1

THE IMPACT OF CLIMATE CHANGE AND TECTONIC EVENTS ON OCEAN CIRCULATION IN THE MIOCENE TO PLIOCENE

By

DERRICK RICHARD NEWKIRK

A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT

OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

2012

2

© 2012 Derrick Richard Newkirk

3

To my parents Pat and Rick Newkirk, and my brother Nick Newkirk

4

ACKNOWLEDGMENTS

I would like to sincerely thank Dr. Ellen Martin for her continued guidance on this

project. Not only is she a great mentor, but also a good friend and a pleasure to work

with. Thanks also go to my committee members Dr. Andrea Dutton, Dr. David Hodell,

Dr. George Kamenov, Dr. John Jaeger, and my external committee member Dr. John

Krigbaum.

I am also very grateful for the financial support that was provided by the NSF SGR

grant awarded to Dr. Ellen Martin, the Department of Geological Sciences, Graduate

Student Council, and the College of Liberal Arts and Sciences.

Special thanks go to Susanna Blair for her guidance and continuous help in the lab

at the beginning of my graduate career, and to Dr. Chandranath Basak for the constant

entertainment in the lab and great team work. Also, I would like to thank Dr. George

Kamenov and Dr. Jason Curtis for their assistance in the lab, help with analysis, and

suggestions for streamlining laboratory procedures. I would like to also thank the many

undergraduates who have helped me in the lab.

Finally, I would like to thank my family and friends. Thanks go to Pat and Rick

Newkirk for their unfaltering love and support. Thanks go to Ryan Newkirk for being a

great friend and a very supportive brother. Thanks go to my girlfriend Emily Pugh for her

support and continued encouragement to finish the dissertation, and for being a great

friend. Thanks go to Scottie Andre for playing a huge role in my upbringing and being a

good role model. Thanks go to my many wonderful friends in the Department of

Geological Sciences at UF, especially Jonathan Hoffman, Dr. Jason Gulley, Dr. Dorsey

Wanless, Dr. Richard MacKenzie, John Ezell, and Dr. PJ Moore. Thanks to Dr. Jennifer

Latimer for her guidance, friendship, and invaluable lab guidance that she gave me as

5

an undergraduate researcher. And last but not least thanks go to all of my wonderful

friends outside of the department and from my hometown, especially Jason Wright,

Adam Faust, Nick Kaufman, Judd Sparks, Ron Wright, and Carlos Zambrano.

6

TABLE OF CONTENTS

page

ACKNOWLEDGMENTS .................................................................................................. 4

LIST OF TABLES ............................................................................................................ 8

LIST OF FIGURES .......................................................................................................... 9

ABSTRACT ................................................................................................................... 11

CHAPTER

1 INTRODUCTION .................................................................................................... 14

2 CIRCULATION THROUGH THE CENTRAL AMERICAN SEAWAY DURING THE MIOCENE CARBONATE CRASH .................................................................. 18

Overview ................................................................................................................. 18

Background ............................................................................................................. 19 Methods .................................................................................................................. 21 Results .................................................................................................................... 22

Discussion .............................................................................................................. 23 Summary ................................................................................................................ 26

3 MIOCENE DEEP WATER CIRCULATION IN THE PACIFIC AND CARIBBEAN: IMPACTS OF THE CENTRAL AMERICAN SEAWAY AND SOUTHERN HEMISPHERE GLACIATION ................................................................................. 36

Overview ................................................................................................................. 36

Background ............................................................................................................. 39

Modern Pacific Deep Water Circulation ............................................................ 39 Shoaling History of the Central American Seaway ........................................... 41

Material and Methods ............................................................................................. 42 Core Descriptions and Age Model .................................................................... 42 Sample Preparation and Nd Isotope Measurements ........................................ 43 δ13C Sample Preparation and Measurements .................................................. 44

Results .................................................................................................................... 44

Discussion .............................................................................................................. 45 Circulation in the Eastern Pacific ...................................................................... 45

The middle to late Miocene transition (14 – 9 Ma) ..................................... 45 Late Miocene to Pliocene (8.5 to 2.5 Ma) circulation ................................. 50

History of the Caribbean Basin ......................................................................... 52

Summary ................................................................................................................ 55

7

4 TRANSCONTINENTAL CONNECTION OF THE AMAZON RIVER BASED ON CEARA RISE SEAWATER RECORDS .................................................................. 76

Overview ................................................................................................................. 76

Material and Methods ............................................................................................. 80 Results .................................................................................................................... 82 Discussion .............................................................................................................. 83

The Seawater Signature ................................................................................... 83 Interpretation of Detrital Isotopes ..................................................................... 87

Summary ................................................................................................................ 90

5 CONCLUSIONS ................................................................................................... 105

LIST OF REFERENCES ............................................................................................. 107

BIOGRAPHICAL SKETCH .......................................................................................... 122

8

LIST OF TABLES

Table page 2-1 Nd isotopic values for modern and Miocene water masses ................................ 31

2-2 Nd isotopic results for Sites 846B, 998A, 999A, and 1241A ............................... 32

3-1 Nd isotopic results for the Pacific (Sites 845, 846, 1237, and 1241) ................... 67

3-2 Nd isotopic results for the Caribbean Basin (Sites 998A and 999A) ................... 72

4-1 Nd isotopic results for Ceara Rise fossil fish teeth .............................................. 98

4-2 Nd isotopic values for detrital silicates .............................................................. 102

4-3 Pb isotopic values for detrital silicates .............................................................. 103

4-4 Pb isotopic values for leachates ....................................................................... 104

9

LIST OF FIGURES

Figure page 2-1 Plate reconstruction of the Caribbean region at 10 Ma. ...................................... 27

2-2 Carbonate MARs and εNd values from Site 846, Site 998, Site 999. ................... 28

2-3 Nd isotopic data for Sites 998 and 999 in the Caribbean Basin, Sites 846 and 1241 in the eastern equatorial Pacific. ................................................................ 29

2-4 Carbonate MARs and εNd values for Site 998 plotted with %NCW. .................... 30

3-1 Bathymetric map of the Pacific Basin and Caribbean Basin illustrating the different basins of the eastern Pacific and the associated sills controlling the flow of deep water. ............................................................................................. 57

3-2 Map of the Pacific Ocean illustrating the flow paths. .......................................... 58

3-3 North-South Seawater profile for the central Pacific determined using the phosphate concentration profile for the modern Pacific ocean. .......................... 59

3-4 Seawater Nd isotopic values for the deep water sites (845, 846, and 1237) of the eastern Pacific compared to published values for the central Pacific and the north Pacific. ................................................................................................. 60

3-5 Nd isotopic values for the eastern Pacific vesus the carbon record. ................... 61

3-6 Seawater Nd isotopic data for Ocean Drilling Program Sites 998 and 999 in the Caribbean Basin Compared to the published values (shaded fields) for the Pacific. .......................................................................................................... 62

3-7 Changes in the Nd isotopic composition of water masses from the beginning of the record to 9 Ma from the central equatorial Pacific, the north Pacific, and the three eastern Pacific sites from this study. ............................................ 63

3-8 North-South Seawater profiles for the central Pacific showing the shifts in the boundaries between Pacific water mass. ........................................................... 64

3-9 North-South Seawater profiles for the eastern Pacific Pacific showing the shifts in the boundaries between Pacific water mass. ........................................ 65

3-10 Seawater Nd isotopic data for the eastern Pacific, the Caribbean Basin, and the Straits of Florida near Blake Nose. ............................................................... 66

4-1 Bathymetric map of the Atlantic Ocean, and the bathymetric profile illustrating the position of Ocean Drilling Program sites 925, 926, 929. ............................... 92

10

4-2 Salinity profile along a north-south transect in the Atlantic overlain by seawater εNd profiles. .......................................................................................... 93

4-3 Plot of εNd from fossil fish teeth vs. age for sites 925, 926, and 929 on Ceara Rise in the western equatorial Atlantic. ............................................................... 94

4-4 εNd vs. age for detrital silicate fractions and fossil fish teeth for sites 925, 926, and 929 on Ceara Rise. ...................................................................................... 95

4-5 206Pb/204Pb vs. age for both detrital silicate fractions and Fe-Mn oxide coatings for sites 925, 926, and 929 on Ceara Rise. .......................................... 96

4-6 Pb isotopic crossplots of the detrital silicate fractions from all three Ceara Rise sites (925, 926, and 929). ........................................................................... 97

11

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy

THE IMPACT OF CLIMATE CHANGE AND TECTONIC EVENTS ON OCEAN

CIRCULATION IN THE MIOCENE TO PLIOCENE

By

Derrick Richard Newkirk

August 2012

Chair: Ellen E. Martin Major: Geology

Two major events occurred in the Miocene following the mid-Miocene Climatic

Optimum: one was a global cooling trend associated with renewed growth of East

Antarctic Ice Sheet, and the other was tectonic closure of the Central American Seaway

(CAS). The response of ocean circulation to Antarctic glaciations has been focused in

the Southern Ocean and the southwest Pacific, yet little is known about how circulation

was affected in the rest of the Pacific, particularly eastern Pacific in the region of the

CAS. Shoaling of the CAS in the middle Miocene has been implicated in local and

global changes in ocean circulation and climate, and in the development of low

carbonate intervals referred to as “carbonate crash” events that are found in the eastern

equatorial Pacific, Caribbean, and possibly the western Atlantic. The effects of an open

CAS on global thermohaline circulation have also been debated. Several modeling

studies have shown that exchange through the CAS would reduce the temperature, and

salinity of the water in the north Atlantic, thereby weakening or shutting down North

Atlantic Deep Water (NADW) production. Reconstructions of paleocirculation are

required to evaluate the impact of closure of this equatorial gateway and high southern

latitude climate change.

12

We reconstructed deep water circulation from the early/middle Miocene to the late

Pliocene using neodymium (Nd) isotopic records for fossil fish teeth/debris from a

longitudinal transect of Ocean Drilling Program sites in the eastern Pacific, two sites in

the Caribbean, and a depth transect in the western Atlantic to assess the impact of

these events on circulation and carbonate deposition in the Pacific, Caribbean, and

Atlantic regions. Nd isotopic records from eastern Pacific sites suggest expansion of

Pacific Deep Water (PDW) through the mid to late Miocene, leading to deepening of the

boundary between PDW and Circumpolar Deep Water in the eastern equatorial region.

This rearrangement of Pacific circulation coincides with increased flow of the Deep

Western Boundary Current into the Pacific in response to intensified Antarctic glaciation.

Expansion of corrosive PDW into the equatorial Pacific and Caribbean appears to

trigger middle Miocene carbonate crash events. Termination of the carbonate crash in

the Pacific is attributed to an increase in equatorial surface productivity and carbonate

rain rates in this region, while termination of the carbonate crash and a

contemporaneous shift from Pacific- to Atlantic-sourced deepwater in the Caribbean

Basin was associated with shoaling of the CAS. Thus, changes in circulation in the

Miocene Pacific appear to be driven by high latitude climate, while changes in the

Caribbean are a response to the low latitude tectonic event.

Results in the Caribbean highlight the flow of Pacific water through the CAS during

times of suggested NADW production, which contradicts many general ocean

circulation models that suggest NADW production did not occur with an open CAS.

Production estimates of NADW throughout the Miocene are based on Atlantic-Pacific

δ13C gradients, which are very subtle for this time interval. To better constrain the

13

relative production of NADW and AABW from the middle Miocene to Pliocene, a

seawater Nd isotopic record was developed using samples from Ceara Rise, which is

located midway between north and south Atlantic deep water sources and covers a

depth range from ~3000-4300 m. The present day position of the boundary between

NADW and AABW lies within the depth transect of Ceara Rise. Miocene to Pliocene Nd

isotopic records of seawater are far less radiogenic than any known water mass in the

Atlantic Ocean, and the record shows a significant shift in values around 8 Ma and a

smaller shift at ~4.5 Ma. Evaluation of the silicate fraction of the deep-sea sediments

illustrated that the Nd and lead (Pb) isotopic values recovered from fossil fish teeth and

ferromanganese oxide coatings on the sediment are very similar to the values of the

detrital silicate fractions. The combined detrital silicate and seawater records suggest

that extensive detrital inputs in this region overwhelmed the record of seawater isotopes

through reversible scavenging, and therefore it was not possible to identify a shift in the

boundary between NADW and AABW in the seawater record. On the other hand, the

isotopic records were interpreted in terms of changes in source material derived from

Amazon lowlands (South American shield material) to material from the Andean

highlands (volcanic arc material), indicating a change in the Amazon drainage basin

associated with known uplift events of the Andes Mountains. These results show that a

transcontinental connection of the Amazon drainage basin occurred at ~8 Ma, and the

small shift at ~4.5 was related to continued Andean uplift.

14

CHAPTER 1 INTRODUCTION

Reconstruction of deep water circulation allows for a more thorough evaluation of

the relationship between changes in ocean circulation and climate. One of the big

questions paleoceanographers and paleoclimatologists are trying to address is the role

ocean circulation plays in driving climate, such as the onset of glaciations, as well as

impact climate has on circulation. Tectonic gateway events have long been suggested

to play a significant role in altering ocean circulation and producing conditions

conducive to onset of glaciations in both hemispheres. The focus of this study revolves

around understanding the relationship of the shoaling of the Central American Seaway

to suggested changes in regional (eastern equatorial Pacific and Caribbean) and global

ocean circulation and ultimately how those changes have affected climate change and

deep sea sedimentation patterns, particularly with reference to low carbonate

accumulation intervals referred to as “carbonate crash” events.

Neodymium (Nd) isotopes recovered from fossil fish teeth were used to

reconstruct circulation in the eastern Pacific, Caribbean, and western equatorial Atlantic

to determine how gateway events and climate change altered circulation in these

regions. Nd isotopes were chosen because they are considered to be quasi-

conservative tracers of water mass, meaning that the cores of different water masses

have distinct Nd isotopic signatures that can only be altered through water mass mixing

or addition of local weathering inputs [Frank, 2002; Goldstein and Hemming, 2003].

Important for this proxy is the fact that the residence time of Nd in the oceans [~600 –

1000 yrs; Elderfield and Greaves, 1982; Piepgras and Wasserburg, 1985; Jeandel et

al., 1995; Tachikawa et al., 1999; Arsouze et al., 2009] is shorter than the modern

15

ocean mixing time of ~1500 years [Broecker and Peng, 1982]. Fossil fish teeth and

debris have been demonstrated to be robust paleoceanographic archives of bottom

water Nd isotopes [Elderfield and Pagett, 1986; Martin and Haley, 2000; Thomas et al.,

2003; Martin and Scher, 2004; Thomas, 2004; Scher and Martin, 2006].

During the middle Miocene the eastern equatorial Pacific [Vincent, 1981; Mayer et

al., 1986; Farrell et al., 1995; Lyle et al., 1995], and Caribbean [Roth et al., 2000]

recorded intervals of extensive carbonate dissolution referred to as “carbonate crash”

events [Lyle et al., 1995]. Lyle et al. [1995] and Roth et al. [2000] suggested observed

carbonate crash events in both the Pacific and Caribbean were related to changes in

circulation coinciding with shoaling of the Central American Seaway (CAS) in the Middle

Miocene, but they could not determine how circulation changed. Chapter 2 attempts to

understand how circulation changed in the Caribbean and determine whether carbonate

dissolution was the result of an influx of corrosive Antarctic Intermediate Water (AAIW)

sourced from the Atlantic or corrosive water sourced from the north Pacific. Chapter 2

was published in the journal titled Geology in January of 2009 [Newkirk and Martin,

2009].

The primary goal of Chapter 3 was to reconstruct deep water circulation in the

eastern Pacific and Caribbean during the Miocene using Nd isotopic records for fossil

fish teeth/debris from a longitudinal transect of four Ocean Drilling Program (ODP) sites

in the eastern Pacific (Site 845, 846, 1231, and 1241) and expanded datasets for

Caribbean records (ODP sites 998 and 999) to improve our understanding of processes

driving circulation in the Pacific and Caribbean following the mid-Miocene Climatic

Optima and shoaling CAS on conditions in the Caribbean Basin.

16

General ocean circulation models suggest NADW production should not occur

with an open CAS; however, Nd isotopic results from Chapters 2 and 3 [Newkirk and

Martin, 2009] indicate the strongest fluxes of Pacific water into the Caribbean during

times of purported strong NADW production. Chapter 4 attempts to evaluate shifts in the

boundary between Northern Component Water (NCW) and Antarctic Bottom Water

(AABW), which has been shown to shoal and deepen depending on relative production

rates of these water masses. To investigate variations in NCW and AABW we analyzed

Nd isotopes of fossil fish teeth from a depth transect spanning from ~3000-4300 m on

Ceara Rise that included ODP sites 925, 926, and 929. The goals was to use variations

in the Nd isotopic records of these sites to establish the position of the NCW/AABW

interface through time and determine if a direct correlation existed between carbonate

dissolution and the influx of AABW over the Ceara Rise.

Our early results of seawater Nd isotopes indicated the values were strongly

influenced by a signal from Amazon sediments. Thus, the study was refocused to

understand the impact of a major river outflow on the seawater Nd isotopes preserved

in fish teeth and to evaluate changes to the composition of terrigenous material

weathered from South America, which improved our understanding of the evolution of

the Amazon River basin and the impact of the extensive sediment supply on the

seawater Nd isotopic signal. Depending on the location of the study (inland, proximal

fan, distal Amazon outflow) and the proxy applied, different studies have come to

different conclusions about the outlet for the Amazon River (Caribbean versus Atlantic),

and the timing and cause of shifts in the composition of material carried by the river.

17

To further evaluate suggested changes in provenances of material reaching the

Ceara Rise [Harris and Mix, 2002; Dobson et al., 2001] and constrain the timing of the

shift in provenance, detrital silicate fractions were separated from bulk sediment

samples from all three sites (925, 926, 929) and analyzed for Nd and lead (Pb). The Nd

and Pb isotopic data help to better constrain the timing of a shift from a system

dominated by South American Shield material to one dominated by Andean material,

and also to determine whether the shift at ~4.5 Ma in the clay mineralogy identified by

Harris and Mix [2002] was the result of continued Andean uplift or changes in

weathering related to changing climatic conditions.

18

CHAPTER 2 CIRCULATION THROUGH THE CENTRAL AMERICAN SEAWAY DURING THE

MIOCENE CARBONATE CRASH

Overview

The Caribbean carbonate crash, an interval of low carbonate mass accumulation

rates (MARs) in the middle to late Miocene, has been linked to the presence of

corrosive waters in the Caribbean associated with changes in circulation patterns and

shoaling of the Isthmus of Panama. For example, Roth et al. [2000] suggested that

North Atlantic Intermediate Water (NAIW) flowed over shallow to intermediate depth sills

on the Atlantic side of the Caribbean Basin during times of enhanced carbonate

preservation, while corrosive Antarctic Intermediate Water (AAIW) overflowed the sills

during intervals of carbonate dissolution. This configuration was based on the

correlation between carbonate crash intervals and periods of more intense Northern

Component Water (NCW) production defined by Wright and Miller [1996], as well as

modern circulation of AAIW into the Caribbean [Haddad and Droxler, 1996]. In contrast,

several general circulation models (GCMs) with an open Central American Seaway

predict west to east flow at depth through the gateway, such that some of the waters

flowing into the Caribbean would have been derived from the Pacific [Mikolajewicz and

Crowley, 1997; Nisancioglu et al., 2003; Prange and Schulz, 2004; Schneider and

Schmittner, 2006; Steph et al., 2006].

Traditional paleoproxy data from this region cannot distinguish between South

Atlantic and Pacific sources of corrosive waters. In contrast, neodymium (Nd) isotopic

compositions of these water masses are highly distinct. The Nd isotopic signature of

seawater is considered to be a conservative tracer of water masses [Goldstein and

Hemming, 2003 and references therein]. End-member Nd isotopic compositions for

19

Miocene water masses potentially flowing into the Caribbean are well constrained by

published data from Fe-Mn crusts and fish teeth, and data in this paper (Table 2-1).

Fossil fish teeth and debris composed of hydroxyfluorapatite have been demonstrated

to be robust paleoceanographic archives of bottom water Nd isotopes [Martin and

Scher, 2004 and references therein], making them powerful archives during the

Caribbean carbonate crash.

In this paper we report Nd isotopic values for fossil fish teeth and debris from

Ocean Drilling Program Sites 998 and 999 in the Caribbean and Sites 846 and 1241 in

the eastern equatorial Pacific (Figure 2-1 and 2-2) in order to identify sources of bottom

waters and basic circulation patterns in the Caribbean during closure of the Central

American Seaway and the Miocene Caribbean carbonate crash. The two Caribbean

sites are ideally located to correlate regional changes in circulation to large-scale

changes in the thermohaline circulation defined by variations in NCW, or proto-North

Atlantic Deep Water, production.

Background

The water masses filling the Caribbean have changed as sills on the eastern and

western margins of the basin evolved. On the Atlantic side, the Windward Passage

(1500 m) and the Anegada-Jungfern Passage (1800 m) are pathways for intermediate

waters to enter the basin (Figure 2-1). On the Pacific side, the Isthmus of Panama

currently blocks all inflow into the Caribbean, but an open Central American Seaway

would have allowed exchange between these basins during the Miocene until final

closure in the Pliocene. Specifically, the Isthmus of Panama is believed to have shoaled

to ~1000 m between 12-10.2 Ma based on benthic foraminiferal assemblages [Duque-

20

Caro, 1990], while final closure has been placed between ~4.2 – 2.4 Ma [Keller et al.,

1989; Haug et al., 2001].

Due to its short oceanic residence time and predominant continental sources,

different water masses have distinct Nd isotopic ratios that reflect the geology of source

regions. Miocene North Atlantic, South Atlantic, and Pacific water masses have distinct

εNd values (Table 2-1) enabling the reconstruction of changing circulation patterns in

response to shoaling of the Isthmus of Panama.

Site 998 is located in the Yucatan Basin at a modern water depth of 3180 m and

Site 999 is located in the Columbian Basin at a modern water depth of 2828 m (Figure

2-1). Although both sites are relatively deep, bottom waters in the Caribbean are

derived from intermediate to shallow waters overflowing sills separating the Caribbean

from the Pacific and Atlantic. Thus, sill depths of these passageways rather than

paleodepths for the specific sites control the composition of bottom waters in the basin.

The end-member composition of water on the Pacific side of the Central American

Seaway was determined using data from two sites in the eastern equatorial Pacific; Site

846 in the Peru Basin (3296 m modern water depth) and Site 1241 on the Cocos Ridge

(2027 m modern water depth). Lyle et al. [1995] calculated that Site 846 subsided ~50

m between 10 Ma and today. Subsidence of several hundred meters proposed for Site

1241 [Mix et al., 2003] would place it at an ideal depth to monitor intermediate waters

flowing through the Central American Seaway during the Miocene. Age-depth models

for all four sites are based on the same biostratigraphic boundaries and age datums

defined by Raffi and Flores [1995] at Site 846 and applied to Sites 998 and 999 by

Kameo and Bralower [2000] and to Site 1241 by Mix et al. [2003].

21

Methods

Sediment samples were oven dried, disaggregated and wet sieved prior to

picking fossil fish teeth from the >125 μm fraction. The fossil fish teeth were then

cleaned using an oxidative/reductive cleaning technique from Boyle [1981] and Boyle

and Keigwin [1985] that removes organic matter and Fe-Mn oxide coatings.

Concentrations of Nd in teeth typically range from 100 to 400 ppm [Martin and Haley,

2000], thus ~100 g of cleaned teeth were processed in order to produce at least 10 ng

Nd for analysis. Cleaned teeth were dissolved in aqua regia and then dried prior to a

two step chemical separation to isolate Nd. Bulk rare earth elements (REEs) were

separated from the sample on a primary quartz column that uses Mitsubishi cation

exchange resin with HCl as the eluent [Scher and Martin, 2004]. Nd was further

isolated using quartz columns packed with Teflon beads coated with bis-ethylhexyl

phosphoric acid and HCl as the eluent. The total blank for this technique is 14 pg Nd.

Nd isotopic ratios were measured on a Nu Plasma Multi-Collector-Inductively

Coupled Plasma-Mass Spectrometer (MC-ICP-MS) at the University of Florida. Dried

samples were re-dissolved with 0.3 ml of 2% Optima HNO3, and then a portion of the

sample was pipetted into a Teflon sampling beaker and diluted 100 times using 2%

Optima HNO3. Additional acid or sample was then added as needed to achieve the

ideal voltage of 2-6 volts for 143Nd. Belshaw et al. [1998] describe the instrument and

the optimal operating conditions for the Nu MC-ICP-MS. JNdi-1 standard was run

between every 4 to 6 samples, depending on the number of analyses acquired. All of

the JNdi-1 values analyzed during one day were averaged and that value was

normalized to the long-term TIMS value of 0.512103 ± 0.000014 (2σ). Individual

22

sample runs were then normalized by the same amount to correct for the daily

variations in running conditions. A drift correction was not applied because variations

throughout a run did not indicate a consistent drift. The 2σ error for the Nu MC-ICP-MS

based on the variability of normalized JNdi-1 analyses is ±0.000015, which is equivalent

to 0.3 εNd units and agrees well with data from replicate analyses (Table 2-2), which

vary by 0.05 to 0.53 εNd units with an average difference of 0.25 εNd units.

Concentrations of Sm and Nd were analyzed on an Element II for three samples

from each site in order to determine 147Sm/144Nd ratios. The average 147Sm/144Nd ratios

are 0.120 for Site 846, 0.134 for Site 998, 0.135 for Site 999, and 0,127 for Site 1241.

These ratios were then used to correct for age-dependent ingrowth of radiogenic 143Nd

(εNd(T)). This correction was minor (0.03 to 0.14 εNd units) for these young samples.

Results

The records for Caribbean Sites 998 and 999 are divided into pre-crash (~ 14 to

12 Ma), crash (12-10 Ma), and post-crash intervals based on carbonate MAR records

from Roth et al. [2000] (Figure 2-2). The εNd values at both sites show distinct values

and patterns associated with these three intervals (Table 2-2). The pre-crash intervals

at Sites 998 and 999 are marked by relatively stable εNd values of approximately -4 and

-3 respectively. The beginning of the pre-crash interval at Site 999 is recorded by εNd

values that increase from -5.5 to –3, while carbonate MARs decrease. The transition

into the crash is marked by a divergence of Nd isotopic values with values increasing to

~0 at Site 998 and decreasing to ~-6.5 at Site 999. However, during the remainder of

the crash interval εNd values exhibit greater variability, ranging between –4.5 and 0, and

the two sites illustrate more similar patterns with a general correlation between low

carbonate MARs and high εNd values. During the post-crash interval, εNd values

23

ultimately decrease to values <-6 at both sites; although, Site 999 records a brief

increase to ~-2 before decreasing abruptly to ~-6.5, while Site 998 records a gradual

decreasing trend.

At Site 846 low carbonate MARs representing a distinct eastern equatorial Pacific

carbonate crash are younger than those of the Caribbean carbonate crash interval. The

Pacific record has been divided into pre-crash and crash intervals. During the pre-crash

from 14.1 to 11.2 Ma, carbonate MARs are ~0.5 to 1.0, while εNd values trend from -3.8

to ~-2.5 (Figure 2-2; Table 2-2). During the crash interval from 11.2 to 8 Ma, carbonate

MARs are lower at ~0 to 0.5 and εNd values remain relatively stable, ranging between ~-

1.7 and -2.9. As with the Caribbean sites, the crash interval at Site 846 is characterized

by lower carbonate MARs and higher εNd values. No carbonate MAR data are available

for Site 1241, but εNd values at this site are even more radiogenic than Site 846, with

values ranging between -3 and +2 from 11.3 to 5 Ma (Figure 2-3).

Discussion

Miocene εNd values within the Caribbean extend to values that are higher than any

known intermediate or deep-water mass from the Atlantic (Figure 2-3). These

radiogenic values could represent water sourced from another region or alteration of the

seawater signal by material introduced into the water column or sediment. Abundant

volcanic ash with εNd values up to +9 [Feigenson et al., 2003] represents an obvious

potential source of radiogenic material in the Caribbean; however, there is no

correlation between εNd values and volcanic ash MARs at Sites 998 and 999. Instead,

observed εNd values during the crash at Sites 998 and 999 are similar to the most

radiogenic seawater values recorded in the Pacific today and during the Miocene

(Figure 2-3).

