Environmental changes and ancient long-term wet field management schemes: A preliminary case study of the Daepyeongri site,locality 3-1-D,South Korea

Abstract

Large-scale excavations between 2012 and 2015 at the Daepyeongri site on the floodplain of the River Geumgang have revealed the presence of an ancient complex settlement comprising houses, extensive agricultural fields, and pits that have been dated to between the Bronze Age and the early Three Kingdoms periods. The beginning and end of the occupation of this site and land-use patterns are assumed to have been associated with its natural environment. The pollen records presented in this study show that wet conditions continued throughout the time that this site was occupied, evidenced by the consistent appearance of Alnus, while there are nevertheless hints of some land reclamation from the Bronze Age onwards. More intensive agricultural activities that took place during this later period are evidenced by an increase in the abundance of NAP pollens related to cultivation including Gramineae, Cyperaceae, Bistorta, Compositae and Fagopyrum. At the end of the occupation period, an abrupt transition to wetter conditions is recognized while dry land condition had been prevalent for some time in other parts of the site. Microscopic examination of buried cultivated soils that evidence multiple phases of ancient field systems has enabled the identification of pedological traces of discrete cultivation patterns and shows that they changed over time. The recognition of micro-structures and associated features shows that seasonally flooded wetland was first utilized during the early phase (the Bronze Age), and that more intensified irrigation management was seen during the late phase (the time span encompassed by the proto-Kingdoms and Three Kingdoms periods). These data reveal evidence for intensive hydromorphic degradation and enable a robust recognition of settlement history and an enhanced understanding of the intensity of various land-use patterns, and landscape changes from both environmental and archaeological perspectives.

Keywords

ancient paddy field cumuli A horizon human–environment interaction mid–late-Holocene palynological analysis settlement history silty clay concentration feature soil micromorphology South Korea

Introduction

Wetlands as sites for human habitation

Wetlands have long provided the fertile sites for human habitation as they are characterized by easy access to water and other abundant natural resources and thus are good places to find food (Brown, 1997). Sites in these areas usually attract a large population and are capable of sustaining long-term occupation. At the same time, however, despite the value of these environments, the dangers that threaten stable life on wetlands can include flooding and rising groundwater levels. Entire settlements and the urban living quarters of past societies are very often discovered buried under flood deposits. Thus, in contrast to the advantages of these areas, populations in wetlands are also subject to the whims of nature and these environments can therefore be considered as frontiers where nature and human society meet and have consistently and actively interacted with one another throughout the history of settlements.

Viewed from an archaeological perspective, such characteristics mean that wetlands are a rich archive of past human societies as they have been favoured locations for settlements and boast excellent preservational conditions. On the Korean peninsula, wetland environments have yielded a great deal of archaeological data and are thus a significant resource. These settings also provide good opportunities for environmental research on past societies and enhance our understanding of how humans and nature have interacted over time.

In terms of the various environmental approaches that can be applied to investigate wetland areas in the archaeological context, macro-scale geomorphological reconstructions, such as river channel meanders and land formation processes, provide one suite of methods while the micro-scale settlement catchment area can be another focus for research. The latter in particular is appropriate for addressing how a settlement unit within a human society lived and utilized its surrounding environment through subsistent patterns, including agriculture. It is also possible to evaluate the experience of the population and determine reactions to environmental changes over the course of the life span of an individual at the generational level (Dincauze, 2000). In this context, recent excavations at the Daepyeongri provide an opportunity to perform such a site-level environmental survey.

The objective of this study is to understand the long-term vegetational and environmental changes that characterized the Daepyeongri settlement from the beginning to the end of its prolonged occupation and to identify how human society, especially agricultural practices, became embedded in the area and adapted to the changing environment over time. The main methods employed to achieve these research aims were palynological and soil micromorphological analyses. Although a wide variety of analytical approaches are available for environmental reconstruction, the most commonly applied environmental proxy is the pollen record, as it can be used to detect vegetational changes and determine human impacts via the appearance of anthropogenic species as cultural indicators. Soils from the activity areas within this settlement were also subjected to analysis to assess modes of cultivation and intensity as well as the evolution of cultural layers. In previous work, Kooistra and Kooistra (2003) emphasized the contribution of distinct disciplines to studies such as this, as palynology can be applied to reveal vegetation history, climate change and settlement contexts, while soil micromorphology is useful for reconstructing ancient land use and for identifying anthropogenic processes in archaeological research. Applied together, these augment one another and enable the generation of multiple and abundant lines of evidence, especially when used in tandem with phytolith analysis.

Study area

The Daepyeongri site (36°29’33”N-36°30’14”N, 127°15’47”E-127°16’18”E) is located within Sejong city on the southern floodplain of the River Geumgang(Geum). In this area, the river flows from northeast to southwest and forms a large floodplain surrounded by low-lying hills. A series of previous archaeological surveys have demonstrated evidence for occupation in this region that dates back to the Paleolithic period; several sites featuring Paleolithic stone tools dated to around 19,000 BP have been found in this region. However, subsequent to the Neolithic where human traces are relatively rare, an abrupt increase in settlement level appears to have occurred from the Bronze Age (between the 15th century BC and the 3rd century BC) and includes burials such as dolmens. During the early Iron Age (between the 3rd century BC and the 1st century BC), small pit burials were constructed in this region, while during the Proto-Three Kingdoms period (between the 1st century BC and the 3rd century AD) and the Three Kingdoms period (between the 3rd century and the 7th century AD), Sejong city developed into a local political centre as a confluence of roads and defensive fortresses. A number of settlements and an urban centre referred to as the Naseongri site are known to encompass a defensive fortress constructed in the outskirts of mountains. It is therefore assumed that this area acted as an important hub as part of both the national land-road system and the route for water transportation along the River Geumgang with local kilns, workshops, a fortress and administrative building complexes known from the Goryeo period in this region (between the10th century and the 14th century AD). This region then consistently maintained its local identity, including the establishment of a governing council established during the subsequent Joseon period (between the 14th century AD and the 19th century AD) (KIAE, 2017).

The levee of the River Geumgang is orientated along a NW to SE axis parallel to the river channel, and 3 m higher than the current water level, because this region has experienced large-scale development related to the Sejong Special Self-Governing City national construction project. A number of commercial archaeological survey organizations have participated in joint rescue investigations at this site. Thus, the northern side of the Daepyeongri wetland, denoted as 3-1-D district, was excavated by the Korean Institute for Archaeology and Environment over 11 sessions between 2012 and 2015 (Table 1).

Table 1.

Excavated archaeological features of each survey stage.

