Reconstructing late Holocene vegetation and fire histories in monsoonal region of southeastern China

Palaeogeography, Palaeoclimatology, Palaeoecology 393 (2014) 102–110

Contents lists available at ScienceDirect

Palaeogeography, Palaeoclimatology, Palaeoecology journal homepage: www.elsevier.com/locate/palaeo

Reconstructing late Holocene vegetation and fire histories in monsoonal region of southeastern China Kangyou Huang a,g,⁎, Zhuo Zheng a, Wenbo Liao b, Linglong Cao a,c, Yanwei Zheng d, Hua Zhang a, Guangqi Zhu a, Zhong Zhang e, Rachid Cheddadi f a

Department of Earth Sciences, Sun Yet-san University, Guangzhou 510275, China School of Life Sciences, Sun Yet-san University, Guangzhou 510275, China South China Sea Marine Engineering and Environment Institute, SOA, Guangzhou 510300, China d Institute of Geography, Guangzhou 510070, China e Administration Bureau of Jinggangshan National Nature Reserve, Jinggangshan 343600, China f Institut des Sciences de l'Evolution de Montpellier, UMR 5554, CNRS-UM2, 34095 Montpellier cedex 5, France g Guangdong Provincial Key Laboratory of Mineral Resources & Geological Processes, Guangzhou 510275, China b c

a r t i c l e

i n f o

Article history: Received 8 January 2013 Received in revised form 8 October 2013 Accepted 4 November 2013 Available online 16 November 2013 Keywords: Late Holocene Pollen analysis Charcoal East Asian Summer Monsoon Subtropical China

a b s t r a c t Eastern subtropical China is a key region for understanding the variability of the East Asian Summer Monsoon (EASM). Multidisciplinary studies in southeastern China have shown that the summer monsoon intensity declined in the mid-late Holocene. We present a high-resolution pollen record of the last 4000 cal yr BP in Jinggang Mts of Jiangxi Province, southeastern China. The identified pollen taxa from the core can be statistically divided into three groups corresponding to evergreen, deciduous and wetland communities. The transitions between evergreen and deciduous-coniferous pollen associations is likely caused by temperature fluctuations, indicating that climate was relatively cool at 3800–3200 cal yr BP and 2200–1300 cal yr BP and warmer at 3200– 2800 cal yr BP and 1300–800 cal yr BP. The vegetation study suggests that an Alnus-dominant association represents a secondary forest that usually takes place and expand after repeated forest fires. The charcoal concentration from the core depicts at least six major forest fire events since 4000 cal yr BP, most of which were followed by the development of an Alnus forest community. This result suggests that the EASM weakened toward the late Holocene and that its related decrease in moisture led to large forest wildfires. Furthermore, the rapid formation of a swamp and the subsequent development of the Alnus wetland community at ~550 cal yr BP suggest a gradual drying up of the lake, which was likely related to “the Little Ice Age”. As a result of a substantial burning related to an intensification of the human cultivation practices, Alnus reached its highest values in the last 200 years along with abundant wetland herbs, pioneer ferns (mainly Dicranopteris) and high charcoal concentrations. The present evidence of several sharp floristic and climate changes coincides with either the collapse or the beginning of some Chinese dynasties, which will need further research on the relationship between natural and human cultural changes. Crown Copyright © 2013 Published by Elsevier B.V. All rights reserved.

1. Introduction The variations of the East Asian Summer Monsoon (EASM) have prominent impacts on the vegetation in southeastern China, where the zonal forests are composed of tropical rainforest, subtropical evergreen broadleaved forest and mixed broadleaved forest. EASM precipitation is a key issue for regional agricultural production and for maintaining natural biodiversity and ecological balance (Zhang et al., 2011). Thus,

⁎ Corresponding author at: Department of Earth Sciences, Sun Yet-san University, Guangzhou 510275, China. Tel.: +86 20 84111068; fax: +86 20 84112390. E-mail address: [email protected] (K. Huang).

the aim of the present study is to contribute to a better understanding of the relationship between ecosystem changes and the monsoon system. Reduced precipitation during late Holocene often induces forest fires over the mountain regions in subtropical China (Sun et al., 2000; Zong et al., 2007; Wu et al., 2008). There are continuous records of the Asian monsoon over the Holocene from δ18O measurements of stalagmite calcite (Dykoski et al., 2005; Wang et al., 2005). However, the response of forests to the EASM remains poorly understood (Yue et al., 2012; Li et al., 2013) due to the paucity of high-resolution subtropical pollen records. Stalagmites from caves in southeastern China have provided many high-resolution δ18O records that primarily reconstruct the amount of precipitation associated with the strength and character of the EASM circulation (Cosford et al., 2008). Correlations among the δ18O records of stalagmites from the Lianhua, Heshang and Dongge caves show that

0031-0182/$ – see front matter. Crown Copyright © 2013 Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.palaeo.2013.11.005

