RICE UNIVERSITY

Evolution of the Antarctic Peninsula continental margin from Late Eocene to present: Seismic stratigraphic analysis related to the development of the

Antarctic Peninsula Ice Sheet (APIS)

by

Russell Tyler Smith

A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE

Doctor of Philosophy

APPROVED, THESIS COMMITTEE:

JohnA Anderson^fofessor, wTMaurice Ewing Professorship in Oceanography, Chair Earth Science

Alan R. Levander, Professor, Carey Croneis Chair in Geology, Earth Science

Dale S. Sawyer, Professor, Earth Science

Brandon Dugan, Assistant Professor, Earth Science

:hard A. Price III, Assistant ] ^Q&

Richard A. Price III, Assistant Professor, Management

HOUSTON, TEXAS JANUARY 2009

UMI Number: 3362408

INFORMATION TO USERS

The quality of this reproduction is dependent upon the quality of the copy

submitted. Broken or indistinct print, colored or poor quality illustrations

and photographs, print bleed-through, substandard margins, and improper

alignment can adversely affect reproduction.

In the unlikely event that the author did not send a complete manuscript

and there are missing pages, these will be noted. Also, if unauthorized

copyright material had to be removed, a note will indicate the deletion.

®

UMI UMI Microform 3362408

Copyright 2009 by ProQuest LLC All rights reserved. This microform edition is protected against

unauthorized copying under Title 17, United States Code.

ProQuest LLC 789 East Eisenhower Parkway

P.O. Box 1346 Ann Arbor, Ml 48106-1346

ABSTRACT

Evolution of the Antarctic Peninsula continental margin from Late Eocene to present: Seismic stratigraphic analysis related to the development of the

Antarctic Peninsula Ice Sheet (APIS)

: by "

Russell Tyler Smith

This investigation into Antarctic Peninsula Ice Sheet (APIS) development

represents research from the stratigraphic record of three geographic areas: The James

Ross Basin (northwestern Weddell Sea), the Pacific continental margin of the Antarctic

Peninsula, and the Joinville Slope (northwestern Weddell Sea).

The stratigraphic architecture of the James Ross Basin, NW Weddell Sea

continental shelf, shows three major phases of deposition: pre-glacial, ice sheet growth,

and ice sheet dominated. Each stratigraphic unit is characterized based upon seismic

facies and stratigraphic architecture, and the ages are inferred from a seismic stratigraphic

age model. A total of 34 grounding events of the Antarctic Peninsula Ice Sheet (APIS)

are recorded on the continental shelf. The seven oldest glacial unconformities are

believed to pre-date all previously identified unconformities on the peninsula continental

shelf. An expanded section of Late Pliocene/Pleistocene deposits show a minimum of 10

grounding events.

Isopachs of sedimentary sequences on the Antarctic Peninsula Pacific continental

margin show shifting depocenters through time. Chronostratigraphic and seismic depth-

converted data from ODP 178 cores allow the calculation of sediment flux for shelf units

S3-S1 and rise units M6-M1. Sediment flux to the margin increases from the Late

u

Eocene until the Late Pliocene and then decreases slightly from Late Pliocene to present.

Significant increases in sediment flux coincide with early development of the APIS and

during the early Pliocene warming period (Barker and Camerlenghi, 2002). Minimum

glacial denudation rates for the Antarctic Peninsula are in the range of 0.06 to 0.13 mm

yr"1.

The Joinville Slope sediment wedge located in the northwestern Weddell Sea

shows seismic stratigraphic evidence of mixed turbidite/contourite/hemipelagic

deposition. A prominent seafloor unconformity and the exposed and eroded basement of

the adjacent continental shelf indicate erosion by grounded ice during the Plio-

Pleistocene. SHALDRIL recovered core at three drill sites, 12A, 5C, and 6D, and

sampled sediments from the upper Oligocene, middle Miocene, and lower and upper

Pliocene which are constrained by diatom and calcareous nannofossil assemblages. The

sediment wedge shows no apparent hiatuses or large unconformities from Late Oligocene

to the Lower Pliocene. Regional sedimentation rates show continuous sedimentation

throughout the Late Paleogene and Neogene.

i i i

ACKNOWLEDGMENTS

There is a saying which teaches, "Give a man a fish; you have fed him for today.

Teach a man to fish; and you have fed him for a lifetime. (Author Unknown)" This

written work is the culmination of a process through which I have been taught to do

meaningful science. The process was only made possible through the steady guidance

and mentoring of my advisor Dr. John Anderson. I am extremely grateful and count

myself lucky to have studied and worked both in the field and the office with such a great

teacher of science.

Along with John, I was guided by the other members of my thesis committee: Dr.

Alan Levander, Dr. Dale Sawyer, and Dr. Brandon Dugan. Their help and enthusiasm

for my research along with constructive criticism has refined and improved this research

greatly. I would also like to thank Dr. Richard Price III for his participation.

During my graduate work, I had the opportunity to participate in five different

field work projects: PICASO with Dr. Andre Droxler, BOLIVAR, with Dr. Alan

Levander and Dr. Colin Zelt, SHALDRILI and II and the Antarctic Peninsula Fjords

cruise with Dr. John Anderson. I am grateful for the funding of these projects by the

National Science Foundation.

Last, but certainly not least, I am eternally grateful for the loving support of my

family who believed in me. My wife and I were married during the writing of this thesis

and she has stood by me throughout my graduate work. I thank her for encouraging and

strengthening me. I would also like to thank my parents and siblings for their

encouragement and for teaching me to always love learning.

IV

TABLE OF CONTENTS

ABSTRACT ii ACKNOWLEDGMENTS . iv INTRODUCTION 1

OVERVIEW 1 LAYOUT OF THE THESIS . 2

CHAPTER 1 - Ice sheet evolution in James Ross Basin, Weddell Sea margin of the Antarctic Peninsula: the seismic stratigraphic record 3

ABSTRACT 3 INTRODUCTION 4

Seismic Analysis of Ice Sheet Evolution 7 REGIONAL SETTING 8 DATA AND METHODS 11 RESULTS 14 RESULTS.. . 15

Seismic Units and Facies 17 Seismic Stratigraphy 21 data from the cores yielded Pleistocene and younger diatom assemblages with some reworked older fossils, however, the base of unit SI was not sampled and could be older than Pleistocene in age 27 Age Model 27

DISCUSSION 29 Pre-Glacial Phase (S5) 29 Initial Ice Sheet Development (Unit S4). 30 Northward Expansion of Ice Sheet (Unit S3) 31 Ice Sheet Dominated Phase (S2-S1) 32 Number of Ice Groundings 33

CONCLUSIONS 35 ACKNOWLEDGEMENTS 37 REFERENCES CITED 38

CHAPTER 2 - Sediment flux to the Antarctic Peninsula margin based on first order stratigraphic sequences of the continental shelf and rise 45

ABSTRACT 45 INTRODUCTION 46 SETTING. 50 DATA 54 METHODS 55

Sediment Yield 55 Denudation Calculation ..56

RESULTS 59 DISCUSSION 63

Denudation Rates 67

v

CONCLUSIONS 70 REFERENCES CITED 71

CHAPTER 3 - Seismic stratigraphic evidence for mixed contourite/turbidite system in the NW Weddell Sea from SHALDRIL cruises 78

ABSTRACT 78 INTRODUCTION 79 BACKGROUND 83 BACKGROUND 84 DATA AND METHODS 87 RESULTS 88

Stratigraphic Architecture and Seismic Facies 88 Ages of Drill Core 94 Sediment Description 99

DISCUSSION 101 Contour Current Deposition 101 Turbidite Current Deposition 103 Hemipelagic Sedimentation 105 Sedimentation Rates 105 Development of the Ice Sheet 107

CONCLUSIONS 109 REFERENCES CITED 111

VI

LIST OF FIGURES

CHAPTER 1

Figure 1. Geographic and tectonic map 6 Figure 2. Peninsula cross section 10 Figure 3. Data location map 12 Figure 4a. Seismic line showing stratigraphic architecture 14 Figure 4b. Seismic facies 14 Figure 5. Isopach contour maps 16 Figure 6. Strike section of unit S4 on seismic line 93-01-D 18 Figure 7. Regional strike line NBP0201-2 20 Figure 8.,Seismic line NBP0602A-02 showing site 3C drill site 22 Figure 9. Diatom biostratigraphy of site 3C. 23 Figure 10. Expanded section of unit SI in strike view 26 Figure 11. Age model interpretation 28

CHAPTER 2

Figure 1. Peninsula cross section 48 Figure 2. Data location map 49 Figure 3. Series of dip lines from north to south.... 51 Figure 4. Isopach contour maps 58 Figure 5a. Shelf to shelf comparison of sediment volume 61 Figure 5b., Shelf to rise comparison of sediment volume 61 Figure 6. Total sediment flux graph... 64

CHAPTER 3

Figure 1. Regional map 81 Figure 2. 3-D seismic rendering of Joinville Slope 82 Figure 3. SHALDRIL core ages summary. ,. 83 Figure 4. Seismic lineNBP0602A-10 showing drill site locations... 89 Figure 5a. Bubble pulse and basement facies 90 Figure 5b. Sediment drape over basement high 90 Figure 6a. Wavy seismic facies 92 Figure 6b. Channel cut and fill 92 Figure 6c. Slump facies ...92 Figure 7. Site 5D summary 95 Figure 8. Site 6C summary 97 Figure 9. Seismic line NBP0602A-10 basement/sediment seafloor 108

vii

LIST OF TABLES

CHAPTER 1

Table 1. Number of ice grounding events 34

CHAPTER2

Table 1. Summary of drill core chronostratigraphy 62 Table 2a. Denudation rates for minimum erosive area .68

Table 2b. Denudation rates for maximum erosive area .68

CHAPTER 3

Table 1. Average sedimentation rates.. 106 Table 2. Maximum sedimentation rates 106

INTRODUCTION

OVERVIEW

The dangers of rapid sea level rise due to modern warming trends and the

destabilization of ice shelves has turned many eyes to the Antarctic with questions

concerning the stability of the West Antarctic Ice Sheet and Antarctic Peninsula Ice Cap.

Answers to current questions may find inspiration by reversing the famous quote by

James Hutton to be, "the past is the key to the present". Understanding the history of ice

development on Antarctica will help to better constrain our models on the future behavior

of the ice sheets. This research attempts to further our understanding of the timing and

extent of ice sheet development on the Antarctic Peninsula.

My approach to understanding ice sheet development on the peninsula has been to

study the accumulation of sedimentary deposits on the continental margins. These

sediments contain a record of the waxing and waning of the ice sheet through time, and

can be used to reconstruct ice volume history. Unlike the proxy deep marine isotope

record (Zachos et al., 2001; Raymo et al., 2006), glacial sedimentary sequences are the

physical remnant of ice sheet grounding on the continental shelf. Using seismic

stratigraphic principles we are able to interpret stratigraphic architecture and facies

changes within the sedimentary column, and therefore, produce regional maps of ice

sheet grounding (Cooper et al., 1991; Bart and Anderson, 1995,1996; Anderson, 1999).

One of the next steps in understanding the evolution of the Antarctic Peninsula

Ice Sheet (APIS) is to quantify the amount of sediment delivered to the margin through

time. Modern (100 yr) glacial sediment fluxes vary wildly geographically (Hallet et al.,

1996). This work presents the first attempt to constrain minimum glacial sediment flux

1

to the Pacific margin of the Antarctic Peninsula during the Neogene (10-15 Ma).

Combined with the current knowledge of the tectonic and climate history of the

peninsula, future ice sheet models can more accurately approximate ice sheet dynamics.

The Antarctic presents unique obstacles to scientific research. Extreme cold, ice

covered waters, and rough seas are common challenges during data collection. The high

costs associated with necessary transport by icebreaker vessels along with a limited

seasonal research window each year, contribute to a lower density of data available from

Antarctic continental margins. Thus, there is great value in trying to get the most from

existing data sets even after the initial analysis has been published. Compiling data

acquired through various organizations and countries was critical to the success of this

work.

LAYOUT OF THE THESIS

The thesis is organized into three major chapters which represent stand alone

projects prepared individually for publication. The first chapter focuses on ice sheet

evolution as recorded by the stratigraphic record of the James Ross Basin, located in the

northwestern Weddell Sea. An age model for the basin is presented along with an

analysis of the timing and frequency of ice sheet grounding on the continental shelf. The

second chapter attempts to quantify sediment flux to the Pacific margin of the Antarctic

Peninsula throughout the development of the APIS. This study combines sedimentation

histories from both the continental shelf and rise, and shows shifting regional depocenters

over the last 36 Ma. The third chapter presents a seismic stratigraphic analysis integrated

with preliminary core data for the Joinville Slope. The data was acquired as part of the

SHALDRIL project which was designed as a new approach to Antarctic margin drilling.

2

CHAPTER 1 - Ice sheet evolution in James Ross Basin, Weddell Sea margin of the Antarctic Peninsula: the seismic stratigraphic record

R. Tyler Smith Rice University Earth Science Dept. 6100 Main St. MS 126 Houston, TX 77005

John B. Anderson Rice University Earth Science Dept. 6100 Main St. MS 126 Houston, TX 77005

ABSTRACT

The stratigraphic architecture of the James Ross Basin, NW Weddell Sea

continental shelf, shows three major phases of deposition: pre-glacial, ice sheet growth,

and ice sheet dominated. The pre-glacial phase shows stratigraphic sequences that are

similar to low latitude margins. During the ice sheet growth phase there is evidence for

initial ice sheet grounding in the southern part of the basin which migrated in steps

northward. Sediment supply was also high during the ice sheet growth phase. During the

ice sheet-dominated phase, both north and south parts of the basin inherited all the classic

features indicative of glacial strata; large glacial troughs 100's of meters deep and 10's of

kilometers wide, truncation of sequences overlain by chaotic seismic facies and topped

with coherent laminated reflections, and an oversteepened continental slope.

The James Ross Basin strata are divided into 5 units (S5-S1). Each unit is

characterized based upon seismic facies and stratigraphic architecture, and the age of

each unit is inferred from a seismic stratigraphic age model. End member strata (S5 and

SI) were drilled and dated during SHALDRIL and help constrain this age model. A total

of 34 grounding events of the Antarctic Peninsula Ice Sheet (APIS) are recorded on the

continental shelf. This is similar to the 31 grounding events imaged on the Pacific side of

3

the peninsula (Bart and Anderson, 1995). Ice grounding on the continental shelf is shown

to have occurred as early as Late Middle Miocene/Early Late Miocene in the southern

part of the basin. The ice sheet advanced northward to encompass the northern peninsula

by Late Miocene time. The seven oldest glacial unconformities are believed to pre-date

all previously identified unconformities on the peninsula continental shelf. An expanded

section of Late Pliocene/Pleistocene deposits show a minimum of 10 grounding events,

which more closely matches the deep sea oxygen isotope record than previous seismic

studies of Antarctic margins have shown.

INTRODUCTION

Proxy records of ice sheet evolution have yielded strong evidence for Antarctic

Ice Sheet evolution culminating in an East Antarctic Ice Sheet as large as present by

Oligocene time (see for a review Anderson, 1999; Zachos and Kump, 2005, Lear et al.,

2008), but the record of West Antarctic Ice Sheet (WAIS) and Antarctic Peninsula Ice

Sheet (APIS) evolution has remained more vague. Relative to the other portions of the

Antarctic cryosphere, the APIS is thought to have the most dynamic ice sheet history and

be the most susceptible to climate change due to its relatively small size and northerly

location. Understanding the timing and frequency of paleo-ice sheet expansions and

retreats through time would allow us to examine the forces that regulate APIS stability

(ie. climate change, sea-level rise and fall). This, in turn, provides constraints for ice

sheet modeling.

A major problem is that the oxygen isotope and eustatic records do not allow us to

determine how ice sheets were actually distributed on Antarctica. One approach to this

4

problem is that of using seismic stratigraphic evidence for ice sheets having grounded on

the continental shelf (see Anderson, 1999 for review). This approach allows not only

geographic constraints on ice sheet evolution, but also tells us when the ice sheet was

larger than present, if age constraints are available, thus, placing constraints on the

magnitude of ice sheet expansions. The basis for seismic stratigraphic studies aimed at

resolving Antarctica's glacial history is that the shift in depositional regime from

fluvial/deltaic to glacial is manifest in the stratigraphic architecture and seismic facies of

the continental margin (Cooper et al., 1991; Bartek et al., 1991).

Several studies have been conducted on the Pacific margin of the peninsula in

which the motivation was to characterize the history and development of the APIS

(Larter and Barker, 1991; Larter and Cunningham, 1993; Bart and Anderson, 1995;

Rebesco et al., 1997; McGinnis et al., 1997; Barker and Camerlenghi, 2002; Rebesco et

al., 2006). The resulting seismic stratigraphic evidence suggests that ice initially

grounded on the continental shelf anywhere between the Late Pliocene to as early as early

Late Miocene time. One of the main objectives of ODP Leg 178 sites 1097,1100,1102,

and 1103 conducted on the continental shelf of the Pacific margin (Fig. 1), was to sample

several ice sheet expansion and retreat cycles and attempt to sample and date the oldest

glacial strata. Unfortunately, these sites never reached the base of the glacial section and

core recovery was poor. The findings from these drilling efforts indicate that ice

grounded on the continental shelf by at least Late Miocene time and possibly as early as

the Middle Miocene (Barker and Camerlenghi, 2002). Bart and Anderson (1995)

concluded that ice sheets had grounded at least 31 times on the Antarctic Peninsula shelf

5

Figure 1.

Geographic and tectonic map for the Antarctic Peninsula with lithotectonic units (Modified from Elliot, 1988).

since the Late Miocene. Older glacial sequences were obscured by the water bottom

multiple or were missing due to erosion, as tectonic uplift of the Pacific margin resulted

in large unconformities during critical time periods when ice sheets took hold as the

dominant depositional mechanism (Larter and Barker, 1991; Bart and Anderson, 1995).

The northwestern Weddell Sea is better suited to study the transition to an ice-dominated

margin because it has remained in a passive margin/back arc basin tectonic setting

throughout Antarctic Peninsula Ice Sheet evolution (Fig. 1).

Seismic Analysis of Ice Sheet Evolution

Larter and Barker (1989) and Cooper et al. (1991) were among the first to attempt

to distinguish between glacial and non-glacial sequences using seismic stratigraphy.

Because their data sets lacked sufficient resolution for seismic facies analysis and

recognition of discrete glacial unconformities, their models were based on overall

changes in stratigraphic architecture revealed by dip sections. Progradational stacking

patterns were assumed to represent glacial episodes when large quantities of sediment

were transported across the shelf and deposited at the shelf break and slope.

Aggradational stacking patterns, in turn, were assumed to represent inter-glacial periods

when continental derived sediment input was low and sedimentation was dominated by

glacimarine processes. Bart and Anderson (1995) took seismic interpretation of glacial

sequences a step farther and identified unconformities in strike-oriented seismic lines

which represent glacial troughs. They argued that glacial and inter-glacial deposits have

distinct seismic facies, represented by massive chaotic units and thin laminated units,

respectively.

7

Studies of the Ross Sea and Antarctic Peninsula shelf have shown that stacked till

sheets bounded by landward sloping glacial unconformities provide conclusive evidence

for sub-ice sheet deposition (Alonso et al., 1992; Bart and Anderson, 1995; De Santis et

al., 1999). Till sheets are characterized by sharp, strongly reflective upper and lower

erosional boundaries, and chaotic internal reflector configurations. Additionally,

evidence of ice streams, in the form of large troughs, usually accompanies till sheet

deposition. Paleo-ice streams produce unconformities with trough geometries that are

later filled by glacimarine and younger sub-ice deposits. While the ice sheet is grounded

on the continental shelf, paleo-ice streams erode troughs 100's of meters deep and 10's of

kilometers wide. Sea floor maps derived from swath bathymetry surveys show modern

glacial troughs and associated geomorphic features formed by ice streams during the Last

Glacial Maximum (O'Cofaigh et al., 2002; Canals et al., 2000; Canals et al., 2002; Heroy

and Anderson, 2005; Wellner et al., 2006).

Criteria for the seismic interpretation of grounded ice sheets on the continental

shelf include: large scale cut and fill pattern, lenticular geometries, truncated reflectors

along boundary surfaces, and large erosional troughs (>30 km width, >50 m deep).

Strike-oriented seismic lines are especially helpful for distiguishing glacial versus inter-

glacial strata because they best image glacial unconformities and ice stream troughs.

REGIONAL SETTING

The James Ross Basin (JRB) is the most northern sub-basin of the greater Larsen

Basin, located on the Antarctic Peninsula continental shelf in the Weddell Sea. Named

for adjacent James Ross Island, the JRB is a back-arc basin which formed in response to

8

subduction of the Phoenix and Aluk Plates on the Pacific side of the peninsula (Fig. 1).