Excavationsession	Excavated archaeological features
1st	Bronze Age: house structures, paddy field; Proto-Three to Three Kingdoms: house structures, overground building, pits, ditches, dryfield, road features; Unified Silla: house structures; Joseon: ditches
2nd and·3rd	Bronze Age: house structures; Proto-Three to Three Kingdoms: house structures, overground buildings, pits, ditches, burials, dryfield, road features; Unified Silla: house structure, dryfield; Joseon: ditches
4th	Proto-Three to Three Kingdoms: burials
5th	Bronze Age: paddy fields; Proto-Three to Three Kingdoms: dryfield, Unified Silla: dryfield; Goryeo: dryfield
6th	Bronze Age: house structure, pits; Proto-Three to Three Kingdoms: house structure, overground buildings, pits, ditches

The basin in this area formed from sediments deposited by the River Geumgang during the last Quaternary and comprises a total of 20 small confluent tributaries. Adjacent to Sejong city area, a stream called the Mihocheon and some tributaries, join the River Geumgang to form a large floodplain. The southern floodplain of the River Geumgang, including the study area is a secondary modified terrace comprising convex and convolute shapes formed from incipient sediments of gravel, sand and silt; these geomorphologies and sediments mean that this meandering river channel has high sinuosity. The channel becomes significantly narrower down the river, 7 km away from the Daepyeongri site. These differences in formation processes explain the lack of a flat and extensive floodplain in this region as well as the presence of a natural levee system. The residential area within the 3-1-D district was situated on this natural levee and onto the backswamp of the southern floodplain. Thus, this complex settlement covering an area of 170.133 m² represents a time span of dwelling and farming activities during the time between the Bronze Age and the Medieval Period (between 1000 BC and 1000 AD). Subsequent to a long period when there was no human activity, archaeological evidence suggests that this site was then used partly as a field plot during the Goryeo period (between the 10th century AD and the 14th century AD) and the Joseon period (between the 14th century AD and the 19th century AD).

Archaeological surveys indicate that land-use patterns varied according to the micro-landscape. The fully excavated area at this site comprises two slightly elevated sections that have relatively extensive flat surfaces and are separated by a large palaeochannel that turns to the base of the valley. Parallel to another deep and large palaeochannel, a series of small braided stream channels were also formed in this area. Infilled old channels have been used as cultivated fields since the Bronze Age. Settlements comprising houses and later phase roads have been discovered on the southern high points, dated to Bronze Age, Proto-Three Kingdoms, Three Kingdoms and Unified Silla periods. On the northern side of the site, a large number of distinctive tombs with surrounding ditches and jar tombs dating to the Proto-Three Kingdoms period and pits have been unearthed. Data show that the distribution patterns of these settlements changed from the crescent shape in the Bronze Age to extended and elongated round shapes as populations increased and building structures became more complex during the Proto-Kingdoms and the Three Kingdoms periods.

Agricultural fields comprising multiple phases have also been recognized within the base of this curve-shaped valley (Lee, 2016). The outlines of the cultivated fields vary during the Bronze Age, field systems comprised the extensive paddy field plots that measured between 96 and 1474 m² and utilized the inner side of the palaeochannel. In contrast, later Proto-Three Kingdoms and Three Kingdoms periods land use comprised dry fields. In particular, field surfaces from the early Three Kingdoms period are characterized by deep furrows and ridges, indicative of cultivation methods used during the terminal phase of occupation. It is assumed, however, that siltation of these palaeochannels prevented paddy rice farming and so these plots were eventually switched to dryfield cropping; the paddy fields used at this time were constructed in the backswamp area of the floodplain, on the slope of the natural levee. Although a variety of analytical reports were contracted and commissioned as a part of the original site report monograph, we only discuss the results of the palaeobotanical and soil micromorphological analyses in this study. The interim results were initially presented in the site report monograph (KIAE, 2017), but we adjusted them and improved the environmental reconstruction with two distinguished lines of research together as well as more supplementary data in this paper.

Materials and methods

Two distinctive outcomes of this research are explored in this paper. First, evidence for agricultural impact is identified based on the preservation of relict archaeological features, while second, a time-span perspective of ancient agricultural impacts on soil and vegetation is developed. The data that provide the basis for this study are the results of a geoarchaeological investigation associated with a rescue excavation at the Daepyeongri site (Figure 1) that was carried out by the Korean Institute for Archaeology and Environment between 2012 and 2015.

Figure 1.

Location of Daepyeongri. (a) Map of South Korea, showing the location of Dapyeongri. (b) Map of the Daepyeongri site 3-1-D district.

Palynological, phytolith and loss-on-ignition analyses

As discussed, pollen is a standard proxy used to identify changes in vegetation. In particular, accumulated data from pollen analyses haves been widely used to reconstruct climate and palaeoenvironmental changes (Kim et al., 2002; Park, 2011) at the macro-scale across Korea. For archaeological use, the higher occurrence of Gramineae pollen is often regarded as a useful parameter for popularization of rice farming, while the appearance of non-arboreal pollen such as Cyperaceae, Chenopodiaceae, Fagopyrum and Artemisia are secondary indicators of anthropogenic activities (Yoon et al., 2008). A detailed palynological analysis was carried out, and although sampling locations ranged from within naturally deposited soil profiles to field plots and the infill of archaeological features, just cross-section profiles from ancient arable field districts containing natural substrates are presented and discussed in this paper. A complete dataset, including the results of analyses from infills and other specific archaeological features investigated to identify domestic use, is attached as supplementary data (available online). Phytolith analysis was also performed to complement the results of palynological analysis. Phytoliths are silica bodies produced in plants, are resilient to weathering and can be dispersed over short distances from parent plants. Complementary to palynological analyses, Gramineae were identified as Oryza, Phramites, Zea, Miscanthus and the like.

Samples were collected from 11 localities over consecutive sessions (Figure 2). Two of these were domestic in context, while nine comprised natural stratigraphies and cross-sectioned soil profiles dug within agricultural fields. Specifically, five of these nine samples were collected from undisturbed soil profiles from the subsoil to the A horizon, while the additional three samples were collected from grey soils that lack clear context and associated agricultural fields (Table in Figure 2). Thus, only samples of five locations (Loc.2-1, 2-2, 5-1, 5-2, 5-3) collected from undisturbed soil profiles (Figure 3a) are discussed in this paper. Each sample was collected within a 10-cm interval and pollen grains were extracted using the method proposed by Erdtman (1934). In brief, 10 g of sediment was treated with 10% KOH and an acetolysis liquid in a 15 ml round centrifuge tube, pollens grains were concentrated via specific gravity using ZnCl₂. Pollen residues were then extracted using Pasteur pipettes dipped in glycerine and were prepared on slides; both pollens and spores were counted using an Olympus BX51, the transmitted light microscope at 200 and 500 times magnification. Pollen identification is based on the reference collection archived by Nakamura (1989) and Chang (1986); a pollen percentage diagram was plotted using the software GRAPHER 3. Pollens and spores are mostly determinable and well preserved with very rare mechanically damaged ones.