K. Huang et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 393 (2014) 102–110

monsoonal circulation weakened throughout the mid- to late-Holocene as the ITCZ shifted southward in response to the decreasing summer insolation at the northern latitudes (Dykoski et al., 2005; Wang et al., 2005; Cosford et al., 2008; Hu et al., 2008). It is generally accepted that both temperature and precipitation continuously rose during the early Holocene and reached their highest levels at the mid-Holocene, followed by a gradual lowering during the late Holocene (An et al., 2000; Chen et al., 2001; Wang et al., 2005). Accordingly, some pollen records in the region reveal that the mixed forest in Chinese subtropical mountains was gradually replaced by the expansion of evergreen broadleaved forest (Xiao et al., 2007) and that evergreen monsoonal forest was the dominant vegetation in subtropical mountains around 8500 yr BP (Yue et al., 2012). Unfortunately, such records remain too scarce in the Chinese subtropical/tropical domain that is largely affected by the EASM, and many vegetation changes in the late Holocene are usually interpreted as related only to human disturbances (Xu et al., 2013). The modern plant distribution is a result of interactions between the environment (abiotic and biotic) and the plants species. Fire, land usage and human activities have a strong impact on natural plant communities (Lubchenco et al., 1991; Hannah et al., 1995; Xu et al., 2013). Understanding the role of plant species in a community or in an ecosystem is a basis for predicting plant responses to climatic changes (Austin, 2002). The present work is based on a ~4000 year record from a wetland in the Jinggang Mts, located in the middle subtropical zone of southern China. Pollen and charcoal analyses were performed to evaluate the vegetation changes and their possible relationship with monsoon variability and to better understand human-nature interactions in southern China during the late Holocene. 1.1. Study area The coring site is a wetland located in the valley of the Jinggang Mts, in southeastern Jiangxi Province (Fig. 1). The site is an oval-shaped swamp with a size of approximately 300 m × 200 m. The bedrock of the sediment basin is composed mostly of Mesozoic granite. The area belongs to the southern limit of the middle subtropical zone. The regional mean annual temperature is approximately 14.2 °C, and the mean annual precipitation is approximately 1800 mm (Lin, 1990). The present-day vegetation in the Jinggang Mts is dominated by a subtropical evergreen broadleaved forest (Fig. 1). The vegetation

103

composition in altitude is as follows: below 1000 m, the evergreen broadleaved forest is dominated by Castanopsis sclerophylla, Castanopsis concinna, Castanopsis fabric, Castanopsis tibetana, Machilus thunbergii, Phoebe hunanensis and Elacocarpus japonicus. Pinus massonia occurs in a few areas. From 1000 m to 1400 m, some deciduous elements are often mixed with the evergreen broadleaved forest. The deciduous elements are Castanopsis eyrei, Schima superva and Fagus lucida. Above 1400 m, the mountain summits are covered by shrub vegetation dominated by Rhododendron simiarum, Pieris formosa and Enkianthus quinqueflorus (Wu, 1980). The present-day vegetation around the coring area is a secondary subtropical evergreen broadleaved forest that is composed of Alnus trabeculosa, Phyllostachys heterocycla, Camellia cuspidata, Litsea elongata, Cleyera pachyphylla, Photinia schneideriana, Malus hupehensis, Rhododendron latoucheae, Daphniphyllum macropodum, Ilex serrata, Indocalamus latifolius, Liriope muscari and Peucedanum praeruptorum. The surrounding mountainside is covered by planted bamboo and pine forest (Pinus massoniana), reflecting a strong human impact on the local vegetation.

2. Material and method A 165-cm core was collected using a Russian Corer at 26°34′51.63″N, 114°04′37.55″E and 1269 m a.s.l. The core lithology may be described as follows: 0–31 cm: dark brown peat with rich plant roots and fragments; 31–72 cm: gray-green silt with little sand; 72–81 cm: gray– yellow medium sand; 81–117 cm: dark gray clay; 117–143 cm: brown silt-clay with plant remains; 143–155 cm: gray–brown clay; 155–165 cm: clay with sandy gravel (Fig. 2). The time frame of the core is based on three AMS 14C dates (Table 1). All radiocarbon ages were measured at the laboratory of Beta Analytic Inc., Miami, USA. The 14C dates were calibrated using the IntCal09 dataset (Reimer et al., 2009). The age model for the core was plotted using the Clam 1.0.2 software package (Blaauw, 2010) (Fig. 3). We used a visible light spectrophotometer (722 s) for peat humification evaluation. A total of 48 samples (3-cm interval) were collected from the core segments, which were mashed and dried in an oven. After filtering through a 240-μm sieve, each sample weighing 0.1 g was put into 100 ml of 8% NaOH and boiled for one hour, then diluted into 100 ml liquid. Five milliliters of liquid was taken out and diluted into 50 ml in a volumetric bottle. In the final step, the absorbency was

Fig. 1. Geographical location of the study core (a) and local vegetation map (b).

104

K. Huang et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 393 (2014) 102–110

Fig. 2. Lithological description and fieldwork photos showing the coring site and the sediment of the SMP core.

measured three times in the spectrophotometer for each sample and the mean value was used to plot curve. The grayscale (GS), used as a proxy indicating the sediment color change, was measured from 8-bit GS images using the Image J software (Rasband, W.S. et al., http://rsb.info.nih.gov/ij/). In general, high GS values indicate lower organic content in the sediment, and low values indicate more abundant organic matter (Fig. 2). A total of 80 samples for pollen analysis were taken at 2 cm intervals from the SMP core. The pollen grains were extracted using the following process:

plotted using Tilia software, and the pollen zones were based on the cluster results of CONISS (Grimm, 1987). The micro-charcoals were counted using the same microscope. Particles smaller than 50 μm were not generally taken into account, constituting background noise, while fragments N50 μm were taken as evidence of region/local fire (Whitlock and Larsen, 2001; Sadori and Giardini, 2007). Charcoal concentration (particles/cm2) values were used to evaluate the dynamics of past fires.

(1) 10% HCL to remove carbonates; (2) hot 10% KOH to remove organic matter; (3) heavy liquid (ZnCl2, density 2.0) to separate pollen grains from heavy minerals; (4) sieving using a 10-μm mesh screen to concentrate pollen grains (Nakagawa et al., 1998). The modern pollen slide collection at Sun Yat-sen University was used as a reference for pollen identification. Some key atlases were also used for palynomorph comparison, such as Pollen Flora of China, 2nd Edition (Wang et al., 1995), and Tropical and Subtropical Angiosperm Pollen Morphology (IBSCIBCAS, 1982). Pollen counting was performed using a Nikon E200 microscope with 1000× magnification. The average number of pollen grains counted was approximately 300 (at least 200 grains for each sample). The pollen percentages were calculated on the sum of trees, shrubs and terrestrial herbs. The pollen diagram was Table 1 AMS 14C dates and calibrated age of the core SMP in Jinggang Mts. Sample No.