The basin began subsiding during the Late Mesozoic (Elliot, 1988) and contains between

4.8 km (Ineson etal., 1986) to 6.4 km (del Valle et al., 1983) of sedimentary strata.

Compilations of articles on the exposed outcrops of James Ross Island and Seymour

Island (Feldmann and Woodburne, 1988; Francis et al., 2006) provide the most complete

descriptions of the stratigraphic succession available. Unfortunately, the stratigraphic

section exposed in outcrop is only complete through the Eocene/Early Oligocene (Ineson

etal., 1986; Elliot, 1988;Huber, 1988; Sadler, 1988). The youngest non-glacial unit

found on Seymour Island is the La Meseta Formation, which has been characterized as an

incised valley deposit (Marenssi et al., 2002). A large hiatus exists between the non-

glacial Eocene/Oligocene fluvial/deltaic deposits of the La Meseta Formation and the

Late Pliocene/Early Pleistocene sub-glacial diamicton overlying it (Gazdzicki et al.,

2004). A recent study by Ivany et al. (2006) led to the discovery of a small outcrop of

Oligocene age diamicton, which they interpreted as tillite. The outcrop is localized,

thereby making it difficult to substantiate whether it is evidence of an ice cap or localized

mountain glaciation. Therefore, the search for evidence for the early evolution of the

Antarctic Peninsula Ice Sheet (APIS) has shifted offshore.

Initial seismic studies of the JRB showed an offlapping succession of

progressively younger strata distinguished by changes in seismic facies and architecture

(Anderson et al., 1992; Sloan et al., 1995). The margin shows similar stratigraphic

architecture to the Pacific side of the peninsula, except the JRB section is more

progradational (Fig. 2). Therefore, the nomenclature assigned follows that used on the

Pacific margin, namely S5-S1, with S5 being the oldest strata and SI the youngest

9

We

dd

ell

Sea

Sid

e

•"1K

V

a~

*%X-

o

Cro

ss se

ctio

n th

roug

h th

e A

ntar

ctic

Pen

insu

la s

how

ing

sim

ilarit

y in

st

ratig

raph

ic a

rchi

tect

ure

of th

e tw

o m

argi

ns o

n ei

ther

sid

e of

the

peni

nsul

a.

The

Pac

ific

mar

gin

deve

lope

d as

a fo

rear

c ba

sin

with

sub

duct

ion

occu

ring

just

bey

ond

the

cont

inen

tal r

ise,

whi

le th

e W

edde

ll Se

a m

argi

n is

a b

ack

arc

basi

n. R

idge

-tren

ch c

ollis

ion

occu

rred

alo

ng th

is p

ortio

n of

the

Paci

fic

mar

gin

durin

g th

e de

posi

tion

of u

nit S

4. B

athy

met

ry a

nd to

pogr

aphi

c da

ta

sour

ced

and

back

grou

nd im

age

crea

ted

by th

e M

arin

e G

eosc

ienc

e D

ata

Syst

em (M

GD

S)(C

arbo

tte e

t al.,

200

7).

(Anderson et al., 1992; later modified in Anderson, 1999). The overall bathymetry of the

shelf is typical of other portions of the Antarctic margin as the shelf deepens landward

(foredeepened) and is unusually deep (500-1000 m).

SHALDRIL cruises I and II were intended to drill short holes through the

overlying till to core the offlapping strata of the JRB. The goal was to sample over a

large time span and to better constrain the onset of glacial conditions. Unfortunately, the

majority of drill sites within the JRB were abandoned due to rapidly flowing multi-year

ice flows, although they did sample Pleistocene and possibly as old as Late Pliocene

shallow strata. However, SHALDRIL II site 3C, located on the northern edge of the

basin (Fig. 3), sampled strata spanning the Latest Eocene/Earliest Oligocene (Anderson et

al., 2007). Other sites located northeast of the JRB sampled Oligocene, Middle Miocene,

Early and Late Pliocene and Pleistocene strata. An attempt to connect these northeasterly

sites to the JRB seismic data set during an NBP 0703 cruise was unsuccessful due to

severe sea ice conditions.

DATA AND METHODS

Two previous seismic grids from the JRB were used in this study. The first was

collected on board the R/V Polar Duke in 1991 and is located in the northern part of the

basin, a description of the seismic data can be found in Anderson et al. (1992). These

data were acquired using a 1638.7 cm3 water gun fired every 6 seconds and a 10 m,

single channel streamer. Ice conditions prevented the use of a long streamer. The second

grid is located to the south of the 1991 data set and was collected on board the RV/IB

Nathaniel B. Palmer in 1993 (Sloan et al., 1995). Additional seismic data was collected

11

eo-O'o-w ss 'ww

65°0'G"S-

Legend /SH^LDRILcor isW

— ^ew^OOS^OOeselsmic

Published seismic (1991.1993)

New 2002 seismic

Bcthymetry (5 )0 m contours)

0 25 50 ' 100 Kilometers i i i\ i i i j i i

•es'wrs

60o0"0"W 55o0'0"W

Figure 3

Seismic and drill core data location map.

12

by Rice University on board the RV/IB Nathaniel B. Palmer during the 2002,2005,2006

and 2007 field seasons (Fig. 3). The Palmer data sets were collected using one or two

210 cm3 or 50 cm3 generator-injector air guns as a source. The air guns were fired at an

interval between 5-7 seconds. A single channel ITI streamer was towed at a distance of

100-300 meters behind the ship depending on ice conditions. Seismic acquisition during

the SHALDRIL cruises (2005-2006 and 2007) was particularly hampered by heavy sea

ice. Ice conditions precluded multi-channel seismic acquisition. Minimal processing was

used on the data. A bandpass filter (50-500 Hz) and overall gain function was applied to

the digital data afterwards to reduce noise and increase signal strength of the deeper

strata. Continuous 3.5 kHz seismic (Bathy 2000), 12 kHz multibeam swath bathymetry

(Simrad EM120), gravity, and magnetics data was also collected during the

aforementioned cruises. Processing of the data was done onboard during the respective

cruises.

Unit thickness maps are measured in two way travel time. Travel time values

were preferred to depth because of the limited velocity information available in the area.

Sparse sonobuoy data is available from around the peninsula, however, resolution is not

sufficient to characterize velocity changes in the soft sedimentary section. A strong water

bottom multiple obscures the stratigraphic section below it. Thickness values of the

oldest stratigraphic units are measured to the water bottom multiple. Isopach thicknesses

were interpolated using an inverse distance weighted algorithm and clipped to the data

area. Minor modifications to the contour lines were made to better illustrate inferred

stratigraphic terminations.

13

a.

, . _ * v t "i . H I '

i PM §£§51

— v Si - \

2 Km

Examples Descriptions

Highly chaotic, aggradational unit which drapes the shelf. Rare continuous reflections within chaotically massive units outline lenticular shaped bodies. Generally exists as a thin unit overlying an angular unconformity at the top of units S2-S5. An expanded section of SI imaged by line NBP0201-2 shows seismic fades similar to S2.

Stacked, well defined lenticular shaped bodies composed of chaotic facies separated by thin draping layers of high amplitude, laminated reflections. Abundant truncation at the tops of individual packages and low lateral continuity is characteristic. Average width of a package is ~20 km.

Similar seismic facies to S4, however this unit is stratigraphically higher in the section. In the northern portion of the basin, S3 rests directly on an angular unconformity at the top of S5.

Stacked sequences of massive chaotic to very discontinuous reflections draped by thin laminated, continuous and high amplitude reflections. Defined at its base by an angular unconformity at the top of S5. This unit exists only in the southern (deeper) portion of the basin.

Highly laminated unit of medium to high amplitude reflections with very regular spacing. Reflections dip at angles between 3-5 degrees and are laterally continous.

Figure 4 a, b

a. Seismic line NBP93-22 shows an offlapping succession of stratigraphic units which is representative of the stratigraphic architecture of the margin.

b. Representative examples and short descriptions of seismic facies from each oftheJRB units SI-S5.

14

RESULTS

The stratigraphic section of the James Ross Basin is divided into 5 separate

seismic units S1-S5 (Fig. 4a). These units were interpreted based on their seismic-

stratigraphic expression. Unit boundaries are primarily defined using strike oriented

seismic lines to identify regional unconformities. The base of S5 is the surface between

chaotic reflections filled with detractions, interpreted as basement, and laminated

reflections which are parallel and continuous. The bottom of S4 is represented by an

onlap surface coinciding with a change in slope which is expressed throughout the basin.

Bounding surfaces between units S4 and S3, and units S3 and S2 are both defined by a

highly reflective regionally extensive unconformity which truncates the topset beds of the

preceding units. The base of unit SI is characterized by a change in stratigraphic

architecture from progradational to aggradational strata and is associated with a regional

unconformity.

Figure 4b summarizes the seismic facies character of each seismic unit. Anderson

et al. (1992) and Sloan et al. (1995) provide more detailed descriptions of these units.

Figure 5 shows in map view the thicknesses of stratigraphic units S1-S5 (in two way

time). The distribution of units, beginning with the oldest unit (S5) to the youngest (SI),

show progressive infilling of the JRB, mainly from continental sources (Fig. 5). The

strata dip basinward with angles decreasing away from the inner edge of the basin. The

oldest strata (S5) rim the edge of the basin close to the continent. Younger strata (S4 -

S2) prograded and filled the accommodation as it migrated toward the center of the basin.

The most recent unit (SI) was deposited on top of the offlapping units as an

aggradational drape (Fig. 4a).

15

6oro

:o°w

ss

wcw

ar

oitr

w

65-0

'0°S

5yo:

o"w

ws

ws

ero'

cs

ww

w

ssw

w

tero

'cs

60

WW

5

5W

W

6ffv

.(rw

55

*0:0

°W

cr\

ero'

o'S'

6(ro

:o'w

sy

oio'

w

tesw

s 6

sws

6(ro

:o"w

ss

-ff.tr

w

tero

'O's

6yo'0

"S

tero

'o's

Fig

ure

5

Con

tour

map

s of

two-

way

tim

e th

ickn

esse

s fo

r un

its S

5-S1

. A

rrow

s re

pres

ent d

irect

ion

of tr

ansp

ort d

urin

g de

posi

tion.

Seismic Units and Fades

The seismic character of the limited part of unit S5 that is above the water bottom

multiple is highly reflective and highly laminated (Fig. 4b). It is composed of high-

amplitude and moderately continuous reflections. The oldest reflections onlap basement

along the inner rim of the basin. S5 reflections dip basinward at angles up to 5° with

younger strata flattening to ~3° (angle measured in twt). Some areas, primarily to the

north, show evidence of slope instability and slumping. The youngest strata of unit S5

display normal and strike slip faulting within a fault zone which rims the basin (fault

zone not shown in Fig. 4, however a more detailed description can be found in Anderson

et al., 1992; Sloan et al., 1995).

The S4 seismic unit consists of highly progradational strata characterized by

chaotic seismic facies alternating with thin high amplitude continuous reflections (Figs.

4, 6). Each progradational package consists of a massive chaotic deposit topped by a thin

laminated deposit. These packages taper landward, onlapping an angular unconformity at

the top of unit S5. Overall, S4 strata dip at lower angles than S5 strata (angles measured

in twt) and the topset beds of the slope sequences of S4 are better preserved. In the

deepest portions of the basin, large portions of S4 are obscured by the water bottom

multiple. Unit S4 does not extend into the northern portion of the basin.

Unit S3 displays similar seismic facies to unit S4, being composed of stacked

prograding packages made up of a thick chaotic unit topped by thin, high amplitude

reflections (Fig. 4). Distinct offlap breaks clearly delineate the paleo-shelf break as it

prograded during S3 deposition. Unit S3 is present throughout the northern and southern

parts of the basin. In the north, S3 sits above an angular unconformity at the top of S5.

17

Figu

re 6

Seis

mic

line

93-

01-D

pro

vide

s a

repr

esen

tativ

e st

rike

sect

ion

for

the

sout

hern

por

tion

of th

e JR

B.

The

sec

tion

exhi

bits

br

oad-

scal

e cu

t and

fill

stru

ctur

es w

ith le

ntic

ular

geo

met

ries

and

alte

rnat

ing

chao

tic a

nd la

min

ated

sei

smic

fade

s. G

laci

al

unco

nfor

miti

es a

re h

ighl

ight

ed w

ith a

rrow

s de

sign

atin

g on

lap

of g

laci

mar

ine

stra

ta w

ithin

pal

eo-tr

ough

s.

In the south, S3 rests on top of S4 and its base rests on a regionally extensive

unconformity represented by a high amplitude continuous reflection. Parts of S3 are

obscured by the water bottom multiple, though not to the extent of unit S4.

Unit S2 is defined at its base by an unconformity at the top of S3. Truncation of

underlying topsets of S3 and high amplitude reflections mark this boundary. S2 is

composed of stacked lenticular bodies, primarily composed of a massive chaotic fades

and outlined by high amplitude reflections (Fig. 4b). The tops of each package are

truncated. Orientation of the deposits suggests erosional troughs that were later filled by

chaotic material and eroded at the top. Trough sizes are anywhere between 30 and 100

km in width and roughly 50 to 100 ms twt (-40-80 m) in depth. In dip section, troughs

are not visible and the unit is largely chaotic with few coherent reflections of limited

lateral extent. Overall, S2 shows a strong cut and fill pattern.

Unit S1 drapes the entire shelf as a thin aggradational unit. The uppermost 30 ms

twt is heavily overprinted by a resonant bubble pulse, caused by the repeated collapse and

expansion of air after being released from the air gun source. Compared to S2, SI is

much more condensed, with few (1 to 2) high amplitude reflections separating massive

chaotic facies (Fig. 4b). The unit is characterized by broad cut and fill geometry,

composed of two to three stacked units. An expanded section of SI with small regional

extent was imaged by line NBP0201-2, which shows a very similar seismic facies to unit

S2 (Fig. 7). Within the expanded section, stacked lenticular packages of chaotic facies

are outlined by high amplitude reflections. The top of SI (sea bed) shows high degrees

of relief (up to > 100 meters) which represents deeply eroded troughs greater than 30 km

in width. Across the shelf and within these troughs, swath bathymetry mapping has

19

Line

NB

P02

01-2

*

«

i t

* !

. i

S

t"

II

t 'I

'H

H

1 -

j!

?.-

Js

?>

V?

i«H

Mi

MIS

'

Line

NB

P02

01-2

ri«il,

ffii«

*iH

r i

' * (

( i*

' *

«S

' **

**

r

' £

i!

! i

u *

£

t i

li

s I

t ]

Hi;-

:•

>',

;tf

;«

Figu

re 7

Seis

mic

line

NB

P020

1-2

prov

ides

a re

pres

enta

tive

strik

e se

ctio

n of

the

inne

r con

tinen

tal s

helf

of t

he JR

B.

Uni

t S3

prog

rade

s fro

m no

rth to

sou

th, w

here

as u

nit S

4 pr

ogra

des f

rom

sout

h to

nor

th.

Uni

t S2

and

SI e

xhib

it st

rong

evi

denc

e of

gr

ound

ed ic

e.

o

imaged mega-scale glacial lineations along with grounding zone wedges (Heroy and

Anderson, 2005; Wellner etal., 2006)

Seismic Stratigraphy

UnitS5

No offlap breaks were imaged within S5 (visible only above the multiple). The

trend of the unit parallels the basin (Fig. 5) and the strata dip at relatively high angles (3-

5° in twt). There is also some evidence of minor slope failure and slumping. These

observations suggest that they are ramp or slope deposits. A highly faulted section rims

the inner portion of the basin and is confined to the uppermost S5 strata. Faulting

therefore appears to pre-date S4 and younger strata.

Several onlap surfaces within S5 are interpreted as representing eustatic lowstands

(Fig. 8). The recent SHALDRIL cruises targeted these onlap surfaces as drilling

objectives because they were assumed to contain condensed sections and, therefore,

offered a greater probability of sampling microfossils for biostratigraphic analysis. Drill

core NBP0602A-3C sampled the oldest onlap surface above the basement unconformity

(Fig. 8). The site sampled very dark grey muddy fine sand that is poorly sorted. The

sediments show no signs of bioturbation and at the base of the core there are faint

laminations. Within the cored interval there are very few dropstones, however, there are

numerous calcareous fossils (gastropods and bivalves). The sediments were dated as

Latest Eocene and/or Early Oligocene based on diatom (Fig. 9) and calcareous

microfossil assemblages (Anderson et al., 2007).

21

22

Hole NBP0602A-3C

°1

2 -

4 -

5 -

6 "

• . 7 -

S „:••

a 10

<=> 11

13

14

15

16

17

18

Cor

e

I l K

2R.

3R,

4Ra

5R.

6R,

•7R.

8R.

Rec

over

y

— ~

Time-Rock Unit

Upper Pleistocene-

Holocene

Upper Eocene-

lowermost Ohgoceae

Rec<

Diatom Zone

T. lentigiriosa-F. ker-giielensis

"b"

Rhizosolenia oligocaenica

Rylcmdsia inaequiradiata

Jo wery

Age (Ma)

0.14

32 to 37

? ' • "

Figured

Diatom biostratigraphy for SHALDRIL drill core NBP0602A-3C located on the northern rim of JRB. Diatom work by Dr. Steven Bohaty. (SHALDRIL II Cruise Report, 2006)

UnUS4

Unit S4 exists only in the southern portion of the basin (Fig. 5). The base of S4 is

an angular unconformity at the top of unit S5. Clinoforms within S4 indicate

progradation to the east and northeast, indicating that accommodation existed within the

basin to the east of Seymour Island in the northern part of the study area (Fig. 5).

Seismic line 93-01-D, collected in the southern portion of the basin, provides a

strike-oriented section for nearly 80 km (Fig. 6). In strike section, S4 exhibits broad-

scale cut and fill structure composed of troughs with lenticular geometries along with

alternating chaotic and laminated seismic fades (Fig. 6). The troughs have an average

width of 30 km and an average depth of 50-100 ms.

Unit S3

Unit S3 is a highly progradational unit that prograded as two distinct lobes, one

from the southwest and another from the northwest (Fig. 5), thus indicating two different

sediment sources. The boundary between the northern and southern sources is in line

with the topographic highs of James Ross, Seymour, and Snow Hill islands. The base of

S3 is marked by an angular unconformity at the top of S5 in the north, and a regional

unconformity at the top of S4 in the south. S3 exhibits similar truncation surfaces and

seismic facies to S4. Identification of broad cut and fill structures is difficult, however,

because of limited areas where profiles are oriented along strike. Overall thickness of S3

is greatest in the north and thins to the south.

24

UnitS2

During the deposition of S2 the shelf margin prograded close to its current

position. The amount of progradation during S2 deposition varied considerably, relative

to the older units, from north to south. The thickness and extent of the chaotic fades

within S2 is greater than in units S4 and S3 units. Sediment accumulation was focused

toward the greatest accommodation, within a depression between lobes of unit S3 (Fig.

5). On the inner shelf, S2 deposits show high amounts of truncation or are completely

removed. The preferential erosion of the inner shelf, resulting from ice sheets scouring

the sediments of the inner shelf and re-depositing them on the outer shelf, produced

landward sloping unconformities and sequences that thicken basinward.

Unit SI

Unit S1 sits atop an angular unconformity that deeply truncates underlying units.

Unlike older units, which are preserved as offlapping packages resting one on top the

other, S1 drapes the entire shelf. The unit is strongly aggradational, with relatively little

progradation of the continental shelf break compared to older units. Similar to S2, there

is an overall thickening of SI in the area of accommodation created during the deposition

of S3. Within the expanded section, SI shows cut and fill structures and large trough

sets, similar to those of S2 (Fig. 7,10). SHALDRILI and II sampled the uppermost

strata of Si in 8 locations within the JRB. These drill cores sampled very poorly sorted,

glacimarine sediment and diamicton with abundant dropstones and no evidence of

layering. Minor bioturbation occurs within the glacimarine sediments. Paleontological

25

Fig

ure

10

Seis

mic

line

NB

P020

1-2

prov

ides

a s

trike

-orie

nted

vie

w o

f the

SI

expa

nded

sec

tion.

A m

inim

um o

f 10

gla

cial

un

conf

orm

ities

are

imag

ed w

ithin

SI

(Lat

e Pl

ioce

ne/P

leis

toce

ne to

pre

sent

).

to

data from the cores yielded Pleistocene and younger diatom assemblages with some

reworked older fossils, however, the base of unit SI was not sampled and could be older

than Pleistocene in age.