24

Pacific-like εNd values and the correlation between radiogenic εNd values and

lower carbonate MARs in the Caribbean are consistent with the idea that corrosive,

radiogenic upper-deep/intermediate Pacific waters filled the basin during intervals of

dissolution, with a greater influx of less corrosive, less radiogenic upper-

deep/intermediate Atlantic waters during intervals of enhanced preservation. In

particular, εNd values during the crash are similar to values recorded at Site 1241, which

represents intermediate waters at the proper depth to flow into the Caribbean. Minor

differences between records at Sites 998 and 999 can be attributed to the presence of

the Nicaraguan Rise (Figure 2-1), which separated the north and south Caribbean until

~12 Ma when faulting initiated a north-south oriented passageway for intermediate

waters [Droxler et al., 1998; Roth et al., 2000].

Before and after the Caribbean carbonate crash, observed εNd values exceed

values recorded in the Atlantic (Figure 2-3), indicating that Pacific throughflow

dominated the deep Caribbean throughout our record, although it was diluted by Atlantic

throughflow during intervals of enhanced preservation. This dominance of Pacific

throughflow when the Central American Seaway was open to at least ~1000 m is also

predicted by several GCMs [e.g., Mikolajewicz and Crowley, 1997; Nisancioglu et al.,

2003; Prange and Schulz, 2004; Schneider and Schmittner, 2006; Steph et al., 2006;

Lunt et al., 2007]. As the Central American Seaway continued to shoal, post-crash data

within the Caribbean is characterized by higher carbonate MARs and decreasing εNd

values, indicating increased input of Atlantic water. This change in source input as a

result of the shoaling of the Central American Seaway is consistent with the idea

proposed by Frank et al. [1999] and Reynolds et al. [1999] to explain decreasing εNd

25

values they reported for an Fe-Mn crust from the Florida Straits, which sampled water

leaving the Caribbean (Figure 2-3). Their data only extend to 8.5 Ma, but data from

Sites 998 and 999 pinpoint the initiation of decreased throughflow at ~10.7 Ma.

Flow of Pacific water into the Caribbean is also supported by data on coccolith and

planktonic foraminiferal assemblages at this time. Nannofossil assemblages at Site 999

are similar to assemblages reported in the eastern equatorial Pacific from 15.9-10.7 Ma.

They begin diverging from Pacific-type assemblages between 13.6-10.7 Ma, and

became completely distinct between 10.7-9.4 Ma [Kameo and Sato, 2000]. In addition,

Chiasson and D’Hondt [2000] attributed temperate latitude foraminiferal assemblages

(Globoconellids) observed at Site 999 until ~10.7 Ma to an influx of cool Pacific surface

water from either the California or Peru Currents.

There are two intervals in our data when higher εNd values do not correspond to

lower carbonate MARs. The first occurs at the transition into the crash, when εNd values

at Site 998 increase dramatically, while values at Site 999 decrease to a value below

the Pacific end-member (Figure 2-3). This less radiogenic value could represent a

mixture of two relatively corrosive water masses, North Pacific Intermediate Water

(NPIW) and AAIW (Figure 2-3), while the more radiogenic values observed at Site 998

may indicate that only the shallowest, Pacific-derived waters could cross the

Nicaraguan Rise. The second deviation occurs at the beginning of the post-crash

interval at Site 999. The observed combination of radiogenic εNd values and enhanced

carbonate preservation suggests this water could represent a mixture of radiogenic

upper NPIW and non-corrosive NAIW, or a system dominated by nutrient-rich Pacific

waters leading to enhanced carbonate rain rates and preservation.

26

A plot of carbonate MAR and εNd at Site 998 compared to estimated NCW

production [Wright and Miller, 1996; Poore et al., 2006] suggests a correlation between

enhanced production of NCW and increased Pacific throughflow from ~12.4 to 9.5 Ma

(Figure 2-4). This relationship suggests that flow patterns in the Caribbean region were

linked to global circulation and that northward flow of low salinity waters due to strong

equatorial exchange did not limit NCW production. Nisancioglu et al. [2003] present the

only GCM to predict significant NCW production with an open Central American Seaway

and geostrophic flow from the Pacific to the Atlantic.

Summary

Nd isotopes from fossil fish teeth and debris at Sites 998 and 999 in the Caribbean

and Sites 846 and 1241 in the eastern equatorial Pacific indicate that waters sourced

from the Pacific dominated flow into the Caribbean during the Miocene Caribbean

carbonate crash. Prior to the Caribbean crash a gradual decrease in carbonate MARs

and an associated increase in εNd values at Site 999 provide evidence for the

introduction of a more corrosive, Pacific intermediate water mass into the Caribbean as

the Central American Seaway shoaled to critical depths for west to east flow. During the

Caribbean carbonate crash (12-10 Ma) highly variable εNd values and carbonate MARs

record pulses of almost pure, corrosive Pacific waters that filled the deep Caribbean.

These pulses of Pacific throughflow correlate well with NCW production, suggesting that

NCW production can occur with an open Central American Seaway and that flow

patterns in the Caribbean region are linked to global circulation patterns. After the

Caribbean carbonate crash, εNd values gradually shift to less radiogenic values

indicating a reduction in the amount of Pacific water flowing into the Caribbean

coincident with the shoaling of the Isthmus of Panama.

27

Figure 2-1. Plate reconstruction of the Caribbean region at 10 Ma [after Pindell, 1994]

illustrating locations of geographic features, as well as ODP sites used in this study (Sites 846, 998, 999, and 1241) and one Fe-Mn crust (BM1963.897) from Reynolds et al. [1999].

28

Nd

(T)

-6

-4

-2

0

0.0

0.5

1.0

1.5

2.0

Age (Ma)

Ca

CO

3 M

AR

(g

/cm

2 p

er

k.y

.)

0.0

0.5

1.0

1.5

2.0

-4

-3

-2

-1

8.0 8.5 9.0 9.5 10.0 10.5 11.0 11.5 12.0 12.5 13.0 13.5 14.0 14.5

-6

-4

-2

0

0.0

0.5

1.0

1.5

2.0

CrashPost-Crash Pre-Crash

Crash Pre-CrashSite 846

Site 998

Site 999

MAR

MAR

MAR

A

B

C

Nd

Nd

Nd

Figure 2-2. Carbonate MARs (small filled circles for all sites) and εNd values from (A) Site 846 (diamonds) in the eastern equatorial Pacific, (B) Site 998 (large circles), and (C) Site 999 (squares) in the Caribbean Basin. Note the εNd scale is slightly different for Site 846. Carbonate MARs are from Farrell et al. [1995] for Site 846 and Roth et al. [2000] for Sites 998 and 999. The carbonate crash interval highlighted in gray is defined by variable, but low carbonate MARs.

29

Figure 2-3. Nd isotopic data for Sites 998 and 999 in the Caribbean Basin, Sites 846 and 1241 in the eastern equatorial Pacific, and data by Reynolds et al. [1999] for crust BM1963.897 from the Straits of Florida. Shaded fields represent the range of εNd values from the South Atlantic [Thomas and Via, 2007], North Atlantic [Burton et al., 1997, 1999; O’Nions et al., 1998; Reynolds et al., 1999], North Pacific [van de Flierdt et al., 2004], and central equatorial Pacific [Ling et al., 1997; Frank et al., 1999]. No data older than 12 Ma are available for North Pacific intermediate/deep waters.

30

Figure 2-4. Carbonate MARs (thin black line) [Roth et al., 2000) and εNd values (filled circles and thick black line) for Site 998 plotted with %NCW (thick gray line) from Wright and Miller [1996]. The %NCW was calculated using interbasin gradients in δ13C. There are large age uncertainties within the compiled data. In addition, low δ13C gradients after 12 Ma translate to large uncertainties in NCW production in the older portion of the record.

Age (Ma)

9.0 9.5 10.0 10.5 11.0 11.5 12.0 12.5 13.0 13.5 14.0

Nd

(T)

-6

-4

-2

0

Carb

onate

MA

R

0.0

0.2

0.4

0.6

0.8

1.0

N

CW

0

20

40

60

80

100

Nd

MAR

NCW

31

Table 2-1. Nd isotopic values for modern and Miocene water masses

Water Mass Modern εNd Miocene εNd

AAIW -7 to -9 (4) -8 (10, 11)

AABW -8 to -9 (4) -8 (10)

UNADW/NAIW -13 (2) -11 (5, 8)

NADW -13.5 (2) -11.5 (7)

PDW -4 (3) -4 (6, 9)

EEPDW* -3.8 (3) -3.6 to -1.6 (10, 13)

NPUDW#/NPIW -3 (3) -2.5 to -1.5 (12)

Equatorial PIW 0 (1) +2 to -0.6 (13)

1Piepgras and Wasserburg, 1982; 2Piepgras and Wasserburg, 1987; 3Piepgras and Jacobsen, 1988; 4Jeandel, 1993; 5Burton et al., 1997; 6Ling et al., 1997; 7O’Nions et al., 1998; 8Burton et al., 1999; 9Martin and Haley, 1999; 10Frank et al., 1999; 11Scher and Martin, 2004; 12van de Flierdt et al., 2004; 13this study. * EEPDW = Eastern Equatorial Pacific Deep Water # NPUDW = North Pacific Upper-Deep Water.

32

Table 2-2. Nd isotopic results for Sites 846B, 998A, 999A, and 1241A

Site Depth (mbsf) Age (Ma) a 143Nd/144Nd b εNd(o) c εNd(t)

d

846B-29X-6W 61-67 272.34 8.09 0.512488 -2.9 -2.9 846B-30X-3W 110-116 278.03 8.43 0.512549 -1.7 -1.7 846B-31X-1W 77-83 284.30 8.81 0.512547 -1.8 -1.7 846B-31X-2W 101-107 286.04 8.92 0.512523 -2.2 -2.2 846B-31X-4W 98-104 289.02 9.10 0.512535 -2.0 -1.9 846B-31X-6W 101-107 292.04 9.29 0.512522 -2.3 -2.2 292.04 9.29 0.512536 -2.0 -1.9 846B-32X-4W 103-109 298.66 9.69 0.512523 -2.2 -2.1 298.66 9.69 0.512545 -1.81 -1.8 846B-32X-6W 27-33 300.90 9.82 0.512526 -2.2 -2.1 300.90 9.82 0.512534 -2.0 -1.9 846B-32X-6W 94-100 301.57 9.86 0.512532 -2.1 -2.0 846B-33X-1W 100-107 303.84 10.00 0.512554 -1.6 -1.5 846B-33X-2W 31-37 304.64 10.05 0.512526 -2.2 -2.1 846B-33X-4W 144-150 308.78 10.30 0.512527 -2.2 -2.1 846B-33X-6W 30-36 310.63 10.42 0.512510 -2.5 -2.4 846B-34X-1W 140-146 313.83 10.61 0.512488 -2.9 -2.8 846B-34X-3W 121-127 316.64 10.78 0.512519 -2.3 -2.2 316.64 10.78 0.512534 -2.0 -1.9 846B-34X-4W 111-117 318.04 10.87 0.512531 -2.1 -2.0 318.04 10.87 0.512512 -2.5 -2.4 846B-34X-7W 21-27 321.64 11.09 0.512541 -1.9 -1.8 846B-36X-2W 128-134 334.51 11.87 0.512500 -2.7 -2.6 334.51 11.87 0.512482 -3.0 -2.9 846B-36X-3W 128-134 336.01 11.96 0.512481 -3.1 -3.0 846B-37X-2W 93-99 343.76 12.43 0.512519 -2.3 -2.2 343.76 12.43 0.512520 -2.3 -2.2 846B-38X-4W 144-150 356.98 13.23 0.512490 -2.9 -2.8 846B-40X-1W 121-127 371.54 14.14 0.512446 -3.7 -3.6 998A-10H-2W 45-50 86.78 4.50 0.512260 -7.4 -7.3 998A-11H-6W 129-134 103.12 5.50 0.512286 -6.9 -6.8 998A-12H-6W 58-63 111.91 6.50 0.512312 -6.4 -6.3 998A-13H-4W 66-71 118.49 7.25 0.512306 -6.5 -6.4 998A-14H-1W 6-11 122.88 7.75 0.512451 -3.7 -3.6 998A-14H-3W 145-150 127.28 8.25 0.512323 -6.2 -6.1 998A-14H-3W 64-69 129.47 8.50 0.512349 -5.6 -5.6 998A-15H-1W 21-26 132.51 8.95 0.512300 -6.6 -6.5 998A-15H-3W 107-113 136.37 9.33 0.512323 -6.2 -6.1 998A-15H-5W 27-32 138.58 9.52 0.512333 -6.0 -5.9 138.58 9.52 0.512314 -6.3 -6.2 998A-16H-1W 54-59 141.85 9.79 0.512379 -5.1 -5.0 998A-16H-3W 126-129 146.05 10.14 0.512351 -5.6 -5.5

33

Table 2-2. Continued

Site Depth (mbsf) Age (Ma) a 143Nd/144Nd b εNd(o) c εNd(t)

d

998A-16H-4W 32-37 146.62 10.18 0.512396 -4.7 -4.6 146.62 10.18 0.512416 -4.5 -4.45 998A-16H-4W 45-50 146.77 10.19 0.512457 -3.5 -3.5 998A-16H-5W 25.5-30 148.11 10.30 0.512378 -5.1 -5.0 998A-16H-6W 32-36 149.65 10.46 0.512400 -4.7 -4.6 998A-16H-6W 125-130 150.58 10.59 0.512491 -2.9 -2.8 998A-17H-1W 21-26 151.53 10.73 0.512496 -2.8 -2.7 998A-17H-1W 32-37 151.62 10.75 0.512545 -1.8 -1.7 998A-17H-1W 77-82 152.08 10.82 0.512498 -2.7 -2.7 998A-17H-1W 105-110 152.35 10.86 0.512466 -3.4 -3.3 998A-17H-2W 25-30 153.05 10.97 0.512440 -3.9 -3.8 998A-17H-2W 54-60 153.36 11.02 0.512476 -3.2 -3.1 998A-17H-2W 126-131 154.06 11.13 0.512438 -3.9 -3.8 154.06 11.13 0.512431 -4.1 -4.0 998A-17H-4W 55-60 156.38 11.50 0.512574 -1.3 -1.2 998A-17H-5W 2-7 157.38 11.66 0.512440 -3.9 -3.8 998A-17H-5W 134-139 158.68 11.82 0.512491 -2.9 -2.8 998A-17H-6W 26-31 159.09 11.87 0.512637 -0.0 0.06 998A-17H-6W 81-87 159.65 11.93 0.512594 -0.9 -0.8 998A-17H-CCW 2-7 160.62 12.03 0.512626 -0.2 -0.2 998A18X-1W 105-111 161.85 12.17 0.512420 -4.3 -4.2 998A-18X-3W 32-36 164.12 12.41 0.512457 -3.5 -3.4 998A-19X-1W 53-58 166.75 12.70 0.512418 -4.3 -4.2 998A-19X-5W 24-28 172.44 13.50 0.512445 -3.8 -3.7 998A-20X-2W 32-36 177.72 14.05 0.512434 -4.0 -3.9 998A-22X-3W 21-26 198.34 16.00 0.512435 -4.0 -3.8 998A-24X-2W 145-150 217.38 17.00 0.512420 -4.3 -4.1 998A-25X-3W 76-81 227.79 17.50 0.512399 -4.7 -4.5 998A-26X-4W 7-12 238.20 18.00 0.512514 -2.4 -2.3 999A-17X-5W 42-47 156.55 5.0 0.512301 -6.57 -6.54 999A-18X-6W 90-95 168.03 5.5 0.512316 -6.09 -6.04 999A-19X-4W 14-19 173.77 5.75 0.512338 -5.85 -5.81 999A-20X-1W 88-93 179.51 6.00 0.512309 -6.42 -6.37 999A-23X-2W 34-39 202.47 7.00 0.512291 -6.77 -6.72 999A-23X-6W 8-13 208.21 7.25 0.512361 -5.41 -5.35 999A-24X-3W 62-67 213.95 7.50 0.512334 -5.93 -5.88 999A-24X-7W 36-41 219.67 7.75 0.512381 -5.02 -4.96 999A-25X-4W 100-105 225.43 8.00 0.512322 -6.17 -6.11 999A-26X-2W 14-19 231.17 8.25 0.512363 -5.37 -5.30 999A-26X-5W 128-133 236/91 8.50 0.512344 -5.74 -5.67 999A-27X-3W 52-57 242.65 8.75 0.512377 -5.09 -5.02 999A-28X-1W 18-23 248.91 8.98 0.512395 -4.75 -4.68 999A-28X-3W 32-36 252.04 9.10 0.512337 -5.88 -5.81

34

Table 2-2. Continued

Site Depth (mbsf) Age (Ma) a 143Nd/144Nd b εNd(o) c εNd(t)

d

999A-28X-4W 84-88 254.06 9.18 0.512360 -5.42 -5.35 999A-28X-5W 55-59 255.27 9.22 0.512474 -3.21 -3.14 999A-29X-1W 72-76 259.14 9.38 0.512546 -1.79 -1.72 999A-29X-2W 51-57 260.44 9.52 0.512498 -2.73 -2.66 999A-29X-4W 25-29 263.17 9.83 0.512555 -1.63 -1.55 999A-29X-6W 5-10 265.98 10.15 0.512414 -4.37 -4.29 265.98 10.15 0.512420 -4.25 -4.17 999A-29X-6W 66-70 266.58 10.22 0.512489 -2.92 -2.84 999A-29X-6W 104-109 266.97 10.26 0.512428 -4.09 -4.01 999A-30X-2W 3-8 269.56 10.46 0.512473 -3.22 -3.13 999A-30X-3W 59-63.5 271.61 10.56 0.512421 -4.24 -4.16 999A-30X-4W 8-14 272.61 10.61 0.512506 -2.58 -2.50 999A-30X-5W 2-7 274.05 10.69 0.512495 -2.79 -2.70 999A-30X-6W 105-110 276.58 10.78 0.512634 -0.07 0.01 999A-30X-7W 28-33 277.31 10.80 0.512508 -2.53 -2.45 999A-31X-2W 34-38 279.46 10.87 0.512628 -0.19 -0.11 999A-31X-3W 7-12 280.70 10.91 0.512506 -2.58 -2.50 999A-31X-4W 53-59 282.66 10.98 0.512494 -2.80 -2.72 999A-32X-1W 28-33 287.51 11.14 0.512550 -1.72 -1.63 287.51 11.14 0.512545 -1.81 -1.73 999A-32X-2W 90-94 289.62 11.21 0.512598 -0.78 -0.69 999A-32X-6W 18-23 294.91 11.39 0.512580 -1.14 -1.05 999A-32X-6W 79-84 295.52 11.41 0.512472 -3.25 -3.16 999A-33X-2W 4-10 298.27 11.50 0.512539 -1.94 -1.85 999A-33X-3W 4-8 299.76 11.55 0.512568 -1.37 -1.27 999A-33X-4W 106-111 302.29 11.63 0.512470 -3.29 -3.20 999A-33X-6W 65-69 304.87 11.72 0.512455 -3.56 -3.47 999A-33X-CCW 13-18 305.96 11.77 0.512514 -2.43 -2.33 305.96 11.77 0.512509 -2.53 -2.52 999A-34X-2W 145-150 309.28 12.01 0.512302 -6.55 -6.46 309.28 12.01 0.512319 -6.22 -6.13 309.28 12.01 0.512302 -6.55 -6.45 999A-34X-3W 63-67 309.95 12.06 0.512486 -2.96 -2.87 999A-34X-6W 100-105 314.83 12.41 0.512448 -3.70 -3.61 999A-34X-6W 118-122 315.00 12.43 0.512447 -3.72 -3.63 999A-35X-3W 109-113 320.11 12.80 0.512486 -2.96 -2.86 999A-35X-5W 54-59 322.57 12.98 0.512514 -2.42 -2.32 322.57 12.98 0.512492 -2.85 -2.74 999A-35X-7W 17-21 325.19 13.17 0.512484 -3.00 -2.90 999A-37X-1W 15-19 335.37 13.39 0.512513 -2.44 -2.33 999A-37X-2W 6-11 336.79 13.42 0.512497 -2.76 -2.66 999A-38X-1W 55-60 345.38 13.69 0.512470 -3.29 -3.18 999A-38X-1W 69-73 345.51 13.70 0.512478 -3.12 -3.01 999A-38X-5W 55-60 351.38 14.01 0.512359 -5.45 -5.34

35

Table 2-2. Continued

Site Depth (mbsf) Age (Ma) a 143Nd/144Nd b εNd(o) c εNd(t)

d

999A-39X-4W 94-99 359.87 14.50 0.512338 -5.86 -5.74 999A-42X-3W 133-138 387.75 16.00 0.512387 -4.90 -4.78 1241A-14H-3W 123-128 122.17 5.00 0.512480 -3.09 -3.04 1241A-16H-3W 47-52 140.40 5.50 0.512679 0.80 0.85 1241A-19H-4W 77-82 170.73 6.00 0.512630 -0.16 -0.11 1241A-23H-3W 29-34 206.71 6.50 0.512685 0.91 0.97 1241A-25H-2W 0-5 223.95 7.00 0.512727 1.74 1.80 1241A-26H-1W 58-63 232.48 7.25 0.512685 0.91 0.98 1241A-27H-6W 79-84 249.76 7.75 0.512675 0.72 0.79 1241A-28H-3W 139-144 255.30 8.00 0.512628 -0.20 -0.13 1241A-29H-5W 49-54 266.91 8.25 0.512649 0.21 0.28 1241A-30H-4W 109-114 275.51 8.50 0.512651 0.25 0.33 1241A-31H-4W 19-24 284.12 8.75 0.512671 0.64 0.72 1241A-33H-2W 139-144 301.33 9.25 0.512671 0.64 0.72 1241A-35X-3W 130-135 318.50 9.75 0.512701 1.23 1.31 1241A-40X-7W 40-45 371.10 11.25 0.512608 -0.59 -0.49

a Age models for all four sites are based on biostratigraphic boundaries and age

datums defined Raffi and Flores [1995] at Site 846 and applied to Sites 998 and 999 by Kameo and Bralower [2000] and to Site 1241 by Mix et al. [2003].

b 143Nd/144Nd values analyzed on a given day were corrected by the difference between the average JNdi-1 value for that day and JNdi-1 = 0.512103 (TIMS average at University of Florida).

b εNd(o) = [143Nd/144Nd(sample)/143Nd/144Nd(CHUR) – 1] × 104, where 143Nd/144Nd(CHUR) =

0.512638. c εNd(t) = [143Nd/144Nd(sample (t))/

143Nd/144Nd(CHUR(t)) – 1] × 104. d The 2σ external uncertainty based on normalized repeat analyses of JNdi-1 is ±0.000015, which is equivalent to 0.3 εNd units. Within run uncertainties were consistently less than this value.

36

CHAPTER 3 MIOCENE DEEP WATER CIRCULATION IN THE PACIFIC AND CARIBBEAN:

IMPACTS OF THE CENTRAL AMERICAN SEAWAY AND SOUTHERN HEMISPHERE GLACIATION

Overview

During the Middle Miocene the eastern equatorial Pacific [Vincent, 1981; Mayer et

al., 1986; Farrell et al., 1995; Lyle et al., 1995], Caribbean [Roth et al., 2000], and

possibly the western Atlantic [King et al., 1997] recorded intervals of extensive

carbonate dissolution referred to as “carbonate crash” events [Lyle et al., 1995]. These

intervals of low carbonate mass accumulation can be attributed to either increased

productivity resulting in enhanced decay (oxidation) of organic matter on the seafloor or

a change in global thermohaline circulation that resulted in more corrosive bottom water

in the equatorial region of the Pacific and the Caribbean. Lyle et al. [1995] argued

against increased surface productivity and associated deep water acidity as a cause of

the eastern equatorial Pacific carbonate crash based on an absence of increased

organic carbon (Corg) or opal Mass Accumulation Rates (MARs), as well as a lack of a

negative covariance between carbonate and opal MARs. Instead, Lyle et al. [1995] and

Roth et al. [2000] suggested carbonate crash events in both the Pacific and Caribbean

were related to changes in circulation coinciding with shoaling of the Central American

Seaway (CAS) in the Middle Miocene.

Neodymium (Nd) isotopes recovered from fossil fish teeth provide a technique to

investigate paleo-circulation patterns and test this theory. Nd isotopes are quasi-

conservative tracers of water mass, meaning that the cores of different water masses

have distinct Nd isotopic signatures that can only be altered through water mass mixing

or addition of local weathering inputs [Frank, 2002; Goldstein and Hemming, 2003].

37

Important for this proxy is the fact that the residence time of Nd in the oceans [~600 –

1000 yrs; Elderfield and Greaves, 1982; Piepgras and Wasserburg, 1985; Jeandel et

al., 1995; Tachikawa et al., 1999; Arsouze et al., 2009] is shorter than the modern

ocean mixing time of ~1500 years [Broecker and Peng, 1982]. Fossil fish teeth and

debris have been demonstrated to be a robust paleoceanographic archives of bottom

water Nd isotopes [Elderfield and Pagett, 1986; Martin and Haley, 2000; Thomas et al.,

2003; Martin and Scher, 2004; Thomas, 2004; Scher and Martin, 2006]. A previous

study of Nd isotopes from the eastern equatorial Pacific and Caribbean during Miocene

carbonate crash intervals demonstrated that the most corrosive waters had radiogenic

isotopic values consistent with a North Pacific source and suggested Pacific water

overflowed intermediate to deep sills in the CAS to fill the deep Caribbean basin during

the Caribbean carbonate crash [Newkirk and Martin, 2009].

Although Nd isotopes support the idea that carbonate crash intervals were

genetically related to changes in ocean circulation during the Middle Miocene, the

ultimate cause of these changes has yet to be identified. Based on the relative timing of

the carbonate crash intervals and the history of the CAS, a number of studies suggest

shoaling of the CAS played a major role in the evolution of Caribbean and Pacific

circulation [e.g., Keigwin, 1978; Farrell et al., 1995; Lyle et al., 1995; Collins et al., 1996;

Ling et al., 1997; Haug and Tiedemann, 1998; Frank et al., 1999; Reynolds et al., 2009;

Roth et al., 2000; Haug et al., 2001; Steph et al., 2006; Jain et al., 2007; Groeneveld et

al., 2008; Newkirk and Martin, 2009]. These studies tend to focus on changes in salinity

related to development of the Isthmus of Panama and the resulting impact on the Gulf

Stream and North Atlantic Deep Water (NADW) production, which ultimately impacts

38

Pacific circulation via the global conveyor. However, there are also documented

changes in Southern Ocean conditions during the Middle Miocene that could influence

Pacific circulation. Specifically, there is evidence for strengthening of the Antarctic

Circumpolar Current (ACC) [Shevenell et al., 2004], a marked increase in the production

and export of Antarctic Bottom Water (AABW) [Wright and Miller, 1993], and increased

flow speeds for the Deep Western Boundary Current (DWBC) exiting the ACC near

Chatham Rise [Hall et al., 2003] during the Middle Miocene climate transition (MMCT;

14.2 to 13.8 Ma [Shevenell et al., 2004]) in conjunction with renewed growth of the East

Antarctic Ice Sheet (EAIS) [Kennett and Barker, 1990].

The primary goal of this study was to reconstruct deep water circulation in the

eastern Pacific and Caribbean during the Miocene based on Nd isotopic records for

fossil fish teeth/debris from a longitudinal transect of four Ocean Drilling Project (ODP)

sites in the eastern Pacific and expanded datasets for Caribbean records (Figure 3-1) in

order to assess the relative roles of tropical CAS shoaling versus high latitude Southern

Ocean climate change on Pacific and Caribbean circulation and development of

carbonate crash intervals. A longitudinal transect of deep water Pacific sites runs from

Site 845 (3704 m water depth) at 9° N through Site 846 (3296 m water depth) at 3°S to

Site 1237 (3212 m water depth) at 16°S. These sites are ideally located to study

changes in the boundary between northern and southern sourced deep Pacific waters

along the eastern margin of the basin, as well as the impact of CAS gateway closure on

the Pacific circulation.

