Holocene environmental changes inferred from sedimentary records in the lower reach of the Yeongsan River,Korea

Abstract

Multidisciplinary paleoproxy data from three sedimentary cores (9.05-m-long MW-2, 16.50-m-long IL-3, and 11.88-m-long JD-1) recovered from the Yeongsan River Estuary of Korea are presented. A marine influence can be observed at 10,400 yr BP (−21 m) and 8600 yr BP (−14 to −12 m) around the Yeongsan River Estuary. However, if we take the macro-tidal setting of the Yeongsan River Estuary into consideration, actual sea level could differ from the depositional surfaces, and might be higher or lower than the sediment elevation. Precipitation at around 6300–5000 yr BP is estimated to have been higher or stronger than in earlier and later periods. The core sites received increased terrigenous sediment input during this period because of intensified rainfall and consequent river activity. Although sea level was estimated to be high enough to have a strong influence on the study area during the mid-Holocene, the sedimentary features suggest that riverine activity was the dominant factor controlling the sedimentation patterns. This implies that even during the Holocene Climate Optimum in Korea, roughly 7000–5000 yr BP, the wetter condition has occurred within the interval of 6300–5000 yr BP. After the Holocene Climate Optimum, sea level decreased to 0 to −1 m around 5000 yr BP and increased again starting around 4000 yr BP. The time-series results presented in this study are coincident with global trends, and show the potential for developing geomorphological and climate histories for this region.

Keywords

climate optimum Holocene precipitation Quaternary sea level temperature Yeongsan River

Introduction

Sea level fluctuation and climate variation throughout the Holocene have been plentifully documented. These changes influenced the lifestyle of prehistoric human beings and affected the formation of present topographic features. However, Holocene climate modes (i.e. timing and magnitude) have varied worldwide (Wanner and Bronnimann, 2012). A reliable data set and a reexamination of existing interpretations are required to elucidate the global climate mode during the Holocene.

The sea level positions in Korea during the Holocene still remain unclear. Several authors arguing for an oscillatory sea level curve reported a maximum Holocene sea level 1–2 m higher than present sea level at around 6000 yr BP, with several sea level fluctuations around the Korean Peninsula (Hwang, 1998; Hwang et al., 1997, 2013; Jo, 1980; Lee, 1987; Yoon, 1997). While this highstand is generally accepted, the details have not reached a consensus. The mid-Holocene Warm Period (also known as the Holocene Climate Optimum, or HCO) has not yet been extensively studied in Korea. The period from roughly 9000 to 5000 yr BP is known to have been warmer than present in the Northern Hemisphere (Winkler and Wang, 1993). The warmest period during the Holocene appears to have been roughly 7000–5000 yr BP, particularly in northern Europe and northwest America (Davis et al., 2003). The HCO ended sometime around 6000–5000 yr BP in accordance with a decrease in Northern Hemisphere summer solar insolation. However, the exact timing of the onset and termination of the HCO is still controversial. The difficulty in clearly delimiting the HCO is partly because of the 2–3°C temperature difference between the HCO and the preceding/following periods, which is a smaller difference than glacial–interglacial or stadial–interstadial changes (Jo et al., 2011).

The HCO is also known as the Hypsithermal, Altithermal, Holocene Thermal Maximum, and Holocene Megathermal (Pielou, 1991). These names focus on the temperature aspect of the HCO period. However, significantly wetter conditions in Asia and Africa were also a part of the HCO. Changes in precipitation are of primary concern, particularly in the agricultural and densely populated areas affected by the East Asian monsoon. However, the spatio-temporal pattern of precipitation during the HCO is not clear (An, 2000; An et al., 2000). This may be because of differences in local hydrological processes and geomorphological characteristics (Nahm et al., 2013; Xiao et al., 2008). However, a comprehensive assessment of the seasonality of temperature and precipitation during the HCO has not yet been done (Ganopolski et al., 1998; Hewitt and Mitchell, 1998). For more detailed study of the HCO, we need a high-resolution, precisely dated, continuous record of the monsoon.

Several studies have mentioned the timing of the HCO in Korea: 8000–4000 yr BP in a human society study (Pearson, 1977), 8000–4500 yr BP in a coastal area pollen study (Kong, 1994), 7000–5000 yr BP in a pollen study in a lowland riverine area (Yi et al., 2008), 8000–4500 yr BP in a pollen study in the shelf zone of the East Korean Bay (Evstigneeva and Naryshkina, 2010), 8400–3700 yr BP in a pollen study of swamp deposits (Chung, 2011), 7100–5000 yr BP in a study of upriver valley sediments (Nahm et al., 2011), 8600–5900 yr BP in a stalagmite study (Jo et al., 2011), and 7300–5000 yr BP in a study of coastal wetland deposits (Nahm et al., 2013). The HCO in Korea has generally been described as a warm and humid period, although the timing of the HCO varied in different localities and archives.

The present study provides a detailed description and chronology of three sediment cores (9.05-m-long MW-2, 16.50-m-long IL-3, and 11.88-m-long JD-1) retrieved from the Yeongsan River Estuary, southwestern Korea. Several changes recognized in the sedimentary features through time provide important data for understanding the regional sedimentary processes related to sea level changes, monsoonal discharge, and the paleo-landscape during the early to middle Holocene.

Study area

The Yeongsan River near the southwestern end of the Korean Peninsula has been a locally important water source for agriculture because it crosses through a wide area of rice fields. The Yeongsan River is about 115 km long, originating on Mt Yongchu (560 m) and flowing southwestward to the Yellow Sea (Figure 1). The river has a drainage area of about 3371 km², with an annual mean water discharge of about 16 × 108 m³ (Eun et al., 1998). About one-third of the river basin area consists of arable land, and over half of the basin is covered with forest. The bedrock of the river basin consists mainly of Precambrian gneiss, Paleozoic sedimentary rocks such as shale and mudstone, and Mesozoic granite and tuff.

Figure 1.

Location of study area and drilling points.

The estuary of the Yeongsan River varies in water depth from 10 to 20 m and has an average width of about 2 km. With a population of about 250,000, Mokpo city is located north of the river mouth. The tides in the estuary are principally semidiurnal. The west coast of Korea is almost a macro-tidal regime. Spring tides range from 4 m near Mokpo city in the south to 9 m near Inchon city in the north part of the Korean Peninsula (Nahm et al., 2008). Freshwater discharge mainly occurs during the summer monsoon season. Annual mean data show that more than 80% of the river water is discharged during the summer (Ryu et al., 2004).

