University of Southern California Dissertations and Theses

Sulfur isotope geochemistry and the end Permian mass extinction

(USC Thesis Other)

Sulfur isotope geochemistry and the end Permian mass extinction

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content SULFUR ISOTOPE GEOCHEMISTRY AND THE END PERMIAN MASS
EXTINCTION

by
Pedro Jose Marenco
__

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(GEOLOGICAL SCIENCES)

August 2007

Copyright 2007 Pedro Jose Marenco
ii
ACKNOWLEDGEMENTS

I would like to extend my immeasurable gratitude to all of those included
within my family and friends that have supported me emotionally and financially in
my educational endeavors throughout my long educational journey. In particular I
thank my father, Pedro Jose Marenco Sr., who sacrificed much to support my
education, even when in my youth, I failed to appreciate it. In more recent years, my
wife, Katherine N. Marenco, has supported my growth as a person and has been a
catalyst for my intellectual creativity. I also thank my brother Paul Marenco for his
laborious efforts as my tireless field assistant.
My accomplishments during higher education, including this dissertation,
could not have been possible without a strong network of support within the USC
Department of Earth Sciences. My fellow graduate students within the USC Sed and
Paleo Labs have helped me in matters both academic and personal, exhibiting a
mutual understanding of the hardships of graduate student life. The administrative
staff of the department, especially John McRaney, Cindy Waite, and Vardui Ter-
Simonian, has helped me fight the logistical forces in life that serve to retard the
progress of education and research. I have been extremely fortunate to have
interacted with a number of Earth Sciences faculty, at USC and elsewhere, who have
engaged me in hours of conversation and have helped me grow as a scientist. In
iii
particular, Doug Hammond of USC and Isabel Montanez of UC Davis have helped
me develop a critical mind for evaluating data.
Financial support for the research presented in this dissertation was granted to
me by a number of agencies. I thank all of those who manage and contribute to
graduate student research funds, especially the USC Department of Earth Sciences,
the Geological Society of America, the Paleontological Society, the American
Association of Petroleum Geologists and the National Science Foundation.
Lastly, I thank my two dissertation advisors, David Bottjer and Frank
Corsetti, without whom I would be a much different scientist, and a much different
person. Their time, effort and encouragement were invaluable to my academic and
personal success.
iv
TABLE OF CONTENTS

Acknowledgements ii
List of Tables v
List of Figures vi
Abstract ix
Chapter 1: Introduction 1
Chapter 2: Environmental and diagenetic variations in carbonate associated
sulfate: an investigation of cas in the lower triassic of the western United
States 24

Chapter 3: Oxidation of pyrite during extraction of carbonate associated sulfate 52
Chapter 4: Combined Permo-Triassic sulfur and carbon isotope variations and
their implications for the End-Permian mass extinction 64

Chapter 5: Early Triassic sulfur isotopes from the western United States 92
Chapter 6: Conclusions 112
Bibliography 116
Appendix: Localities, measured sections and data 126

v
LIST OF TABLES

Table 4-1: Model parameters and initial values 83
Table A-1: Stable isotopic analyses used in this work 195
Table A-2: Trace elemental and petrographic analyses used in this work 201
vi
LIST OF FIGURES

Figure 1-1: Generalized sulfur cycle diagram showing the isotopic 4
composition of the different reservoirs of sulfur as well as the fluxes
between reservoirs

Figure 1-2: CAS and evaporite δ
34
S curve for the Phanerozoic 11

Figure 1-3: Early Triassic time-scale and Western U.S. Lower Triassic 14
formations studied in this work

Figure 1-4: The Early Triassic world 15

Figure 1-5: Locality map showing the locations of the Western U.S. sections 16
Studied

Figure 2-1: Map showing the localities used in this study 28

Figure 2-2: Spathian δ
34
S values 29

Figure 2-3: Early Triassic strontium isotopes 31

Figure 2-4: Early Triassic carbon isotopes 32

Figure 2-5: Chemostratigraphy from Beyond Lost Cabin 33

Figure 2-6: Chemostratigraphy from Blue Diamond 34

Figure 2-7: Chemostratigraphy from Rainbow Gardens 35

Figure 2-8: Stable isotope and trace element cross plots versus Ca/Mg 38

Figure 2-9: δ
34
S of evaporites and carbonates vs.
87
Sr/
86
Sr 40

Figure 2-10: Model results of Rayleigh-type
34
S depletion in a hypothetical brine 47
and precipitate as evaporation progresses

