Khim.indd 341Khim & Yoon 2003: Polar Research 22(2), 341–353 During the last decade, marine geological and geophysical researchers have carried out exten- sive bathymetric surveys around the continen- tal shelves of the Antarctic Peninsula which con- tain a portion of the marine-based West Antarctic Ice Sheet (Domack & Ishman 1993; Pudsey et al. 1994; Pudsey & Camerlenghi 1998; Shipp et al. 1999; Canals et al. 2000). The incised glacial troughs observed on the continental shelves of the Antarctic Pen insula preserve thick sediment fi lls derived from grounded ice shelves which built up during the last glacial period. Sugden & Clapperton (1977) have suggested that during the Last Glacial Maximum many ice caps formed on the South Shetland Islands, which were separat- ed from the Antarctic Peninsula Ice Sheet by the deep Bransfi eld Strait. The timing of the onset of deglaciation in the Postglacial marine environmental changes in Maxwell Bay, King George Island, West Antarctica Boo-Keun Khim & Ho Il Yoon Sediment textural properties and total organic carbon (TOC) contents of three sediment cores from Maxwell Bay, King George Island, West Antarctica, record changes in Holocene glaciomarine sedimentary envi- ronments. The lower sedimentary unit is mostly composed of TOC-poor diamictons, indicating advanced coastal glacier margins and rapid ice- berg discharge in proximal glaciomarine settings with limited productiv- ity and meltwater supply. Fine-grained, TOC-rich sediments in the upper lithologic unit suggest more open water and warm conditions, leading to enhanced biological productivity due to increased nutrient-rich meltwater supply into the bay. The relationship between TOC and total sulfur (TS) indicates that the additional sulfur within the sediment has not originated from in situ pyrite formation under the reducing condition, but rather may be attributed to the detrital supply of sand-sized pyrite from the hydro- thermal-origin, quartz–pyrite rocks widely distributed in King George Island. The evolution of bottom-water hydrography after deglaciation was recorded in the benthic foraminiferal stable-isotopic composition, corroborated by the TOC and lithologic changes. The δ18O values indi- cate that bottom-water in Maxwell Bay was probably mixed gradually with intruding 18O-rich seawater from Bransfi eld Strait. In addition, the δ13C values refl ect a spatial variability in the carbon isotope distribution in Maxwell Bay, depending on marine productivity as well as terrestrial carbon fl uxes by meltwater discharge. The distinct lithologic transition, dated to approximately 8000 yr BP (uncorrected) and characterized by textural and geochemical contrasts, highlights the postglacial environ- mental change by a major coastal glacier retreat in Maxwell Bay. B.-K. Khim, Dept. of Marine Science and Marine Research Institute, Pusan National University, Pusan 609-730, South Korea, bkkhim@ pusan.ac.kr; H. I. Yoon, Korea Polar Research Institute, Korea Ocean Research and Development Insti tute, Box 29, Ansan 425-600, South Korea. 342 Postglacial marine environmental changes in Maxwell Bay, King George Island region of the Antarctic Peninsula, particularly the South Shetland Islands, is still subject to debate. Based on the sediment accumulation in the sev- eral freshwater lakes of the ice-free basin, differ- ent opinions on the Holocene glacial and palaeo- climate history of the South Shetland Islands have been offered. Some investigators argue that the present-day warm environment dates back to about 5000 - 6000 yr BP (Mäusbacher et al. 1989; Björck et al. 1991; Björck et al. 1993). Martinez- Macchiavello et al. (1996) corroborated the com- plete deglac iation on King George Island by ca. 6000 yr BP. In contrast, it has been reported that deglaciation generally commenced in the South Shetland Islands before at least 10 000 yr BP (John 1972; Sugden & Clapperton 1986). Clap- perton & Sugden (1982) proposed that the exten- sive glaciers in the South Shetland Islands were developed at about 9500 yr BP. In addition, Mäus- bacher et al. (1989) reported that the last degla- ciation occurred between 9000 and 5000 yr BP in King George Island, based on the dates con- strained with the lake sediments. During the glacial period, the ice sheet expand- ed to the continental shelves, leading to the for- mation of consecutive prograding shelf edge clinoforms indicatative of subglacial and pro- glacial sediment gravity fl ow deposition (Larter & Vanneste 1995). In contrast, during deglaci- ation, the ice sheet rapidly retreated from the continental shelves, as substantiated by subgla- cial geomorphic features. Postglacial warming caused the collapse of grounded ice followed by increased meltwater discharge from the neigh- bouring coastal margins. This meltwater deliv- ered large amounts of fi ne-grained particles to the continental shelves via shallow marine bays and fjords (Pudsey 2000; Yoon et al. 2000; Taylor et al. 2001). Thus, the shallow marine bays and fjords record such palaeoclimatic changes. How- ever, information on the deglacial palaeoceano- graphic features of these fjords in King George Island is still limited. Fjord sediments typically contain the poten- tial records of past biotic, climatic and glacial fl uctuations (Syvitski 1989). The shallow areas of the Antarctic Peninsula region, especially in fjords, bays and coves, are the appropriate place to investigate deglacial marine environmental change. Advances and retreats of grounded ice in such environments are key factors determin- ing depositional processes and sediment distribu- tion (Yoon et al. 1997; Kirby et al. 1998; Pud sey 2000). Recently, Li et al. (2000) investigated sur- face textures of foraminiferal tests in Maxwell Bay, King George Island, located in the South Fig. 1. Core sites and bathymetry of Maxwell Bay, King George Island, South Shetland Islands. Arrows represent the current direction of glacier movement. Abbreviations: CB–Collins Bay, MC–Marian Cove, PC–Potter Cove. The bathymetry is in metres. 