We also extended the records from Newkirk and Martin [2009] for Sites 999 (12°N

in the Colombian Basin; 2839 m water depth) and Site 998 (19°N in the Yucatan Basin;

39

3101 m water depth) in the Caribbean Basin to further evaluate the impact of the CAS

on conditions in the Caribbean.Data from Site 1241 (5°N on the Cocos Ridge; 2027 m

water depth) located just across the Isthmus of Panama from Site 999 represent the

endmember for more intermediate depth waters that might have flowed into the

Caribbean as the isthmus shoaled (Figure 3-1). All of these data are used to improve

our understanding of processes driving circulation in the Pacific and Caribbean following

the Mid-Miocene Climatic Optima.

Background

Modern Pacific Deep Water Circulation

Bottom water enters the Pacific Basin from the south as a deep western boundary

current (DWBC) that flows northward below 3000 m depth through deep passageways

in the Southwest Pacific to fill the deep Pacific Ocean basin (Figure 3-2) [Warren, 1973,

1981; Lonsdale, 1976; Carter et al., 1996, 1999; Carter and McCave, 1997; Tsuchiya

and Talley, 1998; Carter and Wilkin¸1999; Orsi et al., 1999]. There is also a branch off

of the ACC that flows north up the eastern side of the Pacific [Lonsdale, 1976; Tsuchiya

and Talley, 1998]. Both of these northward flowing water masses are composed of

Circumpolar Deep Water (CDW), a combination of AABW, Antarctic Intermediate Water

(AAIW), NADW, and Pacific Deep Water (PDW) that mixes in the ACC [Orsi et al.,

1995, 1999; Rintoul et al., 2001].

The Nd isotopic composition of the Pacific has been shown to become

progressively more radiogenic from south to north and from deep to intermediate water

depths [van de Flierdt et al., 2004a]. While transiting northward, CDW, which starts with

an εNd value of -8 [Piepgras and Wasserburg, 1982], accumulates nutrients and loses

oxygen as a result of organic matter decay, and becomes more radiogenic due to

40

mixing with southward flowing PDW [van de Flierdt et al., 2004a]. Near the equator its

Nd isotopic composition is further modified to a value of ~-5.2 [Horikawa et al., 2010] by

boundary exchange with sediment from Papua New Guinea and adjacent volcanic

islands in the Western Equatorial Pacific [Lacan and Jeandel, 2001]; it is also influenced

by reversible scavenging in the water column [Horikawa et al., 2011] leading to the

formation of modified CDW (mCDW).

Once in the North Pacific, most mCDW upwells and flows southward along the

eastern side of the basin at mid-depths (1000 and 3000 m) as PDW (Figure 3-2)

[Schmitz, 1995; Reid, 1997; Kawabe and Fujio, 2010]. During this transit, it continues to

accumulate nutrients and lose oxygen [Matsumoto et al., 2002]. Today re-circulated

PDW obtains a radiogenic Nd isotopic value of ~-2 as a result of reversible scavenging

[Horikawa et al., 2011] and interactions with young circum-Pacific island arcs in the

subarctic North Pacific. This water mass has the lowest δ13C values in the Pacific

[Kroopnick, 1985] and becomes progressively more corrosive with age and southward

flow as a result of organic matter decay (remineralization) and associated increasing

CO2 concentrations.

Overlying PDW is the northward flowing Antarctic Intermediate Water (AAIW) in

the southern portion of the basin and southward flowing North Pacific Intermediate

Water (NPIW) in the northern portion of the basin with the dividing line located in the

equatorial region today [Tsuchiya and Talley, 1998]. These water masses flow at

intermediate depths between ~500 to 1000 m (Figure 3-3) and can be distinguished by

the fact that AAIW has lower nutrients and higher oxygen as a result of substantial

atmospheric interaction and incomplete nutrient depletion in the Southern Ocean

41

[Kroopnick, 1985]. In contrast, NPIW forms from upwelling in the northwest Pacific and

experiences relatively little to no atmospheric interaction [Talley, 1993].

The Peru and Guatemala Basins of the eastern Pacific (Figure 3-1) are isolated

from northward flowing mCDW by the East Pacific Rise (EPR), which extends up to

~3000 m water depth. These two basins are separated by the Cocos Ridge, and the

Galapagos Spreading Center. In these basins the deepest waters are composed of a

northward flowing branch of CDW that originates from the circumpolar current in the

southeast Pacific Basin and mixes with PDW on its northward transit, creating a water

mass with intermediate phosphate values between CDW and PDW and εNd values of ~-

3 [Horikawa et al., 2010]. Some of this water continues northward into the southern

portion of the Peru Basin, flowing through the Peru-Chile Trench (4900 m) near the

Nazca Ridge and across a 3900 m sill [Lonsdale, 1976; Tsuchiya and Talley, 1998].

Today there is a hydrologic divide in the Peru Basin; the deep water in the

northern portion of the basin is PDW, while the deep water in the southern portion is

CDW. These northern and southern sourced deep water masses in the Peru Basin have

similar temperatures and salinities, but PDW is slightly less dense and tends to override

CDW, which has higher oxygen concentrations and lower phosphate and nitrate

concentrations [Tsuchiya and Talley, 1998]. Like the northern portion of the Peru Basin,

the Guatemala Basin (Figure 3-1) to the north is filled with PDW.

Shoaling History of the Central American Seaway

Although the closure history of the CAS is controversial [Keigwin, 1978; Marshall,

1985; Webb, 1985; Keller et al., 1989; Coates et al., 1992; Haug and Tiedemann, 1998;

Haug et al., 2001; Steph et al., 2006; Groeneveld et al., 2008, Montes et al., 2012], the

general consensus is that closure started with uplift and arc formation approximately

42

~17 to 15 Ma [Coates et al., 1992, 2003, 2004], with shoaling to a depth of ~2000 m by

~15.9–15.1 Ma [Duque-Caro, 1990; ages adjusted to Shackleton et al., 1995], and

continued shoaling to an upper bathyal depth of ~1000 m between 12-10.2 Ma [Duque-

Caro, 1990]. At this point, it is suggested that the minimum width of the gateway was

~200 km [Montes et al., 2012]. Unlike Duque-Caro [1990], Coates et al. [2003]

suggested the seaway was shallower, and that an archipelago had developed by ~12

Ma. Oceanographic evidence suggests the seaway restricted deep and intermediate

water flow from the Pacific into the Caribbean beginning at ~7.9-7.6 Ma based on shifts

in benthic foraminiferal diversity and paleoproductivity differences between the

Caribbean and eastern equatorial Pacific [Jain et al., 2007]. Coates et al. [2004] also

suggested continued shoaling related to collision of the volcanic arc and northern

Colombia created a barrier to deep and intermediate waters by 7 Ma, consistent with

the observed divergence of benthic organisms on either side of the CAS [Keller et al.,

1989; Collins et al., 1996]. Continued shoaling to a depth of <100 m has been dated at

4.6 Ma [Haug and Tiedemann, 1998], with final closure occurring between ~4.2 – 2.4

Ma [Keigwin, 1978; Marshall, 1985; Webb, 1985; Keller et al., 1989; Coates et al., 1992;

Haug and Tiedemann, 1998; Haug et al., 2001; Steph et al., 2006; Groeneveld et al.,

2008].

Material and Methods

Core Descriptions and Age Model

The age models applied to all six ODP sites in the study are based on the

biostratigraphy of Sites 845 and 846 by Raffi and Flores [1995], which used zonal

boundaries defined by Martini [1971] and Bukry [1973]. The ages of these zonal

boundaries and magnetic reversals [Mayer et al., 1992] were recalibrated to ages

43

determined by Shackleton et al. [1995] using orbital tuning. The age/depth models

based on this biostratigraphy for Sites 1237 and 1241 come from Mix et al. [2003], while

age models for Caribbean Sites 998 and 999 come from Kameo and Bralower [2000].

Pacific sites were sampled at a 0.25 Ma interval from 2.5 to 14.6 Ma for Sites 845 and

846, 2.5 to 14 Ma at Site 1237, and 2.5 to 11.5 Ma (maximum depth) at Site 1241. The

Caribbean sites were sampled at an interval of 0.25 Ma from 2.5 to ~18 Ma, with a

higher resolution (~0.1 Ma) from 10 to 12 Ma. Ten samples were analyzed for Site 999

from 0.015 to 0.13 Ma.

Sample Preparation and Nd Isotope Measurements

Sediment samples were oven dried, disaggregated and wet sieved prior to hand-

picking fossil fish teeth from the >125 μm fraction. Fish teeth and debris were dissolved

in aqua regia, without prior cleaning [Martin et al., 2010] and then dried prior to a two

step chemical separation to isolate Nd. Bulk rare earth elements (REEs) were

separated on primary quartz columns using Mitsubishi cation exchange resin [Scher

and Martin, 2004] or Teflon columns using Eichrom TRUspecTM Resin, both used HCl

as the eluent. Nd was then isolated from bulk REEs using Eichrom LNspecTM resin with

HCl as the eluent on volumetrically calibrated Teflon columns [Pin and Zalduegui,

1997]. The total blank for these techniques is 14 pg Nd.

Nd isotopic ratios were measured on a Nu Plasma Multi-Collector-Inductively

Coupled Plasma-Mass Spectrometer (MC-ICP-MS) at the University of Florida. The

standard JNdi-1 was run between every 4 to 6 unknown samples. All of the JNdi-1

values analyzed during a given day were averaged and compared to the published

value of JNdi-1 (0.512115 ± 0.000007) [Tanaka et al., 2000] to determine the amount of

correction to apply to the unknown samples analyzed that day. A drift correction was not

44

applied to the data because variations throughout a day of analysis did not indicate a

consistent drift. The 2σ error for the Nu MC-ICP-MS based on the variability of

normalized JNdi-1 analyzed over the past several years is ± 0.000014, which is

equivalent to 0.27 εNd units (εNd represents the deviation in parts per 104 of the

143Nd/144Nd ratio of the sample relative to the chondritic uniform reservoir with

143Nd/144Nd=0.512638 [Jacobsen and Wasserburg, 1980]).

δ13C Sample Preparation and Measurements

Carbon isotopic ratios were measured on benthic foraminifera (Cibicidoides)

picked from the >63 μm fraction of the same samples picked for fossil fish teeth. These

foraminifera were cleaned in 15% H2O2, sonicated, and dried with methanol. Samples

were then reacted in 100% orthophosphoric acid at 70°C using a Finnigan-MAT Kiel III

carbonate preparation device. The evolved CO2 gas was measured online with a

Finnigan-MAT 252 mass spectrometer at the University of Florida. Isotopic results are

reported in standard delta notation relative to Vienna Pee Dee Belemnite (VPDB).

Analytical precision is estimated to be ± 0.018 ‰ for δ13C (1 standard deviation of

standards run with samples).

Results

From 14 – 9 Ma the Nd isotopic ratios at all three eastern Pacific deepwater sites

increased from values of ~-3.5 at 14 Ma to -1.7 at 9 Ma (Figure 3-4; Table 3-1). This

shift in Nd isotopes is accompanied by a decrease from ~1.5 to ~0.5 in δ13C values from

14 to 9 Ma (Figure 3-5). Starting at 9 Ma εNd values of the two more northerly sites

(Sites 845 and 846) diverge from values of the more southerly site (Site 1237). From 9

to 2.5 Ma values for the northern sites fluctuate between -0.4 and -1.5 with an average

45

value of ~1.3. For Site 1237 εNd values decrease to ~ -3 (Figure 3-4) and fluctuate

between -2 and -3.3 (Figure 3-4).

Records from Sites 998 and 999 in the Caribbean start at values of ~-2 to -3 at 18

Ma, decrease to ~-5.5 by ~15 Ma, and then increase to values of -2 to -3.5 at ~13.5 Ma

(Figure 3-6; Table 3-2). The εNd values in the Caribbean from ~14 to ~9 Ma range from -

5 to 0.4 with the most radiogenic values occurring from 12 to 10.7 Ma and overlapping

with values recorded at intermediate depth Site 1241 from the Pacific. From ~10.7 to 2.5

Ma, Nd isotopic values gradually decrease from values of -2 to values of -8 to -9. The

values in the youngest section of the record (0.13 to 0.01 Ma) fluctuate between -10.5

and -5.7.

Discussion

Circulation in the Eastern Pacific

Based on records of Nd isotopes from fossil fish teeth/debris there are two distinct

patterns of circulation in the eastern Pacific: 1) the period of increasing εNd values at all

sites in the middle to late Miocene (14 to 9 Ma), and 2) the late Miocene to Pliocene

interval (9 to 2.5 Ma), which is characterized by distinct values at the northern versus

the southern sites (Figure 3-4).

The Middle to Late Miocene Transition (14 – 9 Ma)

The interval of increasing εNd values at all three sites in the eastern Pacific

coincides with increases in the value of mCDW observed in Fe-Mn crust VA13/2 in the

equatorial Central Pacific [Ling et al., 1997] and PDW from crust D4-13A Alaska in the

North Pacific [van de Flierdt et al., 2004a] (Figure 3-7). However, the magnitude of the

increase is only 0.65 εNd units in deep waters of the North Pacific (D4-13A Alaska [van

de Flierdt et al., 2004b]) and 0.8 εNd units in the deep Central Pacific (VA 13/2 [Ling et

46

al., 1997]), compared to 1.5 εNd units in the deep eastern Pacific, indicating the shift in

the eastern Pacific may in part reflect changes in the composition of the water sourcing

PDW, but additional processes are also required. Initially, Ling et al. [1997] attributed

the isotopic increase in the Central Pacific to a decrease in the supply of a less

radiogenic, presumably Atlantic-sourced, water mass flowing into the Pacific through the

CAS. Although some General Circulation Model’s (GCM’s) support this interpretation

[Maier-Reimer et al., 1990; Mikolajewicz et al., 1993; Mikolajewicz and Crowley, 1997;

Nof and van Gorder, 2003], Nd isotopic results of Newkirk and Martin [2009]

demonstrate continued flow from the Pacific into the Caribbean Basin during this time,

in agreement with other GCM results [Nisancioglu et al., 2003; Klocker et al., 2005;

Schneider and Schmittner, 2006; Lunt et al., 2007].

Assuming the input value of the CDW end-member from the Southern Ocean did

not change throughout this interval, as illustrated by Frank et al. [2002] and O’Nions et

al. [1998], alternative explanations for the increase in εNd values in the Central and

North Pacific include either (1) the value of mCDW flowing into the North Pacific

changed as a result of increased weathering inputs with radiogenic values, and/or (2)

the mixing boundary between northward flowing CDW and southward flowing PDW

shifted to a position farther south and deeper in the water column. The first option is

consistent with Ling et al. [2005], who revised their earlier ideas on the source of the

seawater Nd isotopic increase [Ling et al., 1997] and instead attributed it to increased

volcanic activity and weathering of volcanic arcs around the Pacific. Increased Pacific

Island Arc production is observed from 18-11 Ma and 6-0 Ma [Lee et al., 1995], but this

timing is not consistent with the pattern of change observed throughout the Pacific.

47

Also, an increase in εNd driven by volcanic inputs within the North Pacific should

produce the largest change in the north, but that region records the smallest change,

with a slightly larger change in the Central Pacific and an even larger change in the

eastern Pacific (Figure 3-7). Therefore, it is plausible that ~0.65 εNd units of the shift

observed in the Central and eastern Pacific could be accounted for by increased

volcanic input in the north Pacific, but the remainder of the documented increases must

be the result of another process.

Given that only part of the Nd isotopic increase observed in all three eastern

Pacific deepwater sites can be attributed to a change in the isotopic value of the source

waters, the remainder of the increase must be due to a change in the relative

contributions of more and less radiogenic components that comprise the deep water in

this area. One potential scenario is southward progression and deepening of the mixing

boundary between northward flowing CDW and southward flowing PDW (Figures 3-8

and 3-9), which could be related to a documented increase in the flux of DWBC entering

the southern Pacific over the 5 my interval of increasing Nd isotopes [Wright et al.,

1991; Hall et al., 2003]. Evidence supporting this increased flux includes strengthening

of the ACC at the middle Miocene climate transition (MMCT; 14.2 to 13.8 Ma)

[Shevenell et al., 2004], increased production and export of AABW based on Southern

Ocean hiatuses [Wright and Miller, 1993], and increased flow speeds for the DWBC

exiting the ACC near Chatham Rise based on sortable silt [Hall et al., 2003]. All of these

changes in the Southern Ocean coincided approximately with renewed growth of the

EAIS [Kennett and Barker, 1990; Flower and Kennett, 1995; Shevenell et al., 2004], as

well as increased NADW production [Wright and Miller, 1996; Poore et al., 2006]. All of

48

these changes would have further strengthened the ACC, leading to an increased flux

of southern sourced waters into Pacific Ocean.

This explanation is a bit counterintuitive because an increased flux of CDW with

εNd values of ~-7.5 to -8 [O’Nions et al., 1998; Frank et al., 2002] into the Pacific would

presumably decrease the Nd isotopic value of deep water in the equatorial region

[Horikawa et al., 2011; van de Flierdt et al., 2004a], yet Nd isotopic data from the central

equatorial region became more radiogenic at this time [Ling et al., 1997]. This paradox

can be reconciled if the increased flux of southern sourced deep waters into the Pacific

led to a greater return flow of radiogenic (εNd ~ -2) southward flowing water out of the

north Pacific, thereby shifting the boundary with CDW deeper in the water column and

further southward (Figures 3-8 and 3-9). Flower and Kennett [1995] document the

encroachment of a low-δ13C water mass (PDW) from the north to ~41°S at 2000 to 3000

m depth in the southwest Pacific beginning ~ 14 to 13.6 Ma. A similar idea was

proposed by van de Flierdt et al. [2004b], who attributed a shift in the Nd isotopic

composition of seawater in the southwestern Pacific to the introduction of southward

flowing waters from the equatorial region as a result of increased intensity of Southern

Ocean circulation along with the shoaling of the Indonesian Seaway. Greater southward

penetration of PDW would result in a more southerly onset of mixing between CDW and

PDW, thereby altering the initial Nd isotopic composition of northward flowing CDW

prior to modification in the equatorial region, producing even more radiogenic mCDW

that ultimately upwells in the north Pacific.

The coherent shift to more radiogenic values at all three deep water sites (Sites

845, 846, and 1237) suggests that the sites were bathed by a water mass with a greater

49

proportion of PDW by the end of the transition (Figure 3-9). This scenario suggests the

core of PDW shifted to a depth of ~3000 m, which is the approximate depth of the sills

controlling the flow of deep water into the eastern equatorial Pacific basins. This shift

would have diminished the proportion of CDW that could enter these basins. These

changes on the eastern side of the Pacific are consistent with expansion of the core of

PDW in the southwestern Pacific documented by Flower and Kennett [1995] in the late

Miocene.

Southward encroachment of corrosive PDW from the north Pacific also agrees

well with the observed progression of the carbonate crash in the eastern equatorial

Pacific. Specifically, sites 845 and 1241 located just north of the equator recorded a

decrease in carbonate MARs beginning ~12 Ma, while extensive dissolution did not start

until ~11.2 Ma at Site 846 just south of the equator [Farrell et al., 1995; Lyle et al.,

1995], and Site 1237 at about 20˚S never recorded extensive carbonate dissolution [Mix

et al., 2003] (Figure 3-1). Similar to sites just north of the equator, carbonate dissolution

began at ~12 Ma in the Caribbean Basin, but the Pacific event lasted ~1 my longer. In

both basins, onset of carbonate crash events is marked by a shift to more radiogenic Nd

isotopic values, indicating the wedge of corrosive PDW arrived in the eastern equatorial

Pacific, bathed Sites 845 and 1241, and flowed through the CAS before the deeper

portion of the wedge reached Site 846 (Figure 3-9). Therefore, the distribution of the Nd

isotopic shift in the EEP, the shift in δ13C values, and the southward progression of the

onset of the carbonate crash suggest that the eastern Pacific basins were dominated by

a mixture of CDW and PDW that was dominated by CDW prior to 14 Ma, and the

50

proportion of PDW increased progressively as the boundary between the two water

masses became deeper and extended southward (Figure 3-9).

Late Miocene to Pliocene (8.5 to 2.5 Ma) Circulation

The Late Miocene to Pliocene circulation pattern was established around 9 to 8.5

Ma with the divergence of the more southerly site (Site 1237) to less radiogenic values

(Figure 3-4). Plausible causes for the ~1 εNd unit decrease at Site 1237 include changes

in volcanic ash input, increased dust from the Atacama Desert, and/or a change in

ocean circulation. Site 1237 recorded an increase in ash layer frequencies in samples

younger than ~9 Ma, with intervals of amplified frequency from ~7.5 to 6, and from 5 to

1 Ma, possibly due to a major uplift of the northern Andes [Mix et al., 2003]. The range

of εNd values reported for Andean volcanic material is highly variable and surprisingly

some samples could drive down the seawater εNd value [Futa and Stern, 1988; Rogers

and Hawkesworth, 1989]; however, changes in the isotopic record do not coincide with

changes in volcanic ash layer frequency and/or ash layer thickness reported by Mix et

al. [2003]. This is especially true after ~5 Ma when all of the sites in the eastern Pacific

recorded an increase in ash deposition, while the seawater εNd signals are relatively

constant (Figure 3-4).

In terms of dust impacts, Jones et al. [1994] and Ling et al. [2005] demonstrated

that Fe-Mn nodules and crusts preserve the overlying water mass Nd isotopic

composition even when it is distinct from the signal of eolian material in the sediment.

Data from the Peru Basin can also be used to argue the observed variations in

seawater Nd isotopes were not driven by eolian inputs. The Atacama Desert is the

prevailing source of dust to the Peru Basin today [Molina-Cruz, 1977]. Increases in dust

accumulation are observed at ~14 Ma at Site 1237, with a more prominent increase

51

after ~8 Ma. These increases are related to aridification of the Atacama Desert [Alpers

and Brimhall, 1988], possibly in response to Andean uplift. Observations of modern

dispersal of dust throughout the Peru Basin [Molina-Cruz, 1977] suggest that Sites 846

and 1237 receive similar dust inputs, yet the offset in Nd isotopes observed in the late

Miocene is also observed today [Horikawa et al., 2011]. Therefore, volcanic and dust

inputs cannot explain the divergence in Nd isotopes; instead we evaluate this

divergence in terms of water mass distribution and circulation.

The resulting distribution of εNd produces a pattern similar to the modern eastern

Pacific in which the northern portion of the Peru Basin is dominated by more radiogenic

PDW, while the southern portion of the basin is dominated by a mixture of PDW and the

less radiogenic CDW sourced from the southeast Pacific [Horikawa et al., 2011],

suggesting development of modern hydrographic conditions in the late Miocene. The

deeper boundary between PDW and CDW in the northern Peru Basin established by

8.5 Ma appears to become a permanent hydrographic feature of the eastern Pacific. In

this configuration, the CDW-PDW boundary was shallower than Site 1237 so the

mixture of water was dominated by CDW in the southern portion of the Peru basin in the

Miocene and probably had lower nutrients and higher dissolved oxygen to accompany

the less radiogenic Nd isotopic composition compared to the mixture dominated by

PDW in the northern portion of the basin, as it does today [Tsuchiya and Talley, 1998;

Horikawa et al., 2011] (Figure 3-9). The increased flow of CDW into the southern portion

of the eastern Pacific occurred as a result of NADW production and the establishment of

the Western Antarctic Ice Sheet (WAIS) [Kennett and Barker, 1990] which led to further

52

strengthening of the ACC and resulted in a shift of the boundary between PDW and

CDW to a shallower depth in the southern portion of the Peru Basin.

While onset of the carbonate crash in the Pacific and Caribbean was driven by the

introduction of corrosive north Pacific waters due to deepening of the PDW/CDW

boundary, termination of the crash intervals appears to have different causes in the two

regions. In the Pacific, carbonate deposition recovered in some of the sites as they

moved under the equatorial productivity belt, despite the fact they were still bathed by

corrosive PDW. Increased carbonate MARs and Corg MARs at ~8 Ma at Pacific Site 846

mark the end of the carbonate crash and record enhanced surface productivity and an

associated increase in carbonate rain rate which was large enough to allow for

carbonate deposition within a corrosive environment [Lyle et al., 1995]. In contrast, Site

845 in the Guatemala basin was never located in the productivity belt and never

recovered from the carbonate crash.

History of the Caribbean Basin

The Nd isotopic values of ~ -2 to -5.5 at sites 998 and 999 prior to 14 Ma (Figure

3-10) plot between known values for Pacific and Atlantic waters, suggesting the

Caribbean Basin was open to exchange with both ocean basins with waters entering

over sills at intermediate to upper-deepwater depths. As noted in Newkirk and Martin

[2009], the shift to more radiogenic values after ~14 Ma suggests a decrease in the

contribution from the Atlantic and an increase in the flux from the Pacific with the basin

filling almost completely from the Pacific, or possibly with a small fraction of Atlantic

waters plus a very radiogenic shallow Pacific source.

General ocean circulation models of the Caribbean show a west to east flow

pattern with the bulk of waters shallower than 2000 m water depth flowing from the

53

Pacific into the Caribbean with an open CAS due to steric height differences between

the Pacific and Atlantic [Lunt et al., 2007; Mikolajewicz and Crowley, 1997; Nisancioglu

et al., 2003; Prange and Schultz, 2004; Schneider and Schmittner, 2006; Steph et al.,

2006]. Shoaling of the CAS to a depth of ~2000 m occurred between ~15.9 – 15.1 Ma

(ages adjusted to Shackleton et al., [1995]) based on benthic foraminiferal assemblages

from Atrato Basin located in northwest South America, which roughly coincides with the

timing of the shift to more radiogenic Nd isotopic values.

The εNd values in the Caribbean from ~14 to ~9 Ma are similar to those recorded in

the Miocene Pacific (Figure 3-4) [Abouchami et al., 1997; Ling et al., 1997, 2005;

O’Nions et al., 1998; Frank et al., 1999; Reynolds et al., 1999; van de Flierdt et al.,

2004a; Newkirk and Martin, 2009, and this study]. In particular, the Caribbean sites

record a general increase in Nd isotopes that is similar to the increase observed at Sites

845 and 846 in the Pacific and has been attributed to encroachment of northern-

sourced PDW that ultimately flowed into the Caribbean Basin.

From 12 to 10.7 Ma Caribbean εNd values are highly variable with several spikes to

values of ~0. Similar radiogenic values are recorded at intermediate depth Site 1241,

located directly across the CAS in the Pacific (Figure 3-1). These high εNd values in the

Caribbean Basin have been suggested to reflect an influx of upper-PDW to intermediate

water through the CAS [Newkirk and Martin, 2009] and, therefore, a change in depth of

the source waters from within the Pacific, rather than an increase in the relative

proportion of Pacific water entering the basin. Duque-Caro [1990] suggested the CAS

had shoaled to upper bathyal depths (~1000 m) by this time, and Montes et al. [2012]

54

suggested the opening between Panama and Columbia was restricted to less than ~200

km.

Given that Pacific values remain high and largely unchanged throughout this

interval (Figure 3-10), decreasing Nd isotopic values from ~10.7 to 2.5 Ma in the

Caribbean argue for a decreased flux of Pacific water into the Caribbean, consistent

with the ending of the carbonate crash in the Caribbean Basin at ~10 Ma as a result of

shoaling of the CAS. Shoaling ultimately created the barrier to flow of corrosive Pacific

deep and intermediate waters into the Caribbean Basin and allowed for the recovery

from the carbonate crash. Therefore, the changes in circulation and the carbonate crash

were the results of changes in southern hemisphere climate, while the recovery of the

carbonate crash in the Caribbean was the result of shoaling of the CAS.