Age Model

An age model for the James Ross Basin, revised from the first model published by

Anderson et al. (1992), was developed using results from the SHALDRIL drill sites to

constrain the end member ages (units S5 and SI). The oldest strata that rim the northern

portion of the basin were originally estimated to be of late Eocene to early Oligocene age

based on along-strike correlation with strata of Seymour Island (Anderson et al. 1992).

The ages of younger sequences were estimated by comparing the stratigraphic

architecture and stacking pattern to various portions of the Antarctic margin.

Bartek et al. (1991) demonstrated similar continental shelf stacking patterns for

the Ross Sea, Antarctic Peninsula (Pacific margin), Gulf of Mexico and several other

margins which they named the Stratigraphic Signature of the Neogene (Fig. 11).

Significant changes in stratal geometry were assigned ages based on this global

correlation scheme. The main assumption in this correlation scheme is that eustasy was

the principle factor regulating the stratigraphic architecture on continental margins with

modest subsidence rates (Vail et al., 1977). Key stratigraphic horizons sampled by drill

core during DSDP Leg 28 in the Ross Sea and ODP Leg 178 on the Pacific Antarctic

margin support the age estimates derived from the seismic stratigraphic age model.

We use correlations in the stratigraphic architecture and sequence geometries to

create an initial age model for the JRB (Fig. 11). There is good correlation between the

stacking pattern from the idealized signature of the Neogene (Bartek et al., 1991) and that

Line

NB

P93

-22

i§i§

Ji

Str

atig

raph

ic S

igna

ture

of t

he N

eoge

ne

—^

Seq

uenc

e B

ound

ary

- - N

M

axim

um F

lood

ing

Sur

face

E

I^

Fluv

ial/D

elta

ic S

edim

ents

|

| Sha

le

[jjg

j] S

lope

and

Bas

in F

loor

Fan

s ^

B

Sha

llow

Wat

er C

arbo

nate

Figu

re 1

1

to

00

Stra

tigra

phic

arc

hite

ctur

e in

terp

reta

tion

and

an a

ge m

odel

for

the

JRB

usi

ng s

eism

ic li

ne N

BP9

3-22

. T

he a

ge m

odel

is b

ased

on

the

Stra

tigra

phic

Sig

natu

re o

f the

Neo

gene

mod

el ta

ken

from

Bar

tek

et a

l. (1

991)

.

imaged in our seismic lines. In general, the overall stacking pattern is as follows:

progradation of the continental shelf between 15 to 8.2 Ma, followed by aggradation from

8.2 to 4.2 Ma, then another progradational package from 4.2 to 2.4 Ma, and lastly,

aggradation from 2.4 Ma to present. These interpretations of stratigraphic architecture

changes are assigned based upon the entire seismic data set, and may not be fully

represented on a single line. We believe these are good initial estimates for the basin fill

of the JRB, however, this age model requires testing with additional drill core.

DISCUSSION

Pre-Glacial Phase (S5)

The base of unit S5 is roughly the basinward equivalent of the La Meseta

Formation on Seymour Island, based on along strike correlation to seismic records

(Anderson et al., 1992). Sedimentation spans from Late Eocene to Late Oligocene with

SHALDRIL drill core data confirming the age of a regional onlap surface within S5 as

being 32-37 Ma (SHALDRIL Scientific Party, 2007). Cored sediments show evidence of

fluvial/deltaic deposition. Specifically, sediments show a higher degree of sorting than

glacimarine deposits and there is an abundance of calcareous fossils, which indicates that

this unit pre-dates the full development of glacial conditions. The laminated seismic

facies records fairly continuous deposition with very little erosion, which is inconsistent

with glacial processes. Our interpretation is that these strata represent an outer shelf to

upper slope or ramp environment which pre-dates the APIS grounding on the continental

shelf. Scattered dropstones in the section likely record an initial ice cap on the continent.

29

Prior studies have argued against the existence of ice sheets in the northern

Antarctic Peninsula during the Late Eocene (see Anderson, 1999 for summary). The

Paleocene shows evidence of temperate climate conditions which supported a diverse

vascular flora. This flora shifted to one dominated by Nothofagus by Eocene time

(Askin, 1992). Using fossil evidence from the La Meseta Formation, Case et al. (1988)

described a setting where marsupials inhabited Nothofagus forests during the Late

Eocene/Early Oligocene. In addition, no sedimentological evidence, such as ice-rafted

debris, is found in the Eocene deposits on Seymour Island (Zinsmeister, 1982). In

contrast, outcrops of till on King George Island and Seymour Island (JRB) provide

evidence to support initial continental/alpine glaciation on the peninsula during the

Middle Eocene-Early Oligocene (49 to 32 Ma) and Early Oligocene respectively

(Birkenmajer, 1991; Ivany et al., 2006). The glacial outcrop found on Seymour Island

(Early Oligocene) would be roughly coincident with cored sediments of unit S5.

However, from our survey there is no evidence to support grounded marine ice sheets on

the continental shelf during this time.

Initial Ice Sheet Development (Unit S4)

Seismic shelf and slope facies of S4 bear a stark difference to the slope facies of

S5. S4 strata display prominent prograding packages of chaotic facies bounded by high

amplitude, continuous reflections. We interpret these packages to be till sheets

interbedded with glacimarine deposits. Erosional surfaces separate these packages (Fig.

4, 8) and are interpreted to be glacial troughs based on their size and depth (>30 km wide,

50-100 ms deep). The seismic facies within the troughs are seismically massive, chaotic

30

deposits indicative of subglacial deposition (King et al., 1991). The change in seismic

fades and stratigraphic architecture represented by S4 most likely represent the early

expansions of the ice sheet onto the continental shelf in the southern half of the study area

(Fig.5). Individual advances were followed by retreat of the ice sheet and deposition of

glacimarine sediments. Ice sheets advanced onto the shelf from the southwest, depositing

clinoforms that prograded to the northeast (Fig. 5). The lack of S4 deposits in the north

indicates that during S4 time, the APIS was too thin to flow over topographic highs, such

as James Ross, Snow Hill and Seymour Islands. These highs must have acted as a

boundary which prevented the northward expansion of the ice sheet. According to our

age model (Fig. 11), the first expansion of the ice sheet onto the continental shelf

occurred during the latest Middle Miocene/early Late Miocene.

Northward Expansion of Ice Sheet (Unit S3)

According to our age model, S3 is Late Miocene in age. At this time, the paleo-

shelf break experienced a noticeable change in orientation that was caused by increased

sediment input from the northern peninsula (Fig. 5). Seismic facies (similar to S4)

indicate a glacial origin for both the southern and northern lobes. This also indicates a

northward expansion of the ice sheet from S4 to S3 time. However, the separation of S3

into two directionally distinct depositional lobes (Fig. 5) implies the ice sheet was still

relatively thin or characterized by small ice caps, causing it to flow around the islands off

the peninsula.

31

Ice Sheet Dominated Phase (S2-S1)

Our age model places the S2 unit within the Early to Late Pliocene. During this

time, sediments continued to fill accommodation in the northern part of the basin (Fig. 5).

Shelf deposits thin to the south, owing to a lack of accommodation. Sedimentation was

focused on the shelf edge and slope. It is also likely that by this time the ice sheet had

eroded inland sedimentary basins of their easily eroded sediments. Truncation and total

removal of S2 deposits from the inner shelf, as well as landward sloping unconformities,

support the idea that the ice sheet began foredeepening the shelf during S2 time. Strike-

oriented profiles show dramatic examples of glacial troughs (>30 km wide, >50 meters

deep) filled with chaotic facies which strongly indicates a sub-ice sheet depositional

environment (Figs. 4 and 7). Preservation of the glacimarine deposits diminished during

S2 time. This may indicate more frequent ice sheet expansion onto the continental shelf

that resulted in shorter inter-glacial episodes and erosion of glacimarine deposits from

previous cycles.

Unit S1 contrasts in its geometry with previous units, particularly its drape-like

geometry and aggradational stacking pattern. The cause of this change in stratigraphic

architecture is not clear. Two prevailing hypotheses have been suggested for a very

similar change on the Pacific side of the peninsula. Bart and Anderson (1995) suggested

that this change is coincident with the establishment of Northern Hemisphere ice sheets,

which caused more rapid fluctuations in sea level. This may have had the effect of

increasing the frequency, but shortening the duration, of grounding events on the shelf.

A separate hypothesis was proposed by Rebesco et al. (2006) which links changes on the

continental shelf to changes on the rise on this and other margins, and suggests that these

were caused by a transition of the ice sheet from wet to dry based (temperate/sub-polar to

polar).

SI shows strong evidence of ice sheet-dominated processes in the form of broad

troughs filled with chaotic seismic facies. SHALDRIL drill core confirm a glacial origin

for SI deposits. Diatom assemblages conclusively date these sediments as Pleistocene or

younger. This agrees with our age model which shows unit SI as Latest

Pliocene/Pleistocene to present.

Number of Ice Groundings

Table 1 shows our interpretation of the number of ice grounding events

within each of the previously described units. Also included in this table is a comparison

between this study and that of Bart and Anderson (1995), conducted on the continental

shelf of the Pacific side of the peninsula. The comparison of these two interpretations

shows good correlation during the Pliocene, but large differences for the Late Miocene

and Pleistocene. As mentioned in the stratigraphy section, the S3 unit (Late Miocene)

was poorly imaged in our data set. This limitation leads us to believe that more

glacial/interglacial cycles potentially exist within S3 than were resolvable. During the

Pleistocene, at least 7 additional glacial unconformities are imaged in the JRB than were

observed on the Pacific side of the peninsula. However, these additional 7 events (total

of 10) are resolvable only within the SI expanded section (Fig. 10). Thus, conditions

present on the shelf during the Pleistocene led to poor preservation of glacial/interglacial

cycles for most of the JRB and Pacific side of the peninsula. This is also true for the

Ross Sea where a total of 7 glacial/interglacial cycles are interpreted from Pliocene to

33

Table 1.

Number of Ice Grounding Events (IGE's)

Unit

S1

S2

S3

S4

S5

Age

Pleistocene

Pliocene

Late Miocene

Middle to Late Miocene

Late Eocene/Oligocene

Bart and Anderson (1995)

3

10

18

? (Obscured)

0

James Ross Basin

10

10

>6

8

0

34

present (Anderson and Bartek, 1992; Alonso et al., 1992). The previous evidence for

only 3 glacial/interglacial cycles on the peninsula during the Pleistocene has been

problematic when trying to correlate seismic records with deep marine oxygen isotope

records. Our interpretation of 20 glacial/interglacial cycles during the Plio-Pleistocene is

more consistent (albeit still less) with the deep marine record, which shows cycles about

every 100 ka during the Pleistocene, and roughly every 41 ka during the Pliocene

(Raymo et al., 2006; Raymo and Huybers, 2008).

The oldest glacial unconformities imaged in the JRB are seen within the S4 unit,

which according to our age model is latest Middle Miocene/Early Late Miocene in age.

A total of 8 grounding events occurred during this time. No grounding events were

reported from the Pacific side of the peninsula for this time interval because of a high

degree of erosion and deformation following ridge-trench collision along the Pacific

margin during the Middle Miocene (S4). Futhermore, the mid-late Miocene section was

poorly imaged along strike and greatly obscured by the water bottom multiple on the

Pacific side of the Peninsula (Bart and Anderson, 1995).

CONCLUSIONS

The James Ross Basin has one of the most complete continental shelf successions

of Neogene strata anywhere in Antarctica. A seismic stratigraphic analysis of the area

utilized a fairly dense data set acquired during five separate cruises to the area. The

nomenclature and bounding surfaces of stratigraphic packages S5-S1 was modified from

prior studies by Anderson et al. (1992) and Sloan et al. (1995). An age model is

presented using general stratigraphic stacking patterns based upon the Stratigraphic

Signature of the Neogene (Bartek et al., 1991) (Fig. 11). The age model assumes eustasy

as the primary driver of stratigraphic architecture. Future drill core is needed to test this

hypthosis, however, these are the best estimates for the chronostratigraphy of the basin.

The oldest and youngest units (S5 and SI) have been confirmed by SHALDRIL drill core

as being Latest Eocene/Earliest Oligocene and Pleistocene in age, respectively.

According to the age model, the ages of the seismic units are as follows: S5 (Late

Eocene/Early Oligocene-Early Miocene), S4 (Middle to Late Miocene), S3 (Late

Miocene), S2 (Pliocene), SI (Pleistocene to Present).

Evidence from the JRB shows that the onset of ice sheet development on the

Antarctic Peninsula was gradual, and that the first ice sheets to advance onto the

continental shelf during the late Middle Miocene/early Late Miocene (unit S4) came from

the south. This implies that the APIS formed not long after the establishment of the West

Antarctic Ice Sheet in the Middle Miocene (see for a review Anderson, 1999; Zachos et

al., 2001). During the Middle-Late Miocene there was pronounced margin progradation

and a change in orientation of the shelf break relative to the pre-glacial configuration.

Prior to this time, descrete depocenters rimmed the basin, implying numerous glacio-

fluvial-deltaic sources.

During the Late Miocene the ice sheet progressively advanced northward (units

S4/S3). Initially its northward advance was blocked by togographic highs and islands of

the northern Peninsula. Full glaciation of the peninsula and the establishment of

grounded ice in the northernmost part of the basin occurred approximately in late Late

Miocene time when the APIS became thick enough to override the islands of the northern

peninsula.

36

Pliocene to present (S2-S1) deposits show evidence of ice-dominated deposition,

with margin strata composed of stacked glacial and glacimarine deposits bounded by

landward sloping unconformities. Prior to the late Pliocene, sediment supply was high

and the shelf prograded (S4-S2). The stratigraphic stacking pattern of the shelf changed

from progradational to aggradational during the Late Pliocene. Similar trends of shelf

erosion and aggradation occur in other areas of the Antarctic margin (Anderson and

Bartek, 1992; Alonso et al., 1992; Bart and Anderson, 1995; De Santis et al., 1999;

Rebesco et al., 2006).

An expanded section of Pleistocene age sediments shows evidence of at least 10

Pleistocene glacial/interglacial cycles (Fig. 10). This is the highest number of Pleistocene

grounding events recorded using seismic techniques on any Antarctic margin and is in

closer agreement with ice sheet history indicated by deep marine oxygen isotope records.

Furthermore, our results demonstrate that the magnitudes of ice sheet expansion during

these ten events were comparable to the Last Glacial Maximum in that the northernmost

portion of the ice sheet advanced across the continental shelf.

ACKNOWLEDGEMENTS

This work was made possible through the National Science Foundation, Grant #

0125922. We acknowledge Dr. Larry Lawver and the University of Texas for the use of

seismic records which they collected and processed. We would like to thank the crews of

the R/V Polar Duke, and R/V Nathaniel B. Palmer for their work during data collection.

37

REFERENCES CITED

Alonso, B., Anderson, J.B., Diaz, J.I., and Bartek, L.R., 1992, Pliocene-Pleistocene

seismic stratigraphy of the Ross Sea: Evidence for multiple ice sheet grounding

episodes, in Elliot, D.H., ed., Contributions to Antarctic Research III: Washington D.C.,

American Geophysical Union, Antarctic Research Series, v. 57, p. 93-103.

Anderson, J.B., 1999, Antarctic Marine Geology: New York, Cambridge University

Press,p.289.

Anderson, J.B., and Bartek, L.R., 1992, Cenozoic glacial history of the Ross Sea revealed

intermediate resolution seismic reflection data combined with drill site information, in

J.P. Kennett and Warnke, D.A., eds., The Antarctic Paleoenvironment: A Perspective

on Global Change I: Washington, D.C., American Geophysical Union, p. 231-263.

Anderson, J.B., Wellner, J.S., Wise, S.W., Bohaty, S., Manley, P.L., Smith, T., Weaver,

F., and Kulhanek, D, 2007, Seismic and chronostratigraphic results from SHALDRIL

II, northwestern Weddell Sea: U.S. Geological Survey and the National Academies,

USGSOOF-2007-1047, Short Research Paper 094.

Anderson, J.B., Shipp, S.S., and Siringan, F.P., 1992, Preliminary seismic stratigraphy of

the northwestern Wedell Sea continental shelf, in Y. Yoshida, et al., eds., Recent

Progress in Antarctic Earth Science: Tokyo, Terra Scientific Publishing Co., p. 603-

612.

Askin, R.A., 1992, The palynological record across the Cretaceous/Tertiary transition on

Seymour Island, Antarctica, in Feldmann, R.M., and Woodburne, M.O., eds., Geology

and Paleontology of Seymour Island, Antarctic Peninsula: Boulder, Colorado, The

Geological Society of America, Inc., Memoir 169, p. 131-154.

38

Barker, P.F., and Camerlenghi, A., 2002, Synthesis of Leg 178 results: Glacial history of

the Antarctic Peninsula from Pacific margin sediments, in Barker, P.F., et al., eds.,

Proceedings of the Ocean Drilling Program, Scientific Results: College Station, Texas,

Ocean Drilling Program, v. 178, p. 1-40.

Bart, P.J., and Anderson, J.B., 1995, Seismic record of the glacial events affecting the

Pacific margin of the northwestern Antarctic Peninsula, in Cooper, A.K., et al., eds.,

Geology and Seismic Stratigraphy of the Antarctic Margin: Washington D.C.,

American Geophysical Union, Antarctic Research Series, v. 68, p. 75-95.

Bartek, L.R., Vail, P.R., Anderson, J.B., Emmet, P.A., and Wu, S., 1991, Effect of

Cenozoic ice sheet fluctuations in Antarctica on the Stratigraphic Signature of the

Neogene: Journal of Geophysical Research, v. 96, no. B4, p. 6753-6778.

Birkenmajer, K., 1991, Tertiary glaciation in the South Shetland Islands, West

Antarctica: Evaluation of data: World and Regional Geology, v. 1, p. 629-632.

Canals, M., Casamor, J.L., Urgeles, R., Calafat, A.M., Domack, E.W., Baraza, J., Farran,

M, and De Batist, M., 2002, Seafloor evidence of a subglacial sedimentary system off

the northern Antarctic Peninsula: Geology, v. 30, no. 7, p. 603-606.

Carbotte, S. M., Ryan, W. B. F., O'Hara, S., Arko, R., Goodwillie, A., Melkonian, A.,

Weissel, R.A., Ferrini, V.L., 2007, Antarctic Multibeam Bathymetry and Geophysical

Data Synthesis: An On-Line Digital Data Resource for Marine Geoscience Research in

the Southern Ocean, U.S. Geological Survey Open-File Report 2007-1047, DOI:

10.3133/of2007-1047.srp002.

Case, J.A., and Woodburne, M.O., 1988, A new genus of polydolopid marsupial from

Antarctica, in Feldmann, R.M., and Woodburne, M.O., eds., Geology and Paleontology

39

of Seymour Island, Antarctic Peninsula: Boulder, Colorado, The Geological Society of

America, Inc., Memoir 169, p. 505-521.

Cooper, A.K., Barrett, P.J., Hinz, K., Traube, V., Leitchenkov, G., and Stagg, H.M.J.,

1991, Cenozoic prograding sequences of the Antarctic continental margin: A record of

glacio-eustatic and tectonic events: Marine Geology, v. 102, p. 175-213.

del Vaile, R.A., Elliot, D.H., and MacDonald, D.I.M., 1983, Sedimentary basins on the

east flank of the Antarctic Peninsula: Proposed nomenclature: Antarctic Science, v. 4, p.

477-478.

Elliot, D.H., 1988, Tectonic setting and evolution of the James Ross Basin, northern

Antarctic Peninsula, in Feldmann, R.M., and Woodburne, M.O., eds., Geology and

Paleontology of Seymour Island, Antarctic Peninsula: Boulder, Colorado, The

Geological Society of America, Inc., Memoir 169, p. 541-555.

Feldmann, R.M., and Woodburne, M.O., eds., 1988, Geology and Paleontology of

Seymour Island, Antarctic Peninsula: Boulder, Colorado, The Geological Society of

America, Inc., Memoir 169, p. 566.

Francis, J.E., Pirrie, D., and J.A. Crame, J.A., eds., 2006, Cretaceous-Tertiary high-

latitude paleoenvironments: James Ross Basin, Antarctica: London, Geological Society

Special Publication, no. 258, p. 206.

Gazdzicki, A., Tatur, A., Hara, U., and del Valle, R.A., 2004, The Weddell Sea

formation: Post-late Pliocene terrestrial glacial deposits on Seymour Island, Antarctic

Peninsula: Polish Polar Research, v. 25, p. 189-204.

40

Heroy, D.C., and Anderson, J.B., 2005, Ice-sheet extent of the Antarctic Peninsula region

during the Last Glacial Maximum (LGM) - Insights from glacial geomorphology:

Geological Society of America Bulletin, v. 117, p. 1497-1512.