Figure 3-1: Plot of sulfate concentration from CAS extraction versus weight 55
percent pyrite

vii
Figure 3-2: Plot of δ
34
S results from CAS extraction versus weight percent 56
Pyrite

Figure 3-3: Plot of modeled [CAS]
app
values versus the M
pyrite
/M
lime
values 58
from this study

Figure 3-4: Plot of modeled δ
34
S
app
values versus [SO
4
]
pyrite
/[CAS]
app
values 60
from this study

Figure 3-5: Results of modeling using the slope of the line in Figure 3-3 with 62
a hypothetical [SO
4
]
lime
of 300 ppm and different δ
34
S
pyrite
values

Figure 4-1: Map showing the location of Çürük Da ğ 66

Figure 4-2: Chemostratigraphy of the Çürük Da ğ P-T boundary section 70

Figure 4-3: δ
34
S of carbonate associated sulfate and δ
13
C
bulk
of carbonate versus 74
height at Çürük Da ğ

Figure 4-4: Cross plot of δ
34
S of carbonate associated sulfate and δ
13
C
bulk
of 76
Carbonate

Figure 4-5: Model results with conditions that are explained in the text 83

Figure 5-1: Sulfur isotope chemostratigraphy and generalized lithology from 97
the Batten and Stokes, San Rafael Swell, Utah locality

Figure 5-2: Sulfur isotope chemostratigraphy and generalized lithology from 98
the Beyond Lost Cabin, Nevada locality

Figure 5-3: Sulfur isotope chemostratigraphy and generalized lithology from 99
the Blacktail Creek, Montana locality

Figure 5-4: Sulfur isotope chemostratigraphy and generalized lithology from 100
the Blue Diamond, Nevada locality

Figure 5-5: Sulfur isotope chemostratigraphy and generalized lithology from 101
the Cascade Springs, Utah locality

Figure 5-6: Sulfur isotope chemostratigraphy and generalized lithology from 102
the Grasshopper Creek, Montana locality

viii
Figure 5-7: Sulfur isotope chemostratigraphy and generalized lithology from 103
the Hidden Pasture, Montana locality

Figure 5-8: Sulfur isotope chemostratigraphy and generalized lithology from 104
the Rainbow Gardens, Nevada locality

Figure 5-9: Sulfur isotope chemostratigraphy and generalized lithology from 105
the Road Cut, San Rafael Swell, Utah locality

Figure 5-10: Early Triassic sulfur isotopic results separated by stage. 107

Figure A-1: Index of lithologic symbols used in this text 127

Figure A-2: Stratigraphic column for the Batten and Stokes, San Rafael Swell, 128
Utah locality, and the Road Cut, San Rafael Swell, Utah locality

Figure A-3: Stratigraphic column for the Beyond Lost Cabin, Nevada locality 131

Figure A-4: Stratigraphic column for the Blacktail Creek, Montana locality 137

Figure A-5: Stratigraphic column for the Blue Diamond, Nevada locality 141

Figure A-6: Stratigraphic column for the Cascade Springs, Utah locality 143

Figure A-7: Stratigraphic column for the Darwin Hills, California locality 146

Figure A-8: Stratigraphic column for the Grasshopper Creek, Montana locality 156

Figure A-9: Stratigraphic column for the Hidden Pasture, Montana locality 160

Figure A-10: Stratigraphic column for the Rainbow Gardens, Nevada locality 168

Figure A-11: Stratigraphic column for the Union Wash, California locality 172

ix
ABSTRACT

The use of carbonate associated sulfate (CAS) to study sulfur isotope
chemostratigraphy is investigated in detail. Results suggest that middle-shelf
limestones are better suited for sulfur isotopic analysis than proximal evaporites or
dolostones because of possible facies-related factors that preclude the latter phases
from recording true seawater sulfate δ
34
S values. Carbonate samples with pyrite
should be avoided because of evidence of pyrite oxidation during the CAS extraction
process.
Coupled sulfur and carbon isotopic fluctuations associated with the end
Permian mass extinction in Turkey suggest that the mass extinction was caused by
H
2
S and CO
2
resulting from shallow-water euxinia following a prolonged period of
deep ocean anoxia in the Late Permian. Extensive deposition of evaporites during
the Permian may have contributed to the global anoxia by burying the oxidized form
of sulfur as sulfate. Extreme volcanism from the Siberian Traps may have expanded
the deep-ocean euxinia into the shallow ocean, where toxic levels of H
2
S and CO
2

