Record of the early Holocene warming in a laminated sediment core from Cape Hallett Bay (Northern Victoria Land, Antarctica

www.elsevier.com/locate/gloplacha

Global and Planetary Chang

Record of the early Holocene warming in a laminated sediment

core from Cape Hallett Bay (Northern Victoria Land, Antarctica)

Furio Finocchiaroa,*, Leonardo Langoneb, Ester Colizzaa, Giorgio Fontolana,

Federico Gigliob, Eva Tuzzia

aDipartimento Scienze Geologiche, Ambientali e Marine, Universita di Trieste, v. E. Weiss, 2-34127 Trieste, ItalybISMAR-CNR, Sezione Geologia Marina, Via Gobetti, 101-40129 Bologna, Italy

Received 16 September 2003; accepted 28 September 2004

Abstract

This paper presents an integrated multiproxy approach study (sedimentological, geochemical, preliminary smear-slides

diatom assemblages, and 14C ages analyses) performed on a sediment core collected in Cape Hallett Bay (Ross Sea, Antarctica).

Sediments record the early Holocene rapid climate changes: buried varved diatomaceous ooze on the base of core (N9.5–9.4 ka

BP) are linked to the early Holocene warming and open marine conditions. From 9.4 ka BP, the climate starts to cool (massive

mud). From 8.0 to 7.8 ka BP, sandy mud sediment suggests a rapid landward recession of the local/regional glaciers, with

relevant underflow inputs, together with the onset of seasonal sea-ice formation. The ages and the characteristics of the

youngest sediments are related to the changed oceanographic conditions linked to the retreat of the calving front of the Ross Ice

Shelf.

D 2004 Elsevier B.V. All rights reserved.

Keywords: Paleoenvironmental reconstruction; Laminated sediments; Early Holocene; Ross Sea; Antarctica

1. Introduction

Recently, research on paleoclimatic reconstruction

in Antarctic areas has focused on short-term climatic

changes during the Holocene. This period is crucial

for understanding the present climatic system and

0921-8181/$ - see front matter D 2004 Elsevier B.V. All rights reserved.

doi:10.1016/j.gloplacha.2004.09.003

* Corresponding author. Tel.: +39 040 5582025; fax: +39 040

5582048.

E-mail address: [email protected] (F. Finocchiaro).

predicting future global changes. Review of ice-core

data has identified two climatic optima during the

Holocene, the first between 11.5 and 9 ka BP, and a

second between 6 and 3 ka BP in the Eastern

Antarctic sector and between 7 and 5 ka BP for the

Ross Sea area (Masson et al., 2000). According to

Cias et al. (1992), the period of maximum Holocene

warmth in East Antarctica ice cores is between 10

and 7.5 ka BP; Siegert (2001) indicates a warm peak

at 9.4 ka BP in Southern Hemisphere. Timing of the

early climatic optimum coincides with a peak in

e 45 (2005) 193–206

F. Finocchiaro et al. / Global and Planetary Change 45 (2005) 193–206194

abundance of foraminifera in South Atlantic sedi-

ments (Hodell et al., 2001). Baroni and Orombelli

(1994) reported a mid-Holocene warm phase on the

basis of penguin rookery occupation along Victoria

Land coast.

The study of this restricted time interval needs

high-resolution geological records, and for marine

sediments, continuous and expanded sedimentary

successions. Marine successions characterized by

high-sedimentation rates in late Quaternary shelf

sediments around Antarctica have been found mainly

in embayments and fjords along the Antarctic

Peninsula coast. There, sedimentation rates are

Fig. 1. Location of core ANTA02-CH41. Bathymetric data and land morph

edited by U.S. Geological Survey.

higher than 100 cm ka�1 (Lallemand Fjord, Domack

et al., 1995; Palmer Deep, Leventer et al., 1996;

Domack et al., 2001) with maximum values of 1500

cm ka�1 (Brialmont Cove; Domack and McClennen,

1996). Recently, another site of very high accumu-

lation rate in laminated diatomaceous ooze (290 cm

ka�1) was found in George V Land continental shelf

(Harris et al., 2001). Marine sediments in this sector

also record two climatic optima: the first in the early

Holocene, the second one in the mid-Holocene

(Pudsey et al., 1994; Leventer et al., 1996; Domack

et al., 2001; Taylor et al., 2001; Taylor and McMinn,

2001; Presti et al., 2003; Goodridge, 1999–2000).

ology from map SS 58-60/2 (Cape Hallett), original scale 1:250,000,

F. Finocchiaro et al. / Global and Planetary Change 45 (2005) 193–206 195

Concerning the mid-Holocene optimum, authors are

not in agreement about exact timing and period of

this phase.

In the Ross Sea, information on Holocene

climatic changes based on the sediment record is

very scarce. Offshore basins of the Ross Sea

continental shelf show generally low sedimentation

rates: the highest values (ca. 20–24 cm ka�1)

characterize modern biogenic mud in the deepest

and central part of the Joides Basin (Frignani et al.,

1998; Finocchiaro et al., 2000). Only inside Granite

Harbor, where the thickness of diatomaceous ooze

can be at least 10 m (Domack et al., 1999), the

sedimentation rate is very high (up to 250 cm ka�1,

DeMaster et al., 1996), reinforcing the observation

that bays and fjords have excellent potential to

preserve high-resolution sedimentary records.

Many authors recognized that the retreat of the

grounding line in the Ross Sea began about 14–13 ka

BP (Stuiver et al., 1981; Denton et al., 1989; Licht et

al., 1996; Brambati et al., 1997). The general ice

withdrawal was affected by minor climatic fluctua-

tions (Steig et al., 1998; Brambati et al., 1997; Orsini

et al., 2003). Recently, detailed studies on diatom

assemblages have identified a warmer period between

6 and 3 ka (Cunningham et al., 1999), while high

resolution in late Holocene paleoclimatic reconstruc-

tion has been obtained from Granite Harbor sediments

(Leventer et al., 1993).

Evidence of the early Holocene climatic warm

phase is herein reported based on results from a

gravity core (ANTA02-CH41) collected in Cape

Hallett Bay (Lat. 72817.49VS–Long. 170809.05VE;water depth: 416 m) during the 2001–2002 austral

summer (Fig. 1). The early Holocene time interval is

well documented due to very high sediment accu-

mulation rates at the coring site. A succession of

climatic events was tentatively reconstructed based

on paleoenvironmental inferences.

