Download A Paleoceanographic Reconstruction of the

A Paleoceanographic Reconstruction of the Mediterranean Sea: Stable Isotope Analysis
of Early Pliocene Foraminifera, Il Trave Sud, Italy
Tiffany A. Larsen
Senior Integrative Exercise
March 11, 2003
Submitted in partial fulfillment of the requirements for a Bachelor of Arts degree from
Carleton College, Northfield, Minnesota.
An Early Pliocene Paleoceanographic Reconstruction of the Mediterranean Sea:
Stable Isotope Analysis of Foraminifera, Il Trave Sud, Italy
Tiffany A. Larsen
Carleton College
Senior Integrative Exercise
March 11, 2003
Professor David M. Bice, Advisor
Abstract
Early Pliocene marine marls from Il Trave Sud, located along the Adriatic coast
of Italy, consist of rhythmic light-grey and dark-grey color variations. These variations
have previously been interpreted as cycles of warm, wet, estuarine climates, and cool,
dry, anti-estuarine climates, based on whole-rock carbon and oxygen stable isotope
analysis, calcimetry, and magnetic susceptibility by Borkowski (2002). Milankovitch
cycles of precession and obliquity were determined by to be the dominant driving forces
behind these climate alterations.
Proxies from marine sediments offer useful data for paleoenvironment
interpretations. Planktonic and benthic foraminifera provide detailed analysis of
paleoceanographic conditions due to their habitat in surface and deep water reservoirs.
Carbon and oxygen isotope analysis from foraminifera are used to reconstruct
paleoceanographic conditions including ocean circulation, biological marine productivity,
precipitation-evaporation rates, salinity, global ice volume, and paleotemperatures.
In this study, carbon and oxygen isotope analyses of Orbulina universa and
Planulina wuellerstorfi from Il Trave Sud generally support and add to the model
proposed by Borkowski (2002). A warm, wet, climate with a stratified water column
consists of low surface water δ13O and CaCO3 content, high surface δ13C, and high deep
water δ13O and δ13C. A cool, dry climate with a vertically mixed water column was
interpreted from high CaCO3 content, high surface δ13O and δ13C, and low deep water
δ13O and δ13C values. A third climate system was interpreted as a warm, wet climate with
some ocean mixing. This climate contains low surface δ13O and δ13C values, high CaCO3
content, and low deep water δ13O and δ13C values due to mixing of the water column.
Keywords: calcium carbonate, foraminifera, Italy, Orbulina universa, paleoclimatology,
Pliocene, stable isotopes
TABLE OF CONTENTS
Introduction……………………………………………………………………………..…1
Background…………………………………………………………………..……………4
Foraminifera as Paleoceanographic Proxies………………………..……………4
Oxygen Isotopes………………………………………………..……….…………5
Carbon Isotopes………………………………………………………..……….…9
Carbonate Content...…………………………………………………………..…11
Previous Work…………………………………………………….……………...12
Field Study Site…………………………………………………………………………..14
Methods…………...……………………………………………………………………...15
Results…………………………………………………………………………………....18
Discussion………………………………………………………………………………..22
Paleoceanographic Reconstructions…………………………………………….23
Conclusions………………………………………………………………………………31
Acknowledgements…………………………………………………………………....…34
References Cited…………………………………………………………………………35
1
INTRODUCTION
The Early Pliocene (~4.8-3.5 Ma) has been established as a warmer geological
time period relative to current global temperatures. This warm climate system is based
on previous stable isotope analyses from ocean drilling cores and stratigraphic studies
(Billups et al., 1997; Crowley, 1991; Faure, 1977; Haywood et al., 2000; Hodell et al.,
1986; Kennett, 1993; Kump, 1991; Zachos et al., 2001). Understanding the Pliocene
climate is integral to the study of global climate interactions, specifically during warmer
periods. Studies and models of warmer climate periods are of particular interest when
analyzing current global conditions.
Sedimentary sequences from the Mediterranean basin are commonly used for
studying Cenozoic climate variability (Foucault et al., 2000; Haywood et al., 2000;
Hodell et al., 1986; Williams et al., 1978) Marine sedimentary outcrops located in Italy
reflect climate changes in the rock record through alternating, rhythmic color and
lithological variations. The marl outcrop of Il Trave Sud (Fig. 1) located along the
western coast of the Adriatic Sea exemplifies such cycles with light-grey and dark-grey
color alternations.
Borkowski (2002) studied whole-rock carbon and oxygen isotope analysis,
calcimetry, and magnetic susceptibility data with the light-grey to dark-grey marl
alternations at Il Trave Sud (Fig. 2, 3). Borkowski (2002) concluded that Milankovitch
cycles were the overall driving forces behind these variations, recording two climate
conditions: a cool, dry period with strong vertical mixing within the water column, and a
warm, wet period with a stratified water column.
Planktonic and benthic foraminifera are often used as proxies for comparison of
2
Figure 1. Location of Il Trave Sud outcrop along the Adriatic
Sea, Italy (from Borkowski, 2002).
Figure 2. Photograph of Il Trave Sud outcrop. Orange dots increase in stratigraphic
sequence from right to left. The bars accentuate the alternating light-grey and dark-grey
color cycles, and are approximately 5.5 m in width. The thin, lighter layers are turbidites
(adapted from Borkowski, 2002).
3
Figure 3. Stratigraphic column of Il Trave Sud outcrop. The width of the marl beds
reflects the average CaCO3 content. The foraminifera and sample names represent the
beds analyzed for carbon and oxygen isotope analysis. Brackets at left exhibit
Milankovitch cycles correlated with the strata. The eccentricity cycle of 305 kyr spans
551 cm; the obliquity cycle of 43 kyr spans 74 cm; and the precession cycle of 23 kyr
spans 41 cm (adapted from Borkowski, 2002).
4
surface and deep water ocean conditions (Billups et al., 1998; Loutit et al., 1979; Zachos,
1993). Orbulina universa and Planulina wuellerstorfi (also classified under genus names
Cibicides, Cibicidoides, and Fontbotia) are common planktonic and benthic genera,
respectively, used in carbon and oxygen stable isotope analysis for paleoenvironment
studies (Billups et al., 1997; Hodell et al., 1986; Mackensen et al., 1993; Spero et al.,
1997; Vidal et al., 2002; Williams et al., 1978; Zachos, 1993). Carbon isotope values
derived from paleoceanographic proxies are useful for interpreting changes in organic
marine productivity, ocean circulation patterns and general carbon cycle changes.
Oxygen isotope values are useful for interpreting paleotemperatures, ice volume changes,
salinity, and therefore precipitation-evaporation rates and freshwater input changes
(Beyer, 1999; Billups et al., 1997; Faure, 1977; Hodell et al., 1986; Kennett et al., 1993;
Mackensen et al., 1993; Zachos, 1993).