Huber, B.T., 1988, Upper Campanian- Paleocene foraminifera from the James Ross

Island region, Antarctic Peninsula, in Feldmann, R.M., and Woodburne, M.O., eds.,

Geology and Paleontology of Seymour Island, Antarctic Peninsula: Boulder, Colorado,

The Geological Society of America, Inc., Memoir 169, p. 163-252.

Ineson, J.R., Crame, J.R., and Thomson, M.R.A., 1986, Lithostratigraphy of the

Cretaceous strata of west James Ross Island: Cretaceous Research, v. 7, p. 141-159.

Ivany, L.C., Van Simaeys, S., Domack, E.W., and Samson, S.D., 2006, Evidence for an

earliest Oligocene ice sheet on the Antarctic Peninsula: Geology, v. 34, p. 377-280.

King, L.H., Rokoengen, K., Fader, G.B., and Gunleiksrud, T., 1991, Till-tongue

stratigraphy: Geological Society of America Bulletin, v. 103, p. 637-659.

Larter, R.D., and Barker, P.P., 1989, Seismic stratigraphy of the Antarctic Peninsula

Pacific margin: a record of Pliocene-Pleistocene ice volume and paleoclimate: Geology,

v. 17, p. 731-734.

Larter, R.D., and Barker, P.F., 1991, Effects of ridge crest-trench interaction on

Antarctic-Phoenix spreading: forces on a young subducting plate: Journal of

Geophysical Research, v. 96, p. 19583-19607.

Larter, R.D., Cunningham, A.P., 1993, The depositional pattern and distribution of

glacial-interglacial sequences on the Antarctic Peninsula: Marine Geology, v. 109, p.

203-219.

41

Lear, C.H., Bailey, T.R., Pearson, P.N., Coxall, H.K., and Rosenthal, Y., 2008, Cooling

and ice growth across the Eocene-Oligocene transition: Geology, v. 36, p. 251-254.

Marenssi, S.A., Net, L.I., and Santillana, S.N., 2002, Provenance, environmental and

paleogeographic controls on sandstone composition in an incised-valley system: The

Eocene La Meseta Formation, Seymour Island, Antarctica: Sedimentary Geology, v.

150, p. 301-321.

McGinnis, J.P., Hayes, D.E., and Driscoll, N.W., 1997, Sedimentary processes across the

continental rise of the southern Antarctic Peninsula: Marine Geology, v. 141, p. 91-109.

O' Cofaigh, CO., Pudsey, C.J., Dowdeswell, J.A., and Morris, P., 2002, Evolution of

subglacial bedforms along a paleo-ice stream, Antarctic Peninsula continental shelf:

Geophysical Research Letters, v. 29, no. 8, p. 41-1-41-4.

Raymo, M.E., Lisiecki, L.E., and Nisancioglu, K.H., 2006, Plio-Pleistocene ice volume,

Antarctic climate, and the global 818 O Record: Science, v. 313, p. 492-495.

Raymo, M.E., and Huybers, P., 2008, Unlocking the mysteries of ice ages: Nature, v.

451, p. 284-285.

Rebesco, M., Larter, R.D., Barker, P.R., Camerlenghi, A., and Vanneste, L.E., 1997,

History of sedimentation on the continental rise west of the Antarctic Peninsula, in

Barker, P.F. and Cooper, A.K., eds., Geology and Seismic Stratigraphy of the Antarctic

Margin: Washington D.C., American Geophysical Union, Antarctic Research Series v..

71, p. 29-49.

Rebesco, M., Camerlenghi, A., Geletti, R., and Canals, M., 2006, Margin architecture

reveals the transition to the modern Antarctic ice sheet ca. 3 Ma: Geology, v. 34, p. 301-

304.

Sadler, P.M., 1988, Geometry and stratification of uppermost Cretaceous and Paleogene

units on Seymour Island, northern Antarctic Peninsula, in Feldmann, R.M., and

Woodburne, M.O., eds., Geology and Paleontology of Seymour Island, Antarctic

Peninsula: Boulder, Colorado, The Geological Society of America, Inc., Memoir 169, p.

541-555.

SHALDRILII Cruise Report, 2006, Anderson, J.B., et al., eds., Report of cruise number

NBP0602A to the northwestern Weddell Sea, 369 p.

Sloan, B.J., Lawver, L.A., and Anderson, J.B., 1995, Seismic stratigraphy of the Palmer

Basin, in Cooper, A.K., et al., eds., Geology and seismic stratigraphy of the Antarctic

Margin: Washington, D.C., American Geophysical Union, Antarctic Research Series, v.

68, p. 235-260.

Vail, P.R., Mitchum, R.M., Jr., Todd, R.G., Widmier, J.M., Thompson, S., Ill, Sangree,

J.B., Bubb, J.N., and Hatelid, W.G., 1977, Seismic stratigraphy and global changes of

sea level, in Clayton, C.E., ed., Seismic Stratigraphy-Applications to Hydrocarbon

Exploration: American Association of Petroleum Geologists, Memoir 26, p. 49-212.

Wellner, J.S., Heroy, D.C., Anderson, J.B., 2006, The death mask of the Antarctic Ice

Sheet: Comparison of glacial geomorphic features across the continental shelf:

Geomorphology, v. 75, p. 157-171.

Zachos, J., Pagani, M., Sloan, L., Thomas, E., and Billups, K., 2001, Trends, rhythms,

and aberrations in global climate 65 Ma to present: Science, v. 292, p. 686-693.

Zachos, J.C., and Kump, L.R., 2005, Carbon cycle feedbacks and the initiation of

Antarctic glaciation in the earliest Oligocene: Global and Planetary Change, v. 47, p.

51-66.

43

Zinsmeister, W.J., 1982, Review of the Upper Cretaceous Lower Tertiary sequence on

Seymour Island, Antarctica: Journal of the Geological Society, v. 139, p. 779-785.

44

CHAPTER 2 - Sediment flux to the Antarctic Peninsula margin based on first order stratigraphic sequences of the continental shelf and rise

R. Tyler Smith Rice University Earth Science Dept. 6100 Main St. MS 126 Houston, TX 77005

John B. Anderson Rice University Earth Science Dept. 6100 Main St. MS 126 Houston, TX 77005

Michele Rebesco Istituto Nazionale di Oceanografia e di Geofisica Sperimentale, Borgo Grotta Gigante 42/C, 34010 Sgonico (TS), Italy

ABSTRACT

Isopachs of the first order stratigraphic units from the continental shelf and rise

were created within a regional area from south of Adelaide Island to north of Brabant

Island. ODP Leg 178 cores were used to define chronostratigraphic intervals as well as

convert two-way-time seismic profiles to depth. Sediment flux to the Antarctic Peninsula

Pacific margin increased from the Late Eocene until the Late Pliocene when they

decreased slightly. Sediment flux is compared between the shelf and the rise, as well as

the opposing continental shelves of the James Ross Basin (NW Weddell Sea) and the

Pacific side. Comparisons show diverging trends in the last ~3 Ma. Large increases in

sediment flux coincide with the initial growth of the Antarctic Peninsula Ice Sheet and

the Early Pliocene warm period (Barker and Camerlenghi, 2002). Denudation rates for

the Antarctic Peninsula are calculated to be in the range of 0.06-0.13 mm yr"1.

INTRODUCTION

The power of ice sheets to erode and transport sediment is apparent throughout

the world, however, estimating rates of erosion and sediment flux by ice over geologic

time has proven challenging. This is particularly important in Antarctica where ice sheets

are the primary mechanism for sediment creation and transport. Some modern day

temperate glaciers, such as those in Alaska, show very high rates of denudation (10-100

mm yr"1) over the last century (Hallet et al., 1996). However, comparisons between

modern (100 yr) and longer (Ma) term glacial sediment flux have been poorly

constrained. One approach to calculating the erosion rate by ice sheets in Antarctica is to

estimate the rate of deposition of sedimentary deposits on the continental margins.

Factors such as climate, underlying substrate, topography, and catchment source

area have a large impact in the capacity of ice to erode and carry sediment (Benn and

Evans, 1998; Anderson, 1999). In Antarctica, the ice sheet has experienced episodes of

dramatic change in all of the above factors. These changes should have produced

variations in the overall sediment flux to the margins, the timing of which can help to

better constrain ice sheet evolution models.

The Antarctic Peninsula, the Ross Sea Embayment, and the Weddell Sea are three

locations around Antarctica where there is adequate data coverage with which to study

sediment flux to the margin. In the Ross Sea, calculating sediment flux coming from a

single source is difficult because of the convergence of both the East Antarctic and West

Antarctic ice sheets. In addition, much of the sediment delivered from the Transantarctic

Mountains is sequestered within the West Antarctic Rift System, situated just north/east

at the base of the mountains, before reaching the Ross Sea Basin. The southeastern

46

Weddell Sea has a separate issue caused by the narrow convergence of fast moving ice

within Crary trough where the majority of sediments have been reworked several times

and ultimately delivered to the Crary Trough Mouth Fan and the Weddell Fan (Johnson et

al., 1981; LaBrecque and Keller, 1982; Elverhed and Maisey, 1983; Miller et al., 1990;

Kuvaas and Kristoffersen, 1991; Moons et al., 1992; De Batist et al., 1994). This makes

it hard to quantify the amount of sediment that was deposited in each glacial cycle.

For these reasons, the Antarctic Peninsula is considered to be the best place to

estimate sediment flux through time. Due to its northerly location, the shallow waters

around the peninsula remain ice free during the summer. This allowed for a dense grid of

seismic data to be obtained on both sides of the peninsula as well as 3 marine drilling

operations, DSDP Leg 325, ODP Leg 178, and SHALDRIL (Fig. 1).

The Antarctic Peninsula margin is an example of an isolated continental shelf and

rise which was sourced by a northward trending mountain range connected to the greater

Antarctic continent. Figure 2 shows a cross section through the Antarctic Peninsula

highlighting the large sediment wedges on the continental shelf. The differences in

sediment wedge geometry on either side are striking. The shelf on the Weddell Sea side

of the peninsula is much wider, however, studies have shown that both margins share

similarities in their stratigraphic architecture (Anderson et al., 1992; Sloan et al., 1995;

Smith and Anderson, submitted).

A key to estimating total sediment flux is to understand the relationship between

sediment distribution to the shelf and the rise. Previous studies of the Antarctic

Peninsula margin have been focused on either the shelf or the rise and have not presented

47

70WW I

60WW

65iO'0"S+

yo'B'O'S-

^ C V *

ft

?E» • Jamefc Rossis, 1102

SHALDRIU Jite3d

7(ro'crw 6ffO'0"W

Figure 1

Location map showing seismic lines, drill core sites, oceanic transform boundaries and the interpreted age of ridge trench collision for each segment, and the position of the modern shelf break.

48

Cro

ss s

ectio

n th

roug

h th

e Pe

nins

ula

com

parin

g th

e co

ntin

enta

l she

lves

of

the

Paci

fic m

argi

n an

d th

e W

edde

ll Se

a m

argi

n (J

RB

). T

he sh

elve

s sh

ow si

mila

ritie

s in

stra

tigra

phic

arc

hite

ctur

e, b

ut d

iffer

gre

atly

in

over

all w

idth

.

an intergrated study of both. The relationship between stratigraphic architecture and

sediment accumulation on both the shelf and the rise will help to identify changes in the

sediment flux, such as times of bypass versus a decrease in overall sediment coming into

the system.

SETTING

The Antarctic Peninsula is a north to south trending mountain range with a mean

elevation of 1-2 km and a width of 100-300 km which was created by a volcanic arc

formed during the subduction of the Phoenix Plate (later named the Aluk mini-plate)

(Barker, 1982; Elliot, 1988). It is composed of igneous and metamorphic rock complexes

(Saunders et al., 1982). There is very little bedrock exposure on the peninsula due to

thick ice cover.

The Pacific margin of the Antarctic Peninsula has a complex tectonic history of

time-trangressive rift-trench collision. Latter and Barker (1991) identified and dated a

sequence of transform boundaries separating segments of oceanic crust extending into the

subduction zone (Fig. 1). The stratigraphy of the shelf shows regional unconformities

which formed in response to the uplift caused by the heat of the ridge as it was subducted

(Larter and Barker, 1991; Bart and Anderson, 1995). Along the northern portion of the

margin, where ridge-trench collision is currently occurring, the uplifted shelf has been

recently scoured by the ice sheet and only bedrock remains (Fig. 3).

The Antarctic Peninsula continental shelf can be divided into two sections, the

sediment barren inner shelf, and the prograded sedimentary outer shelf (Fig 3). Bart and

Anderson (1995) mapped the main stratigraphic units and regional unconformities on the

shelf and counted 31 shelf-wide glacial unconformities. The thick chaotic intervals of

50

Pacific margin PD 88-9

Pacific margin PD 88-5 ~ :..".•. qi/8,~,~v"oi/»,,.,,"',•-. .01/3: - ~ : . ~ j , . . ~ : ' , . :oi«

r-* r

Pacific margin PD 90-58 la

James Ross Basin NBP 93-22

===s3ES^@a;

Figure 3

Series of seismic lines from the Pacific margin continental shelf showing the northward narrowing of the shelf due to later uplift of the shelf caused by a time-trangressive ridge-trench collision. A line from the JRB on the other side of the peninsula is shown for comparison.

51

units S3-S1 (Fig. 3) represent deposition during ice sheet advance and early retreat

separated by high-amplitude, continuous reflections representing intervals of hemipelagic

deposition during glacial minima (King et al., 1991; Anderson, 1999). The stratigraphic

architecture varies from south to north across the shelf and is divided into mostly

aggradational strata within unit (S3) overlain by a primarily progradational unit (S2).

The youngest unit (SI) sometimes includes progradational strata at the base, but is

primarily an aggradational drape that covers the entire outer shelf with exception of the

northernmost portion of the peninsula (Bart and Anderson, 1995). The characteristics of

each unit were interpreted as the general stratigraphic architecture and may not be

represented on a single line.

The continental rise consists of a series of sediment mounds which were deposited

at the toe of the slope. Rebesco et al. (1997,2002) interpreted the mounds as "drifts"

derived from sediment sourced from sediment gravity flows that were subsequently re-

distibuted by contour currents. They subdivided the drifts into six stratigraphic units M6-

Ml and three developmental stages: pre-drift (M6-M5), drift growth (M4-M3), and drift

maintenance (M2-M1).

Both the continental shelf and rise of the Pacific side of the peninsula were drilled

in 2002 during ODP Leg 178. Previously the only other drill site in the survey area was

on a distal continental rise location drilled during DSDP Leg 35 (Site 325).

Unfortunately, recovered sediments at DSDP Site 325 represent only -4.8% of the entire

drilled sequence (Hollister et al., 1976). These sites provide the chronostratigraphic

framework for the area. The base of the glacial section, however, was not reached at any

of the drill sites, so questions remain as to the timing of the onset of full glaciation of the

52

peninsula. The ODP Leg 178 cores show that ice has grounded on the shelf since at least

the early Late Miocene (Barker and Camerlenghi, 2002).

On the Weddell Sea side of the peninsula, the James Ross Basin (JRB) formed as

a back-arc basin in response to the subduction on the Pacific side of the peninsula. The

western flank of the basin includes James Ross Island and Seymour Island (Fig. 1) which

are two of the most intensely studied continental locations in Antarctica (Feldman and

Woodburn, 1988; Francis et al., 2006; Smellie et al., 2008). They remain relatively ice

free during summer months and have produced the most complete paleontological history

of Antarctica through the Paleogene. Strata within the James Ross Basin show

similarities in stratigraphic architecture and seismic facies to strata on the Pacific side of

the peninsula (Anderson et al., 1992; Sloan et al., 1995; Smith and Anderson, submitted).

Smith and Anderson (submitted) argue that the James Ross Basin may provide a more

complete history of sedimentation for the peninsula because it has provided prolonged

subsidence and generation of accommodation, unlike the Pacific side of the peninsula

where sediments were more intensely eroded due to tectonic uplift. Smith and Anderson

(submitted) proposed an age model for the JRB based on correlations between its

stratigraphic architecture to that of the Pacific side of the peninsula. Unfortunately, there

is a lack of published seismic data from the continental rise on the Weddell Sea side of

the peninsula.

An aspect that must be considered when conducting a study on both sides of the

Antarctic Peninsula is the difference in climate. An orographic effect caused by the

Antarctic Peninsula mountains has a profound effect on the annual mean temperatures

53

and precipitation on either side (Morris and Vaughan, 2003). The western margin is

warmer and receives more precipitation than the cooler and dryer eastern margin.

DATA

Because of sea-ice cover, it is often impossible to collect mulitchannel seismic

data in Antarctica, especially on the continental shelf. Both multi-channel and single-

channel seismic data were incorporated into this study. A collection of seismic data sets

obtained during several different years and by different countries provide a relatively

dense seismic grid along a portion of the continental margin. Details of the seismic

acquisition parameters and processing can be found in the original publications of the

data (Anderson et al., 1992; Bart and Anderson, 1995; 1996; Sloan et al., 1995; Rebesco

et al., 1997; Bart et al., 2005; Rebesco et al., 2002; Smith and Anderson, submitted).

Several seismic lines were shot on the continental rise by the Institute Nazionale

di Oceanografia e di Geofisica Sperimentale (OGS) with the R/V OGS-Explora within

the Programma Nazionale di Ricerche in Antartide (PNRA). Data sets are identified by

the year of acquisition and paired with the resultant publication: 1992 (Rebesco et al.,

1997), 1995,1997 (Rebesco et al., 2002). Data sets on the continental shelves on both

sides of the peninsula were acquired in 1988,1990 and 1991 by Rice University using the

R/V Polar Duke (Anderson et al., 1992; Bart and Anderson, 1995; 1996). Additional

data sets on the shelf were acquired using the R/V Nathaniel B. Palmer in 1993 by the

University of Texas (Sloan et al., 1995), and in 2001 and 2006 by Rice University (Bart

et al., 2005; Smith and Anderson, submitted).

ODP Leg 178 drilled nine sites along the Pacific margin of the Antarctic

Peninsula (Fig. 1). Two sites (1098 and 1099) targeted an expanded Holocene

sedimentary section located at Palmer Deep and were not included in the study. Sites

1100,1102, and 1103 were drilled to the north of the study area on the continental shelf.

Included within our study area are sites 1095,1096, and 1101 located on the continental

rise, and site 1097 located on the continental shelf (Fig. 1). Also included within the

study area is DSDP Leg 35 site 325, drilled on the distal portion of the continental rise.

Lastly, sites drilled during SHALDRIL provide the only chronostratigraphic control for

our age model for the JRB (Smith and Anderson, submitted). Of these, only Site 3c was

used in this study.

METHODS

Sediment Yield

The various seismic data sets from different years and cruises were compiled into

a geographic information system. The interpretations of the stratigraphic horizons on the

shelf and rise were stored as discrete data points approximately 100 shot points apart,

totaling 17,280 individual points along the seismic lines. Isochron maps (TWT) of unit

thicknesses were created using a natural neighbor interpolation algorithm and was

clipped to the edge of the data set. All volume calculations were done using surfaces

created by a triangulated irregular network (TIN) interpolation function (see below in

Denudation Calculation). Sediments below the bottom water multiple were not included

or measured. Ages for the bounding surfaces between units and the age duration of each

unit were estimated using magnetostratigraphy and biostratigraphy of sediments cored

•55

during ODP Leg 178 sites 1095,1096,1101, and 1097 (Iwai et al., 2002). Later studies

have refined some of the age ranges for the bounding surfaces between S1/S2 and S2/S3

(Bart et al., 2005; Rebesco et al., 2006). Bounding surfaces on the rise below the

ultimate depth reached by drilling operations were approximated by Rebesco et al.

(1997). The base of S3 on the shelf was estimated to coincide with the base of M3

located on the rise. Unit S4 on the shelf was omitted from the study due to the lack of

exposure and poor chronostratigraphic estimates for its bounding surfaces.

Age constraints for the oldest unit (S5) within the JRB come from SHALDRIL

site 3C, which sampled sediments from the Eocene/Oligocene boundary (Anderson et al.,

2007). Other sites to the northeast sampled Late Oligocene, Middle Miocene and Early

Pliocene sediments, however, using available seismic data we were unable to make

stratigraphic ties to the JRB. Therefore, the age model for the JRB interprets

chronostratigraphic surfaces by matching similarities in stratigraphic architecture from

the western side of the peninsula to those of Bartek et al. (1991) and using similarities in

stratigraphic architecture on both sides of the Peninsula (Smith and Anderson, submitted)

(Fig-2).