2. Material and methods

Cape Hallett is located along the northern Victoria

Land coast, about 110 km north of Coulman Island

(Fig. 1). The sediment core was collected at the

entrance of the Edisto Inlet, along the conjunction

between Cape Hallett and Cape Christie. Edisto Inlet

is a small bay, about 15 km long and 4 km wide,

deeper than 500 m and separated by a sill from the

larger Moubray Bay (USGS, 1968). The Edisto

Glacier is small and flows into the inner part of the

bay, whereas a saddle at only 800 m above sea level

separates (a few kilometers southward) the bay from

the terminal section of the Tucker Glacier.

The core site is seaward (north) of the sill, and a

SBP 3.5 kHz seismic reflection profile shows parallel

and subparallel reflectors of a stratified sequence about

8 m thick (Bussi et al., 2003). The 408-cm-long

sediment core was collected from the site using a 2.3-

ton gravity corer. The core was scanned for magnetic

susceptibility, split, X-radiographed, described, and

sub-sampled.

The upper 3 m of core are massive and were

sampled in 1-cm-thick slices taken at 10-cm intervals.

Below 3 m, the sediment has alternating dark and

light layers. Forty-two layers were sampled along this

interval. Samples were dried at 60 8C and then

slightly disaggregated for the following analyses.

Porosity was calculated based on water content

according to Berner (1971) and assuming a mineral

density of 2.55 g cm�3.

Particle size was determined after treatment with

H2O2 and wet sieving at 63 Am. Sandy samples were

then analyzed by a MacroGranometer sedimentation

balance and muddy samples by a Sedigraph 5100 ET

(Micromeritics). A Coulter Multisizer (100-Am ori-

fice tube) was used for very small muddy samples

from the laminated section.

Organic carbon content of each sample was

determined using a FISONS NA2000 Elemental

Analyzer (EA) after removal of the carbonate

fraction by adding HCl 1.5 N. The errors associated

with determinations are around 1%. Biogenic silica

content was determined through a progressive dis-

solution method (DeMaster, 1981), followed by

colorimetric analysis. We used NaOH 0.5 M as an

extractant in view of the significant concentrations of

biogenic silica usually found in Antarctic samples

(DeMaster, 1981).

Forty-three smear slide observations provided

preliminary information on diatom assemblages. The

chronology of the sediment core was defined by

means of 7 AMS 14C ages determined on the bulk

organic fraction at Geochron Laboratories (Cam-

bridge, MA, USA).

F. Finocchiaro et al. / Global and Planetary Change 45 (2005) 193–206196

3. Results

Preliminary data of core ANTA02-CH41 were

reported by Finocchiaro et al. (2003).

Two main lithological units can be recognized

in the core (Fig. 2). In the lower unit (Unit A,

408–207 cm), two subunits (A1 and A2) can be

distinguished.

Sub-Unit A1 (408–307 cm) is characterized by a

rhythmic sequence of parallel light and dark mud

laminae, from 0.5 to 19.5 mm thick (Fig. 3). Only

10% of laminae has a thickness between 10.5 and

19.5 mm, and we report them as dthick laminaeT,following the terminology of Reineck and Singh

(1973). Colors of laminae vary from black-gray to

several shades of olive (dark olive to pale olive). In

general, contacts between light laminae and over-

lying dark laminae are gradational, whereas the

contacts from dark to light laminae are sharp. From

preliminary observations of smear slides, light

laminae are almost entirely composed of diatom

frustules, whereas volcaniclastic silt is a subordinate

component of dark laminae. The compositional

difference between light and dark laminae is also

testified by the slightly higher values of organic C

Fig. 2. Core ANTA02-CH41: Lithostratigraphy and unit identification; m

content (%) on wet weight; concentration of organic carbon and biogenic

(0.85 vs. 0.73 wt.% for light and dark laminae,

respectively), but their biogenic silica values are

similar (34.1 vs. 33.3 wt.%).

The particle size of laminae of different color was

measured: dark laminae are poorly sorted with a

coarse-grained tail, whereas olive laminae have the

same modal diameter (12–15 Am), although it is

more leptokurtic. Pale olive laminae have better

sorting and a finer mode (6–7 Am; Fig. 4). The

different particle size distributions of light laminae

are probably related to different diatom assemblages.

Some light laminae (pale olive) show a bfluffyQtexture, very similar to the description of the

bcottonyQ layer found in cores from Granite Harbor

and MacRobertson Bay (Leventer et al., 1993;

Taylor and McMinn, 2001).

Number and thickness of laminae were measured.

Two hundred and sixteen laminae have been defined,

well recognizable by both direct visualization and

different X-ray beam attenuation in radiographs.

Total thickness of 108 pairs of laminae (light+dark

lamina couplets) varies from 3 to 23 mm (mean

value of 8.9 mm) with frequent oscillations, but the

trend is to thin upward (Fig. 5a). The light laminae

are generally thicker than those dark in color. The

ass magnetic susceptibility (from Finocchiaro et al., 2003); water

silica (wt.%).

Fig. 3. Photograph (left) and positive X-ray (right) of the parallel

laminated interval Sub-Unit A1. Scale in centimeters, from the core

top.

F. Finocchiaro et al. / Global and Planetary Change 45 (2005) 193–206 197

ratio between dark lamina and total thickness of

light–dark pair tends to increase upwards, due to

reducing thickness of the light laminae, whereas the

dark laminae remain almost constant in thickness

(Fig. 5b).

Sub-Unit A2 (307–207 cm) is a massive dark

olive-grey mud with wavy and irregular laminae

concentrated between 253 and 207 cm. In compar-

ison with Sub-Unit A1, organic carbon and biogenic

silica decrease (mean values 0.77 and 18 wt.%,

respectively), whereas the sand content increases.

Grain size of massive sediment shows a prevalence

of silt, with a modal diameter between coarse silt and

very fine sand (Fig. 4).

The upper unit (Unit B, 207–0 cm) is charac-

terized by sand (Figs. 2 and 4) with dispersed gravel

clasts as dropstones. Two subunits can be distin-

guished in this section: the lower subunit (Sub-Unit

B1, 70–207 cm) is a dark olive-grey muddy sand

with a sand content varying between 33% and 85%.