Figure 2.

Plan of the Daeyeongri site, showing the locations of all the archaeological features and samples for analyses. Insert: table showing the analysis applied for each location (modified from KIAE, 2017).

Figure 3.

Soil profiles and sampling locations in this study. (a) Soil profiles and sampling locations for palynological, phytolith and loss on ignition (LOI) analyses discussed in the text. (b) A Soil profile and drawing of sampling locations for soil micromorphological analysis (see Table 2 for more details).

The glass bead method was used for phytolith extraction (Hiroshi, 1976). In brief, 2 g of sediment was taken and a contamination treatment was carried out by repeatedly adding distilled water and removing and adding 0.01 g glass beads per gram for counting. Residues were then dehydrated in a dry oven and prepared on slides with mounting mediums. The protocol used for loss-on-ignition followed the method of Zhong et al. (2010) and Xu et al. (2013) and was modified from Ball (1964). Water content was measured after drying at 100°C, and the total organic carbon (TOC), organic contents and loss on ignition was measured at 550°C in an electronic furnace for 6 h (for more detailed measuring process, see supplementary data S7, available online).

Soil micromorphological analysis

Soil micromorphology plays a crucial role as a proxy for identifying the processes of soil formation and the post-depositional alteration of soil-sediment sequences. It is used to identify the various components of the soil matrix as well as associated pedogenic processes (Schaetzl and Anderson, 2005; Stoops et al., 2010). Thus, this approach provides an opportunity to examine the soil at a higher resolution in addition to field observations. Micromorphology is therefore commonly utilized in geomorphological research to understand pedogenesis, landforms and hydrological changes (Srivastava and Sauer, 2014). Interpretations of ancient agricultural practice can be based on analogies between the pedofeatures produced by modern experiments, and soil conditions as well as the examination of ancient buried agricultural soils (Gebhardt, 1993; Goldberg and Berna, 2010; Goldberg and Macphail, 2006; Simpson, 1997; Usai, 2001). Sample locations were selected, based on stratigraphic units identified in the field that were related to cultivation features (Table 2, Figure 3b); block samples from sample sites were processed into thin sections by a petrographic company and were examined using Olympus Bx 51 microscope. Soil micromorphological descriptions were made under the petrographic polarizing microscope following the guidelines presented by Bullock et al. (1985) and Stoops (2003). A summary of the results of this analysis is presented in Table 3.

Table 2.

Field Descriptions of the soil profile (location of soil micromorphological samples).

Depth(cm)	Soil	Soil classification	Archaeological features
0–100	10 yr 6/8 (brown, moist) silty clay, sharp boundary	A
100–130	N 6 (grey, moist) clay loam, smooth boundary, buried feature with depression, covered with pale clay lens (few centimeters)	2Ab	Early Baekje period horizon
130–170	10 yr 4/6 (reddish brown, moist) silty clay, sharp boundary	Hf
170–210	N 6 (gray, moist) clay loam, smooth boundary	GC
210–250	N 5 (dark, moist) clay loam, sharp boundary	1Ab	BA horizon paddy horizon
250–300	2.5 yr 8/6 (yellow) brown loam	C2
300–350	2.5 yr 5/6 (yellowish brown) silty loam	C1

Table 3.

Summary of the results of soil micromorphological analysis.

			Thin section(Depth: cm)		MS#		Voids				Groundmass				Iron/depletion features
			Ch^##	Vg^###	c/f ratio (10 ㎛)	Rel. Distr.^## ^##	Fe-Mn nodule	Fe-Mn staining	Dep. Gr mass^### ^###	H/Q coatings^#######	Dusty silty clay	Clay intercal-ation	Com. Infilling^########	SCCF	Macro remains	Micro remains (phytolith)	Micro charcoal	Humic matter
5		M	-	pc + + (4 mm)cv + + + (5 mm)	20:80	p	+ + + + (5 mm)	+ + + +		+ + (5 mm)	+	+ +	+ +	+ + + (1 cm)	+	+ + + +	+ + + +	+ +
4	u	m	-	p + pc +	80:20	e	+	+ +		+	+ +			+ +	–	+ + +	+ +	–
	l	m		pc + + p + cv + + cb + +	70:30	e	+	+ + + + (5 mm)			+ + + (4 mm)	+ +	+ + + (8 mm)	+ + +	+	+ + +	+	+
3	u	m		pc + + (4.5 mm)cv + p +	20:80	p	+ + + + +	+ + + +		+		+ +	+	+ + + +	+	+ + +	+ + +	+
	l	m		p +	10:90	p	+ + + +	+ + + (1 cm)		+		+ + +		+ + +	+ +	+ + +	+ + + +	+ +
2		m	+ (4 mm)	pc + v +	70:30	p	+ + + (1 mm)	+ + (3 mm)	+ +	+ + (3 mm)	+ + (3 mm)	+ +		+	+	+ + +	+ + +	+ +
1		m	+	pc + + p +	70:30	p	+ + (3 mm)	+	+ +	< +	+				+ +	+ + +	+ +	+ + +

MS: microstructure; pt: platy; ms: massive.

＃＃

CH: channel.

###

p: planar; bk: blocky; pc: polyconcave; vs: versicle; cv: closed vugh; cb: chamber.

####

p: porphyric; e: enaulic.

#####

notable b: fabric; g-s: grano-striated; p-s: parallel-striated; r-s: random-striated.

######

Dep. GrMass: depleted groundmass.

#######

H/Q coatings: hypo/quasi coatings.

########

Com. Infilling: compound layered infilling.

Degree of occurrence: + – rare (<5%); + + – occasional (<10%); + + + – many (<20%); + + + + – abundant (20–30%), + + + + + – very abundant (>30%).

The samples used in this study were collected during the second survey during the year of 2012 (KIAE, 2017). Archaeological features sampled include Bronze Age dwelling pits and associated field systems that have a convoluted surfaces and are dated to 3210 ± 30 BP (Beta-346255) as well as sediment samples, dwelling pits and dry field system which have been dated to 2050 ± 30 BP (Beta - 346254). This horizon belongs to the Proto-Three Kingdoms to early Three Kingdoms period (0–3rd century AD) with road features, ditches and dwelling pits of later phases from the Unified Silla (between the 7th century AD and the 9th century AD) to the Joseon period (between the 14th century AD and the 19th century AD). The sampling locality comprised cross-sectioned soil profile situated on the boundary of the excavation unit (Table 2, Figure 3b); the lowermost portion of this profile comprises a brown sandy deposit (C horizon), while above it, a reduced grey clay rich layer containing Bronze Age field plots (1Ab horizon) is deposited, overlain by the organic rich clayey layer (GC horizon). A brighter, thick silty clay horizon that grades to a reddish brown layer (hf horizon) was formed beneath a horizon that preserves early Three Kingdoms (possibly Baekje) period ridges and furrows (2Ab horizon). This surface of ridges and furrows is covered with a thick whitish clay lens, with later, brownish sandy clay loams thickly deposited above (A horizon). Five block samples for micromorphological analysis were collected from key layers and transitions.