343Khim & Yoon 2003: Polar Research 22(2), 341–353 Shetland Islands (Fig. 1). They have reported that the sea level rise and concomitant infl ux of open ocean waters during deglaciation led to a strong dissolution of foraminiferal tests. The retreat of coastal glaciers increased meltwater fl ux into the bay, resulting in higher biological productivity (Domack & Ishman 1993; Domack & McClennen 1996; Shevenell et al. 1996; Kirby et al. 1998). Thus, higher rates of organic matter accumula- tion may have played an additional role in the car- bonate dissolution due to porewater acidifi cation by organic matter oxidation. In this study, we widen the palaeoenvironmen- tal study of postglacial sediments by using new data concerning grain size and the geochemis- try of postglacial sediments, as well as stable- isotope composition of benthic foraminifera, and by establishing the chronology with accelerated mass spectrometry radiocarbon dates obtained from three sediment cores in Maxwell Bay (Fig. 1). These results contribute to the reconstruction of the deglacial and Holocene environments of the coastal area of King George Island. The study area King George Island’s Maxwell Bay is 14 km long and 6 - 14 km wide (Fig. 1). It is one of the deep, U-shaped fjords found along the southern margin of the South Shetland Islands. It is sur- rounded by ice cliffs creeping from the low pro- fi le ice cap of Fildes Peninsula and Nelson Island. Maxwell Bay is separated from Bransfi eld Strait by a deep (> 430 m water depth) submarine sill. Water depth increases gradually from the coast- line to the 200 m isobath, and then steeply to the 550 m isobath. At 65° S, the study area experiences a slightly more moderate climate (cold temperate to sub- polar) compared to the main Antarctic Peninsu- la region. The surface waters of Maxwell Bay freeze regularly in winter, from late July to mid- September. A surface layer of warmer (1.04 to 0 °C), less saline (33.85 to 34.0 psu) and lower δ18O (− 0.44 to − 0.30 ‰) water overlies a colder (0 to − 0.34 °C), more saline (34.0 to 34.53 psu) and higher δ18O (− 0.30 to − 0.17 ‰) subsurface water mass (Khim et al. 1997). The temperature and salinity differences between the surface and subsurface waters are approximately 1 °C and 0.6 psu, respectively. The δ18O values are below – 0.30 ‰ in the surface water and above –0.20 ‰ in the subsurface water. Because Maxwell Bay is connected directly to Bransfi eld Strait over a relatively deep submarine sill, Bransfi eld Strait water (open ocean water) is well circulated with the subsurface water in the bay (Hong et al. 1991; Khim et al. 1997). The surface water is governed by the freshwater input from glacial melting and surface runoff. Several tributaries were developed in Maxwell Bay, including Marian Cove and Potter Cove in the north-east and Collins Bay in the north (Fig. 1). From late October, sea ice begins rapidly to break up and numerous icebergs are scattered in Maxwell Bay. Active iceberg calving from the tidewater glacier terminus was observed in Col- lins Bay during the austral summer (Yoon et al. 1998). Small valley glaciers draining into Marian Cove also deliver icebergs and large volumes of turbid meltwater during the summer months (Yoon et al. 1998). Potter Cove, a tributary inlet near the entrance of Maxwell Bay, is character- ized by large meltwater discharge, but no calv- ing of icebergs has been observed (Klöser et al. 1994). Most ice-rafted debris therefore origi- nates from the calving of icebergs off the edges of glaciers, whereas fi ne-grained particles are dis- charged by meltwater during the summer season. The bay becomes completely ice-free during the summer (November–February), leading to increased primary biological production. Materials and methods In 1997, three gravity cores were collected from Collins Bay (A10-02), Marian Cove (A10-01) and Potter Cove (A10-08) in Maxwell Bay aboard the RV Yuzhmorgeologiya during the Tenth Korea Antarctic Research Expedition (Fig. 1, Table 1). Each core was cut in the laboratory and then subsampled at 5-cm intervals for analyses of grain size, total organic carbon (TOC) and total sulfur (TS). Grains larger than 63 µm (gravel and sand) were separated by wet sieving and classi- Table 1. Location, water depth, and core length of sediment gravity cores that were obtained in the Maxwell Bay. Core Latitude Longitude Water depth Core length A10-01 62° 13.0' W 58° 47.5' S 85 m 235 cm A10-02 62° 11.3' W 58° 49.7' S 105 m 270 cm A10-08 62° 13.7' W 62° 39.7' S 45 m 105 cm 344 Postglacial marine environmental changes in Maxwell Bay, King George Island fi ed by dry sieving. Grains smaller than 63 µm (silt and clay) were analysed using a Micromet- rics Sedigraph 5100D. After removing CaCO3 by 10 % HCl from the bulk sediment powders, TOC and TS were measured using a Carlo-Erba NA- 1500 Elemental Analyzer at KORDI. Oxygen and carbon stable isotopic values were determined on the tests of benthic foraminifera (Globocassidulina biora), ranging in size from 150 to 250 µm. Because some foraminiferal tests show the dissolution texture (Li et al. 2000), spec- imens for analyses were carefully selected. After treatment at 350 °C for 1 hour under vacuum to remove organic matter, samples were then react- ed with H3PO4 at 90 °C in an on-line automated carbonate CO2 preparation device connected to a VG PRISM II mass spectrometer at the Universi- ty of California, Santa Barbara. The instrumental precisions of isotopic measurement are ± 0.11 ‰ for δ18O and ± 0.09 ‰ for δ13C based on the inter- nal standards. All values are expressed in the δ notation in accordance with the V-PDB standard (Craig 1957; Coplen 1996). Accelerator mass spectrometry (AMS) radi- ocarbon (14C) ages were measured at the Insti- tute of Geological and Nuclear Sciences, New Zealand, using the acid-insoluble organic matter fraction of bulk sediments. Results Chronology of core sediments Because we did not collect box cores