Organic carbon and biogenic silica contents of Sub-

Unit B1 are the highest of Unit B (mean values, 0.37

and 5 wt.%, respectively). The Sub-Unit B2 (0–70

cm) is very dark grey, slightly muddy sand, with

sand content varying from 68% to 70%, and the

lowest values of organic carbon (0.24 wt.%) and

biogenic silica (2%).

Preliminary diatom analyses show that Corethron

pennatum (=C. criophilum: Crawford et al., 1998),

Fragilariopsis curta, together with Chaetoceros rest-

ing spores account for up to 96% of the diatom

assemblages in the whole core. C. pennatum occurs

only in the laminated levels (Sub-Unit A1 and only

partially in Sub-Unit A2), whereas F. curta and

Chaetoceros r.s. are found in different percentages

throughout the core. F. curta becomes dominant in the

upper, coarse-grained Unit B.

Finally, we tried to determine if the total

thickness of the light+dark pair and dark/light

lamina ratio have cyclical trends through time,

considering each couplet as a 1-year deposit.

Although the autocorrelation and Fourier analyses

did not give statistically significant results, total

thickness data and dark/light thickness ratios show

11- and 13-year major periodicity cycles, respec-

tively. Standard Z statistics was applied on auto-

correlation coefficients for the lag=11 and 13

years, giving Z t(11 years) =1.60 and Z t(13

Fig. 4. Particle-size data: (a) Sub-Unit A1: average frequency curves of dark, pale olive and dark olive laminae measured by Coulter Multisizer;

(b,c,d) from Sub-Units A2, B1, and B2, respectively, frequency curves of samples, measured by sedimentation balance and Sedigraph.

F. Finocchiaro et al. / Global and Planetary Change 45 (2005) 193–206198

years)=1.85, slightly lower than 95% probability

limit (Zt=1.96).

4. Age model and sediment accumulation rates

Due to the absence of carbonate, it was necessary

to date the bulk organic matter in the sediment. Seven

radiocarbon dates of bulk organic matter (Table 1)

were used to set the chronological constraint (Fig. 6)

and estimate sedimentation rates (Fig. 7).

Radiocarbon ages from Antarctic material must be

interpreted with great caution because of the uncer-

tainty of several factors, such as: reservoir effect, vital

effect among different organisms, and other minor

effects (Domack et al., 1999). In addition, sediments

may be variably contaminated with reworked carbon

(Harris et al., 1996). In this regard, diatomaceous mud

and ooze units provide the most accurate determina-

tions of radiocarbon age, because the material has the

highest concentration of autochthonous organic matter

of all the lithofacies. Ages at or prior to the last glacial

maximum (LGM) are interpreted as mixed (Domack et

al., 1999). Most authors use a reservoir correction

ranging between 1.2 and 1.5 ka for the Ross Sea region

(Stuiver et al., 1981; Licht and Andrews, 1997;

Berkman, 1997; Licht et al., 1996, 1999) and for other

Antarctic shelf areas (Pudsey et al., 1994; Shipp and

Anderson, 1994; Gingele et al., 1997). Nevertheless,

the problems that plague radiocarbon dating and the

high spatial variability of surface sediment ages in the

Antarctic marine system mean that it is unlikely that a

reliable absolute age for a single-dated horizon within a

sediment core can be obtained (Domack et al., 1999).

Adjustments could be made by subtracting the age of

the organic matter at the sediment–water interface for

each core, but one needs to be certain that the

sediment–water interface has been recovered and

sediment mixing also must be evaluated (Domack et

al., 1999).

By comparing the magnetic susceptibility profiles

with a companion box-core collected in the same site

(data not shown), the uppermost 2 cm of the

sedimentary succession of core ANTA02-CH41 seems

to have been lost during coring operations. If a

sediment constant accumulation rate of 12.6 cm ka�1

is assumed between the levels 0–1 and 34–35 cm, then

interpolated age at the sediment–water interface is 1630

Fig. 5. Thickness variation of laminae of Sub-Unit A1 (307–408 cm) relative to: (a) 108 dark+light laminae couplets; (b) dark laminae, and (c)

light laminae.

F. Finocchiaro et al. / Global and Planetary Change 45 (2005) 193–206 199

years, which is very near to the most commonly used

value for reservoir corrections.

Radiocarbon results were consequently corrected by

subtracting the calculated age of the organic carbon at

the core top and the bias due to the top loss during core

sampling. According to the corrected data, the studied

core spans from early Holocene to the Recent. There is

a progression of ages down core, with the exception of

three dates measured in Sub-Unit A1, which are not

completely consistent with values above. Only part of

Table 1

AMS 14C ages of ANTA91-CH41 core, on bulk organic matter

Cruise Core Level Code 14C age (year BP) F(1r) d13C (x vs. PDB)

ANTA02 CH41 00–01 GX-29188 1790 40 �25.8

ANTA02 CH41 34–35 GX-30574 4490 50 �25.5

ANTA02 CH41 70–71 GX-29991 9400 40 �28.6

ANTA02 CH41 232 GX-29189 9970 50 �27.3

ANTA02 CH41 300 GX-29190 10,920 50 �26.7

ANTA02 CH41 369 CX-29992 9130 40 �31.9

ANTA02 CH41 402 GX-29191 10,070 50 �28.4

F. Finocchiaro et al. / Global and Planetary Change 45 (2005) 193–206200

the difference can be also accounted for by reproduci-

bility variability (Andrews et al., 1997, 1999). More

likely, the inversion could be due to the release of 14C

depleted glacial meltwater, a variable reservoir effect

through time (van Beek et al., 2002) or contamination

during subsampling. Alternatively, if the lamina

couplets are an annual record of surface productivity,

the Sub-Unit A1 comprises the sediment accumulating

in a very short time interval (about one century).

To overcome the problem of the lower three dates

and to calculate sediment accumulation rates, an age

model was developed.

For the uppermost 300 cm, ages of samples were

estimated assuming a constant sedimentation rate

Fig. 6. Age-depth model of the core ANTA02-CH41. Conventional14C dates were corrected for a reservoir age of 1630 years. The age

model was developed using an integrated approach: in the upper-

most part of the core (0–300 cm), ages were calculated based on a

constant sediment accumulation rate for each lithological unit; below

300 cm, an assumption was made that each couplet was a varve.

within each lithological unit. In detail, in the Sub-Unit

A2, a sediment accumulation rate of 71.6 cm ka�1 was

calculated based on corrected 14C dates measured at

232 and 300 cm. Then, the ages at the boundaries of

Sub-Unit A2 (207 and 307 cm) were extrapolated,

assuming for the whole subunit the same sedimentation

rate measured between 232 and 300 cm.