Depth Material (cm)

SMP01 SMP04 SMP02 SMP03

48 63 95 142

Conventional 13C/12C age (yr BP) (‰)

Peat 110 Sediment 380 Sediment 910 Sediment 2780

± ± ± ±

30 30 30 30

−26.6 −25.6 −28.1 −27.4

Calibrated date (yr BP) (2δ)

Central cal. age (yr BP)

13–148 440–450 761–915 2837–2952

80 445 840 2895

Fig. 3. Age/depth model based on calibrated 14C dates of the SMP core.

K. Huang et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 393 (2014) 102–110

Fig. 4. Pollen diagram of the SMP core and cluster analysis (shadow is 2× magnification; con: conifer, dbt: deciduous broadleaf tree, ebt: evergreen broadleaf tree, shr: shrub).

105

106 K. Huang et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 393 (2014) 102–110 Fig. 5. Variation of ecological pollen groups (deciduous elements, evergreen elements, wetland herbs and Alnus) and charcoal concentration and correlation with other climate proxies: summer insolation of 30ºN (June, Berger and Loutre, 1991), moisture index from δ18O carb-based integration (Zhang et al., 2011) and δ18O of the Dongge cave (Dykoski et al., 2005; Wang et al., 2005).

K. Huang et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 393 (2014) 102–110

3. Result of pollen and charcoal analysis A total of 78 pollen/spore types were identified in more than 80 samples. The main arboreal pollen types (AP) include coniferous taxa (Cupressaceae and Pinus) and broadleaved trees and shrubs (Alnus, Castanopsis/Lithorcarpus, Quercus-evergreen, Quercusdeciduous/ Cyclobalanopsis, Acer, Ericaceae, Betula, Liquidambar, Rosaceae, Hamamelis, Fagus, Ilex, Eurya, Theaceae, Celtis and Ulmus). The main herbaceous taxa are Cyperaceae, Poaceae, Araceae, Umbelliferae, Compositae/Artemisia and Polygonum. The identified spores are mainly composed of various Bryophyta, Dicranopteris, Polypodiaceae, Pteris, Adiantum, Trilete and Monolete spores. According to the cluster analysis, the pollen assemblages can be divided into 8 zones from bottom to top (Fig. 4): Zone 1 165–150 cm (ca. 4000–3200 cal yr BP). The pollen taxa are principally composed of deciduous broadleaved and coniferous types, such as Fagus, Liquidambar, Betula, Carpinus, Acer, Alnus, Cuperaceae and Pinus. The general percentage of each taxa is approximately 5–13%. The mountain shrub and evergreen broadleaved components, such as Ericaceae and Quercus, are approximately 10%. The pollen content of Alnus is the lowest in the whole profile. The shrub is composed of Ericaceae, Rosaceae and Ilex. The herbs and spores are dominated by Poaceae, Bryophytes and Triletes. Zone 2 150–139 cm (ca. 3200–2800 cal yr BP). The deciduous broadleaved pollen taxa abruptly decrease and give place to the evergreen broadleaved components, such as Quercusevergreen (10–15%), Liquidambar (8–10%) and Ilex (5–10%). The ferns of the Trilete spores are more abundant. Zone 3 139–125 cm (ca. 2800–2200 cal yr BP). The characteristic feature of this zone is that both evergreen and deciduous broadleaved trees are all in low proportion (less than 8%). Alnus, considered a secondary forest plant, dominates the pollen assemblage, increasing from b 10% in zone 2 to 35%. Zone 4 125–107 cm (ca. 2200–1300 cal yr BP). This zone coincides with light color sediment interval, and there is a sharp decrease of Alnus from 35% to b10%. Deciduous broadleaved trees such as Quercus-evergreen (ca. 8%) and Fagus (ca. 10%) and bushes such as Ericaceae (15–25%) are more abundant.

Fig. 6. Principal correspondence analysis (PCA) based on pollen data from the SMP core. Only the principal taxa are shown.

107

Zone 5 107–75 cm (ca. 1300–550 cal yr BP). The broadleaved pollen taxa have higher percentages, and the most representative components are Castanopsis, Liquidambar and Quercusevergreen. The above three taxa cover more than 40% of the total pollen. In the same way, the Alnus content is also higher (20–35%) than that of zone 4. By contrast, the deciduous trees abruptly decline in this zone. Zone 6 75–52 cm (ca. 550–200 cal yr BP). The taxa belonging to subtropical broadleaved forests decline obviously. The wetland herb Cyperaceae increases to ~15%. The Alnus pollen decreases to ~10%. Zone 7 52–38 cm (ca. 200–90 cal yr BP). The percentage of Cyperaceae continues to increase, and the sediments of the cores contain large amounts of plant remains. The content of the broadleaved pollen taxa decreases greatly. The percentage of Dicranopteris spores reaches its highest level in this zone. Zone 8 38–0 cm (ca. 90 cal yr BP–present). Most part of this zone corresponds to the top peat layer, and there are a gradual and evident increase in Alnus and Cyperaceae. The percentage of Alnus is generally ~50%, with a maximum level as high as 70%. However, the percentage of other arboreal trees decreases sharply, especially the taxa of subtropical evergreen forest, such as Castanopsis, Quercus-evergreen, Photinia and Symplocos. The temperate deciduous taxa, such as Fagus and Betula, almost completely disappear. At the same time, large numbers of Poaceae, Bryophyta and Dicranopteris mark this zone. The total charcoal concentrations vary between 1.5 × 103 and 21.36 × 103 grains/cm2, with at least six peaks in the SMP profile (Fig. 5). The peak at approximately 3100 cal yr BP has a concentration of approximately 10.04 × 103 grains/cm2. The second peak, at 2600 cal

Fig. 7. Environmental changes of the last 2000 years indicated by key pollen taxa from the SMP core, and their correlation with the estimations of the Northern Hemisphere mean temperature (Moberg et al., 2005). a, Previous multi-proxy reconstruction (Mann and Jones, 2003, black line). b, Multi-proxy reconstruction (Moberg et al., 2005). c, Percentage variation of wetland herbs (black), temperate taxa (red) and Alnus (blue).