In this study, carbon and oxygen isotope analyses of planktonic and benthic
foraminifera from Il Trave Sud are compared with whole-rock carbon and oxygen isotope
results from Borkowski (2002) in order to test the model based on the whole-rock results.
This comparison offers a more specific interpretation of paleoceanographic and
paleoenvironmental conditions of the alternating, rhythmic marls of Il Trave Sud, and
detailed data for surface and deep water conditions from foraminiferal proxies.
BACKGROUND
FORAMINIFERA AS PALEOCEANOGRAPHIC PROXIES
Stable isotope analyses of foraminifera are commonly used for interpretation of
paleoceanographic conditions. Foraminifera are single-celled protists with outer shells
5
called tests, and live in all marine environments. Planktonic foraminifera are surface
dwelling organisms, while benthic foraminifera live within or atop the sea floor.
Foraminifera use CaCO3 from ambient seawater to construct their tests, recording
the isotopic composition of the seawater during this test formation (Faure, 1977). The
fractionation between water and calcite during test formation is temperature dependent;
there is an approximate 0.25‰ decrease in carbonate δ18O per 1°C increase (Kim et al.,
1997). There are several other factors that may create deviations in isotope records of
foraminifera, such as vital effects, disequilibrium of foraminiferal shells and seawater,
test development and growth at varying depths within the water column, and postdepositional recalcification of tests (Emiliani, 1971; Wefer et al., 1999).
Planktonic foraminifera record conditions of the surface waters they inhabit
during test formation, and benthic foraminifera record bottom water conditions.
Planktonic foraminifera photosynthesize, while benthic foraminifera live on organic
material that falls to the seafloor. Since these two organisms live in different oceanic
reservoirs, a comparison of carbon and oxygen isotope values from their tests is used to
recreate paleoceanographic conditions at the time of deposition.
OXYGEN ISOTOPES
The ratio between 18O and 16O of from marine sediment is commonly used to
determine paleoclimate conditions such as evaporation, precipitation, salinity,
paleotemperature, and ice volume changes (Faure, 1977; Hodell et al., 1986; Spero et al.,
1997; Vidal et al., 2002). This ratio is expressed as the following:
6
(18 O/16O)sample - (18 O/16O)standard
 ×1000
δ O =


(18 O/16O)standard
18
16
O has a much larger abundance in our earth system than 18O, 99.756% and 0.205%
respectively, and a lighter mass than18O. This mass and abundance difference affects the
δ18O through fractionation processes between the ocean and atmosphere. The standard
commonly used for isotope analyses is the Pee Dee Belemnite (PDB), having δ18O and
δ13C values of zero (Epstein et al., 1953).
Evaporation, Precipitation, and Salinity
During evaporation of surface waters, there is a preferential transfer of the lighter
oxygen isotope, 16O, into the atmosphere. When the rate of evaporation exceeds
precipitation, surface waters increase in δ18O (Faure, 1977). This leads to a correlation
between high salinity when the rate of evaporation is greater than precipitation rates and
high δ18O.
During precipitation, high δ18O water vapor is initially removed from the
atmosphere due to the greater mass of 18O. This high δ18O water vapor, relative to the
remaining water vapor in the atmosphere, is low in δ18O with respect to normal ocean
water (Faure, 1977). As condensation continues, the atmosphere quickly decreases in
δ18O due to the initial removal of high δ18O water vapor (Faure, 1977; Wefer et al., 1999).
If precipitation persists for an extended period of time, the surface waters of the ocean or
where precipitation occurs shift to lower δ18O values. Isotopic fractionation during
evaporation and precipitation is temperature dependent because the saturation of water
7
vapor pressure is temperature dependent, meaning that fractionation decreases with an
increase in temperature (Faure, 1977).
Freshwater can be introduced into bodies of water directly by precipitation or by
continental runoff. Since continental freshwater typically has low (approximately –10‰)
δ18O values, the addition of freshwater though river systems influences the isotopic
composition of ocean surface waters toward a lighter δ18O value (Paul et al., 1999). This
freshwater creates a low salinity layer on the surface of the ocean due to its low density
and dilution of surface waters. This further contributes to lower δ18O values.
Paleotemperatures and global ice volume changes
δ18O in sediment records can reflect changes in paleotemperature and global ice
volume (Faure, 1977; Zachos, 1993). During cool, glacial periods, low δ18O concentrated
snowfall on ice sheets at low temperatures preferentially precipitates (Faure, 1977). This
leaves ocean waters high in δ18O. During warmer, interglacial periods, glacial melt of
these ice sheets adds more freshwater into the oceans, introducing low δ18O into surface
waters. Deep waters generally contain less 18O than surface waters due to dilution from
high 16O concentrated sinking polar waters (Millero, 1996).
δ18O records from the North Atlantic Ocean reflect temperature change associated
with glacial and interglacial periods (Fig. 4). During previous glacial periods, δ18O
values have ranged from 4‰ to 6‰. Previous interglacial δ18O values have ranged
between 3‰ and 4.5‰.
Important factors to consider when analyzing oxygen isotopes in the rock record
are post-depositional diagenetic influences and freshwater contamination. Diagenetic
8
Figure 4. δ18O and δ13C values from benthic foraminifera against depth in a core from the
eastern margin of the North Atlantic Ocean. δ13C isotopic values in the benthic
foraminifera are lower during glacial time periods and higher during interglacial periods
(from Broecker and Peng, 1982).
9
signals of oxygen isotope alteration in carbonate rock usually are characterized by light
δ18O values from cementation and low porosity (Arthur et al., 1991). Freshwater is
typically δ18O depleted, which lowers oxygen isotope values as a result of contamination.
It is difficult, however, to prove if diagenetic or freshwater contamination alterations
have or have not occurred (Faure, 1977). Since groundwater typically contains low
concentrations of carbon, there is not enough carbon to alter carbon isotope values.
CARBON ISOTOPES
The ratio between 13C and 12C is used for interpretation of marine productivity,
ocean circulation patterns, and general carbon cycle changes. The abundance of 12C and
13
C is 98.89% and 1.11% respectively. This ratio is defined as the following:
(13 C/12C)sample - (13 C/12C)standard
 ×1000
δ C =
13
12


( C/ C)standard
13
δ13C values from the North Atlantic Ocean vary slightly between previous glacial
and interglacial periods (Fig. 4). δ13C values range between 0‰ and –1.5‰ overall.
Glacial periods are lower in δ13C than interglacial periods.