Denudation Calculation

Depth to the stratigraphic boundaries were linearly interpolated using a

triangulated irregular network (TIN) function. This method differs from the natural

neighbor interpolation method used to create the isopach contour maps. A linear TIN

interpolation was preferred in order to avoid additional error added by interpolation

algorithms. Continental rise and shelf surfaces were converted to depth using

56

velocity/depth profiles acquired at ODP Leg 178 sites 1095,1096,1101 and 1097

respectfully (Volpi et al., 2002; Bart et al., 2005). The velocity profiles are Consistent

with sonar buoy velocity data for the peninsula continental margin (Hayes et al., 1991).

Calculations made to convert sediment volumes to overall average erosion rate

were done using the methodology of Koppes and Hallet (2006): Vs- Volume of sediments ps-Density of sediments

(Vs)(ps) pp- Density of parent rock — — — — •= Average Erosion Rate A - Erosive area

(Ae)(T)(pp) T- Time span

Average values for each sedimentary unit's rock density and percent of biogenic volume

were derived from sites 1095,1096,1101 and 1097 and compared with neighboring sites

1100,1102, and 1103. The average value assigned for the density of parent source rock

from its continental source was 2.75 g/cm3 (mixed igneous and metamorphic source).

Two end members, a maximum and a minimum area, were calculated to represent the

erosive source. The maximum area represents everything from the ridge line of the

Antarctic Peninsula mountains to the landward edge of the sediment wedge located on

the outer continental shelf. The minimum area was an estimation of the area of bedrock

erosion which excluded the inner shelf where sediments are being reworked. This makes

the assumption that there was minimal basement rock removal on the inner shelf. The

rugged bathymetry of the inner shelf basement rock supports minimal bedrock erosion.

No approximation is made for the quantity of sediment transported beyond the seismic

coverage interpolation area. This was not a closed sedimentary system, therefore the

percentage of sediment "lost" out of the system may be significant. The values

calculated with respect to sediment flux and continental denudation are thus minimum

estimates constrained by the study area.

58

RESULTS

Isopach maps (Fig. 4) of the individual units first interpreted by Rebesco et al.

(1997) for the continental rise (M1-M5) and Bart and Anderson (1995) for the continental

shelf (S1-S3) show an uneven distribution of sediment. On the continental rise,

sedimentation became more focused on the tops of the drifts through time. Rebesco et al.

(1997) identified a general focusing of sedimentation through a characterization of three

development stages: the pre-drift stage (M6-M5), the drift growth stage (M4 and M3),

and the drift maintenance stage (M2 and Ml). Figure 4 shows unit M6 on the rise to be

confined to the southern portion of the study area. Sediments deposited during M6 to the

north of the Tula Fracture Zone (F.Z.) were later subducted during the ridge-trench

collision of the plate segment (Larter and Barker, 1991). A portion of sediments

deposited during unit M5 were also subducted to the north of the Adelaide F.Z. Units M6

and M5 show depocenters which fill basement lows and onlap onto basement highs. The

isopach map of unit M4 shows sedimentation was fairly uniform across the rise. During

M3 sediment accumulation began to be focused on the bathymetric highs, but remained

much more uniform in distribution than later units. The drift maintenance stage (Ml and

M2) shows a gradual change in sediment distribution and the development of strong

depocenters located on bathymetric highs.

On the shelf, most if not all of the older units (S5 and S4) were eroded during the

uplift of the shelf during ridge-trench collision. Isopachs of units S3-S1 show the

majority of the inner shelf was scoured down to bedrock and devoid of sediments (Fig.

4). An inner shelf mini-basin (not shown in isopach maps) contains sediments of

unknown age and no direct stratigraphic ties to the outer shelf wedge. The irregular

59

distribution of S3 has been studied in detail by Bart et al. (2005). Their findings call for

areas of thin S3 to be caused by erosion by ice streams. In the southern portion of the

data set, the thinner area of S3 represents the erosion of Margarite trough. Unit S2 was

interpreted as a progradational unit with more than 2 to 3 km of slope progradation (Bart

and Anderson, 1996) which is demonstrated by the concentrated accumulations at the

shelf edge. Unit SI represents a change from dominantly progradational stratigraphic

architecture to aggradational stacking patterns. The isopach of S1 shows a nearly

uniform, thin sediment package that covers almost all of the outer shelf. On the eastern

side of the peninsula within the James Ross Basin, Smith and Anderson (submitted)

constructed isopach maps of units S5-S1 which based on an age model for the basin, are

believed to correlate to units on the western side of the peninsula. Thus, these isopach

maps were used to calculate sediment volumes.

Figure 5a presents shelf sediment volume accumulation data (Table 1) for both

sides of the peninsula. The opposing sides show dramatically different trends in overall

accumulation. The western margin shows higher accumulation for units S3 and S2 with a

sharp decrease during SI time. The opposite trend occurs in the JRB, where overall

accumulations decrease during S3 and S2 time and then increase dramatically during SI

time. The sediment volumes on either side of the peninsula show relatively similar

values, despite the greater width of the shelf on the eastern margin (Fig. 2).

A similar comparison of sediment accumulation is made between the continental

shelf and the rise for the western side of the peninsula (Fig. 5b). The continental rise

shows increasing volumes from M6-M4, then a decrease in overall sediment volume until

unit M2 deposition, followed by a significant increase in volume during Ml time. On the

60

Shelf Accumulation 9000

8000

7000

6000

, 5000 Km3

4000

3000

2000

1000

0 SI S2 S3 S4 S5

Unit

Pacific Margin Accumulation 50000

40000

30000

20000

10000

0

SI Ml S2M2 S3M3 M4 M5 M6 Unit

Figure5

a. Comparison of first order unit volume accumulations from the Pacific and Weddell Sea continental shelves.

b. Comparison of first order unit volume accumulations of the continental shelf and rise on the Pacific margin.

61

Q Pacific Shelf

• JRB Shelf

• Pacific Shelf

• Pacific Rise

Table 1

Unit

S1

S2

S3

M1

M2

M3

M4

M5

M6

Location

Shelf

Shelf

Shelf

Rise

Rise

Rise

Rise

Rise

Rise

Aae(Ma)

0-2.9

2.9-5.12

5.12-7.94+

0-2.9

2.9-5.5

5.5 - 9.45

9.45-15

15-25

25-36

Reference

(Rebesco et al. (2006)

Iwai and Winter (2002) and Bart et al. (2005)

Iwai and Winter (2002) and Bart et al. (2005)

(Rebesco et al. (2006)

Iwai et al. (2002) and Volpi et al. (2002)

Iwai et al. (2002) and Volpi et al. (2002)

Rebesco etal. (1997)

Rebesco etal. (1997)

Rebesco etal. (1997)

shelf, units S3 and S2 are similar in total volume, whereas there is a significant decrease

in volume during SI time. The patterns of sediment accumulation on the shelf and rise

seem to follow opposing trends during the transition between units S2/M2 and Si/Ml.

Table 1 gives total accumulation volumes and outlines the chronostratigraphy for

the continental shelf and rise units obtained from ODP Leg 178 sites 1095,1096, and

1101 on the continental rise, and site 1097 on the shelf. The bases of units S3 on the

shelf and M4 on the rise were not reached during drilling, therefore ages for the bases of

these units and below are estimated based upon regional unconformities coinciding with

the uplift of the margin during ridge-trench collision (Larter and Barker, 1991; Bart and

Anderson, 1995; Rebesco et al., 1997).

DISCUSSION

Sediment accumulation on the Antarctic Peninsula continental margin over the

last 25 Ma has varied with respect to the tectonic environment, climate, and sediment

delivery mechanism. Figure 6 summarizes sediment flux to the margin through time.

The graph on the left shows the total volume of sediment accumulated on the western

margin. There, accumulation rates increased through time until approximately 2.9 Ma

(base of unit SI) when rates decreased slightly. The highest accumulation rates occurred

during unit S2 depositon, which spans' a period of early Pliocene warming identified

during ODP Leg 178 (Hillenbrand and Futterer, 2001; Barker and Camerlenghi, 2002).

Clay mineralogy and IRD concentrations from the drill core indicate that the ice sheet

continued to expand across the continental shelf during this time (Hillenbrand and

Ehrmann, 2001; Hassler and Cowan, 2001; Pudsey, 2001). A later decrease in

63

50000 100000 Knf

150000

Km 4to 200000 0 2000 4000 6000 8000 100001200014000

Transition to Cold Based (Polar Conditions) Ice Sheet

Early Pliocene Warming!

rliest APIS Expands onto the Continental shelf!

+•* Ridge-Trench Collision South of Biscoe F.Z.

l]iidge-Trench Collision South of Adelaide F.Z.

Ridge-Trench lolfision South t f Tula FZ.

(This curve only includes Pacific side accumulations)

Figure 6

(Left) Total sediment accumulation through time on the Pacific margin of the Antarctic Peninsula. This graph excludes volume calculations from the Weddell Sea. The timing of major tectonic and climatic events is shown. (Right) Sediment flux through time for the Pacific shelf and rise, Pacific rise, Pacific shelf, and JRB shelf (Weddell Sea) show opposing trends during the last 5 Ma.

64

accumulation rate during the Late Pliocene and Quaternary coincided with what has been

described as the transition to polar (cold based) ice sheet conditions around 3 Ma on the

peninsula (Rebesco et al., 2006). Rates of accumulation and sediment flux increased

significantly during the initial glaciation (-15-10 Ma) and growth of the ice sheet (-10-5

Ma) on the peninsula. The exact timing of glaciation is disputed (Anderson, 1999), but

drill core on the shelf from ODP Leg 178 site 1097 sampled diamicton fades interpreted

to be sub-ice sheet deposits as old as 8 Ma. Evidence from continental volcanic deposits

concurs with the marine deposits and records ice thicknesses of up to 750 m for at least

the last 6 Ma (Smellie et al., 2008).

The graph to the right on Figure 6 shows flux of the individual stratigraphic units

with respect to time. The sediment flux to opposing continental shelves display similar

total volumes, but very different trends. The highest flux on the shelf of the western side

of the peninsula occurred during S2 time, whereas within the JRB the rates slightly

decreased during S2 and increased dramatically during SI. This could be explained three

different ways. First, sediment delivery to the shelf may be controlled by different ice

sheet conditions and therefore different ice flow dynamics. This is improbable because

the stratigraphic architecture and seismic facies of the units are so closely matched on

both sides of the peninsula. Second, the timing of glacial onset, and therefore the

increase in sediment delivery, may differ due to the more northerly location of the data

set within the JRB. Smith and Anderson (submitted) argued that the ice sheet migrated

northward over time and that the northernmost peninsula may not have been fully

glaciated until Pliocene time. The later growth of the ice sheet may explain the delayed

increase in sediment flux to the shelf.

65

Lastly, the tectonic environment of the western (Pacific) margin may have caused

increased bypass of the shelf to the rise during SI, whereas the JRB continued to act as a

sediment trap. Figure 2 shows the differences in shelf geometry across the peninsula.

The narrower shelf on the Pacific side may have allowed sediments to bypass more

efficiently as the ice sheet became larger and more established. Evidence for this is

shown by the increase in sediment accumulation during Ml on the continental rise being

coincident with the decreased rate during SI on the shelf. On the eastern side of the

peninsula, the JRB continued to subside as a back-arc basin which provided added

accommodation through time.

Finally, the cause of the differences in continental shelf accumulation was most

likely caused by differences in shelf physiography and tectonics of either side of the

peninsula, as well as the different climate setting and the timing of ice sheet growth in the

two regions.

Volumes of the shelf units on either side of the peninsula are similar in

magnitude, despite differences in accumulation across these margins. This is

contradictory to what would be expected given the lower temperature and precipitation

on the eastern margin, a product of the orographic effect of the peninsula mountains.

Unfortunately, volumes of sediments on the continental rise of the eastern margin were

not calculated, therefore, it is not possible to determine whether the total sediment flux to

the margin is similar to the western margin. It is likely, however, that sediment bypass of

the shelf was lower on the eastern side than on the western margin because of increased

accommodation within the JRB.

66

Isopachs of the continental rise units (Ml -M5) of the western margin show a

trend of greater thickness on the tops of sediment mounds (Fig. 4). Sediment flux also

increased over time (to the continental rise) (Fig. 6). This demonstrates that while

sedimentation was becoming more focused into depocenters, overall accumulation rates

were increasing. Therefore, we propose that the nomenclature to characterize drift

development should be changed from the original interpretation (Rebesco et al., 1997) of

"drift growth" during units M4 and M3 to "drift formation", and "drift maintanence"

during units M2 and Ml to "drift enhancement".

Denudation Rates

A final step, though a significant one, is to convert the sediment volumes back

into the parent rock equivalent volumes and calculate a minimum rate of denudation for

the Antarctic Peninsula. The system is not closed, so a considerable amount of sediment

may have been lost to the outer continental rise and abyssal plain. This is indicated by

the fact that DSDP Leg 35 site 325 sampled large sections of stacked turbidites on the

continental rise located at the edge of the data set (Tucholke and Houtz, 1976). However,

our calculations should provide reasonable estimates to within an order of magnitude of

the overall denudation rate. Table 2 presents the results of volume to rock equivalent to

denundation rates. Two different areas were used to represent maximum and minimum

source areas.

Halletet al (1996) and Koppes and Hallet (2002, 2006) have shown that

estimation of modern erosion rates for mountain glaciers are within the range of 10"1 to

102 mm yf1 depending on the drainage basin area, the bedrock lithology, precipitation,

67

Table 2

Minimum Erosive Source Area = 70600 Km2

Unit

S1+M1

S2+M2

-S3+M3

M4

M5

M6

Volume (Km*3)

37800

35000

46200

43100

35700

26600

Time (Ma)

2.90

2.36

4.19

5.55

10.0

11.0

Average Erosion Rate (mm/yr)

0.115

0.132

0.100

0.079

0.039

0.026

Maximum Erosive Source Area= 120000 Km2

Unit

S1+M1

S2+M2

-S3+M3

M4

M5

M6

Volume (KmA3)

37800

35000

46200

43100

35700

26600

Time (Ma)

2.90

2.36

4.19

5.55

10.0

11.0

Average Erosion Rate (mm/yr)

0.067

0.077

0.059

0.046

0.023

0.015

and tectonic environment. Koppes and Hallet (2006) estimated that long term

(millennial) erosion rates may be less than 25% of the modern rates. Patagonian glaciers

show estimations of even larger differences between modern and millennial scale erosion

rates, with values in the range of 1-8% of the modern for millennial scale erosion rates.

Meaning that for temperate glaciers values on the order of 10"1 to 10° rnm/yr may be

expected when considering several advance and retreat cycles in a millennial timescale.

Therefore, values of 10"1 mm yr"1 or less might be expected for long term erosion rates of

sub-polar to polar glaciers of the Antarctic Peninsula.

Based on our estimated sediment yields for the Antarctic Peninsula continental

margin we obtained erosion rates of 0.06-0.13 mm yr"1 for the APIS. These values fall

within the expected order of magnitude. Our rates correspond closest to a study by

Bogen (1989) which measured erosion rates of 0.037-0.48 mm yr"1 for 16 glaciers in

Norway.

This represents a significant decrease in the rate of exhumation across a

longitudinal transect from Patagonian Chile to the Antarctic Peninsula, and raises

questions with respect to our understanding of glacial erosion and denudation of

mountains. The first is whether modern, short term rates of erosion by glaciers can be

extrapolated into the geologic past as an accurate estimate of continental denudation?

This has implications for models attempting to estimate the exhumation of mountains.

Second, it confirms that climate plays a major role in the effectiveness of glaciers to

erode rock and transport sediment, and begs the question of whether temperature (with an

accurate understanding of the substrate) can be used to model erosion rates.

69

CONCLUSIONS

Sediment flux to the Antarctic Peninsula continental margin has varied over time.

Accumulations on the continental shelf and rise were defined using previously interpreted

stratigraphic units and interpolated within an integrated grid of multi-national seismic and

drill core data. Isopach maps of the units reveal shifting depocenters through time. On

the shelf, the creation of Margarite Trough during unit S3 time (5.12 to -10 Ma) and the

progradation of the shelf edge during unit S2 time (2.9 to 5.12 Ma) is evident. On the

continental rise, depocenters became more focused on the tops of the bathymetric highs

through time. Units M6 and M5 (36 to 15 Ma) show filling of basement controlled mini-

basins. During units M4 and M3 deposition (15 to 5.5 Ma) there was more uniform

sedimentation across the rise. Drift growth was accelerated during units M2 and Ml

deposition (5.5 Ma to present), and this coincides with the highest rates of sedimentation

on the rise.

Our findings show that rates of accumulation on the continental shelf and rise

have increased through time up until the Late Pliocene. The highest accumulation rates

coincided with climate warming which occurred during the Early Pliocene. A slight

decrease in accumulation rates during the Late Pliocene and Quaternary occurred during

the transition of the ice sheet to a more polar (cold based) condition.

Opposing trends between the stratigraphic units of the opposing continental

shelves of the western and eastern sides of the peninsula indicate delayed ice sheet

development to the north, or increased bypass of the shelf on the western side of the

Peninsula during the last ~3 Ma years. Increased sedimentation on the continental rise of

70

the Pacific side of the peninsula during that time supports the theory that sediment bypass

of the shelf was becoming more efficient.

Denudation rates for the Antarctic Peninsula over the last 15 Ma (-0.06-0.13 mm

yr"1) show more than an order of magnitude difference compared to shorter term rates

derived from studies of temperate glaciers in similar tectonic environments such as

Patagonian Chile (0.42-1.2 mm yr"1) and Alaska (10-100 mm yr"1) (Hallet et al, 1996;

Fernandez pers. comm.). This underscores the need for further study of the relationship

between short- and long-term erosion rates by ice sheets, and the relationship of

temperature to the erosive potential and sediment transport by ice.

REFERENCES CITED

Anderson, J.B., 1999, Antarctic Marine Geology: New York, Cambridge University

Press., p. 1-286.

Anderson, J.B., Shipp, S.S., and Siringan, F.P., 1992, Preliminary seismic stratigraphy of

the northwestern Wedell Sea continental shelf, in Y. Yoshida, et al., eds., Recent

Progress in Antarctic Earth Science: Tokyo, Terra Scientific Publishing Co., p. 603-

612.

Anderson, J.B., Wellner, J.S., Wise, S.W., Bohaty, S., Manley, P.L., Smith, T., Weaver,

F., and Kulhanek, D, 2007, Seismic and chronostratigraphic results from SHALDRIL

II, northwestern Weddell Sea: U.S. Geological Survey and the National Academies,

USGS0OF-2007-1047, Short Research Paper 094.

Barker, P.F., 1982, The Cenozoic subduction history of the Pacific margin of the

71

Antarctic Peninsula: ridge crest-trench interactions: Journal of the Geologic

Society of London, v. 139, p. 787-802.

Barker, P.F., and Camerlenghi, A., 2002, Synthesis of Leg 178 results: Glacial history of

the Antarctic Peninsula from Pacific margin sediments, in Barker, P.F., et al., eds.,

Proceedings of the Ocean Drilling Program, Scientific Results: College Station, Texas,

Ocean Drilling Program, v. 178, p. 1-40.

Bart, P.J., and Anderson, J.B., 1995, Seismic record of the glacial events affecting the

Pacific margin of the northwestern Antarctic Peninsula, in Cooper, A.K., et al., eds.,

Geology and Seismic Stratigraphy of the Antarctic Margin: Washington D.C.,

American Geophysical Union, Antarctic Research Series, v. 68, p. 75-95.

Bart, P.J., and Anderson, J.B., 1996, Seismic expression of depositional sequences

associated with expansion and contraction of ice sheets on the northwestern Antarctica

Peninsula continental shelf, in De Batist, M. and Jacobs, P., eds., Geology of

Siliciclastic Shelf Seas: Oxford, UK, London, Geological Society Special Publication,

no. 117, p. 171-187.

Bart P.J., Egan, D., and Warny, S.A., 2005, Direct constraints on Antarctic Peninsula Ice

Sheet grounding events between 5.12 and 7.94 Ma: Journal of Geophysical Research, v.

110,p.F04008.

Bartek, L.R., Vail, P.R., Anderson, J.B., Emmet, P.A., and Wu, S., 1991, Effect of

Cenozoic ice sheet fluctuations in Antarctica on the Stratigraphic Signature of the

Neogene: Journal of Geophysical Research, v. 96, no. B4, p. 6753-6778.

Benn, D.I., and Evans, D.J.A., Glaciers and Glaciation: London, Oxford University Press

Inc., pp. 1-15,178-193.

Bogen, J., 1989, Glacial sediment production and development of hydro-electric power in

glacierized areas: Ann. Glaciol., v. 13, p. 6-11.