for pre- cise age determination of surfi cial sediments, the core-top ages of each gravity core are estimated by the extrapolation of the regressions based on the measured 14C ages (Table 2). The extrapolat- ed ages of core-tops are approximately 5061 yr BP for core A10-01, 3555 yr BP for core A10-02 and 5128 yr BP for core A10-08 (Fig. 2). These extrapolated surface ages are signifi cantly larger than the typical reservoir correction for Antarc- tic marine environments, although the ocean res- ervoir correction may have changed during the glacial–interglacial cycles. In general, the ocean reservoir effect for the Antarctic marine car- bonates accounts for about 1200 to 1300 years (Gordon & Harkness 1992; Berkman & Forman 1996). However, it has been reported that radio- carbon ages obtained from acid-insoluble organ- ic fraction of surfi cial sediments are substan- tially older and often range between 2000 and 5000 years (Domack 1992; Andrews et al. 1999; Pudsey & Evans 2001). Such large discrepancies between the apparent reservoir correction and the measured 14C ages from sediment organics may be attributed to the old carbon contamina- tion caused either by the liberation of CO2 stored in the ice sheets (Domack et al. 1989), along with the uptake of “old” dissolved inorganic carbon by diatoms (Gibson et al. 1999), or by the assimi- lation of dissolved inorganic carbon by diatoms during CO2 limiting conditions (Tortell et al. 1997). Otherwise, the core-top sediments seem to be lost during the coring process. Therefore, in this study, we give the uncorrected measured radiocarbon ages in describing the downcore var- Fig. 2. Relationship between measured AMS 14C ages (bulk sediments) and core depths (see Table 2). Estimated linear sedimentation rates are shown for each core. The dotted line represents the sedimentation rate of core A10-01 averaged from two segments. 9.0 cm/Ky 22.7 cm/Ky 27.8 cm /Ky Table 2. AMS 14C radiocarbon ages measured on bulk sedi- ment organic carbon. Core Depth (cm) Measured 14C age (yr BP) δ13C (‰) Lab code A10-01 5 5313 ± 71 –31.6 NZA8021 103 8649 ± 70 –35.8 NZA8030 232 13461 ± 98 –40.0 NZA8045 A10-02 4 3688 ± 73 –27.7 NZA8025 120 8851 ± 73 –34.1 NZA8026 A10-08 10 6234 ± 69 –32.3 NZA8022 38 9365 ± 93 –38.6 NZA8023 345Khim & Yoon 2003: Polar Research 22(2), 341–353 iation in sedimentological and geochemical prop- erties. Sedimentation rates of core sediments Linear sedimentation rates were calculated by inter polation between the 14C-dated levels, although sedimentation rates might vary during depos ition (Fig. 2). The apparent linear sedimen- tation rates are 27.8 cm/Ky (thousand years) for core A10-01, 22.7 cm/Ky for core A10-02 and 9.0 cm/Ky for core A10-08. The sedimentation rates in cores A10-01 and A10-02 are comparable to those of glacimarine sediments in Marguerite Bay and Gerlache Strait in the Antarctic Penin- sula (20 - 50 cm/Ky; Harden et al. 1992), but still one order of magnitude lower than those in Lal- lemand Fjord, Andvord Bay and Granite Harbor (100 - 250 cm/Ky; Leventer et al. 1993; Domack & McClennen 1996; Kirby et al. 1998; Taylor et al. 2001). The sedimentation rate in Porter Cove (core A10-08) is one order of magnitude lower than in other areas of Maxwell Bay, in spite of the meltwater discharge from many small creeks (Klöser et al. 1994). Such a low sedimentation rate may be due to low amounts of suspended particles delivered by meltwater from the dis- charge basin. Alternatively, the suspended par- ticles delivered into Potter Cove may transit to Maxwell Bay because of a small depositional accommodation and the tidal action. Grain size of core sediments Downcore variation of sediment properties, including granulometric components and geo- chemical properties such as TOC and TS, are shown in Fig. 3. Measured AMS radiocarbon ages are also added on the core profi les; there is no reversal in the age throughout the cores. Based on sediment properties, all three cores can be divided clearly into two lithologic units: (1) an upper unit composed of fi ne-grained mud with scattered, minor ice-rafted debris; and (2) a lower unit composed mostly of diamicton. X- radiographs of core sediments clearly identify the distinct textural contrast of the two litholog- ic units (Fig. 4). The upper unit is dominated by silt and clay in cores A10-01 and A10-02, but sand and gravel occur in core A10-08—as much as 25 % in con- tent (Fig. 3). The coarse fraction in core A10-08 is probably accounted for by the core’s location close to the coast, where terrigenous clastic par- ticles are common. The lower unit of diamicton is characterized by mixed sediments with large amounts of gravel and sand (Fig. 3). The near absence of gravel in core A10-02 may be attrib- uted to the limited activity of the coastal gla- ciers. Although the vertical profi le of mean grain size shows an overall decreasing trend, more fre- quent fl uctuations occur in the lower unit due to the diamictic sediment composition. The distinct textural contrast between the upper and lower lithologic units is also identifi ed by the mode of change in individual grain size classes (Fig. 5). Such a change in grain size mode refl ects varia- tion with respect to the sediment transport mech- anisms. The formation of the lower unit can be explained by advanced glacier margins with rapid iceberg discharge and limited meltwater supply. Because it is generally diffi cult to distin- guish subglacial tills from glacial marine diamic- tons based solely on visual appearance or a single attribute of the sediments (see Licht et al. 1999), further analytical research on diamictons is nec- essary to constrain the interpretation of deposi- tional environments. TOC, TS and their relationship with grain size data Downcore variation of TOC content shows up wardly increasing values, matching the litho- logic changes (Fig. 3). In the diamicton, TOC contents are less than 0.1 %, whereas the fi ne- grained sediments contain as much as 0.5 %. Such a high TOC value is similar to that of the report- ed modern surface sediments from Andvord Bay (Domack & Ishman 1993). TOC contents usually depend on sediment grain size. Most of the organ- ic matter in shallow marine sediments is closely linked to the mineral matrix, occurring as organ- ic coating adsorbed on the mineral surfaces (Keil et al. 1994). Thus, a negative correlation between TOC and sediment coarseness is almost univer- sally observed in the noncarbonate sediments in marine and lacustrine environments. Figure 6 shows the relationship between TOC contents and mean grain size. TOC contents of the core sediments are largely determined by the amounts of clay particles in the sediments, although over- all TOC content is mostly low. The positive line- arity between the two parameters is expected in spite of the different correlation coeffi cient in the size fractions. 