The study area is governed by a typical monsoon climate. A cold and dry wind from the north to northeast predominates in the autumn and winter. In summer, warm winds with high humidity blow from the southeast, occasionally catastrophically intensified by typhoons. Storms (>14 m/s maximum wind speed) occur 28.8 days a year, on average, primarily during winter. In contrast, precipitation occurs exclusively during the summer between June and August, with an annual average of 1300 mm (Ryu et al., 2004).

Materials and methods

A standard stationary piston sampler (74-mm inner diameter) equipped with an 85-cm-long brass tube was used for recovering three undisturbed sedimentary cores, MW-2 (34°46′36″N, 126°29′14″E), IL-3 (34°49′04″N, 126°33′15″E), and JD-1 (34°52′19″N, 126°33′31″E) from the Yeongsan River Estuary. Cores were subsampled in the laboratory into 1-cm intervals for grain-size analysis and organic element analysis. Grain-size analysis of the treated bulk sediment was performed using sodium hexametaphosphate as the dispersing agent, with a Mastersizer 2000 laser particle analyzer (Malvern Instruments Ltd, Worcestershire, UK) using the laser diffraction method, which is capable of detecting a 0.02- to 2000-µm particle-size range. Carbonates were removed with 10% HCl, and organic matter was removed using 30% H₂O₂.

For measurements of total carbon (TC), total organic carbon (TOC), total nitrogen (TN), and total sulfur (TS), an aliquot of each of the sediment subsamples was dried to a constant weight at 60°C, ground to a particle size of <0.063 mm, decalcified using 1 N HCl, and then homogenized using an agate mortar and pestle. The concentrations of organic elements were analyzed by combustion at 900°C in a FlashEA-1112 (Thermo Finnigan S.p.A., Milan, Italy), with 2- to 3-mg samples placed in aluminum capsules. TOC/TN and TOC/TS ratios were calculated on an atomic basis.

AMS ¹⁴C ages from bulk carbon in the sediments and plant remains from several levels in the sediment sections were measured at the AMS laboratory of Seoul National University, Korea (irf.snu.ac.kr, SNU), the Korea Institute of Geoscience and Mineral Resources (KIGAM), Korea (http://www.kigam.re.kr, ICa), and Geochron Laboratories, Chelmsford, MA, USA (http://www.geochronlabs.com, GX). The CalPal07 program (http://www.calpal-online.de) was used to convert radiocarbon ages (¹⁴C yr BP) to calibrated years before present (yr BP). Ages of sample levels were determined with linear interpolation between the ¹⁴C dates.

Results and interpretation

Core MW-2

The 9.05-m-long (−15.06 to −6.01 m above sea level, m a.s.l.) core MW-2 was taken in the middle of the main channel of the Yeongsan River. Five sedimentary units were distinguishable in the core based on grain size, color, and sedimentary structures and textures (Figures 2 and 3).

Figure 2.

Simplified radiocarbon-dated sediment strata of core MW-2 with multiproxy data.

Figure 3.

Representative core MW-2 photographs: (a) −15.06 to −14.76 m, (b) −10.37 to −10.07 m, (c) −8.91 to −8.61 m, (d) −7.41 to −7.11 m, (e) −6.31 to −6.01 m; and X-radiographs: (f) −14.76 to −14.52 m, (g) −13.56 to −13.26 m, (h) −8.61 to −8.31 m, (i) −7.28 to −6.98 m.

MW-2 unit 1 (−15.06 to −14.39 m a.s.l., 0.67 m thick, ?−8630 yr BP)

Unit 1 consists of brown (2.5Y 4/2) to dull-yellow (2.5Y 5/3) massive clayey silt. Unit 1 is relatively stiff and semi-consolidated, with abundant vertical color variations, rootlets with iron encrustations, yellowish-brown mottle structures, and vegetal remains in the stiff clayey silt. Unit 1 represents a paleosol. Nahm et al. (2008) reported a paleosol unit in the core MW-1 taken around the Yeongsan River and discussed the age of deposition and pedogenesis.

MW-2 unit 2 (−14.39 to −9.34 m a.s.l., 5.05 m thick, 8630–6220 yr BP)

Unit 2 is a monotonous sequence mostly composed of greenish-gray mud with scattered plant fragments. Several centimeters thick sand layers containing shell fragments often interrupt the mud in unit 2. Cross and parallel laminae are occasionally recognizable in the soft x-radiographs, especially in the lower part. Bidirectional ripple cross laminae can be also observed, but most of the sedimentary structures are disturbed because of strong bioturbation. The occurrence of mud and intercalated sand with these internal features is generally associated with tidally influenced sedimentation and is interpreted as representing tidal flat to muddy bay fills deposited during a period of sea level rise. The TOC/TN (5–7) and TOC/TS (2–3) ratios in the sediments of unit 2 show marine influences and support the interpretation that unit 2 was deposited in a tide-influenced environment.

MW-2 unit 3 (−9.34 to −7.99 m a.s.l., 1.35 m thick, 6220–5010 yr BP)

Unit 3 is composed of horizontally alternating fine-grained sand and mud layers (sand–mud couplets). The thickness of the sand layers ranges from 1 to 3 cm, decreasing upward. The base of the sand layers is sharp and erosional into the underlying clay. The sand layers display parallel to cross laminae. Some shell fragments and plant fragments are found in the sand layers, but evidence of bioturbation is only observed in the upper portions of unit 3. The lack of bioturbation indicates high-energy and high-sedimentation rates. These horizontal sand–mud couplets are interpreted as representing point bar deposits, a part of the fluvial system, and they may have been deposited within a channel belt (Hori et al., 2001; Lin et al., 2005).

MW-2 unit 4 (−7.99 to −6.37 m a.s.l., 1.62 m thick, 5010–4080 yr BP)

Unit 4 primarily consists of massive silty clay. Interbedded layers of medium to fine sands, typically several millimeters to centimeters thick, and oyster shell accumulations occur irregularly. Abundant plant fragments and organic debris are also characteristic of this unit. The Pacific oysters (Crassostrea gigas) are clear indicators of shallow marine tidal waters.