To calculate the sediment accumulation rate of Sub-

Unit B1, the measured and calculated ages of 70.5 and

207 cm depth, respectively, were used. Downcore, for

Sub-Unit A1 (307–408 cm), we assumed an annual

varved-like sedimentation (see previous discussion).

By assuming this age model (Fig. 6), we calculated

linear sediment accumulation rates and fluxes of bulk

mass, biogenic (organic carbon and biogenic silica) and

lithic components (Fig. 7).

In early Holocene, represented by Sub-Unit A1,

the linear accumulation rate is very high (926 cm

ka�1). Sediment accumulation rates decrease (71.6

cm ka�1) in Sub-Unit A2, increase again in Sub-Unit

B1 (615.7 cm ka�1), and then becomes low (9.3 cm

ka�1) during the last 7.8 ka�1 in Sub-Unit B2. A quite

different picture is obtained when we observe the

variation of mass accumulation rates with time (Fig.

7b) due to the different porosity and bulk dry density

between the sediment above and below 300 cm. In

fact, the mass accumulation rate shows the highest

values in Unit B1 (8.0–7.8 ka BP) and the lowest

(12.2 g cm�2 ka�1) in the last 7.8 ka BP (Unit B2).

However, the strong variability of sediment compo-

sition that characterizes the time interval between 9.5

and 7.8 ka results in a partition in three time intervals:

the oldest, equivalent to the laminated section (9.5–

9.4 ka), is characterized by the highest biogenic

fluxes of both organic carbon and biogenic silica (Fig

7c,d), whereas the youngest period (8.0–7.8 ka)

shows the highest fluxes of the lithogenic fraction

Fig. 7. (a) Linear (dashed line) and mass (solid line) accumulation rate vs. sediment depth (cm); (b) mass accumulation rate (g cm�2 ka�1) vs.

time (ka). Temporal variations of the mass accumulation rate (MAR) of biogenic silica (BSi), Corg and lithic fraction were plotted with more

detail between 10 and 7.5 ka BP, in (c), (d), and (e), respectively.

F. Finocchiaro et al. / Global and Planetary Change 45 (2005) 193–206 201

(Fig. 7e). The accumulation of both lithogenic and

biogenic components (Fig. 7c–e) are relatively low in

the intermediate time interval (9.4–8.0 ka).

5. Discussion and conclusions

5.1. Laminated sediments (Sub-Unit A1)

The most common stratigraphic succession on the

Ross Sea shelf has a basal diamicton unit of glacial

origin at its base (Anderson, 1999; Domack et al.,

1999; Brambati et al., 2002). Overlying facies are

composed of glacimarine silt and sandy intervals. The

uppermost portion of most cores collected in the Ross

Sea is a diatomaceous ooze (Dunbar et al., 1985; Licht

et al., 1996; Frignani et al., 1998; Langone et al.,

1998; Domack et al., 1999) interpreted as having been

deposited in seasonally open water conditions.

The presence of fine diatomaceous mud buried

under about 2 m of sandy sediment is peculiar of core

ANTA02-CH41, and indicates open marine condi-

tions during early Holocene. Despite the fact that

buried diatomaceous sediment was recently recog-

nized by Colizza et al. (2003) south of the Drygalski

basin, this stratigraphic sequence is quite rare in the

Ross Sea.

Fine-grained laminated sediments are widely

reported in Late Quaternary Antarctic sequences.

For example, highly laminated sediments characterize

the drift area of the Pacific margin of the Antarctic

Peninsula (Pudsey and Camerlenghi, 1998; Lucchi et

al., 2002), Palmer Deep and Gerlache Strait (Leventer

et al., 2002; Goodridge, 1999–2000), the MacRo-

bertson Shelf (Harris, 2000), the George V continen-

tal shelf (Wilkes Land Margin: Domack and

Anderson, 1983; Domack, 1988; Brancolini and

Harris, 2000; Presti et al., 2003), and outer slope

areas of the Ross Sea embayment (Bonaccorsi et al.,

2000). In the inner shelf of the Ross Sea, laminated

sediments are reported by Nishimura et al. (1998)

(north of Ross Island), Leventer et al. (1993) (Granite

F. Finocchiaro et al. / Global and Planetary Change 45 (2005) 193–206202

Harbour) and Colizza et al. (2003) (Wood Bay area).

In general, the preservation of parallel lamination in

marine sediment is considered a proxy for anoxic

conditions, that exclude the presence of a benthic

community, or a very high sedimentation rate in areas

characterized by upwelling or by favorable oceano-

graphic conditions (Grigorov et al., 2002; Cofaigh

and Dowdswell, 2001).

5.2. Facies interpretation and paleoenvironmental

reconstruction

Based on physical, geochemical, and biological

proxies, and using the age-depth model, a sequence of

depositional events driven by paleoclimate and

paleoenvironmental changes was tentatively recon-

structed. Particular attention was devoted to the early

Holocene, a time interval well preserved in sediments

of Cape Hallett Bay.

5.2.1. Sub-Unit A1 (N9.5–9.4 ka BP)

The high accumulation rate and high organic

carbon and biogenic silica contents in the Unit A

(about 2 m thick) of core ANTA02-CH41 suggest a

period of high biological productivity. Based on the

modern assemblage distribution (Cunningham and

Leventer, 1998), the Corethron occurrence is

associated with a well-stratified water column

linked to weak wind, thermal warming and/or sea-

ice meltwater (Leventer et al., 1996). The olive

laminae are mainly composed by Corethron and

Chaetoceros r.s. Few pale-olive, fluffy, thick lam-

inae contain a nearly monospecific Corethron

assemblage, which produces pulses of rapidly

sinking algal flocs during early spring blooms. In

the following dark lamina, the contribution of fine

glacial debris mainly from overflow plumes

increases through summer. Thus, the pair of

laminae depicts a varve-like sedimentation with a

seasonal alternation of productivity events (light

laminae) and deposition of terrigenous debris (dark

laminae).

All these proxies, together with the age model,

constrain the deposition of the lowest sedimentary

unit to the early Holocene warming.