346 Postglacial marine environmental changes in Maxwell Bay, King George Island Fig. 3. Vertical profi les of sedi- ment properties (granulometric content, mean grain size, total organic carbon and total sulfur contents) from cores (a) A10-01, (b) A10-02 and (c) A10-08. Measured radiocarbon ages are indicated at the right side of column. The hatched area represents the transition zone from diamicton to mud. (a) (b) (c) 347Khim & Yoon 2003: Polar Research 22(2), 341–353 It has been suggested that the ratio of organ- ic carbon to pyrite sulfur (TOC/TS) can be used to distinguish between marine and freshwater sediments (Berner & Raiswell 1984). In normal marine sediments, the sulfur mainly originates from organic matter and/or pyrite minerals. The limiting factor for pyrite formation under normal oxic seawater conditions is the amount of organ- ic matter, which determines reducing conditions in the near-surface sediments (Berner 1982). Figure 6 also shows the interesting relationships between TOC and TS in core sediments. Two divided opposite trends between TOC and TS are observed in core A10-08 (Fig. 6f), which is dif- fi cult to explain. In contrast, the other two rela- tionships of cores A10-01 and A10-02 are more hyperbolic (Fig. 6d, e). In the case of the hyper- bolic relationship, the pyrite seems to be the main contributor of surplus sulfur because of the low TOC content and therefore the insuffi cient for- mation of sulfi de minerals in situ in the reduc- ing environment. In core sediments of A10-01 and A10-02, high TS contents are closely relat- ed to the sand fraction (Figs. 4, 6), implying that terrestrial material seems to be a main source for sulfur. The increase of TS content can be attrib- uted to an input of terrestrial pyrite of hydrother- mal origin—quartz–pyrite rocks, which have been recognized and described by many inves- Fig. 4. X-radiograph of core A10-01 showing the lithologic feature of upper (fi ne-grained mud with scattered ice-rafted debris) and lower sedimentary (mostly diamictons) units. Fig. 5. Variation of grain size class distribution in sediment core A10-01. Histogram of grain size classes represents the change of mode from lower to upper lithologic units. 348 Postglacial marine environmental changes in Maxwell Bay, King George Island tigators on King George Island (Littlefair 1978; So et al. 1995). Stable-isotope composition of benthic foraminifera The downcore variation in δ18O and δ13C values of benthic foraminifera shows the deglacial tran- sition, similar to the lithologic variation (Fig. 7). The hatched areas mark the transition zone between the lower and upper sedimentary units (Fig. 3). Both δ18O and δ13C values vary in a small range (Fig. 8), indicating relatively stable environmental conditions for the stable-isotope compositions. The δ18O values lie in the small range, between 3.7 ‰ and 4.1 ‰. However, some upcore increase in δ18O values can be identifi ed in the upper unit (Fig. 7), although the increasing trend in δ18O values is not distinct in core A10- 08. The differences of δ18O values between the upper and lower units are about 0.1 ‰ for cores A10-01 and A10-02. The δ13C foraminiferal values in cores A10- 01 and A10-02 vary slightly between 0 and 1 ‰ (Fig. 8). Clustering of core A10-08 is clearly dis- tinguishable. The δ13C values show a larger vari- Fig. 6. (a - c) Relationships between total organic carbon (TOC) and mean grain size, (d - f ) between TOC and total sulfur (TS), and (g - i) between TS and sand content. The surplus sulfur with low TOC is observed in cores A10-01 and A10-02. High TS contents are closely related to the sand fraction in A10-01 and A10-02, implying that terrestrial material is the main source for sulfur. (a) (b) (c) (d) (e) (f) (g) (h) (i) 349Khim & Yoon 2003: Polar Research 22(2), 341–353 ation and lighter values of –1.5 to 0 ‰ in this core. This may be due to a depth difference between the core sites, with the shallowest depth of A10- 08 infl uenced by meltwater discharge. All three cores show a gradual increase in δ13C values in the upper unit, possibly refl ecting the infl ow of shelf water with sea level rise and glacier retreat (Yoon et al. 2000; Khim et al. 2001). In addition, the higher productivity during the warm post- glacial period enhanced the content of heavy carbon in the seawater carbonate pool (Arthur et al. 1983). Fluctuations of δ13C values through- out the lower unit of cores A10-02 and A10-08 are likely due to redeposition of pre-glacial sedi- ments, through reworking of the coarser particles. Otherwise, the poor preservation of foraminife- ral specimens and their secondary alteration may affect the isotope signals. Discussion Postglacial marine environmental change in Maxwell Bay Sedimentary and geochemical properties show that all three cores in Maxwell Bay are charac- terized by two lithologic units (Fig. 3). The upper unit consists of fi ne-grained mud with scattered, minor ice-rafted debris. The lower unit is mostly diamicton. The main boundaries of the lithologic changes from diamictons to the fi ne-grained mud are at downcore depths of approximately 105, 