Results

Palynological and phytolith analyses including loss on ignition values

The results of pollen and phytolith analyses are presented in Figure 4a and b. The results show that the dominant species are deciduous broad-leaved trees such as Alnus, Quercus and Castanea. The results from the five localities are summarized below. Descriptions of pollen assemblages Zones (I–Ⅳ) are hereafter reported from the oldest/lowest substrate to the youngest/uppermost substrate.

Figure 4.

The results of analyses in Loc 2-1, 2-2, 5-1, 5-2, and 5-3. (a) Diagrams for selected taxa of pollens and LOI contents, denoted in percentage, are presented throughout the sediment core (Loc 2-1 and Loc 2-2) with respect to local pollen zones (PZ) and locations for radiocarbon dating samples. (b) Diagrams for selected taxa of pollens(only spores beyond 100 counts with the total numbers) and LOI contents, denoted in percentage, and phytoliths, denoted in percentage with the total number, are presented throughout the sediment core (Loc 5-1, 5-2, and 5-3) with respect to local pollen zones (PZ).

Levels of pollen concentration vary depending on the preservation capacity of each soil layer (see details in Figure 4a and b). High occurrence of pollens is derived from fine sediments of clay and silt with high organic contents, whereas coarse sediments of sands often yield low pollen counts. In contrast, phytoliths, which are likely less affected by soil-sediment in which they were buried, show the consistent high occurrence in the layers where the pollen counts are relatively low (i.e. PZ. III and IV of Loc. 5). Thus the concentration level of pollen appears to have been affected by acidification of organic matter rather than the abundance level of vegetation. In addition, we assume that the origin of pollen is likely to be local because the pollen diagram shows the peculiarity of this site, different from the general patterns in this region (see more in discussion). On the other hand, the influence of human activity on pollen data and phytoliths assemblage is highly plausible because the locations are inside the ancient field systems (Figure 2).

PZ I stage

The clayey horizon with a gradation of colours (2.5 Y 3/1 and 5 Y 3/1) is characterized by a low concentration of pollen grains and a dominance of spores (97.06–100%). Dominant arboreal pollens are Alnus, Quercus, Castanea/Castanopsis, with the former occurring in highest proportions (16.67–100%). Artemisia (0–100 < %) is the only non-arboreal plant which also reaches the significant percentage. Spores are all monolete types.

Specifically, Loc.2-2 can be subdivided into PZ Ia and PZ Ib stages. The first phase (PZ Ia) was an unstable sedimentary environment and is characterized by the dominance of spores and a few Alnus pollen, while the second phase (PZ Ib) exhibits an increase in Alnus, a species noted for its hygropreference. Over the course of the time period encompassed by these samples, the presence of just a small quantity of pollen grains indicates an unstable land surface that would have been unsuitable for woody plants to become established. Samples yielded the highest concentration of phytholiths, including Andropogoneae (17.11%), Paniceae (18.42%) and phytoliths identified as sponge spicules.

A sponge spicule is a zoophyte and inhabits in places where water circulation is good (Lee et al., 2001). Identifying and counting of sponge spicules was designed to recognize episodes such as influx of water from the River Guemgang. However, its low occurrence and absence do not indicate any meaningful and excessive flooding events.

The LOI (loss on ignition) content ranges between 4.87 and 8.92, TOC ranges between 1.83% and 3.68%, organic contents range between 3.16% and 6.35% and water content ranges between 19.23% and 24.12%.

PZ Ⅱ stage (3,210 BP ~)

The results reveal a significant change in this stage, documented by an increase in various non-arboreal pollens. It is marked by an increase in pollen concentration and the highest values of non-arboreal pollen in the sequence. This time period is characterized by a cross-fluctuating dominance of Alnus (0–100%) and Quercus (0–100%) alongside the appearance of Castanea/Castanopsis. Gramineae (0–100%), Compositae (0–100 < %) and Artemisia (0–100 < %) also occur at significant percentages.

Cultural indicators such as Gramineae, Cyperaceae, Persicaria, Fagopyrum and Chenopodiaceae occur. Monolete type spores are abundant and trilete type spore also occurs. Despite the low concentration of phytoliths, Oryza appears for the first time in this stage. In particular, Loc. 5-1 and 5-2 samples yielded the high concentration of Oryza (0–3.45%) Andropogoneae (3.23–21.21%) and Paniceae (6.90–15.79%). LOI values range between 4.49% and 10.5% (TOC: 1.66–4.41%, organic contents: 2.86–7.6%). The difference between the lowest value and the highest value of LOI and TOC exhibits the greatest range. Water content ranges between 19.56% and 24.52%.

PZ III stage (~2050 BP)

This phase Ⅲ is characterized by an increase in pollen concentration and the highest values of non-arboreal pollen in the sequence. Some localities exhibit more drastic vegetational changes determined by the type of arboreal pollen. Thus, this part is divided into Ⅲa and Ⅲb at localities Loc.2-1, 2-2 and 5-1.

PZ III a stage

Results show that the dominant arboreal pollen is Quercus (0-51.85%) accompanied by a gradual upward increase in Alnus (18.52–100%). NAP reaches 29.73–35.0% while spore occurrence falls back to 55–65%. This phase is defined by the occurrence of the most varied species. Castanea/Castanopsis, Corylus and Pinus occur. Artemisia reaches its highest value. Cultural indicators such as Gramineae, Cyperaceae, Persicaria, Fagopyrum and Chenopodiaceae are identified. Monolete type spores are abundant and trilete type spore also occur. The phytoliths extracted are predominately phytoliths of weeds such as Andropogoneae, Miscanthus, Leersia and Oryza, while woody species were not recovered. Values of LOI range between 4.47% and 6.96% (TOC: 1.65–2.79%, organic contents: 2.84–4.81%), and water contents range between 23.29% and 26.71%.

PZ III b stage

This phase is distinguished by the dominance of Alnus as well as constant occurrence of Quercus and an increase in the abundance of Corylus. Although scarce, the cultural indicators Gramineae, Artemisia and the Chenopodiaceae are present, while the absence of anthropogenic indicator pollens, recovered until the stage below is noticeable. Abundant charcoal fragments are also present in this stage, but these were not quantified at locality Loc 5-1. Only monolete type spores occur throughout this phase and the phytolith assemblage comprises Miscanthus, Phragmites and Andropogonaceae. Phytoliths of Oryza are present in Loc 5-1. Values of LOI range between 3.75% and 6.13% (TOC: 1.32–2.41%, organic contents: 2.27–4.15%). Water contents range between 23.90% and 28.26%.