108

K. Huang et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 393 (2014) 102–110

yr BP, has a similar concentration of 8.97 × 103 grains/cm2. During the period 1300–700 cal yr BP, there are several peaks varying between 3.2 × 103 and 9.19 × 103 grains/cm2. During 600–400 cal yr BP, the charcoal concentration reaches 10.83 × 103 grains/cm2. The last period, with a highest concentration of charcoal (21.36 × 103 grains/cm2, at least 2–3 times the previous peaks), occurred at approximately 160– 60 cal yr BP. 4. Discussion 4.1. Ecological significance of pollen taxa In previous studies, the environmental significance for the fossil pollen of Alnus was controversial. It was interpreted as reflecting cold and wet conditions in some works of southern China (Xiao et al., 2007), and it was also used as an intrazonal taxon of wetland shrub (Zong et al., 2007). Based on a great number of vegetation investigation, the Alnus genus contains four major species in China which are distributed along the creek or sub-mountain in modern subtropical shrub ecosystem. Botanical and ecological works have proved that many Alnus communities in the southwestern and subtropical mountains belong to secondary forests that resulted from disforestation and fire (Jiang, 1980; Li et al., 2008; Peng, 2010). Peng et al. (2010) have stated that Alnus nepalensis communities in Yunnan province represent an unstable successional forest. Wu (1980) also mentioned that some deciduous broadleaved elements including Alnus occur in the disforestation areas of typical potential evergreen broadleaved forest zone. In consequence, the fossil pollen of Alnus reflects the secondary ecological system rather than cold climate indicator in the work region. To statistically infer an ecological relationship between the pollen taxa identified in the fossil record, we used principal component analysis (PCA). This multivariate analysis reveals three distinct ecological groups (Fig. 6). The first group (G1) is composed of evergreen taxa including Castanopsis, Quercus-evergreen, Hamamelidaceae, Symplocos, Liquidambar, Celastraceae and Pterocarya. G1 may be assigned to the subtropical broadleaved forest. The second group (G2) consists of Fagus, Acer, Betula, Quercus-deciduous, Ulmus, Pinus, Cupressaceae and Ericaceae. G2 gathers deciduous and coniferous mixed forests, which are adapted to colder conditions. The Ericaceae (mostly Rhododendron) scores are close to those of the deciduous group, most likely because of the vicinity of the mountain range. The third group (G3) has only one arboreal taxon, namely, Alnus. The associated taxa are Cyperaceae, Typha, Poaceae, Dicranopteris and Bryophyta, which are often characteristic of wetland communities. According to our fieldwork investigation, the climax forest community in the vicinity of the Jinggang mountain summit is characterized by evergreen-deciduous mixed broadleaved forest with Quercus-evergreen, Castanopsis, Fagus and Ericaceae. The Alnus community is very local. 4.2. Vegetation response to the climate change during late Holocence Current evidence has demonstrated that the Holocene thermal maximum occurred during the early to mid-Holocene (9000–4000 ka BP). This thermal maximum affected those regions from southern China to the northwestern monsoon boundary in northern China (An et al., 1993, 2000; Yang and Scuderi, 2010). Some authors have suggested that the forest decline observed in the Nanling Mts, southern China, after 6000 cal yr BP was due to drying conditions (Xiao et al., 2007). Most of the Holocene records tend to agree on a temperature decrease after 4200 cal yr BP (Xia et al., 2000; Jin and Liu, 2001; Wu and Liu, 2004). It is now widely believed that the climate in southern China became colder and dryer in the late Holocene, possibly because of a weakening of the summer monsoon. The δ18O records of speleothems from the Sanbao and Dongge caves confirm that the weakening of the summer monsoon took place at approximately 4000–3000 cal yr BP in China (Dykoski et al., 2005; Wang et al., 2005; Shao et al., 2006). A