may have been introduced to the atmosphere.
Following the mass extinction, elevated δ
34
S values throughout the entire
Early Triassic suggest that anoxia continued for at least five million years after the
end Permian mass extinction. However, the relegation of lithologic features
characteristic of anoxia to deeper-water environments of deposition suggest that the
x
anoxia was once again restricted to the deep ocean following the end Permian mass
extinction.
1
CHAPTER 1: INTRODUCTION
The End-Permian mass extinction
At the end of the Permian Period, an estimated 80-96% of Earth’s marine
species (Raup, 1979; Stanley and Yang, 1994), as well as a significant proportion of
the terrestrial biota (Retallack, 1995), disappeared from the fossil record. The End
Permian mass extinction was unusual in that it took until the end of the Early
Triassic (Erwin, 1993), an interval of about 5 m.y. (Mundil et al., 2004; Lehrmann et
al., 2006), for life to recover. Proposed extinction mechanisms include bolide impact
(Becker et al., 2001; Becker et al., 2004), CO
2
poisoning due to the overturn of a
stratified ocean (Knoll et al., 1996), widespread marine and atmospheric anoxia
(Wignall and Hallam, 1992; Isozaki, 1997; Huey and Ward, 2005), a global salinity
crisis (Fischer, 1964), prolonged release of methane (Krull and Retallack, 2000),
extreme volcanism from the Siberian traps and related climate change (Renne et al.,
1995; Kamo et al., 2003), and H
2
S poisoning due to oceanic euxinia (Nielsen and
Shen, 2004; Grice et al., 2005; Kump et al., 2005; Marenco et al., In Prep.).
The evidence for bolide impact at the P-T transition has been widely
questioned (Farley and Mukhopadhyay, 2001; Isozaki, 2001; Glickson, 2004; Renne
et al., 2004; Wignall et al., 2004) and has yet to be reproduced (Dalton, 2004).
However, a growing body of evidence provides support for anoxia-related extinction
mechanisms. Sedimentological evidence, such as the prevalence of black shales and
black cherts, the absence of bioturbation, and the prevalence of small framboidal
2
pyrite (Wignall and Twitchett, 1996; Isozaki, 1997; Wignall and Newton, 2003)
suggests the presence of widespread deep ocean anoxia from the Late Permian
through the Early Triassic (Isozaki, 1997) with shallow marine anoxia beginning in
the latest Permian and lasting through the first stage of the Early Triassic (Wignall
and Twitchett, 1996). Geochemical evidence for a prolonged biotic crisis is found in
the P-T carbon isotopic record, which reveals a strong negative excursion at the mass
extinction level (a decrease of up to ~5‰ (Corsetti et al., 2005)), as well as multiple
and large δ
13
C excursions that punctuate the Early Triassic (Payne et al., 2004).