120 and 45 cm in cores A10-01, A10-02 and A10-08, respectively. Among three cores, the transition- al boundaries of A10-01 and A10-02 are com- parable in age: 8649 and 8951 yr BP, respective- ly. Based on previous studies in the Antarctic Peninsula, the fi ne-grained sediments compris- ing the upper unit document open water condi- tions of an ice-distal environment, whereas the Fig. 7. Downcore variation of stable isotopes in benthic foraminifera. The hatched area represents the transition zone from diamicton to mud. Fig. 8. Biplot of δ18O and δ13C values of benthic foraminif- era. 350 Postglacial marine environmental changes in Maxwell Bay, King George Island coarse diamictons in the lower unit record an ice-proximal environment (see Shevenell et al. 1996; Yoon et al. 1997; Kirby et al. 1998; Pudsey 2000; Taylor et al. 2001). Thus, the accumulation of fi ne-grained mud in the upper unit was likely infl uenced by sediment-deriven meltwater infl ux during a warm climate regime after the comple- tion of deglaciation (Kirby et al. 1998). This indi- cates fast deglaciation, probably due to a sea level determined ice break-up and subsequent calving (Yoon et al. 2000). The resulting upper unit in the sediment cores is characterized by relatively high TOC, derived from enhanced productivity con- comitant with nutrient enrichment (Fig. 3). In polar fjord sediments, organic carbon enrichment may result from the increased verti- cal settling of organic carbon due to high produc- tivity in surface water (Domack & McClennen 1996; Kirby et al. 1998). Otherwise, less dilu- tion by more terrigenous input associated with ice-proximal meltwater plumes and/or rework- ing of bottom sediments may enhance the pres- ervation of TOC within the sediments (Shevenell et al. 1996). The low TOC contents in the lower unit may be attributed to the limit and reduction of organic carbon production mainly due to the advanced coastal glacier margin, with extensive sea ice coverage inhibiting primary production in surface water. This period corresponds roughly with a cold condition recorded on nearby Living- stone Island and in lake sediments on James Ross Island (Björck et al. 1996). It also matches a cold event documented in the sediment cores from Lallimand Fjord, Antarctic Peninsula (Shevenell et al. 1996). In contrast, the high TOC contents in the upper unit may be caused by the enhanced primary production in the surface water, due to a higher delivery of nutrients by meltwater dis- charge during the warm period (Yoon et al. 2000; Khim et al. 2001; Taylor et al. 2001). The increased productivity in the upper unit is also supported by the biogenic silica (Bsi) content and positive relationship between the Bsi and the TOC (Fig. 9). Such high TOC is largely account- ed for by increased diatom abundance. Thus, the clear disparity in TOC contents corresponding to lithologic changes refl ects a range of deposition- al environments, from ice-proximal to ice-distal condition (Fig. 3). The environmental transition from diamic- ton to fi ne-grained sediments is also supported by δ18O variation in benthic foraminifera from shallow marine environments, although it is ten- uous because of complex hydrographic param- eters determining the stable-isotope composi- tions (Arthur et al. 1983). In particular, dominant incorporation of oxygen from freshwater to the carbonate precipitation complicates the deter- mination of the temperature effect and the δ18O content of seawater. Furthermore, stable temper- atures in Antarctic Ocean waters could hardly add any noticeable temperature-controlled vari- ation in the carbonate stable-isotope composition (Charles & Fairbanks 1990). Today, the bottom- water temperature in Maxwell Bay is stable and consistently lower than 0 °C (Khim et al. 1997). During the deglacial period, the bottom-water temperature may have been higher. Here, we Fig. 9. Comparison of vertical variation between total organic carbon (TOC) and biogenic silica (Bsi) (modifi ed from Kim et al. 1999; printed with permis- sion of Geosciences Journal). The strong positive relationship indicates a high biological productivity in the upper unit, largely composed of diatoms. 351Khim & Yoon 2003: Polar Research 22(2), 341–353 assume that benthic foraminifera in this envi- ronment mostly represent bottom-water stable- isotope properties by inheriting the δ18O and δ13C compositions of ambient seawater at the time of precipitation. During the postglacial time, the increasing δ18O values may be due to the marine seawater intruding into Maxwell Bay from Brans- fi eld Strait, not the decrease of bottom-water tem- perature. Although the surfi cial meltwater dis- charge increases with warming, the bottom-water in Maxwell Bay is still connected with the saline water of Bransfi eld Strait. Higher δ18O values (− 0.2 ‰) of shelf water in the strait are contrasted with those (− 0.3 ‰) of seawater in Maxwell Bay (Khim et al. 1997). We infer that with the retreat of grounded coastal glaciers and the reduction of sea ice coverage, more seawater from Bransfi eld Strait entered Maxwell Bay, resulting in heavi- er δ18O values in benthic foraminifers. Howev- er, later in the Holocene, δ18O values decreased, probably due either to higher temperatures or to stronger dilution by meltwater. Holocene marine environmental change in South Shetlands Islands Marine geological and geophysical studies have shown that the shallow marine environments around the Antarctic Peninsula were covered by grounded ice sheets during the last glacia- tion (Domack & Ishman 1993; Pudsey et al. 1994; Pud sey & Camerlenghi 1998; Shipp et al. 1999). In particular, sedimentary and lithologic inves- tigations have demonstrated that Lapeyrere Bay (Anvers Island), Cierva Cove (Danco Coast), Lal- lemand Fjord, Andvord Bay and Brialmont Cove preserve a characteristic change in sediment lithology type from glacier-proximal coarse- grained sediments to distal sandy, biosiliceous muds (Domack & McClennen 1996; Kirby et al. 