Marine Productivity
Organisms that live within the photic zone of the oceans, such as planktonic
foraminifera, photosynthesize. In this process, organisms use 12C rather than 13C during
primary production (Wefer et al., 1999). Due to this preference, surface waters tend to
have high δ13C values, and deep waters are generally low in δ13C due to low δ13C
10
descending organic matter. Therefore, high productivity in oceans results in high δ13C,
and low productivity results in low δ13C.
Ocean Circulation
Carbon isotope data from planktonic and benthic organisms provide information
about vertical mixing and stratification within a water column. Surface waters normally
contain higher δ13C values than deep waters due to surface productivity, and deep waters
normally contain lower δ13C values due to deposition of organic matter (Kump, 1991).
Vertical mixing tends to homogenize the surface and deep waters, minimizing the effects
of productivity in the surface water. Therefore in a water column with strong vertical
mixing, δ13C values are usually similar between the two oceanic reservoirs.
Stratified water columns are a result of freshwater input creating density
differences between surface and deep waters; there is essentially no vertical mixing
between surface and deep waters. Freshwater introduces low δ13C input (approximately
–5‰) into surface waters (Derry et al., 1996).
Organic production in surface waters continue to consume 12C, resulting in higher
δ13C values. Without vertical mixing, low δ13C particles descend to the deep waters, but
no mixing between the surface and deep waters occurs. Surface waters increase in δ13C,
and deep waters decrease in δ13C, creating a distinct difference between carbon isotope
values between the two reservoirs (Kump, 1991).
11
The Carbon Cycle
Carbon supply in the ocean depends on the exchange of CO2 between the ocean
and atmosphere through molecular diffusion and input of terrigenous material from
erosional processes. CO2 in the surface waters of the ocean is consumed during
photosynthesis of marine organisms. Changes in the δ13C of atmospheric CO2 are
recorded in the marine sedimentary record. A decrease in atmospheric CO2 results in a
high δ13C record, and an increase in atmospheric CO2 results in low δ13C values.
Atmospheric CO2 content is associated with global temperatures, where decreases
in CO2 concentrations correspond with cooler, glacial periods, and increases in CO2
correspond with warmer, interglacial periods (Einsele et al., 1991). Therefore, warmer,
interglacial periods contain low δ13C oceanic and atmospheric values. Cooler, glacial
periods contain high δ13C oceanic and atmospheric values.
CARBONATE CONTENT
Calcimetry measurements of marine sediment determine the CaCO3 content
compared to the rock’s whole composition. CaCO3 content of marine sediment can vary
due to variations in surface productivity of carbonate producing organisms, changes in
sediment supply or dilution of the carbonate from other sediment, or from dissolution of
carbonates within the water column (de Boer, 1991).
Periods of high productivity generally result in high CaCO3 content, whereas low
CaCO3 reflects low productivity. Increases in sediment supply via fluvial processes can
dilute carbonate sediment, resulting in a lower CaCO3 content even though productivity
was at a steady rate.
12
PREVIOUS WORK
Borkowski (2002) concluded that the rhythmic color variations at Il Trave Sud are
due to orbital forcings, mainly precessional and obliquity. From CaCO3 content and
magnetic susceptibility data, precession was determined as the dominant cycle for the
first part of the outcrop, with a 41 cm cycle, and then changes to obliquity cycles, with 74
cm cycles.
Two climate conditions were derived from whole-rock stable isotope analysis,
calcimetry, and correlation of Milankovitch cycles (Fig. 5). Borkowski (2002) was able
to calculate a sedimentation rate of approximately 16.82 to 18.29 m/Ma by conducting
spectral analysis on CaCO3 values. Borkowski (2002) interpreted carbon isotope values
as representing surface water isotopic concentrations.
Borkowski (2002) interpreted a contemporaneous increase in δ13C and decrease in
δ18O and CaCO3 as wet, warm, estuarine climate periods. This climate period has a
higher rate of precipitation than evaporation, which caused an increase in fluvial transport
of low δ13C and δ18O concentrated sediment and nutrients into the Mediterranean basin.
The low δ13C freshwater input and increase in nutrients supplied surface waters with a
larger amount of consumable matter for surface primary productivity. The consumption
of such matter resulted in a high δ13C concentrated surface waters. A decrease in CaCO3
content and salinity is attributed to dilution from the increase in precipitation and
terrigenous sediment carried by freshwater entering the sea. This also created a low
density freshwater layer on the ocean surface, resulting in ocean stratification and lower
δ18O values.
13
Figure 5. δ13C against % CaCO3 and δ18O against δ13C from Borkowski (2002)
stable isotope analysis and calcimetry. This data shows δ13C is generally out
of phase with δ18O and % CaCO3 (adapted from Borkowski, 2002).
14
Borkowski (2002) interpreted a decrease in δ13C and an increase in δ18O and
CaCO3 as cool, dry, anti-estuarine climate periods. Arid conditions lead to a higher rate
of evaporation than precipitation and a decrease in continental runoff into the
Mediterranean Sea. This increase in evaporation and salinity resulted in higher δ18O
values. A reduction of precipitation and runoff decreases the amount of nutrients and 12C
supplied by fluvial processes, thus decreasing surface productivity and δ13C. A decrease
in fluvial input also reduces the amount of dilution of the seawater, creating a higher
CaCO3 content.
FIELD STUDY SITE
Depositional Environment
The marls at Il Trave Sud were deposited during the Early Pliocene (~4.8-3.5
Ma); a time of reopening and refilling of the Mediterranean Sea, following the Messinian
Salinity Crisis. This reintroduction of flow from the Atlantic Ocean into the
Mediterranean Sea is attributed to eustatic sea-level rise (Kennett et al., 1993). The
refilling of the Mediterranean basin occurred rapidly over a 100 yr period, at a rate of
40,000 cubic kilometers per year (Hsü, 1983).
From past stable isotope studies, Early Pliocene paleotemperatures have been
determined to be approximately 5°C warmer than current global mean temperatures
(Billups et al., 1997, 1998; Crowley, 1991; Hermoyian et al., 2001; Kennett et al., 1993;
Thunell, 1979). There is an overall trend throughout the Pliocene from a warm, humid
climate to a cooler, dry climate (Fauquette et al., 1998). Shorter secondary warm and
cool climate variations exist within the overall cooling trend. The end of the Pliocene has
15
been recorded as rapid glacial-interglacial cycles leading into the Pleistocene (Fauquette
et al., 1998).
Stratigraphy
Il Trave Sud outcrop is located along the Adriatic coast of central Italy (Fig. 1),
south of Ancona. The outcrop of marine carbonate exposes approximately 60 m of strata,
and is considered to be from the Zanclean (~4.8-3.5 Ma), the earliest age of the Pliocene
(Borkowski, 2002). There are large-scale alternations of light- and dark-grey marls
between 5 and 6 m thick (Fig. 2, 3). One large light-grey and dark-grey cycle (~11 m)
has been determined to represent 650 kyr (Borkowski, 2002).