De Batist, M., Bart, P.H., Kuvaas, B., Moons, A.-M. and Miller, H., 1994, Detailed

seismic stratigraphy of Crary Fan, Weddell Sea: Terra Antartica, v. 1, no. 2, p. 321-323.

Elliot, DEL, 1988, Tectonic setting and evolution of the James Ross Basin, northern

Antarctic Peninsula, in Feldmann, R.M., and Woodburne, M.O., eds., Geology and

Paleontology of Seymour Island, Antarctic Peninsula: Boulder, Colorado, The

Geological Society of America, Inc., Memoir 169, p. 541-555.

Elverhtfi, A., and Maisey, G., 1983, Glacial erosion and morphology of the eastern and

southeastern Weddell Sea shelf, in Oliver, R.J.,et al., eds., Antarctic Earth Science:

Cambridge University Press, New York, p. 483-487.

Feldmann, R.M., and Woodburne, M.O., eds., 1988, Geology and Paleontology of

Seymour Island, Antarctic Peninsula: Boulder, Colorado, The Geological Society of

America, Inc., Memoir 169, p. 566.

Francis, J.E., Pirrie, D., and J.A. Crame, J.A., eds., 2006, Cretaceous-Tertiary high-

latitude paleoenvironments: James Ross Basin, Antarctica: London, Geological Society

Special Publication, no. 258, p. 206.

Hallet, B., Hunter, L., and Bogen, J., 1996, Rates of erosion and sediment evacuation by

glaciers: A review of field data and their implications: Global and Planetary Change, v.

12, p. 213-235.

Hassler, L.E., and Cowan, E.A., 2001, Characteristics of Ice-Rafted Pebbles from the

Continental Rise Sediment Drifts West of the Antarctic Peninsula (Sites 1095,1096,

73

and 1101), in Barker, P.F., et al., eds., Proceedings of the Ocean Drilling Program,

Scientific Results: College Station, Texas, Ocean Drilling Program, v. 178, p. 1-23.

Hayes, D.E., ed., 1991, Marine Geological and Geophysical Atlas of the Circum-

Antarctic to 30 degrees S.: Antarctic Research Series, vol. 54. American

Geophysical Union, Washington, D.C., 56 p.

Hillenbrand, CD., and Ehrmann, W., 2005, Late neogene to Quaternary environmental

changes in the Antarctic Peninsula region: evidence from drift sediments: Global

and Planetary Change, v. 45, p. 165-191.

Hillenbrand, C.-D., and Fiitterer, D.K., 2001, Neogene to Quaternary Deposition of Opal

on the Continental Rise West of the Antarctic Peninsula, ODP Leg 178, Sites 1095,

1096, and 1101, in Barker, P.F., et al., eds., Proceedings of the Ocean Drilling Program,

Scientific Results: College Station, Texas, Ocean Drilling Program, v. 178, p. 1-33.

Hollister, C. D., Craddock, C, et al., 1976, Initial Reports of the Deep Sea Drilling Project,

Volume 35: Washington (U.S. Government Printing Office), 930 p.

Iwai, M., Acton, G.D., Lazarus, D., Ostermann, L.E., and Williams, T., 2002,

Magnetobiochronologic synthesis of ODP Leg 178 rise sediments from the Pacific

Sector of the southern ocean: Sites 1095,1096, and 1101, in Barker, P.F., etal.,eds.,

Proceedings of the Ocean Drilling Program, Scientific Results: College Station, Texas,

Ocean Drilling Program, v. 178.

Johnson, G.L., Vanney, J.-R., Elverhai, A., and LaBrecque, J.L., 1981, Morphology of

the Weddell Sea and southwest Indian Ocean: Deutsche Hydrographische Zeitschrift, v.

34, p. 995-1002.

King, L.H., Rokoengen, K., Fader, G.B., and Gunleiksrud, T., 1991, Till-tongue

stratigraphy: Geological Society of America Bulletin, v. 103, p. 637-659.

Kuvaas, B., and Kristoffersen, Y., 1991, The Crary Fan: a trough mouth fan on the

Weddell Sea continental margin, Antarctica: Marine Geology, v. 97, p. 345-362.

LaBrecque, J.L., and Keller, M., 1982, A geophysical study of the Indo-Atlantic Basin, in

Craddock, C, ed., Third Symposium on Antarctic Geology and Geophysics: Madison,

WI, University of Wisconsin Press.

Larter, R.D., and Barker, P.F., 1991, Effects of ridge crest-trench interaction on

Antarctic-Phoenix spreading: forces on a young subducting plate: Journal of

Geophysical Research, v. 96, p. 19583-19607.

Miller, H., Henriet, J.P., Kaul, N., and Moons, A., 1990, A fine-scale seismic stratigraphy

of the eastern margin of the Weddell Sea, in Bleil, U., and Thiede, J., eds., Geological

History of the Polar Oceans: Arctic Versus Antarctic: Kluwer Academic Publishers,

Boston, MA, p. 131-161.

Moons, A., De Batist, M., Henriet, J.P., and Miller, H., 1992, Sequence stratigraphy of

the Crary Fan, southeastern Weddell Sea, in Yoshida, Y., et al., eds., Recent Progress in

Antarctic Science: Terra Scientific Publishing Co., Tokyo, Japan, p. 613-618.

Morris, E.M., and Vaughan, D.G., 2003, Spatial and temporal variation of surface

temperature on the Antarctic Peninsula and the limit of viability of ice shelves, in

Domack, E., et al., eds., Antarctic Peninsula Climate Variability: Washington D.C.,

American Geophysical Union, Antarctic Research Series, v. 79, p. 61-68.

Pudsey, C.J., 2001, Neogene record of Antarctic Peninsula glaciation in continental rise

sediments: ODP Leg 178, site 1095, in Barker, P.F., et al., eds., Proceedings of the

Ocean Drilling Program, Scientific Results: College Station, Texas, Ocean Drilling

Program, v. 178, p. 1-25.

75

Rebesco, M., Larter, R.D., Barker, P.R., Camerlenghi, A., and Vanneste, L.E., 1997,

History of sedimentation on the continental rise west of the Antarctic Peninsula, in.

Barker, P.F. and Cooper, A.K., eds., Geology and Seismic Stratigraphy of the Antarctic

Margin: Washington D.C., American Geophysical Union, Antarctic Research Series v..

71, p. 29-49.

Rebesco, M, Pudsey, C.J., Canals, M., Camerlenghi, A., Barker, P.F., Estrada, F., and

Giorgetti, A., 2002, Sediment drifts and deep-sea channel systems, Antarctic

Peninsula Pacific margin, in Stow, D.A.V., et al.,eds., Deep-Water Contourite Systems:

Modern Drifts and Ancient Series, Seismic and Sedimentary Characteristics, Geological

Society Memoir, v. 22, p. 353-371.

Rebesco, M., Camerlenghi, A., Geletti, R., and Canals, M., 2006, Margin architecture

reveals the transition to the modern Antarctic ice sheet ca. 3 Ma: Geology, v. 34, p. 301-

304.

Saunders, A.D., Weaver, S.D., and Tarney, J., 1982, The pattern of Antarctic Peninsula

plutonism, in Craddock, C, ed., Antarctic Geoscience: University of Wisconsin

Press, Madison, p. 305-314.

Sloan, B.J., Lawver, L.A., and Anderson, J.B., 1995, Seismic stratigraphy of the Palmer

Basin, in Cooper, A.K., et al., eds., Geology and seismic stratigraphy of the Antarctic

Margin: Washington, D.C., American Geophysical Union, Antarctic Research Series, v.

68, p. 235-260.

Smellie, J.L., Johnson, J.S., Mcintosh, W.C., Esser, R., Gudmundsson, M.T., Hambrey,

M.J., and van Wyk de Vries, B., 2008: Six million years of glacial history recorded in

76

volcanic lithofacies of the James Ross Island Volcanic Group, Antarctic Peninsula:

Palaeogeography, Palaeoclimatology, Palaeoecology, v. 260, p. 122-148.

Smith, R.T., and Anderson, J.B., submitted, Geological Society of America Bulletin

Tucholke, B.E., and Houtz, R.E., 1976, Sedimentary framework of the Bellingshausen

Basin from seismic profiler data, in CD. Hollister, and C. Craddock, eds., Initial

Reports of the Deep Sea Drilling Project: U.S. Government Printing Office,

Washington, D.C., v. 35, p. 197-228.

Volpi, V., Camerlenghi, A., Moerz, P., Gorubolo, P., Rebesco, M., and Tinivella, U.,

2002, Data Report: Physical Properties Relevant to Seismic Stratigraphic Studies,

Continental Rise Sites 1095,1096, and 1101, ODP Leg 178, Antarctic Peninsula, in

Barker, P.F., et al., eds., Proceedings of the Ocean Drilling Program, Scientific Results:

College Station, Texas, Ocean Drilling Program, v. 178, p. 1-36.

77

CHAPTER 3 - Seismic stratigraphic evidence for mixed contourite/turbidite system in the NW Weddell Sea from SHALDRIL cruises

R. Tyler Smith Rice University Earth Science Dept. 6100 Main St. MS 126 Houston, TX 77005

John B. Anderson Rice University Earth Science Dept. 6100 Main St. MS 126 Houston, TX 77005

Julia Wellner University of Houston Dept. of Earth and Atmospheric Sciences 4800 Calhoun Rd., Houston, Texas 77204

Steve Bohaty Southampton Oceanography Centre, U.K. European Way, Southampton S014 3ZH,

Fred Weaver Former Curator Antarctic Research Facility P.O. Box 3064100, Florida State University, Tallahassee, FL 32306

Woody Wise Florida State University Dept. of Geological Sciences P.Q. Box 3064100, Tallahassee, FL 32306

Patricia Manley Middlebury College Dept. of Geology McCardell Bicentennial Hall Middlebury, Vermont 05753

ABSTRACT

Seismic acquired during SHALDRIL I and II along with preliminary drill core

data reveals a mixed depositional (turbidite/contourite/hemipelagic) system located on

the Joinville Slope, northwestern Weddell Sea. Sites 12A, 5C, and 6C sampled upper

Oligocene, middle Miocene, and lower and upper Pliocene respectively.

Chronostratigraphic ages were interpreted using diatom and calcareous nannpfossil

assemblages. Four first order stratigraphic units are defined by new chronostratigraphic

constraints. Regional sedimentation rates show continuous sedimentation throughout the

Late Paleogene and Neogene. The sediment wedge is highly laminated and shows no

apparent hiatuses or large unconformities from Late Oligocene to the Lower Pliocene. A

prominent sea floor unconformity and a strongly eroded and sediment barren continental

shelf are evidence of ice grounding during the Plio-Pleistocene. The formation of a large

channel on the slope as well as the occurrence of slumps indicates a change in deposition

beginning during the Miocene and upper Pliocene Which may indicate initial ice growth

on the Joinville Plateau.

INTRODUCTION

Our knowledge of the Antarctic marine record of climate evolution is impeded by

a lack of continuous drill core from the shallow continental margins. Nearly all

continental shelves surrounding Antarctica have been gouged by ice streams and heavily

eroded by repeated cycles of ice sheet expansion and contraction. Sediments are often

barren of fossil material necessary for establishing a chronostratigraphic framework and

climate reconstruction. Treacherous icebergs and severe sea ice on some continental

shelves hinder the use of conventional drill ships close to the continent. For the past 40

years active source seismic imaging combined with gravity and piston coring methods

have predominated the research efforts. Currently, a new strategy is being applied which

allows for relatively short (<100 m) drill cores, strategically placed based upon a robust

knowledge of the seismic stratigraphy.

SHALDRIL cruises I and II successfully targeted and drilled key Tertiary

chronostratigraphic targets along the continental shelf and slope of the northwestern

79

Weddell Sea. As part of a new concept in Antarctic marine research, SHALDRIL

mounted a drill rig on the deck of the icebreaking vessel the R/V Nathaniel B. Palmer.

The goal was to use relatively short drill holes to core a transect of offlapping shelf strata

in order to sample a range of stratigraphic horizons. Drilling is necessary to penetrate

through the overlying glacial deposits and access the older underlying strata, which has

proved virtually impossible to reach with conventional piston and gravity coring

methods.

SHALDRIL identified a sediment wedge located on the Joinville Slope (Fig. 1) as

an ideal drilling location because of the wedge geometry and relative accessibility in ice-

free waters. The configuration (dip direction) of the wedge is oblique with respect to the

basement edge orientation. This is particularly significant because it produces a situation

where increasingly older stratigraphic packages are exposed along strike (Fig. 2). Thus,

sampling of the entire stratigraphic column is possible with relatively short drill cores.

Here we present results from an analysis of seismic data obtained during SHALDRIL and

identify the stratigraphic framework for the region. The strata sampled by SHALDRIL

represent important periods in the climate and cryosphere evolution of the Antarctic

Peninsula, including; the Latest Eocene/Early Oligocene, Late Oligocene, Middle

Miocene, Early Pliocene, and Pleistocene (Fig. 3). The intent of this paper is to provide a

sequence stratigraphic framework for the area for future sedimentological, geochemical

and paleontological research.

80

55°0'0"W

65°0'0"S-

JL

\ or? y

7 '110' » I

so-co-w

( •? / ' ' y la i r ^ ip iene j? ) _

/fAM4 > / /25 iSo j ) c-; 100-Kilonfiet^rs; / ' / i

H5°0'0

Figure 1 55°6i0"W 50o0'0"W

This regional map shows the location of the SHALDRILII cores as well as the cores studied by Gilbert et al. (1998). The large gray arrows indicate ocean current direction (von Gyldenfeldt et al., 2002). The dashed lines show changes in the continental shelf break. South of the large basement high the dashed lines show the progradation of the JRB shelf break with the position shown at these times: early Oligocene, Late Miocene, Early Pliocene, Late Pliocene, Pleistocene (Smith and Anderson, submitted). To the north of the basement high the rifting away of the South Orkney Plateau and the opening of the Bransfield Strait are shown with dashed lines and arrows indicating direction of the movement. The prominent channel discussed in the text is shown in blue with the direction of sediment flux to the south.

81

m

Figure 2

Seismic profiles rendered three dimensionally to demonstrate slope architecture. The dashed lines represent isochrons based on ages from the drill core.

82

Epo

ch

SH

ALD

RIL

D

rillc

ores

Ple

isto

cene

Plio

cene

k

i

1- 2- 3 4- 10-

15

- M

ioce

ne

CO

^ 2

0

CD

O)

25

<

30-1

35

-

40

-

45

50

H

55

Olig

ocen

e

Eoc

ene

• H

ole3

C

• H

ole5

D

• H

ole

6C a

nd 6

D

DH

ole1

2A

Jam

es

Ro

ss/

Se

ymo

ur

Isla

nd

Str

atig

rap

hy

Ma

jor

Clim

atic

Eve

nts

in t

he

An

tart

ica P

en

insu

la R

eg

ion

Vol

cani

cs (J

ames

R

oss

Isla

nd G

roup

)

.Coc

kbum

Fm

.

La M

eset

a Fo

rmat

ion

(hia

tus)

Cro

ss V

alle

y Fm

.

Ext

rem

e cl

imat

e va

riabi

lity

and

rapi

d ch

ange

in ic

e sh

eet e

xten

t.

Wes

t Ant

arct

ic Ic

e S

heet

as

larg

e or

la

rger

than

pre

sent

.

Mou

ntai

n gl

acia

tion

in th

e P

enin

sula

re

gion

with

rela

tivel

y w

arm

inte

rgla

cial

in

terv

als

whe

n riv

ers

drai

ned

the

regi

on.

A/o

fhor

agus

-dom

inat

ed fl

ora

in

Pen

insu

la re

gion

. Coo

ler c

limat

e.

Div

erse

ang

iosp

erm

-dom

inat

ed fl

ora

in P

enin

sula

regi

on. W

arm

tem

pera

ture

.

Age

518

0(%

o)

(Ma)

5

4 3

2 n u

™ M

. -10

- • • •

20

- • • • • 30

- * • • •

40

- • •

50

- " -eg

6 K s>

e 4» ioc

2 u 8

g^

^^

^^

^p

rT

TT

jT

Tn

ll^

^B

L*

«3|j^

^^

••I

Er

z5r ife

T

BL

. *

^T

SK

jSr"

3P

•I

SF

_]js

tet

«M

fl^

^|^

^S

20$^

^

K«M

» .s

f-

^aSi

-

o o e Eoce

_

*"Bf

e SO

HK

Sw^

j—^

yg

^L ^ i

•

g Pa

rtia

l or

Eph

emer

al

1 Fu

ll Sc

ale

and

Perm

anen

t

- " -

1 0

-1?"

s

_l

l —

T3

3" S ce-she VI

*"

F

1 |

1 1

> r+

0>

ret s » In k = 2

fe

^u

t

rtE

j* C ^

t J0>>

*a

WU

Fig

ure

3

Sum

mar

y of

sed

imen

t age

s co

llect

ed b

y SH

AL

DR

IL.

The

wid

th o

f the

age

bar

s is

to s

how

the

rang

e of

err

or a

nd n

ot th

e to

tal r

ange

sam

pled

. M

ajor

clim

atic

eve

nts

alon

g w

ith th

e de

ep-s

ea m

arin

e ox

ygen

isot

ope

reco

rd (Z

acho

s et

al.,

200

1)

illus

trate

the

dram

atic

cha

nges

occ

urrin

g du

ring

the

sam

pled

inte

rval

s. T

he J

ames

Ros

s/Se

ymou

r Isl

and

stra

tigra

phy

show

s th

e ex

tent

of t

he c

ontin

enta

l rec

ord

avai

labl

e.

00

BACKGROUND

The study area (Fig. 1) is located east of Joinville Island which we will hereafter

refer to as the Joinville Plateau and the Joinville Slope. Seismic records reveal that the

Joinville Plateau is composed of acoustic basement that has been eroded at the top. The

Joinville Slope is composed of a sedimentary succession that onlaps the basement edge of

the Joinville Plateau. The Joinville Slope is bounded to the northeast by an E-W

elongated morphological high, which is called by several names including the South

Powell Ridge or the Joinville Ridge. The ridge narrows and deepens as it extends

eastward until 49°30'W where it slopes down to the basin floor (Maldonado et al., 1998).

The ridge is composed of continental blocks, most likely of metamorphic composition as

evidenced by low magnetic anomalies (Maldonado et al., 1998; Surinach et al., 1997). It

is considered to be the eastward extension of the Joinville Plateau basement high. The

southern flank of the Joinville Ridge, eastward of Joinville Slope, is occupied by a

southward dipping sedimentary succession that is generally thinner than the strata of the

Joinville slope (Maldonado et al., 2005). To the southwest, the study area is bounded by

the James Ross Basin, which contains a thick stratigraphic succession that is bounded to

the north by the southern basement rim of the Joinville Plateau (Anderson et al., 1991).

Modern oceanographic conditions of the Joinville Slope are characterized by

strong contour currents generated by the cyclonic motion of the Weddell Gyre (Fahrbach

et al., 1995; von Gyldenfeldt et al., 2002), (Fig. 1). Currents diverge in the region with a

surface current veering to the west across the plateau and encircling the upper tip of the

Antarctic Peninsula into the Bransfield Strait. Another current diverges from the Weddell

Gyre and flows eastward and then northward into the Powell Basin.

84

Tectonically, the region is part of a back-arc basin which formed in response to

subduction of the Phoenix and Aluk plates on the Pacific side of the Antarctic Peninsula

during the Cretaceous (Elliot, 1988). Additionally, to the northeast of Joinville Plateau,

the South Orkney Plateau rifted away from the Antarctic Peninsula during the Oligocene

forming the Powell Basin (Fig. 1) (Eagles and Livermore, 2002). A fault zone within the

Late Eocene-Oligocene stratigraphic section of the James Ross Basin shows evidence of

transtensiond movement associated wim mis rifting event (Smith and Anderson,

submitted).

A single regional northwest-southeast multi-channel seismic reflection profile

across the Joinville Plateau and Slope was collected in 1985 by the R.R.S. Discovery.

Barker and Lonsdale (1991) reported sediment thicknesses of 4.7-5.0 seconds twt on the

continental rise and seismic stratigraphic indications of a sediment drift on the slope.

Gilbert et al. (1998) analyzed several gravity and kasten cores (Fig. 1) from the

continental slope along with CTD casts and 3.5 kHz seismic data. Their results show

compositional and textural cyclicity. Two distinct lithologies were recognized: 1) fine­

grained sands and silty sands, with common foraminifera and some diatoms and

radiolaria, and 2) silty-clayey sands with little to no biogenic material. They concluded

that the sandy, biogenic-rich units represent times of strong bottom currents during

glacial minima, whereas fossil barren, finer-grained units indicate weak bottom currents

that existed during glacial maxima.. Their analysis also showed the dominant modern

sedimentary processes to be controlled by: 1) ice rafting; 2) turbidites and other sediment

gravity flows; and 3) resuspension of slope sediments by bottom currents. Minor

biogenic material, volcanic ash, and wind-blown dust were also identified.