PZ IV stage

This phase is characterized by the lowest concentration of pollen grains, but a relatively high concentration of phytoliths in the sequence. Thus Alnus (0–100%) is dominant and shares a continuous curve relationship with Corylus. However, Alnus pollen is absent or very rare in Loc 5-2 and 5-3. Except for Artemisia, non-arboreal pollen has definitively decreased. Monolete spores tend to fluctuate according to the depth and particle size of a deposit, Oryza phytoliths were recovered from all localities younger than the middle IV phase. Weeds, woody species and sponge spicules were observed. These data reveal an abrupt increase in Phragmites and sponge spicule phytoliths. Values of LOI range between 2.5% and 6.1% (TOC: 0.75–2.39%, organic contents: 1.29–4.13%) and water contents range from 14.49% to 29.23%. In particular, Loc. 5-2. shows the lowest value.

Soil micromorphology

Although thin sections reveal massive microstructure with a clayey and silty groundmass, certain textural pedofeatural changes depending on the stratiphigraphic units are observed. Thin section DP (Daepyeong) 1 (C2/1Ab horizon; PZ Ⅱ; Figure 5a) shows a mixed and homogenized groundmass with abundant humic matter and micro-charcoal (Figure 6a). In thin section DP 2 (1Ab horizon) textural pedofeatures with more dusty clay features are formed with less humic matter but a similar groundmass. Iron nodules tend to form as smaller nodules and occurred at a higher frequency. Thin section DP 3 (GC horizon; Figure 5b) comprises two parts. The lower part is a silty loam with iron stainings and nodules while the upper part is a sandy silty loam with numerous iron nodules. The convolute boundary between the two parts is marked by aligned iron nodules and stainings. The groundmass is a silt loam and very homogenized. Silty clay concentration features occur as a form of iron stained clay intercalation (Figure 6b). Iron nodules are also occasionally formed. In the upper part, the groundmass texture is sandier and becomes highly heterogeneous with silty clay concentration features as forms of aggregates. Many iron nodules are formed.

Figure 5.

Thin sections: (a) DP1 showing the reduced groundmass with aggregates of iron nodules; (b) DP3 exhibiting various iron nodules.

Figure 6.

Microscopic photographs: (a) groundmass of thin section with some dusty clay features (DC) DP 1 (PPL); (b) aggregate of silty clay concentration feature (SCCF) from thin section DP 3 (PPL); (c) aggregate of SCCF with impregnative iron staining and nodules (Fe) from thin section DP 4 (PPL); (d) textural changes of the groundmass with SCCF from thin section DP 5 (PPL).

Thin section DP 4 (GC horizon/hf horizon; PZⅢ) comprises two parts. The lower part is an iron rich sandy loam and the upper part is a highly homogenized groundmass. The boundary between these two parts is defined by a crust (8 mm) consisting of lenses of dusty clay intercalations and sand grains (Figure 6c). Numerous iron stained silty clay concentration features and dusty clay infillings are present in the lower part. Various pore spaces are formed and there are many polyconcave vughs and vesicles. In contrast, the groundmass in the upper part is homogenized with more sand grains that conform to enaulic c/f-related distribution. The planar voids (0.5 cm) are dominant. The iron rich and clay intercalated aggregates also occur occasionally. The contrast between the lower part and the upper part in terms of sediment size influencing the organic contents and might explain the fluctuation of LOI values (c. 3%) in the corresponding pollen phase of PZⅢ. Thin section DP 5 (2Ab horizon; PZⅢ) also contains abundant iron features including an iron quasi-coating which has developed into coarse nodules in the very homogeneous sand dominated groundmass. Compound infillings of iron clay and sandy lenses occur, while clay intercalations (Figure 6d) occur as simple and serrated infillings, and some vughs are almost completely infilled with sand grains.

Discussion

The pollen and phytolith data presented in this paper from the Daepyeongri site provide a detailed context for changes in vegetation and the surrounding environment and are indicative of ebbing and waning cultivation practices. Soil micromorphological analysis also detects certain soil changes associated with various wetland cultivation modes. An integrated interpretation of multiple analyses was therefore attempted to reconstruct natural environmental conditions and to determine interactions with human society. The results show deciduous broad leaved trees and pollen and spore from land plants dominated since the PZ I stage, although the presence of reclaimed land characterized by yellow sandy silt loam points to stable water conditions and a sediment supply that is more consistent with wetlands. These conditions would have meant that various trees would have grown well.

The onset of cultivation, evidenced by the excavated Bronze Age field at this site, is also reflected by the pollen assemblage of the PZ II stage. Notable changes, such as anthropogenic indicators of human interference like Gramineae, Cyperaceae, Bistorta and Compositae corroborates the fact that wetland was cleared at this time; in particular, the presence of Gramineae is indicative of the possibility of rice farming (Atahan et al., 2007). In addition, incipient edaphic traits of soils characterized by separately deposited sand grains and fine aggregates, rich humic organic matter and abundant iron-manganese features reflects repeated redoximophic conditions (D’amore et al., 2004; Vepraskas and Wilding, 1983). A mixed groundmass of coarse grains and fine particles, possibly the result of churning, as well as dusty clay features and abundant micro-charcoal (thin sections 1 and 2) point to intensive surface disturbances and mixing by cultivation (Macphail et al., 1990). This leads to the assumption that the gley horizon within the sequence is cumulic soil, which had been cultivated while sediment accumulating at the same time resulted in the absence of buried surface of previously cultivated field plots.

It appears that the PZ IIIa stage corresponding to the Proto-Three Kingdoms and gleyed horizon is marked by a sudden increase in NAP pollens. However, the lower part of this section is indicative of a short-lived change in vegetation evidenced by an increase in broad-leaved trees Quercus amid the dominance of deciduous Alnus within the arboreal pollen diagram. The layer between 240 and 140 cm indicates a strong influence of cultivation on pollen assemblages as the proportion of Gramineae increase and common weeds such as Persicaria, Fagopyrum, Compositae and Artemisia. The appearance of Pinus pollen at this site, however, is disputable. Yoon (2012) and Park and Yi (2008) both noted a change from early Neolithic deciduous broad-leaved forest to coniferous-deciduous mixed forest from the middle-Neolithic onwards. In particular, Alnus japonica and Quercus serrata dominated in the early stage of the Neolithic, while Pinus densiflora had spread by later stages. Alnus japonica and Quercus serrata indicate a humid environment and gradually reduced in abundance during the Bronze Age. An alternative explanation is that the abundance of these pollen grains might indicate open landscape rather than signaling the change to a cooling climate because the distribution of Pinus densiflora, previously confined to the hilly area, had expanded to the low lands in the Iron Age (Choi et al., 2005; Yoon, 2012). The trend is in agreement with that seen generally within flora of the Korean peninsula, although there is a discrepancy between general patterns and vegetation changes in Daepyeongri.