MW-2 unit 5 (−6.37 to −6.01 m a.s.l., 0.36 m thick, 4080–3870 yr BP)

Unit 5 is composed largely of massive greenish-gray (5GY 2/1) sandy mud. Shell fragments are not found in this unit, indicating a relatively deep-water coastal environment.

Core IL-3

The 18.30-m-long (−25.13 to −6.83 m a.s.l.) core IL-3 was taken in the present tidal flat near Mokpo city, near the confluence point of the Yeongsan River and the Youngam tributary. The core was stopped by weathered bedrock at −25.13 m a.s.l. Five sedimentary units were distinguished based on grain size, color, and sedimentary structures and textures (Figures 4 and 5).

Figure 4.

Simplified radiocarbon-dated sediment strata of core IL-3 with multiproxy data.

Figure 5.

Representative core IL-3 photographs: (a) −21.72 to −21.42 m, (b) −21.30 to −21.00 m, (c) −16.21 to −15.91 m, (d) −15.21 to −14.91 m, (e) −12.04 to −11.74 m; and X-radiographs: (f) −19.23 to −18.93 m, (g) −18.16 to −17.86 m, (h) −16.77 to −16.47 m, (i) −14.65 to −14.35 m.

IL-3 unit 1 (−25.13 to −21.61 m a.s.l., 3.52 m thick, ?−10,420 yr BP)

Unit 1 is composed of olive-gray, poorly sorted medium to coarse sand. The sands are often intercalated with black organic-rich layers. Faint planar laminae are the only sedimentary structures identified in this unit. The TOC values vary between 0.1 and 0.2 weight percent (wt%), and TN values vary between 0.01 and 0.02 wt% within this unit. TOC/TN ratios vary between 10.2 and 23.4, and TOC/TS ratios are generally about 5.8.

IL-3 unit 2 (−21.61 to −16.92 m a.s.l., 4.69 m thick, 10,420–10,310 yr BP)

Unit 2 is characterized by thinly interbedded or thickly interlaminated inclined gray mud and light gray, fine to medium sand (sand–mud couplets). An overall upward-fining and upward-thinning trend is recognized. The sand layers are 1–3 cm thick in the lower part and millimeters thick in the upper part. These sand layers are commonly ripple laminated or planar laminated, locally showing bidirectional cross laminae. Bivalve shells are intermittently distributed throughout unit 2. Small burrows <1 cm in diameter are found in some sand layers. The TOC content ranges from 0.2 to 0.5 wt%. TOC/TN and TOC/TS ratios range from 7.1 to 12.3 and 0.8 to 4.2, respectively. Inclined heterolithic strata have been described from modern intertidal point bars, modern tidally influenced rivers, and ancient channel-fill sequences (Thomas et al., 1987). They are generally interpreted as lateral accretion deposits on point bar surfaces within tidally influenced rivers.

IL-3 unit 3 (−16.92 to −15.58 m a.s.l., 1.34 m thick, 10,310–9720 yr BP)

Unit 3 is composed of gray clayey silt with thin shelly layers, with a gradational contact with the underlying layer. No internal sedimentary structures were observed with the naked eye. X-radiographs show parallel, wavy, and cross laminae in some intervals, but these are strongly bioturbated and largely destroyed. Bidirectional cross laminae also occur intermittently. Plant fragments are sporadically distributed throughout unit 3. TOC/TN ratios are generally between 6.1 and 9.7, and TOC/TS ratios range from 0.8 to 3.3.

IL-3 unit 4 (−15.58 to −14.99 m a.s.l., 0.59 m thick, 9720–9530 yr BP)

Shells of the Pacific oysters (Crassostrea gigas) are highly concentrated in unit 4, although most are fragmentary. TOC/TN ratios vary between 5.7 and 11.5 in unit 4, indicating a significant change in sediment source.

IL-3 unit 5 (−14.99 to −6.83 m a.s.l., 8.16 m thick, 9530–7330 yr BP)

Unit 5 is largely composed of massive greenish-gray mud and lacks shell fragments.

Core JD-1

The 11.88-m-long (−12.79 to −0.91 m a.s.l.) core JD-1 was taken in the present tidal flat near Naju city, near the confluence point of the Yeongsan River and the Sampo tributary. Six sedimentary units were distinguishable based on grain size, color, and sedimentary structures and textures (Figures 6 and 7).

Figure 6.

Simplified radiocarbon-dated sediment strata of core JD-1 with multiproxy data.

Figure 7.

Representative core JD-1 photographs: (a) −11.90 to −11.60 m, (b) −8.16 to −7.86 m, (c) −5.91 to −5.61 m, (d) −1.96 to −1.66 m, (e) −1.62 to −1.32 m; and X-radiographs: (f) −12.01 to −11.71 m, (g) −8.26 to −7.96 m, (h) −5.67 to −5.37 m, (i) −1.64 to −1.34 m.

JD-1 unit 1 (−12.79 to −12.05 m a.s.l., 0.74 m thick, ?−8520 yr BP)

Unit 1 is composed of olive-gray, medium to coarse sand, generally showing an upward-fining succession. Layers rich in plant debris are present in the upper part. Unit 1 is interpreted as a succession of river channel infill deposits. The lack of marine or brackish-water mollusk shells also suggests a lack of tidal influence in unit 1.

JD-1 unit 2 (−12.05 to −11.20 m a.s.l., 0.85 m thick, 8520–8390 yr BP)

Unit 2 is primarily composed of irregular or quasi-rhythmical inclined alternations of fine- to medium-grained sand and mud (sand–mud couplets). The sand layers in this succession range in thickness from 0.5 to 5.0 cm and the mud layers are 0.5–1.0 cm thick. The dominant sedimentary structures in unit 2 are planar cross laminae, dipping approximately 10–20°. X-radiographs show that some cross laminae are bidirectional. Fragments of plants and bivalve shells are sometimes found within the sand layers. TOC/TN and TOC/TS ratios range from 5.3 to 12.8 and from 1.2 to 3.2, respectively. The succession in the upper part of unit 2 is clearly upward-fining, with sand content decreasing and mud content increasing. Rhythmic sand–mud couplets are especially common in tide-influenced coastal environments (Allen and Posamentier, 1993). The sand–mud couplets in unit 2 are interpreted as tidal estuarine sand and mud sediments forming tidal sand bars or tidal estuarine point bars. An overall upward-fining lithological succession and the occurrence of bidirectional cross laminae indicate that the sediments were deposited as a result of lateral accretion in a tide-influenced meandering river (Miall, 1992).

