerting peat decay in the uppermost peat layer in the southern Altai Mountains.

Figure 8.

Comparison of the hematite-stained grains (%) in the North Atlantic (b) (Bond et al., 1997) with the averaged light absorbance residuals (a) of four peat cores (NRX, GHZ, TLHT, and KLSZ) in the southern Altai Mountains (this study). Vertical bands indicate the corresponding relationships between the lower-value stages of averaged residuals in the southern Altai Mountains and the higher-value stages of hematite-stained grains (%) in the North Atlantic.

Conclusions

Holocene humification records were derived from four hydrologically independent ombrotrophic peatlands (Narenxia, Tuolehaite, Ganhanzi, and Kelashazi) to investigate the millennial-scale and decadal-centennial-scale variability of peat decomposition in the southern Altai Mountains. The results show that the individual differences among four humification records at short-term timescales basically resulted from differences in botanical compositions and local topographies. The millennial-scale humification variations in the southern Altai Mountains was basically moisture-dependent. That is, dry conditions during the middle Holocene (between ~8200 and ~4000 cal. yr BP) promoted the peat decay, and wet conditions in the early Holocene (before ~8200 cal. yr BP) and in the late Holocene (after ~4000 cal. yr BP) suppressed the peat decay. The decadal-centennial-scale humification variations in the southern Altai Mountains have been primarily paced by SST in the North Atlantic Ocean. It means that both the lower temperature and the lower temperature-elevated moisture level were most likely responsible for inerting the decay activity in the uppermost peat layer.

Footnotes

Acknowledgements

This research was financially supported by National Natural Science Grants of China (No. 41771234 and No. 41803024), Postdoctoral Innovative Talent Support Program of China (No. BX20190363) and Western Young Scholar Program-B of Chinese Academy of Sciences (2018-XBQNXZ-B-020).

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Dongliang Zhang

Yunpeng Yang

References

Aaby

Tauber

(2008) Rates of peat formation in relation to degree of humification and local environment, as shown by studies of a raised bog in Deninark. Boreas 4: 1–17.

Aizen

Melack

, et al. (2001) Precipitation and atmospheric circulation patterns at mid-latitudes of Asia. International Journal of Climatology 21(5): 535–556.

Amesbury

Barber

Hughes

(2012) The relationship of fine-resolution, multi-proxy palaeoclimate records to meteorological data at Fågelmossen, Värmland, Sweden and the implications for the debate on climate drivers of the peat-based record. Quaternary International 268: 77–86.

Barber

(1981) Peat stratigraphy and climatic change: a palaeoecological test of the theory of cyclic peat bog regeneration. Rotterdam: Balkema.

Blaauw

Christen

(2011) Flexible paleoclimate age-depth models using an autoregressive gamma process. Bayesian Analysis 6: 457–474.

Blackford

Chambers

(1993) Determining the degree of peat decomposition for peat-based palaeoclimate studies. International Peat Journal 5: 7–24.

Bond

Shower

Cheseby

, et al. (1997) A pervasive millennial-scale cycle in North Atlantic Holocene and Glacial climates. Science 278: 1257–1266.

Borgmark

Wastegård

(2008) Regional and local patterns of peat humification in three raised peat bogs in Värmland, south-central Sweden. GFF 130: 161–176.

Burrows

Fenner

Haberle

(2014) Humification in northeast Australia: Dating millennial and centennial scale climate variability in the late Holocene. The Holocene 24: 1707–1718.

10.

Caseldine

Baker

Charman

, et al. (2000) A comparative study of optical properties of NaOH extracts: implications for humification studies. Holocene 10: 649–658.

11.

Castro

Souto

Garcia-Rodeja

, et al. (2015) Climate change records between the mid- and late Holocene in a peat bog from Serra do Xistral (SW Europe) using plant macrofossils and peat humification analyses. Palaeogeography Palaeoclimatology Palaeoecology 420: 82–95.

12.

Chambers

Barber

Maddy

, et al. (1997) A 5500-year proxy-climate and vegetation record from blanket mire at Talla Moss, Borders, Scotland. The Holocene 7: 391–399.

13.

Chambers

Beilman

(2011) Methods for determining peat humification and for quantifying peat bulk density, organic matter and carbon content for palaeostudies of climate and peatland carbon dynamics. Mires and Peat 7(7): 1–10.

14.

Charman

(2007) Summer water deficit variability controls on peatland water-table changes: implications for Holocene palaeoclimate reconstructions. The Holocene 17: 217–227.

15.

Chen

(2010) Geography of Chinese arid Region. Beijing: Science Press (in Chinese).

16.