Leventer et al. (1993) documented a layer of

Corethron ooze that was related to deposition during

the Medieval Warm Period.

5.2.2. Sub-Unit A2 (9.4–8.0 ka BP)

Climatic conditions were not constant during the

time of fine-grained sediment accumulation (Unit A):

the reduction of light lamina thickness in the upper

part of Sub-Unit A1 implies a general decrease in

productivity over the period, thus suggesting a

progressive climate cooling. The massive mud of

Sub-Unit A2 further indicates the end of the optimal

climatic conditions of the early Holocene. However,

the shifting is punctuated by several oscillations

marked by an alternation between massive and

irregularly laminated levels. In Sub-Unit A2, both

biogenic silica and sand contents show intermediate

values from the laminated interval and the Sub-Unit

B1, above.

The low lithic mass accumulation rate (about one

third of the Sub-Unit A1), together with the

significant increase of F. curta and Chaetoceros in

diatom assemblages, imply a climate cooling which

is likely to be associated to a more persistent sea-ice

covering. Subsequently, during a shorter and cooler

summer, a low amount of meltwater prevents the

formation of stratification along the water column; in

the same manner, fine sediment release from over-

flow plumes is sensibly lower than during the

Holocene warming.

The sedimentary signature is therefore given by the

superimposition of fine detrital material settled during

summer, arranged in amalgamated and structureless

(by bioturbation) layers, only occasionally interrupted

by laminated biosiliceous mud, typical of warmer

phases.

5.2.3. Sub-Unit B1 (8.0–7.8 ka BP)

The abrupt change at level 207 cm depth

(corresponding to 8.0 ka BP) can be related to a

dramatic environmental change. All paleoproducti-

vity markers decrease, particle-size becomes coarser

and consequently, the mass accumulation rate of

lithics exceeds the biogenics. Corethron disappears,

whereas F. curta becomes the prevailing diatom

species. High percentages of F. curta have been

associated with marginal sea-ice zones (Leventer et

al., 1996; Cunningham et al., 1999). The high

accumulation of the terrigenous sediment could

document the rapid landward recession of the local

and/or regional glaciers and the onset of seasonal sea-

ice formation.

F. Finocchiaro et al. / Global and Planetary Change 45 (2005) 193–206 203

In the Ross Sea region, the ice sheet retreated by 11

ka BP (Domack et al., 1999). A possible trigger of the

fast retreat of local glaciers could be linked to a

relative sea level rise following the regional retreat of

the East Antarctica Ice Sheet (EAIS). The rapid EAIS

retreat and related sea level rise may well have forced

instability of the local glacier grounding line pinned

on the morphological high at the entrance of Edisto

Inlet (Fig. 1). Consequently, the glacier grounding line

rapidly migrated to inner positions of the inlet.

In some cores from the continental shelf of the

Ross Sea, the transition from glacial (diamicton) to

glacimarine sediment and/or siliceous mud is marked

by a sorted muddy sand layer. The occurrence of this

layer was first pointed out by Kellogg et al. (1979)

and then described as a bgranulated faciesQ by

Domack et al. (1999). This coarse grained sediment

has been interpreted as a meltwater facies, related to

decoupling and lift-off of a recessional melting line of

the ice sheet. The depositional model interprets the

granulated facies as resulting from melting of basal ice

near the grounding zone, linked to strong bottom tidal

currents that have winnowed out the fine fractions,

thus increasing sand percentages and sorting (Domack

and Harris, 1998). The sandy level recognized at the

base of this subunit, may be the equivalent of the

granulated facies, representing the lift-off of local

glaciers in the Cape Hallett area.

Despite the moderate increase of the mud content,

the remnant upper part of the Sub-Unit B1 features

grain size distribution with marked sorting character-

istics (Fig. 4). These data, together with the high

sedimentary rates and increase in ice-rafted debris

(IRD), suggest a progressive landward recession of

the glacier front, which caused an abundant release of

dense underflow plumes. A new circulation pattern

inside the fjord, supported by high meltwater fluxes

and sea-level rise, is likely to be the main triggering

factor for bottom currents able to winnow and

transport sediments toward the deepest part of the

bay, far from the supply zone of coarse debris (glacier

margin subaqueous fan; Boulton, 1990).

5.2.4. Sub-Unit B2

Proxies at the base of Sub-Unit B1 (140–207 cm)

and in Sub-Unit B2 (0–70 cm) are quite similar,

suggesting that similar processes affected sedimenta-

tion processes, but the sedimentation rate dramatically

decreased to values typical of offshore Holocene

sediments of the Ross Sea: 9.1 cm ka�1 (Brambati et

al., 1997; Frignani et al., 1998) during the last 7.9 ka.

This difference may indicate the retreat of the calving

front of the Ross Ice Shelf over the study site at about

8.0 ka BP and the reduction of the glacial debris

supplied only by small glaciers flowing into the

Edisto Inlet.

Retreat of the ice sheet from most of the Western

Ross Sea area had as a consequence the onset of the

Ross Sea and Terra Nova polynyas and the formation

of High Salinity Shelf Water (HSSW), the densest

waters of the Ross Sea continental shelf. The HSSW

feeds bottom currents flowing out the Ross Sea along

the coast of the Victoria Land. Establishment of these

conditions also determined a more pronounced

intrusion of the Circumpolar Deep Water onto the

shelf (Denton et al., 1989) that, in turn, contributed to

raise to present-day levels the diatom productivity that

characterizes the southwestern Ross Sea and the

JOIDES basin.

The uppermost lithofacies (0–70 cm) of core

ANTA02-CH41 is a common feature in the north-

western Ross Sea. In fact, coarse sediment was also

observed in surficial sediments collected on the

continental shelf between Coulman Island and Cape

Adare (Melis et al., 2002), supporting the hypothesis

of a similar origin.

5.3. Concluding remarks

This study sheds new light on the sedimentary

record of rapid climatic changes affecting the early

Holocene in the Antarctic region. An integrated

multiproxy approach has allowed the documentation

of:

! A possible varved biogenic ooze, which implies a

recovery of the record of the early Holocene

warming in sediments of the Ross Sea. Sedimen-

tological and geochemical parameters, together

with floral assemblage observations, indicate this

time period (N9.5–9.4 ka BP) as being character-

ized by weak winds and thermal warming of the

strongly stratified upper water column. Biological

productivity was highly enhanced with substantial

early spring algal blooms, which produced epi-

sodic events of sediment deposition of rapidly

F. Finocchiaro et al. / Global and Planetary Change 45 (2005) 193–206204

sinking biogenic material. During the following

summers, thermal warming released a peak of fine

glacial debris. The result was the accumulation of

an annual couplet of alternating light and dark

laminae.