JD-1 unit 3 (−11.20 to −6.25 m a.s.l., 4.95 m thick, 8390–6390 yr BP)

Unit 3 is characterized by greenish-gray to gray (10Y5/1) clayey silt. X-radiographs show that unit 3 is heavily bioturbated. However, parallel laminae, planar cross laminae, and bidirectional ripple cross laminae can be identified in the x-radiographs. Sand layers are not intercalated in unit 3. Small plant fragments are scattered intermittently throughout the sediment. TOC/TN and TOC/TS ratios are generally between 3.7 and 7.3 and between 0.6 and 2.2, respectively. Overall, the sedimentary structures imply transportation and deposition of mud by flood and ebb tidal currents. The ripple cross laminae are bidirectional, suggesting an oscillatory flow. Unit 3 is interpreted as having formed in an intertidal to subtidal tide-influenced environment.

JD-1 unit 4 (−6.25 to −4.08 m a.s.l., 2.17 m thick, 6390–4950 yr BP)

Unit 4 is composed mainly of finely laminated grayish mud (silty clay). The internal fine laminae are nearly horizontal, consisting of rhythmic alternations of millimeter-thick light and dark laminae that are distinguishable in the x-radiographs. Several erosional surfaces can be identified, and no bioactivity was observed. These features may reflect seasonal fluctuations in suspended load or variations in the amount and components of supplied material, as well as high-energy currents. The absence of bioturbation in unit 4 suggests a shallow intertidal setting with riverine influence and weak organismal activity.

JD-1 unit 5 (−4.08 to −1.66 m a.s.l., 2.42 m thick, 4950–4040 yr BP)

Unit 5 contains the highest concentration of the Pacific oyster (Crassostrea gigas) shells in this core. Sedimentary structures are rarely observed because of the shell fragments. Unit 5 also contains abundant plant fragments and shows slightly increased TOC values and TOC/TN ratios. The sediments of unit 5 were likely deposited in an intertidal to supratidal environment.

JD-1 unit 6 (−1.66 to −0.91 m a.s.l., 0.75 m thick, 4040–3890 yr BP)

Unit 6 is characterized by gray (10Y5/1) clayey silt with some bioturbation. Overall organic content including plant fragments decreases in this unit, as confirmed by decreased TOC and TN values. The absence of shell fragments indicates that unit 6 accumulated in an intertidal to subtidal zone.

Discussion

Early Holocene sea level rise and geomorphological development

All three cores (IL-3, MW-2, and JD-1) can be divided at the boundary between units 1 and 2 into a lower portion lacking a seawater influence and an upper portion with a seawater influence (Figure 8). The first evidence of seawater influence appears at 10,420 yr BP (−21.61 m) in core IL-3, 8630 yr BP (−14.39 m) in core MW-2, and 8520 yr BP (−12.05 m) in core JD-1. Previous studies have reported the position of sea level at around 10,400 yr BP, including −25 m in the Mekong River Delta, Vietnam (Ta et al., 2002); −21 m in the Mekong River lowland, Cambodia (Tamura et al., 2007); and −18 m in the Song Hong (Red River) delta plain, Vietnam (Funabiki et al., 2007). These data are almost coincident with the results of this study (i.e. −21.61 m in core IL-3). However, several deeper positions have also been found, for example, −40 m in the Changjiang (Yangtze) River delta (Hori et al., 2002); −38 m in Tahiti (Peltier and Fairbanks, 2006); −37 m in the Echigo plain, Japan (Tanabe et al., 2010); −36 m in Barbados (Bard et al., 2010); and −31 m in the Toyooka basin, Japan (Tanigawa et al., 2013). Sea level positions around 10,400 yr BP in East Asia including Vietnam, Cambodia, China, Japan, and Korea can be largely grouped into two categories (i.e. −21 to −25 m and −36 to −40 m), taking into account the possible error from age-dating and depth-measuring methods.

Figure 8.

Correlation diagram for the main features of the different profiles (core MW-2, IL-3, and JD-1).

Several intervals of rapid sea level rise resulting from outbursts of glacial meltwater into the ocean (Meltwater Pulses, or MWP) have been reported after the Last Glacial Maximum (LGM): MWP 1a at 14,200 yr BP, MWP 1b at 11,500 yr BP, and MWP 2 at 7600 yr BP (Fairbanks, 1989; Smith et al., 2011; Woodroffe and Horton, 2005). Sea level rises during such events would significantly alter sedimentation and erosion patterns as well as the topography of coastal areas. It would therefore be difficult to find reliable sea level indicators such as geomorphological features (paleo-shoreline notches and terraces) and fixed biological indicators (rock-clinging oyster beds and fossil tubeworm encrustations) (Yim and Huang, 2002). Furthermore, even in far-field locations, sea level positions could vary because of hydro-isostatic adjustment, local tectonic conditions, sediment flux and compaction, and coastal topography and basin physiography (Chen and Liu, 2000; Lambeck and Nakada, 1990; Zong, 2004). As so many factors affect sea level position, sea level at around 10,400 yr BP should be more scattered, not concentrated into two groups. To explain the approximately 15- to 20-m difference between the two groups, we have to thoroughly examine the drilling locations, age-dating samples, and sedimentation and erosion patterns.

The possibility that sedimentation was unstable at the time of the sea level rise cannot be ruled out in this study as well. The sedimentary environments could have been unstable at the non-marine and marine boundary between units 1 and 2 in all three cores. Erosion and re-deposition could occur frequently in coastal areas, meaning that the age dating for the lower part of unit 2 might be inaccurate. Furthermore, the age of the boundary of units 1 and 2 in the three cores was assumed by linear interpolation using the age-dating results from the lower part of unit 2. This is based on the assumption of a constant sedimentation rate from the boundary to the lower part of unit 2. Therefore, this method may not reflect the complex pattern or rate of sedimentation during the period of sea level rise.