Clarkson

Schipper

Lehmann

(2004) Vegetation and peat characteristics in the development of lowland restiad peat bogs, North Island, New Zealand. Wetlands 24: 133–215.

17.

Clymo

(1984) The limits to peat bog growth. Philosophical Transactions of The Royal Society of London. Series B, Biological Sciences 303: 605–654.

18.

Davis

Brewer

Stevenson

, et al. (2003) The temperature of Europe during the Holocene reconstructed from pollen data. Quaternary Science Reviews 22(15–17): 1701–1716.

19.

Feng

Sun

Abdusalih

, et al. (2017) Vegetation changes and associated climatic changes in the southern Altai Mountains within China during the Holocene. The Holocene 27(5): 683–693.

20.

Gehrels

Lowe

Hazell

, et al. (2006) A continuous 5300-yr Holocene cryptotephrostratigraphic record from northern New Zealand and implications for tephrochronology and volcanic hazard assessment. The Holocene 16: 173–187.

21.

Hong

Wang

Jiang

, et al. (2001) A 6000-year record of changes in drought and precipitation in northeastern China based on a δ13C time series from peat cellulose. Earth and Planetary Science Letters 185: 111–119.

22.

Huang

Cheng

Mao

, et al. (2013) Humification degree of peat and its implications for Holocene climate change in Hani peatland, Northeast China. Chinese Journal of Geochemistry 32(4): 406–412.

23.

Huang

Peng

Rudaya

, et al. (2018) Holocene vegetation and climate dynamics in the Altai Mountains and surrounding areas. Geophysical Research Letters 45(13): 6628–6636.

24.

Hughes

Mallon

Essex

, et al. (2012) The use of k-values to examine plant ‘species signals’ in a peat humification record from Newfoundland. Quaternary International 268: 156–165.

25.

Jara

Newnham

Alloway

, et al. (2017) Pollen-based temperature and precipitation records of the past 14,600 years in northern New Zealand (37°S) and their linkages with the Southern Hemisphere atmospheric circulation. The Holocene 27(11): 1756–1768.

26.

Zhang

Andreeva

, et al. (2020a) Temporal–spatial variability of modern climate in the Altai Mountains during 1970–2015. PLoS One 15(3): e0230196.

27.

Chen

Ran

(2020b) Holocene peat humification and carbon dynamics in the Westerlies-influenced Northwest China. Environmental Research Letters 15(12): 124014.

28.

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(S1): 28–37.

29.

Marcott

Shakun

Clark

, et al. (2013) A reconstruction of regional and global temperature for the past 11,300 years. Science 339(6124): 1198–1201.

30.

McGlone

(2009) Postglacial history of New Zealand wetlands and implications for their conservation. New Zealand Journal of Ecology 33: 1–23.

31.

Meeker

Mayewski

(2002) A 1400-year high-resolution record of atmospheric circulation over the North Atlantic and Asia. Holocene 12: 257–266.

32.

Morgan

Herod

Brain

, et al. (2005) Examination of soil contaminated by coal-liquids by size exclusion chromatography in 1-methyl-2-pyrrolidinone solution to evaluate interference from humic and fulvic acids and extracts from peat. Journal of Chromatography A 1095: 81–88.

33.

Newnham

Hazell

Charman

, et al. (2019) Peat humification records from Restionaceae bogs in northern New Zealand as potential indicators of Holocene precipitation, seasonality, and ENSO. Quaternary Science Reviews 218: 378–394.

34.

Nurbayev

Niels

Altai

(2008) Peanlands of the Chinese Altay. Peatlands International 2: 25–28.

35.

Payne

Blackford

(2008) Peat humification and climate change: a multi-site comparison from mires in south-east Alaska. Mires and Peat 3: 1–11.

36.

Ran

Chen

(2019) The 4.2 ka BP climatic event and its cultural responses. Quaternary International 521(30): 158–167.

37.

Rao

Huang

Xie

, et al. (2019) Long-term summer warming trend during the Holocene in central Asia indicated by alpine peat α-cellulose δ13C record. Quaternary Science Reviews 203: 56–67.

38.

Rao

Shi

, et al. (2020) Long-term winter/summer warming trends during the Holocene revealed by α-cellulose δ18O/δ13C records from an alpine peat core from central Asia. Quaternary Science Reviews 232: 106217.

39.

Reimer

Bard

Bayliss

, et al. (2013) IntCal13 and Marine13 radiocarbon age calibration curves, 0–50,000 years cal BP. Radiocarbon 55: 1869–1887.

40.

Sun

(2012) Holocene environmental changes inferred from the plant macrofossil records from the peatlands in the Altai Mountains. Lanzhou: Lanzhou University (in Chinese).

41.