Within these large colored beds, small-scale beds vary in color from light- to
dark-grey and red-beige with thickness ranging from 1-20 cm. There is no apparent
overall pattern of the small-scale colored beds. There is a general lack of bioturbation
throughout the section. Thin (<1 cm) layers of shell-like fossils are present throughout,
and are more abundant in the large, light-grey sections. Laminated beds and turbidites
(12-95 cm) exist throughout the section.
METHODS
Sample Collection and Preparation
To construct a comparative analysis, samples were collected from the same strata
that Borkowski (2002) studied. Five vertical sections spanning approximately one meter
in length were sampled based on individual bed thickness, ranging from 1.5 to 22 cm (fig.
3). Turbidites were excluded from sampling due to their abnormal representation of
paleoenvironmental conditions.
16
Calcimetry
Calcimetry measurements were conducted in order to determine the carbonate
content of each sample. Approximately two grams of each sample were crushed with a
mortar and pestle and sieved to 177 µm. An average of 705 mg per sample was
combined with a solution of 10% HCl/90% tap water in a calcimeter. A standard sample
of pure marble was run every 20 samples.
Foraminifera Collection
From calcimetry results, one high and one low percent CaCO3 layer from each
meter-length section sampled was selected for foraminiferal stable isotope analysis.
Samples were soaked in tap water for at least twelve hours until they softened and
disaggregated. Samples were wet sieved to 63 µm and dried beneath a heat lamp. Siltsized particles were collected during sieving by allowing the silt to settle out of solution
while clays remaining in suspension were decanted. Silt was collected for stable isotope
analysis to obtain isotope values from what are presumed to be nannofossils.
One planktonic (Orbulina universa) and one benthic (Planulina wuellerstorfi)
species was selected for stable isotope analysis from the residue (Fig. 6, 7). O. universa
was present throughout all samples with two exceptions, and the size of individual
foraminifera ranged from 200 µm to 480 µm. P. wuellerstorfi was present only in the
higher CaCO3 content samples, and the width across the individual foraminifera ranged
from 380 µm to 550 µm. Size selection was to ensure that the specimen was in an older,
more developed stage, approaching equilibrium with ambient conditions (Beyer, 1999).
Between 6 and 21 individual foraminifera per sample were collected for stable isotope
17
Figure 6. Orbulina universa. Scale bar is 1mm in length.
Figure 7. Planulina wuellerstorfi. Scale bar is 1mm in length.
18
analysis. An unknown benthic foraminifer was collected from samples of both high and
low CaCO3 content, in an attempt to correlate the unknown benthic isotope values with
the P. wuellerstorfi isotope values. This was conducted to create a more complete data
set since so few benthic foraminifera were present in low CaCO3 samples.
Carbon and oxygen stable isotope analysis was performed on each sample with an
Optima duel inlet mass spectrometer. Data are recorded in parts per mil (‰) relative to
the Pee Dee Belemnite (PDB) standard. Ratios of carbon and oxygen isotopes are
expressed as δ13C and δ18O.
RESULTS
Sample Observations
High CaCO3 sediment samples of ≥63 µm under a binocular microscope
contained a large variety of planktonic and benthic foraminifera species, but did not
appear to contain much else in composition. Low CaCO3 content sediment samples were
composed of a larger content of terrigenous material, and had a fairly low abundance and
specie diversity of planktonic and benthic foraminifera than in the higher CaCO3 samples.
Due to the lack of P. wuellerstorfi in several sediment samples, only five samples
contain both planktonic and benthic δ18O and δ13C values. These full data sets are
considered for creating paleoceanographic interpretations and for comparison with
Borkowski (2002) paleoclimate interpretations. Only one sets used the calibrated
unknown benthic foraminifera to P. wuellerstorfi isotope values.
It is common for benthic foraminiferal isotope values to be adjusted for vital
effects due to their endobiontic nature. This is conducted because benthic foraminifera
19
living within the sediment in the seafloor results in recordings of isotopic compositions of
the sediment rather than the seawater. P. wuellerstorfi, however, is considered an
epibenthic organism, which lives atop the bottom sediment and calcifies its test nearest to
equilibrium bottom water compositions (Mackensen et al., 1993; Spero et al., 1997).
Therefore, no vital effect adjustments were made on the P. wuellerstorfi isotope values.
Calcimetry
As previously mentioned, from the five one-meter sections sampled, one high and
one low CaCO3 sediment layer was selected for stable isotope analysis. Overall,
percentages of CaCO3 are relatively low, ranging from 11.26 to 44.03 percent (Table 1).
CaCO3 percentages from the larger light-grey strata are slightly higher than those from
the larger dark-grey strata.
Oxygen Isotopes
Oxygen isotope ratios for O. universa are generally positive, but range from
–0.9‰ to 0.9‰ (Table 1). δ18O values for P. wuellerstorfi are predominantly negative
and range from –0.78‰ to 0.545‰. Average δ18O values for the entire water column for
samples containing both planktonic and benthic foraminifera are recorded in Table 1.
Oxygen isotope ratios for the silt-sized samples are all negative with values falling
between –3.71‰ to –1.57‰. The silt δ18O values are in phase with CaCO3 content,
higher δ18O values correlating with higher CaCO3 content (Fig. 8).
Meter Level
100.6-100.64
100.84-100.9
101.67-101.79
101.95-102.05
104.26-104.35
104.5-104.53
108.61-108.66
108.77-108.81
110.74-110.77
110.96-111.02
% CaCO3
16.37
33.55
19.91
36.47
44.03
24.33
28.36
18.19
11.62
30.16
0.47
-0.13
-0.78
-0.61
-0.505
-0.46
δ18O PDB
-0.14
-0.9
δ18O PBD
0.545
0.41
0.46
0.9
0.6
-0.9
O. universa
P. wuellerstorfi
-1.13
-0.45
-0.55
-0.16
δ13C PBD
0.44
0.6
P. wuellerstorfi
-0.27
0.4
0.62
-1.74
0.84
-0.14
δ13C PBD
0.89
-0.54
O. universa
-0.504
-0.655
-0.562
-1.191
-0.321
-0.509
δ13C PBD
0.485
0.486
Average Water Column*
δ18O PBD
0.447
0.36
* Average water column values were calculated using a 1:9 ratio between surface and deep water reserviors, respectively.