record of the last 4200 years in the Arabian Sea shows a drought event that lasted between 300 years (Cullen et al., 2000) and 600 years, with different ranges in other regions (Bond et al., 1997; Guo et al., 1999; De menocal et al., 2000; Perry and Hsu, 2000). In China, drought events since 4000 cal yr BP have been widely recorded (Tang et al., 1993; Fang and Sun, 1998; Guo et al., 1999). The multidisciplinary results of current SMP core demonstrates that both sediment feature and pollen assemblages vary evidently during the last 4000 cal yr BP. The core bottom (~165 cm) began at ~4000 cal yr BP, and the earliest sediment (165–155 cm) containing some gravels is a type of alluvial deposit, which is followed by clayish lacustrine sediment at ~3500–2900 cal yr BP (155–142 cm). Two main peaks of peat humification (zone 3 and zone 8) imply two intervals of peatland formation that are likely related with climate and/or fire events. Our fossil record from the Jinggang Mts shows that the local vegetation has changed between deciduous and evergreen broadleaved forests and secondary Alnus forest. The variations between deciduous and evergreen forest are likely due to temperature fluctuations. The high proportion of the deciduous group indicates a cooler climate at 3800–3200 cal yr BP and 2200–1300 cal yr BP (Fig. 5). By contrast, the evergreen group shows an optimum climate with higher temperature at 3200–2800 cal yr BP and 1300–800 cal yr BP. The above changes in forest climaxes indicate that subtropical mountain broadleaved forests are sensitive to climate changes. For instance, the cooling of 2200–1300 cal yr BP marked by the high deciduous taxa may be correlated (an average of low resolution) with tree rings, revealing that the temperature was relatively low prior to 1400 cal yr BP (Moberg et al., 2005). Evidence from historical documents also indicates that there was a cold climate episode at approximately AD 310 (1640 cal yr BP) (Shi et al., 1999). How did these climate changes affect the subtropical ecosystems and their tree species composition? Today in the Jinggang Mts, Fagus deciduous mixed forest is found at a higher elevation (1200–1500 m) than the studied site. The high proportion of Fagus and other deciduous elements until ~ 1300 cal yr BP (Fig. 5) indicates cooler climate conditions than those of today. Similar pollen assemblages were identified in southern subtropical China, for example, in the SZY core collected in Fujian Province (Yue et al., 2012), where the expansion of Fagus forest occurred during the LGM, indicating a colder climate. The second-highest peak of the evergreen group occurs during the period centered on approximately 1000 cal yr BP, where a lowering of the lake level led to the formation of a shallow swamp at approximately 500–400 cal yr BP (Figs. 5 and 7). Our data are coherent with the northern hemisphere climate reconstructed from tree-rings, which show high and then low temperatures between 1000 and 900 cal yr BP and approximately 400 cal yr BP, respectively (Moberg et al., 2005). A striking feature that one may observe in the forest community reconstructed from the pollen record is the periodic variation between mixed broadleaved forests and Alnus secondary forest. The latter is considered a temporary community that likely developed after destructive forest fires at a time when dry condition was caused by a decrease of rainfall. The replacement by the Alnus community in the forest succession implies a strong regression of the local forest that may be caused by natural fire or human activity. The first peak of charcoal occurs at 3100 cal yr BP, indicating the first fire event recorded in the SMP core. This event did not lead to the complete destruction of the local mixed forest, but it indicates at least the first potential lowering of atmospheric humidity triggered by a weakening of the monsoon rainfall. The interval of 140–124 cm (2800–2200 cal yr BP) with high values of Alnus and high charcoal concentrations can be interpreted as a forest fire event caused by a dry condition that favored the rapid development of a secondary community. Furthermore, a remarkable peak in the humification curve is observed in the same time interval (Depth: 140– 124 cm) (Fig. 2), which matches the above rise of Alnus. This event is concordant with that of the Qian Mutian peat record (Yin et al., 2006). A synthetic study of δ13C records in southeastern China reveals that EASM weakening toward the late Holocene led to a decrease in the

K. Huang et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 393 (2014) 102–110

moisture availability and precipitation (Zhang et al., 2011). The drier climate during the late Holocene also led to large forest wildfires in northern China (Li et al., 2003; Huang et al., 2006). In the SMP core, there are at least six important fire events marked by high concentrations of charcoal (Fig. 5). Most of these forest fire events were followed by the development of a secondary forest community (Alnus in particular). We also observe that the development of these secondary forests generally has a 50–150 year delay after each fire event. The phenomenon of late Holocene aridity can quite possibly be related to the weakened EASM when the ITCZ migrates southward (Dykoski et al., 2005; Fleitmann et al., 2007; Yancheva et al., 2007; Wang et al., 2008). The δ18O results from the Dongge and Heshang caves demonstrate that the summer monsoon not only has a clear weakening tendency toward the late Holocene (Dykoski et al., 2005; Wang et al., 2005; Hu et al., 2008) but also has some high-resolution variability in the short-term changes, many of which are concordant with the charcoal events of the present study. A study of a lacustrine record form Hainan Island (Zheng et al., 2003) suggests that a significant climate change took place at approximately 2700 cal yr BP, with a decrease in temperature that is synchronous in time with our first development of Alnus secondary forest. Another study of the SZY core in nearby Fujian indicates an abrupt decrease of the evergreen broadleaved trees at 2.6 cal ka BP, suggesting a cooling climate driven by the weakening of the EASM, itself a result of the orbitally induced reduction in the Northern Hemisphere summer insolation (Yue et al., 2012). Many peat profiles in eastern China also show that a dryer and cooler climate appeared at 3200–630 cal yr BP (Yin et al., 2006). Human activities are also an important factor triggering forest fires and the subsequent secondary forest communities. For instance, the high proportion of Dicranopteris (fern species), that occurred after 200 cal yr BP, is usually considered as a pioneer plant species that develops after a severe disturbance of vegetation by human activities or forest fires. To sum up, the last 200 years were characterized by the highest values of charcoal concentration (highest throughout the record), an abrupt regression of the evergreen forests, the rapid growth of Alnus secondary forest and the rapid formation of a shallow water swamp. The most realistic explanation for these assemblages is that the climax forest was completely destroyed by repeated fires from deforestation-and-burn cultivation, which led to a rapid development of a shallow swamp from a natural lake, where a secondary community of Alnus developed from the lake front to the center of the basin. Many climate proxies, such as stalagmites from the Dongge Cave in southern China, indicate that the EASM intensified rapidly in the latest Holocene (Wang et al., 2005). This change means that the precipitation and humidity during the last 200 years must be relatively high due to the strengthening of the EASM, which has reduced the fire frequency. However, the recent strong forest fires in the Jinggang Mts must have been caused by human activities and the local increase in human population since the later Qing Dynasty. It should be noted that other studies have found evidence that climate change may have played a role in the collapse of historical cultures (Hodell et al., 1995; Ge et al., 2003; Yancheva et al., 2007). Our data also show coincidences with the collapse of some Chinese dynasties. For instance, the collapse of the Xia-Shang Dynasty may well coincide with the sudden disappearance of the climax forest replaced by the Alnus secondary forest (fire event 2). The Spring and Autumn Warring States Period coincide with pollen zone 3, when the climate was dry with frequent forest fires. Finally, the Han and Sui Dynasties ended during pollen zone 5 (fire event 3). Although we cannot directly relate these major Chinese societal events with our vegetation record, it is interesting to state that there are striking parallels in chronology. 5. Conclusion Although there is a great number of paleoclimate studies for the Holocene period in the monsoon region of eastern China, few of them