It is interesting to note that even after estimated dates for final closure of the CAS,

(~4.2 – 2.4 Ma) [Keigwin, 1978; Keller et al., 1989; Coates et al., 1992; Haug et al.,

2001; Steph et al., 2006; Groeneveld et al., 2008] εNd values at Sites 998 and 999 do

not appear to represent a pure north Atlantic source. Even Pleistocene samples that are

less than 0.13 Ma continue to record values up to -6. The depth of flow into the

Caribbean Basin on the Atlantic side is controlled by the Lesser Antilles and Aves Swell,

which subsided from ~600 m to a modern depth of ~1200 m in the middle Miocene

[Pinet et al., 1985], producing two deep passageways, the Windward and Anegada-

Jungfern Passages with modern sill depths of 1540 m and 1800 m respectively [Pinet et

al., 1985]. These passages limit Atlantic inflow to intermediate depth and shallower

water masses. Possible sources therefore include North Atlantic Intermediate Water

(NAIW) with εNd values of -13 to -12.5 at 2000 m on the Atlantic side of the Lesser

55

Antilles [Foster et al., 2007]; Glacial North Atlantic Intermediate Water (GNAIW) with

values of -9.7 at 1790 m at Blake Nose [Gutjahr et al., 2008]; and Antarctic Intermediate

Water (AAIW) with values of -6 to -9 [Jeandel, 1993; Pahnke et al., 2008].

Using a composite dissolution index, Haddad and Droxler [1996] noted a shift

between carbonate preservation and dissolution during the Pleistocene in the

Caribbean Basin as a result of shifts in circulation on glacial/interglacial time scales. In

this scenario, times of increased NADW production during interglacials result in

encroachment of corrosive AAIW into the Caribbean Basin leading to in carbonate

dissolution, while decreased/diminished NADW production during glacials resulted in

introduction of less corrosive glacial North Atlantic Intermediate Water (GNAIW) into the

Caribbean and enhanced carbonate preservation. Unfortunately at this time we do not

have the resolution to fully evaluate this theory, but the range of Pleistocene Nd

isotopes in the Caribbean do suggest variations between northern and southern

sourced waters. By 7 Ma the decreasing Nd isotopic trend in the Caribbean had

reached the -6 εNd unit value obtained during times of known closure; thus, our data

suggest continued flow of Pacific water through the CAS until at least 7 Ma, but beyond

that the data do not allow us to more precisely define the age of final closure.

Summary

Nd isotopic analyses of Miocene to Pliocene fossil fish teeth/debris from a depth

transect in the eastern equatorial Pacific and Caribbean Basin reveal that the dominant

deep water mass in the eastern equatorial Pacific at ~14 Ma was southern-sourced

CDW that was over-ridden by northern-sourced, PDW. At that time the Caribbean

Basin was filled with an undefined mixture of Pacific and Atlantic waters. Decreasing

δ13C values from 14 Ma to 9 Ma at all three eastern Pacific deepwater sites are

56

accompanied by increasing Nd isotopic values, consistent with increased contributions

from an older water mass sourced from the north, indicating enhanced southward flow

of PDW. Timing of this reorganization of circulation coincides with intensification of

circulation in the Southern Ocean as a result of southern hemisphere glaciation.

The Nd isotopic values and the pattern of onset of the carbonate crash in the

northern equatorial Pacific sites and the Caribbean indicate that the introduction of the

PDW into the equatorial region was responsible for the observed changes in carbonate

MARs in these region. These findings suggest the increased depth and southward

progression of corrosive, radiogenic PDW into the equatorial region of the Pacific

appears to be a response to strengthening of fluxes of circumpolar water into the

Southern Pacific following climate change in the southern hemisphere. Therefore, this

interpretation attributes changes in circulation in the eastern equatorial Pacific, as well

as some of the changes in the Caribbean and the onset of the carbonate crash in both

regions to climate-driven changes in the Southern Ocean rather than shoaling of the

CAS.

57

Figure 3-1. Bathymetric map of the Pacific Basin and Caribbean Basin [Schlitzer, R.,

Ocean Data View, http://odv.awi.de, 2012] illustrating the different basins of the eastern Pacific and the associated sills controlling the flow of deep water.

58

Figure 3-2. Map of the Pacific Ocean [Schlitzer, R., Ocean Data View, http://odv.awi.de, 2012] illustrating the flow paths of Kawabe and Fujio [2010]. Dark blue represents the northward flowing Circumpolar Deep Water (CDW), while the green represents the southward flowing Pacific Deep Water (PDW). Open circles represent areas of upwelling of CDW to form PDW. The Ocean Drilling Program sites (Sites 845, 846, 1237 and 1241) are represented with red circles. Sites VA13/2, CD29-2, D4-13A Alaska, and 13D-27A Kamchatka are represented with the purple circles.

59

Figure 3-3. North-South Seawater profile for the central Pacific determined using the phosphate concentration profile for the modern Pacific ocean (modified from Horikawa et al. [2010]).

60

Figure 3-4. Seawater Nd isotopic values for the deep water sites (845, 846, and 1237) of the eastern Pacific compared to published values (shade fields) for the central Pacific [Ling et al., 1997; ] and the north Pacific [van de Flierdt et al., 2004a]. The middle Miocene Climatic Optimum (MMCO) [Woodruff and Savin, 1989], middle Miocene Climate Transition (MMCT) [Shevenell et al., 2004], East Antarctic Ice Sheet development (EAIS) [Kennett and Barker, 1990], Caribbean Carbonate Crash (Caribbean C.C.) [Roth et al., 2000], Pacific Carbonate Crash (Pacific C.C.) [Lyle et al., 1995], and West Antarctic Ice Sheet development [Kennett and Barker, 1990] are illustrated at the top of the figure.

North Pacific

Central Pacific

Age (Ma)2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Nd

(T)

-4

-3

-2

-1

0

Site 846

Site 845

Site 1237

Pliocene Late Miocene Middle Miocene

MMCO

MMCT

EAISWAIS

Caribbean C.C.Pacific C.C.

61

Figure 3-5. Nd isotopic values for the eastern Pacific versus the carbon record. Shaded box represents the carbonate crash interval(s) for the Pacific. Site 845 recorded the onset of carbonate dissolution at ~12 Ma. Dissolution at Site 846 began at ~11.2 Ma, while extensive carbonate dissolution was not recorded at Site 1237.

62

North Pacific deep water

Atlantic Deep Water

2

No Data

Age (Ma)

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Nd

(T)

-10

-8

-6

-4

-2

0

2Site 999

Site 998

Site 1241

Central Pacific deep water

Figure3- 6. Seawater Nd isotopic data for Ocean Drilling Program Sites 998 and 999 in the Caribbean Basin Compared to the published values (shaded fields) for the Pacific [Ling et al., 1997, Frank et al., 1999; van de Flierdt et al., 2004a,b] and the Atlantic [Burton et al., 1997, 1999; O’Nions et al., 1998; Reynolds et al., 1999; Thomas and Via, 2007].

63

Figure 3-7. Changes in the Nd isotopic composition of water masses from the beginning of the record to 9 Ma from the central equatorial Pacific (Va 13/2) [Ling et al., 1997], the north Pacific (D4-13A Alaska) [van de Flierdt et al., 2004b], and the three eastern Pacific sites from this study (Sites 845, 846, and 1237).

Age (Ma)6 8 10 12 14 16

Nd

-5

-4

-3

-2

Nd = 0.8

Nd = 0.65

-4

-3

-2

-1

0

Nd = 1.5

Site 1237

Site 845

Site 846

D4-13A Alaska (North Pacific)

VA 13/2 (Central Pacific)

64

Figure 3-8. North-South Seawater profiles for the central Pacific showing the shifts in the boundaries between Pacific water mass. Boundaries were determined using the phosphate concentration profiles of the modern oceans and then adjusted based on the Nd isotopic composition observed in the Pacific from 14 to 2.5 Ma, with Nd isotopic values for Va 13/2 [Ling et al., 1997], CD29-2 [Ling et al., 1997],and D4-13A Alaska and 13D-27A Kamchatka [van de Flierdt et al., 2004]. The Nd isotopic value recorded at each site is listed next to the site name for each interval. (a) Seawater profile illustrating the position of PDW for the Late Miocene to Pleistocene (8.5 to 2.5 Ma) which is the most similar to the modern distribution of water masses, (b) profile for the beginning of the Middle to Late Miocene (14 Ma), and (c) profile for the end of the Middle to Late Miocene Mode (9 Ma).

65

Figure 3-9. North-South Seawater profiles for the eastern Pacific showing the shifts in the boundaries between Pacific water mass. Boundaries were determined using the phosphate concentration profiles of the modern oceans and then adjusted based on the Nd isotopic composition observed in the Pacific from 14 to 2.5 Ma, with Nd isotopic values for Va 13/2 [Ling et al., 1997], CD29-2 [Ling et al., 1997], D4-13A Alaska and 13D-27A Kamchatka [van de Flierdt et al., 2004], and sites 845, 846, 1237, 1241 from this study. The Nd isotopic value recorded at each site is listed next to the site name for each interval. (a) Seawater profile illustrating the position of PDW during the Late Miocene to Pleistocene (8.5 to 2.5 Ma) which is the most similar to the modern distribution of water masses, (b) profile for the end of the Middle to Late Miocene Mode (9 Ma), and (c) profile for the beginning of the Middle to Late Miocene (14 Ma).

66

Figure 3-10. Seawater Nd isotopic data for Ocean Drilling Program Sites 845, 846, 1237, and 1241 in the eastern Pacific, Sites 998 and 999 in the Caribbean Basin, and BM1969.897 [Reynolds et al., 1999] located in the Straits of Florida near Blake Nose. The shaded fields represent published values for the Pacific [Ling et al., 1997, Frank et al., 1999; van de Flierdt et al., 2004a,b] and the Atlantic [Burton et al., 1997, 1999; O’Nions et al., 1998; Reynolds et al., 1999; Thomas and Via, 2007].

Pliocene

North Pacific deep water

Atlantic Deep Water

2

Site 999BM1969.897Site 998Site 1241Site 846Site 1237Site 845

Late Miocene Middle Miocene Late MiocenePleistocene

No Data

Age (Ma)

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Nd(T

)

-10

-8

-6

-4

-2

0

2

Central Pacific deep water

67

Table 3-1. Nd isotopic results for the Pacific (Sites 845, 846, 1237, and 1241)

Site Depth (mcd)

Age (Ma)a 143Nd/144Ndb εNd(0)

c εNd(T)d 2σ

845A-5H-6W 126-128 48.38 2.50 0.512569 -1.3 -1.3 0.3

845A-6H-1W 138-140 51.07 2.75 0.512583 -1.1 -1.0 0.3

845A-6H-3W 91-93 53.60 3.00 0.512574 -1.2 -1.2 0.3

845A-6H-5W 39-41 56.08 3.25 0.512563 -1.5 -1.4 0.3

845A-6H-6W 140-142 58.59 3.50 0.512580 -1.1 -1.1 0.3

845A-7H-1W 116-118 61.13 3.75 0.512564 -1.4 -1.4 0.3

845A-7H-3W 64-66 63.61 4.00 0.512541 -1.9 -1.9 0.3

845A-7H-5W 15-17 66.12 4.25 0.512570 -1.3 -1.3 0.3

845A-7H-6W 115-117 68.62 4.50 0.512557 -1.6 -1.5 0.3

845A-8H-2W 123-125 73.64 5.00 0.512572 -1.3 -1.2 0.3

845A-8H-4W 91-93 76.32 5.25 0.512571 -1.3 -1.3 0.3

845A-9H-1W 35-37 82.47 5.60 0.512558 -1.6 -1.5 0.3

845A-9H-3W 12-14 85.24 5.75 0.512558 -1.6 -1.5 0.3

845A-9H-5W 92-94 89.04 6.00 0.512569 -1.3 -1.3 0.3

845A-10H-1W 33-35 92.42 6.25 0.512535 -2.0 -1.9 0.3

845A-10H-3W 64-66 95.73 6.50 0.512535 -2.0 -1.9 0.3

845B-10H-3W 48-50 102.42 7.00 0.512580 -1.1 -1.1 0.3

845A-11H-2W 140-142 105.77 7.25 0.512577 -1.2 -1.1 0.3

845A-11H-5W 80-82 109.67 7.50 0.512537 -2.0 -1.9 0.3

845A-12H-1W 20-22 114.14 7.76 0.512586 -1.0 -0.9 0.3

845A-12H-3W 120-122 118.14 8.00 0.512538 -2.0 -1.9 0.3

845A-12H-6W 90-92 122.34 8.25 0.512546 -1.8 -1.7 0.3

845A-13H-2W 80-82 127.02 8.50 0.512571 -1.3 -1.2 0.3

845A-13H-5W 145-147 132.17 8.75 0.512540 -1.9 -1.8 0.3

845A-14H-2W 23-25 137.32 9.00 0.512544 -1.8 -1.7 0.3

845A-14H-5W 57-59 142.16 9.24 0.512550 -1.7 -1.6 0.3

845B-14H-2W 137-139 145.88 9.50 0.512522 -2.3 -2.2 0.3

845A-15H-2W 97-99 149.44 9.75 0.512524 -2.2 -2.1 0.3

845A-15H-5W 5-7 153.02 10.00 0.512513 -2.4 -2.4 0.3

845A-15H-7W 67-69 156.64 10.25 0.512510 -2.5 -2.4 0.3

845A-16H-2W 45-47 160.29 10.51 0.512500 -2.7 -2.6 0.3

845A-16H-4W 96-98 163.80 10.75 0.512492 -2.8 -2.7 0.3

845A-16H-6W 128-130 167.12 10.98 0.512495 -2.8 -2.7 0.3

845A-17H-2W 142-144 171.84 11.25 0.512500 -2.7 -2.6 0.3

845A-18H-3W 13-15 182.02 11.75 0.512473 -3.2 -3.1 0.3

845A-18H-6W 98-100 187.37 12.00 0.512493 -2.8 -2.7 0.3

845A-19H-4W 61-63 194.88 12.25 0.512508 -2.5 -2.4 0.3

845A-20H-2W 1-3 202.45 12.50 0.512481 -3.1 -2.9 0.3

845A-20H-7 7-9 210.01 12.75 0.512465 -3.4 -3.2 0.3

845A-21H-5W 14-16 217.48 13.00 0.512469 -3.3 -3.2 0.3

68

Table 3-1. Continued

Site Depth (mcd)

Age (Ma)a 143Nd/144Ndb εNd(0)

c εNd(T)d 2σ

845A-22H-2W 69-71 225.03 13.25 0.512470 -3.3 -3.2 0.3

845A-22H-6W 41-43 230.75 13.50 0.512466 -3.4 -3.2 0.3

845A-23X-3W 20-22 235.54 13.75 0.512479 -3.1 -3.0 0.3

845A-23X-6W 48-50 240.32 14.00 0.512438 -3.9 -3.8 0.3

845A-25X-1W 5-7 251.79 14.60 0.512478 -3.1 -3.0 0.3

846B-10H-1W 41-43 95.97 2.50 0.512584 -1.1 -1.0 0.3

846B-11H-2W 35-37 107.86 2.75 0.512572 -1.3 -1.3 0.3

846B-13H-1W 110-112 129.06 3.25 0.512568 -1.4 -1.3 0.3

846B-14H-1W 22-24 138.38 3.48 0.512612 -0.5 -0.5 0.3

846B-14H-1W 89-91 139.05 3.50 0.512611 -0.5 -0.5 0.3

846B-15H-4W 137-139 156.08 4.00 0.512569 -1.4 -1.3 0.3

846B-16H-3W 77-79 163.98 4.25 0.512618 -0.4 -0.3 0.3

846B-17H-2W 52-54 172.78 4.50 0.512578 -1.2 -1.1 0.3

846B-18H-3W 90-92 184.76 4.75 0.512614 -0.5 -0.4 0.3

846B-19H-2W 115-119 194.97 5.00 0.512622 -0.3 -0.3 0.3

846B-19H-2W 117-119 194.98 5.00 0.512624 -0.3 -0.2 0.3

846B-20H-1W 48-50 204.09 5.25 0.512575 -1.2 -1.2 0.3

846B-21H-1W 100-102 215.46 5.50 0.512572 -1.3 -1.2 0.3

846B-22H-2W 58-60 229.54 5.75 0.512560 -1.5 -1.5 0.3

846B-23X-3W 8-10 243.59 6.00 0.512511 -2.5 -2.4 0.3

846B-24X-2W 143-145 254.54 6.25 0.512592 -0.9 -0.8 0.3

846B-25X-3W 143-145 265.44 6.50 0.512567 -1.4 -1.3 0.3

846D-25X-6W 62-64 272.03 6.75 0.512587 -1.0 -0.9 0.3

846B-27X-6W 47-49 288.28 7.25 0.512564 -1.4 -1.4 0.3

846B-28X-4W 102-104 295.53 7.43 0.512603 -0.7 -0.6 0.3

846B-29X-2W 132-134 302.13 7.75 0.512607 -0.6 -0.5 0.3

846B-29X-6W 61-67 307.44 8.09 0.512575 -1.2 -1.2 0.3

846B-30X-3W 110-116 313.13 8.43 0.512561 -1.5 -1.4 0.3

846B-31X-1W 77-83 319.40 8.81 0.512559 -1.5 -1.5 0.3

846B-31X-2W 101-107 321.14 8.92 0.512535 -2.0 -1.9 0.3

846B-31X-4W 98-104 324.11 9.10 0.512547 -1.8 -1.7 0.3

846B-31X-6W 101-107 327.14 9.29 0.512541 -1.9 -1.8 0.3

846B-32X-4W 103-109 333.76 9.69 0.512546 -1.8 -1.7 0.3

846B-32X-6W 27-33 336.00 9.82 0.512542 -1.9 -1.8 0.3

846B-32X-6W 94-100 336.67 9.87 0.512544 -1.8 -1.7 0.3

846B-33X-1W 101-107 338.94 10.06 0.512566 -1.4 -1.3 0.3

846B-33X-2W 31-37 339.74 10.12 0.512538 -2.0 -1.9 0.3

846B-33X-4W 144-150 343.87 10.45 0.512539 -1.9 -1.8 0.3

846B-33X-6W 30-36 345.73 10.60 0.512522 -2.3 -2.2 0.3

69

Table 3-1. Continued

Site Depth (mcd)

Age (Ma)a 143Nd/144Ndb εNd(0)

c εNd(T)d 2σ

846B-34X-1W 140-146 348.93 10.78 0.512500 -2.7 -2.6 0.3

846B-34X-3W 121-127 351.74 10.93 0.512538 -2.0 -1.8 0.3

846B-34X-4W 111-117 353.14 11.01 0.512533 -2.0 -1.9 0.3

846B-34X-7W 21-27 356.74 11.21 0.512553 -1.7 -1.6 0.3

846B-35X-1W 37-39 357.48 11.25 0.512510 -2.5 -2.4 0.3

846B-35X-4W 40-42 362.01 11.50 0.512509 -2.5 -2.4 0.3

846B-36X-2W 128-134 369.61 11.92 0.512503 -2.6 -2.5 0.3

846B-36X-3W 128-134 371.11 12.00 0.512493 -2.8 -2.7 0.3

846B-36X-6W 132-134 375.63 12.25 0.512529 -2.1 -2.0 0.3

846B-37X-2W 93-99 378.86 12.43 0.512532 -2.1 -2.0 0.3

846B-38X-1W 14-16 386.25 12.83 0.512505 -2.6 -2.5 0.3

846B-38X-4W 144-150 392.07 13.16 0.512502 -2.7 -2.5 0.3

846B-40X-1W 123-125 406.64 14.14 0.512458 -3.5 -3.4 0.3

846B-40X-4W 42-44 410.33 14.56 0.512453 -3.6 -3.5 0.3

1237C-5H-5W 116-120 47.88 2.50 0.512499 -2.7 -2.7 0.3

1237B-6H-2W 118-120 51.92 2.75 0.512485 -3.0 -3.0 0.3

1237C-6H-3W 115-117 55.95 3.00 0.512504 -2.6 -2.6 0.3

1237C-6H-6W 75-77 60.08 3.25 0.512494 -2.8 -2.8 0.3

1237C-7H-5W 35-37 68.15 3.75 0.512510 -2.5 -2.5 0.3

1237B-8H-3W 97-99 72.23 4.00 0.512505 -2.6 -2.6 0.3

1237D-5H-1W 58-60 76.06 4.24 0.512479 -3.1 -3.1 0.3

1237D-5H-4W 78-80 80.79 4.50 0.512490 -2.9 -2.9 0.3

1237C-9H-3W 143-145 86.43 4.75 0.512472 -3.2 -3.2 0.3

1237D-6H-4W 37-39 92.28 5.01 0.512518 -2.3 -2.3 0.3

1237C-10H-4W 73-75 97.78 5.25 0.512467 -3.3 -3.3 0.3

1237D-7H-4W 112-114 103.47 5.50 0.512520 -2.3 -2.2 0.3

1237D-8H-3W 28-30 110.68 5.75 0.512496 -2.8 -2.7 0.3

1237C-12H-4W 117-119 118.62 6.00 0.512494 -2.8 -2.8 0.3

1237B-13H-4W 87-89 126.47 6.25 0.512500 -2.7 -2.6 0.3

1237C-13H-6W 48-50 133.04 6.50 0.512522 -2.3 -2.2 0.3

1237B-14H-5W 77-79 138.05 6.75 0.512509 -2.5 -2.5 0.3

1237C-14H-3W 133-135 140.71 7.00 0.512536 -2.0 -1.9 0.3

1237C-14H-5W 48-50 142.87 7.25 0.512515 -2.4 -2.3 0.3

1237B-15H-3W 43-45 145.01 7.50 0.512512 -2.5 -2.4 0.3

1237C-15H-3W 62-64 149.50 7.75 0.512496 -2.8 -2.7 0.3

1237C-15H-6W 52-54 153.93 8.00 0.512511 -2.5 -2.4 0.3

1237C-16H-4W 37-39 162.10 8.50 0.512499 -2.7 -2.6 0.3

1237B-17H-4W 97-99 168.30 9.09 0.512550 -1.7 -1.6 0.3

1237B-17H-6W 63-65 170.97 9.34 0.512558 -1.6 -1.5 0.3

70

Table 3-1. Continued

Site Depth (mcd)

Age (Ma)a 143Nd/144Ndb εNd(0)

c εNd(T)d 2σ

1237C-17H-5W 143-145 175.13 10.01 0.512541 -1.9 -1.8 0.3

1237B-18H-3W 63-65 176.54 10.20 0.512533 -2.1 -2.0 0.3

1237B-18H-4W 97-99 178.39 10.45 0.512538 -2.0 -1.9 0.3

1237B-18H-5W 133-135 180.26 10.71 0.512512 -2.5 -2.4 0.3

1237C-18H-3W 62-64 182.45 11.01 0.512529 -2.1 -2.0 0.3

1237C-18H-4W 88-90 184.22 11.25 0.512521 -2.3 -2.2 0.3

1237C-18H-5W 122-124 186.07 11.50 0.512492 -2.9 -2.7 0.3

1237B-19H-4W 2-4 188.92 11.75 0.512506 -2.6 -2.5 0.3

1237C-19H-2W 142-144 193.03 12.24 0.512518 -2.4 -2.2 0.3

1237B-20H-1W 17-19 194.83 12.46 0.512511 -2.5 -2.4 0.3

1237C-19H-5W 66-68 196.77 12.69 0.512501 -2.7 -2.6 0.3

1237B-20H-2W 77-79 196.94 12.71 0.512485 -3.0 -2.9 0.3

1237B-20H-3W 136-138 199.04 12.96 0.512496 -2.8 -2.6 0.3

1237B-20H-5W 51-53 201.20 13.21 0.512490 -2.9 -2.8 0.3

1237B-20H-6W 111-113 203.31 13.46 0.512473 -3.2 -3.1 0.3

1237C-20H-4W 62-64 205.76 13.75 0.512493 -2.8 -2.7 0.3

1237C-20H-5W 121-123 207.86 14.00 0.512497 -2.8 -2.6 0.3

1241A-7H-4W 115-116 64.10 2.55 0.512676 0.7 0.8 0.3

1241A-8H-2W 94-96 72.05 2.85 0.512705 1.3 1.3 0.3

1241A-8H-6W 62-64 77.76 3.06 0.512671 0.6 0.7 0.3

1241A-9H-4W 12-14 84.68 3.29 0.512645 0.1 0.2 0.3

1241A-10H-1W 107-108 91.70 3.52 0.512696 1.1 1.2 0.3

1241A-10H-4W 126-128 96.39 3.68 0.512621 -0.3 -0.3 0.3

1241A-11H-3W 54-56 104.77 3.96 0.512658 0.4 0.4 0.3

1241A-12H-1W 2-4 112.13 4.16 0.512698 1.2 1.2 0.3

1241A-12H-5W 4-6 118.19 4.32 0.512670 0.6 0.7 0.3

1241A-13H-4W 15-16 126.68 4.55 0.512647 0.2 0.2 0.3

1241A-14H-3W 124-126 136.34 4.81 0.512694 1.1 1.1 0.3

1241A-14H-3W 125-126 136.35 4.81 0.512492 -2.9 -2.8 0.3

1241A-15H-3W 86-88 146.01 5.06 0.512731 1.8 1.9 0.3

1241A-16H-3W 47-48 156.66 5.31 0.512691 1.0 1.1 0.3

1241A-17H-3W 9-10 166.35 5.54 0.512740 2.0 2.0 0.3

1241A-19H-4W 80-81 190.02 6.06 0.512642 0.1 0.1 0.3

1241A-22H-3W 119-121 221.39 6.52 0.512665 0.5 0.6 0.3

1241A-23H-3W 29-30 232.12 6.68 0.512697 1.1 1.2 0.3

1241A-25H-2W 4-5 252.80 6.98 0.512739 2.0 2.0 0.3

1241A-26H-1W 58-59 262.52 7.24 0.512697 1.1 1.2 0.3

1241A-26H-1W 59-61 262.53 7.24 0.512716 1.5 1.6 0.3

1241A-26H-7W 19-20 271.18 7.49 0.512760 2.4 2.4 0.3

71

Table 3-1. Continued

Site Depth (mcd)

Age (Ma)a 143Nd/144Ndb εNd(0)

c εNd(T)d 2σ

1241A-26H-7W 20-22 271.19 7.49 0.512697 1.2 1.2 0.3

1241A-27H-6W 83-84 280.82 7.76 0.512687 1.0 1.0 0.3

1241A-28H-3W 140-141 287.76 7.96 0.512640 0.0 0.1 0.3

1241A-29H-5W 49-50 300.11 8.32 0.512661 0.4 0.5 0.3

1241A-30H-4W 109-110 309.86 8.61 0.512663 0.5 0.6 0.3

1241A-31H-4W 19-20 319.41 8.90 0.512683 0.9 1.0 0.3

1241A-32H-3W 80-82 331.05 9.16 0.512630 -0.2 -0.1 0.3

1241A-33H-2W 143-144 342.73 9.39 0.512683 0.9 1.0 0.3

1241A-34H-2W 54-56 352.58 9.58 0.512664 0.5 0.6 0.3

1241A-35X-3W 130-131 361.96 9.76 0.512713 1.5 1.5 0.3

1241A-36X-3W 91-93 371.74 9.96 0.512726 1.7 1.8 0.3

1241A-38X-2W 30-31 391.43 10.40 0.512688 1.0 1.1 0.3

1241A-39X-1W 91-93 401.28 10.63 0.512642 0.1 0.2 0.3

1241A-40X-7W 41-43 420.74 11.07 0.512610 -0.5 -0.4 0.3

1241A-41X-5W 41-42 428.59 11.25 0.512620 -0.4 -0.3 0.3

1241A-42X-5W 31-32 439.45 11.50 0.512651 0.3 0.4 0.3 a Age models are described in the methods section. b 143Nd/144Nd values analyzed on a given day were corrected by the difference between

the average JNdi-1 value for that day and JNdi-1 = 0.512103 (TIMS average at University of Florida).

b εNd(o) = [143Nd/144Nd(sample)/143Nd/144Nd(CHUR) – 1] × 104, where 143Nd/144Nd(CHUR) =

0.512638. c εNd(t) = [143Nd/144Nd(sample (t))/

143Nd/144Nd(CHUR(t)) – 1] × 104. d The 2σ external uncertainty based on normalized repeat analyses of JNdi-1 is ±0.000015, which is equivalent to 0.3 εNd units. Within run uncertainties were consistently less than this value.