Sulfur isotopes and the Permo-Triassic transition
Sulfur isotope systematics
There are four stable isotopes of sulfur:
32
S (95.02% abundance),
33
S (0.75%
abundance),
34
S (4.21% abundance) and
36
S (0.02% abundance). The most widely
studied sulfur isotopes are the two most abundant,
32
S and
34
S, which are reported as
the ratio of
34
S to
32
S relative the Canyon Diablo Troilite (CDT) using the ‰
(permil) notation where:
()
1000
Ratio
Ratio Ratio
‰
(CDT)
(CDT) Sample
×
−
= δ
Most labs today report δ
34
S values relative to the international Vienna Canyon
Diablo Troilite (VCDT) standard, which is an isotopic replica of the original Canyon
Diablo troilite.
3
The ocean and sedimentary rocks comprise the largest reservoirs of sulfur
(Fig. 1-1). Although the sedimentary sulfur reservoir is much larger than the oceans
(about 300 X 10
18
mols of sulfur in sediments versus 40 X 10
18
mols in the ocean)
(Holser et al., 1988), the flux of sulfur from the sedimentary reservoir (via rivers) to
the oceans is seven orders of magnitude smaller than the number of mols of sulfur in
the ocean. Consequently, the residence time of sulfur in the oceans today is on the
order of 10
7
years.
The largest fractionation in the sulfur cycle occurs during the production of
pyrite via bacterial sulfate reduction under anaeroibic conditions. The observed
fractionation between sulfide and sulfate during bacterial sulfate reduction varies
from -4 to -46‰ although much larger fractionations have been observed between
sedimentary sulfides and sulfates (e.g., Canfield and Thamdrup, 1994). There is a
fractionation of about +2‰ during the precipitation of gypsum. Because the
fractionation during bacterial sulfate reduction dwarfs all other known fractionations
of sulfur, the formation and burial of
34
S-depleted pyrite relative to the burial of
sulfate evaporites is largely considered to be the dominant control on the δ
34
S
composition of seawater through time (e.g., Strauss, 1997).

Sulfur isotopes and mass extinctions
The sulfur isotopic composition of oceanic sulfate provides an important test
for extinction mechanisms that involve changes in global redox conditions. The δ
34
S
composition of sulfate in the oceans is largely controlled by a) the amount of sulfate
4

Figure 1-1: Generalized sulfur cycle diagram showing the isotopic composition of
the different reservoirs of sulfur as well as the fluxes between reservoirs. The
flux of sulfur to the carbonates reservoir is estimated by taking the flux of
carbonate deposition (51 X 10
12
mols/year) and assuming that carbonate
sediments will have from 200 to 5000 ppm of carbonate associated sulfate.
The average isotopic composition of the carbonate rock reservoir is not yet
known. As a first approximation I estimate that the average δ
34
S of the
carbonate reservoir is between the average of the evaporite reservoir (+16‰)
and the current ocean δ
34
S value (+20‰). (Data taken from Chapter 4;
Claypool et al., 1980; Holser et al., 1988; Canfield and Thamdrup, 1994;
Marenco et al., In Prep.)
5
in the ocean and b) the fraction of sulfur buried as pyrite resulting from anaerobic
bacterial sulfate reduction, which preferentially removes the lighter isotope
32
S,
leaving oceanic sulfate enriched in
34
S. In a well-oxygenated ocean (such as the
present ocean), bacterial sulfate reduction occurs within ocean sediments, where
oxygen concentrations decrease to zero(Berner and Raiswell, 1983). However, in a
poorly oxygenated ocean, the overall mass of SO
4
reduced by anaerobes would
increase as the anoxic zone expands into the water column, thus having a stronger
effect on the δ
34
S composition of oceanic sulfate, resulting in the enrichment of
34
S
in residual sulfate. Because anaerobes use organic substrates to reduce sulfate,
changes in redox conditions affect both the δ
34
S of sulfate and the δ
13
C of dissolved
carbon. Oxidation of organic matter during bacterial sulfate reduction results in
increased input of
13
C-depleted CO
2
into the TCO
2
pool while
34
S-enriched SO
4