! Starting at 9.4 ka BP, the climate cooled. The algal

assemblages show more stressed environmental

conditions, and also, glacial melting diminished.

The shift to these conditions is quite well marked,

although interrupted by several short-lived warm

oscillations.

! From 8.0 to 7.8 ka BP, the sediment core records a

high flux of terrigenous material, coarse and well-

sorted at the beginning and finer at the end of this

time interval, which was interpreted as the product

of a rapidly receding glacier. The diatom assemb-

lages suggest that seasonally open marine con-

ditions were established, associated with marginal

sea-ice zones.

! The onset of the current general circulation system

of the Ross Sea was tentatively set at 7.8 ka BP, as a

consequence of the polynya and HSSW formation.

We demonstrate that the coastal bay near Cape

Hallett is a promising location to obtain detailed

paleoenvironmental records for paleoclimate recon-

structions. In this regard, the database should be further

improved: the sediment core did not recover the base of

the laminated layer, equivalent to the onset of the early

Holocene warming. Based on the SBP seismic profile,

the maximum thickness of the laminated sequence is

expected to be 4 to 5m thick. In addition, this study was

based on the analyses and interpretation of a single

sediment core. It is necessary to sample further sites in

the bay in order to understand to what extent of

confidence our paleoenvironmental interpretations can

be extrapolated on a spatial scale. Finally, our findings

have to be more closely integrated with regional

information from the Ross Sea.

Acknowledgments

Research carried out within the framework of

the Project 4.5 (P.I. Prof. A. Brambati) of the

Italian Programma Nazionale di Ricerche in Antar-

tide, and financially supported by ENEA. We thank

the crew and scientific party onboard R/V Italica

for their help with fieldwork during the ANTA02

cruise. The two referees Amy Leventer and Ross

D. Powell are fully acknowledged for the critical

review of the manuscript. The authors wish to

dedicate this paper to the seaman David Basciano,

who has left us too early. This is contribution No.

1434 of the ISMAR-CNR, Sezione Geologia

Marina di Bologna, Italy.

References

Anderson, J.B., 1999. Antarctic Marine Geology. Cambridge

University Press, Cambridge (UK), 289 pp.

Andrews, J.T., Jull, A.J.T., Leventer, A., 1997. Replication of

accelerator mass spectrometry carbon-14 dates on the acid-

insoluble fraction of Ross Sea surface sediments. Antarctic

Journal of the United States 32, 37–38.

Andrews, J.T., Domack, E.W., Cunningham, W.L., Leventer, A.,

Licht, K.J., Jull, A.J.T., DeMaster, D.J, Janning, A.E., 1999.

Problems and possible solutions concerning radiocarbon dating

of surface marine sediments, Ross Sea, Antarctica. Quaternary

Research 52, 206–216.

Baroni, C., Orombelli, G., 1994. Abandoned penguin rockeries as

Holocene paleoclimatic indicators in Antarctica. Geology 22,

23–26.

Berkman, P.A., 1997. Retreat of the West Antarctic Ice Sheet and

subsequent persistence of marine macrofaunal assemblages

along the Victoria Land coast during the Holocene. Antarctica

and Global Change: Interactions and Impacts, Hobart, Tasmania

Australia, 13–18 July 1997, Abstract, vol. 0805.

Berner, R.A., 1971. Principles of Chemical Sedimentology.

McGraw-Hill, New York, 240 pp.

Bonaccorsi, R., Brambati, A., Busetti, M., Fanzutti, G.P., 2000.

Relationship among X-Rays lithofacies, magnetic susceptibility,

P-waves velocity, and bulk density in core ANTA95-89 (Ross

Sea, Antarctica). Terra Antartica Reports 4, 241–258.

Boulton, G.S., 1990. Sedimentary and sea level changes durino

glacial cyclesand their control on glacimarine architecture. In:

Dowdeswell, J.A., Scourse, J.D. (Eds.), Glacimarine Environ-

ments: Processes and Sediments. Geological Soc. Spec. Publ.,

vol. 53, pp. 15–52.

Brambati, A., Fanzutti, G.P., Finocchiaro, F., Melis, R., Frignani,

M., Ravaioli, M., Setti, M., 1997. Paleoenvironmental record in

Core ANTA91-30 (Drygalski Basin, Ross Sea, Antarctica). In:

Cooper, A.K., Barker, P.F., Brancolini, G. (Eds.), Geology and

Seismic Stratigraphy of the Antarctic Margin, Part 2. Antarctic

Res. Series, vol. 71. Am. Geophys. Union, Washington, DC,

pp. 137–151.

Brambati, A., Corradi, N., Finocchiaro, F., Giglio, F., 2002. The

position of the last glacial maximum grounding line in the Joides

Basin. Royal Society of New Zealand Bulletin 35, 365–372.

Brancolini, G., Harris, P.T., Shipboard Party, 2000. Post-cruise

report: joint Italian/Australian marine geoscience expedition

aboard the R.V. Tangaroa to the George V Land Region of East

F. Finocchiaro et al. / Global and Planetary Change 45 (2005) 193–206 205

Antarctica during February–March 2000. AGSO Record

No.2000/19, 181 pp.

Bussi, M., Colizza, E., Corradi, N., Finocchiaro, F., Fontolan, G.,

Ivaldi, R., Landucci, C., Nicotra, G., Pitta’, A., Salvi, G., 2003.

Late Quaternary palaeoclimatic evolution of marine sediments

in the Southern Ocean project: preliminary results of the

20012002 PNRA cruise in the Ross Sea. Terra Antartica

Reports 9, 89–94.

Cias, P., Petit, J.R., Jouzel, J., Lorius, C., Barkov, N.I., Lipenkov, V.,

Nicolaiev, V., 1992. Evidence from an early Holocene climatic

optimum in the Antarctic deep ice-core record. Climate

Dynamics 6, 169–177.