72

Table 3-2. Nd isotopic results for the Caribbean Basin (Sites 998A and 999A)

Site Depth (mbsf)

Age (Ma) a 143Nd/144Nd b εNd(o)

c εNd(T) d 2σ

998A-1H-1W 10-16 0.13 0.15 0.512102 -10.5 -10.4 0.3

998A-6H-3W 35-38 50.18 2.50 0.512204 -8.5 -8.4 0.3

998A-6H-6W 43-46 54.76 2.75 0.512164 -9.2 -9.2 0.3

998A-7H-2W 146-149 59.29 3.00 0.512208 -8.4 -8.4 0.3

998A-7H-6W 5-8 63.88 3.25 0.512200 -8.5 -8.5 0.3

998A-8H-2W 116-119 68.49 3.50 0.512190 -8.7 -8.7 0.3

998A-8H-5W 122-125 73.05 3.75 0.512215 -8.2 -8.2 0.3

998A-9H-2W 80-83 77.63 4.00 0.512229 -8.0 -7.9 0.3

998A-9H-5W 88-91 82.21 4.25 0.512248 -7.6 -7.6 0.3

998A-10H-2W 46-49 86.79 4.50 0.512272 -7.1 -7.1 0.3

998A-10H-5W 54-57 91.37 4.75 0.512228 -8.0 -8.0 0.3

998A-11H-2W 12-15 95.95 5.00 0.512285 -6.9 -6.8 0.3

998A-11H-5W 20-23 100.53 5.25 0.512294 -6.7 -6.7 0.3

998A-11H-6W 130-133 103.13 5.39 0.512298 -6.6 -6.6 0.3

998A-12H-2W 1-4 105.34 5.75 0.512324 -6.1 -6.1 0.3

998A-12H-3W 67-70 107.50 6.00 0.512328 -6.0 -6.0 0.3

998A-12H-4W 140-143 109.73 6.25 0.512287 -6.8 -6.8 0.3

998A-12H-6W 60-63 111.93 6.50 0.512324 -6.1 -6.1 0.3

998A-13H-1W 76-79 114.09 6.75 0.512324 -6.1 -6.1 0.3

998A-13H-2W 146-149 116.29 7.00 0.512402 -4.6 -4.5 0.3

998A-13H-4W 67-70 118.55 7.26 0.512318 -6.2 -6.2 0.3

998A-13H-5W 138-141 120.76 7.51 0.512451 -3.6 -3.6 0.3

998A-14H-1W 7-10 122.90 7.75 0.512463 -3.4 -3.4 0.3

998A-14H-2W 73-76 125.05 8.00 0.512324 -6.1 -6.1 0.3

998A-14H-3W 146-149 127.28 8.25 0.512335 -5.9 -5.8 0.3

998A-14H-5W 66-69 129.48 8.50 0.512361 -5.4 -5.3 0.3

998A-14H-6W 137-140 131.69 8.87 0.512507 -2.6 -2.5 0.3

998A-15H-1W 21-26 132.53 8.95 0.512312 -6.4 -6.3 0.3

998A-15H-3W 107-113 136.40 9.33 0.512335 -5.9 -5.8 0.3

998A-15H-5W 27-32 138.60 9.52 0.512328 -6.0 -6.0 0.3

998A-16H-1W 54-59 142.37 9.83 0.512391 -4.8 -4.8 0.3

998A-16H-3W 125-129 146.08 10.14 0.512363 -5.4 -5.3 0.3

998A-16H-4W 32-37 146.65 10.18 0.512413 -4.4 -4.3 0.3

998A-16H-4W 45-50 146.78 10.19 0.512469 -3.3 -3.2 0.3

998A-16H-5W 25.5-30 148.07 10.30 0.512390 -4.8 -4.8 0.3

998A-16H-6W 32-36 149.64 10.46 0.512412 -4.4 -4.3 0.3

998A-16H-6W 125-130 150.58 10.59 0.512503 -2.6 -2.6 0.3

998A-17H-1W 21-26 151.54 10.73 0.512508 -2.5 -2.5 0.3

998A-17H-1W 32-37 151.65 10.75 0.512557 -1.6 -1.5 0.3

998A-17H-1W 77-82 152.10 10.82 0.512510 -2.5 -2.4 0.3

73

Table 3-2. Continued

Site Depth (mbsf)

Age (Ma) a 143Nd/144Nd b εNd(o)

c εNd(T) d 2σ

998A-17H-1W 105-110 152.38 10.86 0.512478 -3.1 -3.0 0.3

998A-17H-2W 25-30 153.08 10.98 0.512452 -3.6 -3.6 0.3

998A-17H-2W 54-60 153.37 11.02 0.512488 -2.9 -2.8 0.3

998A-17H-2W 126-131 154.09 11.14 0.512461 -3.5 -3.4 0.3

998A-17H-4W 55-60 156.38 11.50 0.512586 -1.0 -0.9 0.3

998A-17H-5W 2-7 157.35 11.65 0.512452 -3.6 -3.5 0.3

998A-17H-5W 134-139 158.67 11.82 0.512503 -2.6 -2.5 0.3

998A-17H-6W 26-31 159.09 11.87 0.512649 0.2 0.3 0.3

998A-17H-6W 81-87 159.64 11.93 0.512606 -0.6 -0.5 0.3

998A-17H-CCW 2-7 160.65 12.04 0.512644 0.1 0.2 0.3

998A-18X-1W 105-111 161.88 12.17 0.512432 -4.0 -3.9 0.3

998A-18X-3W 32-36 164.14 12.42 0.512469 -3.3 -3.2 0.3

998A-19X-1W 53-58 166.76 12.70 0.512430 -4.1 -4.0 0.3

998A-19X-5W 24-28 172.46 13.50 0.512457 -3.5 -3.4 0.3

998A-20X-2W 32-36 177.74 14.05 0.512446 -3.8 -3.6 0.3

998A-20X-5W 60-63 182.52 14.50 0.512518 -2.3 -2.2 0.3

998A-21X-5W 56-59 192.08 15.41 0.512358 -5.5 -5.3 0.3

998A-22X-1W 13-16 195.25 15.71 0.512408 -4.5 -4.4 0.3

998A-22X-3W 22-25 198.34 16.00 0.512447 -3.7 -3.6 0.3

998A-23X-2W 64-66 206.96 16.50 0.512448 -3.7 -3.6 0.3

998A-24X-2W 146-149 217.38 17.00 0.512432 -4.0 -3.9 0.3

998A-25X-3W 77-79 227.79 17.50 0.512411 -4.4 -4.3 0.3

998A-26X-4W 8-10 238.19 18.00 0.512526 -2.2 -2.0 0.3

999A-1H-1W 59-61 0.60 0.00 0.512346 -5.7 -5.7 0.3

999A-1H-1W 122-124 1.23 0.00 0.512221 -8.1 -8.1 0.3

999A-1H-1W 132-134 1.33 0.00 0.512242 -7.7 -7.7 0.3

999A-1H-1W 142-144 1.43 0.00 0.512308 -6.4 -6.4 0.3

999A-1H-2W 40-42 1.91 0.00 0.512290 -6.8 -6.8 0.3

999A-1H-2W 93-95 2.44 0.01 0.512248 -7.6 -7.6 0.3

999A-1H-3W 131-132 4.32 0.01 0.512291 -6.8 -6.8 0.3

999A-1H-3W 133-135 4.34 0.01 0.512262 -7.3 -7.3 0.3

999A-1H-4W 12-14 4.63 0.01 0.512237 -7.8 -7.8 0.3

999A-1H-4W 73-75 5.24 0.01 0.512187 -8.8 -8.8 0.3

999A-9H-6W 93-97 82.55 2.51 0.512247 -7.6 -7.6 0.3

999A-10H-5W 69-71 90.30 2.75 0.512206 -8.4 -8.4 0.3

999A-11H-4W 58-60 98.19 3.00 0.512247 -7.6 -7.6 0.3

999A-12H-3W 55-58 106.17 3.25 0.512226 -8.0 -8.0 0.3

999A-13H-2W 47-50 114.09 3.50 0.512291 -6.8 -6.7 0.3

999A-14H-1W 38-41 122.00 3.75 0.512251 -7.5 -7.5 0.3

74

Table 3.2 Continued

Site Depth (mbsf)

Age (Ma) a 143Nd/144Nd b εNd(o)

c εNd(T) d 2σ

999A-14H-6W 82-85 129.97 4.00 0.512322 -6.2 -6.1 0.3

999A-15H-5W 70-72 137.84 4.25 0.512346 -5.7 -5.7 0.3

999A-16H-3W 147-150 145.09 4.50 0.512335 -5.9 -5.9 0.3

999A-17H-1W 69-71 150.79 4.75 0.512285 -6.9 -6.8 0.3

999A-17H-5W 42-46.5 156.56 5.00 0.512313 -6.3 -6.3 0.3

999A-18H-2W 117-120 162.29 5.25 0.512341 -5.8 -5.8 0.3

999A-18H-6W 90-95 168.03 5.50 0.512338 -5.9 -5.8 0.3

999A-19H-4W 14-19 173.77 5.75 0.512350 -5.6 -5.6 0.3

999A-20H-1W 88-93 179.51 6.00 0.512321 -6.2 -6.1 0.3

999A-20H-5W 63-65 185.24 6.25 0.512787 2.9 3.0 0.3

999A-21H-1W 137-139 189.48 6.43 0.512222 -8.1 -8.1 0.3

999A-21H-6W 108-110 196.69 6.75 0.512320 -6.2 -6.2 0.3

999A-23X-2W 34-39 202.47 7.00 0.512300 -6.6 -6.5 0.3

999A-23X-6W 8-13 208.21 7.25 0.512370 -5.2 -5.2 0.3

999A-24X-3W 62-67 213.95 7.50 0.512343 -5.8 -5.7 0.3

999A-24X-7W 36-41 219.69 7.75 0.512393 -4.8 -4.7 0.3

999A-25X-4W 100-105 225.43 8.00 0.512334 -5.9 -5.9 0.3

999A-26X-2W 14-19 231.17 8.25 0.512372 -5.2 -5.1 0.3

999A-26X-5W 128-131 236.91 8.50 0.512353 -5.6 -5.5 0.3

999A-27X-3W 52-57 242.65 8.75 0.512387 -4.9 -4.8 0.3

999A-28X-1W 18-23 248.91 8.98 0.512411 -4.4 -4.4 0.3

999A-28X-3W 32-36 252.04 9.10 0.512347 -5.7 -5.6 0.3

999A-28X-4W 84-88 254.06 9.18 0.512372 -5.2 -5.1 0.3

999A-28X-5W 55-59 255.27 9.22 0.512484 -3.0 -2.9 0.3

999A-29X-1W 72-76 259.14 9.38 0.512558 -1.6 -1.5 0.3

999A-29X-2W 51-57 260.44 9.52 0.512519 -2.3 -2.3 0.3

999A-29X-4W 25-29 263.17 9.83 0.512565 -1.4 -1.3 0.3

999A-29X-6W 5-10 265.98 10.15 0.512427 -4.1 -4.0 0.3

999A-29X-6W 66-70 266.58 10.22 0.512499 -2.7 -2.6 0.3

999A-29X-6W 104-109 266.97 10.26 0.512432 -4.0 -3.9 0.3

999A-30X-1W 128-132 269.30 10.45 0.512427 -4.1 -4.0 0.3

999A-30X-2W 3-8 269.56 10.46 0.512483 -3.0 -2.9 0.3

999A-30X-3W 59-63.5 271.61 10.56 0.512431 -4.0 -4.0 0.3

999A-30X-4W 8-14 272.61 10.61 0.512516 -2.4 -2.3 0.3

999A-30X-5W 2-7 274.05 10.69 0.512505 -2.6 -2.5 0.3

999A-30X-6W 105-110 276.58 10.78 0.512659 0.4 0.5 0.3

999A-30X-7W 28-33 277.31 10.80 0.512518 -2.3 -2.3 0.3

999A-31X-2W 34-38 279.46 10.87 0.512640 0.0 0.1 0.3

999A-31X-3W 7-12 280.70 10.91 0.512516 -2.4 -2.3 0.3

999A-31X-4W 53-59 282.66 10.98 0.512506 -2.6 -2.5 0.3

75

Table 3.2 Continued

Site Depth (mbsf)

Age (Ma) a 143Nd/144Nd b εNd(o)

c εNd(T) d 2σ

999A-32X-1W 28-33 287.51 11.14 0.512551 -1.7 -1.6 0.3

999A-32X-2W 90-94 289.62 11.21 0.512610 -0.5 -0.5 0.3

999A-32X-6W 18-23 294.91 11.39 0.512590 -0.9 -0.8 0.3

999A-32X-6W 79-84 295.52 11.41 0.512482 -3.0 -3.0 0.3

999A-33X-2W 4-10 298.27 11.50 0.512565 -1.4 -1.3 0.3

999A-33X-3W 4-8 299.76 11.55 0.512580 -1.1 -1.0 0.3

999A-33X-4W 106-111 302.29 11.63 0.512480 -3.1 -3.0 0.3

999A-33X-6W 68-69 304.87 11.72 0.512463 -3.4 -3.3 0.3

999A-33X-CCW 13-18 305.96 11.77 0.512512 -2.5 -2.4 0.4

999A-34X-2W 145-150 309.28 12.01 0.512318 -6.3 -6.2 0.4

999A-34X-3W 63-67 309.95 12.06 0.512498 -2.7 -2.6 0.3

999A-34X-6W 100-105 314.83 12.41 0.512462 -3.4 -3.3 0.3

999A-34X-6W 118-122 315.00 12.43 0.512459 -3.5 -3.4 0.3

999A-35X-3W 109-113 320.11 12.80 0.512498 -2.7 -2.6 0.3

999A-35X-5W 54-59 322.57 12.98 0.512508 -2.5 -2.4 0.4

999A-35X-7W 17-21 325.19 13.17 0.512496 -2.8 -2.7 0.3

999A-37X-1W 15-19 335.37 13.39 0.512525 -2.2 -2.1 0.3

999A-37X-2W6-11 336.79 13.42 0.512507 -2.6 -2.5 0.3

999A-38X-1W 55-60 345.38 13.69 0.512480 -3.1 -3.0 0.3

999A-38X-1W 69-73 345.51 13.70 0.512490 -2.9 -2.8 0.3

999A-38X-5W 55-60 351.38 14.01 0.512369 -5.2 -5.1 0.3

999A-39X-4W 94-99 359.87 14.50 0.512347 -5.7 -5.6 0.3

999A-40X-4W 64-66 369.15 15.00 0.512352 -5.6 -5.5 0.3

999A-41X-4W 21-23 378.42 15.50 0.512407 -4.5 -4.4 0.3

999A-42X-3W 133-138 387.76 16.00 0.512396 -4.7 -4.6 0.3

999A-44X-1W 82-84 403.33 16.50 0.512400 -4.6 -4.5 0.3

999A-45X-3W 107-110 416.19 17.00 0.512426 -4.1 -4.0 0.3

999A-46X-5W 122-125 429.04 17.50 0.512414 -4.4 -4.2 0.3

999A-47X-5W 140-142 438.81 17.88 0.512468 -3.3 -3.2 0.3 a Age models are described in the methods section. b 143Nd/144Nd values analyzed on a given day were corrected by the difference between

the average JNdi-1 value for that day and JNdi-1 = 0.512103 (TIMS average at University of Florida).

b εNd(o) = [143Nd/144Nd(sample)/143Nd/144Nd(CHUR) – 1] × 104, where 143Nd/144Nd(CHUR) =

0.512638. c εNd(t) = [143Nd/144Nd(sample (t))/

143Nd/144Nd(CHUR(t)) – 1] × 104. d The 2σ external uncertainty based on normalized repeat analyses of JNdi-1 is ±0.000015, which is equivalent to 0.3 εNd units. Within run uncertainties were consistently less than this value.

76

CHAPTER 4 TRANSCONTINENTAL CONNECTION OF THE AMAZON RIVER BASED ON CEARA

RISE SEAWATER RECORDS

Overview

The Ceara Rise in the topical Atlantic is located midway between northern and

southern deep-water sources. Ocean Drilling Program (ODP) sites on the rise cover a

range of depths from ~3000-4300 m (Figure 4-1), and are frequently used to track the

boundary between Northern Component Water (NCW; proto North Atlantic Deep Water)

and Antarctic Bottom Water (AABW) in the Oligocene/Miocene [Flower et al., 1997] and

early Pliocene [Billups et al., 1998] based on carbon isotopes. King et al. [1997] used

magnetic susceptibility, natural gamma, and digital reflectance of lithology to determine

the percent carbonate deposition at Ceara Rise. Episodes of carbonate dissolution

between 14 and 10.5 Ma were observed and were suggested to be the result of a

shallower lysocline related to a shift in the boundary between NCW and AABW. More

dissolution is consistent with a greater proportion of older, more corrosive AABW at

these depths. The return of carbonate deposition starting ~10.5 Ma was attributed to a

shift in the AABW-NCW boundary as a result of increased NCW production based on

results from Wright and Miller [1992] [King et al., 1997]. Most basin-wide to global

studies of the production rate of NCW have been based on the gradient between δ13C in

the Atlantic and Pacific [Wright et al., 1996; Poore et al., 2006], but that proxy is

complicated in the Miocene by the fact that the gradient is very small. In addition, δ13C

is altered by changes in the nutrient signals which changes along the flow path of a

water mass and therefore indicates the age rather than the composition of a water mass

[Kroopnick, 1985]. To better constrain how the boundary between AABW and NCW

changed as a result of changes in the production rate of both NADW and AABW we

77

used Nd isotopes which are considered to be quasi-conservative tracers of water mass

[Frank, 2002; Goldstein and Hemming, 2003]. Like a true conservative tracer of water

mass, the modern residence time of Nd in the oceans [~600 – 1000 yrs; Elderfield and

Greaves, 1982; Piepgras and Wasserburg, 1985; Jeandel et al., 1995; Tachikawa et al.,

1999; Arsouze et al., 2009] is shorter than the ocean mixing time [~1500 years;

Broecker and Peng, 1982]; however, they reflect the initial signal of the source region,

but can be modified by weathering inputs during circulation. Fossil fish teeth and debris

were analyzed because they have been shown to be robust paleoceanographic

archives of bottom water Nd isotopes [Elderfield and Pagett, 1986; Martin and Haley,

2000; Thomas et al., 2003; Martin and Scher, 2004; Thomas, 2004; Scher and Martin,

2006].

The depth transect of ODP Sites 925, 926 and 929 (~3000 to 4300 m) on Ceara

Rise encompasses the modern mixing boundary between AABW and NADW (based on

temperature and salinity), which falls close to the deepest site (929) (Figure 4-2). These

two water masses have distinct Nd isotopic compositions with modern values of -13.5

for NADW and -8 for AABW [Piepgras and Wasserburg, 1982; Piepgras and

Wasserburg, 1987; Jeandel, 1993] and Miocene values of ~-11 for NCW and ~-8 for

AABW [O’Nions et al., 1998; Frank et al., 2002]. Documentation of seawater Nd isotopic

records at this depth transect should allow us to reconstruct the position of this interface

through time and determine whether there is a direct correlation between carbonate

dissolution and the influx of AABW over the Ceara Rise. It would also provide

information about relative NCW and AABW production rates.

78

Early results of seawater Nd isotopic analyses indicated the seawater values at

these open ocean sites were strongly influenced by a signal from Amazon sediments.

Thus, this study was refocused to understand the impact of Amazon outflow on

seawater Nd isotopes and the history of the Amazon basin preserved in the isotopic

record. Andean uplift events during the Late Miocene have been identified as important

factors in the evolution of the Amazon River drainage basin [Hoorn, 1993; Hoorn, 1994;

Hoorn et al., 1995; Campbell et al., 2006; Figueiredo et al., 2009, 2010; Hoorn et al.,

2010]. Prior to these events, Hoorn (1995) argued that the early Miocene proto-

Amazon River drained into the Caribbean Basin rather than connecting to the Atlantic

and Andean uplift is credited with producing a transcontinental connection with

development of the Amazon Basin and outflow to the Atlantic [Castro et al., 1978;

Campbell, 1992; Hoorn, 1993; Hoorn, 1994; Hoorn et al., 1995; Campbell et al., 2006;

Figueiredo et al., 2009, 2010; Campbell, 2010; Hoorn et al., 2010]; however, the timing

of this evolution is debated.

In its modern configuration the sediments that comprise the Amazon Fan are

derived from both the Andes and the lowlands of Brazil, but the detrital fraction is

dominated by the Andean highlands [Gibbs, 1967; Milliman, 1979; Damuth et al., 1988;

McDaniel et al., 1997; Dobson et al., 1997; Dobson et al., 2001]. Estimates for

development of the Amazon River and Fan range from the end of the middle/late

Miocene to the late Pliocene [Castro et al., 1978; Campbell, 1992; Hoorn, 1993; Hoorn,

1994; Hoorn et al., 1995; Campbell et al., 2006; Figueiredo et al., 2009, 2010;

Campbell, 2010; Hoorn et al., 2010]. Figueiredo et al. [2009] used biostratigraphic,

isotopic, and well log data to show that during the early to middle Miocene the sediment

79

reaching the Amazon Fan region was derived from the lowland cratons, but from the

late Miocene to present the sediment has predominantly been derived from the Andes

and attributed the fan development to the onset of a transcontinental connection of the

Amazon. Prior to the transcontinental connection during the late middle Miocene, Hoorn

et al. [1995] argues Andes uplift altered the northwest Amazonia region causing proto-

Amazon River formation but argues this proto-Amazon does not actually flow east to the

Atlantic until continued Andes uplift causes a shift in the Amazon drainage basin

allowing for a transcontinental connection in the late Miocene. Harris and Mix [2002]

used shifts in the clay mineralogy sourced from the Amazon River and reaching Ceara

Rise to determine the timing of changes in the Amazon drainage basin. The clay

mineralogy showed a shift from material sourced from a region dominated by chemical

weathering (interpreted as sourced from the Amazon Lowlands) to material sourced

from a region dominated by physical weathering (interpreted as sourced from the

Andean Highlands) at 8 Ma and another event at ~4.5 Ma. The shift at 8 Ma was

attributed to Andean uplift, which resulted in the transcontinental connection of the

Amazon River, while the shift at ~4.5 Ma was associated with either continued uplift or

climate change [Harris and Mix, 2002].

To confirm the change in provenances of material reaching the Ceara Rise

suggested by Harris and Mix [2002], the detrital silicate fraction was separated from

bulk sediment samples from all three sites (925, 926, and 929) and analyzed for Nd and

lead (Pb). The Nd and Pb isotopic data will help to better constrain when a shift from

South American Shield material to one dominated by Andean material occurred, and

whether the second shift in the clay mineralogy of Harris and Mix [2002] at ~4.5 Ma was

80

the result of continued Andean uplift or changes in weathering related to changing

climatic conditions.

Material and Methods

The Ceara Rise is an aseismic ridge located in the western equatorial Atlantic

outside of the direct influence of the Amazon Fan [Curry et al., 1995]. Ocean Drilling

Program Sites 925 (4°12’N, 43°29’W; 3042 m water depth), 926 (3°43’N, 42°55’W; 3598

m water depth), 929 (5°58’N, 43°44’W; 4356 m water depth) used for this study (Figure

4-1) were sampled at ~0.25 Ma for fossil fish teeth/debris to reconstruct seawater Nd

isotopic values from 18 to 2.5 Ma. A total of eight detritral silicate fractions were taken

from each site, with three detrital silicate fractions analyzed representing the time period

of 16.25 to 11 Ma, and five samples from 7.25 to 2.75 Ma. Age models for the 2.5 to 14

Ma sections of these sites were based on the astrochronologically tuned age models,

while the 14 to 18 Ma interval is based on Shipboard biostratigraphy [Curry, Shackleton,

Richter, et al., 1995]. The age model for the youngest portion of site 925 (2.5-5.0 Ma)

is from Tiedemann and Franz [1997], for 5.0-14 Ma at site 925 and 0-14 Ma at site 926

the age models are from Shackleton and Crowhurst [1997]), and Shipboard

biostratigraphic age models have been applied to the 14 to 18 Ma interval at all three

sites. The ages of biostratigraphic datums at site 929 were adjusted to the

astronomically calibrated ages of site 926 for the 0-14 Ma interval.

Fossil fish teeth and debris were handpicked from the >125 μm size fraction and

dissolved in aqua regia, without prior cleaning based on Martin et al. [2010].

Additionally, a comparison was made between Nd isotopes recovered from cleaned

fossil fish teeth and Fe-Mn oxide coatings at site 926 (Figure 4-2). For analysis of

81

detrital silicate fractions, approximately 1 g of bulk sediment was processed through the

sequential extraction procedure of Basak et al. [2011]. The initial step of the process is

decarbonation using 1 M Na acetate in 2.7% optima glacial acetic acid (buffered to

pH=5). Fe-Mn oxide coatings were removed from the bulk sediment using Chelex

cleaned 0.02 N Hydroxylamine Hydrochloride (HH) in 25% acetic acid solution. A

fraction (0.05 g) of the remaining detrital silicate material was heated to >100°C for 48

hours in a 4:1 mixture of concentrated optima grade HF:HNO3 for dissolution. Bulk rare

earth elements (REEs) from the fossil fish teeth/debris, the HH fraction, and the detrital

silicates were separated on primary quartz columns using Mitsubishi cation exchange

resin [Scher and Martin, 2004], or Teflon columns using Eichrom TRUspecTM Resin.

Both used optima grade HCl as the eluent. Nd was isolated from the bulk REEs using

Eichrom LNspecTM resin with HCl as the eluent in volumetrically calibrated Teflon

columns [Pin and Zalduegi, 1997]. The total Nd blank for both techniques is 14 pg.

Nd and Pb isotopic ratios were measured on a Nu Plasma Multi-collector

Inductively Coupled Plasma Mass Spectrometer (MC-ICP-MS) using a DSN1000

nebulizer at the University of Florida. The JNdi-1 standard was run between every 4 to 6

unknown samples. Dilutions of the samples were adjusted to achieve a 2 to 6 V beam

for 142Nd. All of the JNdi-1 values analyzed during a given day were averaged and

compared to the published value of JNdi-1 (0.512115 ± 0.000007) [Tanaka et al., 2000]

to determine the amount of correction to apply to unknown samples analyzed that day.

A drift correction was not applied to the data because variations throughout a day of

analysis did not indicate a consistent drift. The 2σ error for the Nu MC-ICP-MS based

on the variability of normalized JNdi-1 analyzed over the past several years is

82

±0.000014, which is equivalent to 0.27 εNd units (εNd represents the deviation in parts

per 104 of the 143Nd/144Nd ratio of the sample relative to the chondritic uniform reservoir

with 143Nd/144Nd=0.512638 [Jacobsen and Wasserburg, 1980]). For Pb analyses, a Tl

normalization technique was used following Kamenov et al. [2004]. The Pb

concentrates collected from the Pb columns were dissolved with Tl spiked 2% Optima

grade HNO3, and the dilutions were adjusted to obtain a 2-5 V beam on 208Pb. The NBS

981 standard was used and has long term average values over several years of

analyses at UF of 206Pb/204Pb=16.937 (2σ=0.004), 207Pb/204Pb=15.489 (2σ=0.003), and

208Pb/204Pb=36.695 (2σ=0.008).

Results

From 18 to 8 Ma, the seawater Nd isotopic values recorded using fossil fish teeth

for the two shallowest sites (925 and 926) yield very similar records with values ranging

between -14 to -18 and a general decreasing trend from 18 to 14 Ma followed by an

increasing trend from 14 to 8 Ma (Figure 4-3; Table 4-1). In contrast, there are higher

seawater εNd values (-12 to -14) and less variability at the deepest site (929) (Figure 4-

3). At 8 Ma, seawater εNd values increase rapidly to a peak at ~-11.5 at all three sites

(Figure 4-3). Values for all three sites are similar for the remainder of the record. They

decrease gradually from εNd values of ~-11.5 to ~-13 between 8 to 4.5 Ma (Figure 4-3),

exhibit a small, but rapid increase to ~-11 at 4.5 Ma, and stay around that value until 2.5

Ma (Figure 4-3).