109

have focused on forest successions with high-resolution records. Our data allow us to analyze the vegetation response to the drying climate trend that resulted from the late Holocene weakening of the EASM. The current pollen record from the Jinggang Mts, located in subtropical China, reveals a series of vegetation changes over the past 4000 cal yr BP that clearly show that mountain forests are sensitive to climate fluctuations. The conclusions of this study can be summarized as follows: (1) there are three coherent ecological groups from the fossil pollen taxa. These three groups provide a coherent interpretation of the past ecosystem changes with fire events and climate changes. (2) The transitions between evergreen, deciduous and coniferous forests were likely caused by temperature fluctuations. The rise of the deciduous group occurred during two cooler intervals at 3800–3200 cal yr BP and 2200–1300 cal yr BP. Conversely, the increase in the evergreen group indicates warmer periods at 3200–2800 cal yr BP and 1300–800 cal yr BP. The above changes in natural climax forests indicate that the subtropical mountain broadleaved forests were sensitive to the climate changes during the late Holocene. In particular, the abrupt lowering of the lake level at approximately 500–400 cal yr BP led to the formation of a shallow swamp and the rapid development of an Alnus community. Such environmental changes are synchronous with the “little ice age”, e.g., the minimum temperatures during the last 2000 years that occurred at approximately 400 cal yr BP (Moberg et al., 2005). (3) The forest succession observed in our record suggests that Alnusdominant association represents a secondary forest growing on intrazonal mire, usually after forest fires or, more recently, strong human activity. The Alnus peak and humification recorded at 2800–2200 cal yr BP may be the result of a fire event. We suggest that the EASM weakening toward the late Holocene and the related decrease in moisture led to large forest wildfires. The Alnus-dominant intervals are basically concordant with the variations of δ18O from the Dongge and Heshang caves. At least six forest fire events have been well identified in the last 4000 years, most of them were followed by the development of a secondary forest community (Alnus mainly). The delay time of secondary forest development after a forest fire is approximately 50–200 years. (4) Our record shows that the local climax forests regressed gradually at 600 cal yr BP, followed by an increase in Alnus, wetland herbs such as Cyperaceae and many other aquatic plants. The above taxa reached their highest level in the last 200 years, which is associated with the pioneer fern Dicranopteris after deforestation. This pollen assemblage reflects that the natural climax forest was completely destroyed due to the forest fires caused by slash-and-burn cultivation. (5) The present record shows a good correlation between environmental changes, increasing population and Chinese dynasties. The collapse of the Xia-Shang Dynasty coincides with the sharp decrease of evergreen broadleaved forest and its replacement by a secondary forest, indicating the beginning of a drier period. The end of the Han and Sui Dynasties also coincides with a fire event (pollen zone 5).

Acknowledgments This work was supported by the following grants: the National Natural Science Foundation of China (NSFC: 41001118, 41072128 and 41230101), the Research Fund for the Doctoral Program of Higher Education of China (Grant No. 20100171120002), Fundamental Research Funds for the Central Universities of Sun Yat-sen University (111gpy53, 101gzd08 and 111gjc13), Young Scientist's Fund of the State Oceanic Administration of China (2012114) and partially by the integrated scientific