remains in the sulfate pool. Consequently, changes in δ
34
S due to changing redox
conditions should show a negatively correlated response in δ
13
C if these changes
occur over time scales that are short compared to the oceanic residence times of
carbon (~100 k.y. for the modern ocean) and sulfur (~10 m.y.)
Today’s oceans are extremely resistant to sulfur isotopic change because of
the long residence time of sulfur. Because of the size of the oceanic sulfate
reservoir, changes in sulfur input from weathering and volcanism have little effect on
the isotopic composition of sulfate. However, there is some evidence that the oceanic
sulfate reservoir may have been much smaller in the geologic past (Chapters 2 and 4;
Canfield et al., 2000; Horita et al., 2002; Hurtgen et al., 2002; Kah et al., 2004;
6
Marenco et al., In Prep.; Marenco et al., In Review). If ancient oceans contained less
sulfate than the modern ocean, then it is likely that the sulfur isotopic composition of
ancient oceans was susceptible to changes in weathering rates and styles, as well as
significant episodes of volcanism. For example, the Permian was a time of extensive
evaporite deposition (e.g., Fischer, 1964; Zharkov, 1981). It is likely that the
increased and prolonged deposition and removal of the heavier sulfur isotope during
evaporite deposition played a role in the significant lowering of oceanic sulfate δ
34
S
towards the end of the Paleozoic (Claypool et al., 1980; Holser, 1984; Kampschulte
and Strauss, 2004). The increased burial of sulfate during the Permian may have led
to lower sulfate concentrations that allowed for the extreme sulfur isotopic
fluctuations discussed in Chapter 4 and elsewhere (Newton et al., 2004; Riccardi et
al., 2006; Marenco et al., In Prep.) Furthermore, the extensive removal of the
oxidized form of sulfur may have set the stage for the geochemical and biological
events at the end of the Permian by burying oxygen in evaporite deposits.
The sulfur isotopic analysis of ancient seawater sulfate was traditionally
performed on sulfate-bearing evaporite minerals. More recently, techniques have
been developed to analyze trace amounts of sulfate that are incorporated into the
carbonate lattice (CAS, carbonate associated sulfate) of limestones and dolostones
(Kaplan et al., 1963; Mekhtiyeva, 1974; Burdett et al., 1989).
Low-resolution sulfur isotopic analyses of bedded marine evaporites and
CAS through the Phanerozoic demonstrate that the δ
34
S composition of oceanic
sulfate increased dramatically (from +11 to +27‰ CDT) from the Late Permian to
7
the latest Early Triassic (Figure 1-1; Holser and Magaritz, 1984; Kampschulte and
Strauss, 2004). Such an increase in δ
34
S is in agreement with a prolonged period of
oceanic anoxia. CAS studies across the Permo-Triassic boundary have revealed a
highly variable δ
34
S composition of oceanic sulfate associated with the End Permian
mass extinction (Newton et al., 2004). In Chapter 4, I present results from the
highest-resolution and most temporally extensive Permo-Triassic boundary CAS
study to date, from the Çürük Da ğ section in Turkey. I also present an interpretation
of coupled δ
34
S and δ
13
C analyses and their implications for the Permo-Triassic
Earth and the End Permian mass extinction. In Chapter 5, I present the results of a
combined evaporite and CAS investigation of the Lower Triassic of the Western US
and discuss its implications for the biotic recovery from the End Permian mass
extinction.