Radiocarbon dates obtained from plant material are subject to uncertainties about the timing of plant death, deposition, and burial (Pekar et al., 2004). The source of the plant remains is often unknown, as a large amount of drifting plant fragments may have been present along the shore because of rapid marine transgression during the Early Holocene (Hormes et al., 2004; Josenhans et al., 1995). Radiocarbon dating using bulk carbon is also problematic because carbon in sediments comes from a variety of sources (Colman et al., 2002). Bjorck and Wohlfarth (2001) explained that bulk carbon samples tend to result in ages that are too old, primarily because of the hard water reservoir effect. However, although sediments in estuarine environments are frequently reworked on a large scale, they can be regarded as palimpsest sediments (Swift et al., 1971). The AMS ¹⁴C dating results in this study are in successive stratigraphic order, so the dates can generally be considered reliable (Tables 1 –3).

Table 1.

AMS ¹⁴C dates from core MW-2.

Core	Sample ID	Material	Elevation	¹⁴C ages (¹⁴C yr BP)	δ¹³C (per mil)	Calibrated ages (yr BP)
MW-2	SNU01-328	Shell	−6.25 m	3670 ± 60	−1.8	4010 ± 80
MW-2	SNU01-329	Shell	−8.80 m	4760 ± 60	−1.3	5470 ± 100
MW-2	SNU01-330	Plant fragment	−9.35 m	5400 ± 30	−25.4	6240 ± 30
MW-2	SNU01-334	Shell	−9.56 m	5820 ± 70	−2.4	6630 ± 90
MW-2	SNU01-335	Plant fragment	−12.77 m	7270 ± 150	−28.8	8120 ± 150
MW-2	SNU01-338	Plant fragment	−14.12 m	7770 ± 40	−24.8	8540 ± 50

Table 2.

AMS ¹⁴C dates from core IL-3.

Core	Sample ID	Material	Elevation	¹⁴C ages (¹⁴C yr BP)	δ¹³C (per mil)	Calibrated ages (yr BP)
IL-3	ICa080011	Bulk carbon	−7.90 m	6730 ± 60	−21.2	7590 ± 50
IL-3	GX-29586	Bulk carbon	−11.72 m	7720 ± 70	−21.9	8510 ± 60
IL-3	GX-29587	Bulk carbon	−14.45 m	8340 ± 70	−22.0	9350 ± 90
IL-3	GX-29588	Bulk carbon	−17.38 m	9130 ± 60	−22.7	10,320 ± 70
IL-3	GX-29589	Bulk carbon	−21.10 m	9220 ± 100	−23.3	10,410 ± 120

Table 3.

AMS ¹⁴C dates from core JD-1.

Core	Sample ID	Material	Elevation	¹⁴C ages (¹⁴C yr BP)	δ¹³C (per mil)	Calibrated ages (yr BP)
JD-1	SNU-03-385	Plant fragment	−1.940 m	3720 ± 80	−29.9	4090 ± 120
JD-1	SNU-03-386	Plant fragment	−3.710 m	3970 ± 80	−28.9	4430 ± 120
JD-1	SNU-03-387	Plant fragment	−4.730 m	4560 ± 60	−25.8	5210 ± 110
JD-1	SNU-03-388	Plant fragment	−8.480 m	7330 ± 40	−20.0	8120 ± 60
JD-1	SNU-03-389	Plant fragment	−9.980 m	7570 ± 40	−30.0	8390 ± 20
JD-1	SNU-03-390	Plant fragment	−12.250 m	7740 ± 80	−28.6	8530 ± 70

The elevations of sediment in sediment cores represent depositional surfaces, so actual sea level could be higher or lower than the sediment elevation. Particularly in macro-tidal regions having tides greater than 4 m like the study area, seawater can intrude upstream along the river channel bottom for more than several kilometers because of tidal forcing in the form of a salt wedge, resulting in a marine signal in an inland area located higher than the mean sea level position. Units 1 in cores IL-3 and JD-1 consist of river channel terrestrial sediment, and units 2 in all three cores are interpreted to be the lateral accretion of point bars under a tidal influence. Therefore, the sediment of units 2 were influenced by seawater, and mean sea level at that time may have been lower than the depositional surfaces. Meanwhile, MW-2 unit 1 is a paleosol that was subaerially exposed for a period of time prior to marine inundation, and the sediment of MW-2 unit 2 was deposited in seawater. It would therefore be reasonable to estimate that sea level was higher than the depositional surface of MW-2 unit 2. The topography around the study area during the early Holocene can be inferred from these observations. The channels of the Yeongsan River and its tributaries ran through the locations of cores IL-3 and JD-1. In contrast, the location of MW-2 is now the middle of the main channel, but it was not the main river channel at that time.

Precipitation during the HCO

An archive sensitive to precipitation changes yet insensitive to temperature changes is required for the study of precipitation during the HCO. Most importantly, the continuity, immediacy, and preservability of samples are key requirements for assessing paleoenvironmental changes. Previously reported archives and proxies include lake-level fluctuations, lacustrine and swamp deposits, pollen-spore sequences, and the magnetic susceptibility of loess–paleosol sequences (An et al., 2000, 2006; Chen et al., 2001; Feng et al., 2004; Xiao et al., 2004).

As for the speleothems, the path of rainwater through several surficial soil and rock layers above the cave should be taken into consideration for the growth of stalagmites and stalactites. Physicochemical conditions in the cave such as temperature and partial pressure of CO₂ are also important. However, the growth of speleothems and their relationship to surrounding environments are not yet well defined (Frisia et al., 2000). Pollen studies are useful for examining the impact of rapid climate change on terrestrial ecosystems because the response of vegetation to climate fluctuation is pronounced (Tinner and Lotter, 2001). However, several researchers have pointed out that pollen records show an apparent gradual response or a 1000-year time lag because mature trees are less sensitive to climate change (Williams et al., 2002; Zhao et al., 2009). In comparison, river sediments respond rapidly to intensified rainfall and are extensively altered by precipitation (Dean and Schmidt, 2013; Nahm et al., 2011). Although unconsolidated riverine sediments are vulnerable to erosion, many researchers have already found records of flooding in riverine sediment sections. Riverine sediments are therefore suitable for assessing past changes in rainfall intensity.