The detrital silicate fraction is less radiogenic than the seawater from 16.5 to 11

Ma with values ranging from -17.2 to -18.7 at Sites 925 and 926, and -16.8 to -17.7 at

Site 929, such that there is a greater separation between seawater and silicate values

at site 929 (Figure 4-4; Table 4-2). In the younger section of the record (2.75 to 7.25

83

Ma), silicate values are more radiogenic (-11.2 to -13.4) and they are very similar to

seawater values (Figure 4-4).

The Pb isotopic values of the detrital silicate fractions are more radiogenic than

seawater Pb isotopic values at all three sites for the entire record (Figure 4-5; Table 4-3

and 4-4). The most radiogenic portion of the curve for both archives occurs in the older

interval (11 to 16.5 Ma) with isotopic values of 19.43 to 19.57 (206Pb/204Pb), 15.856 to

15.865 (207Pb/204Pb), and 39.73 to 39.80 (208Pb/204Pb) for the detrital silicates from all

three sites and values of 19.22 to 19.13 (206Pb/204Pb), 15.748to 15.765 (207Pb/204Pb),

39.31 to 39.37 (208Pb/204Pb) for seawater (Figure 4-5). For the younger interval detrital

values range from 19.01 to 19.08 (206Pb/204Pb), 15.701to 15.719 (207Pb/204Pb), 39.14 to

39.28(208Pb/204Pb), while seawater values range between 18.87 to 18.92 (206Pb/204Pb),

15.683 to 15.695 (207Pb/204Pb), 38.94 to 38.99 (208Pb/204Pb) (Figure 4-5).

Discussion

The Seawater Signature

The seawater Nd isotopic composition at all three Ceara Rise sites was less

radiogenic than published data from any other deep water site in the Miocene, (Figure

4-3). At ~8 Ma, the Nd isotopic composition at all three sites shifted to values that are

more similar to typical values reported from Fe-Mn crusts and fish teeth from the deep

Atlantic [Burton et al., 1997; O’Nions et al., 1998; Palmer and Elderfield, 1986; Reynolds

et al., 1999; Frank et al., 2003; Thomas and Via, 2007]. Although the values younger

than ~8 Ma plot within the range of published Atlantic values, the nonradiogenic values

in the older section suggest the seawater isotopic composition in the Ceara Rise region

is being altered either through boundary exchange or reversible scavenging as a result

of its proximity to Amazon outflow.

84

Alteration of the seawater Nd isotopic composition toward values of the sediment

has been observed along ocean margins in the modern oceans, and the process of this

alteration has been termed “boundary exchange” [Lacan and Jeandel, 2005]. Boundary

exchange occurs as a water mass flows over sediment-rich continental margins and

interacts with the sediment composed of weathered continental detritus as well as

volcanic and authigenic material [Lacan and Jeandel, 2001, 2005a, 2005b; Tachikawa

et al., 2004]. These interactions alter the isotopic composition without increasing the

concentration of Nd in seawater [Lacan and Jeandel, 2005]. Lacan and Jeandel [2005]

acknowledge the mechanisms which drive the process of boundary exchange remains

unclear.

Boundary exchange is unlikely to be the cause of the anomalous seawater Nd

isotopic values at Ceara Rise since this process typically occurs at shallow depths on

the continental margin where there is abundant continentally derived sediment. Ceara

Rise receives some material from the Amazon, but the sites receive typical deep sea

sediments with carbonate percentages between 50 to 80% except at the deepest site

(929) [King et al., 1997] which probably reflects dissolution below the lysocline rather

than more input. In addition, there is no known mechanism to bring seawater altered on

the margin to depth because deep water masses are not forming in this region. Also,

the differences in carbonate contents suggest that Site 929 should be the most

susceptible to boundary exchange, yet this is the site with the least alteration toward

silicate values (Figure 4-4).

Reversible scavenging, on the other hand is the process of adsorption of REE

onto particulates near the surface of the ocean along with desorption within the water

85

column or at the sediment-water interface as a result of concentration contrasts

between the water and absorbed REE [Sidall et al., 2008] or dissolution of Fe-Mn

oxyhydroxides and/or particulates carrying the adsorbed REE to depth (Opal, CaCO3,

dust). Reversible scavenging has mainly been used to explain why the modern ocean

Nd concentrations behave like nutrients with low concentrations in surface waters in the

Atlantic, and higher concentrations in deep waters in the Pacific despite the fact that the

Nd isotopic composition of the seawater does not change systematically with the

increase in concentration. This discrepancy has been termed the “Nd Paradox”

[Goldstein and Hemming, 2003; Lacan and Jeandel, 2001; Jeandel et al., 1995, 1998;

Tachikawa et al., 1999a, 1999b; Bertram and Elderfield, 1993]. Siddall et al. [2008] used

reversible scavenging to try to explain the radiogenic Nd isotopic composition of

seawater in the North Pacific at depth, which he attributed to dissolution of highly

reactive volcanic arc material at the surface of the ocean. The released REEs were then

adsorbed onto particulates (suggested as dust or opal) that sank carrying the REE to

depths where the particulates and/or material carrying Nd down to depth have a

tendency to dissolve [Siddall et al., 2008]. Arsouze et al. [2009] also found it necessary

to include vertical cycling (reversible scavenging) in models to reconstruct modern Nd

concentration and isotopic distribution in seawater. Reversible scavenging appears to

be a better explanation for Ceara Rise since there appears to be a depth discrepancy

with the alteration being the strongest at the two shallow sites (925 and 926).

Although the potential for reversible scavenging would appear to be high near a

large river system, such as the Amazon, Albarède et al. [1998] and O’Nions et al. [1998]

documented that Nd isotopes behaved conservatively in the Indian Ocean, which

86

receives a large amount of terrigenous material from Ganges and Brahmaputra Rivers.

Specifically, there is no Himalayan Nd signature in the seawater records collected from

Indian Ocean Fe-Mn nodules and crusts, even though Frank and O’Nions [1998]

showed that Pb isotopes from the same Fe-Mn crusts recorded Himalayan input. The

discrepancy between the behavior of Pb and Nd isotopes in the Indian Ocean may be

attributed to trapping of Nd in the Bengal Fan [O’Nions et al., 1998] or estuaries [Frank,

2002] or was the result the long residence time of Nd versus Pb. On the other hand,

Bayon et al. [2004] showed that preformed oxides and oxyhydroxides adsorbed onto

sediment reaching the Atlantic via the Congo River altered the seawater Nd isotopic

values through reversible scavenging, and in turn showed that Nd behaved non-

conservatively near a large input of terrigenous material.

All three sites along the Ceara Rise received the same terrigenous material based

on the Nd and Pb isotopic composition of the detrital silicate fraction (Figure 4-4 and 4-

5), but the amount of terrigenous material reaching each of the sites is controlled by

their proximity to the Amazon Fan and water depth [King et al., 1997]. King et al. [1997]

showed that site 925 receives the most terrigenous material followed by site 929, while

site 926 receives the least terrigenous material of the three sites. Yet, despite the fact

that sites 925 and 926 receive different amounts of terrigenous material, these two

shallow sites appear to be the most strongly affected by weathering inputs from the

Amazon River. This depth discrepancy is consistent with reversible scavenging, where

most of the desorption of Nd into the seawater as a result of dissolution or concentration

differences between the seawater and adsorbed Nd appears to have occurred closer to

the depth of the two shallower sites. Site 929 was well below the lysocline and the

87

sediments accumulation prior to 5.7 Ma at this site are described as red clay [King et al.,

1997]). Therefore, the seawater Nd isotopic composition at Site 929 would be less

impacted by reversible scavenging that occurred at a shallower depth. In summary, the

proximal location of Ceara Rise to the Amazon has resulted in alteration of the seawater

Nd isotopic composition in this region, presumably through reversible scavenging.

Unlike most deep sea sites, the sites at Ceara Rise are located close enough to a major

source of continental detrital input to be impacted by sediment from that source. In

addition, the detrital material includes highly reactive Fe-oxide coatings that are

common byproducts of tropical weathering. The net result appears to be that seawater

Nd isotopes in this region have been altered by interaction with this sediment. Thus,

seawater Nd isotopes in this specific region do not reflect the composition of the water

mass advected into the region and cannot be used to track the boundary between

AABW and NCW.

Interpretation of Detrital Isotopes

The modern Amazon River drains the Andes and the lowlands of Brazil, which is

composed of the Brazil and Guyana cratons, and ultimately flowing into the Atlantic

forming the Amazon Fan. The sediment reaching the Amazon Fan today is dominantly

sourced from the Andean highlands [Gibbs, 1967; Milliman, 1979; Damuth et al., 1988;

McDaniel et al., 1997; Dobson et al., 1997; Dobson et al., 2001]. Estimates for the

timing of initiation of a dominant Andean source, and development of the Amazon River

and Fan range from the middle/late Miocene to the end of the Pliocene [Castro et al.,

1978; Campbell, 1992; Hoorn, 1993; Hoorn, 1994; Hoorn et al., 1995; Campbell et al.,

2006; Figueiredo et al., 2009, 2010; Campbell, 2010; Hoorn et al., 2010]. Studies

indicate that the continental shelf underlying the Amazon Fan was a carbonate platform

88

that only received moderate amounts of Amazon River sediment until the late Miocene.

Figueiredo et al. [2009] showed that during the early to middle Miocene the sediment

reaching the Amazon shelf was derived from the lowland cratons, but from the late

Miocene to present the sediment was predominantly derived from the Andes, which

they attributed to the onset of a transcontinental connection of the Amazon. The timing

of this switch and initiation of the fan was initially placed between 11.8 to 11.3 Ma

[Figueiredo et al., 2009], and later adjusted to ~10.5 Ma [Figueiredo et al., 2010] with a

fully developed Amazon River by ~6.8 Ma [Figueiredo et al., 2009]. Prior to the

transcontinental connection during the late middle Miocene, Hoorn et al. [1995] argued

early Andes uplift altered the northwest Amazonia region generating a proto-Amazon

River which flowed north instead of flowing east to the Atlantic until continued Andes

uplift in the late Miocene created the modern Amazon drainage basin complete with a

transcontinental connection.

Harris and Mix [2002] used the ratio of chlorite/kaolinite to determine whether the

Ceara Rise was receiving physically weathered material from the Andean Highlands

(chlorite) or the chemically weathered material from the Amazon Lowlands (kaolinite).

Prior to 8 Ma, the chlorite/kaolinite ratio was low, suggesting a lack of material sourced

from the Andean region. After 8 Ma, the chlorite/kaolinite ratio increased along with a

notable increase in terrigenous mass accumulation rates as a result of an increase in

the supply of Andean sourced sediments. Another noted shift in the chlorite/kaolinite

record occurred at ~4.5 Ma, which Harris and Mix [2002] interpreted as additional input

of Andean derived sediment in response to another Andean uplift event, but they could

not rule out a change in paleoclimate as the cause for this observed shift. The timing of

89

the second chlorite/kaolinite shift was also accompanied by an increase in terrigenous

mass accumulation rates, and Harris and Mix [2002] argued the two events could be

correlated with two Andean tectonic events: Quecha 2 and 3, which are well-known

from the Peruvian Andes [Megard et al., 1984]) and have been credited with the

development of the transcontinental Amazon connection [Harris and Mix, 2002]. The

findings of Harris and Mix [2002] were similar to the findings of Dobson et al. [2001] who

looked at the terrigenous mass accumulation rates of the material reaching Ceara Rise

and the geochemical composition of the terrigenous material. Dobson et al. [2001]

noted that the composition of Ceara Rise changed from shield material to sediment

predominately sourced from the Andes at ~10 Ma, which coincided with an increase in

the terrigenous mass accumulation rate at Ceara Rise.

The Pb isotopic compositions of the detrital silicates provide additional evidence

to support this shield to Andes source transition. Pb isotope cross plots illustrate that the

11 to 16.5 Ma silicates from all three Ceara Rise sites plot near the field of Older

Cratonic Rock, while on values for the younger silicates (2.75 Ma to 7.25 Ma) are clearly

distinct and plot within the Amazon Delta and Amazon Fan mud fields of McDaniel et al.

[1997] (Figure 4-6) as defined by Pleistocene to modern sediment. The shift observed in

the Pb isotopic composition of the detrital silicate fraction further demonstrates that the

sediment reaching the Ceara Rise in the older section was sourced from the Amazon

lowlands (South American Shield material), while the sediment reaching the Ceara Rise

during the younger section of the record was sourced from the Andean highlands

(volcanic arc material).

90

The Nd isotopic data also support a shift from detrital silicates sourced from the

South American Shield material, which has εNd value of ~-20 [Allegre et al., 1996],

during the older section (16.5 to 11 Ma) to more radiogenic values in the younger

section (7.25 to 2.75 Ma) with values slightly less radiogenic than Amazon Fan

Pleistocene silicate values until ~4.5 Ma when they become slightly more radiogenic

and overlap with the Nd isotopic values of the Amazon Fan silicate values of McDaniel

et al. [1997]. Although the detrital silicate data set is relatively low resolution, the

changes appear to track the variations in values recorded in the record from fish teeth,

which provides better constraint on the exact timing. These data argue for a large shift

at ~8 Ma, and a smaller shift at ~4.5 Ma, which agrees well with observed shifts in the

clay mineralogy at Ceara Rise [Harris and Mix, 2002], and supports their conclusion that

the shift is generated by tectonic events in the Andes and changes in the Amazon

drainage basin rather than climate-controlled weathering.

Summary

Miocene ‘seawater’ Nd and Pb isotopes preserved in fossil fish teeth/debris at

Ceara Rise record similar patterns of change as the detrital silicate fractions. Therefore,

the observed shifts appear to be driven by local weathering inputs from the Amazon

River that are transmitted into the seawater through reversible scavenging and reflect

major changes in the provenance of the source material. The impact of reversible

scavenging has been documented in the Pacific, but appears to be less common in the

Atlantic; however, the large amount of sediment transported down the Amazon River

that reaches the Ceara Rise is unusual for a deep sea location. Overprinting of

seawater Nd isotopic values by weathering inputs prevented reconstruction of

circulation to identify shifts in the boundary between NCW and AABW at Ceara Rise.

91

The Nd and Pb isotopic composition of the detrital silicate fraction recorded a

dramatic shift at ~8 Ma. Prior to 8 Ma the nonradiogenic Nd isotopic values and

radiogenic Pb isotopic values are similar to South American cratonic shield material.

These results suggest the detrital silicate material reaching Ceara Rise was sourced

from the Amazon lowlands, which appear to dominate the Amazon drainage basin at

this time. After 8 Ma, the Nd isotopic values shift to more radiogenic values and Pb

isotopic values shift to less radiogenic values, both of which are more representative of

volcanic arc or Andean material. The timing of the observed shifts in the Pb and Nd

isotopes agrees with shifts in the mineralogy and geochemical data of the silicate

material reaching Ceara Rise and has been linked to the development of a

transcontinental connection of the Amazon River and an overall change in the

Amazonian drainage basin following Andean tectonic events. After this connection the

Amazon lowland (South American shield) material was overwhelmed by younger

Andean highland material (volcanic arc) consistent with modern and Pleistocene detrital

sources in the Amazon River and on the Amazon Fan.

92

Figure 4-1. Bathymetric map of the Atlantic Ocean [Schlitzer, R., 2010], and the bathymetric profile illustrating the position of Ocean Drilling Program sites 925, 926, 929 [modified from Curry et al., 1995].

93

Figure 4-2. Salinity profile along a north-south transect in the Atlantic overlain by seawater εNd profiles illustrating that Nd isotopes are also a conservative tracker of water mass in the Atlantic. The gray rectangle outlines the location and depths of sites on Ceara Rise [modified from von Blackenburg, 1999].

94

Age (Ma)

2 4 6 8 10 12 14 16 18

Nd

(0)

-18

-16

-14

-12

-10

-8

-6

Atlantic Deep Waters

2

NADW

AABW

Site 929 (4,360 m water depth)

Site 925 (3,040 m water depth)

Site 926 (3,600 m water depth)

Figure 4-3. Plot of εNd from fossil fish teeth vs. age for sites 925, 926, and 929 on Ceara Rise in the western equatorial Atlantic. The gray shadow box represents the range of published εNd values for the Atlantic [Burton et al., 1997, 1999; Ling et al., 1997; O’Nions et al., 1998; Frank et al., 1999; Reynolds et al., 1999; Thomas and Via, 2007]. The Nd isotopic values for the Ceara Rise sites plot well below these values from 18 to 8 Ma, and in the lower portion of the field after 8 Ma.

95

Age (Ma)

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

d(0

)

-18

-16

-14

-12

-10

Site 925 Detrital Silicate

Site 926 Detrital Silicate

Site 929 Detrital Silicate

Site 929 Seawater (4,360 m)

Site 926 Seawater (3,600 m)

Site 925 Seawater (3,040 m)

Figure 4-4. εNd vs. age for detrital silicate fractions (stars) and fossil fish teeth (lines) for sites 925, 926, and 929 on Ceara Rise. The detrital silicate fractions are less radiogenic than seawater values from 16.5 to 11 Ma, with the largest offset observed at the deepest site (929). From 7.25 to 2.75 Ma, the detrital silicate fraction and seawater values record similar values for all three sites, with no consistent offset observed.

96

Figure 4-5. 206Pb/204Pb vs. age for both detrital silicate fractions (circles) and Fe-Mn oxide coatings, which are interpreted to represent seawater values (squares) for sites 925, 926, and 929 on Ceara Rise. Seawater values remain less radiogenic than detrital silicate fractions for the entire record, and both archives show a shift from radiogenic to less radiogenic values between the older section (16.5 to 11 Ma) and the younger section (7.25 to 2.75 Ma).

Age (Ma)

2 4 6 8 10 12 14 16 18

206P

b/2

04P

b

18.8

19.0

19.2

19.4

19.6

19.8

925 Detrital Silicate

925 Seawater

926 Detrital Silicate

926 Seawater

929 Detrital Silicate

929 Seawater

97

Figure 4-6. Pb isotopic crossplots of a) 207Pb/204Pb vs. 206Pb/204Pb, and b) 208Pb/204Pb vs. 206Pb/204Pb for the detrital silicate fractions from all three Ceara Rise sites (925, 926, and 929) illustrating that older detrital silicate fractions (16.5 to 11 Ma; open symbols) plot close to the Old Cratonic Rock Field, while younger detrital silicate fractions (7.25 to 2.75 Ma; filled symbols) plot within the Pleistocene Amazon Fan sediment and just outside the Andean Igneous Rocks field. Shaded zones and reference arrows (which indicate the direction of published data for older cratonic rocks from South America) from McDaniel et al. [1997; and references therein].

98

Table 4-1. Nd isotopic results for Ceara Rise fossil fish teeth

Sample Depth (mcd) Age (Ma) 143Nd/144Nda εNd(0)b

154-925C-9H-4, W, 6-8 84.14 2.50 0.512085 -10.78

154-925D-10H-2, W, 118-120 100.14 3.00 0.512064 -11.20

154-925C-11H-3, W, 87-89 108.21 3.25 0.512053 -11.41

154-925D-11H-6, W, 99-101 116.23 3.50 0.512099 -10.52

154-925D-12H-5, W, 32-34 124.26 3.75 0.512036 -11.75

154-925C-13H-2, W, 142-144 128.98 4.00 0.512076 -10.97

154-925D-13H-5, W, 78-80 135.73 4.25 0.512073 -11.02

154-925D-14H-3, W, 102-104 142.42 4.50 0.512022 -12.02

154-925C-15H-2, W, 115-117 149.16 4.75 0.511968 -13.06

154-925D-15H-4, W, 107-109 155.80 5.00 0.511957 -13.28

154-925D-16H-3, W, 3-5 162.54 5.25 0.511977 -12.90

154-925C-17H-2, W, 81-83 169.21 5.50 0.511982 -12.79

154-925C-18H-4, W, 87-89 182.56 6.00 0.512009 -12.26

154-925B-19H-4, W, 132-134 189.12 6.25 0.511999 -12.47

154-925C-19H-5, W, 119-121 195.62 6.50 0.512015 -12.15

154-925C-20H-3, W, 62-64 202.22 6.75 0.511997 -12.50

154-925D-20H-3, W, 109-111 208.78 7.00 0.511981 -12.82

154-925C-21H-5, W, 13-15 215.34 7.25 0.512038 -11.70

154-925D-21H-5, W, 100-102 221.90 7.50 0.512042 -11.63

154-925D-22H-3, W, 25-27 228.27 7.74 0.512028 -11.90

154-925B-23H-5, W, 52-54 234.99 8.00 0.511999 -12.46

154-925C-23H-5, W, 142-144 241.57 8.25 0.511884 -14.71

154-925D-24H-3, W, 111-113 249.12 8.54 0.511905 -14.29

154-925C-25H-4, W, 19-21 254.70 8.75 0.511891 -14.56

154-925D-25H-4, W, 127-129 261.26 9.00 0.511953 -13.35

154-925D-26H-2, W, 30-32 267.81 9.25 0.511865 -15.08

154-925D-26H-6, W, 28-30 273.79 9.50 0.511825 -15.86

154-925D-27H-4, W, 1-3 278.93 9.75 0.511861 -15.16

154-925B-28H-3, W, 119-121 284.02 10.00 0.511915 -14.11

154-925B-29H-2, W, 32-34 294.22 10.50 0.511934 -13.73

154-925D-29H-4, W, 80-82 304.42 11.00 0.511821 -15.94

154-925B-30H-4, W, 98-100 309.56 11.25 0.511849 -15.38

154-925D-30H-5, W, 0-2 314.89 11.50 0.511840 -15.57

154-925C-31H-2, W, 75-77 324.92 12.25 0.511836 -15.64

154-925B-32H-2, W, 33-35 328.25 12.50 0.511795 -16.44

154-925B-32H-4, W, 63-65 331.51 12.74 0.511765 -17.02

154-925D-32H-3, W, 109-111 334.95 13.00 0.511767 -16.99

154-925D-32H-5, W, 147-149 338.29 13.25 0.511791 -16.51

154-925B-33H-4, W, 17-19 341.63 13.50 0.511776 -16.81

154-925D-33H-3, W, 46-48 344.98 13.75 0.511908 -14.23

154-925D-33H-5, W, 80-82 348.31 14.00 0.511725 -17.81

154-925B-34H-1, W, 97-99 351.57 14.24 0.511823 -15.89

99

Table 4-1. Continued Sample Depth (mcd) Age (Ma) 143Nd/144Nda εNd(0)

b

154-925B-34H-3, W, 142-144 355.03 14.50 0.511883 -14.72

154-925A-5R-5, W, 61-63 368.41 15.50 0.511869 -15.00

154-925C-36X-3, W, 28-30 371.72 15.75 0.511845 -15.47

154-925D-36H-1, W, 77-79 376.71 16.00 0.511873 -14.92

154-925A-7R-3, W, 15-17 384.19 16.25 0.511878 -14.83

154-925A-8R-1, W, 98-100 391.69 16.50 0.511921 -13.99

154-925C-38X-2, W, 21-23 399.21 16.75 0.511860 -15.17

154-925A-9R-5, W, 32-34 406.72 17.00 0.511768 -16.96

154-925A-10R-2, W, 100-102 412.58 17.20 0.511931 -13.78

154-925A-11R-2, W, 63-65 421.77 17.50 0.511780 -16.74

154-925A-12R-1, W, 13-15 429.25 17.75 0.511791 -16.52

154-926C-8H-3, W, 131-133 77.83 2.48 0.512003 -12.39

154-926A-10H-3, W, 18-20 93.69 3.00 0.512079 -10.91

154-926C-10H-4, W, 44-46 99.75 3.20 0.512068 -11.12

154-926C-11H-5, W, 59-61 112.56 3.62 0.512094 -10.61

154-926C-13H-3, W, 49-51 132.95 4.28 0.512079 -10.91

154-926C-15H-1, W, 41-43 140.43 4.50 0.511992 -12.61

154-926B-15H-5, W, 54-56 153.88 4.99 0.511953 -13.36

154-926C-15H-4, W, 67-69 156.11 5.10 0.511969 -13.04

154-926A-16H-3, W, 44-46 158.39 5.20 0.511989 -12.65

154-926A-16H-6, W, 34-36 162.79 5.40 0.511961 -13.20

154-926C-16H-3, W, 22-24 164.87 5.50 0.511967 -13.08

154-926C-16H-4, W, 92-94 167.07 5.60 0.512009 -12.26

154-926C-16H-6, W, 2-4 169.17 5.69 0.511995 -12.54

154-926B-17H-2, W, 104-106 171.50 5.80 0.512005 -12.34

154-926B-17H-4, W, 44-46 173.90 5.87 0.512006 -12.32

154-926C-17H-3, W, 67-69 175.89 5.99 0.511999 -12.46

154-926C-17H-5, W, 147-149 179.69 6.21 0.512013 -12.19

154-926B-18H-3, W, 114-116 183.19 6.41 0.512004 -12.36

154-926B-19H-4, W, 94-96 195.28 7.11 0.512061 -11.25

154-926B-19H-6, W, 14-16 197.48 7.23 0.512046 -11.54

154-926A-20H-3, W, 5-7 200.03 7.38 0.512077 -10.94

154-926B-20H-3, W, 104-106 204.73 7.69 0.512044 -11.58

154-926A-21H-3, W, 75.5-80 210.80 7.98 0.511859 -15.20

154-926B-21H-4, W, 121-126 215.36 8.20 0.511936 -13.69

154-926C-22H-2, W, 114-119 219.71 8.41 0.511849 -15.39

154-926C-22H-3, W, 117-119 221.22 8.51 0.511888 -14.62

154-926C-22H-4, W, 94-99 222.51 8.59 0.511878 -14.83

154-926C-22H-4, W, 97-99 222.52 8.59 0.511846 -15.44

154-926B-22H-4, W, 14-16 224.00 8.69 0.511957 -13.28

154-926B-22H-5, W, 10.5-16 225.48 8.80 0.511940 -13.62

154-926A-23H-1, W, 82.5-88 228.40 8.99 0.511924 -13.93

100

Table 4-1. Continued Sample Depth (mcd) Age (Ma) 143Nd/144Nda εNd(0)

b

154-926A-23H-1, W, 88-90 228.44 9.00 0.511889 -14.60

154-926A-23H-3, W, 73.5-79.5 231.32 9.19 0.511875 -14.88

154-926B-23H-3, W, 10.5-16 234.19 9.39 0.511868 -15.02

154-926B-23H-5, W, 0-6 237.09 9.58 0.511890 -14.59

154-926A-24H-2, W, 141-146 240.53 9.82 0.511914 -14.12

154-926A-24H-4, W, 131-136 243.43 10.00 0.511917 -14.06

154-926B-24H-4, W, 11-16 246.15 10.22 0.511856 -15.25

154-926B-24H-5, W, 69.5-76 248.23 10.38 0.511844 -15.49

154-926A-25H-3, W, 0-6 251.37 10.63 0.511903 -14.34

154-926A-25H-4, W, 81-86 253.68 10.81 0.511884 -14.71

154-926B-25H-4, W, 124-129 257.27 11.10 0.511847 -15.43

154-926B-25H-5, W, 129-135 258.82 11.22 0.511866 -15.06

154-926A-26H-1, W, 110-116 261.48 11.43 0.511869 -15.00

154-926A-26H-3, W, 51-56 263.89 11.62 0.511857 -15.23

154-926B-26H-4, W, 120-125.5 268.99 12.00 0.511870 -14.98

154-926B-27H-3, W, 31-36 277.39 12.58 0.511795 -16.44

154-926B-27H-5, W, 61-65 280.68 12.80 0.511787 -16.60

154-926A-28H-3, W, 131.5-136 284.52 13.02 0.511789 -16.56

154-926A-28H-5, W, 90-96 287.11 13.16 0.511803 -16.29

154-926B-29H-2, W, 89-93 290.62 13.35 0.511812 -16.11

154-926B-29H-5, W, 38-43 294.61 13.57 0.511789 -16.57

154-926B-29H-7, W, 29-34 297.53 13.73 0.511792 -16.50

154-926B-30X-4, W, 100-102 303.34 14.00 0.511945 -13.51

154-926B-30X-6, W, 74-76 306.08 14.25 0.511808 -16.18

154-926B-31X-1, W, 127-129 308.82 14.50 0.511803 -16.29

154-926B-31X-3, W, 92-94 311.47 14.74 0.511846 -15.45

154-926B-31X-5, W, 71-73 314.27 15.00 0.511834 -15.68

154-926B-31X-7, W, 42-44 316.98 15.25 0.511888 -14.63

154-926B-32X-2, W, 97-99 319.73 15.50 0.511830 -15.75

154-926B-32X-6, W, 42-44 325.19 16.00 0.511952 -13.38

154-926B-33X-1, W, 105-107 327.93 16.25 0.511815 -16.06

154-926B-33X-5, W, 51-53 333.40 16.75 0.511963 -13.16

154-926B-34X-4, W, 5-7 341.15 17.46 0.511953 -13.35

154-926B-34X-CC, W, 25-27 345.88 17.89 0.511846 -15.45

154-926B-35X-1, W, 85-87 347.06 18.00 0.511850 -15.37

154-929A-9H-1, W, 104-106 80.82 2.50 0.512086 -10.76

154-929C-9H-3, W, 38-40 87.93 2.75 0.512065 -11.17

154-929A-10H-4, W, 41-43 94.89 3.00 0.512088 -10.73

154-929A-11H-2, W, 41-43 101.99 3.25 0.512074 -10.99

154-929C-11H-4, W, 33-35 109.05 3.50 0.512085 -10.78

154-929A-12H-4, W, 125-127 116.12 3.75 0.512076 -10.96

154-929C-12H-5, W, 133-135 121.24 4.00 0.512107 -10.36

101

Table 4-1. Continued Sample Depth (mcd) Age (Ma) 143Nd/144Nda εNd(0)