The fidelity of the Carbonate Associated Sulfate signal
The distribution of sulfate-bearing evaporite minerals, the most commonly
used source of ancient sulfate for analysis, is sporadic in time and space. Evaporites
are often deposited in restricted shallow marine settings and are easily dissolved
subsequent to their deposition. Therefore, evaporites tend to form in settings that
may not be representative of the open ocean and have a poor preservation potential.
Consequently, evaporite-based studies may be of limited value in the study of long-
term atmospheric and oceanographic trends that may have had important
8
consequences for the history of life. Marine barites have been used, as well (Paytan
et al., 1998), but it is not clear that they occur in enough abundance in shallow
marine settings to be useful in the ancient rock record.
It has been well documented that trace amounts of seawater sulfate are
incorporated into the calcium-carbonate lattice of carbonate minerals (Kaplan et al.,
1963; Mekhtiyeva, 1974; Burdett et al., 1989; Staudt and Schoonen, 1995; Lyons et
al., 2004). Because the rock record of ancient carbonates is more complete than the
record of marine evaporites, the analysis of carbonate associated sulfate (CAS) is
becoming an important tool for studying ancient ocean-atmosphere chemistry.
Sulfur isotopic analysis of CAS has been used for a variety of geochemical studies
on recent and ancient carbonates (Kaiho et al., 2001; Kampschulte et al., 2001;
Hurtgen et al., 2002; Kah et al., 2004; Kampschulte and Strauss, 2004; Newton et al.,
2004; Riccardi et al., 2006). Kaplan et al. (1963) first reported δ
34
S compositions
comparable to that of seawater from sulfate bound in modern mollusc shells.
Mekhtiyeva (1974) used the δ
34
S
CAS
of fossil mollusc shells as an indicator of
geochemical conditions in ancient basins. Burdett et al. (1989) further studied the
δ
34
S composition of CAS in modern mollusc shells and foraminiferal tests from the
Miocene to the Recent, and reported that δ
34
S
CAS
in both forms was similar to that of
modern seawater and the evaporite sulfate curve, respectively. Lyons et al. (2004)
demonstrated that the δ
34
S of CAS in recent micrite sediments agrees with the δ
34
S
of modern seawater sulfate.
9
Although CAS in recent sediments has been shown to agree with modern
seawater sulfate, the fidelity of the CAS signals in rocks that have undergone
variable degrees of alteration (including dolomitization) remains uncertain, and
diagenetic studies are warranted. One way to test the CAS method in ancient rocks
is to compare CAS with sulfur isotopic data from interbedded marine sulfates. One
such test revealed that δ
34
S
CAS
from Mesoproterozoic (1.2 Ga) dolostones agreed
closely with δ
34
S of associated evaporite deposits (Kah et al., 2004). However, a
systematic investigation of CAS in coeval limestones and dolostones has yet to be
reported. Likewise, there have been no published investigations of CAS from
different but coeval environments of deposition.
Chapter 2 presents the results of an investigation of the CAS method using
interbedded carbonates (limestones and dolostones) and evaporites from Lower
Triassic rocks across a spectrum of well-constrained shallow marine
paleoenvironments. The goal of the study was to investigate the dynamics of CAS
abundance and isotopic composition with changing environment and diagenetic
grade. The results of a detailed petrographic, isotopic (sulfur, carbon, oxygen and
strontium), and trace element analysis of interbedded carbonate and evaporite units
are reported in order to investigate the veracity of the CAS record across an onshore-
offshore transect in a well-constrained depositional system.
An important concern regarding the use of CAS is the possibility that pyrite
present in whole rock samples would be oxidized to sulfate during the CAS
extraction process. To date, a systematic controlled investigation of pyrite oxidation
10
during CAS extraction has not been available. In Chapter 3, I present unambiguous
data that demonstrates the oxidation of pyrite during CAS extraction, using both a
strong acid (HCl) and a weak acid (acetic). I also present a mixing model that can be
used to quantify the effect of pyrite oxidation of CAS analyses.

Early Triassic Stratigraphy and Timing
The GSSP for the base of the Early Triassic is located at Meishan, China, and
is defined as the first occurrence of the conodont Hindeodus parvus (Yin et al.,
2001). Unfortunately, the H. parvus and other basal Triassic conodont zones are
absent in the western U.S. (Clark, 1979), and so there are no known Permo-Triassic
boundary sections in the western U.S.
In this work, I adopt a hybrid of the North American and European/Chinese
stage and substage-level Early Triassic classification schemes (Figure 1-2). The
European stage Induan is used here because of the difficulty in differentiating
between the North American Griesbachian and Dienerian at the localities studied.
Thus, Induan is used to refer to the combined Griesbachian stage and the Dienerian
substage of the Nammalian. Similarly, because the base of the Smithian substage is
easy to define (based on the diagnostic occurrence of the Meekoceras ammonoid
fauna), the Smithian is treated as a stage-level classification in this work. The North
American Spathian is used as the last stage of the Early Triassic.
11

Figure 1-2: CAS and evaporite δ
34
S curve for the Phanerozoic (modified from
Kampschulte and Strauss, 2004).
12
Recent ash bed discoveries in China and advances in U-Pb radiometric dating
have allowed for a refining of the Early Triassic time-scale. The development of a
method for eliminating the effects of lead loss by pre-treating zircons by annealing
and attack with hydrofluoric acid, have allowed for more robust age dates at a
number of key biostratigraphic intervals (Mundil et al., 2004). The Permo-Triassic
mass extinction is now widely accepted to have occurred at 252.6 ± 0.2 Ma (Mundil
et al., 2004), which dates the actual Permo-Triassic boundary as slightly younger.
The boundary between the Early and Middle Triassic has been dated using U-Pb as
247.8 ± 0.4 Ma (Lehrmann et al., 2006), giving the Early Triassic, and the biotic
recovery from the End Permian mass extinction, a duration of ~5 m.y. The base of
the early Spathian Tirolites ammonoid zone has been dated as 250.6 ± 0.5 Ma
(Ovtcharova et al., 2006), which drastically changes our understanding of the length
of the Spathian stage. It was previously thought that the Induan was the longer Early
Triassic stage (Palmer and Geissman, 1999), but it is now evident that the Spathian
was much longer, lasting about 3 m.y.
The localities studied in this work are described in detail, including measured
stratigraphic sections, in Appendix A. The combined eleven localities from North
America include the entire Lower Triassic of the Western U.S. and collectively
consist of the Dinwoody, Woodside, Thaynes, Moenkopi and Union Wash
Formations. The Lower Triassic formations of the Western U.S. are discussed in
detail within Chapters 2 and 4. The Permo-Triassic of Turkey is discussed in
Chapter 4.
13