The odd combination seen in this study where Pinus appears as secondary forest along with the consistent dominance of Alnus is in contrast with the floodplains of other rivers exhibiting a noticeable decline of Alnus in the Korean Peninsula. This phenomenon can be explained by the local geography of the river Guemgang floodplain as the bottlenecked channel and raised river bed have caused the subsequent discharge of massive deposits leading to heavy flooding in this area (Moon et al., 2017). On the other hand, the occurrence of Fagopyrum pollen can be regarded as anthropogenic in origin (Yi, 2009), and provides further evidence for intensive agricultural activities at this level along with the increase of Pinus.

A discrepancy remains between ₁₄C dates and excavated archaeological features. Major changes are visible in soils. 2Ab horizon containing the ridge and the furrow features belonging to the Proto-Three Kingdoms to Early Baekje (the Three Kingdoms) periods has exhibited a highly iron rich fabric indicating redoximorphic conditions and a significant textural change with dominant silty clay concentration features. These silty clay concentration features exhibiting a pale clay groundmass and speckled silt size particles that lack clear laminations are known to be associated with more controlled irrigated fields as a form of extreme soil hydrological degradation by intensive wetland cultivation, probably paddy rice farming. This specific kind of dusty clay feature also appears to correspond with the development of rice agricultural technology and the start of advanced wet cultivation involving controlled intermittent irrigation schemes (Zhuang et al., 2014) for enhancing the growth of rice roots during the Three Kingdoms and the United Silla periods in the Korean peninsula (Lee et al., 2014). In addition, in some sections (DP 1, 3 and 5), aggregates of iron nodules are observed. Semi-aquatic plants including rice plants generally have respiring root channels resulting in hypocoating forms and nodules are common due to water fluctuation (Lindbo et al., 2010). In particular, it is often observed that iron nodules are formed in a bunch of aggregates in paddy soils (Macphail and Goldberg, 2017: 338–339).

However, excavated field plots with ridge and furrow features point to dryfield cultivation and are not consistent with the paddy rice farming suggested by soil analysis. This discrepancy might be explained by the fact that this horizon occurs under the paddy rice farming regimes and so this field may have been switched to a dry field before burial. Conversion between two different modes of field plots as a form of crop rotation to replenish soil nutrition has been common in Korea since ancient times.

The fact that the underlying gleyed horizon exhibits strong surface disturbance is probably associated with the result of cultivation. There are indications of more intense farming practice in 2bA horizon, as agricultural activities here had been ongoing long-term. These data match with the pollen assemblage of PZ Ⅱ and Ⅲ with a high frequency of anthropogenic indicators. This conclusion is also supported by the presence of a large settlement complex revealed by excavations that spanned the Proto-Three Kingdoms to Three Kingdoms periods. The existence of extensive and intensive land use in this area is clear (Lee et al., 2014).

Data from the PZ IV stage pollen assemblage might provide further clue as to the plausible causes for settlement abandonment. In addition, data from some localities indicate a change to a more flood-prone environment based on an increase in Alnus, whereas open land surface continued to dry, perhaps as a result of both macro-and micro environmental conditions of this site. The peculiar local sedimentation seen within the basin of the River Guemgang is also responsible for creating a prolonged wetland suitable for rice farming, while at the same time, this region is both vulnerable and over-sensitive to even slight changes in water level fluctuation. On the other hand, the more diverse vegetation at some localities, not seen in any underlying horizons, likely indicates that the large-scale construction of a settlement complex and extensive land clearance may have led to diversified micro-landscapes that experienced both wet and dry conditions. We assume that this area was likely affected by increased water levels overall, and that some places (i.e. Loc 2) were likely to flood more easily; thus, the latter was too humid and wet to grow crops while other locations would have tended to stay dry longer. The conflict between cultivation and living conditions thus might have contributed to the abandonment of fields.

Over the transition from post-abandonment to present-day, we assume a higher abundance of plant species that tend to favour open lands compared with cultivated fields, because Gramineae and Artemisia were identified among the NAP pollens over this time period. Large-scale environmental change, especially flooding, can affect the growth of plants, because a rising river level resulted from local geomorphic properties and sediment supply (Moon et al., 2017) might have restrained the growth of plants and eventually might have contributed to decrease in pollen concentration.

Conclusion

The Daepyeongri site study case presented in this paper documents long-term wetland landscape changes associated with the local environment and land-use patterns. Extensive excavations in this region have provided the opportunity to detect environmental changes within the site based on preserved micro-remains, including pollens and phytoliths. Soil micromorphological investigation adds information augmenting the identification of various levels of agricultural practices. This research therefore provides an important empirical dataset of soils altered by ancient wetland cultivation practices.

The results of palynological analysis can be matched with previously established hypothesis of the extensive environmental reconstruction. Data on the Neolithic vegetational conditions of Chungcheongnam Province where the Daepyeongri site belongs to, were previously presented by Yoon (2012), who constructed a Holocene isopollen map for Korea. In even earlier work, Park and Yi (2008) noted that the flora of the Korean peninsula has changed from deciduous broad-leaved forest in a cool and temperate climate in the early stage of the Neolithic to coniferous-deciduous mixed forest since the middle-Neolithic onwards. Such a broad scale of vegetation assemblage is also similarly observed in this site-level palynological analysis.

Summing up, pollen records outlined here are indicative of oscillations in wet conditions derived from high water contents due to its regional geomorphological characteristics such as the bottlenecked river channel resulting in a raised riverbed and increased sedimentation while Alnus remained the dominant tree species for most of the study period. The Bronze Age horizon discussed here is marked by a reclaimed wetland and is evidenced by increase in Gramineae, in contrast to the pre-Bronze Age phase when only Alnus was dominant. The horizon corresponds to the Proto-Three Kingdom–the Three Kingdoms (Baekje) period, the focus of this article; despite the lack of excavated field features in some areas, both the results of soil micromorphological analysis and pollen data point to the presence of the cultivated fields at this time. The mixed groundmass with a high frequency of silty clay concentration features observed in thin section 3 may indicate that wetland cultivation is consistent with an increase in Gramineae and common weeds such as Persicaria, and Fagopyrum, Compositae and Artemisia in the pollen record of the IIIa stage. At the same time, textural changes including silty clay concentration features are first seen in the early historical cultivated layer, whereas the Bronze Age field horizon does not contain any evidence for this transition. The fact that this gleyed cumulic horizon is constantly present in the upper part of the section suggests enhanced water control and drainage, a hypothesis that is corroborated by excavated ridges and furrows present in the same horizon.