110

K. Huang et al. / Palaeogeography, Palaeoclimatology, Palaeoecology 393 (2014) 102–110

survey project of biodiversity from the Jinggang Mts of Jiangxi Province, China. References An, Z., Porter, S.C., Kutzbach, J.E., Wang, S., Liu, X., Li, X., Wang, J., Zhou, W., Xiao, J., Liu, J., Lu, J., 1993. Holocene optimum in mid- and eastern China and the vicissitude of the East Asian summer monsoon. Chin. Sci. Bull. 38, 1302–1305 (in Chinese). An, Z., Porter, S.C., Kutzbach, J.E., Wu, X., Wang, S., Liu, X., Li, X., Zhou, W., 2000. Asynchronous Holocene optimum of the East Asian monsoon. Quat. Sci. Rev. 19, 743–762. Austin, M.P., 2002. Spatial prediction of species distribution: An interface between ecological theory and statistical modelling. Ecol. Model. 157, 101–118. Berger, A., Loutre, M.F., 1991. Insolation values for the climate of the last 10 million years. Quat. Sci. Rev. 10, 297–317. Blaauw, M., 2010. Methods and code for ‘classical’ age-modelling of radiocarbon sequences. Quat. Geochronol. 5, 512–518. Bond, G., Showers, W., Cheseby, M., et al., 1997. A pervasive millennial-scale cycle in North Atlantic Holocene and glacial climates. Science 278, 1257–1266. Chen, F.H., Zhu, Y., Li, J.J., Shi, Q., Jin, L.Y., Wünemann, B., 2001. Abrupt Holocene changes of the Asian monsoon at millennial-and centennial-scales: evidence from lake sediment document in Minqin Basin, NW China. Chin. Sci. Bull. 46 (23), 1942–1947. Cosford, J., Qing, H., Yuan, D., Zhang, M., Holmden, C., Patterson, W., Hai, C., 2008. Millennial-scale variability in the Asian monsoon: evidence from oxygen isotope records from stalagmites in southeastern China. Palaeogeogr. Palaeoclimatol. Palaeoecol. 266, 3–12. Cullen, H.M., de Menocal, P.B., Hemming, S., Brown, F.H., Guilderson, T., Sirocko, F., 2000. Climate change and the collapse of the Akkadian empire. Geology 28 (4), 379–382. De Menocal, P.B., Ortiz, J., Guilderson, T., Sarnthein, M., 2000. Coherent high- and lowlatitude climate variability during the Holocene warm period. Science 288 (23), 2198–2202. Dykoski, C.A., Edwards, R.L., Cheng, H., Yuan, D., Cai, Y., Zhang, M., Lin, T., Qing, J., An, Z., Revenaugh, J., 2005. A high-resolution, absolute-dated Holocene and deglacial Asian monsoon record from Dongge Cave, China. Earth Planet. Sci. Lett. 233, 71–86. Fang, X.Q., Sun, N., 1998. Cold event: a possible cause of the interruption of the Laohushan Culture. Hum. Geogr. 13 (1), 71–76 (in Chinese with English abstract). Fleitmann, D., Burns, S.J., Mangini, A., Mudelsee, M., Kramers, J., Villa, I., Neff, U., AlSubbarye, A.A., Buettner, A., Hippler, D., Matter, A., 2007. Holocene ITCZ and Indian monsoon dynamics recorded in stalagmites from Oman and Yemen (Socotra). Quat. Sci. Rev. 26, 170–188. Ge, Q., Zheng, J., Fang, X., 2003. Winter half-year temperature reconstruction for the middle and lower reaches of the Yellow River and Yangtze River, China, during the past 2000 years. The Holocene 13, 933–940. Grimm, E.C., 1987. CONISS: a FORTRAN 77 program for stratigraphically constrained cluster analysis by the method of incremental sum of squares. Comput. Geosci. 13, 13–35. Guo, Z.T., Petit-Marie, N., Liu, T.S., 1999. Holocene abrupt environmental changes in the arid regions in Africa and Asia. J. Palaeogeogr. 1 (1), 68–74 (in Chinese with English abstract). Hannah, L., Carr, J.L., Lankerani, A., 1995. Human disturbance and natural habitat: a biome level analysis of a global data set. Biodivers. Conserv. 4, 128–155. Hodell, D.A., Curtis, J.H., Brenner, M., 1995. Possible role of climate in the collapse of Classic Maya civilization. Nature 375, 391–394. Hu, C., Henderson, G.M., Huang, J., Xie, S., Sun, Y., Johnson, K.R., 2008. Quantification of Holocene Asian monsoon rainfall from spatially separated cave records. Earth Planet. Sci. Lett. 266, 221–232. Huang, C.C., Pang, J.L., Chen, S.E., et al., 2006. Charcoal records of fire history in the Holocene loess soil sequences over the southern Loess Plateau of China. Palaeogeogr. Palaeoclimatol. Palaeoecol. 239, 28–44. IBSCIBCAS (Institute of Botany and South China Institute of Botany, Chinese Academy of Sciences), 1982. Angiosperm Pollen Flora of Tropic and Subtropic China. Science Press, Beijing 26–453 (in Chinese). Jiang, H.Q., 1980. Distriutional features and zonal regularity of vegetation in Yunnan. Acta Bot. Yunnanica 2, 142–151. Jin, G.Y., Liu, T.S., 2001. Cold event and the prehistoric cultural evolution during Holocene climatic optimum in north of China. Chin. Sci. Bull. 46 (20), 1725–1730. Li, Y.H., Yin, H.N., Zhang, X.Y., et al., 2003. The environment disaster events and the evolution of man-land relation in the west Liaoning during 5000 a BP. J. Glaciol. Geocryol. 25, 19–26 (in Chinese with English abstract). Li, D.W., Chen, H.W., Shi, F.Q., Yang, B., Li, J., 2008. Biological and ecological characteristics and geographical distribution of Alnus nepalensis yunnanensis. For. Invent. Plan. 33, 25–28. Li, J., Zheng, Z., Huang, K.Y., Yang, S.X., Chase, B.M., Valsecchi, V., Cheddadi, R., 2013. Vegetation changes during the past 40,000 years in Central China from a long fossil record. Quat. Int 310, 221–226. Lin, Y., 1990. Study on the Scientific Survey of Jinggang Mts, Nature Reserve in China. Xinhua Press, Beijing 140–145. Lubchenco, J., Olson, A.M., Brubaker, L.B., Carpenter, S.R., Holland, M.M., Hubbell, S.P., Levin, S.A., MacMahon, J.A., Matson, P.A., Melillo, J.M., Mooney, H.A., Peterson, C.H., Pulliam, H.R., Real, L.A., Regal, P.J., Risser, P.G., 1991. The sustainable biosphere research initiative: an ecological research agenda. Ecology 72, 371–412. Mann, M.E., Jones, P.D., 2003. Global surface temperature over the past two millennia. Geophys. Res. Lett. 30 (15), 1820. http://dx.doi.org/10.1029/2003GL017814.