Geologic Setting
The Early Triassic world was considerably different from that of today. The
continents were assembled into the supercontinent Pangea, and the majority of
Earth’s surface was covered by the vast Panthalassa Ocean (Figure 1-3). Permo-
Triassic localities studied in this work consist of those in the Lower Triassic of the
Western United States (from the eastern Panthalassa Ocean) and the Çürük Da ğ
Permo-Triassic boundary section in Turkey (from the Neo-Tethys Ocean; Figures 1-
3 and 1-4).
The western U.S. sections were deposited in two marine basins, separated by
the Oquirrh-Uinta Uplift (Figure 1-5; Carr and Paull, 1983). The northern of the two
basins was a reactivation of the Upper Permian Phosphoria Basin, following a
regional hiatus. The northern basin sedimentary package paraconformably overlies
the Upper Permian rocks of the Gerster Formation (Clark, 1979). Sedimentary rocks
from the southern basin were deposited after a long interval of non-deposition onto
the underlying Lower Permian rocks of the Kaibab Formation (Clark, 1979). The
contact between the Permian and Lower Triassic strata of the southern basin is
characterized by the Timpoweap Conglomerate Member of the Moenkopi
Formation, which contains clasts of the underlying Permian Kaibab Formation.
14

Figure 1-3: Early Triassic time-scale and Western U.S. Lower Triassic formations
studied in this work.
15

Figure 1-4: The Early Triassic world (modified from Corsetti et al., 2005), showing
the supercontinent Pangea, the generalized location of the Western U.S. and
Turkey localities studied in this text, and the four major oceans of the time,
the Panthalassa, Boreal, Paleo-Tethys and Neo-Tethys oceans.
16

Figure 1-5: Locality map showing the locations of the Western U.S. sections studied
in this text relative to the generalized location of the Cordilleran miogeocline
and the northern and southern marine basins (shaded in gray). UW = Union
Wash; DH = Darwin Hills; BLC = Beyond Lost Cabin; BD = Blue Diamond;
RG = Rainbow Gardens; BS = Batten and Stokes locality, San Rafael Swell;
RC = Road Cut locality, San Rafael Swell; CS = Cascade Springs; HP =
Hidden Pasture; BTC = Blacktail Creek; GC = Grasshopper Creek.

17
Methods
Sampling and Preparation
Visibly well-preserved samples (those with the least amount of visible veins
and fractures) were preferentially collected from each locality. In the laboratory,
carbonate samples for isotopic analyses were first prepared by trimming obvious
diagenetic phases, such as large veins and surface weathering. The carbonate
samples were then cut into smaller pieces; one billet from each sample was used for
thin sectioning and micro-drilling, the rest were powdered using a Rock Labs
standard ring mill. For evaporite samples, cut faces were drilled for isotopic
analysis.

Petrographic Analysis and Micro-drilling of Carbonates
Detailed petrographic analysis was carried out on each sample in order to
assess the effects of diagenesis. The petrographic analyses included staining (for a
visual determination of mineralogy and iron content), cathodoluminescence, and the
construction of a paragenetic sequence for each sample. Samples were thin-
sectioned and stained with Alizarin red S and potassium ferricyanide (Dickson,
1966). The staining was used to distinguish calcite from dolomite and to indicate Fe-
rich carbonate phases. For the analyses presented in Chapter 2, the luminescence of
each sample was compared to that of other samples from the same study and ranked
on a scale of 1-9 (where 1 is dark red to black in color and 9 is bright orange). This
18
combined petrographic approach was used to determine the least altered carbonate
phases for micro-drilling directly on the thin section billet. Micro-drilled powders
were used for carbon, oxygen and strontium isotopic analyses, as well as sulfur,
manganese, strontium, calcium, magnesium and iron trace element analyses. For the
carbon isotopic analyses presented in Chapter 4, bulk powders were used in order to
obtain a time-averaged δ
13
C result comparable to that of the bulk δ
34
S analysis of
CAS.

