Holocene paleoclimatic variation in the Schirmacher Oasis, East Antarctica: A mineral magnetic approach

Available online at www.sciencedirect.com

ScienceDirect Polar Science 8 (2014) 357e369 http://ees.elsevier.com/polar/

Holocene paleoclimatic variation in the Schirmacher Oasis, East Antarctica: A mineral magnetic approach Binita Phartiyal Birbal Sahni Institute of Palaeobotany, 53-University Road, Lucknow 226007, UP, India Received 17 July 2013; revised 27 May 2014; accepted 9 June 2014 Available online 14 June 2014

Abstract An analysis of remanent magnetism and radiocarbon ages in the dry lacustrine/sediment fills of the Schirmacher Oasis (SO) in East Antarctica was conducted to reconstruct past climatic condition. The statistically run mineral magnetic data on paleontological statistics software package (multivariate cluster analysis) placed on accelerator mass spectrometer radiocarbon chronology of the three sediment sections, trace 6 phases of climatic fluctuation between 13 and 3 ka, (Phases 1, 3 and 5 represent cold periods while Phases 2, 4, and 6 represent warm periods). One short warm period (Phase 2, ca. 12.5 ka) occurred in the late Pleistocene, and two marked warm periods (Phase 4, 11e8.7 ka; Phase 6, 4.4e3 ka) occurred in the Holocene. High magnetic susceptibility (c), saturation isothermal remanent magnetism (SIRM), and soft isothermal remanent magnetism (soft IRM) values correspond to colder periods and low values reflect comparatively warmer lacustrine phases. Holocene Optima (Phase 4) and Mid Holocene Hypsithermal (Phase 6) are distinguished by decreased values of concentrations dependent parameters. Remanence is preserved in the low-coercive minerals. Heavy metals in the sediments include, Fe, Rb, Zn, Mo, Co, Pb, Mn, Cu, and As in order of decreasing abundance. © 2014 Elsevier B.V. and NIPR. All rights reserved.

Keywords: Holocene; Schirmacher Oasis; Lacustrine sediments; Mineral remanent magnetism; Paleoclimate

1. Introduction Reconstructions of past environment changes yield insights into the natural variability of the Earth's climate system and provide a context for understanding present and future global changes. The universal presence of magnetic minerals in sediments and their sensitivity to climatic conditions allows for investigations of mineral magnetism (also known as environmental magnetism or rock magnetism) in a E-mail addresses: [email protected], binitaphartiyal@ bsip.res.in. http://dx.doi.org/10.1016/j.polar.2014.06.001 1873-9652/© 2014 Elsevier B.V. and NIPR. All rights reserved.

wide range of geographic locations and depositional environments (Thompson and Oldfield, 1986; Maher and Thompson, 1999; Evan and Heller, 2003). Magnetic parameters provide rapid, robust, and economic measures of qualitative, and in some cases quantitative, compositions of magnetic minerals and their relative associations under different climate conditions and depositional environments (Thompson and Oldfield, 1986; Maher and Thompson, 1999; Evan and Heller, 2003; Peters and Dekkers, 2003; Phartiyal et al., 2003; Warrier et al., 2011). The proxies used in this study include concentration-dependent parameters (magnetic susceptibility (c) and saturation isothermal

358

B. Phartiyal / Polar Science 8 (2014) 357e369

remanent magnetism (SIRM)), grain-size-dependent parameters (cARM,, SIRM/c, cfd%), and a mineralogy-dependent parameter (S-ratio). Magnetic parameters are sensitive indicators of temporal variations in the concentrations and grain sizes of magnetic minerals deposited in lakes (Jelinowska et al., 1997). They can also provide information about the circumstances surrounding the deposition of sediments, as well as about post-depositional processes (Berner, 1981; Maher and Taylor, 1988; Mann et al., 1990). This study presents the mineral magnetic parameters in sediments of Schirmacher Oasis (SO), Antarctica, as proxies to decipher the palaeoclimate of this region. The sediment records in the SO preserve the last period of late Pleistocene and nearly the entire Holocene (~13e3 ka). 2. Geographic setting and geology of the study area The study was conducted in the SO, an ice-free 35km2 area situated in Dronning Maud Land (East Antarctica) at 11
220 4000 e11
540 2000 E, 70
430 5000 e 70
460 4000 S (Fig. 1). The region is located between the margin of the ice sheet and areas of shelf ice. The SO consists of low-lying hills up to 203 m in elevation, and with an average elevation of 100 m; it contains 118 interspersed glacial lakes classified as pro-glacial, landlocked, and epi-shelf lakes (Ravindra, 2001) (Fig. 1). On the northern margin of the SO, towards the ice shelf, steep cliffs with ~100 m of relief border the SO, whereas the southern margin is mostly covered by the inland ice sheet. Extensive debris cover, moraine deposits, hillocky relief, and former lakes form a part of the present-day landscape. The SO is a recent periglacial region that has been unaffected by anthropogenic activity (Krause et al., 1997). The present lakes are mostly remnants of larger glacial lakes that occupied the valleys in the past (Phartiyal et al., 2011). The Precambrian crystalline basement of the SO, which forms a part of the East Antarctic craton (Sengupta, 1991), is polymetamorphic. The rocks have been strongly and repeatedly deformed under granulite- and amphibolite-facies conditions (Hoch, 1999). The rock types consist of charnockites, enderbites, garnetesillimanite gneisses, garnetebiotite gneisses, quartzo-feldspathic augen gneisses with minor amounts of foliated lamprophyres, amphibolites, dolerites, metagabbros, and metabasalts (Sengupta, 1991). The rock suites dominantly represent Grenvillean (1000 Ma) and PanAfrican (550 Ma) events.