85

The only published information about the older strata of the region comes from

initial analyses of DSDP Leg 113 and from drill core collected in the James Ross Basin

(JRB) during SHALDRIL I. The DSDP Leg 113 results include three sites drilled on the

southern flank of the South Orkney Plateau (Barker et al., 1988). The scientific objective

of these sites was to obtain a sedimentary record of changes in late Paleogene and

Neogene water masses located in the northern portion of the the Weddell Gyre. ODP Leg

178 sites 695,696,697 form a transect that sampled continental shelf and slope

sediments ranging back to Eocene age. The deposits consist of mixed terrigenous,

volcanic, and pelagic sediments with a biologic component consisting mostly of diatoms,

but also silicoflagellates, radiolarians, and sponges. Site 697 showed evidence of

cyclicity at the lamina, layer, and unit scales. All sites indicate high variability in

sedimentation rates, which likely result from variations in bottom current strength

(Barker et al, 1988). Little information about the early development of the Antarctic

Peninsula Ice Sheet (APIS) was derived from these sites.

Seismic records from southwestern portion of the JRB show thick subglacial

deposits bounded by glacial unconformities (Anderson et al., 1991; Sloan et al., 1993).

Based on a recent seismic stratigraphic age model for the basin, glaciation in the southern

portion of the basin occurred during the latest Middle Miocene/early Late Miocene, with

expansion of the APIS onto the continental shelf in the northern portion of the basin by

Late Miocene time (Smith and Anderson, submitted). Progradation of the JRB shelf

occurred from Late Eocene to present. Prominent onlap surfaces within the Late Eocene-

Middle Miocene strata are interpreted to represent eustatic events that can be used for

regional stratigraphic correlation (Smith and Anderson, submitted).

86

DATA AND METHODS

A total of ~112 meters of drill core were recovered during SHALDRILII. All but

4 of the target locations were drilled within the JRB or the adjoining area named the

Joinville Slope. The sediments were sampled using a push core system which averaged

<60% core recovery overall. Most holes were drilled using seawater to lubricate the hole,

however, when drilling sand-rich layers guar gum was added to make a more viscous

drilling mud that was biodegradable and safe to the environment. All the cores were

sampled in 3 meter sections and later sectioned and split into 1.5 meter archive and

sample halves. Each core (before splitting) was run through the Multi-Sensor Gore

Logger (MSCL) which took multiple simultaneous measurements of density/porosity, p-

wave velocity, magnetic susceptibility, and electrical resistivity. Photographs were taken

and descriptions made of each archive half. On board, sample halves were described and

sampled for biostratigraphic analysis. The cores are housed at the Antarctic Research

Facility (ARF) located at Florida State University. During a recent field season (2007

NBP Palmer cruise) we attempted to acquire jumbo piston cores to obtain additional

material, however, we were not successful in recovering core.

Several single-channel seismic lines were shot during SHALDRIL II in

conjunction with drilling operations as a means of identifying additional targets in more

ice-free waters. The data were acquired with a 210 cm3 air gun source and a shot spacing

of 5 seconds. We recorded a 2 second time interval using an ITI hydrophone streamer.

A band-pass filter was applied to the digital data with a range of 4 to 100 Hz, along with

an overall gain function.

87

SHALDRIL drill core chronostratigraphy was tied to the seismic to create

isochron horizons and then extrapolated between the lines by a nearest neighbor

interpolation function to create regional surfaces. Isopach maps were then created based

on the regional surfaces. A minimum velocity function was developed by taking the

average interval velocity values of the SHALDRIL cores recorded by the multi-sensor

core logger and applying a linear relationship between the velocity values with respect to

stratigraphic depth. The velocity values were also compared to regional sonobuoy

velocity data (Hayes, 1991). Using the velocity function the isopach twt thickness values

were converted to depth.

RESULTS

Stratigraphic Architecture and Seismic Facies

The Joinville Plateau is virtually barren of sediment cover. Southeast of the

Plateau there is a thick (greater than 1.5 seconds twt), southeastward dipping wedge of

strata that onlaps at high angles acoustic basement to the north and west (Fig. 4). The

acoustic basement is characterized by a high amplitude, chaotic reflection pattern with

strong diffractions (Fig. 5a). Exposed basement shows very low bathymetric relief, in

contrast to the high relief exhibited along the flanks of the plateau (Fig. 5b). Locally,

coherent reflections occur within the acoustic basement indicating a possible

sedimentary/meta-sedimentary or volcanic composition.

The sedimentary section dips at low angle (<4°) and is highly aggradational. No

definable shelf break is present at the seafloor; rather the geometries of the wedge

indicate a ramp depositional environment. The stacking pattern shows fairly consistent

88

NBP0602ALine10

J ^ ^ ^ v A 7 S = ^

Figure 4

Seismic line NBP0602A-10 from SHALDRIL showing cores locations and overall seismic facies and architecture.

5 km

Figure 5

a) Example of the bubble pulse at the seafloor and the underlying basement facies.

b) Example of a basement high buried by sediments. Sediments drape the upper portion of the basement high and show very little thickness change.

90

lateral thicknesses. Truncation of sequences occurs primarily at a seafloor unconformity

or around the large channel identified in the southwestern portion of the study area.

Overall the sediment wedge is relatively undeformed. The deepest strata show evidence

of some faulting adjacent to the basement contacts, most likely associated with basement

movement during the rifting away of the South Orkney Plateau.

The overall seismic character of the stratigraphic succession is highly laminated,

with individual reflections showing little thickness variations and with alternating units

composed of high and low amplitude reflections (Fig. 4). Laminated sediments are

correlated across the area with confidence due to nearly identical reflection stacking

thicknesses (Fig. 5b). Individual reflections onlap basement at high angles. Condensed

sections evidenced by onlap surfaces exist within the thinning stratigraphic section as it

nears the basement contact. These onlap surfaces are not regionally correlative and

therefore represent localized tWnning most likely due to tectonic movement of the

basement rather than eustasy. Near the top of the stratigraphic section, just below the

seafloor unconformity, a regional unconformity truncates the uppermost topset beds.

Above this unconformity exists a chaotic facies. A resonant bubble pulse completely

obscures the uppermost 30 msec and partially obscures up to 60 msec (Fig. 5a).

Although some high amplitude reflectors approaching the seafloor can be identified

within the partially obscured region from 30 to 60 msec TWT. Drill core, piston cores

and grab samples taken in the region indicate a lag deposit at the seafloor causing a hard

reflection bottom.

Within the lowermost part of the stratigraphic section, a localized but distinct

wavy seismic facies exists near the point where reflections onlap basement (Figs. 4, 6a).

91

5 km

Figure 6

a) Example of wavy seismic facies. Possible evidence of asymmetric sediment waves. Theoretical direction of flow is shown by the arrow at the top based upon contour current deposition.

b) Cross section of the channel. The channel formed during JP3 and underwent several morphological changes during JP2.

c) An example of a slump occurring at the JP3/JP2 bounding isochron. Slump geometry is highlighted.

92

The wave forms have an amplitude of 10-20 msec twt and a wavelength of ~2 km. Wave

geometry is asymmetric with the steeper side of the waves facing to the north. Above

and below the wavy fades are flat, laminated reflections.

A prominent canyon exists in the southwestern portion of the study area and

generally follows the contact between sedimentary strata and basement (Fig. 1). Gilbert

et al. (1998) first described this feature to the south of our study where it no longer

parallels the basement contact. They interpreted it to be of turbiditic origin, citing

evidence from 3.5 kHz seismic profiles and cored sediments from the region. Associated

with the canyon is a positive bathymetric feature that they interpreted as a sediment drift.

Our seismic lines cross the canyon in several (5) locations and show that the canyon has

been filled with ~400 ms of sediment (Fig. 6b). High-angle onlap of canyon walls exists

only in the base of the canyon (Fig. 6b).

The modern canyon geometry changes with position on the slope. Up dip, in the

northern part of the wedge, the canyon is shallow and relatively narrow. Basinward it

widens and deepens. The canyon geometry also changes at depth within the stratigraphic

section. In the deeper part of the section the canyon is broad and shallow, approximately

10-15 km wide and 30-50 m deep. In the middle of the section it narrows and deepens,

but then becomes broad and shallow once more near the top of the section. The canyon

fill is characterized by low amplitude, chaotic/transparent reflections in the narrow and

deep portions of the canyon and high amplitude, semi-continuous reflections in the broad

and shallow portions of the canyon.

In the upper part of the section there is evidence for localized slumping. Slumps

consist of chaotic units within strongly laminated packages and with surfaces that exhibit

93

tens of meters relief (Fig. 6c). In some cases the internal structure of the slumps show

preservation of original stratigraphy, as in the example shown in Figure 6c.

Ages of Drill Core

The following is a summary of SHALDRIL preliminary results, for a detailed

summary of methods and descriptions see SHALRILII Cruise Report (2006). The drill

site locations were chosen along two seismic lines shot during SHALDRIL II (Figs. 1,4).

Site 5D was the first attempt to sample in the area. It was chosen in an attempt to sample

a prominent high amplitude and continuous reflection which terminates near the seafloor.

There is no faulting within the targeted stratigraphic section (Figs. .1,4), which gave a

preliminary indication during SHALDRIL that the strata was at least younger than

Eocene/Early Oligocene age and post-dated rifting away of the South Orkney Plateau.

Diatom analysis from smear slides of hole 5D subdivides the age of sediments

into 4 diatom zones, representing the Pleistocene, Lower Pliocene, and Middle Miocene

(Fig. 7). The top of the core to 8.8 mbsf is assigned to the Thalassiosira lentiginosa-

Fragilariopsis kerguelensis "b" Zone which constrains the age to < 140 ka. The presence

of Actinocyclus actinochilus, Fragilariopsis curta, F. cylindrus, F. vanheurckii, and

Thalassiosira antarctica indicates the predominance of a sea-ice environment. Diatoms

are well preserved and moderately abundant.

The interval from 8.85-16.25 mbsf is assigned to the Thalassiosira inura Zone

due to the presence of T. inura and the absence of Fragilariopsis barronii. Diatoms

within the interval are poorly preserved, highly fragmented and in relatively low

abundances therefore the absence of Fragilariopsis barronii needs greater confirmation

94

NBP0602ASite: 5 Hole: D WD: 506m

Top:0mbsf

I E,

Dep

th I

£ IR

eco

very

(

10—

20-

30—

40-

1%

*«

£ IR

eco

very

(i grain _

D size 1 o S Q.O

S £ h

lml S R - , ^ ^ 1

«| "S

&'

?RS

—

'

• - »

• Z. mT 4

• •

*

*-:•

JJ.*;:5 !•.• • . [_• r

1 ?-••-• ^m-& &,*

2 | I . **«

*'•--I K i L l i

1 o

1 t l ! ^ a i J

Lithoiogic

description

^ 4>

5"

lb

II

III

IV

"v"

VI

Very Dark Greenish Gray to Very Dari Gray <GLEV1 3fW• GLEYi 3/N) Pebbly Muddy Sand (-Ji to. 2cm in diameter) and rare macro fossils, interbedded deaner sands and gravel layers

lb Sandy Mud, same color as above. more stiffand without pebbles

_ » _ 1 _ J — ! _ ( « . ! _ ! _

Olive (5Y 4/3) to Black (5Y 2.5/1) Pebbly Sandy Mud, poorly sorted, fine to medium sands sub-angular grains, no visible macro fossils

Very Park Gray {SY 3/1J Pebbly Muddy Sand, fine sand dominant poorly sorted. pebbles and sands are sub-angular to surrounded,smells of hydrogen sulfide (orgahk rich), no visible macro fossils, very firm to stiff

Greenish Black (GLEY1 3/1OY) Sandy. Mud, very few pebbles and very little very fine sand, no visible macro fossils, very firm

Dark Gray EGLEY14/NJ Homogeneous Gray Clay, very rare pebWes.no layering, no bioturbation, no visible macro fossils firm to very firm

Greenish Black (GLEY13/10Y) Pebbly Sandy Mud, fine to very fine sand, no visible macro fossils, very firm to stiff

<0ll4]

4.4

to

S.1

— — . 11.8

to

12.1

12.1

to

12.7

Time-Rock

Unit

' "upper"/

Pleistocene-

Holocene

Lower

Pliocene

. _ _ _ .

Middle

Miocene

Diatom

Zone

i T.lentiginoio-y

Rkerguelensis •b"

Thalassloiira

inura

._ — ___ D.ovata-

N.denticuloides

Dentiqiiopsis

dimorpha

(tegnetfe SiafcprititSty

en £

*~ i

J > r "•

f

0 .

0

V f » vw V •»

SB

•

Electric Resistivity (oftm-m).

0 . - 3

9fjf' . ~

\<v

%$Z •••

V - •

l e f e •

—

«fl&

^E* -}£•"" l^K^J^y.

^"fcK,1^!.

j fcT-•^pr -

• • • -

-

Density (gm/cc)

s.o •

' * '

•

• ^

4 »

3B

a * .

"3'-—if.

—ajx • ^ ^ F ^

^ i ' *-

3

* -r

* • —

• •

,

•

•

, " —

,

-.

• •

-

-

•

Figure 7 Core description sheet for site 5D.

95

The wave forms have an amplitude of 10-20 msec twt and a wavelength of -2 km. Wave

geometry is asymmetric with the steeper side of the waves facing to the north. Above

and below the wavy facies are flat, laminated reflections.

A prominent canyon exists in the southwestern portion of the study area and

generally follows the contact between sedimentary strata and basement (Fig. 1). Gilbert

et al. (1998) first described this feature to the south of our study where it no longer

parallels the basement contact. They interpreted it to be of turbiditic origin, citing

evidence from 3.5 kHz seismic profiles and cored sediments from the region. Associated

with the canyon is a positive Pliocene age with the bottom of the hole possibly reaching

Upper Miocene sediments. The hole is subdivided into 3 intervals based the diatom

analysis (Fig. 8).

The core was washed down to 4.0 mbsf. The first interval spans -4.0-16.0 mbsf

and contains few to common, moderately well preserved diatoms. Several samples are

very rich in diatomaceous material, however, there is a high degree of fragmentation. A

broad range of early Pliocene is assigned to the interval, but by using the Southern Ocean

zonal scheme for the Pliocene this interval can be assigned to the F. barronii Zone based

upon the presence of F. barronii and the absence of Fragilariopsis interfrigidaria. This

gives an age calibration of between -3.8 and 4.4 Ma for the interval.

The second interval from -16.0-17.0 mbsf is tentatively assigned the T. inura

Zone based on the presence of T. inura and the absence of F. barronii. This places the

age of these sediments at between-4.4 and 5.1 Ma.

The lowermost interval of hole 6C between -17.0-20.5 mbsf contains very rare

and poorly preserved diatoms. No zonal assignment was given to this interval, however,

96

NBP0602ASite: 6 Hole: C WD: 532m

Top: 0 mbsf

1 HS i grain size

Hi

Lithologic description

Time-Rock Unit

Diatom Zone

Susceptibility

en

Electric

Resistivity

( o h m - m )

Density (gm/cc)

1.5

30-

I40-I

*i

- 1

Dark Greenish Gray (GLEY 1 4/1 5GY) Pebbly Sand with a few clayey laminations, no bioturbatfbn, no macro fossils

Pebbly Muddy Sand, with fine rounded to subrpunded sand grains, variable color (Dark Gray to Very Dark Greenish Gray), no bioturbation, no macro fossils, firm

3&

t o

4 . 4

Lower

Pliocene FragHariopsis

barronii

4 .4 t o 5.1

Thalassiosira inura

m *

Figure 8 Core description sheet for site 6C.

the presence of D. delicata and T. torokina fragments constrains the age to the late

Miocene to early Pliocene. Future higher density sampling may identify higher diatom

content sections within this interval and thus, better constrain age estimates.

Site 12A was identified by mapping the deepest continuous reflections along

strike to seismic line 10 where these strata are located within a few meters of the seafloor

(Fig. 4). Based on our seismic stratigraphic correlation, the strata sampled at site 12A are

stratigraphically located -200 meters below the strata sampled at Site 5D.

Both diatom and calcareous nannofossil assemblages were analyzed from smear

slides from site 12 A. Diatoms from the uppermost 15 cm of core contains pre-

Quaternary assemblages showing modern sediment deposition. Below modern sediments

the rest of the core spanning the interval 0.15-7.2 mbsf contains an Oligocene diatom

assemblage with rare to few diatoms which are poorly to moderately preserved. An age

range of-28.6-24.0 Ma is estimated for this section because of the presence of Cavitatus

jouseanus, C. rectus, Kisseleviella cicatricata, and K. tricoronata (See Scherer et al.,

2000; Wilson et al., 2002; Olney et al., 2005 for age classifications).

Calcareous nannofossils from site 12A were also analyzed. While some samples

were barren, many contain rare to few, moderately preserved calcareous nannofossils. A

precise age calculation is impossible due to the limited diversity of species, however, the

presence of D. bisecta which has a last occurrence at the end of the Oligocene (-23.8-

24.2 Ma), indicates that the interval from 0.15-7.2 mbsf is late Oligocene or older in age.

98

Sediment Description

Below is a brief overview of the lithologic descriptions from SHALDRIL

sediment cores 5D, 6C, and 12A (SHALDRIL II Cruise Report, 2006). A summary of

cores 5D and 6C is presented along with a biostratigraphic summary in Figures 7 and 8.

Site 5D (Summary in Fig. 7)

Hole NBP0602A-5D cored an interval between 8.0-31.4 mbsf. The first 8 meters

Were washed in order to drill through the surface sediments faster and spud the drill string

into the bottom. The cored sediments are subdivided into 5 lithologic units. Unit 1 (8.0-

12.15 mbsf) consists of an uppermost diamicton layer with an underlying stiff muddy

sand with no pebbles. Unit 2 (12.15-15.0 mbsf) is a firm sandy, silty, diatomaceous mud

which smelled strongly of hydrogen sulfide when opening the core. Unit 3 (15.0-22.18

mbsf) is composed almost entirely of stiff, pebbly, muddy fine to medium sand which

also smelled of hydrogen sulfide upon opening. Within this unit diatom biostratigraphy

confirms the existence of a condensed section from 16.24-16.28 mbsf where a slight

color change exists and the sand has a lower clay component. This thin interval

represents the contact between lower Pliocene and Middle Miocene sediments. Unit 4

(22.18-25.07 mbsf) consists of an upper silty mud with no sand grains, and a lower

homogenous clay unit with no sand or silt. Unit 5 (25.07-31.40 mbsf) becomes sandier

again, the upper portion being a stiff, sandy, silty mud with pebbles and the lower portion

changing to a sandy diatomaceous mud. Unit 5 also notes the return of the hydrogen

sulfide smell mentioned also in units 2 and 3.

99

Sites 6C (Summary in Fig. 8)

Site NBP0602A-6C was drilled to a depth of 20.5 mbsf. The first 4 mbsf were

washed, however a small amount of cobbles and gravel with a small amount of mud was

recovered. The core was not subdivided into units, but the upper section (4.0-8.0 mbsf)

recovered a pebbly sand with few clay laminations and no evidence for bioturbation or

macrofossils. The lower section is noticeably firmer and is composed of a pebbly muddy

sand with the larger grains showing evidence of greater rounding. No macrofossils were

found along with no indications of bioturbation.

Sitel2

Site NBP0602A-12 was drilled to a depth of 7.2 mbsf. The hole is subdivided

into 6 lithologic units. Unit 1 (0-0.14 mbsf) consists of soft sandy mud with coarse

pebbles. At the base of this unit exists a 1 mm thick lamina of medium sand with well-

rounded and well-sorted grains. Unit 2 (0.14-2.90 mbsf) is a sandy mud with no pebbles,

which is considerably firmer than the overlying Unit 1. Unit 3 (2.90-3.76 mbsf) is a

muddy, fine to very fine sand. Unit 4 (3.76-4.20 mbsf) has an upper section (3.76-4.00

mbsf) which is a muddy fine sand, and a lower section (4.00-4.20 mbsf) which is a sandy

(fine grained) mud. Small clay laminations occur throughout. Unit 5 (4.20-5.35 mbsf) is

composed of muddy fine sand. Clay lenses appear throughout the unit. There are also

shell fragments present. This is the only unit within the core to show the presence of

macrofossils. Unit 6 (5.35-7.20 mbsf) is a sandy (fine grained ) mud. One large cobble

was recovered near the base of the unit.