1998; Khim et al. 2001; Taylor et al. 2001). Down- core variation in TOC complements the interpre- tation of depositional environments and allows the reconstruction of palaeo environmental con- ditions. The elevation of TOC contents represents a time when the vertical fl ux of organic carbon throughout the fjord was higher, under favourable preservation conditions. The lithologic transition from coarse-grained diamicton to homogeneous mud in Maxwell Bay refl ects a major coastal-ice retreat, dated to approximately 8000 yr BP. A similar date was obtained from the lake sediment cores in King George Island (Mäusbacher et al. 1989). Assum- ing that the ice sheet break-up was bathymetri- cally controlled by ice retreat toward the island, we infer a similar time for the glacimarine– marine transition in King George Island; this is also supported by dates from the marine cores in Admiralty Bay, King George Island (Khim et al. 2001). During postglacial period, the ground-line retreating of the coastal glacier and the dimin- ished sea ice coverage resulted in rapid and enhanced primary production in Maxwell Bay. With the increased productivity, abundant benth- ic foraminifera were deposited in the postglacial sediment, with an upcore trend of increasing TOC (Fig. 3; Li et al. 2000). Judging from the steady increase in TOC, the climate became gradually warmer. Regionally, this interpretation is cor- roborated by lake sediments from King George Island (Schmidt et al. 1990) and James Ross Island (Björck et al. 1996). It also corresponds to the period of open marine conditions recorded in the sediment core from Lallimand Fjord, Antarc- tic Peninsula (Shevenell et al. 1996). The postglacial records of South Shetland Islands, including Maxwell Bay of King George Island, are comparable to those of the fjords of the Antarctic Peninsula, and East and West Ant- arctica, revealing both similarities and anoma- lies between the regions (Domack & McClennen 1996). The sedimentological and geochemical data generally represent a transition from ice- proximal to open marine deposition. This is strongly supported by the presence of the ice- assoc iated diatom assemblage in open marine envi ron ments (Leventer et al. 1993; Taylor et al. 2001). In addition, the increasing TOC suggest increasing primary production and may refl ect dilu tion of this signal due to increased siliclastic sed imentation. During the mid-Holocene, the cli- mate generally became warmer and more humid, showing high productivity peaks in the marine records. In particular, lake sediment data from South Georgia (Rosqvist et al. 1999) and James Ross Island (Björck et al. 1996), and even marine sed iments in the Ross Sea, suggest that atmos- pheric temperatures were warmer than present during that time. Conclusions Sediment textures and geochemical properties of three sediment cores from Maxwell Bay, King 352 Postglacial marine environmental changes in Maxwell Bay, King George Island George Island, record changes in glacimarine environments, which occurred during coastal gla- cier retreat, accompanied by increased meltwa- ter discharge and postglacial warming. Coupled lithologic and geochemical changes demonstrate that the lower diamictons represent advanced glacier margins and/or extensive sea ice cover, which limited primary biological production and meltwater supply. In contrast, the younger unit, composed mainly of fi ne-grained particles, indi- cates more open water and warmer conditions, favourable for enhanced production. Comparison between TOC and TS suggests that the additional sulfur, characteristic of the lower unit, originat- ed from the nearby hydrothermal-origin, quartz– pyrite rocks of King George Island, but not from the in situ pyrite formation under reducing condi- tions. Benthic foraminiferal stable-isotope com- position indicates that bottom-water conditions in Maxwell Bay were affected by shelf water inputs from Bransfi eld Strait after deglaciation. The lithologic transition occurring at about 8000 yr BP (14C age), in the mid-Holocene, supports pre- vious results reported for King George Island. Acknowledgements.⎯This research was supported by the Korea Polar Research Program (PP03106). We thank the cap- tain and crew of the RV Yuzhmorgeologiya. Outdoor sampling of gravity cores and laboratory work were specially assist- ed by C. Y. Kang and Y. S. Yang, both of the Korea Ocean Research and Developement Institute. Thanks go to dr J. M. Kim (University of California, Santa Barbara) for analysing the stable-isotope composition of the benthic foraminifera. Dr H. J. Lee’s (Korea Ocean Research and Developement Insti- tute) painstaking efforts reading and editing the manuscript are appreciated. We thank Dr B. Diekmann and an anony- mous reviewer for their critical reviews, which improved our interpretation. References Andrews, J. T., Domack, E. W., Cunningham, W. L., Levent- er, A., Licht, K. J., Jull, A. J. T., DeMaster, D. J. & Jenning, A. 1999: Problems and possible solutions concerning radio- carbon dating of surface marine sediments, Ross Sea, Ant- arctica. Quat. Res. 52, 206–216. Arthur, M. A., Anderson, T. F., Kaplan, I. R., Veizer, J. & Land. L. S. 1983: Stable isotopes in sedimentary geology. SEPM Short Course Notes 10. Tulsa, OK: Society for Sedi- mentary Geology. Berkman, P. A. & Forman, S. L. 1996: Pre-bomb radiocarbon and the reservoir correction for calcareous marine species in the Southern Ocean. Geophys. Res. Lett. 23, 363–366. Berner, R. A. 1982: Burial of organic carbon and pyrite sulfur in the modern ocean: its geochemical and environmental signifi cance. Am. J. Sci. 278, 451–473. Berner, R.A. & Raiswell, R. 1984: C/S method for distin- guishing freshwater from marine sedimentary rocks. Geol- ogy 12, 365–368. Björck, S., Håkansson, H., Olsson, S., Barnekow, L. & Jans- sens, J. 1993: Paleoclimatic studies in South Shetland Islands, Antarctica, based on numerous stratigraphic varia- bles in lake sediments. J. Paleolimnol. 