Pollen data alone reveal environmental conditions subsequent to the burial of Baekje period cultivated fields. It is devoid of high concentrated anthropogenic indicators and it reflects the decreased or occasional human impacts until the modern times. We assume that this land reverted to conditions that did not favour intensive farming, so that yielding of pollen is very low. It is noteworthy that there are some variations in the dry/wet conditions by locality, while at the same time, reflecting the fact that adjacent fields were available for land use and the presence of Pinus indicates the formation of secondary forest.

The problem with recognizing ancient land-use patterns associated with agriculture has been mitigated in recent years via achievements in excavation and archaeobotanical research. Nevertheless, one key line of evidence for understanding agricultural land use is often limited to the preserved buried field systems. This is, in part, due to the complexity of agro-soil systems such that agricultural soil is often accreted, and thus forms an overthickened A horizon (Homburg and Sandor, 2011). This process means that often just the last field plots prior to the burial are recognized in archaeological surveys and long-term cultivation just beneath is ignored. Such a perception constrains our understanding of the reality of ancient agricultural practices, our detection of the remnants of dynamic land-use schemes and information on the various levels of technological development in agricultural practices imprinted on soils.

The combined approach of soil micromorphology, palynological and phytolith analyses has proven to be useful in reconstructing site-level environmental changes as their data complement the results of excavation by providing information on cumulic soils, with or without, preserved buried field features. Thus, the combined use of microscopic and palynological analyses is proposed as an effective research tool to obtain multiple lines of evidence regarding soil and vegetational changes as well as to complement the excavation data. These approaches might even help to build understanding of how extensive pedosequences are shaped by various land use and cultivation schemes changes. However, more integrated palaeo-environmental research will be required to establish a robust environmental reconstruction for individual settlements and to identify the conditions necessary for human inhabitation, land-use patterns, the intensity of human impacts and site abandonments due to environmental or anthropogenic factors.

Supplemental Material

S1._DPR_Loc.1_180723 – Supplemental material for Environmental changes and ancient long-term wet field management schemes: A preliminary case study of the Daepyeongri site, locality 3-1-D, South Korea

Supplemental material, S1._DPR_Loc.1_180723 for Environmental changes and ancient long-term wet field management schemes: A preliminary case study of the Daepyeongri site, locality 3-1-D, South Korea by Youngrong Moon, Hong-Jong Lee and Heejin Lee in The Holocene

Supplemental Material

S2.DPR_Loc.3_180723 – Supplemental material for Environmental changes and ancient long-term wet field management schemes: A preliminary case study of the Daepyeongri site, locality 3-1-D, South Korea

Supplemental material, S2.DPR_Loc.3_180723 for Environmental changes and ancient long-term wet field management schemes: A preliminary case study of the Daepyeongri site, locality 3-1-D, South Korea by Youngrong Moon, Hong-Jong Lee and Heejin Lee in The Holocene

Supplemental Material

S3._DPR_Loc.4-1_180723 – Supplemental material for Environmental changes and ancient long-term wet field management schemes: A preliminary case study of the Daepyeongri site, locality 3-1-D, South Korea

Supplemental material, S3._DPR_Loc.4-1_180723 for Environmental changes and ancient long-term wet field management schemes: A preliminary case study of the Daepyeongri site, locality 3-1-D, South Korea by Youngrong Moon, Hong-Jong Lee and Heejin Lee in The Holocene

Supplemental Material

S4._DPR_Loc.4-2_180723 – Supplemental material for Environmental changes and ancient long-term wet field management schemes: A preliminary case study of the Daepyeongri site, locality 3-1-D, South Korea

Supplemental material, S4._DPR_Loc.4-2_180723 for Environmental changes and ancient long-term wet field management schemes: A preliminary case study of the Daepyeongri site, locality 3-1-D, South Korea by Youngrong Moon, Hong-Jong Lee and Heejin Lee in The Holocene

Supplemental Material

S5._DPR_Loc.6-1_180723 – Supplemental material for Environmental changes and ancient long-term wet field management schemes: A preliminary case study of the Daepyeongri site, locality 3-1-D, South Korea

Supplemental material, S5._DPR_Loc.6-1_180723 for Environmental changes and ancient long-term wet field management schemes: A preliminary case study of the Daepyeongri site, locality 3-1-D, South Korea by Youngrong Moon, Hong-Jong Lee and Heejin Lee in The Holocene

Supplemental Material

S6.DPR_Loc.6-2_180723 – Supplemental material for Environmental changes and ancient long-term wet field management schemes: A preliminary case study of the Daepyeongri site, locality 3-1-D, South Korea

Supplemental material, S6.DPR_Loc.6-2_180723 for Environmental changes and ancient long-term wet field management schemes: A preliminary case study of the Daepyeongri site, locality 3-1-D, South Korea by Youngrong Moon, Hong-Jong Lee and Heejin Lee in The Holocene

Supplemental Material

S7.Supplementary_LOI – Supplemental material for Environmental changes and ancient long-term wet field management schemes: A preliminary case study of the Daepyeongri site, locality 3-1-D, South Korea

Supplemental material, S7.Supplementary_LOI for Environmental changes and ancient long-term wet field management schemes: A preliminary case study of the Daepyeongri site, locality 3-1-D, South Korea by Youngrong Moon, Hong-Jong Lee and Heejin Lee in The Holocene

Footnotes

Acknowledgements

The authors are thankful to the Korean Institute for Archaeology and Environment’s excavation team and Ji-hoon Yang and In-gun Choi, who were managing the Dapyeongri site as senior researchers for their support for sampling and providing excavation data. They appreciate Prof Soon-Ock Yoon (Kyung Hee University) and Prof Sangill Hwang (Kyungpook National University) for academic advice on geomorphological interpretation. In addition, they express their sincere gratitude to the reviewers and editors for their advice which improved the quality of this paper.

Funding

This article is supported by the Korean Institute for Archaeology and Environment.

ORCID iD

Moon Youngrong

References

Atahan

Grice

Dodson

(2007) Agriculture and environmental change at Qingpu, Yangtze delta region, China: A biomarker, stable isotope and palynological approach. The Holocene 17: 507–515

Ball

(1964) Loss-on-ignition as an estimate of organic matter and organic carbon in non-calcareous soil. Journal of Soil 15: 84–92.

Brown

(1997) Alluvial Geoarchaeology: Floodplain Archaeology and Environmental Change. Cambridge: Cambridge University Press.

Bullock

Fedoroff

Jongerius

et al . (1985) Handbook for Soil Thin Section Description. Wolverhampton: Waine Research Publications.