3. Materials and methods

The sedimentary sequences were mapped over the entire 35 km2 of the SO. Two cores, 30 and 75 cm in length (DLL and PDL, respectively), were manually collected using 4.000 diameter plastic Polyvinyl Chloride pipes and sliced into 19 and 55 samples, respectively, using wooden knives to avoid contamination. In addition, 23 samples were collected from the eastern part of the SO (from the 30-cm-thick SWDL section) by the pit-and-channel method. Standard techniques were used for sample preparation (Walden, 1999a). The samples were air dried in the laboratory under normal temperature conditions (28e30
C), and gently crushed to powder; a 10-g sample was obtained by coning and quartering, and was transferred to nonmagnetic sample holders for analysis. Several magnetic parameters were obtained. Magnetic susceptibility (c) was determined on susceptibility meter (Bartington Susceptibility meter; Model MS2; with a dual sensor; Bartington Instruments Ltd, Oxford, England). The susceptibility used here are of low frequency (0.47 kHz; clf). Anhysteric Remanent Magnetization (ARM) was induced in a constant biasing field of 0.1 mT superimposed on a decaying alternating field (a.f.) with a peak field strength of 100 mT at the decay rate of 0.001 mT per cycle (using Molspin Alternating Field Demagnetizer with ARM attachment, Molspin Ltd., (presently owned by Bartington Instruments Ltd., Oxford, England)). The susceptibility of ARM (cARM) was calculated by dividing the mass-specific ARM by the magnitude of the biasing field (0.1 mT ¼ 79.6 A m1; Walden, 1999b). Isothermal remanent magnetization (IRM) was induced in the samples at different field strengths (20, 50, 70, 100, 200, 300, and 1500 mT in 100-mT increments) and backfields of up to 300 mT using an impulse magnetizer (ASC Scientific IM 10e30, Carlsbad, CA, USA). Remanence was measured using the Minispin Fluxgate magnetometer [Molspin Ltd., (presently owned by Bartington Instruments Ltd., Oxford, England)]. Three interparametric ratios were used in the analysis: frequency-dependent susceptibility (cfd%), the Sratio, and the SIRM/clf ratio. The cfd% value gives a broad estimate of the average size of ferromagnetic grains (Dearing, 1999). The S-ratio is the negative of the ratio of the IRM induced at 300 mT to the SIRM induced at 1500 mT (eIRMe300mT/SIRM1500mT). The SIRM/clf ratio gives a determination of grain size. For magnetic mineralogy measurements, the IRM acquisition was performed on all samples. Room

B. Phartiyal / Polar Science 8 (2014) 357e369

359

Schirmacher Oasis 800 Kms An

DRONNING MAUD LAND

ta

rc

tic

Pe

Larseman Hills

nin

su

la

Vestfold Hills Bunger Hills

QUEEN MARY LAND Windmill McMurdo Sound WILKES LAND

Islands

A 5’

11 º2

’

’

’

0 11º3

’

’

0 11º4

5 11º3

0 11º5

5 11º4

SWDL

N

Ice shelf

DLL

PDL

Streaky gneiss

1 km 70º45’

L

Augen gneiss Calc gneiss, khondalite Alaskite Garnet–biotite gneiss

P G

Continental ice sheet

Banded gneiss Pro-glacial lakes Land-locked lakes Epi-shelf lakes L-Long Lake; P-Pridarshani (Zub) Lake; G- Globokoye Lake Extent of lakes during the Late Quaternary (Phartiyal et al., 2011)

Core site

Pit/trench site

B

Fig. 1. A. Location map. B. Map of Schirmacher Oasis, showing geological and geomorphological details (modified after Sengupta, 1988 and Ravindra, 2001) and sampling sites.

Table 1 Details of AMS chronology after Phartiyal et al. (2011), based on dating of bulk sediment organic carbon. Silesian University of Technology, Gliwice, Poland 14

No.

Sample name

Lab. No.

Calibrated age range 68%

Calibrated age range 95%

Age

1 2 3

A1 A2 A3

GdA-1236 GdA-1237 GdA-1238

11093BC (95.4%) 10812BC 10773BC (95.4%) 10492BC 6226BC (95.4%) 5894BC

12900 ± 70 12580 ± 70 7230 ± 80

4

B1

GdA-1239

5

C1

GdA-1240

11022BC (68.2%) 10881BC 10702BC (68.2%) 10561BC 6205BC (3.4%) 6192BC 6183BC (3.0%) 6171BC 6159BC (3.9%) 6143BC 6104BC (54.3%) 5982BC 5942BC (3.6%) 5928BC 6372BC (55.5%) 6207BC 6188BC (0.8%) 6185BC 6169BC (2.0%) 6161BC 6141BC (10.0%) 6106BC 5711BC (51.4%) 5606BC 5596BC (16.8%) 5560BC

C (BP)

6412BC (95.4%) 6071BC

7430 ± 80

5742BC (95.4%) 5484BC

6770 ± 80

National Ocean Sciences AMS Facility (NOSAMS), Woods Hole Oceanographic Institute, USA No.

Sample name

Lab no.

d13C

F Modern

Fm error

Age

14

6 7

AS01 AS02

OS-69329 OS-69330

24.46 13.26

0.3346 0.6786

0.0019 0.0023

8790 ± 45 3110 ± 25

C (BP)

Sra t

)

RM

m

SI

(A

(1 0

io

/X

) kg A

SI R

M

kg m

(1 0

X

fd

X

%

AR M

)

) kg

lf

X

m

(1 0

lo g Li th o 0

C (A hro M no S) lo yr gy BP

B. Phartiyal / Polar Science 8 (2014) 357e369

y

360

C1-

8,790 ± 45 5

Depth (cm)

10

15

20

25

A S01-

6,770 ± 80 30

Base not exposed

87.5

Clay

92.5

97.5

Sand

0

4

Silt

8

12 540

620

700 34

42

50

0.3

0.4

0.5

0.6 0.85

0.89

0.93

Rock fragments

Fig. 2. Mineral magnetic parameters and inter-parametric ratios plotted against the SWDL section. AMS, accelerator mass spectrometry.