The deposition of MW-2 unit 3 (sand–mud couplet) and JD-1 unit 4 (laminated mud) occurred roughly during 6300–5000 yr BP. The influence of rivers is higher in these units than in the units above or below. Sand–mud couplets in MW-2 unit 3 show horizontal layering, a lack of bidirectional cross laminae, and a slightly higher TOC/TN ratio (7.90–13.48). This indicates point bar deposits within a channel belt. IL-3 unit 2 (10,420–10,310 yr BP) and JD-1 unit 2 (8520–8390 yr BP) show similar sand–mud couplets, but the layers dip 10°–20°, bidirectional cross laminae are apparent, and TOC/TN ratios are low (4.6–9.3). Laminated mud in JD-1 unit 4 was basically deposited under a tidal setting, but the light and dark laminae express seasonal changes in the supply of terrestrial material. Possible explanations for the riverine influence in MW-2 unit 3 and JD-1 unit 4 include sea level fall, river channel changes, and intensified river activity. According to previous work, sea level was higher than the present position at around 6300–5000 yr BP (Nahm et al., 2013). Many researchers have estimated that the sea level fall occurred only after about 5000 yr BP, meaning that a fall in sea level may not be the cause of the riverine influence in these units. In core JD-1, unit 2 shows evidence of a riverine influence, unit 3 represents a tidal flat environment, and unit 4 is again a river setting. This means that the meandering river channel moved back to the core site when units 2 and 4 were deposited. However, there is no evidence for paleo-channels from the complementary cores around the location of core JD-1. The most plausible explanation for the riverine influence in MW-2 unit 3 and JD-1 unit 4 is increased amount or strength of rainfall at around 6300–5000 yr BP, with subsequent active riverine transportation of sediments. Although sea level was estimated to be high enough to have a strong influence on the study area during the mid-Holocene, the sedimentary features suggest that riverine activity was the dominant factor controlling the sedimentation patterns. This finding shows that even during the HCO in Korea, roughly 7000–5000 yr BP, the wetter condition has occurred within the interval of 6300–5000 yr BP.

Mid-Holocene sea level fluctuations

MW-2 unit 4 and JD-1 unit 5, which were deposited around 5000–4000 yr BP, contain a characteristic concentration of oyster shells. These Pacific oysters (Crassostrea gigas) are well known along the west and south coasts of the Korean Peninsula and are characteristic of the intertidal and supratidal zone in Korea (Frey et al., 1989). They usually attach themselves to rocks in tidal flat environments and live mainly in shallow water <5 m in depth. Kim et al. (1999) reported oyster shell layers in Gomso Bay, located 100 km north of this study area, and interpreted these layers as storm deposits. These shell layers are several centimeters thick, and show sharp erosional bases that cut the tops of the underlying layers. The shell layers in MW-2 unit 4 and JD-1 unit 5, in elevations of −7.99 to −6.37 m and −4.08 to −1.66 m, respectively, are too thick (1.6–2.4 m) to have been deposited by a storm event. If these elevations represented the living environment of the oysters, then the sea level at that time would have been around 0 to −1 m. This means that the sea level decreased after the HCO, and the locations of cores MW-2 and JD-1 became a suitable habitat for the oysters. A sea level fall around 5000 yr BP is known from the west (Hwang et al., 1997; Jo, 1987), east (Yum et al., 2003), and south (Yang, 2011) coasts of the Korean Peninsula. A sea level fall of similar timing has also been reported in the Yangtze River delta, China, based on topographic development modeling (Xin and Xie, 2006). The post-HCO sea level fall that occurred at about 5000 yr BP is termed a ‘Neoglacial’ (Miller et al., 2005; Porter, 2000). Milne et al. (2005) mentioned glacial isostatic adjustment processes including local hydro-isostatic loading (termed continental levering) and global sea level fall because of both hydro- and glacio-isostatic loading of the Earth’s surface (termed equatorial ocean syphoning) as possible causes for the Neoglacial sea level fall.

MW-2 unit 5 and JD-1 unit 6 were deposited around 4000 yr BP and are composed of marine-gray or greenish-gray mud, indicating repeated sea level rise. Jo (1980), Hwang (1998), and Hwang et al. (2013) previously reported a sea level rise starting about 4000 yr BP around the Korean Peninsula, although the estimate of the elevation of sea level differs with location. The timing of the sea level rise coincides with the global trends. For example, the second phase of sea level rise during the Holocene in the Indian Ocean started around 4300 yr BP (Banerjee, 2000).

Conclusion

This study presents detailed regional records of Holocene paleoenvironments in three sediment cores from the Yeongsan River Estuary. Our sedimentological and geochemical study showed the following: (1) A marine influence can be observed at 10,400 yr BP (−21 m) and 8600 yr BP (−14 to −12 m) around the Yeongsan River Estuary. If we take the macro-tidal setting of the Yeongsan River Estuary into consideration, mean sea level may have been higher or lower than the depositional surfaces. More age data from different places and depths are required to develop a more complete history of relative sea level. (2) Precipitation at around 6300–5000 yr BP is estimated to have been higher than in earlier and later periods. The core sites received increased terrigenous sediment input during this period because of intensified rainfall and river activity. (3) Sea level decreased to 0 to −1 m around 5000 yr BP and increased again starting around 4000 yr BP. These fluctuations are coincident with global trends. The time-series results presented in this paper show the potential for developing geomorphological and climate histories for this region, and they provide a framework for further research.

Footnotes

Funding

This research was supported by the Basic Research Project of Korea Institute of Geoscience and Mineral Resources funded by the Korea government Ministry of Science, ICT and Future Planning, and the Radioactive Waste Management of the Korea Institute of Energy Technology Evaluation and Planning (KETEP) grant funded by the Korea government Ministry of Knowledge Economy (2012171020001B).

References

Allen

Posamentier

(1993) Sequence stratigraphy and facies model of an incised valley fill: The Gironde Estuary, France. Journal of Sedimentary Petrology 63: 378–391.

Feng

Barton

(2006) Dry or humid? Mid-Holocene humidity changes in arid and semi-arid China. Quaternary Science Reviews 25: 351–361.

(2000) The history and variability of the East Asian paleomonsoon climate. Quaternary Science Reviews 19: 171–187.