8, 233–272. Björck, S., Hjort, C., Ingólfsson, O. & Skog, G. 1991: Radio- carbon dates from the Antarctic Peninsula—problems and potential. In J. J. Lowe (ed): Radiocarbon dating: recent applications and future potential. Pp. 55–65. Cambridge: Cambridge University Press. Björck, S., Olsson, S., Ellis-Evans, C., Håkansson, H., Hum- lum, O. & Manuel de Lirio, J. 1996: Late Holocene palaeo- climatic records from lake sediments on James Ross Island, Antarctica. Palaeogeogr. Palaeoclimatol. Palaeoecol. 121, 195–220. Canals, M., Urgeles, R. & Calafat, A.M. 2000: Deep sea-fl oor evidence of past ice streams off the Antarctic Peninsula. Geology 28, 31–34. Charles, C. D. & Fairbanks, R. G. 1990: Glacial to interglacial changes in the isotopic gradients of Southern Ocean sur- face water. In U. Bleil & J. Thiede (eds.): Geologic history of the polar oceans: Arctic versus Antarctic. Pp. 519–538. Dordrecht, The Netherlands: Kluwer. Clapperton, C. M. & Sugden, D. E. 1982: Late Quaternary glacial history of George VI Sound area, West Antarctica. Quaternary Research 18, 243-267. Coplen, T. 1996: New guidelines for reporting stable hydro- gen, carbon, and oxygen isotope-ratio data. Geochim. Cos- mochim. Acta 60, 3359–3360. Craig, H. 1957: Isotopic standards for carbon and oxygen and correction factors for mass spectrometric analysis of CO2. Geochim. Cosmochim. Acta 12, 133–149. Domack, E. W. 1992: Modern carbon-14 ages and reservoir corrections for the Antarctic Peninsula and Gerlache Strait area. Antarct. J. U. S. 27, 63–64. Domack, E. W. & Ishman, S. 1993: Oceanographic and physi- ographic controls on modern sedimentation within Antarc- tic fjords. Geol. Soc. Am. Bull. 105, 1175–1189. Domack, E. W., Jull, A. J. T., Anderson, J. B., Linick, T. W. & Williams, C. R. 1989: Application of tandem accelera- tor mass spectrometer dating to late Pleistocene–Holocene sediments of the East Antarctic continental shelf. Quat. Res. 31, 277–297. Domack, E. W. & McClennen, C. E. 1996: Accumulation of glacial marine sediments in fjords of the Antarctic Penin- sula and their use as Late Holocene palaeoenvironmental indicators. Antarct. Res. Ser. 70, 135–154. Gibson, J. A. E., Trull, T., Nichols, P. D., Summons, R. W. & McMinn, A. 1999: Sedimentation of 13C-rich organic matter from Antarctic sea-ice algae: a potential indicator of past sea ice extent. Geology 27, 331–334. Gordon, J. E. & Harkness, D. E. 1992: Magnitude and geo- graphic variation of the radiocarbon content in Antarctic marine life: implications for reservoir corrections in radio- carbon dating. Quat. Sci. Rev. 11, 697–708. Harden, S. L., DeMaster, D. J. & Nittrouer, C. A. 1992: Devel- oping sediment geochronologies for high-latitude con- tinental shelf deposits: a radiochemical approach. Mar. Geol. 103, 69–97. Hong, G. H., Kim, D. Y., Chung, H. & Pae, S. 1991: Coast- 353Khim & Yoon 2003: Polar Research 22(2), 341–353 al and inshore water interaction, mixing and primary pro- ductivity in the Bransfi eld Strait, Antarctica during austral summer, 1989/1990. Korean J. Polar Res. 2, 43–59. John, B. S. 1972: Evidence from the South Shetland Islands towards a glacial history of West Antarctica. In D. E. Sugden & R. J. Price (eds.): Polar geomorphology. Pp. 75– 92. London: Institute of British Geographers. Keil, R. D., Tsamakis, E., Fuh, C. B., Giddings, J. C. & Hedges, J. I. 1994: Mineralogical and textural controls on the organic composition of coastal marine sediments: hydro dynamic separation using SPLITT-fractionation. Geochim. Cosmochim. Acta 58, 879–893. Khim, B. K., Park, B. K. & Yoon, H. I. 1997: Oxygen isotop- ic compositions of seawater in the Maxwell Bay of King George Island, West Antarctica. Geosci. J. 1, 115–121. Khim, B. K., Yoon, H. I., Kim, Y. & Shin, I. C. 2001: The Holocene stable isotope chronology and meltwater dis- charge event in King George Island of South Shetland Islands, Antarctica. Antarct. Sci. 13, 167–173. Kim, D., Park, B. K., Yoon, H. I. & Kang, C. Y. 1999: Geo- chemical evidence for Holocene paleoclimatic changes in Maxwell Bay of South Shetland Islands, West Antarctica. Geosci. J. 3, 55–62. Kirby, M. E., Domack, E. W. & McClennen, C. E. 1998: Mag- netic stratigraphy and sedimentology of Holocene glacial marine deposits in the Palmer Deep, Bellingshausen Sea, Antarctica: implications for climate change? Mar. Geol. 152, 247–259. Klöser, H., Ferrerya, G., Schloss, I., Mercuri, G., Laturnus, F. & Curtosi, A. 1994: Hydrography of Potter Cove, a small fjord-like inlet on King George Island (South Shetlands). Estuar. Coast. Shelf Sci. 38, 523–537. Larter, R. D. & Vanneste, L. E. 1995: Relict subglacial deltas on the Antarctic Peninsula outer shelf. Geology 23, 33–36. Leventer, A., Dunbar, R. B. & DeMaster, D. J. 1993: Diatom evidence for late Holocene climatic events in Granite Harbor, Antarctica. Paleoceanography 8, 373–386. Li, B., Yoon, H. I. & Park, B. K. 2000: Foraminiferal assem- blages and CaCO3 dissolution since the last deglaciation in the Maxwell Bay, King George Island, Antarctica. Mar. Geol. 169, 239–257. Licht, K. J., Dunbar, N. W., Andrews, J. T. & Jennings, A. E. 1999: Distinguishing subglacial till and glacial marine diamictons in the western Ross Sea, Antarctica: implica- tions for a Last Glacial Maximum grounding line. Geol. Soc. Am. Bull. 111, 91–103. Littlefair, M. J. 1978: The quartz-pyrite rock of the South Shetland Islands, western Antarctic Peninsula. Econ. Geol, 73, 1184–1189. Martinez-Macchiavello, J. C., Tatur, A., Servant-Vildary, S. & Del Valle, R. 1996: Holocene environmental change in a marine–estuarine sediment sequence, King George Island, South Shetland Islands. Antarct. Sc. 8, 313–322. Mäusbacher, R., Muller, J., Munnich, M. & Schmidt, R. 1989: Evolution of postglacial sedimentation in Antarctic lakes (King George Island). Z. Geomorphol. 