Sample Name
TS1
TS2
TS3
TS4
TS5
TS6
TS7
TS8
TS9
TS10
TABLE 1. % CaCO3 AND OXYGEN AND CARBON ISOTOPE VALUES FOR BENTHIC AND
PLANKTONIC FORAMINIFERA FROM IL TRAVE SUD
21
Figure 8. Carbon and oxygen stable isotope values from silt in comparison with calcium
carbonate percentage (% CaCO3) from each meter level sampled for foraminifera stable
isotope analysis.
22
Carbon Isotopes
Carbon isotope ratios for O. universa ranged from –1.74‰ to 0.89‰ (Table 1).
δ13C for P. wuellerstorfi are generally negative, ranging from –1.13‰ to 0.6‰. Average
δ13C values for the entire water column for samples containing both planktonic and
benthic foraminifera are recorded in Table 1. Carbon isotope ratios for the silt-sized
samples are negative, ranging from –2.08‰ to –0.20‰. These values are out of phase in
comparison with the silt δ18O and CaCO3 content (Fig. 8).
Typical Pliocene Isotope Values from the Atlantic Ocean
Early Pliocene carbon and oxygen isotope values are comparable between Il
Trave Sud and the Atlantic Ocean. δ18O values reported from the North Atlantic Ocean
from benthic foraminifera range from 0.3 to 1.4‰, and δ13C values range from 0 to 1.1‰
(Venz et al., 1999).
δ18O and δ13C values from the North Atlantic Ocean are higher than δ18O and δ13C
values from Il Trave Sud. Lower δ18O values at Il Trave Sud could be due to postdepositional groundwater alteration or interaction. Higher δ18O values from the North
Atlantic can presumably be due to salinity differences between the North Atlantic Ocean
and the Mediterranean Sea.
DISCUSSION
δ13C and δ18O values and % CaCO3 are proxies used to determine changes in
global ice volume and evaporation-precipitation rates, temperature, fluvial input, salinity,
productivity, and ocean circulation patterns. Such conditions affect the isotopic
23
composition of both surface and deep water reservoirs, including the Mediterranean Sea.
In the following discussion, Early Pliocene Mediterranean Sea paleoceanographic
interpretations of stable isotope data from the two oceanic reservoirs and CaCO3 content
are examined and compared to the whole-rock model interpretations from Borkowski
(2002).
PALEOCEANOGRAPHIC RECONSTRUCTIONS
The foraminiferal isotope values that were analyzed in this study provide a more
precise and detailed interpretation of paleoceanographic conditions of the Mediterranean
Sea during the Early Pliocene. Three climate conditions were derived from carbon and
oxygen isotope data from benthic and planktonic foraminifera from Il Trave Sud. Two of
the climate conditions correspond with the warm, wet, estuarine and the cool, dry, antiestuarine climate systems proposed by Borkowski (2002). The third climate condition
represents a warm, wet climate with some oceanic mixing and stratification within the
water column, which was not established in the study conducted by Borkowski (2002).
Carbon and oxygen isotope values from silt-sized nannofossil samples correspond
with the same climate trends derived from the whole-rock stable isotope data studied by
Borkowski (2002). In both studies, CaCO3 content and δ18O values are in phase, while
δ13C values are out of phase (Fig. 5, 8). This relationship represents a correlation
between the climate systems determined by Borkowski (2002) and the climate conditions
in this study, as expected.
24
Warm, wet, estuarine climate (TS1)
P. wuellerstorfi
O. universa
P. wuellerstorfi
O. universa
% CaCO3
δ18O
δ18O
δ13C
δ13C
Avg. δ18O
0.447
0.545 ‰
-0.14
0.44
0.89
16.37
δ13C
0.485
Precipitation
(-8‰)
Surface Productivity
δ13C
Dilution
CaCO3
Salinity
δ18O
Stratified surface and deep waters
Organic and inorganic Salinity
carbon burial
δ18O
δ13C
River input
-nutrients
-δ13C (-5‰)
-δ18O (-10‰)
Figure 9. Schematic illustration of sample TS1, a warm, wet, estuarine climate, during the
Early Pliocene in the Mediteranean. Isotopic concentrations are recorded per mil (‰) PDB.
Warm, wet, estuarine climate (TS2)
P. wuellerstorfi
O. Universa
P. wuellerstorfi
O. Universa
% CaCO3
δ18O
δ18O
δ13C
δ13C
0.41 ‰
-0.09
0.60
-0.54
33.55
Avg. δ18O
0.36
δ13C
0.486
Precipitation
(-8‰)
Surface Productivity
δ13C
Salinity
δ18O
Stratified surface and deep waters
Organic and inorganic
Salinity
carbon burial
δ18O
δ13C
Dilution
CaCO3
River input
-nutrients
-δ13C (-5‰)
-δ18O (-10‰)
Figure 10. Schematic illustration of sample TS2, a warm, wet, estuarine climate, during
the Early Pliocene in the Mediteranean. A larger amount of δ13C depleted fluvial input
results in low δ13C values of surface waters and a higher % CaCO3. Isotopic concentrations
are recorded per mil (‰) PDB.
25
Cool, dry, anti-estuarine climate (TS4)
Avg. δ18O δ13C
-0.655 -0.321
Evaporation
Surface Productivity
δ13C
-0.78 ‰
0.47
-0.45
0.84
36.47
δ18O
δ18O
δ13C
δ13C
P. wuellerstorfi
O. Universa
P. wuellerstorfi
O. Universa
% CaCO3
Dilution
CaCO3
Salinity
δ18O
River input
-nutrients
-δ13C
-δ18O
Mixing of surface
and deep waters
δ13C
δ18O
Figure 11. Schematic illustration of sample TS4, a cool, dry, anti-estuarine climate, during
the Early Pliocene in the Mediteranean. Isotopic concentrations are recorded per mil (‰)
PDB.
Warm, wet, and some ocean mixing climate (TS5 and TS10)
P. wuellerstorfi
O. Universa
P. wuellerstorfi
O. Universa
% CaCO3
δ18O
δ18O
δ13C
δ13C
TS5
-0.61
-0.13
-0.55
-0.14
44.03
TS10
-0.46 ‰
-0.90
-1.13
-1.74
30.16
Avg. δ18O δ13C
TS5 -0.562 -0.509
TS10 -.504 -1.191
Surface Productivity
δ13C
Salinity
δ18O
Dilution
CaCO3
Some mixing and stratification
of surface and deep waters
δ13C
Precipitation
(-8‰)
River input
-nutrients
-δ13C (-5‰)
-δ18O (-10‰)
δ18O
Figure 12. Schematic illustration of sample TS5 and TS10, warm, wet climate with some
ocean circulation, during the Early Pliocene in the Mediteranean. Low surface δ13C due
to large input of carbon-rich freshwater. High productivity obscures dilution effects
resulting in high CaCO3 content. Ocean circulation results in similar δ13C and δ18O in
surface and deep water reservoirs. Isotopic concentrations are recorded per mil (‰) PDB.