b

154-929A-13H-4, W, 97-99 126.32 4.25 0.512105 -10.39

154-929C-13H-4, W, 133-135 131.32 4.50 0.512011 -12.22

154-929A-15H-6, W, 15-17 144.37 5.25 0.511980 -12.84

154-929A-16X-1, W, 62-64 146.84 5.49 0.512007 -12.31

154-929C-15X-4, W, 15-17 149.47 5.75 0.511987 -12.70

154-929C-16X-5, W, 102-104 160.99 6.25 0.512024 -11.97

154-929B-18X-2, W, 117-119 174.78 6.75 0.512010 -12.24

154-929B-19X-5, W, 6-8 188.59 7.25 0.512082 -10.84

154-929A-21X-2, W, 61-63 202.37 7.75 0.511960 -13.22

154-929B-21X-6, W, 54-56 209.27 8.00 0.511961 -13.20

154-929A-22X-5, W, 50-52 216.15 8.25 0.511985 -12.74

154-929A-23X-1, W, 112-114 220.63 9.50 0.511970 -13.03

154-929A-23X-3, W, 66-68 223.17 10.24 0.511964 -13.14

154-929A-23X-4, W, 89-91 224.90 10.75 0.511971 -13.00

154-929A-23X-5, W, 26-28 225.77 11.01 0.511951 -13.40

154-929A-23X-5, W, 109-111 226.60 11.25 0.511996 -12.52

154-929A-23X-6, W, 46-48 227.47 11.51 0.511987 -12.70

154-929A-23X-6, W, 129-131 228.30 11.75 0.511992 -12.59

154-929A-24X-1, W, 32-34 229.13 11.99 0.511937 -13.67

154-929A-24X-1, W, 117-119 229.98 12.24 0.511955 -13.32

154-929A-24X-2, W, 57-59 230.88 12.51 0.511949 -13.44

154-929A-24X-2, W, 139-141 231.70 12.75 0.511994 -12.55

154-929A-24X-4, W, 9-11 233.40 13.25 0.511945 -13.51

154-929A-24X-4, W, 92-94 234.23 13.49 0.511976 -12.91

154-929A-24X-5, W, 28-30 235.09 13.75 0.511957 -13.28

154-929A-24X-5, W, 109-111 235.90 13.98 0.511928 -13.84

154-929A-24X-6, W, 49-51 236.80 14.25 0.511938 -13.65

154-929B-24X-5, W, 110-112 237.63 14.49 0.512011 -12.22

154-929B-24X-6, W, 130-132 239.34 15.00 0.512012 -12.21

154-929A-25X-1, W, 77-79 240.20 15.25 0.511947 -13.47

154-929A-25X-2, W, 12-14 241.05 15.50 0.511976 -12.91

154-929A-25X-2, W, 100-102 241.93 15.76 0.511969 -13.04

154-929A-25X-4, W, 0-2 243.93 16.00 0.511957 -13.28

154-929A-26X-2, W, 32-34 250.85 16.75 0.512013 -12.18

154-929A-26X-3, W, 37-39 252.40 16.91 0.511909 -14.22

154-929A-26X-5, W, 53-55 255.56 17.25 0.511969 -13.04

154-929A-26X-6, W, 80-82 257.33 17.44 0.511987 -12.70

154-929A-27X-1, W, 133-135 260.16 17.75 0.512083 -10.82

154-929A-27X-3, W, 67-69 262.50 18.00 0.511942 -13.57 a 143Nd/144Nd values analyzed on a given day were corrected by the difference between the average JNdi-1 value for that day and Tanaka et al. [2000]. b εNd(0) = [143Nd/144Nd(sample)/

143Nd/144Nd(CHUR) – 1] × 104, where 143Nd/144Nd(CHUR) = 0.512638. The 2σ is ±0.000015, which is equivalent to 0.3 εNd units.

102

Table 4-2. Nd isotopic values for detrital silicates

Sample Depth (mcd) Age (Ma) 143Nd/144Nd εNd

154-925B-10H-4W 130-131 92.10 2.75 0.512025 -12.0 154-925D-13H-5W 76-77 136.21 4.27 0.512031 -11.8 154-925C-15H-2W 117-118 149.18 4.75 0.511961 -13.2 154-925C-17H-2W 83-84 169.23 5.50 0.511999 -12.5 154-925C-21H-5W 16-17 215.37 7.25 0.511951 -13.4 154-925D-29H-4W 79-80 304.39 11.00 0.511728 -17.8 154-925D-32H-5W 149-150 338.30 13.25 0.511698 -18.3 154-925A-7R-3W 14-15 384.17 16.25 0.511749 -17.3

154-926C-29H-2W 32-33 86.28 2.76 0.512029 -11.9 154-926C-13H-3W 47-49 132.93 4.28 0.512052 -11.4 154-926C-14H-3W 13-14 143.31 4.62 0.511980 -12.8 154-926C-16H-3W 21-23 164.86 5.50 0.511975 -12.9 154-926B-25H-4W 122-123 257.24 11.09 0.511755 -17.2 154-926B-29X-2W 88-89 290.59 13.35 0.511711 -18.1 154-926B-33X-1W 107-108 327.94 16.25 0.511679 -18.7

154-929C-9H-3W 36-38 87.91 2.75 -11.946052 -11.9 154-929A-13H-4W 95-97 126.30 4.25 -11.258293 -11.3 154-929C-13H-4W 131-133 131.30 4.50 -12.292160 -12.3 154-929A-16X-1W 61-62 146.83 5.49 -12.019063 -12.0 154-929A-19X-2W 85-86 181.66 7.00 -12.916383 -12.9 154-929A-23X-5W 24-26 225.75 11.00 -17.660285 -17.7 154-929A-24X-4W 8-9 233.39 13.25 -17.422486 -17.4 154-929A-25X-7W 10-11 248.53 16.50 -16.759250 -16.8

103

Table 4-3. Pb isotopic values for detrital silicates

Sample Depth (mcd)

Age (Ma) 208Pb/204Pb 207Pb/204Pb 206Pb/204Pb

154-925B-10H-4W 130-131 92.10 2.75 39.172 15.704 19.083 154-925D-13H-5W 76-77 136.21 4.27 39.144 15.701 19.026 154-925C-15H-2W 117-118 149.18 4.75 39.241 15.715 19.017 154-925C-17H-2W 83-84 169.23 5.50 39.208 15.711 19.011 154-925C-21H-5W 16-17 215.37 7.25 39.282 15.720 19.052 154-925D-29H-4W 79-80 304.39 11.00 39.726 15.858 19.428 154-925D-32H-5W 149-150 338.30 13.25 39.790 15.865 19.573 154-925A-7R-3W 14-15 384.17 16.25 39.802 15.856 19.557

154-926C-29H-2W 32-33 86.28 2.76 39.199 15.715 19.114 154-926C-13H-3W 47-49 132.93 4.28 39.180 15.704 19.040 154-926C-14H-3W 13-14 143.31 4.62 39.159 15.697 18.905 154-926C-16H-3W 21-23 164.86 5.50 39.262 15.722 19.032 154-926B-19H-4W 94-96 195.28 7.11 39.200 15.712 19.012 154-926B-25H-4W 122-123 257.24 11.09 39.591 15.824 19.358 154-926B-29X-2W 88-89 290.59 13.35 39.794 15.864 19.585 154-926B-33X-1W 107-108 327.94 16.25 39.877 15.884 19.660

154-929C-9H-3W 36-38 87.91 2.75 39.247 15.717 19.144 154-929A-13H-4W 95-97 126.30 4.25 39.230 15.712 19.096 154-929C-13H-4W 131-133 131.30 4.50 39.248 15.709 19.010 154-929A-16X-1W 61-62 146.83 5.49 39.252 15.719 19.010 154-929A-19X-2W 85-86 181.66 7.00 39.259 15.722 19.067 154-929A-23X-5W 24-26 225.75 11.00 39.599 15.825 19.419 154-929A-24X-4W 8-9 233.39 13.25 39.722 15.844 19.546 154-929A-25X-7W 10-11 248.53 16.50 39.889 15.872 19.690

104

Table 4-4. Pb isotopic values for leachates

Sample Depth (mcd)

Age (Ma) 208Pb/204Pb 207Pb/204Pb 206Pb/204Pb

154-925B-10H-4W 130-131 92.10 2.75 38.955 15.683 18.911

154-925D-13H-5W 76-77 136.21 4.27 38.957 15.685 18.901

154-925C-15H-2W 117-118 149.18 4.75 38.990 15.695 18.925

154-925C-17H-2W 83-84 169.23 5.50 38.941 15.686 18.866

154-925C-21H-5W 16-17 215.37 7.25 39.003 15.692 18.901

154-925D-29H-4W 79-80 304.39 11.00 39.313 15.748 19.144

154-925D-32H-5W 149-150 338.30 13.25 39.374 15.765 19.223

154-925A-7R-3W 14-15 384.17 16.25 39.323 15.758 19.131

154-926C-29H-2W 32-33 86.28 2.76 38.995 15.685 18.935

154-926C-13H-3W 47-49 132.93 4.28 38.953 15.677 18.871

154-926C-14H-3W 13-14 143.31 4.62 38.983 15.680 18.844

154-926C-16H-3W 21-23 164.86 5.50 38.986 15.671 18.894

154-926B-19H-4W 94-96 195.28 7.11 38.970 15.688 18.884

154-926B-25H-4W 122-123 257.24 11.09 39.270 15.737 19.108

154-926B-29X-2W 88-89 290.59 13.35 39.284 15.745 19.186

154-926B-33X-1W 107-108 327.94 16.25 39.325 15.758 19.183

154-929C-9H-3W 36-38 87.91 2.75 39.024 15.685 18.946

154-929A-13H-4W 95-97 126.30 4.25 38.900 15.660 18.913

154-929C-13H-4W 131-133 131.30 4.50 38.997 15.663 18.868

154-929A-16X-1W 61-62 146.83 5.49 39.127 15.695 18.896

154-929A-19X-2W 85-86 181.66 7.00 38.965 15.681 18.908

154-929A-23X-5W 24-26 225.75 11.00 39.196 15.723 19.083

154-929A-24X-4W 8-9 233.39 13.25 39.202 15.725 19.087

154-929A-25X-7W 10-11 248.53 16.50 39.093 15.708 19.015

105

CHAPTER 5 CONCLUSIONS

The results of this study showed that the neodymium (Nd) isotopic values were

representative of Pacific Deep Water (PDW) and the onset of the carbonate crash in the

eastern equatorial Pacific sites and the Caribbean is consistent with a reorganization of

the Pacific which resulted in enhanced flow of PDW in the equatorial region. Although

many studies suggest closure of the Central American Seaway (CAS) played a major

role in driving these changes, the timing of events supports climate change in the

Southern Ocean and increased Deep Western Boundary Current flow as the primary

factors leading to changes in Miocene Pacific circulation. Although the onset of both the

Pacific and Caribbean carbonate crash intervals is attributed to encroachment of PDW,

termination of these events has a different cause at each location. Shoaling of the CAS

ultimately created a barrier to flow of corrosive Pacific deep and intermediate waters

into the Caribbean Basin and allowed for the recovery from the Caribbean carbonate

crash. Although equatorial Pacific continued to be exposed to corrosive PDW, the

recovery from the Pacific carbonate crash coincides with increased surface water

productivity and carbonate deposition.

The Caribbean Basin was filled with PDW while the CAS was open for exchange

from the middle to late Miocene. The peak flow of Pacific-sourced waters through the

CAS occurred over the same time as suggested times of North Atlantic Deep Water

(NADW) production, and is in disagreement with most Ocean General Circulation

Models. Unfortunately, an attempt to study NADW production using seawater Nd

isotopes on the Ceara Rise in the tropical western Atlantic was unsuccessful because

the Miocene Nd isotopes from fish teeth and Pb isotopes from oxide coatings recorded

106

seawater values which were strongly altered by detrital outputs from the Amazon River.

The sites on the Ceara Rise record shifts in the composition of the detrital silicate

fraction rather than shifts in the position in the boundary between NADW and Antarctic

Bottom Water. Observed shifts in the seawater signals indicate the sediment on Ceara

Rise records changes in the sediment supply from the Amazon River despite the fact

that these sites are located beyond the position of the Amazon Fan. Alteration of the

seawater values by the detrital inputs appear to be driven by reversible scavenging and

the major changes record shifts in the provenance of the source material that can be

used to understand the evolution of the Amazon drainage system. Prior to ~8 Ma Nd

and Pb isotopic compositions of detrital silicate fractions suggest this material was

derived from an Amazon lowlands source. This shift in isotopic values at ~8 Ma

documents a shift to detrital silicates dominantly sourced from the Andean Highlands.

The timing of this transition supports previous arguments for the development of a

transcontinental connection to the Andes at that time.

107

LIST OF REFERENCES

Abouchami, W., S. L. Goldstein, S. J. G. Galer, A. Eisenhauer, and A. Mangini (1997), Secular changes of lead and neodymium in central Pacific seawater recorded by a Fe-Mn crust, Geochim. Cosmochim. Acta, 61(18), 3957-3974.

Albarède F., S. L. Goldstein, and D. Dautel D. (1997), The neodymium isotopic

composition of manganese nodules from the Southern and Indian oceans, the global oceanic neodymium budget, and their bearings on the deep ocean circulation, Geochim. Cosmochim. Acta, 61, 1277-1291.

Alpers, C., and G. Brimhall (1988), Middle Miocene climate change in the Atacama

Desert, northern Chile: evidence from supergene mineralization at La Escondida, Geol. Soc. Am. Bull., 100, 1640-1656.

Arsouze, T., J. -C. Dutay, F. Lacan, and C. Jeandel (2009), Reconstructing the Nd

oceanic cycle using a coupled dynamical-biogeochemical model, Biogeosciences, 6, 2829-2846.

Basak C., E. E. Martin, G. D. Kamenov (2011), Seawater Pb isotopes extracted from

Cenozoic marine sediments, Chemical Geology, 286, 94-108. Bayon, G., C. R. German, K. W. Burton, R. W. Nesbitt, and N. Rogers (2004),

Sedimentary Fe-Mn oxyhydroxides as paleoceanographic archives and the role of aeolian flux in regulating oceanic dissolved REE, Earth Planet. Sci. Lett., 224, 477-492.

Belshaw, N. S.,P. A. Freedman, R. K. O’Nions, M. Frank, and Y. Guo, Y. (1998), A new

variable dispersion double-focusing plasma mass spectrometer with performance illustrated for Pb isotopes, International Journal of Mass Spectrometry, 181, 51-58.

Bertram C. J., and H. Elderfield (1993), The geochemical balance of the rare earth

elements and Nd isotopes in the oceans, Geochim. Cosmochim. Acta, 57, 1957-1986.

Billups, K., A. C. Ravelo, and J. C. Zachos (1998), Early Pliocene deep water circulation

in the western equatorial Atlantic: Implications for high-latitude climate change, Paleoceanography, 13, 84-95

Broecker, W. S., and T. –H. Peng (1982), Tracers in the Sea, Eldigio Press. Buckry, D. (1973), Low-latitude coccolith biostratigraphic zonation, in Initial Reports

DSDP 15, edited by N. T. Edgar, J. B. Saunders et al., pp. 685-703, U.S. Govt. Printing Office, Washington.

108

Burton, K. W., H. –F. Ling, and R. K. O’Nions (1997), Closure of the Central American Isthmus and its effect on deep-water formation in the north Atlantic, Nature, 386, 382-385, doi: 10.1038/386382a0.

Burton, K. W., D. –C. Lee, J. N. Christensen, A. N. Halliday, and J. R. Hein (1999),

Actual timing of neodymium isotopic variations recorded by Fe-Mn crusts in the western North Atlantic, Earth Planet. Sci. Lett., 171, 149-156, doi: 10.1016/S0012-821X(99)00138-7.

Campbell, K. E. Jr., C. D. Frailey, and L. Romero-Pittman (2006), The Pan-Amazonian

Ucayali Peneplain, late Neogene sedimentation in Amazonia, and the birth of the modern Amazon River system, Palaeogeogr., Palaeoclimatol., Palaeoecol., 239, 166-219.

Campbell, K. E. (2010), Late Miocene onset of the Amazon River and the Amazon

deep-sea fan: Evidence from the Foz do Amazonas Basin: Comment, Geology, 38, E212-E212.

Carter, L., R. M. Carter, I. N. McCave, and J. Gamble (1996), Regional sediment

recycling in the abyssal southwest Pacific Ocean, Geology, 24, 735-738. Carter, R. M., et al. (1999) Proceedings of the Ocean Drilling Program, Initial Reports,

vol. 181, Ocean Drill. Program, College Station, Tex. Carter, L., and I. N. McCave (1997), Development of sediment drifts approaching an

active plate margin under the SW Pacific Deep Western Boundary Current, Paleoceanography, 9(6), 1061-1085.

Carter, L., and J. Wilkin (1999), Abyssal circulation around New Zealand-A comparison

between observations and a global circulation model, Marine Geology, 159, 221-239.

Castro, J. C., K. Miura, and J. A. E. Braga (1978), Stratigraphic and structural

framework of the Foz do Amazonas Basin, Annual Offshore Technology Conference Proceedings, 3, 1843-1847

Chiasson, W. P., and S. L. D’Hondt (2000), Neogene Planktonic Foraminifer

Biostratigraphy at Site 999, Western Caribbean Sea, in Proceedings of the Ocean Drilling Program, Scientific Results,165, edited by Leckie, R. M., H. Siquardssom,, G. D. Acton, and G. Draper, pp. 19-56, Ocean Drilling Program, College Station, Texas.

Coates A. G., J. B. Jackson, L. S. Collins, T. M. Cronin, H. J. Dowsett, L. M. Bybell, P.

Jung, and J. A. Obando (1992), Closure of the Isthmus of Panama: the near-shore marine record of Costa Rica and Western Panama, Geol. Soc. Am. Bull., 104, 814-828.

109

Coates, A. G., M. -P. Aubry, W. A. Berggren, L. S. Collins, and M. Kunk (2003), Early Neogene history of the Central American arc from Bocas del Toro, western Panama. Geol. Soc. Am. Bull., 115(3), 271-287.

Coates, A. G., L. S. Collins, M. -P Aubry, and W. A. Berggren (2004), The Geology of

the Darien, Panama, and the late Miocene-Pliocene collision of the Panama arc with northwestern South America, Geol. Soc. Am. Bull., 116(11/12), 1327-1344; doi: 10.1130/B25275.1.

Collins, L. S., A. F. Budd, and A. G. Coates (1996), Earliest evolution associated with

closure of the Tropical American Seaway, Proc. Natl. Acad. Sci., 93, 6069-6072. Curry, W. B., N. J. Shackleton, C. Richter, and Shipboard Scientific Party (1995), Leg

154 Introduction, In Proc. ODP, Int. Reports, 154, edited by Curry, W. B., N. J. Shackleton, C. Richter, et al., pp. 5-10, Ocean Drilling Program, College Station, Texas.

Damuth, J. E., R. D. Flood, R. O. Kowsmann, R. H. Belderson, and M. A. Gorini (1988),

Anatomy and growth pattern of Amazon deep-sea fan as revealed by long-range side-scan sonar (GLORIA) and high-resolution seismic studies, American Association of Petroleum Geologists Bulletin, 72, 885-911

Dobson, D. M., G. R. Dickens, and D. K. Rea (1997), Terrigenous Sedimentation at

Ceara Rise, in Proc. Ocean Drilling Program, Sci. Results, 154, edited by Shackleton, N. J., W. B. Curry, C. Richter, and T. J. Bralower, pp. 465-473, Ocean Drilling Program, College Station, Texas.

Dobson, D. M., G. R. Dickens, and D. K. Rea (2001), Terrigenous sediment on Ceara

Rise: A Cenozoic record of South American orogeny and erosion, Palaeogeogr., Palaeoclimatol., Palaeoecol., 165, 215-229.

Droxler, A. W., K. C. Burke, A. D. Cunningham, A. C. Hine, E. Rosencrantz, D. S.

Duncan, P. Hallock, and E. Robinson (1998), Caribbean constraints on circulation between Atlantic and Pacific Oceans over the past 40 million years, in Tectonic Boundary Conditions for Climate Reconstructions, edited by Crowley, T. J., and K.C. Burke, pp. 160-190, New York, Oxford Univ. Press.

Duque-Caro, H. (1990), Neogene stratigraphy, paleoceanography and

paleobiogeography in northwest South America and the evolution of the Panama Seaway. Palaeogeogr., Palaeoclimatol., Palaeoecol., 77, 203-234.

Elderfield, H., and M. J. Greaves (1982), The rare earth elements in seawater, Nature,

296, 214-219. Elderfield, H., and R. Pagett (1986), Rare earth elements in icthyoliths: Variations with

redox conditions and depositional environments, Sci. Total Environ., 49, 175-197.

110

Farrell, J. D., I. Raffi, T. R. Janecek, D. W. Murray, M. Levitan, K. A., Dadey, K. –C. Emeis, M. Lyle, J. –A. Flores, and S. Hovan (1995), Late Neogene sedimentation patterns in the eastern equatorial Pacific Ocean, in Proceedings of the Ocean Drilling Program, Scientific Results, 138, edited by Pisias, N.G., L.A. Mayer, T. R. Janecek, A. Palmer-Julson, and T. H. van Andel, pp. 717-756, Ocean Drilling Program, College Station, Texas.

Feigenson, M. D., M. J. Carr, S. V. Maharaj, S. Juliano, and L. L. Bolge (2004), Lead

isotope composition of Central American volcanoes: Influence of the Galapagos plume, Geochemistry, Geophysics, Geosystems, 5, doi:10.1029/2003GC000621.

Figueiredo, J., C. Hoorn, P. van der Ven, and E. Soares (2009), Late Miocene onset of

the Amazon River and the Amazon deep-sea fan: Evidence from the Foz do Amazonas Basin, Geology, 37, 619-622.

Figueiredo, J., C. Hoorn, P van der Ven, and E. Soares (2010), Late Miocene onset of

the Amazon River and the Amazon deep-sea fan: Evidence from the Foz do Amazonas Basin: Reply, Geology, 38, E213-E213.

Flower, B. P., and J. P. Kennett (1995), Middle Miocene deepwater paleoceanography

in the southwest Pacific: Relations with East Antarctic Ice Sheet development, Paleoceanography, 10(6), 1095-1112.

. Flower, B. P., J. C. Zachos, and E. Martin (1997), Latest Oligocene through early

Miocene isotopic stratigraphy and deep-water paleoceanography of the western equatorial Pacific: Sites 926 and 929, in Proc. Ocean Drilling Program, Sci. Results, 154, edited by Shackleton, N. J., W. B. Curry, C. Richter, and T. J. Bralower, pp. 451-462, Ocean Drilling Program, College Station, Texas.

Foster, G. L., D. Vance, and J. Prytulak (2007), No change in the neodymium isotope

composition of deep water exported from the North Atlantic on glacial-interglacial time scales, Geology, 35, 37-40.

Frank, M., B. C. Reynolds, and R. K. O'Nions (1999), Nd and Pb isotopes in Atlantic and Pacific water masses before and after closure of the Panama gateway, Geology 27, 1147-1150.

Frank, M., N. Whiteley, S. Kasten, J. R. Hein, and R. K. O'Nions (2002), North Atlantic

Deep Water export to the Southern Ocean over the past 14 Myr: Evidence from Nd and Pb isotopes in ferromanganese crusts, Paleoceanography,17, doi:10.1029/2000PA000606.

Frank, M., T. van de Flierdt, A.N. Halliday, P.W. Kubik, B. Hattendorf, and D. Gunther

(2003), Evolution of deepwater mixing and weathering inputs in the central Atlantic Ocean over the past 33 Myr, Paleoceanography, 18.

111

Frank, M. (2002), Radiogenic isotopes: tracers of past ocean circulation and erosional input, Reviews in Geophysics, 40, 1-38.

Futa, K., and C. R. Stern (1988), Sr and Nd isotopic trace-element compositions of

Quaternary volcanic centers of the southern Andes, Earth and Planetary Science Letters, 88(3-4), 253-262, doi: 10.1016/0012-821X(88)90082-9.

Gibbs, R. J. (1967), The geochemistry of the Amazon River system. Part I. The factors

that control the salinity and the composition and concentration of the suspended solids, Geol. Soc. Am. Bull., 78, 1203-1232.

Goldstein, S. L., and S.R. Hemming (2003), Long-lived isotopic tracers in

oceanography, paleoceanography and ice sheet dynamics, in Treatise on Geochemistry, 6, Edited by H. Elderfield, pp. 453-489, Elsevier.

Groeneveld, J., D. Nürnberg, R. Tiedemann, G. –J. Reichart, S. Steph, L. Reuning, D.

Crudeli, P. Mason (2008), Foraminiferal Mg/Ca increase in the Caribbean during the Pliocene: Western Atlantic Warm Pool formation, salinity influence, or diagenetic overprint?,Geochem. Geophys. Geosys., 9(1), doi: 10.1029/2006GC001564.

Gutjahr, M., M. Frank, C. H. Stirling, L. D. Keigwin, and A. N. Halliday (2008), Tracing

the Nd isotope evolution of North Atlantic Deep and Intermediate Waters in the western North Atlantic since the Last Glacial Maximum from Blake Ridge sediments, Earth Planet. Sci. Lett,. 266, 61-77.

Haddad, G. A., and A. W. Droxler (1996), Metastable CaCO3 dissolution at intermediate

water depths of the Caribbean and western North Atlantic: Implications for intermediate water circulation during the past 200,000 years, Paleoceanography, 11, 701-716.

Hall, I. R., I. N. McCave, R. Zahn, L. Carter, P. C. Knutz, and G. P. Weedon (2003),

Paleocurrent reconstruction of the deep Pacific inflow during the middle Miocene: Reflections of East Antarctic Ice Sheet growth, Paleoceanography, 18(2), doi: 10.1029/2002PA000817.

Harris, S. E., and A. C. Mix (2002), Climate and tectonic influences on continental

erosion of tropical South America, 0-13 Ma, Geology, 30, 447-450 Haug, G. H., and R. Tiedemann (1998), Effect of the formation of the Isthmus of

Panama on Atlantic Ocean thermohaline circulation, Nature, 393, 673-676. Haug, G. H., R. Tiedemann, R. Zahn, and A. C. Ravelo, (2001), Role of Panamanian

uplift on oceanic freshwater balance, Geology, 29, 207-210.

112

Hildreth, W., and S. Moorbath (1988), Crustal contributions to arc magmatism in the Andes of Central Chile, Contributions to Mineralogy and Petrology, 98(4), 455-489, doi: 10.1007/BF00372365.

Henry, F., C. Jeandel, and J. -F. Minster (1994), Particulate and dissolved Nd in the

western Mediterranean Sea: Sources, fates and budget, Marine Chemistry, 45, 283-305.