33, 219–234. Pudsey, C. J. 2000: Sedimentation on the continental rise west of Antarctic Peninsula over the last three glacial cycles. Mar. Geol. 167, 313–338. Pudsey, C. J., Barker, P. F. & Larter, R. D. 1994: Ice sheet retreat from the Antarctic Peninsula shelf. Cont. Shelf Res. 14, 1647–1675. Pudsey, C. J. & Camerlenghi, A. 1998: Glacial–interglacial deposition on a sediment drift on the Pacifi c margin of the Antarctic Peninsula. Antarct. Sci. 10, 286–308. Pudsey, C. J. & Evans, J. 2001: First survey of Antarctic sub- ice shelf sediments reveals mid-Holocene ice shelf retreat. Geology 29, 787–790. Rosqvist, G. C., Rieti-Shati, M. & Shemesh, A. 1999: Late glacial to middle Holocene climatic record of lacustrine biogenic silica oxygen isotopes from a Southern Ocean island. Geology 27, 967–970. Schmidt, R., Mäusbacher, R. & Muller, J. 1990: Holocene diatom fl ora and stratigraphy from sediment cores of two Antarctic lakes (King George Island). J. Paleolimnol. 3, 55–74. Shevenell, A. E., Domack, E. W. & Kernan, G. M. 1996: Record of Holocene paleoclimate change along the Antarc- tic Peninsula: evidence from glacial marine sediments, Lal- lenmand Fjord. Pap. Proc. R. Soc. Tasm. 130, 55–64. Shipp, S. Anderson, J. B. & Domack, E. W. 1999: Seismic sig- nature of the late Pleistocene fl uctuation of the West Ant- arctic Ice Sheet System in the Ross Sea: a new perspec- tive. Part I—geophysical results. Geol. Soc. Am. Bull. 111, 1486–1516. So, C. S., Yun, S. T. & Park, M. E. 1995: Geochemistry of a fossil hydrothermal system at Barton Peninsula, King George Island. Antarct. Sci. 7, 63–72. Sudgen, D. E. & Clapperton, C. M. 1977: The maximum ice extent on island groups in the Scotia Sea, Antarctica. Quat. Res. 7, 268–282. Sudgen, D. E. & Clapperton, C. M. 1986: Glacial history of the Antarctic Peninsula and South Georgia. S. Afr. J. Sci. 82, 508–509. Syvitski, J. P. M. 1989: On the deposition of sediment within glacier-infl uenced fjords: oceanographic controls. Mar. Geol. 85, 301–329. Taylor, F., Whitehead, J. & Domack, E. 2001: Holocene pale- oclimate change in the Antarctic Peninsula: evidence from the diatom, sedimentary and geochemical record. Mar. Micropaleontol. 41, 25–43. Tortell, P. D., Reinfelder, J. R. & Morel, F. M. M. 1997: Active uptake of bicarbonate by diatoms. Nature 390, 243–244. Yoon, H., Han, M. W., Park, B. K., Oh, J. K. & Chang, S. K. 1997: Glaciomarine sedimentation and paleo-glacial set- ting of Maxwell Bay and its tributary embayment, Marian Cove, in the South Shetland Islands, West Antarctica. Mar. Geol. 140, 265–282. Yoon, H., Park, B. K., Domack, E. W. & Kim, Y. 1998: Dis- tribution and dispersal pattern of suspended particulate matter in Maxwell Bay and its tributary, Marian Cove, in the South Shetland Islands, West Antarctica. Mar. Geol. 152, 261–275. Yoon, H. I., Park, B. K., Kim, Y. & Kim, D. 2000: Glacioma- rine sedimentation and its paleoceanographic implications along the fjord margins in the South Shetland Islands, Ant- arctica during the last 6000 years. Palaeogeogr. Palaeocli- matol. Palaeoecol. 157, 189–211. << /ASCII85EncodePages false /AllowTransparency false /AutoPositionEPSFiles true /AutoRotatePages /All /Binding /Left /CalGrayProfile (Dot Gain 20%) /CalRGBProfile (sRGB IEC61966-2.1) /CalCMYKProfile (U.S. Web Coated \050SWOP\051 v2) /sRGBProfile (sRGB IEC61966-2.1) /CannotEmbedFontPolicy /Warning /CompatibilityLevel 1.4 /CompressObjects /Tags /CompressPages true /ConvertImagesToIndexed true /PassThroughJPEGImages true /CreateJDFFile false /CreateJobTicket false /DefaultRenderingIntent /Default /DetectBlends true /ColorConversionStrategy /LeaveColorUnchanged /DoThumbnails false /EmbedAllFonts true /EmbedJobOptions true /DSCReportingLevel 0 /SyntheticBoldness 1.00 /EmitDSCWarnings false /EndPage -1 /ImageMemory 1048576 /LockDistillerParams false /MaxSubsetPct 100 /Optimize true /OPM 1 /ParseDSCComments true /ParseDSCCommentsForDocInfo true /PreserveCopyPage true /PreserveEPSInfo true /PreserveHalftoneInfo false /PreserveOPIComments false /PreserveOverprintSettings true /StartPage 1 /SubsetFonts true /TransferFunctionInfo /Apply /UCRandBGInfo /Preserve /UsePrologue false /ColorSettingsFile () /AlwaysEmbed [ true ] /NeverEmbed [ true ] /AntiAliasColorImages false /DownsampleColorImages true /ColorImageDownsampleType /Bicubic /ColorImageResolution 300 /ColorImageDepth -1 /ColorImageDownsampleThreshold 1.50000 /EncodeColorImages true /ColorImageFilter /DCTEncode /AutoFilterColorImages true /ColorImageAutoFilterStrategy /JPEG /ColorACSImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /ColorImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /JPEG2000ColorACSImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /JPEG2000ColorImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /AntiAliasGrayImages false /DownsampleGrayImages true /GrayImageDownsampleType /Bicubic /GrayImageResolution 300 /GrayImageDepth -1 /GrayImageDownsampleThreshold 1.50000 /EncodeGrayImages true /GrayImageFilter /DCTEncode /AutoFilterGrayImages true /GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /GrayImageDict << /QFactor 0.15 /HSamples [1 1 1 1] /VSamples [1 1 1 1] >> /JPEG2000GrayACSImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /JPEG2000GrayImageDict << /TileWidth 256 /TileHeight 256 /Quality 30 >> /AntiAliasMonoImages false /DownsampleMonoImages true /MonoImageDownsampleType /Bicubic /MonoImageResolution 1200 /MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50000 /EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode /MonoImageDict << /K -1 >> /AllowPSXObjects false /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false /PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox true /PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ] /PDFXOutputIntentProfile () /PDFXOutputCondition () /PDFXRegistryName (http://www.color.org) /PDFXTrapped /Unknown /Description << /FRA /ENU (Use these settings to create PDF documents with higher image resolution for improved printing quality. The PDF documents can be opened with Acrobat and Reader 5.0 and later.) /JPN /DEU /PTB /DAN /NLD /ESP /SUO /ITA /NOR /SVE >> >> setdistillerparams << /HWResolution [2400 2400] /PageSize [612.000 792.000] >> setpagedevice