26
Warm, wet, estuarine climate
Oxygen and carbon isotope values and CaCO3 content of samples TS1 and TS2
represent similar paleoenvironmental conditions (Fig. 9, 10). Both samples record high
deep water δ18O and δ13C values and low surface δ18O values. High surface δ13C and low
CaCO3 content of sample TS1, however, are opposite those values in TS2. Overall, TS1
and TS2 represent a wet, warm, estuarine climate with a stratified water column.
The δ18O surface water values for both samples are lower than the average δ18O
water column values, while deep water δ18O values are higher. This difference between
surface and deep water δ18O values represents a stratified water column. Stratification is
a result of density differences in the amount of freshwater input between surface and deep
water reservoirs. Freshwater input decreases the salinity in the surface waters, lowering
δ18O values, and increases the salinity in the deep waters, raising δ18O values.
The surface δ13C of sample TS1 is higher than the average δ13C water column
composition, but surface δ13C is abnormally negative for TS2, lower than the average
δ13C. High δ13C values in surface waters are expected as a result of marine productivity
as shown in TS1. The unusually low δ13C value for TS2 can be attributed to a greater
influx of negative δ13C freshwater at this time.
Also unlike TS1, TS2 has a high CaCO3 content, which contradicts the low
surface δ13C value and previous interpretation of low CaCO3 content due to strong
dilution effects. This phenomenon can be attributed to a high productivity rate of surface
dwelling organisms, which obscures the effects of dilution, resulting in high CaCO3
values.
27
Stratification of the water column minimizes communication between surface and
deep water reservoirs. The lack of circulation between the two reservoirs can lead to
anoxic bottom water conditions. Absence of oxygen in deep waters can inhibit the
decomposition of falling organic matter. Decompostion converts organic matter to
inorganic carbon by microbial respiration in the deep water, lowering δ13C of deep water
reservoirs. Since decomposition is prohibited due to low oxygen levels in TS1 and TS2,
contain high δ13C deep water compositions.
Previous studies from Pliocene Mediterranean marine strata have found an
association between warm, wet climate conditions, and cool, dry climate conditions
(Filippelli, 1997; Foucault et al., 2000; Van Os et al., 1994). These oscillating climate
cycles are determined as a result of changes in African and Asian monsoon intensities,
which strongly influence the climate conditions of the Mediterranean.
Carbon and oxygen isotope data from TS1 and TS2 reflect increases in freshwater
input, thus representing wet climate conditions within the Mediterranean. As mentioned
above, wet climate conditions can be associated with warmer temperatures from monsoon
influences. Increases in freshwater input also results in stratification between surface and
deep water reservoirs. This interpreted warm, wet climate with a stratified water column
supports the warm, wet, estuarine climate proposed by Borkowski (2002).
Cool, dry, anti-estuarine climate
TS4 contains high δ18O and δ13C surface water values, low δ18O and δ13C deep
water values, and a high CaCO3 content (Fig. 11). This isotopic and CaCO3 trend
represents a cool, dry, anti-estuarine climate system.
28
The surface δ18O value is significantly higher than the water column average, and
the deep water δ18O value is slightly lower than the water column average. High δ18O
surface water is due to an increase in evaporation rates and a decrease in low δ18O
freshwater input. This increase in surface δ18O also increases the salinity, and therefore
density, of the surface waters. This high density of surface waters drives vertical mixing
of the water column.
In a mixed water column, δ18O surface and deep water values are expected to be
similar. Deep water δ18O in TS4, however, is considerably lower than surface water δ18O.
This could be due to a mass flow of lower δ18O concentrated water from the Atlantic
Ocean into the Mediterranean basin across the Strait of Gibraltar. Or, a strong
evaporation rate may increase surface δ18O faster than vertical mixing can respond,
creating δ18O differences between the two reservoirs.
Although a decrease in freshwater input lowers the amount of nutrient and low
δ13C concentrated input, surface waters in TS4 record high δ13C values. However,
vertical mixing of the water column recycles carbon-rich matter from deep to surface
waters. This transportation of low δ13C material to surface waters results in high surface
δ13C values.
The low deep water δ13C value in TS4 reflects the decomposition of organic
matter in deep waters, which converts organic matter to inorganic matter, lowering the
δ13C value. Decomposition indicates a presence of oxygen in the bottom waters,
signifying circulation between surface and deep water reservoirs.
The high CaCO3 content of TS4 reflects the reduction of freshwater input into the
surface waters. A decrease in freshwater input from river systems and precipitation
29
reduces the dilution effects of the surface waters. The high CaCO3 content thus reflects
the surface productivity and carbonate deposition in TS4.
The carbon and oxygen isotope data from TS4 represent drier climate conditions
than TS1 and TS2. This dry climate can also be associated with cooler temperatures from
monsoon influences within the Mediterranean as previously mentioned. Decreases in
deep water δ13C values and increases in surface water δ18O values support circulation
between the two reservoirs, resulting in a mixed water column. This cool, dry climate
system with a mixed water column supports the cool, dry, anti-estuarine climate system
proposed by Borkowski (2002).
Wet, warm climate with some ocean mixing
Samples TS5 and TS10 have similar CaCO3 content and carbon and oxygen
isotopic trends (Fig. 12). Both samples are low in δ13C and δ18O surface and deep water
values, and are high in CaCO3 content. Samples TS5 and TS10 represent wet, warm
climates with some ocean mixing and stratification.
Low surface δ18O values in TS5 and TS10 reflect increases in low δ18O
concentrated freshwater input from fluvial systems and precipitation into surface waters.
An increase in freshwater input also explains the low surface δ13C values of TS5 and
TS10. These low δ13C values can be attributed to highly negative δ13C freshwater
flowing into the surface waters, decreasing surface δ13C values.
Deep water δ18O and δ13C values are also low. Similarities in carbon and oxygen
isotopic compositions of surface and deep water reservoirs are indicative of vertical
mixing within the water column. Deep water δ18O values are low due to the
30
transportation of low δ18O from the surface waters. Deep water δ13C values are low also
from vertical mixing and decomposition of organic matter.
The high CaCO3 content of TS5 and TS10 are similar to the interpretation of high
CaCO3 in TS2. Since there is an increase in freshwater input, surface waters should
become diluted, lowering the CaCO3 content. However, the surface production rate
obscures the dilution effects, resulting in a high CaCO3 content in TS5 and TS10.