Porter

Kutzbach

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

Banerjee

(2000) Holocene and Late Pleistocene relative sea level fluctuations along the east coast of India. Marine Geology 167: 243–260.

Bard

Hamelin

Delanghe-Sabatier

(2010) Deglacial meltwater pulse 1B and Younger Dryas sea levels revised with boreholes at Tahiti. Science 327: 1235–1237.

Bjorck

Wohlfarth

(2001) 14C Chronostratigraphic techniques in paleolimnology. In: Last

Smol

(eds) Tracking Environmental Change Using Lake Sediments. Volume 1: Basin Analysis, Coring, and Chronological Techniques. Dordrecht: Kluwer Academic Publishers, pp. 1–41.

Chen

Liu

(2000) Holocene uplift and subsidence along an active tectonic margin southwestern Taiwan. Quaternary Science Reviews 19: 923–930.

Chen

Shen

. (2001) Yangtze River of China: Historical analysis of discharge variability and sediment flux. Geomorphology 41: 77–91.

10.

Chung

(2011) Holocene vegetation dynamics and its climatic implications inferred from pollen record in Boseong area, South Korea. Geosciences Journal 15: 257–264.

11.

Colman

Baucom

Bratton

. (2002) Radiocarbon dating, chronologic framework, and changes in accumulation rates of Holocene estuarine sediments from Chesapeake Bay. Quaternary Research 57: 58–70.

12.

Davis

BAS

Brewer

Stevenson

. (2003) The temperature of Europe during the Holocene reconstructed from pollen data. Quaternary Science Review 22: 1701–1716.

13.

Dean

Schmidt

(2013) The geomorphic effectiveness of a large flood on the Rio Grande in the Big Bend region: Insights on geomorphic controls and post-flood geomorphic response. Geomorphology 201: 183–198.

14.

Eun

GYN

Chung

Kim

(1998) A comparative study of depositional environments through coring analysis in Yeongsan and Nakdong estuaries. Journal of Korean Environmental Sciences Society 7: 889–893. (In Korean with English abstract).

15.

Evstigneeva

Naryshkina

(2010) The Holocene climatic optimum at the Southern Coast of the Sea of Japan. Paleontological Journal 44: 1262–1269.

16.

Fairbanks

(1989) A 17,000-year glacio-eustatic sea level record: Influence of glacial melting rates on the Younger Dryas event and deep-ocean circulation. Nature 342: 637–642.

17.

Feng

Tang

. (2004) Stratigraphic evidence of megahumid mid-Holocene climate in the western part of the Chinese Loess Plateau. Global and Planetary Change 43: 145–155.

18.

Frey

Howard

Han

. (1989) Sediments and sedimentary sequences on a modern macrotidal flat, Inchon, Korea. Journal of Sedimentary Petrology 59: 28–44.

19.

Frisia

Borsato

Fairchild

. (2000) Calcite fabrics, growth mechanisms, and environments of formation in speleothems from the Italian Alps and southwestern Ireland. Journal of Sedimentary Research 70: 1183–1196.

20.

Funabiki

Haruyama

Quy

. (2007) Holocene delta plain development in the Song Hong (Red River) delta, Vietnam. Journal of Asian Earth Sciences 30: 518–529.

21.

Ganopolski

Kubatzki

Claussen

. (1998) The influence of vegetation-atmosphere-ocean interaction on climate during the mid-Holocene. Science 280: 1916–1919.

22.

Hewitt

Mitchell

JFB

(1998) A fully coupled GCM simulation of the climate of the mid-Holocene. Geophysical Research Letters 25: 361–364.

23.

Hori

Saito

Zhao

. (2001) Sedimentary facies of the tide-dominated paleo-Changjiang (Yangtze) estuary during the last transgression. Marine Geology 177: 331–351.

24.

Hori

Saito

Zhao

. (2002) Evolution of the coastal depositional systems of the Changjiang (Yangtze) river in response to late Pleistocene-Holocene sea-level changes. Journal of Sedimentary Research 72: 884–897.

25.

Hormes

Karlen

Possnert

(2004) Radiocarbon dating of palaeosol components in moraines in Lapland, northern Sweden. Quaternary Science Reviews 23: 2031–2043.

26.

Hwang

(1998) The Holocene depositional environment and sea-level change at Ilsan area. Journal of the Korean Geographical Society 33: 143–163. (In Korean with English abstract).

27.

Hwang

Kim

Yoon

(2013) Sea Level Change during the Middle Holocene at Bibong-ri, Changnyeong-gun, Gyeongsangnam-do, South Korea. Journal of the Korean Geographical Society 48: 837–855. (In Korean with English abstract).

28.

Hwang

Yoon

(1997) The change of the depositional environment on the Dodaecheon River basin during the middle Holocene. Journal of the Korean Geographical Society 32: 403–420. (In Korean with English abstract).

29.

Woo

Lim

. (2011) Holocene and Eemian climatic optima in the Korean Peninsula based on textural and carbon isotopic records from the stalagmite of the Daeya Cave, South Korea. Quaternary Science Reviews 30: 1218–1231.

30.

(1980) Holocene sea-level changes on the east coast of Korea Peninsula. Geographical Review of Japan 53: 317–328. (In Japanese with English abstract).

31.

(1987) Alluvial Plain in Korea. Seoul: Kyohakyongusa, 220 pp. (In Korean).

32.

Josenhans

Fedje

Conway

. (1995) Post glacial sea levels on the Western Canadian continental shelf: Evidence for rapid change, extensive subaerial exposure, and early human habitation. Marine Geology 125: 73–94.

33.

Kim

Lee

Chun

. (1999) Holocene transgressive stratigraphy of a macrotidal flat in the southeastern Yellow Sea: Gomso Bay, Korea. Journal of Sedimentary Research 69: 328–337.

34.

Kong

(1994) The vegetational history of Korea during the Holocene period. Korean Journal of Quaternary Research 8: 10–26.

35.

Lambeck

Nakada

(1990) Late Pleistocene and Holocene sea-level change along the Australian coast. Palaeogeography, Palaeoclimatology, Palaeoecology 89: 143–176.

36.

Lee

(1987) Stratigraphical research of the Quaternary deposits in the Korean Peninsula. Korean Journal of Quaternary Research 1: 3–20.

37.