Hoorn, C., J. Guerrero, G. A. Sarmiento, and M. A. Lorente (1995), Andean tectonics as

a cause for changing drainage patterns in Miocene northern South America, Geology, 23, 237-240

Hoorn, C., F. P. Wesselingh, H. ter Steege, M. A. Bermudez, A. Mora, J. Sevink, I.

Sanmartín, A. Sanchez-Meseguer, C. L. Anderson, J. P. Figueiredo, C. Jaramillo, D. Riff, F. R. Negri, H. Hooghiemstra, J. Lundberg, T. Stadler, T. Särkinen, and A. Antonelli (2010), Amazonia Through Time: Andean Uplift, Climate Change, Landscape Evolution, and Biodiversity, Science, 330, 927.

Hoorn, C. (1993), Marine Incursions and the Influence of Andean Tectonics on the

Miocene Depositional History of Northwestern Amazonia-Results of Palynostratigraphic Study, Palaeogeogr., Palaeoclimatol., Palaeoecol., 105, 267-309.

Hoorn, C. (1994), An environmental reconstruction of the palaeo-Amazon River system

(middle to late Miocene, NW Amazonia), Palaeogeogr., Palaeoclimatol., Palaeoecol., 112, 187-238.

Horikawa, K., E. E. Martin, Y. Asahara, and T. Sagawa (2011), Limits on conservative

behavior of Nd isotopes in seawater assessed from analysis of fish teeth from Pacific core tops, Earth Planet. Sci. Lett., 310, 119-130.

Huber, M., H. Brinkhuis, C. E. Stickley, K. Doos, A. Sluijs, J. Warnaar, S. A.

Schellenberg, and G. L. Williams (2004), Eocene circulation of the Southern Ocean: Was Antarctica kept warm by subtropical waters?, Paleoceanography, 19(4), PA4026, doi: 10.1029/2004PA001014.

Huber, M., and D. Nof (2006), The ocean circulation in the southern hemisphere and its

climatic impacts in the Eocene, Paleogeography paleoclimatology paleoecology, 231(1-2), 9-28, doi: 10.1016/j.palaeo.2005.07.037.

Jacobsen, S. B., and G. J. Wasserburg (1980), Sm-Nd isotopic evolution of chondrites,

Earth Planet. Sci. Lett., 50, 139-155 Jain, S., L. S. Collins, and L.-A. C. Hayek (2007), Relationship of benthic foraminiferal

diversity to paleoproductivity in the Neogene Caribbean, Palaeogeogr., Palaeoclimatol., Palaeoecol., 255, 223-245.

113

Jeandel, C. (1993), Concentration and isotopic compositions of Nd in the South Atlantic Ocean, Earth Planet. Sci. Lett., 117, 581-591.

Jeandel, C., J. K. Bishop, and A. Zindler (1995), Exchange of Nd and its isotopes

between seawater small and large particles in the Sargasso Sea, Geochim. Cosmochim. Acta, 59, 535-547.

Jeandel, C., D. Thouron, M. Fieux (1998), Concentrations and isotopic compositions of

Nd in the Eastern Indian Ocean and Indonesian Straits, Geochim. Cosmochim. Acta, 62, 2597-2607.

Jones, C. E., A. N. Halliday, D. K. Rea, R. M. Owen (1994), Neodymium isotopic

variations in North Pacific modern silicate sediment and the insignificance of detrital REE contributions to seawater, Earth Planet. Sci. Lett., 127, 55-66.

Kamenov, G. D., P. Mueller, M. Perfit (2004), Optimization of mixed Pb-Tl solutions for

high precision isotope analyses by MC-ICP-MS, Journal of Analytical Atomic Spectrometry, 19, 1262-1267

Kameo, K., and T. J. Bralower (2000), Neogene nannofossil biostratigraphy of Sites

998, 999, and 1000, in Proceedings of the Ocean Drilling Program, Scientific Results, 165, edited by Leckie, R. M., H. Siquardssom, G. D. Acton, and G. Draper, pp. 3-18., Ocean Drilling Program, College Station, Texas.

Kameo, K., and T. Sato (2000), Biogeography of Neogene calcareous nannofossils in

the Caribbean and the eastern equatorial Pacific-floral response to the emergence of the Isthmus of Panama, Marine Micropaleontology, 39, 201–218, doi: 10.1016/S0377-8398(00)00021-9.

Keigwin, L.D. (1978), Pliocene closing of the Isthmus of Panama, based on

biostratigraphic evidence from nearby Pacific Ocean and Caribbean Sea cores, Geology, 31, 291-294.

Keigwin, L.D. (1982), Isotopic Paleoceanography of the Caribbean and East Pacific:

Role of Panama Uplift in Late Neogene Time, Science, 217, 350-353. Keller G., C. E. Zenker, and S. M. Stone (1989), Late Neogene history of the Pacific-

Caribbean gateway, Journal of South American Earth Science, 2, 73-108. Kennett, J.P. (1977), Cenozoic evolution of Antarctic Glaciation, the Circum-Antarctic

Ocean and their impact on global paleoceanography, Journal of Geophysical Research, 82, 3843-3860.

114

Kennett, J. P., and P. F. Barker (1990), Latest Cretaceous to Cenozoic climate and oceanographic developments in the Weddell Sea, Antarctica: An ocean-drilling perspective, in Proceedings Ocean Drilling Program, Scientific Results, 113, 937-960.

King, T.A., W. G Jr. Ellis, D. W. Murray, N. J. Shackleton, and S. Harris, (1997),

Miocene evolution of carbonate sedimentation at the Ceara Rise: a multivariate date/proxy approach, In Proc. ODP, Sci. Results, 154, edited by Shackleton N. J., W. B. Curry, C. Richter, and T. J. Bralower, pp. 349-366, Ocean Drilling Program, College Station, Texas.

Klocker, A., M. Prange, and M. Schulz (2005), Testing the influence of the Central

American Seaway on orbitally forced Northern Hemisphere glaciation, Geophys. Res. Lett., 32, L03703.

Kroopnick, P. M. (1985), The distribution of 13C in the world oceans, Deep Sea

Research, 32, 57-84. Kawabe, M., and S. Fujio (2010), Pacific Ocean Circulation Based on Observation,

Jour. of Oceangraphy, 66, 389-403. Lacan, F., and C. Jeandel (2001), Tracing Papua New Guinea imprint on the central

Equatorial Pacific Ocean using neodymium isotopic compositions and Rare Earth Element patterns, Earth Planet. Sci. Lett., 186, 497-512.

Lacan, F., and C. Jeandel (2005a), Acquisition of the neodymium isotopic composition

of the North Atlantic Deep Water, Geochem. Geophys. Geosyst., 6, doi: 10.1029/2005GC000956

Lacan, F., and C. Jeandel (2005b), Neodymium isotopes as a new tool for quantifying

exchange fluxes at the continent-ocean interface, Earth Planet. Sci. Lett., 232, 245-257.

Lee, J., R. J. Stern, and S. H. Bloomer (1995), Forty million years of magmatic evolution

in the Mariana arc: The tephra glass record, Journal of Geophysical Research, 100(B9), 17,671-17,687.

Lindsay, J. M., A. K. Schmitt, R. B. Trumbull, S. L. De Silva, W. Siebel, and R.

Emmerman (2001), Magmatic Evolution of the La Pacana Caldera System, Central Andes, Chile: Compositional Variation of Two Cogenetic, Large-Volume Felsic Ignimbrites, Journ. of Petrology, 42(3), 459-486.

Ling, H. -F., K. W. Burton, R. K. O'Nions, B.S. Kamber, F. von Blankenburg, A.J. Gibb,

and J.R. Hein (1997), Evolution of Nd and Pb isotopes in Central Pacific seawater from ferromanganese crusts, Earth Planet. Sci. Lett., 146, 1-12.

115

Ling, H. -F., S. -Y. Jian, M. Frank, H. –Y. Zhou, F. Zhou, Z. –L. Lu, X. –M. Chen, Y. –H. Jiang, and C. –D. Ge (2005), Differing controls over the Cenozoic Pb and Nd isotope evolution of deepwater in the central North Pacific Ocean, Earth and Planetary Science Letters, 232, 345-361.

Lonsdale, P. (1976), Abyssal circulation of the southeastern Pacific and some

geological implications, Journal of Geophysical Research, 81, 1163-1176. Lunt, D.J., P.J. Valdes, A. Haywood, and I.C. Rutt (2008), Closure of the Panama

Seaway during the Pliocene: implications for climate and Northern Hemisphere glaciation, Clim. Dyn., 30, doi:10.1007/s00382-007-0265-6.

Lyle, M., K. A. Dadey, and J. W. Farrell (1995), The late Miocene (11-8 Ma) eastern

Pacific carbonate crash: Evidence for reorganization of deep-water circulation by the closure of the Panama Gateway, in Proceedings of the Ocean Drilling Program, Scientific Results, 138, edited by Pisias, N.G., L.A. Mayer, T. R. Janecek, A. Palmer-Julson, and T. H. van Andel, pp. 821-838, Ocean Drilling Program, College Station, Texas.

Maier-Reimer, E., U. Mikolajewicz, and T. J. Crowley (1990), Ocean general circulation

model sensitivity experiment with an open Central American Isthmus, Paleoceanography, 5, 349-366.

Marshall, L. G. (1985), Geochronology and land-mammal biochronology of the

transamaerican faunal interchange, in The Great American Biotic Interchange, edited by Stelhi, F. G., and S. D. Webb, pp. 49-85, Springer, New York.

Martin, E. E., and B. A. Haley (2000), Fossil fish teeth as proxies for seawater Sr and

Nd isotopes, Geochim. Cosmochim. Acta, 64, 835-847. Martin, E. E., and H. Scher (2004), Preservation of seawater Sr and Nd in fossil fish

teeth: bad news and good news, Earth Planet. Sci. Lett., 220, 25-39. Martin, E.E., S. W. Blair, G. D. Kamenov, H. D. Scher, E. Bourbon, C. Basak, and D. R.

Newkirk (2010), Extraction of Nd isotopes from bulk deep sea sediments for paleoceanographic studies on Cenozoic time scales, Chemical Geology, 269, 414-431.

Martini, E. (1971), Standard Tertiary and Quaternary calcareous nannoplankton

zonation, In: Proc. 2nd Planktonic Conf. Roma, edited by Farinacci, Rome. Matsumoto, K., T. Oba, J. Lynch-Stieglitz, and H. Yamamoto (2002), Interior

hydrography and circulation of the glacial Pacific Ocean, Quaternary Science Reviews, 21, 1693-1704.

116

Mayer, L. A., T. H. Shipley, and E. L. Winterer (1986), Equatorial Pacific seismic reflectors as indicators of global oceanographic events, Science, 233, 761-764.

Mayer, L., N. Pisias, T. Janecek et al. (1992), Proceeding of the Ocean Drilling

Program, Initial Reports, Ocean Drilling Program, College Station, TX. McDaniel, D. K., S. M. McLennan, and G. N. Hanson (1997), Provenance of the

Amazon Fan muds: Constraints from Nd and Pb isotopes, in in Proceedings of the Ocean Drilling Program, Scientific Results,155, edited by Flood, R. D., D. J. W. Piper, A. Klaus, and L. C. Peterson, pp. 169-176, Ocean Drilling Program, College Station, Texas.

Megard, F. (1984), The Andean orogenic period and its major structures in central and

northern Peru, J. Geol. Soc. London, 141, 893-900 Mikolajewicz, U., and T. J. Crowley (1997), Response of a coupled ocean/energy

balance model to restricted flow through the Central American isthmus, Paleoceanography, 12, 429-441.

Mikolajewicz, U., E. Maier-Reimer, T. J. Crowley, and K. –Y. Kim (1993), Effect of Drake

and Panamanian Gateways on the circulation of an ocean model, Paleoceanography, 8, 409-426.

Milliman, J. D. (1979), Morphology and Structure of Amazon Upper Continental-Margin,

AAPG Bulletin- American Association of Petroleum Geologists, 63, 934-950. Mix, A. C., R. Tiedemann, P. Blum et al. (2003), Proceedings of the Ocean Drilling

Program,Initial Reports. Ocean Drilling Program, College Station, TX. Molina-Cruz, A. (1977), The relation of the southern trade winds to upwelling processes

during the last 75,000 years, Quat. Res., 8, 324-339. Molnar, P. (2008), Closing of the Central American Seaway and the Ice Age: A critical

review, Paleoceanography, 23, doi:10.1029/2007PA001574. Montes, C., A. Cardona, R. McFadden, S. E. Morón, C. A. Silva, S. Restrepo-Moreno,

N. Ramírez, Hoyos, J. Wilson, D. Farris, G. A. Bayona, C. A. Jaramillo, V. Valencia, J. Bryan, and J. A. Flores (2012), Evidence for middle Eocene and younger land emergence in central Panama: Implications for Isthmus closure, Geological Society of America Bulletin, doi: 10.1130/B30528.1

Murdock, T. Q., A. J. Weaver, and A. F. Fanning (1997), Paleoclimatic response of the

closing of the Isthmus of Panama in a coupled ocean-atmosphere model, Geophys. Res. Lett., 24, 253-256.

117

Newkirk, D. R., and E. E. Martin (2009), Circulation through the Central American Seaway during the Miocene carbonate crash, Geology, 37, 87-90., doi:10.1130/G25193A.1.

Nisancioglu, K. H., M. E. Raymo, and P. H. Stone (2003), Reorganization of Miocene

deep water circulation in response to the shoaling of the Central American Seaway, Paleoceanography, 18(1), 1006, doi:10.1029/2002PA000767.

Nof, D., and S. van Gorder (2003), Did an open Panama Isthmus correspond to an

invasion of Pacific water into the Atlantic?, Journ. Phys. Oceanogr., 33, 1324-1336.

O'Nions, R. K., M. Frank, F. von Blanckenburg, and H. –F. Ling (1998), Secular

variation of Nd and Pb isotopes in ferromanganese crusts from the Atlantic, Indian and Pacific Oceans, Earth Planet. Sci. Lett., 155, 15-28.

Orsi, A. H., T. Whitworth, and W. D. Nowlin (1995), On the meridional extent and fronts

of the Antarctic Circumpolar Current, Deep-Sea Research Part 1-Oceanographic Research Papers, 42(5), 641-673.

Orsi, A. H., G. C. Johnson, and J. L. Bullister (1999), Circulation, mixing, and production

of Antarctic Bottom Water, Progress in Oceanography, 43, 55-109. Pahnke, K., S. L. Goldstein, and S. R. Hemming (2008), Abrupt changes in Antarctic

Intermediate Water circulation over the past 25,000 years, Nature Geoscience, 1, 870-874, doi: 10.1038/ngeo360.

Palmer, M. R., and H. Elderfield (1985), Variations in the Nd isotopic composition of

foraminifera from Atlantic Ocean sediments, Earth Planet. Sci. Lett., 73, 299-305 Piepgras, D. J., and G. J. Wasserburg (1982), Isotopic composition of neodymium in

waters from the Drake Passage, Science, 217, 207-214. Piepgras, D. J., and G. J. Wasserburg (1985), Strontium and neodymium isotopes in hot

springs on the East Pacific Rise and Guaymas Basin, Earth Planet. Sci. Lett., 72, 341-356.

Piepgras, D. J., and G. J. Wasserburg (1987), Rare earth element transport in the

western North Atlantic inferred from Nd isotopic observations, Geochim. Cosmochim. Acta, 51, 1257-1271.

Pin, C., and J. F. S. Zalguegui (1997), Sequential separation of light rare-earth

elements, thorium and uranium by miniaturized extraction chromatography: Application to isotopic analyses of silicate rocks, Analy. Chim. Acta, 339, 79-89.

118

Pindell, J. L. (1994), Evolution of the Gulf of Mexico and Caribbean, in Caribbean Geology: An Introduction, edited by Donovan, S., and T. A. Jackson, pp. 13-39, University of West Indies Publ. Assoc., Kingston, Jamaica.

Pinet, B., D. Lajat, P. Le Quellec, and P. Bouysse (1985), Structure of Aves Ridge and

Grenada Basin from multichannel seismic data, in Geodynamique des Caraibes., edited by Mascle, A, Editions Technip., Paris.

Poore, H. R., R. Samworth, N. J. White, S. M. Jones, and I. N. McCave (2006),

Neogene overflow of Northern Component Water at teh Greenland-Scotland Ridge, Geochemistry, Geophysics, Geosystems, 7.

Prange, M., and M. Schulz (2004), A coastal upwelling seesaw in the Atlantic Ocean as

a result of the closure of the Central American Seaway, Geophys. Res. Lett., 31, L17207.

Raffi, I., and J. –A. Flores (1995), Pleistocene through Miocene calcareous nannofossils

from eastern equatorial Pacific Ocean (Leg 138), in Proceedings of the Ocean Drilling Program, Scientific Results, 138, edited by Pisias, N.G., L.A. Mayer, T. R. Janecek, A. Palmer-Julson, and T. H. van Andel, pp. 233-286, Ocean Drilling Program, College Station, Texas.

Reid, J.L. (1997), On the total geostrophic circulation of the pacific ocean: flow patterns,

tracers, and transports, Progress in Oceanography, 39, 263-352. Reynolds, B. C., M. Frank, and R. K. O'Nions (1999), Nd- and Pb- isotope time series

from Atlantic ferromanganese crusts: implications for changes in provenance and paleocirculation over the last 8 Myr, Earth Planet. Sci. Lett., 173, 381-396.

Roth, J. M., A. W. Droxler, and K. Kameo (2000), The Caribbean carbonate crash at the

middle to late Miocene transition: linkage to the establishment of the modern global ocean conveyor, in Proceedings of the Ocean Drilling Program, Scientific Results,165, edited by Leckie, R. M., H. Siquardssom,, G. D. Acton, and G. Draper, pp. 249-273, Ocean Drilling Program, College Station, Texas.

Rogers, G., and C. J. Hawkesworth(1989), A geochemical traverse across the North

Chilean Andes- evidence for crust generation from the mantle wedge, Earth Planet. Sci. Lett., 91(3-4), 271-285.

Scher, H. D., and E. E. Martin E.E. (2004), Circulation in the Southern Ocean during the

Paleogene inferred from Nd isotopes, Earth Planet. Sci. Lett., 28, 391-405. Scher, H. D., and E. E. Martin (2006), Timing and climatic consequences of the opening

of Drake Passage, Science, 312, 428-430. Schlitzer, R. (2010), Ocean Data View, http://odv.awi.de.

119

Schmitz, W. J. (1995), On the Interbasin-Scale Thermohaline Circulation, Reviews of Geophysics, 33(2), 151-173.

Schneider, B., and A. Schmittner (2006), Simulating the impact of the Panamanian

seaway closure on ocean circulation, marine productivity and nutrient cycling, Earth Planet. Sci. Lett., 246, 367-380.

Shackleton, N. J., S. Crowhurst, T. Hagelberg, N. G. Pisias, and D. A. Schneider

(1995a) A new Late Neogene time scale: application to Leg 138 sites, in Proceedings of the Ocean Drilling Program, Scientific Results, 138, edited by Pisias, N.G., L.A. Mayer, T. R. Janecek, A. Palmer-Julson, and T. H. van Andel, pp. 73-101, Ocean Drilling Program, College Station, Texas.

Shackleton, N. J., J. G. Baldauf, J. -A.Flores, M. Iwai, T. C. Moore Jr., I. Raffi, and E.

Vincent (1995b), Biostratigraphic Summary for Leg 138, in Proceedings of the Ocean Drilling Program, Scientific Results, 138, edited by Pisias, N.G., L.A. Mayer, T. R. Janecek, A. Palmer-Julson, and T. H. van Andel, pp. 517-536, Ocean Drilling Program, College Station, Texas, doi:10.2973/odp.proc.sr.138.127.1995.

Shackleton, N.J., and S. Crowhurst (1997), Sediment fluxes based on an orbitally tuned

time scale 5 Ma to 14 Ma, Site 926, In Proc. ODP, Sci. Results, 154, edited by Shackleton N. J., W. B. Curry, C. Richter, and T. J. Bralower, pp. 69-82, Ocean Drilling Program, College Station, Texas.

Shevenell, A.E., J. P. Kennett, and D. W. Lea, (2004), Middle Miocene Southern Ocean

Cooling and Antractic Cryosphere Expansion, Science, 305, 1766, doi: 10.1126/science.1100061.

Siddall, M., S. Khatiwala, T. van de Flierdt, K. Jones, S. L. Goldstein, S. Hemming, and

R. F. Anderson (2008), Towards explaining the Nd paradox using reversible scavenging in an ocean general circulation model, Earth Planet. Sci. Lett., 274, 448-461

Sigurdsson, H., R. M. Leckie, G. D. Acton et al. (1997), Proceeding of the Ocean Drilling

Program, Initial Reports, 165, Ocean Drilling Program, College Station, Texas. Sijp, W. P., and M. H. England (2004), Effect of the Drake Passage throughflow on

global climate, Journal of Physical Oceanoagraphy, 34(5), 1254-1266. Steph, S., R. Tiedemann, M. Prange, J. Groeneveld, D. Nürnberg, L. Reuning, M.

Schulz, and G. H. Haug (2006), Changes in Caribbean surface hydrography during the Pliocene shoaling of the Central American Seaway, Paleoceanography, 21, PA4221.

120

Tachikawa, K., C. Jeandel, and M. Roy-Barman (1999a), A new approach to the Nd residence time in the ocean: the role of atmospheric inputs, Earth Planet. Sci. Lett., 170, 433-446.

Tachikawa, K., C. Jeandel, A. Vangriesheim, and B. Dupre (1999b), Distribution of rare

earth elements and neodymium isotopes in suspended particles of the tropical Atlantic Ocean (EUMELI site), Deep-Sea Res. I: Oceanogr. Res. Pap., 46, 733-755.

Tachikawa, K. C., M. Roy-Barman, A. Michard, D. Thouron, D. Yeghicheyan, C. Jeandel

(2004), Neodymium isotopes in the Mediterranean Sea: comparison between seawater and sediment signals, Geochim. Cosmochim. Acta, 68, 3095-3106.

Talley, L. D. (1993), Distribution and formation of North Pacific intermediate water,

Journal of Physical Oceanography, 23, 517-537. Tanaka, T., S. Togashi, H. Kamioka, H. Amakawa, H. Kagami, T. Hamamoto, M.

Yuhara, Y. Orihashi, S. Yoneda, H. Shimizu, T. Kunimaru, K. Takahashi, T. Yanagi, T. Nakano, H. Fujimaki, R. Shinjo, Y. Asahara, M. Tanimizu, and C. Dragusanu (2000), JNdi-1: a neodymium isotopic reference in consistency with LaJolla neodymium, Chemical Geology, 168, 279-281.

Thomas, D. J. (2004), Evidene for deep-water production in the North Pacific Ocean

during the early Cenozoic warm interval, Nature, 430, 65-68. Thomas, D. J., T. J. Bralower, and C. E. Jones (2003), Nd isotopic reconstruction of

Late Paleocene-Early Eocene thermohaline circulation, Earth Planet. Sci. Lett., 290, 309-322.

Thomas, D. J., and R. K. Via (2007), Neogene evolution of Atlantic thermohaline

circulation: Perspective from Walvis Ridge, southeastern Atlantic Ocean, Paleoceanography, 22, PA2212, doi: 10.1029/2006PA001297.

Tiedemann, R., and S. O. Franz (1997), Deep-Water Circulation, Chemistry, and

Terrigenous Sediment Supply in the Equatorial Atlantic During the Pliocene, 3.3-2.6 Ma and 5-4.5 Ma, In Proc. ODP, Sci. Results, 154, edited by Shackleton N. J., W. B. Curry, C. Richter, and T. J. Bralower, pp. 299-318, Ocean Drilling Program, College Station, Texas.

Toggweiler, J.R., and B. Samuels (1995), Effect of Drake Passage on the global

thermohaline circulation, Deep-Sea Research I, 42(4), 477-500. Trumbull, R.B., R. Wittenbrink, K. Hahne, R. Emmermann, W. Büsch, H. Gerstenberger,

and W. Siebel (1999), Evidence for Late Miocene to Recent contamination of arc andesites by crustal melts in the Chilean Andes (25-26°S) and its geodynamic implications, Journal of South American Earth Sciences, 12, 135-155.

121

Tsuchiya, M., and L. D. Talley (1998), A Pacific hydrographic section at 88°W: Water-property distribution, Journal of Geophysical Research, 103(C6), 12,899-12,918.

van de Flierdt, T., M. Frank, A. N. Halliday, J. R. Hein, B. Hattendorf, D. Gunther and P.

W. Kubik (2004a), Deep and bottom water export from the Southern Ocean to the Pacific over the past 38 million years, Paleoceanography, 19, PA1020, doi: 10.1029/2003PA000923

van de Flierdt, T., M. Frank, D. –C. Lee, A. N. Halliday, B. C. Reynolds, and J. R. Hein

(2004b), New constraints on the sources and behavior of neodymium and hafnium in seawater from Pacific Ocean ferromanganese crusts, Geochim. Cosmochim. Acta., 68(19), 3827-3843.

Vincent, E. (1981), Neogene carbonate stratigraphy of Hess Rise (central North Pacific)

and paleoceanographic implications, in Deep Sea Drilling Project, Initial Reports, 62, edited by Thiede, J., T. L. Vallier et al., p. 571-606, Washington, U.S. Govt. Printing Office.

Warren, B. A. (1973), TransPacific hydrographic sections at latitudes 43°S and 28°S;

the SCORPIO Expedition-deep water, Deep Sea Research, 20, 9-38. Warren, B. A. (1981), Deep circulation of the world ocean, in Evolution of Physical

Oceanography, edited by Warren, B. A., and C. Wunsch, pp. 6-41, MIT Press, Cambridge, Mass.

Webb, S. D. (1985), Late Cenozoic of mammal dispersals between the Americas, in the

Great American Biota Interchange, edited by Stehli, F. G., and S. D. Webb, pp. 357-386, Springer, New York.

Wright, J. D., and K. G. Miller (1993), Southern Ocean influence on late Eocene to

Miocene deep-water circulation, in The Antarctic Paleoenvironment: A Perspective on Global Change (Part 2), edited by Kennett, J. P., Antarctic Research Series, 60, 1-25.

Wright, J. D., and K. G. Miller (1996), Control on the North Atlantic Deep Water,

Paleoceanography, 11, 157-170. Wright, J. D., K. G. Miller, and R. G. Fairbanks (1991), Evolution of Modern Deepwater

Circulation: Evidence from the Late Miocene Southern Ocean, Paleoceanography, 6(2), 275-290, doi: 10.1029/91PA00346.

122

BIOGRAPHICAL SKETCH

Derrick Richard Newkirk was born in Indianapolis, Indiana. He is the eldest son of

Patricia and Richard Newkirk, and the older brother of Ryan Newkirk. His primary

education, elementary through high school, was completed in Greenwood, Indiana in

the Center Grove School District. While attending Indiana University-Purdue University

at Indianapolis he became interested in geology after taking an introductory course

taught by Bob Barr. After completion of his four years of eligibility for collegiate soccer,

he turned his focus to geology. During his undergraduate education he worked as a lab

assistant for Dr. Gabriel Filippelli, and helped Dr. Filippeli’s Ph.D. student at the time,

Dr. Jennifer Latimer. While working under Dr. Gabriel Filippelli and Dr. Jennifer Latimer

he worked on his own research project looking at human impacts on the watershed of

Laguna Zoncho, Costa Rica using phosphorus geochemistry. This invaluable

experience doing scientific research led him to graduate school. He completed his

degree in the summer of 2004 with a Bachelor of Science with a focus in geology. At

the University of Florida his research focused on the Miocene carbonate crash in the

Caribbean using Nd isotopes in fossil fish teeth to reconstruct ocean circulation. After

completion of the Master of Science degree in July of 2007, he continued on at the

University of Florida pursuing his Ph.D. under the guidance of Dr. Ellen Martin. While

continuing his education at the University of Florida, he tried to understand the global

effects of the formation of Central America on the global pattern of ocean circulation.

After graduation, he plans on pursuing a career in either academia or industry

depending on job availability.