temperature hysteresis loops for selected samples were measured on 20e25-mg samples using a maximum field strength of 500 mT, using a Princeton Measurements alternating gradient force magnetometer (noise level, 106 A m2 kg1 for a 20-mg sample). Hightemperature susceptibility (ceT) was measured with an Agico KLY 2 Kappabridge in combination with a CS-2 heating unit. The analyses were performed at the Wadia Institute of Himalayan Geology, Dehradun, India, and at the Institut fu¨r Geologie und Pal€aontologie, Universitat Tu¨bingen, Germany. To identify mineral magnetic zonation in the profiles, the values of all parameters were entered into the paleontological statistics software package (PAST) statistical program, and analyzed using multivariate cluster analysis, following Ward's method (Hammer et al., 2001) for choosing the most appropriate parameter for assignment of magnetic zones. Elemental abundances were measured using an Xray fluorescence spectrometer (Model a-S4000S, Innov-X Systems) at the Directorate of Mining, Lucknow, India. The measurements were conducted in soil mode, with a precision of 10 ppm. Organic carbon in the bulk sediments dated by Phartiyal et al. (2011) were used in this paper. Seven accelerator mass spectrometer (AMS) dates were obtained from Silesian University of Technology, Gliwice, Poland on total organic carbon from bulk sediment samples. Standard sample preparation

methods were adapted to make graphite targets which were analyzed by mass accelerator along with primary and secondary standard process (Piotrowska, 2013). Details of chronologies are given in Table 1. 4. Results: litho-stratigraphy and mineral magnetism Periglacial lakes in the SO include several each of epi-shelf, landlocked, and proglacial lakes, and several dry lake beds are also present. Sedimentation rates in the periglacial lakes are very low (Shen et al., 1998). Three sedimentary sections (SWDL, DLL, PDL) were selected for the present study. The SWDL section, obtained from the western part of the SO (Fig. 2), is a 30-cm section composed of silty sand and clay, and rock fragments in the basal 2 cm. An age inversion in the AMS radiocarbon ages was detected in the sediments between the depths of 28 cm below the sediment surface (6770 ± 80 yr BP) (Sample No-SO1; Lab. No.GdA-1240) and 2 cm below the surface (8790 ± 45 yr BP) (Sample No-C1; Lab. No.-OS-69329). This inversion can be attributed to periglacial cryoturbation e a processes related to freezeethaw cycles, that mixes materials from various horizons in the sedimentary column. The mineral magnetic parameters and inter-parametric ratios for the SWDL section are plotted against lithology in Fig. 2. For the 23 levels of the studied section, the mean initial c, which is a

Mag (A m kg )

361

Normalized Intensity

Normalized Intensity

Normalized Intensity

B. Phartiyal / Polar Science 8 (2014) 357e369

Mag (A m kg )

Cooling curve Magnetisation (normalised)

Magnetisation (normalised)

Heating curve

Cooling curve

D 8-at 10.6-cm level

Heating curve

D 35-at 43.1 cm level

E

D

Fig. 3. AeC. Isothermal remanent magnetism (IRM) acquisition curves of samples at different depths in the SWDL, DLL, and PDL profiles, respectively. D. Hysteresis loops of samples from the DLL section; E e c-temperature runs of selected samples in the DLL section.

top of the section, and values of nearly 10% in basal layers (25e30 cm) are indicative of super paramagnetic (SP) grains. The SP grains contribute to c but not to the IRM; thus, their presence may be detected by lower SIRM/c values, which are observed at depths of 24e30 and 9e18 cm, indicating changes in the contents of ferro- and ferri-magnetic grains. (Thompson and Oldfield, 1986). The IRM acquisition curves show nearly 95% saturation at 300 mT (Fig. 3A),

representation of the bulk ferromagnetic content (along with the paramagnetic component), was 94.41  108 m3 kg1, with higher values being detected at depths of 0e9 and 16e24 cm. The SIRM values (1500 mT-SIRM values), which primarily represent the concentration of ferromagnetic materials (and which, unlike c, are unaffected by diamagnetic and paramagnetic minerals) are high at depths of 0e10 and 16e24 cm. Values of cfd% decrease towards the SWDL Section

Fe

Lithology

Rb

C18,790 ± 45

Co

10

Mo

15

Mo

Co

5

Depth (cm)

5

Depth (cm)

Depth (cm)

0

Fe

Rb

Mo Zn 15

20

Mo

25

30

25

Mo

A S016,770 ± 80 12000

ppm

22000

A

30

ppm

75

80

ppm

B

ppm

ppm

C

ppm

Chronology (AMS) yr BP

Fig. 4. Relative abundances of elements plotted against the lithology of the SWDL (A), the DLL (B), and the PDL (C) sections.

B. Phartiyal / Polar Science 8 (2014) 357e369

Table 2 Distribution of heavy metals, as measured by a portable X-ray fluorescence spectrometer, in the SWDL section (values in ppm; a hyphen indicates values below the detection range).

4.1. DLL Core The DLL section dates to 12,900 ± 70 yr BP at a depth of 65 cm (Sample No.-A1; Lab. No.-GdA-1236), 12,580 ± 70 yr BP at a depth of 52 cm (Sample No.A2; Lab No.-GdA-1237), and 7230 ± 90 yr BP in the upper 2 cm of the core (Sample No.-A3; Lab. No.GdA-1238). The 75-cm-thick core consists of alternating silty clay and sand (Fig. 5). The surface and basal layers are gravely, and the section sits on bedrock. The average susceptibility of the 52 studied levels is 53.59  108 m3 kg1. High susceptibilities, which indicate increased ferrimagnetic contents, occur at depths of 0e21, 42e49, and 58e73 cm. The SIRM values are higher at depths of 0e21 and 60e75 cm.