Moberg, A., Sonechkin, D.M., Holmgren, K., Datsenko, N.M., Karlen, W., 2005. Highly variable Northern Hemisphere temperatures reconstructed from low- and highresolution proxy data. Nature 433, 613–617. Nakagawa, T., Brugiapaglia, E., Digerfeldt, G., Reille, M., de Beaulieu, J.L., Yasuda, Y., 1998. Dense-media separation as a more efficient pollen extraction method for use with organic sediment samples: comparison with the conventional method. Boreas 25, 15–24. Peng, H.M., 2010. Study on community characteristics of Alnus nepalensis natrual forest in Pu'er region. (Master thesis) Southwest University 1–4 (in Chinese with English abstract). Perry, C.A., Hsu, K.J., 2000. Geophysical, archaeological, and historical evidence support a solaroutput model for climate change. Proc. Natl. Acad. Sci. U. S. A. 97, 12433–12438. Reimer, P.J., Baillie, M.G.L., Bard, E., Bayliss, A., Beck, J.W., Blackwell, P.G., Bronk Ramsey, C., Buck, C.E., Burr, G.S., Edwards, R.L., Friedrich, M., Grootes, P.M., Guilderson, T.P., Hajdas, I., Heaton, T.J., Hogg, A.G., Hughen, K.A., Kaiser, K.F., Kromer, B., McCormac, F.G., Manning, S.W., Reimer, R.W., Richards, D.A., Southon, J.R., Talamo, S., Turney, C.S.M., van der Plicht, J., Weyhenmeyer, C.E., 2009. IntCal09 and Marine09 radiocarbon calibration curves, 0–50,000 years cal BP. Radiocarbon 4, 1111–1150. Sadori, L., Giardini, M., 2007. Charcoal analysis, a method to study vegetation and climate of the Holocene: the case of Lago di Pergusa (Sicily, Italy). Geobios 40, 173–180. Shao, X., Wang, Y., Cheng, H., Kong, X., Wu, J., Edwards, R.L., 2006. Long-term trend and abrupt events of the Holocene Asian monsoon inferred from a stalagmite d 18 O record from Shennongjia in Central China. Chin. Sci. Bull. 51, 221–228. Shi, Y.F., Yao, T.D., Yang, B., 1999. Decadal climatic variations recorded in Guliya ice core and comparison with historical documentary data from east China during the last 2000 years. Sci. China Ser. D 29 (Suppl. 1), 79–86. Sun, X.J., Li, X., Chen, H.C., 2000. Natural fires and climate record of the 37 ka from the north of the South China Sea. Sci. China Ser. D 30 (2), 163–168. Tang, L.Y., Shen, C.M., Zhao, X.T., Xiao, J.Y., Yu, G., Han, H.Y., 1993. Vegetation and climate since Holocene based on Qingfeng Section, Jianhu, Jiangsu Province. Sci. China Ser. B 23 (6), 637–643. Wang, F.H., Chie, N.F., Zhang, Y.L., Yang, H.Q., 1995. Pollen Flora of China, Second ed. Science Press, Beijing 461 (in Chinese). Wang, Y.J., Cheng, H., Edwards, L.R., He, Y.Q., Kong, X.G., An, Z.S., Wu, J.Y., Kelly, M.J., Dykoski, C.A., Li, X., 2005. The Holocene Asian Monsoon: links to solar changes and North Atlantic climate. Science 308, 85–857. Wang, Y., Cheng, H., Edwards, R.L., Kong, X., Shao, X., Chen, S., Wu, J., Jiang, X., Wang, X., An, Z., 2008. Millennial- and orbital-scale changes in the East Asian monsoon over the past 224,000 years. Nature 451, 1090–1093. Whitlock, C., Larsen, C., 2001. Charcoal as a fire proxy. In: Smol, J.P., Briks, H.J.B., Last, W.M. (Eds.), Tracking Environmental Change Using Lake Sediments. 3. KluwerAcademic Publishers, Dordrecht, pp. 43–67. Wu, Z.Y., 1980. Vegetation of China. Science Press, Beijing 825 (in Chinese). Wu, W.X., Liu, T.S., 2004. Variations in East Asia Monsoon around 4000 a B. P. and the collapse of Neolithic cultures around Central Plain. Quat. Sci. 24 (3), 278–284 (in Chinese with English abstract). Wu, L., Wang, X.Y., Zhang, G.S., Xiao, X.Y., 2008. Vegetation evolution and climate change since the Holocene recorded by pollen-charcoal assemblages from lacustrine sediments of Chaohu Lake in Anhui Province. J. Palaeogeogr. 10 (2), 183–192. Xia, Z.K., Deng, H., Wu, H.L., 2000. Geomorphologic background of the prehistoric cultural evolut ion in the Xar Moron River basin, Inner Mongolia. Acta Geograph. Sin. 55 (3), 329–338 (in Chinese with English abstract). Xiao, J.Y., Lü, H.B., Zhou, W.J., Zhao, Z.J., Hao, R.H., 2007. Evolution of vegetation and climate since the last glacial maximum recorded at Dahu peat site, South China. Sci. China Ser. D Earth Sci. 50, 1209–1217. Xu, J.X., Zheng, Z., Huang, K.Y., Yue, Y.Y., Li, J., Chase, B.M., Ledru, M.P., Carré, M., Cheddadi, R., 2013. Impacts of human activities on ecosystems during the past 1300 years in Pingnan area of Fujian Province, China. Quat. Int. 286, 29–35. Yancheva, G., Nowaczyk, N.R., Mingram, J., Dulski, P., Schettler, G., Negendank, J.F.W., Liu, J.Q., Sigman, D.M., Peterson, L.C., Huang, G.H., 2007. Influece of the intertropical convergence zone on the East Asian monsoon. Nature 445, 74–77. Yang, X., Scuderi, L., 2010. Hydrological and climatic changes in deserts of China since the Late Pleistocene. Quat. Res. 73, 1–9. Yin, Q., Zhu, C., Ma, C.M., Wang, F.B., Zhao, Z.P., Zheng, C.G., Huang, L.Y., Tian, X., 2006. Holocene climate change recorded in peat humification in Tianmu Mountain region. Mar. Geol. Quat. Geol. 26 (6), 117–122 (in Chinese with English abstract). Yue, Y.F., Zheng, Z., Huang, K.Y., Chevalier, M., Chase, B.M., Carré, M., Ledru, M.-P., Cheddadi, R., 2012. A continuous record of vegetation and climate change over the past 50,000 years in the Fujian Province of eastern subtropical China. Palaeogeogr. Palaeoclimatol. Palaeoecol. 365–366, 115–123. Zhang, J.W., Chen, F.H., Holmes, J.A., Li, H., Guo, X.Y., Wang, J.L., Li, S., Lü, Y.B., Zhao, Y., Qiang, M.R., 2011. Holocene monsoon climate documented by oxygen and carbon isotopes from lake sediments and peat bogs in China: a review and synthesis. Quat. Sci. Rev. 30, 1973–1987. Zheng, Z., Wang, J.H., Wang, B., Liu, C.L., Zou, H.P., Zhang, H., Deng, Y., Bai, Y., 2003. Highresolution records of Holocene from the Maar Lake Shuangchi in Hainan Island. Chin. Sci. Bull. 48 (5), 497–502. Zong, Y., Chen, Z., Innes, J.B., Chen, C., Wang, Z., Wang, H., 2007. Fire and flood management of coastal swamp enabled first rice paddy cultivation in east China. Nature 449, 459–462.