100

DISCUSSION

Mixed Depostional System

The Joinville Slope stratigraphy shows evidence of a mixed

contourite/turbidite/hemipelagic depostional system. Below we detail the various

evidences for each.

Contour Current Deposition

First recognized as an area of strong contour current influence by Hollister and

Elder (1969) using bottom photographs, the modern northwestern Weddell Sea

continental margin is clearly influenced by contour currents. Our interpretation of

contourite deposition is based on the following diagnostic criteria from Faugeres et al.

(1999). The principle identifiable features include four main types of drift geometry;

sheets, elongate mounds, channel-related drifts, and confined drifts. These geometries

are not unique to only contourite deposits, therefore additional features are necessary for

identification. Contourite deposits generally have abundant discontinuities marked by

high-amplitude, continuous reflections that correlate across the deposit. Depositional

units are lenticular in shape and convex upward. The deposits tend to prograde in the

direction of current flow or oblique to it. In some cases lateral migration of the deposits

can distinguish contour vs. turbidite deposits based upon direction of flow and

hemisphere location. Seismic facies are also non-unique to contourites but include wavy

reflections, and well stratified, horizontal, parallel, high amplitude reflections.

Unlike other Antarctic margins, the Joinville Slope shows little evidence for drift

formation. The bathymetry of the slope is a ramp geometry with a fairly smooth dipping

101

seafloor surface. Sedimentation throughout the area is nearly uniform and sheet-like,

which indicates a regional depositional mechanism such as hemipelagic settling or

oceanographic currents. The single anomaly to these sheet-like deposits is the large

channel identified along the southwestern basement contact. This may represent a moat

channel associated with initial drift formation, however, moat channels are usually non-

depositional features rather than erosional (Faugeres et al., 1999). The characteristics of

the channel show periods of deep incision and erosion which indicate a turbiditic origin

rather than scouring by contour currents.

Within the deepest part of the stratigraphic section, a wavy seismic facies (Fig.

6a) may be indicative of sediment waves caused by contour currents. The geometry of

the waves would seem to indicate a bottom current rather than turbiditic origin based

upon the criteria set forth by Wynn and Stow (2002). Specifically, higher wave

irregularity, crest alignment in the direction of modern oceanographic currents and

oblique to the regional slope, and the lack of evidence of decreasing wave size downslope

are evidence of bottom currents. Alternatively, Rebesco and Stow (2001) reported that

sediment waves are not necessarily diagnostic of contourites. They identified slope

deformation, sediment creep, and large-scale water escape as possible mechanisms that

can create ahummocky seismic facies that can be misinterpreted as sediment waves.

To the east of our survey area, Maldonado et al. (2005) recognized 7 styles of

contourite drift morphology in the NW Weddell Sea close to our study area. The

morphology of the different drifts is thought to be caused by basin physiography

(Maldonado et al., 2005). Our analysis of the sediment wedge shows that drifts never

102

fully formed on the Joinville Slope, however, the large-scale morphology most closely

matches the giant elongated drift type identified by Rebesco and Stow (2001).

Additional evidence for contour currents is found in the preliminary work done on

the SHALDRIL cores. Lithologic descriptions of the cores note an abundance of very

fine to fine sand indicating considerable sorting either before deposition or winnowing of

sediments post deposition. The physical property measurements taken on the cores show

little downhole increase, especially in density (Anderson et al., 2007). This is caused

when fines have been winnowed away by currents and the residual larger grains are held

apart by grain to grain contacts which prevent downhole compaction. On the other hand,

many samples from site 6C (Early Pliocene age) contain abundant diatoms, approaching

diatomaceous sand, the preservation of which argues against strong winnowing.

Turbidite Current Deposition

One of the more distinguishing features of contourite vs. turbidites deposits is the

orientation of erosional scours. Contour currents often form moat channels which run

parallel to basement highs and scour large troughs along the rim of the basin. Opposite

of this, turbidites usually create troughs which extend downslope rather than parallel to

bathymetric highs. Additionally, turbidites create levee deposits composed of fine

grained material ejected during the flow and deposited on either side of the channel.

Contourites, on the other hand, often form assymetrical or even single lobes adjacent to

their channels (Faugeres et al., 1999). The seismic records show characteristics of

contourite and turbidite influence. The prominent canyon identified (Fig. 1,6b) follows

the basement contact in the north, but then extends downslope farther south. Gilbert et al.

103

(1998) interpreted some features as turbiditic in origin based on its morphology and deep

erosional incision. Our interpretation is that the channel may have been influenced by

different depositional processes at different times. The modern channel is broad and

shallow, showing non-deposition and relatively little recent incision (Fig. 6b). However,

deeper in the section the canyon is deeply incised. This is shown by the abundant

truncation of reflectors from previous channel deposits. During periods of deep incision

the canyon was filled with low-amplitude, highly chaotic and nearly transparent seismic

fades, whereas, during less active periods the canyon was filled with more laminated and

high amplitude seismic facies, similar to the seismic facies outside but adjacent to the

channel.

Other studies confirm the possibility of alternating depositional processes through

time within the Weddell Sea. South of the study area on the continental rise of the

Weddell Sea, Michels et al. (2001) showed seismic evidence that sedimentation

approximately during the Late Miocene was strongly influenced by channelized currents.

Pudsey and Howe (1988) identified through gravity core analyses that the Weddell Gyre

has been the dominant depositional force throughout the last few glacial and interglacial

cycles, but circulation was sluggish during the Late Pliocene to Early Pleistocene.

The most conclusive evidence for sediment mass movement is the occurence of

slumps in the upper stratigraphic section (Fig. 6c). No evidence for slumping occurs in

pre- Middle Miocene strata. The highest concentration of slumps occurs in the upper part

of the section. The slumps appear to be small, isolated and localized, however their

presence indicates mass movement.

104

Hemipelagic Sedimentation

In general, hemipelagic deposits are characterized in seismic records by consistent

lateral thickness of reflections and by draping over basement highs (Fig. 5b). Also the

abundance of diatoms within layers of otherwise fine sediments indicates hemipegic

sedimentation.

Sedimentation Rates

The Joinville Slope sedimentary wedge can be subdivided into 4

chronostratigraphic packages based upon the SHALDRIL drill cores. Each drill core is

correlated to the seismic data to create isochrons and used as bounding surfaces for the

stratigraphic units. We have assigned the following nomenclature to subdivide the

stratigraphic column: JS1, the uppermost unit is bounded by the seafloor at the top and

the Early Pliocene isochron at the base; JS2, bounded by the Early Pliocene isochron at

the top and Middle Miocene isochron at the base, JS3, bounded by the Middle Miocene

isochron at the top and Late Oligocene isochron at the base; and JS4, bounded by the

Late Oligocene isochron at the top and the top of basement at the base. Table 1 shows

the regional average sedimentation rates for the area covered by the SHALDRIL seismic

survey of the Joinville Slope. The regional average sedimentation rate is determined by

taking the average isopach thickness of each unit. This method will produce lower

sedimentation rates overall because it includes areas of stratigraphic thinning and

condensed packages where they onlap basement. The numbers in Table 1 show the

highest sedimentation rates of-2.4 cm/kyr during the Late Oligocene to Middle Miocene,

however, the variation in sedimentation rate is not large (<1 cm/kyr) and shows that

105

Table 1

Unit JP1 JP2 JP3

Average thickness

(TWT) 0.100 0.120 0.295

Unit Velocity

1620 1780 1870

Average Thickness

(Depth) 81

107 276

Time Interval

4750 7650

11600

Sedimentation Rate (cm/kyr)

1.71 1.40 2.38

Table 2

Unit

JP1

JP2

JP3

Max Thickness (TWT)

0.175

0.35

0.67

Unit Velocity

1620

1780

1870

Max Thickness (Depth)

140

310

625

Time Interval

4750

7650

11600

Sedimentation Rate (cm/kyr)

2.95

4.05

5.39

106

sedimentation on the Joinville Slope has remained fairly constant. Even the maximum

sedimentation rates show relatively little variance over time (Table 2). This implies that,

despite evidence for alternating depositional processes, the flux of sediment to the slope

remained relatively constant.

The sedimentation rates on the Joinville Slope (average rates of ~1.4 to 2.4

cm/kyr; and highest rates of ~3 to 5.4 cm/kyr) are similar to published sedimentation

rates on both sides of the Antarctic Peninsula. Michels et al. (2001) reported

sedimentation rates on the channel levees/drift bodies in the western Weddell Sea as 0.7

to 2.8 cm/ka for interglacial periods and 1.6 to 5.5 cm/ka for glacial periods. Pudsey and

Camerlenghi (1988) reported sedimentation rates of drift bodies located on the

continental rise of the Pacific margin of the Antarctic Peninsula as 3.0 to 5.5 cm/ka.

Development of the Ice Sheet

The Joinville Slope provides dual methods for studying the evolution of the

Antarctic Peninsula Ice Sheet (APIS). Firstly, the stratigraphic column shows continuous

sedimentation from the Late Oligocene to Lower Pliocene. This is an anomaly among

Antarctic margins which generally have low preservation of late Paleogene and early

Neogene sediments due to deep erosion of the margin by grounded ice sheets. The

SHALDRIL cores show these sediments to be rich in biological material which can be

used to develop paleoclimate reconstructions for the Antarctic Peninsula.

Secondly, seismic stratigraphic interpretation in combination with drill core

shows the expansion of grounded ice into the region during the Lower Pliocene. At drill

site 6C, a regional unconformity was sampled which truncates Middle Miocene

107

N

NB

P06

02A

Lin

e 9

Figu

re 9

Seis

mic

line

NB

P060

2A-9

. Sh

ows

the

flatn

ess

of th

e ex

pose

d ba

sem

ent a

nd th

e la

ck o

f sea

floor

rel

ief a

t the

bas

emen

t to

sedi

men

t con

tact

.

sediments with Lower Pliocene sediments above. This unconformity represents an early

expansion and grounding of the ice sheet on the Joinville Slope. The flat seafioor

morphology is additional evidence that ice sheets have expanded and grounded on the

Joinville Slope. Figure 9 shows the horizontally flat surface of exposed basement of the

Joinville Plateau and the lack of bathymetric relief at the contact between basement and

soft sediments.

The unconformity sampled at site 6C most likely represents a thinner and shorter

duration grounding event when the ice sheet was just establishing itself in the

northernmost Peninsula. The abundance of diatom group Stephanopyxis spp. within the

core from site 6C during the Early Pliocene indicates the presence of open- water neritic

areas during the summer.

Initial ice growth in the region would have begun much early than the initial

grounding events on the Joinville Slope, possibly as early as the Late Miocene. To the

southwest, seismic interpretations of the JRB shows that ice grounded out to the

northernmost shelf edge by Late Miocene time (Smith and Anderson, submitted). This

roughly coincides with the development of mass wasting features on the Joinville Slope.

Early temperate glaciers may have been the cause of shifts in the depositional regime and

the development of gravity flows and turbidites on the margin.

CONCLUSIONS

The Joinville Slope has great potential for yielding a high resolution climate and

ice sheet history for the Antarctic Peninsula. The seismic stratigraphy presented here

shows a sediment wedge with no apparent hiatuses or large unconformities from the Late

Oligocene to the Lower Pliocene. The regional sedimentation rates calculated using both

the drill core and regional seismic lines show fairly constant rates of sedimentation

throughout the Late Paleogene and Neogene. Drill core samples recovered during

SHALDRIL show an abundance of biological material with which to date and conduct

paleoclimate reconstructions for the Antarctic Peninsula. This is significant because it is

the only continuous record of sedimentation during this time frame on the peninsula. The

Joinville Slope is also ideal for shallow drilling operations because of the configuration

of the sediment wedge geometry oblique to the basement contact. This allows for

sampling of the Late Oligocene through modern sediments nearly at the surface along a

lateral transect of the sediment/basement contact. Unfortunately, an armored surface at

the seafloor most likely due to current winnowing makes it unable to sample older

sediments using conventional gravity coring methods. The Joinville Slope may be the

only accessible location whereby a continuous Antarctic climate record of Paleogene and

Neogene sediments is preserved.

Seismic and core evidence suggests that the Antarctic Peninsula Ice Sheet (APIS)

grounded on the Joinville Plateau and Slope by Pliocene time. Seafloor morphology also

suggests the erosion of ice sheets at the modern day surface. The timing of ice sheet

expansion is correlative with evidence from the James Ross Basin (JRB) to the southwest

which shows the ice sheet reached the northernmost peninsula by late Miocene time

(Smith and Anderson, submitted).

Sediment descriptions and seismic facies show a mixed depositional environment

with alternating periods of mass gravity/turbidite and contour current dominated

deposition. This is similar to the interpretation of modern sediments by Gilbert et al.

110

(1998). Late Oligocene and Early Miocene strata show possible evidence of possibly

greater dominance by contourite processes. The appearance of slumps and the

intensification of channel incision during the Middle Miocene to Late Pliocene is

evidence of changes in the depositional regime with a higher turbidite influence and may

reflect early ice growth on the adjacent Joinville Plateau.

Michels et al. (2001) theorize that a correlation exists between sediment flux to

the region and diminished contour current influence, based on other work in the western

Weddell Sea. Our study shows little evidence for increased sediment input, however, the

initiation of mass wasting process on the Joinville Slope may coincide with the timing of

initial ice growth on the continent. This would fit the hypothesis by Gilbert et al. (1998)

that changes in depositional dominance between turbidity currents and contour currents

occur with the expansion and retreat of the ice sheet, respectively.

REFERENCES CITED

Anderson, J.B., Shipp, S.S., and Siringan, F.P., 1992, Preliminary seismic stratigraphy of

the northwestern Wedell Sea continental shelf, in Y. Yoshida, et al., eds., Recent

Progress in Antarctic Earth Science: Tokyo, Terra Scientific Publishing Co., p. 603-

612.

Anderson, J.B., Wellner, J.S., Wise, S.W., Bohaty, S., Manley, P.L., Smith, T., Weaver,

F., and Kulhanek, D, 2007, Seismic and chronostratigraphic results from SHALDRIL

II, northwestern Weddell Sea: U.S. Geological Survey and the National Academies,

USGSOOF-2007-1047, Short Research Paper 094.

I l l

Barker, P.E, Kennett, J.P., et al., 1988. Proc. ODP, Init. Repts., 113: College Station, TX

(Ocean Drilling Program), doi: 10.2973/odp.proc.ir. 113.1988.

Barker, P.F., and Lonsdale, M.J., 1991, A multichannel seismic profile across the

Weddell Sea margin of the Antarctic Peninsula: regional tectonic implications, in

Thomson, J. A., et al., eds., Geological Evolution of Antarctica: Cambridge, Cambridge

University Press, p. 237-241.

Eagles, G., and Livermore, R.A., 2002, Opening history of Powell Basin, Antarctic

Peninsula: Marine Geology, v. 185, p. 195-205.

Elliot, D.H., 1988, Tectonic setting and evolution of the James Ross Basin, northern

Antarctic Peninsula, in Feldmann, R.M., and Woodburne, M.O., eds., Geology and

Paleontology of Seymour Island, Antarctic Peninsula: Boulder, Colorado, The

Geological Society of America, Inc., Memoir 169, p. 541-555.

Fahrbach, E., Rohardt, G., Scheele, N., Schroder, M., Strass, V., and Wisotzki, A., 1995,

Formation and discharge of deep and bottom water in the northwestern Weddell Sea:

Journal of Marine Research, v. 53, p. 515-538.

Faugeres, J.-C, Dorrik, A.V., Stow, P.I., Viana, A.j 1999, Seismic features diagnostic of

contourite drifts: Marine Geology, v. 162, p. 1-38.

Gilbert, I.M., Pudsey, C.J., and Murray, J. W., 1998, A sediment record of cyclic bottom-

current variability from the northwest Weddell Sea: Sedimentary Geology, v. 115, p.

185-214.

von Gyldenfeldt, A.-B., Fahrbach, E., Garcia, M.A., and Schroder, M., 2002, Flow

variability at the tip of the Antarctic Peninsula: Deep-Sea Research II, v. 49, p. 4743-

4766.

112

Hayes, D.E., ed., 1991, Marine Geological and Geophysical Atlas of the Circum-

Antarctic to 30 degrees S.: Antarctic Research Series, vol. 54. American

Geophysical Union, Washington, D.C., 56 p.

Hollister, CD., and Elder, R.B., 1969, Contour currents in the Weddell Sea: Deep-Sea

Research, v. 16, p. 99-101.

Maldonado, A., Zitellini, N., Leitchenkov, G., Balanya, J.G., Coren, F., Galindo-Zaldivar,

J.,Lodolo,E, Jabaloy, A.,Zanolla, C.,Rodriguez-Fernandez, J.,andVinnikovskaya,

O., 1998, Small ocean basin development along the Scotia-Antarctica plate boundary

and in the northern Weddell Sea: Tectonpphysics, v. 296, p. 371-402.

Maldonado, A., Barnolas, A., Bohoyo, F., Galindo-Zaldivar, J., Hernandez-Molina, F.J.,

Jabaloy, A., Lobo, F.J., Nelson, C.H., Rodriguez-Fernandez, J., Somoza, L., and

Vazquez, J.T., 2005, Miocene to recent contourite drifts development in the northern

Weddell Sea (Antarctica): Global Planet Change, v. 45, p. 99-129.

Michels, K.H., Kuhn, G., Hillenbrand, C.-D., Diekmann, D., Ftitterer, D.K., Grobe, H.,

and Uenzelmann-Neben, G., 2001, The southern Weddell Sea: combined contourite-

turbidite sedimentation at the southeastern margin of the Weddell Gyre, in Stow,

D.A.V., et al., eds., Deep-Water Contourite Systems: Modern Drifts and Ancient

Series, Seismic and Sedimentary Characteristics: London, Geological Society Memoirs,

v. 22, p. 305-323.

Olney, M.P., Scherer, R.P., Bohaty, S.M., and Harwood, D.M., 2005, Eocene-Oligocene

paleoecology and the diatom genus Kisseleviella Sheshukova-Poretskaya from the

Victoria Land Basin, Antarctica: Marine Micropaleontology, v. 58, p. 56-72.

113

Pudsey, C.J., and Camerlenghi, A., 1988, Glacial-interglacial deposition on a sediment

drift on the Pacific margin of the Antarctic Peninsula: Antarctic Science, v. 10, p. 286-

308.

Pudsey, C.J., and Howe, J. A., 1988, Quaternary history of the Antarctic Circumpolar

Current: evidence from the Scotia Sea: Marine Geology, v. 148, p. 83-112.

Rebesco, M., and Stow, D., 2001, Seismic expression of contourites and related deposits:

a preface: Marine Geophysical Researches, v. 22, p. 303-308.

Scherer, R.P., Bohaty, S.M., and Harwood, D.M., 2000, Oligocene and lower Miocene

siliceous microfossil biostratigraphy of Cape Roberts Project Core CRP-2/2A, Victoria

Land Basin, Antarctica: Terra Antartica, v. 7, n. 4, p. 417-442.

SHALDRILII Cruise Report, 2006, Anderson, J.B., et al., eds., Report of Cruise number

NBP0602A to the northwestern Weddell Sea, 369 p.

Sloan, B.J., Lawver, L.A., and Anderson, J.B., 1995, Seismic stratigraphy of the Palmer

Basin, in Cooper, A.K., et al., eds., Geology and seismic stratigraphy of the Antarctic

Margin: Washington, D.C., American Geophysical Union, Antarctic Research Series, v.

68, p. 235-260.

Smith, R.T., and Anderson, J.B., submitted, Geological Society of America Bulletin

Surinach, E., Galindo-Zaldivar, J., Maldonado, A., and Livermore, R., 1997, Large

amplitude magnetic anomalies in the northern sector of the Powell Basin, NE Antarctic

Peninsula: Marine Geophysical Researches, v. 19, p. 65-80.

Wilson, G.S., Lavelle, M., Mcintosh, W.C., Roberts, A.P., Harwood, D.M., Watkins,

D.K., Villa, G., Bohaty, S.M., Fielding, C.R., Florindo, F., Sagnotti, L., Naish, T.R.,

Scherer, R.P., and Verosub, K.L., 2002, Integrated chronostratigraphic calibration of

114

the Oligocene- Miocene boundary at 24.0 ±0.1 Ma from the CRP-2A drill core, Ross

Sea, Antarctica: Geology, v. 30, n. 11, p. 1043-1046.

Wynn, R.B., and Stow, D.A.V., 2002, Classification and characterisation of deep-water

sediment waves: Marine Geology, v. 192, p. 7-22.

115