In comparison to average water column δ18O and δ13C values of TS5 and TS10,
surface and deep water δ18O and δ13C values fluctuate due to changes in mixing rates,
intensity of freshwater input, and amount of carbonate transported between reservoirs.
The freshwater input indicates wet climate conditions, along with warmer temperatures
due to the influence of monsoons. Similar carbon and oxygen isotopic values in surface
and deep water reservoirs represent some vertical mixing occurring within the water
column. From carbon and oxygen isotope data, TS5 and TS10 represent a wet, warm
climate with both ocean stratification and vertical mixing.
The paleoenvironments derived from stable isotope analysis and CaCO3 content
in this study are interpreted as responses to African and Asian monsoon intensities during
the Early Pliocene within the Mediterranean Sea. From the limited data and lack of
stratigraphic continuity of the samples analyzed in this study, no correlations with the
precessional, obliquity, or eccentricity cycles can be made. Sample analysis from
continous intervals throughout Il Trave Sud would supply sufficient information for
correlating the strata with the Milankovitch cycles determined by Borkowski (2002).
31
CONCLUSIONS
Carbon and oxygen isotope analysis of benthic and planktonic foraminifera act as
fairly informative paleoceanographic proxies. Stable isotope analysis from O. universa
and P. wuellerstorfi from Il Trave Sud represents a more detailed interpretation of Early
Pliocene climate variations than the whole-rock stable isotope analysis of Il Trave Sud
conducted by Borkowski (2002).
Samples TS1 and TS2 contain low δ13O surface waters and high deep water δ13O
and δ13C values. TS1 has a low CaCO3 content with a high surface δ13C value. TS2 has a
high CaCO3 content with a low surface δ13C value. Sample TS4 consists of high surface
water δ18O and δ13C values, high CaCO3 content, and low deep water δ18O and δ13C
values. Samples TS5 and TS10 contain low surface and deep water δ18O and δ13C values
and high CaCO3 content.
Samples TS1 and TS2 are interpreted as wet, warm climates with stratified water
columns. Oceanic conditions of wet, warm climates generally contain low surface δ13O
and CaCO3 content due to dilution effects from low δ13O freshwater input, and high
surface δ13C from an increase in input of low δ13C concentrated freshwater and nutrients,
which is used during primary production. Deep waters contain high δ13O as a result of
higher salinity levels due to stratification of the water column. Deep waters also contain
high δ13C due to anoxic water conditions preventing organic carbon decomposition.
Sample TS4 is interpreted as a cool, dry climate with a vertically mixed water
column. A decrease freshwater input due to an increase in evaporation rates is reflected
in the low surface δ18O value and high CaCO3 content. A high surface δ13C value is a
result of upwelling and mixing of the water column transporting carbon-rich matter.
32
Deep water δ13O and δ13C values are low due to ocean circulation patterns mixing the
different isotopic compositions of the two oceanic reservoirs and decomposition of the
transported organic carbon matter.
Samples TS5 and TS10 are though to represent a warm, wet climate with some
ocean circulation and stratification. Low surface δ18O and δ13C values reflect an increase
in freshwater input with low δ18O and δ13C concentrations. Low δ18O and δ13C deep
water values similar to surface water isotope values indicates vertical mixing of the water
column. A high CaCO3 represents a high surface production rate due to an increase in
carbon-rich material from freshwater input and the recycling of high δ13C deep
water.(Broecker et al., 1982)
The wet, warm, estuarine climates and the cool, dry, anti-estuarine climates
proposed by Borkowski (2002) are comparable to paleoenvironments interpreted from
stable isotope analysis of benthic and planktonic foraminifera from Il Trave Sud. The
wet, warm climate with ocean stratification represented in samples TS1 and TS2
corresponds with the wet, warm, estuarine paleoenvironment proposed by Borkowski
(2002). Sample TS4 represents a cool, dry climate with vertical mixing ocean circulation
patterns, which is consistent with the cool, dry, anti-estuarine climate system established
by Borkowski (2002). Carbon and oxygen isotope data from silt-sized particles in this
study show a strong relationship with the whole-rock stable isotope model by Borkowski
(2002).
The wet, warm climate system with both ocean stratification and vertical mixing
proposed in this study from samples TS5 and TS10, however, does not correlate with
interpretations of whole-rock carbonate examined by Borkowski (2002). This lack of a
33
correlation exhibits how planktonic and benthic foraminfera provide more precise data
compared to data from whole-rock.
34
ACKNOWLEDGEMENTS
I would like to thank the Bernstein Student Research Endowment for their support
in making this research project possible. Many thanks go to Dave Bice for his advising
and assistance in the field; Sandro Montanari and the Coldigioco folk for the use of
laboratory equipment, hospitality, and good times; Dan Shrag at Harvard for conducting
isotope analysis on my samples; Adrienne Hacker and Isaac Larsen for editing my drafts;
and my friends and family for their advice and support throughout the duration of this
project.
Figure 13. Visual aid used for foraminifera identification.
35
REFERENCES CITED
Arthur, M. A., and Dean, W. E., 1991, A Holistic Geochemical Approach to Cyclomania:
Examples from Cretaceous Pelagic Limestone Sequences, in Einsele, G., Ricken,
W., and Seilacher, A., eds., Cycles and Events in Stratigraphy: Berlin, SpringerVerlag, p. 138-139.
Beyer, E., 1999, Using Foraminiferal Stable Isotopes to Reconstruct Paleoceanographic
Changes: a study of the Middle to Late Miocene Mediterranean, Monte dei
Corvi, Italy [Senior Integrative Exercise]: Northfield, Minnesota, Carleton
College, p. 1-68.
Billups, K., Ravelo, A. C., and Zachos, J. C., 1997, Early Pliocene Deep-Water
Circulation: Stable Isotope Evidence for Enhanced Northern Component Deep
Water Formation: Proc. ODP, Sci. Results, v. 154, p. 319-330.
-, 1998, Early Pliocene climate: A perspective from the western equatorial Atlantic warm
pool: Paleoceanography, v. 13, no. 5, p. 459-470.
Borkowski, H., 2002, Early Pliocene Climate in the Mediterranean: Geochemical
Analysis at Trave Sud, Italy [Senior Integrative Exercise]: Northfield, Minnesota,
Carleton College, p. 1-46.
Broecker, W. S., and Peng, T.-H., 1982, Tracers in the Sea: Palisades, NY, Eldigio, 689
p.
Crowley, T. J., 1991, Modeling Pliocene warmth: Quaternary Science Reviews., v. 10, p.
275-282.
de Boer, P. L., 1991, Pelagic Black Shale-Carbonate Rhythms: Orbital Forcings and
Oceanographic Response, in Einsele, G., Ricken, W., and Seilacher, A., eds.,
Cycles and Events in Stratigraphy: Berlin, Springer-Verlag, p. 68-69.