0

kg

)

SI R

A

M

kg M

0

AR

0

(1

X

(1

fd

X

0 (1

m

%

X

l m f kg

)

)

C (A hro M no S) lo yr gy BP

Li

th

ol

og

y

which is indicative of ‘soft’ low-coercive minerals. However, the possibility that hematite is present cannot be ruled out, as the c value of hematite is <102 that of magnetite and large amounts of hematite (10 times

io

34.64 53.17 36.63 71.62 46.96 52.56 54.69 39.51 64.28 59.44 50.92 49.90 32.31 43.50 64.16 45.50 40.23 62.11 56.26 83.04 57.31 47.70

Sra t

Rb

e e e e e e e 47.39 e 56.82 58.81 e e e e e e e e e e e

/X

Mo

15888.58 17679.13 18291.04 17413.46 18570.2 19624.91 17024.96 15220.07 19877.32 18454.81 19447.12 16641.8 13892.15 14683.13 23795.-8 22050.94 22766.64 18830.24 21785.87 21935.26 17761.24 19704.76

)

Fe

e e e 1101.11 e e e e e 1145.31 e e 911.45 e e e e e e 1142.96 e 1129.85

RM

Co

1.0 2.5 4.0 5.5 7.0 8.5 10.0 11.5 13.0 14.5 16.0 17.5 19.0 20.5 22.0 23.5 25.0 26.5 28.0 29.5 31.0 32.5

m

Depth (cm)

S1 S2 S3 S4 S5 S6 S7 S8 S9 S10 S11 S12 S13 S14 S16 S17 S18 S19 S20 S21 S22 S23

SI

SN

more than is currently present) can exist, even if the IRM acquisition curve reaches 95%. The grain size indicator parameter, SIRM/clf, suggests that grain sizes at depths of 0e10 and 16e25 cm are coarser than those at 23e30 and 10e16 cm. The Sratio varies in the range of 0.55e0.9 throughout the section, which is indicative of a mixed mineralogy with both soft and hard components (Bloemendal et al., 1988). The most abundant metals in the profile are Fe and Rb; Mo is present at depths of 11.5 and 14.5e16 cm, and Co is present at a depth of 5.5 cm (Fig. 4A and Table 2).

(A

362

A3-7,230 ± 80

10

Depth (cm)

20

30

40

50

A2-12,580 ± 70

60

A1-12,900 ± 70 70 Bedrock

44

Clay

52

Sand

60

68 0

2

Silt

4

6

228

276

324

120

280

440

2

4

6

0.5 0.6 0.7 0.8 0.9

Rock fragments

Fig. 5. Mineral magnetic parameters and inter-parametric ratios plotted against the DLL section.

B. Phartiyal / Polar Science 8 (2014) 357e369

363

Table 3 Distribution of heavy metals, as measured by a portable X-ray fluorescence spectrometer, in the DLL section (values in ppm; a hyphen indicates values below the detection range). SN

Depth (cm)

As

Co

Cu

Fe

Mn

Mo

Pb

Rb

Ti

Zn

D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D D

1.5 2.8 4.1 5.4 6.7 8.0 9.3 10.6 11.9 13.2 14.5 15.8 17.1 18.4 19.7 21.0 22.3 23.6 24.9 26.2 27.5 28.8 30.1 31.4 32.7 34.0 35.3 36.6 37.9 39.2 40.5 41.8 43.1 44.4 45.7 47.0 48.3 49.6 50.9 52.2 53.5 54.8 56.1 57.4 58.7 60.0 61.3 62.6 63.9 65.2 66.5 67.0

e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e 29.05 e e e e e e e e e e e e e e e

e 613.84 847.66 e e e e e e e e 847.18 e 785.73 e e e e e e e e e e e e 733.91 e e e e e e e e e e e e e e e e e e e e e e e e e

e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e 26.58 e e e e e e e e e e e e e e e

14507.29 9901.77 13717.23 13473.00 10994.24 11599.32 10717.08 12254.25 9986.83 11161.60 15142.70 12649.10 14127.02 13799.82 15964.79 15159.72 13341.18 13410.40 12772.03 12261.12 14502.49 12147.33 15102.96 14031.44 12717.37 14743.05 13421.49 9347.32 12657.91 9864.88 11528.67 26489.22 24274.69 22471.77 29736.44 32254.26 12332.44 11216.31 4108.31 12664.73 7375.64 9406.03 11597.31 9438.45 29070.74 28875.33 12317.10 11838.23 14148.23 13400.97 12961.13 5888.52

e e e e e e e e e e e e e 623.94 e e e e e e e e e e e e e e e e e e e e e e 120.95 e e e e e e e e e e e e e e e

e e e 42.86 e e e e e e e e 45.05 e e e e e e e e 44.92 e e e e 80.19 e e 52.22 e e e e e e 14.93 e e e e e e e e e e e e e e e

e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e 228.27 106.64 e 385.62 e e e e e e e e e e e e e e e

39.19 45.67 50.69 49.47 70.76 39.04 41.30 41.91 50.85 44.70 44.84 45.82 42.37 49.26 63.05 74.12 49.9 46.51 43.19 47.9 55.35 44.34 56.62 55.5 51.48 58.76 42.02 31.88 37.57 37.85 57.35 70.58 48.55 48.75 39.18 55.66 16.60 60.01 27.08 49.63 56.28 38.92 44.30 43.08 37.96 66.41 28.77 44.07 45.21 39.83 44.62 47.8

e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e 1918.01 e e e e e e e 14720 e e e e e e e

e e e e e e e e e e e e e e e e e e e e 116.24 e e e e e e e e e e e e e e e 46.32 167.35 158.28 121.78 233.45 164.26 155.99 253.09 e 157.80 e 144.65 197.85 e 159.46 192.32

1 2 3 4 5 6 7 8 9 10 11 12 13 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 42 43 44 45 46 47 48 49 50 51 52 53 54 55

Sra t

m

SI

(A

(1

io

RM / ) X

)

M

A

0

SI R

kg 0

(1

X

m

AR M

% fd

X

kg

)

) kg

X

l 0 f m

(1

C (A hro M no S) lo yr gy BP

B. Phartiyal / Polar Science 8 (2014) 357e369

Li th ol og y

364

0 AS02-3,110 ± 25 5

Depth (cm)

10

15

20

25 B1-7,430 ± 80 30

Base not exposed

32

Clay

48

64

0.0

Silt

Sand

5.0

10.0

12

20

28

36 44 2.5

12.5

22.5

0.1

0.2

0.3

0.4 0.5 0.6 0.7 0.8 0.9

Rock fragments

Fig. 6. Mineral magnetic parameters and inter-parametric ratios plotted against the PDL section.