Viau

Gajewski

(2009) Reconstructing millennial-scale, regional paleoclimates of boreal Canada during the Holocene. Journal of Climate 22(2): 316–330.

42.

Vorren

Jensen

Nilssen

(2012) Climate changes during the last c. 7500 years as recorded by the degree of peat humification in the Lofoten region, Norway. Boreas 41(1): 13–30.

43.

Walker

MJC

Berkelhammer

Björck

, et al. (2012) Formal subdivision of the Holocene series/epoch: a discussion paper by a working group of INTIMATE (integration of ice-core, marine and terrestrial records) and the subcommission on Quaternary Stratigraphy (international commission on Stratigraphy). Journal of Quaternary Science 27: 649–659.

44.

Wang

Hong

Zhu

, et al. (2004) Humification degrees of peat in Qinghai-Xizang Plateau and palaeoclimate change. Chinese Science Bulletin 49: 514–519.

45.

Wang

Zhang

(2019) Holocene vegetation evolution and climatic dynamics inferred from an ombrotrophic peat sequence in the southern Altai Mountains within China. Global and Planetary Change 179: 10–22.

46.

Wilmshurst

McGlone

Charman

(2002) Holocene vegetation and climate change in southern New Zealand: linkages between forest composition and quantitative surface moisture reconstructions from an ombrogenous bog. Journal of Quaternary Science 17: 653–666.

47.

Cao

Jia

, et al. (2020) Peat brGDGTs-based Holocene temperature history of the Altai Mountains in arid Central Asia. Palaeogeography Palaeoclimatology Palaeoecology 538: 109464.

48.

Xiao

Cheng

Mao

, et al. (2017) Characteristics of peat humification, magnetic susceptibility and trace elements of Hani peatland, northeastern China: paleoclimatic implications. Atmospheric Science Letters 18(3): 140–150.

49.

Yang

Zhang

Lan

, et al. (2019) Peat δ¹³C_celluose-signified moisture variations over the past ∼2200 years in the southern Altai Mountains, northwestern China. Journal of Asian Earth Sciences 174: 59–67.

50.

Yeloff

Mauquoy

(2006) The influence of vegetation composition on peat humification: implications for palaeoclimatic studies. Boreas 35: 662–673.

51.

Zanchetta

Regattieri

Isola

, et al. (2016) The so-called “4.2ka event” in the central Mediterranean and its climatic teleconnections. Alpine and Mediterranean Quaternary 29: 5–17.

52.

Zhang

Feng

Yang

, et al. (2018c) Peat δ¹³C_celluose-recorded wetting trend during the past 8000 years in the southern Altai Mountains, northern Xinjiang, NW China. Journal of Asian Earth Sciences 156: 174–179.

53.

Zhang

Chen

, et al. (2020a) Response of vegetation to Holocene evolution of westerlies in the Asian Central Arid Zone. Quaternary Science Reviews 229: 106138.

54.

Zhang

Chen

, et al. (2020b) Holocene moisture variations in the Arid Central Asia: new evidence from the southern Altai Mountains of China. Science of the Total Environment 735: 139545.

55.

Zhang

Chen

, et al. (2020c) Holocene vegetation dynamics and associated climate changes in the Altai Mountains of the Arid Central Asia. Palaeogeography Palaeoclimatology Palaeoecology 550: 109744.

56.

Zhang

Feng

(2018) Holocene climate variations in the Altai Mountains and the surrounding areas: a synthesis of pollen records. Earth-Science Reviews 185: 847–869.

57.

Zhang

Yang

Lan

(2017) Peat humification- and δ¹³C_celluose-recorded warm-season moisture variations during the past 500 years in the southern Altai Mountains within northern Xinjiang of China. Journal of Mountain Science 14(11): 2200–2211.

58.

Zhang

Yang

Tong

, et al. (2018a) Palynological record of Holocene vegetation and climate changes in a high-resolution peat profile from the Xinjiang Altai Mountains, northwestern China. Quaternary Science Reviews 201: 111–123.

59.

Zhang

Liu

, et al. (2018b) Preliminary study on morphology, development process and peat accumulation rate of palsas during the Altai Mountains, northern Xinjiang Autonomous Region, Northwest China. Quaternary Sciences 38(5): 1221–1232. in Chinese.

60.

Zhang

Meyers

Liu

, et al. (2016) Holocene climate changes in the central Asia mountain region inferred from a peat sequence from the Altai mountains, Xinjiang, northwestern China. Quaternary Science Reviews 152: 19–30.

61.

Zhao

Chen

, et al. (2009) Vegetation response to Holocene climate change in monsoon-influenced region of China. Earth-Science Reviews 97: 242–256.