Chang

(1986) Illustrated Flora & Fauna of Korea (Pollens). Seoul: Ministry of Education (in Korean).

Choi

K-R

Kim

K-H

Kim

J-W

et al . (2005) Vegetation History since the mid-late glacial from Yeongsan River Basin, Southwestern Korea. Journal of Ecology and Environment 28: 37–43.

D’amore

Stewart

Huddleston

(2004) Saturation, reduction, and the formation of iron–manganese concretions in the Jackson-Frazier wetland, Oregon. Soil Science Society of America Journal 68: 1012–1022.

Dincauze

(2000) Environmental Archaeology: Principles and Practice. Cambridge: Cambridge University Press.

Erdtman

(1934) Über die Verwendung von Essigsäureanhydrid bei Pollenuntersuchungen. Svenska Botanica Tiddskrift 28: 354–361.

10.

Gebhardt

(1993) Micromorphological evidence of soil deterioration since the mid-Holocene at archaeological sites in Brittany, France. The Holocene 3: 333–341.

11.

Goldberg

Berna

(2010) Micromorphology and context. Quaternary International 214: 56–62.

12.

Goldberg

Macphail

(2006) Practical and Theoretical Geoarchaeology. Oxford: Blackwell Science Ltd.

13.

Hiroshi

(1976) A basic study on plant opal analysis method(1): Specimens and quantity analysis of Gramineae plants (プラント•オパール分析法の基礎的硏究(1)-數種イネ科植物の珪酸体標本と定量分析法). Archaeology and Natural Science 考古學と自然科學 9: 15–29.

14.

Homburg

Sandor

(2011) Anthropogenic effects on soil quality of ancient agricultural systems of the American Southwest. Catena 85: 144–154.

15.

KIAE (2017) Report of the Dapyeongri site (Excavation for Seojong Governmental Office Complex District 3-1-D). In: Lee

(ed.) Report of Excavation. Daejeon: Hyundae.

16.

Kim

Park

Yang

et al . (2002) Formation environment of quaternary deposits and palynology of Jangheung-ri archaelogical site (Jiphyeon County, Jinju City), Korea. Korean Journal of Quaternary Research 16: 9–21.

17.

Kooistra

(2003) Integrated research in archaeology using soil micromorphology and palynology. Catena 54: 603–617.

18.

Lee

D-J

Kwon

Choi

(2001) Fossil sponges from the Makkol Formation (Lower Ordovician), Taebaek Group, Korea. Journal of Paleontological Society of Korea 17: 77–85.

19.

Lee

H-J

(2016) Landuse patterns of archaeological sites located on the natural levee: A study case of the Daepyeongri site. Jungang Gogo 20: 2–23.

20.

Lee

H-J

French

Macphail

(2014) Microscopic examination of ancient and modern irrigated paddy soils in South Korea, with special reference to the formation of silty clay concentration features. Geoarchaeology 29: 326–348.

21.

Lindbo

Stolt

Vepraskas

(2010) Redoximorphic features. In: Stoops

Marcelino

Mees

(eds) Interpretation of Micromorphological Features of Soils and Regoliths. Oxford: Elsevier, pp. 129–147.

22.

Macphail

Goldberg

(2017) Applied Soils and Micromorphology in Archaeology. Cambridge: Cambridge University Press.

23.

Macphail

Courty

Gebhardt

(1990) Soil micromorphological evidence of early agriculture in north-west Europe. World Archaeology 22: 53–69.

24.

Moon

Yoon

S-O

Hwang

(2017) The geomorphic characteristics and late-Holocene paleoenvironmental reconstruction in the right bank of Geumgang in Sejong-si. Journal of the Korean Geomorphological Association 24: 1–12.

25.

Nakamura

(1989) Diagnostic Characters of Pollen Grains of Japan (In Japanese). Osaka, Japan: Osaka Museum of Natural History.

26.

Park

(2011) A modern pollen–temperature calibration data set from Korea and quantitative temperature reconstructions for the Holocene. The Holocene 21: 1125–1135.

27.

Park

(2008) Review; Postglacial environments of the Chungnam Province inferred from pollen analysis: With emphasis of change in climate and vegetation together with human impact. Journal of Paleontological Society of Korea 24: 55–75.

28.

Schaetzl

Anderson

(2005) Soils: Genesis and Geomorphology. New York: Cambridge University Press.

29.

Simpson

(1997) Relict properties of anthropogenic deep top soils as indicators of infield management in Marwick, West Mainland, Orkney. Journal of Archaeological Science 24: 365–380.

30.

Srivastava

Sauer

(2014) Thin-section analysis of lithified paleosols from Dagshai Formation of the Himalayan Foreland: Identification of paleopedogenic features and diagenetic overprinting and implications for paleoenvironmental reconstruction. Catena 112: 86–98.

31.

Stoops

(2003) Guidelines for Analysis and Description of Soil and Regolith Thin Sections. Madison, MI: Soil Science Society of America.

32.

Stoops

Marcelino

Mees

(2010) Interpretation of Micromorphological Features of Soils and Regoliths. London: Elsevier.

33.

Usai

(2001) Textural pedofeatures and pre-Hadrian’s wall ploughed paleosols at Stanwix, Carlisle, Cumbria, UK. Journal of Archaeological Science 28: 541–553.

34.

Vepraskas

Wilding

(1983) Aquic moisture regimes in soils with and without low chroma colors. Soil Science Society of America Journal 47: 280–285.

35.

Ding

et al . (2013) Speciation of organic phosphorus in a sediment profile of Lake Taihu I: Chemical forms and their transformation. Journal of Environmental Sciences 25: 637–644.

36.

(2009) Archaeopalynological aspect of prehistoric maize pollen from late Holocene of western central Korea: A preliminary report. Journal of the Geological Society of Korea 45: 697–709.

37.

Yoon

S-O

(2012) Holocene vegetation and climatic change inferred from isopollen maps on the Korean Peninsula. Quaternary International 254: 58–67.

38.

Yoon

S-O

Moon

Y-R

Hwang

(2008) Pollen analysis from the Holocene sediments of Lake Gyeongpo, Korea and its environmental implications. Journal of the Geological Society of Korea 44: 781–894.

39.

Zhong

Fan

Liu

et al . (2010) Seasonal variation of potential denitrification rates of surface sediment from Meiliang Bay, Taihu Lake, China. Journal of Environmental Science 22: 961–967.

40.

Zhuang

Ding

French

(2014) Water management and agricultural intensification of rice farming at the late-Neolithic site of Maoshan, Lower Yangtze River, China. The Holocene 24: 531–545.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

1.03 MB

0.13 MB

0.23 MB

0.95 MB

0.86 MB

0.70 MB

0.01 MB