The cfd% value above 55 cm fluctuates around 6%, which is indicative of contributions of higher concentrations of SP grains. The SIRM/c values are low at depths of 20e60 cm, which is indicative of finer grain sizes. The S-ratio is 0.5e0.8, showing variable contributions of magnetic minerals. The maximum saturation was acquired at 300 mT (Fig. 3B). The Table 4 Distribution of heavy metals, as measured by a portable X-ray fluorescence spectrometer, in the PDL section (values in ppm; a hyphen indicates values below the detection range). SN

Depth (cm)

Co

Fe

Mo

Rb

Zn

P1 P2 P3 P4 P5 P6 P7 P8 P9 P10 P11 P12 P13 P14 P15 P16 P17 P18 P19

29.0 27.5 26.0 24.5 23.0 21.5 20.0 18.5 17.0 15.5 14.0 12.5 11.0 9.5 8.0 6.5 5.0 3.5 2.0

e e e e e e e e e e e e e e e e 706.03 e e

16900.37 17480.2 15720.93 16315.48 17248.21 14110.74 15659.02 17887.39 16305.28 17035.95 16646.92 14393.54 14367.75 16444.93 17452.81 16809.1 14905.15 13577.95 15254.42

e 43.59 e e e 40.95 e e e 54.6e e e 61.82 53.17 e e e e

46.77 39.60 60.12 58.60 89.78 70.33 74.75 30.38 69.62 47.16 40.46 80.82 43.19 56.39 80.85 79.71 94.10 75.60 76.52

e e e e e e e e e 125.54 e e 86.74 e e e e e e

hysteresis loops are generally thin (Fig. 3D), which is indicative of a mixture of magnetic constituents with different coercivities (Tauxe et al., 1996). Susceptibility versus temperature curves show a large decrease in susceptibility at 575
C for the majority of the samples, confirming magnetite as the primary remanence carrier (Fig. 3E). However, a sample at a depth of 43.1 cm shows a decrease in susceptibility at temperatures of 575e700
C, perhaps due to a contribution of some hard magnetic mineral. Fig. 4B and Table 3 show the distributions of elements in the section; Fe and Rb constitute the major metals. The Fe and c curves coincide in the lower section (depths of 21e75 cm); however, in the upper part of the section, Fe variations are small, and c values are higher, perhaps due to the contributions of remanence from dia- and paramagnetic minerals instead of Fe minerals. The section is enriched in As, Cu, Mn, Mo, Pb, Rb, T, and Zn at depths of 23e24 cm, and in Cu, Mn, and Mo at depths of 58.5e61 cm; Mo is present at depths of 36.5 and 46.5 cm, Zn is present at 49 cm, and Co is present at 75e77.5 cm (Table 3). 4.2. PDL Core The PDL core, obtained towards the eastern side of the SO, gives an AMS date at the base of the core (depth, 29 cm) of 7430 ± 80 yr BP (Sample No. B1; Lab. No. GdA-1239), while the topmost clays (depth,

365

M a Zo gne ne tic s

%

X

fd

m

)

SI R (A M

)

(1 0

(1 0

m

X

kg

AR M

) lf

m

X

kg

y og ol th Li

/X

lf

B. Phartiyal / Polar Science 8 (2014) 357e369

Depth (cm)

3,110 ± 25

1e

Xfd%

1d

SIRM/X S-ratio

1c

XARM SIRM

Xlf 0.2

0.3

0.4 0.0

5.0

kg

M

M

A

0

SI R

(1

Coph. corr., 0.905

D6

7,230 ± 80

XARM

D5

20

Depth (cm)

10.0

ag Zo net ne ic s

36 44 0.1

) 0 (1

0 (1

0

AR m M kg

X

kg

X

m

lf

)

og ol Li

th

28

%

20

fd

12

X

64

)

48

y

32

SIRM

Xfd%

D4 40

SIRM/X D3

12,580 ± 70

S-ratio

60

D2 12,900 ± 70

B

Xlf

D1

80

44

Coph. corr., 0.9901

52

60

68 228

276

324

120

280

440

0

2

4

6

Fig. 7. Zonation of the PDL section (A) and the DLL section (B) into magnetozones, using best representatives of the mineral magnetic dataset, as calculated by multivariate cluster analysis using Ward's method.

2 cm) date to 3110 ± 25 yr BP (Sample No. AS02; Lab. No. OS-69330). The 30-cm core is composed of clay and sand, with clayey basal layers, while the top layers contain more sand and gravel (Fig. 6). The average c value for the 19 studied levels is 52.84  108 m3 kg1. The concentration-dependent parameters (c and SIRM) show decreasing trends towards the surface, with sharp fluctuations at depths of 10e15 cm. The cfd% value ranges between 0% and 5%, except at depths of 15e24 cm. The cARM also shows sharp fluctuations at depths of 5e15 cm, and higher values at depths of 21e30 cm. The SIRM/clf ratios indicate that grain sizes are finer towards the top of the section (depths, 0e10 cm) and coarser in the lower section. The S-ratios are high near the surface (depths, 0e7 cm); however, overall, the S-ratio is in

the range of 0.6e0.9, which is indicative of mixed mineralogical contributions. A saturation of nearly 90% at 300 mT (Fig. 3C) is indicative of ‘soft’ lowcoercive minerals, as well as of hard magnetic minerals, which is similar to that observed in the other two profiles. Fig. 4 and Table 4 show the distribution of heavy metals in the section. As in the other sections, Fe and Rb are the major elements present; Mo is present at depths of 8e9.5, 21.5, and 27.5 cm, and Co and Zn are present at depths of 5 and 11 cm. 5. Discussion and conclusions High-latitude regions have experienced unusually large climatic and ecological changes since the last glaciation as compared with other regions on Earth,