Cofaigh, C.O., Dowdswell, J.A., 2001. Laminated sediments in

glacimarine environments: diagnostic criteria for their interpre-

tation. Quaternary Science Reviews 20, 1411–1436.

Colizza, E., Finocchiaro, F., Ivaldi, R., Pitta, A., Tolotti, R.,

Brambati, A., 2003. Northern Victoria Land (Western Ross

Sea—Antarctica): inner shelf fine sedimentation. Geo. Res.

Abst. vol. 5. EGS p. 11853.

Crawford, R.M., Hinz, F., Honeywill, C., 1998. Three species of the

diatom genus Corethron Castracane: structure, distribution and

taxonomy. Diatom Research 13 (1), 1–28.

Cunningham, W.L., Leventer, A., 1998. Distribution of diatom

assemblages in surface sediments of the Ross Sea, Antarctica:

relationship to modern oceanographic conditions. Antarctic

Science 10 (2), 134–146.

Cunningham, W.L., Leventer, A., Andrews, J.T., Jennings, A.E.,

Licht, K.J., 1999. Late Pleistocene–Holocene marine conditions

in the Ross Sea, Antarctica: evidence from the diatom record.

The Holocene 9, 129–139.

DeMaster, D.A., 1981. The supply and accumulation of silica in the

marine environment. Geochimica et Cosmochimica Acta 45,

1715–1732.

DeMaster, D.J., Raguenau, O., Nittrouer, C.A., 1996. Preser-

vation efficiencies and accumulation rates for biogenic

silica and organic C, N, and P in high-latitude sediments:

the Ross Sea. Journal of Geophysical Research 101 (C8),

18501–18518.

Denton, G.H., Bockheim, J.G., Wilson, S.C., Stuiver, M., 1989.

Late Wisconsin and early Holocene glacial history, inner Ross

embayment, Antarctica. Quaternary Research 3, 151–182.

Domack, E.W., 1988. Biogenic facies in the Antarctic glaci-

marine environment: basis for a polar glacimarine summary.

Palaeogeography, Palaeoclimatology, Palaeoecology 63,

357–372.

Domack, E.W., Anderson, J.B., 1983. Marine geology of the

George V continental margin: combined results of Deep Freeze

79 and the 1911–1914 Australian expedition. In: Olivier, R.L.,

James, P.R., Jago, J.B. (Eds.), Antarctic Earth Science.

Australian Academy of Science. Cambridge University Press,

Canberra, pp. 402–406.

Domack, E.W., Harris, P., 1998. A new depositional model for

ice shelves, based upon sediment cores from the Ross Sea

and the Mac, Robertson shelf, Antarctica. Annals of

Glaciology 27, 281–285.

Domack, E.W., McClennen, C.E., 1996. Accumulation of glacial

marine sediments in fjords of the Antarctic Peninsula and their

use as late Holocene paleoenvironmental indicators. Antarctic

Research Series, A.G.U. 70, 135–154.

Domack, E.W., Ishman, S.E., Stein, A.B., McClennen, C.E., Jull,

T.A.J., 1995. Late Holocene advance of the Mqller Ice Shelf,

Antarctic Peninsula: sedimentological, geochemical and pale-

ontological evidence. Antarctic Science 7, 159–170.

Domack, E.W., Jacobson, E.A., Shipp, S., Anderson, J.B., 1999.

Late Pleistocene–Holocene retreat of the West Antarctic ice

sheet system in the Ross Sea: part 2. Sedimentological and

stratigraphic signature. GSA Bulletin 111, 1517–1536.

Domack, E.W., Leventer, A., Dunbar, R.B., Taylor, F., Brachfeld,

S., Sjunneskog, C., 2001. Chronology of the Palmer Deep site,

Antarctic Peninsula: a Holocene palaeoenvironmental reference

for the circum-Antarctic. The Holocene 11, 1–9.

Dunbar, R.B., Anderson, J.B., Domack, E.W., Jacobs, S.S., 1985.

Oceanographic influences on sedimentation along the Antarctic

continental shelf. In: Jacobs, S.S. (Ed.), Oceanology of the

Antarctic Continental Shelf. Antarctic Research Series, A.G.U.,

vol. 43, pp. 291–312.

Finocchiaro, F., Melis, R., Tosato, M., 2000. Late Quaternary

environmental events in two cores form southern Joides Basin

(Ross Sea, Antarctica). Terra Antartica Reports 4, 125–130.

Finocchiaro, F., Langone, L., Colizza, E., Busetti, M., Fontolan, G.,

Giglio, F., 2003. Preliminary results on a laminated sediment

core collected from Cape Hallett bay (northern Victoria Land).

Terra Antartica Reports 9, 105–108.

Frignani, M., Giglio, F., Langone, L., Ravaioli, M., Mangini, A.,

1998. Late-Pleistocene–Holocene sedimentary fluxes of organic

carbon and biogenic silica in the northwestern Ross Sea,

Antarctica. Annals of Glaciology 27, 697–703.

Gingele, F.X., Kuhn, G., Maus, B., Melles, M., Schfne, T., 1997.Holocene ice retreat from the Lazarev Sea Shelf, East

Antarctica. Continental Shelf Research 17, 137–163.

Goodridge, C., 1999–2000. Mid Holocene warmth in the Antarctic

Peninsula analog to global warming? Colgate University Journal

of the Sciences 32, 67–84.

Grigorov, I., Pearce, R.B., Kemp, A.E.S., 2002. Southern Ocean

laminated diatom ooze: mat deposits and potential for palaeo-

flux studies, ODP leg 177, Site 1093. Deep Sea Research II 49,

3407–3991.

Harris, P.T., 2000. Ripple cross-laminated sediments on the East

Antarctic Shelf: evidence for episodic bottom water production

during the Holocene? Marine Geology 170, 317–330.

Harris, P.T., O’Brien, P.E., Sedwick, P., Truswell, E.M., 1996. Late

Quaternary history of sedimentation on the McRoberson shelf,

East Antarctica; problems with 14C-dating of marine sediments

cores. Papers and Proceedings of the Royal Society of Tasmania

130, 47–53.

Harris, P.T., Brancolini, G., Armand, L., Busetti, M., Beaman, R.J.,

Giorgetti, G., Presti, M., Trincardi, F., 2001. Continental shelf

drift deposit indicates non-steady state Antarctic bottom water

production in the Holocene. Marine Geology 179, 1–8.