Lin

Zhuo

Gao

(2005) Sedimentary facies and evolution in the Qiantang River incised valley, eastern China. Marine Geology 219: 235–259.

38.

Miall

(1992) Alluvial deposits. In: Walker

James

(eds) Facies Models: Response to Sea Level Change. Waterloo: Geological Association of Canada, pp. 119–139.

39.

Miller

Wolfe

Briner

. (2005) Holocene glaciation and climate evolution of Baffin Island, Arctic Canada. Quaternary Science Reviews 24: 1703–1721.

40.

Milne

Long

Bassett

(2005) Modelling Holocene relative sea-level observations from the Caribbean and South America. Quaternary Science Reviews 24: 1183–1202.

41.

Nahm

Kim

Bong

. (2008) Late Quaternary stratigraphy of the Yeongsan Estuary, Southwestern Korea. Quaternary International 176–177: 13–24.

42.

Nahm

Kim

. (2013) The Holocene climatic optimum in Korea: Evidence from wetland records. Palaeogeography, Palaeoclimatology, Palaeoecology 376: 163–171.

43.

Nahm

Kim

Lim

. (2011) Responses of the upriver valley sediment to Holocene environmental changes in the Paju area of Korea. Geomorphology 133: 80–89.

44.

Pearson

(1977) Paleoenvironment and human settlement in Japan and Korea. Science 197: 1239–1246.

45.

Pekar

McHugh

CMG

Christie-Blick

. (2004) Estuarine processes and their stratigraphic record: Paleosalinity and sedimentation changes in the Hudson Estuary (North America). Marine Geology 209: 113–129.

46.

Peltier

Fairbanks

(2006) Global glacial ice volume and Last Glacial Maximum duration from as extended Barbados sea level record. Quaternary Science Reviews 25: 3322–3337.

47.

Pielou

(1991) After the Ice Age. Chicago, IL: University of Chicago Press, 366 pp.

48.

Porter

(2000) Onset of neoglaciation in the Southern Hemisphere. Journal of Quaternary Science 15: 395–408.

49.

Ryu

Lee

Chang

(2004) Seasonal cycle of sedimentary process on mesotidal flats in the semienclosed Muan Bay, southern west coast of Korea: Culminating summertime erosion. Continental Shelf Research 24: 137–147.

50.

Smith

Harrison

Firth

. (2011) The early Holocene sea level rise. Quaternary Science Reviews 30: 1846–1860.

51.

Swift

DJP

Stanley

Curray

(1971) Relict sediments on continental shelves: A reconsideration. Journal of Geology 79: 322–346.

52.

TKO

Nguyen

Tateishi

. (2002) Sediment facies and Late Holocene progradation of the Mekong River Delta in Bentre Province, southern Vietnam: An example of evolution from a tide-dominated to a tide- and wave-dominated delta. Sedimentary Geology 152: 313–325.

53.

Tamura

Saito

Sieng

. (2007) Depositional facies and radiocarbon ages of a drill core from the Mekong River lowland near Phnom Penh, Cambodia: Evidence for tidal sedimentation at the time of Holocene maximum flooding. Journal of Asian Earth Sciences 29: 585–592.

54.

Tanabe

Nakanishi

Yasui

(2010) Relative sea-level change in and around the Younger Dryas inferred from late Quaternary incised-valley fills along the Japan Sea. Quaternary Science Reviews 29: 3956–3971.

55.

Tanigawa

Hyodo

Sata

(2013) Holocene relative sea-level change and rate of sea-level rise from coastal deposits in the Toyooka Basin, western Japan. Holocene 23: 1039–1051.

56.

Thomas

Smith

Wood

. (1987) Inclined heterolithic stratification: Terminology, deposition, interpretation and significance. Sedimentary Geology 53: 123–179.

57.

Tinner

Lotter

(2001) Central European vegetation response to abrupt climate change at 8.2 ka. Geology 29: 551–554.

58.

Wanner

Bronnimann

(2012) Is there a global Holocene climate mode? PAGES news 20: 44–45.

59.

Williams

Post

Cwynar

. (2002) Rapid and widespread vegetation responses to past climate change in the North Atlantic region. Geology 30: 971–974.

60.

Winkler

Wang

(1993) The late-Quaternary vegetation and climate of China. In: Wright

Jr Kutzbach

Webb

III . (eds) Global Climates since the Last Glacial Maximum. Minneapolis, MA: University of Minnesota Press, pp. 221–261.

61.

Woodroffe

Horton

(2005) Holocene sea-level changes in the Indo-Pacific. Journal of Asian Earth Sciences 25: 29–43.

62.

Xiao

Zhai

. (2008) Hydrology of Dali Lake in central-eastern Inner Mongolia and Holocene East Asian monsoon variability. Journal of Paleolimnology 40: 519–528.

63.

Xiao

Nakamura

. (2004) Holocene vegetation variation in the Daihai Lake region of North-Central China: A direct indication of the Asian monsoon climatic history. Quaternary Science Reviews 23: 1669–1679.

64.

Xin

Xie

(2006) Construction of the simulating model for geomorphic evolution on the Yangtze delta, China. Acta Geographica Sinica 61: 549–560. (In Chinese with English abstract).

65.

Yang

(2011) Holocene sea level reflected from marine terrace in Geoje Island and its influences on coastal morphogenesis. Journal of the Korean Geomorphological Association 18: 101–112. (In Korean with English abstract).

66.

Kim

Yang

. (2008) Mid- and Late-Holocene palynofloral and environmental change of Korean central region. Quaternary International 176–177: 112–120.

67.

Yim

WWS

Huang

(2002) Middle Holocene higher sea-level indicators from the south China coast. Marine Geology 182: 225–230.

68.

Yoon

(1997) The Holocene environmental changes and reconstruction of the geography at Ilsan area with special reference to pollen analysis. Journal of the Korean Geographical Society 32: 15–30. (In Korean with English abstract).

69.

Yum

Takemura

Tokuoka

. (2003) Holocene environmental changes of the Hwajinpo Lagoon on the eastern coast of Korea. Journal of Paleolimnology 29: 155–166.

70.

Zhao

Chen

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

71.

Zong

(2004) Mid-Holocene sea-level highstand along the Southeast Coast of China. Quaternary International 117: 55–67.