B. Phartiyal / Polar Science 8 (2014) 357e369

during colder periods enhanced concentrations of magnetic minerals. During warm periods, on the other hand, fine sediments were deposited, as organic activity in these glacial lakes is negligible, and the magnetic signals of concentration are therefore reduced. Phartiyal et al. (2011) reported increased magnetic susceptibilities in samples with negligible organic carbon contents (<1%). Therefore, it is assumed that high c, SIRM, and soft IRM values correspond to cold intervals, and that low values of these parameters correspond to comparatively warmer climatic conditions. Using the AMS chronology and the ageedepth model of Phartiyal et al. (2011), the DLL and PDL sections are estimated to represent the climatic history of the SO during ca. 13e3 ka. Because of the reversed chronology in the SWDL section, this section was not

Glacial and environmental changes in

Northern islands deglaciate

6 Slow deglaciation Melting of dead ice; expanding avian habitats

Ice retreat on land

Deglaciation of southern part of the oasis

10

Phase 6

Estimate ages (ka)

P1c

6 P1b P1a

7 D6

Southern islands deglaciate; Shelf areas deglaciate

Cold, but warming climate

Deglaciation of Shelf areas Inner shelf areas deglaciate Increased martinity as glaciers and ice shelves retreat from continentl shelf areas

Cold

P1d

Intermediate climatic conditions

Warm

Increased sea marinty as ice shelfs retreat

8 D5

9 10 D4

11

D3

12 D2 D1

Fig. 8. Comparison chart of the circum-Antarctic climate records to that of Schirmacher Oasis.

Holocene Optimum

8

Onset of lake sedimentation

P1e

5 Climatic getting warmer-moss banks and lake sedimnts start to accumulate; penguines colonize coasal rookeries

P2

4

Brief cooling

Brief cooling Glacial readvance

3

Mid-Holocene Hypsithermal

Down-wasting of dead ice

Climatic optimum-Increased insolation controls climate optimum and deglaciation

Phase 5

Glacial readvance

Arid and cool climate

Cool climate

4

Climate readvance Optimum Open marine conditions

Schirmacher Oasis

Antarctic Peninsula Warming climate

Glacial

Climate Optimum

Warmer than present ?

Larseman Hills

Warming

Windmill Islands

Warming climate

(ka)

Bunger Hills

Vestfold Hills

Climatic Phases

and their susceptibility to the impacts of future warming are particularly high on account of cryosphereealbedo feedbacks involving changing sea-ice and glacial cover (CAPE Project Members, 2001; Moritz et al., 2002; Overpeck et al., 1997). Despite the relative stability of present-day climate-forcing mechanisms, as compared with those during glacial periods, the Holocene interval has experienced dramatic environmental changes. The lakes investigated in this study are in low-lying areas, which were part of a more extensive lake system in the past (Phartiyal et al., 2011). In cold arid desserts, lakes are covered by ice and snow for a major portion of each year. The geomorphic and landscape evolution of the SO in eastern Antarctica is such that sedimentation rates in the lakes are low. Moreover, sediment introduced from catchment areas by glacial ice and glacial meltwater

Magnetic zones

366

Phase 4

Phase 3 Phase 2 Phase 1

B. Phartiyal / Polar Science 8 (2014) 357e369

considered in the final interpretations of the region's climatic history. The calculated mean sedimentation rate for the PDL core is ~0.069 mm y1, and that for the DDL core is 0.132 mm y1 Shen et al. (1998) reported sedimentation rates of 0.004e1.1 mm y1 in lacustrine environments in other areas of Antarctica. A multivariate cluster analysis divided the seven parameters into four clusters. Because cfd%, SIRM/clf, and the S-ratio fall within one cluster, cfd% was selected as a representative measure from this cluster. Fig. 7A and B shows that the parameters cfd%, c, cARM, and SIRM explain most of the variation in the PDL and DLL sections. On the basis of variations observed in the above parameters, as well as the relative abundances of elements (Fig. 4C), the PDL section was divided into two major magnetic zones, P1 and P2, with P1 consisting of five subzones. The sediments in the SO are generally super paramagnetic and single domain. A mixed magnetic mineralogy, with both soft and hard components, is evident from the IRM acquisition curves, the thin shape of the hysteresis loops, and ceT values. The sources of Fe, Rb, Mn, Co, and Zn are gneisses of Pre-Cambrian crystalline basement in the SO, especially banded gneiss and Alaskite. The magnetic zones of the PDL (P1eP2) and DLL (D1eD6) exhibit a suite of proxy results that can be used to identify different climatic phases, specifically a set that includes high values of concentrationdependent parameters, representing cold climatic conditions, and a set that includes lower values of such parameters, representing warmer climatic conditions (Fig. 7). Fig. 8 shows the different climatic phases in the SO and their correlation with the data thus far collected from Vestfold Hills, Bunger Hills, Windmill Islands, Larseman Hills, and the Antarctic Peninsula (Wilson and Wellman, 1962, 1998; Kaup et al., 1993; Smith and Friedman, 1993; Doran et al., 1994, 2004; Verleyen et al., 2011; Govil et al., 2012). Phase 1 is a short-duration climatic phase occurring at ca. 12.9 ka, which was a prelude to Phase 2 at 12.5 ka, both formed under comparatively warmer climates. Phase 3, at ca. 11.8e11 ka, shows conditions similar to those of Phase 1, while the Larseman Hills area shows increased marine incursion and glacial retreat during this time. A warmer and longer Phase 4, from 11 to ca. 8.7 ka, recorded in the DLL sediments can be correlated with the worldwide Holocene Optimum, and also an apparent signature of deglaciation of shelf areas in the Antarctic (see Fig. 8). Phase 5, during 8.7e4.4 ka, was a period of an arid and cold climate, which included shorter and comparatively warmer signals