Hodell, D.A., Kanfoush, S.L., Shemesh, A., Crosta, X., Guilder-

son, T.P., 2001. Abrupt cooling of Antarctic surface waters

and sea ice expansion in the South Atlantic sector of the

Southern Ocean at 5000 cal yr B.P. Quaternary Research 56,

191–198.

F. Finocchiaro et al. / Global and Planetary Change 45 (2005) 193–206206

Kellogg, T.B., Osterman, L.E., Stuiver, M., 1979. Late Quaternary

sedimentology and benthic foraminiferal paleoecology of the

Ross Sea, Antarctica. Journal of Foraminiferal Research 9 (4),

322–335.

Langone, L., Frignani, M., Labbrozzi, L., Ravaioli, M., 1998.

Present-day biosiliceous sedimentation in the NW Ross Sea

(Antarctica). Journal of Marine Systems 17, 459–470.

Leventer, A., Dunbar, R., DeMaster, D., 1993. Diatom evidence for

Late Holocene climatic events in Granite Harbor, Antarctica.

Paleoceanography 8, 373–386.

Leventer, A., Domack, E.W., Ishman, S.E, Brachfeld, S., McClen-

nen, C.E., Manley, P., 1996. Productivity cycles of 200–300

years in the Antarctic Peninsula region: understanding linkages

among the sun, atmosphere, oceans, sea ice, and biota. GSA

Bulletin 108, 1626–1644.

Leventer, A., Domack, E.W., Barkoukis, A., McAndrews, B.,

Murray, J., 2002. Laminations from the Palmer Deep: a diatom-

based interpretation. Paleoceanography 17, 1–15.

Licht, K.J., Andrews, J.T., 1997. Diamictons of the east-central

Ross Sea continental shelf: implications for ice sheet extent

during the last glacial maximum (LGM). Antarctica and Global

Change: Interactions and Impacts, Hobart, Tasmania Australia,

13–18 July, Abstract vol. 0378.

Licht, K.J., Jennings, A.E., Andrews, J.T., Williams, K.M., 1996.

Chronology of late Wisconsin ice retreat from the western Ross

Sea, Antarctica. Geology 24, 223–226.

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. GSA

Bulletin 111, 91–103.

Lucchi, R.G., Rebesco, M., Camerlenghi, A., Busetti, M., Tomadin,

L., Villa, G., Persico, D., Morigi, C., Bonci, M.C., Giorgetti, G.,

2002. Mid-late Pleistocene glacimarine sedimentary processes

of a high-latitude, deep-sea sediment drift (Antarctic Peninsula

Pacific margin). Marine Geology 189, 343–370.

Masson, V., Vimeux, F., Jouzel, J., Morgan, V., Delmotte, M., Cias,

P., Hammer, C., Johnsen, S., Lipenkov, V.Y., Mosley-Thomp-

son, E., Petit, J., Steig, E.J., Stievenard, M., Vaikmae, R., 2000.

Holocene climate variability in Antarctica based on 11 ice-core

isotopic records. Quaternary Research 54, 348–358.

Melis, R., Colizza, E., Pizzolato, F., Rosso, A., 2002. Preliminary

study of the calcareous taphocoenoses in Late Quaternary

glacial marine sequences of the Ross Sea (Antarctica). Geobios

24, 207–218.

Nishimura, A., Nakasone, T., Hiramatsu, C., Tanahashi, M., 1998.

Late Quaternary paleoenvironment of the Ross Sea continental

shelf, Antarctica. Annals of Glaciology 27, 275–280.

Orsini, G., Giglio, F., Langone, L., Ravaioli, M., 2003. Palae-

oenvironmental inferences from Core ANTA95-1 (Granite

Harbor, SW Ross Sea—Antarctica). Terra Antartica Reports 8,

133–138.

Presti, M., De Santis, L., Busetti, M., Harris, P.T., 2003. Late

Pleistocene and Holocene sedimentation on the George V

Continental Shelf, East Antarctica. Deep Sea Research II 50

(8/9), 1441–1461.

Pudsey, C.J., Camerlenghi, A., 1998. Glacial–interglacial deposition

on a sediment drift on the Pacific margin of the Antarctic

Peninsula. Antarctic Science 10 (3), 286–308.

Pudsey, C.J., Barker, P.F., Larter, R.D., 1994. Ice sheet retreat from

the Antarctic Peninsula shelf. Continental Shelf Research 14,

1647–1675.

Reineck, H.E., Singh, I.B., 1973. Depositional Sedimentary

Environments. Springer-Verlag, Berlin, 439 pp.

Shipp, S.S., Anderson, J.B., 1994. Late Pleistocene deglaciation of

Ross Sea (Antarctica) inferred from high-resolution seismic

data. Abstracts with Programs-GSA 26, A-364.

Siegert, M.J., 2001. Ice Sheets and Late Quaternary Environmental

Change. John Wiley and Sons, Chichester (UK), 231 pp.

Steig, E.J., Hart, C.P., White, J.W.C., Cunningham, W.L., Davis,

M.D., Saltzman, E.S., 1998. Changes in climate, ocean and ice-

sheet conditions in the Ross embayment, Antarctica, at 6 ka.

Annals of Glaciology 27, 305–310.

Stuiver, M., Denton, G.H., Hughes, T.J., Fastook, J.L., 1981. History

of the marine ice sheet in West Antarctica during the LGM: a

working hypothesis. In: Denton, G.H., Hughes, T.J. (Eds.), The

Last Great Ice Sheets. John Wiley, New York, pp. 319–349.

Taylor, F., McMinn, A., 2001. Evidence from diatoms for Holocene

climatic fluctuation along the East Antarctic margin. The

Holocene 11, 455–466.

Taylor, F., Whitehead, J., Domack, E., 2001. Holocene paleoclimate

change in the Antarctic Peninsula: evidence from the diatom,

sedimentary and geochemical record. Marine Micropaleontol-

ogy 41, 25–43.

U.S.G.S., Map SS 58-60/2 Cape Hallett, 1:250.000 Antarctica

Reconnaisance Series, 1968.

van Beek, P., Reyss, J.-L., Paterne, M., Gersonde, R., van der Loeff,

M., Kuhn, G., 2002. 226Ra in barite: absolute dating of

Holocene Southern Ocean sediments and reconstruction of

sea-surface reservoir ages. Geology 30, 731–734.