Derry, L. A., and France-Lanord, C., 1996, Neogene growth of the sedimentary organic
carbon reservoir: Paleoceanography, v. 11, no. 3, p. 267-275.
Einsele, G., and Ricken, W., 1991, Limestone-Marl Alternation- an Overview, in Einsele,
G., Ricken, W., and Seilacher, A., eds., Cycles and Events in Stratigraphy: Berlin,
Springer-Verlag, p. 23-47.
Emiliani, C., 1971, Depth habitats of Growth Stages of Pelagic Foraminifera: Science, v.
173, p. 1122-1124.
Epstein, S., Buchsbaum, R., Lowenstam, H. A., and Urey, H. C., 1953, Revised
carbonate-water isotopic termperature scale: Geological Society of America
Bulletin, v. 64, p. 1315-1325.
Fauquette, S., Guiot, J., and Suc, J., 1998, A method for climatic reconstruction of the
Mediterranean Pliocene using pollen data: Palaeogeography, Palaeoclimatology,
Palaeoecology, v. 144, p. 183-201.
Faure, G., 1977, Principles of Isotope Geology: New York, John Wiley & Sons, 464 p.
Filippelli, G. M., 1997, Intensification of the Asian monsoon and a chemical weathering
event in the late Miocene-early Pliocene: Implications for late Neogene climate
change: Geology, v. 25, no. 1, p. 27-30.
Foucault, A., and Mélieres, F., 2000, Palaeoclimatic cyclicity in central Mediterranean
Pliocene sediments: the mineralogical signal: Palaeogeography,
Palaeoclimatology, Palaeoecology, v. 158, p. 311-323.
36
Haywood, A. M., Sellwood, B. W., and Valdes, P. J., 2000, Regional warming: Pliocene
(3 Ma) paleoclimate of Europe and the Mediterranean: Geology, v. 28, no. 12, p.
1063-1066.
Hermoyian, C. S., and Owen, R. M., 2001, Late Miocene-early Pliocene biogenic bloom:
Evidence from low-productivity regions of the Indian and Atlantic Oceans:
Paleoceanography, v. 16, no. 1, p. 95-100.
Hodell, D. A., Elmstrom, K. M., and Kennett, J. P., 1986, Latest Miocene benthic d18O
changes, global ice volume, sea level, and the 'Messinian salinity crisis': Nature,
v. 320, p. 411-414.
Hsü, K. J., 1983, The Mediterranean was a desert: a voyage of the Glomar Challenger:
Princeton, Princeton University Press, 197 p.
Kennett, J. P., and Hodell, D.A., 1993, Evidence for relative climatic stability of
Antarctica during the Early Pliocene: A marine perspective: Geografiska
Annaler, v. 75 A, no. 4, p. 205-220.
Kennett, J. P., and Hodell, D. A., 1993, Evidence for relative climate stability of
Antarctica during the Early Pliocene: A marine perspective: Geografiska
Annaler, v. 75 A, no. 4, p. 205-220.
Kim, S. T., and O'Neil, J. R., 1997, Equilibrium and nonequilibrium oxygen isotope
effects in synthetic calcites: Geochemica et Cosmochimica Acta, v. 61, p. 34613475.
Kump, L. R., 1991, Interpreting carbon-isotope excursions: Strangelove oceans:
Geology, v. 19, p. 299-302.
Loutit, T. S., and Kennett, J. P., 1979, Application of Carbon Isotope Stratigraphy to Late
Miocene Shallow Marine Sediments, New Zealand: Science, v. 204, p. 11961199.
Mackensen, A., Hubberten, H.-W., Bickert, T., Fischer, G., and Fütterer, D. K., 1993,
The d13C in benthic foraminiferal tests of Fontbotia wuellerstorfi (Schwager)
relative to the d13C of dissolved inorganic carbon in southern ocean deep water:
Implications for glacial ocean circulation models: Paleoceanography, v. 8, p. 587610.
Millero, F. J., 1996, Chemical Oceanography: Boca Raton, CRC Press, Inc., 469 p.
Paul, A., Mulitza, S., Pätzold, J., and Wolff, T., 1999, Simulation of Oxygen Isotopes in a
Global Ocean Model, in Fischer, G., and Wefer, G., eds., Use of Proxies in
Paleoceanography: Berlin, Springer-Verlag, p. 655-686.
Spero, H. J., Bijma, J., Lea, D. W., and Bemis, B. E., 1997, Effect of seawater carbonate
concentration on foraminiferal carbon and oxygen isotopes: Nature, v. 390, p.
497-500.
Thunell, R. C., 1979, Pliocene-Pleistocene paleotemperature and paleosalinity history of
the Mediterranean Sea: Results from DSDP sites 125 and 132: Marine
Micropaleontology, v. 4, p. 173-187.
Van Os, B. J. H., Lourens, L. J., Hilgen, F. J., and De Lange, G. J., 1994, The Formation
of Pliocene sapropels and carbonate cycles in the Mediterranean: Diagenesis,
dilution, and productivity: Paleoceanography, v. 9, no. 4, p. 601-617.
Venz, K. A., Hodell, D. A., Stanton, C., and Warnke, D. A., 1999, A 1.0 Myr record of
Glacial North Atlantic Intermediate Water variability from ODP Site 982 in the
northeast Atlantic: Paleoceanography, v. 14, p. 42-52.
37
Vidal, L., Bickert, T., Wefer, G., and Röhl, U., 2002, Late Miocene stable isotope
stratigraphy of SE Atlantic ODP Site 1085: Relation to Messinian events: Marine
Geology, v. 180, p. 71-85.
Wefer, G., Berger, W. H., Bijma, J., and Fischer, G., 1999, Clues to Ocean History: a
Brief Overview of Proxies, in Fischer, G., and Wefer, G., eds., Use of Proxies in
Paleoceanography: Berlin, Springer-Verlag, p. 1-68.
Williams, D. F., Thunell, R. C., and Kennett, J. P., 1978, Periodic freshwater flooding
and stagnation of eastern Mediterranean Sea during Late Quaternary: Science, v.
201, p. 252-254.
Zachos, J. C., Lohmann, K.C., Walker, J.C., and Wise, S.W., 1993, Abrupt Climate
Change and Transient Climates During the Paleogene: A Marine Perspective:
The Journal of Geology, v. 101, p. 191-213.
Zachos, J. C., Pagani, M., Sloan, L., Thomas, E., and Billups, K., 2001, Trends, Rhythms,
and Aberrations in Global Climate 65 Ma to Present: Science, v. 292, p. 686-693.