367

during 7.9e7.1 and 6.4e5.6 ka. The trend from 5.3 ka onwards is indicative of increasingly warm conditions, which continued until 3 ka, and which are comparable to conditions during the mid-Holocene Hypsithermal (MHH), which have been recorded in most of the studied sections in Antarctica to date (Bentley et al., 2005). In the SO, the period of 4.6e4.4 ka was a time of intermediate climatic conditions. The recorded timing of the MHH varies in different regions of Antarctica: 4e2.7 ka (Bentley et al., 2005, 2009), 4.5e2.8 ka on the Antarctic Peninsula, and 3.8e1.7 ka in maritime Antarctica (Hodgson and Convey, 2005; Jones and Bowser, 1979), as compared with ca. 4.4e3 ka in the SO. Land-based geological evidence from the Antarctic shows two marked warm periods during the Holocene, at 11.5e9 ka and 4e2 ka. Some marine records also show evidence of a climate optimum at ca. 7e3 ka (www.scar.org/treaty). These land-based and marine records correlate with records from the SO, which show a warmer period during 11 to ca. 8.7 ka and a gradual diminishment of harsh conditions from 5 to 3 ka. The understanding of past climatic trends is of considerable interest, from both social and environmental perspectives, as such an understanding will help predict possible effects of future climate change. As only a few regions in East Antarctica are currently deglaciated, the ice-free area of the SO represents a valuable repository of climatic data. Our ongoing work on chemical and biotic proxies will further strengthen the dataset of climate proxies from this region of the icy continent. Acknowledgments I sincerely thank Directors, Birbal Sahni Institute of Palaeobotany, Lucknow, India and National Centre for Antarctic and Ocean Research (NCAOR), Goa, India for their encouragement. I thank NCAOR, Goa, for selection to participate in the 25th Indian Antarctic Expedition (IAE), 2005e06. I am grateful to the field teams and all members of the 24th and 25th IAEs for help during fieldwork. I also thank Deutscher Akademischer Austauschdienst (DAAD) for the Wiedereinladungen Programme Fellowship, and Professor Erwin Appel of Universitat Tu¨bingen for assistance. I am grateful to Dr. Santosh K. Shah for assistance with the statistical analyses. I also thank the anonymous reviewers for useful suggestions and comments, which helped to bring the manuscript into its present form. Laboratory facilities provided by the Wadia Institute of

368

B. Phartiyal / Polar Science 8 (2014) 357e369

Himalayan Geology, Dehradun, India; the Directorate of Mining, Lucknow, India; and the Institut fu¨r Geologie und Pal€aontologie, Universitat Tu¨bingen, Germany; as well as the chronology provided by the Silesian University of Technology, Gliwice, Poland, and NOSAMS, USA, are duly acknowledged. References Bentley, M.J., Hodgson, D.A., Sugden, D.E., Roberts, S.J., Smith, J.A., Leng, M.J., Bryant, C., 2005. Early holocene retreat of the George VI ice shelf, Antarctic Peninsula. Geology 33 (3), 173e176. Bentley, M.J., Hodgson, D.A., Smith, J.A., Cofaigh, C.O., Domack, E.W., Larter, R.D., Roberts, S.J., Brachfeld, S., Leventer, A., Hjort, C., et al., 2009. Mechanisms of Holocene palaeoenvironmental change in the Antarctic Peninsula region. Holocene 19 (1), 51e69. Berner, R.A., 1981. A new geochemical classification of sedimentary environments. J. Sed. Petrol 51 (2), 359e366. Bloemendal, J., Lamb, B., King, J.W., 1988. Palaeoenvironmental implications of rock-magnetic properties of late Quaternary sediment cores from the eastern equatorial Atlantic. Paleoceanography 3, 61e87. CAPE Project Members, 2001. Holocene paleoclimate data from the Arctic: testing models of global climate change. Quat. Sci. Rev. 20, 1275e1287. Dearing, J.A., 1999. Magnetic susceptibility. In: Walden, J., Oldfield, F., Smith, J. (Eds.), Environmental Magnetism: A Practical Guide. Quaternary Research Association, London, pp. 63e88. Technical Guide, No. 6. Doran, P.T., Wharton Jr., R.A., Lyons, W.B., 1994. Palaeolimnology of the McMurdo dry valleys, Antarctica. J. Paleolimnol 10, 85e114. Doran, P.T., Wharton Jr., R.A., Lyons, W.B., 2004. Paleolimnology of the McMurdo dry valleys, Antarctica. J. Paleolimnol. 10 (2), 85e114. Evan, M.E., Heller, F.H., 2003. Environmental Magnetism. Academic Press, San Diego (311pp.). Govil, P., Asthana, R., Mazumder, A., Ravindra, R., 2012. Grain size distribution and its influence on biological productivity during Holocene in a fresh water lake in Larsemann Hills, Antarctica. Natl. Acad. Sci. Lett. http://dx.doi.org/10.1007/s40009-012-0013-2. Hammer, Ø., Harper, D.A.T., Ryan, P.D., 2001. PAST: paleontological statistics software package for education and data analysis. Palaeontol. Electronica 4 (1), 9. http://palaeo-electronica.org/ 2001_1/past/issue1_01.htm. Hoch, M., 1999. Geochemistry and petrology of ultramafic lamprophyres from Schirmacher Oasis, East Antarctica. Miner. Petrol 65, 51e67. Hodgson, D.A., Convey, P., 2005. A 7000-year record of oribatid mite communities on a maritime-Antarctic island: responses to climate change. Arct. Antarct. Alp. Res. 37 (2), 239e245. Jelinowska, A., Tucholka, P., Wieckowski, K., 1997. Magnetic properties of sediments in a Polish lake: evidence of a relation between the rock-magnetic record and environmental changes in Late Pleistocene and Holocene sediments. Geophys. J. Int. 129, 727e736. Jones, B.F., Bowser, C.J., 1979. The mineralogy and related chemistry of lake sediments. In: Lerman, A. (Ed.), Lakes: Chemistry, Geology, Physics. Springer Verlag, New York, pp. 179e227.

Kaup, E., Haendel, D., Vaikmae, R., 1993. Limnological features of the saline lakes of the Bunger Hills (Wilkes Land, Antarctica). Antarct. Sci. 5 (1), 41e50. Krause, W.E., Krbetschek, M.R., Stolz, W., 1997. Dating of Quaternary lake sediments from the Schirmacher Oasis (East Antarctica) by infra-red stimulated luminescence (IRSL) detected at the wavelength of 560 nm. Quat. Sci. Rev. 16, 387e392. Maher, B.A., Taylor, R.M., 1988. Formation of ultrafine-grained magnetite in soils. Nature 366, 368e370. Maher, B.A., Thompson, R., 1999. Quaternary Environments, Climates and Magnetism. xxiiþ390pp.. Cambridge University Press, Cambridge, New York, Melbourne. Mann, S., Sparks, N.H.C., Frankel, R.B., Bazylinski, D.A., Jannasch, H.W., 1990. Biomineralization of ferrimagnetic greigite (Fe3S4) and iron pyrite (FeS2) in a magnetotactic bacterium. Nature 343, 258e261. Moritz, R.E., Bitz, C.M., Steig, E.J., 2002. Dynamics of recent climate change in the Arctic. Science 297, 1497e1502. Overpeck, J., Hughen, K., Hardy, D., Bradley, R., Case, R., Douglas, M., Finney, B., Gajewski, K., Jacoby, G., Jennings, A., Lamoureux, S., Lasca, A., MacDonald, G., Moore, J., Retelle, M., Smith, S., Wolfe, A., Zielinski, G., 1997. Arctic environmental change of the last four centuries. Science 278, 1251e1256. Peters, C., Dekkers, M.J., 2003. Selected room temperature magnetic parameters as a function of mineralogy, concentration and grain size. Phys. Chem. Earth 28, 659e667. Phartiyal, B., Appel, E., Blaha, U., Hoffmann, V., Kotlia, B.S., 2003. Palaeoclimatic significance of magnetic properties from Late Quaternary lacustrine sediments at Pithoragarh, Kumaun Lesser Himalaya, India. Quat. Int 108, 51e62. Phartiyal, B., Sharma, A., Bera, S.K., 2011. Glacial lakes and geomorphological evolution of schirmacher oasis, east Antarctica, during late Quaternary. Quat. Int. 235, 128e136. Piotrowska, N., 2013. Status report of AMS sample preparation laboratory at GADAM Centre, Gliwice, Poland. Nucl. Instrumet. Meth. B 294, 176e181. Ravindra, R., 2001. Geomorphology of schirmacher oasis, east Antarctica. Proc. Symp. On snow, ice and glacier. Geol. Surv. India 53, 379e390. Sengupta, S., 1991. Structural and petrographical evolution of basement rocks in the Schirmacher hills, Queens Maud land, east Antarctica. In: Thompson, M.R.A., Crame, J.A., Thompson, J.W. (Eds.), Geological Evolution of Antarctica. Cambridge University Press, Cambridge, pp. 95e97. Shen, C., Liu, T., Yi, W., Sun, Y., Jiang, M., Beer, J., Banani, G., 1998. 14C dating of terrestrial moss in tern lake deposits, Antarctica. Radiocarbon 40 (1e2), 849. Smith, G.I., Friedman, I., 1993. Lithology and palaeoclimate implications of lacustrine deposits around Lake Vanda and Don Juan Pond, Antarctica. Antarct. Res. Ser. 59, 83e94. Tauxe, L., Mullender, T.A.T., Pick, T., 1996. Potbellies, wasp-waists and supermagnetism in magnetic hysteresis. J. Geophys. Res.Sol. Ea 101, 571e583. Thompson, R., Oldfield, F., 1986. Environmental Magnetism. Allen Unwin, London. Verleyen, E., Hodgson, D.A., Sabbe, K., Cremer, H., Emslie, S.D., Gibson, J., Hall, B., Imura, S., Kudoh, S., Marshall, G.J., McMinn, A., Melles, M., Newman, L., Roberts, D., Roberts, S.J., Singh, S.M., Sterken, M., Tavernier, I., Verkulich, S., Vyver, E., Van de Nieuwenhuyze, W.V., Wagne, B., Vyverman, W., 2011.

B. Phartiyal / Polar Science 8 (2014) 357e369 Post-glacial regional climate variability along the East Antarctic coastal margindEvidence from shallow marine and coastal terrestrial records. Earth Sci. Rev. 104, 199e212. Walden, J., 1999a. Sample collection and preparation. In: Walden, J., Oldfield, F., Smith, J. (Eds.), Environmental Magnetism: A Practical Guide. Technical Guide, No. 6. Quaternary Research Association, London, pp. 63e88. Walden, J., 1999b. Remanence measurements. In: Walden, J., Oldfield, F., Smith, J. (Eds.), Environmental Magnetism: A

369

Practical Guide. Quaternary Research Association, London, pp. 197e211. Technical Guide, No. 6. Warrier, A.K., Sandeep, K., Harshavardhana, B.G., Shankar, R., Pappu, S., Akhilesh, K., Prabhu, C.N., Gunnell, Y., 2011. A rock magnetic record of Pleistocene rainfall variations at the Palaeolithic site of Attirampakkam, Southeastern India. J. Archaeol. Sci. 38, 3681e3693. Wilson, A.T., Wellman, H.W., 1962. Lake Vanda: an Antarctic lake. Nature 196, 1171e1173.