Extraction of Carbonate Associated Sulfate
The extractions of CAS in this study were a modification of the method of
Burdett et al. (1989). Approximately 150-300 grams of carbonate powder were
subjected to two consecutive eight-hour washes in two liters of 18.2 M Ω distilled
water. After gravitational settling, the fluid was aspirated from the residual powder.
The purpose of the washes was to dissolve and remove any soluble sulfur phases,
including pyrite oxidized on the outcrop and minor evaporite inclusions. The
samples were then washed for eight hours in a solution consisting of 105 ml of 6%
NaOCl (bleach) added to 1895 ml of DDI water. The fluid was then aspirated. The
bleach step was designed to oxidize organic matter and hence remove any organic
sulfur present in the sample. The samples were then subjected to two additional
washes in DDI water to dilute and remove any residual bleach.
The samples were then dissolved using 3M HCl. The amount of HCl to be
added was determined using the stoichiometric ratios necessitated by the starting
19
sample mass (assuming the sample was pure carbonate); the purpose of this
calculation was to dissolve all of the carbonate while minimizing the risk of
dissolving any remaining sulfide phases that may oxidize at lower pH. The samples
were allowed to dissolve for eight hours. The samples were then filtered through a
0.45 micron membrane filter to remove any insoluble particulate matter. The
insoluble residue was then weighed; the mass of insolubles was then subtracted from
the starting mass of sample in order to determine the mass of carbonate dissolved.
An aliquot of 50-100 ml (depending on starting mass of sample) of a 30%
BaCl
2
solution (300 g of anhydrous BaCl
2
powder dissolved into 1000 ml of DDIW)
was then added to each sample. Because BaSO
4
(barite) is very insoluble, ionic
exchange occurs if any SO
4
2-
is present in the sample and barite is precipitated. The
samples were left for three days to ensure that any sulfate present had precipitated as
barite. The barite was then filtered from the sample with a 0.45 micron membrane.
The filters were then dried and weighed. The mass of barite precipitated was used to
calculate the amount of CAS in the sample, taking the mass of insoluble residue into
account. Filters containing insolubles where rinsed with multiple volumes of DDIW
before drying and weighing to remove any residual acid. Based on replicate
samples, the uncertainty in the CAS abundance measurement is ~8%.
For the analyses presented in Chapter 3, prior to CAS extraction, the sample
powder was first split into eight sub-samples ranging from 135 to 150g, four to be
used for CAS extraction using HCl, and four to be used for CAS extraction using
CH
3
COOH. The samples were then combined with granular pyrite (FeS
2
distributed
20
by EMD, guaranteed 85% pure through 50 mesh) to make a total of 150g using
approximately 0, 1.5, 7.5, and 15 g of pyrite. One set of samples containing 0, 1, 5
and 10 % pyrite was used for CAS extraction with HCl using the method described
above, and the other set was used for CAS extraction using CH
3
COOH by the same
procedure.

Isotopic Analyses
I visited Dr. Alan J. Kaufman’s lab at the University of Maryland in the
spring of 2002 and participated in all phases of the analysis for the Darwin Hills
samples. Dr. Kaufman’s group performed subsequent isotopic analyses.
Sulfur: A Eurovector elemental analyzer (EA) was used for on-line
combustion of barite and evaporite powders and the separation of SO
2
on-line to a
GV Isoprime mass spectrometer for
34
S/
32
S analyses following the procedures
outlined by (Grassineau et al., 2001). The effluent from the EA is introduced in a
flow of He (80-120 ml/min) to the IRMS through a SGE splitter valve that controls
the variable open split. Timed pulses of SO
2
reference gas (99.9% purity, ~ 3nA) are
introduced at the beginning of the run using an injector connected to the IRMS with
a fixed open ratio split. The isotope ratios of reference and sample peaks are
determined by monitoring ion beam intensities relative to background values.
Prepared samples (~100 µgrams) are accurately weighed and folded into
small tin cups that are sequentially dropped with a pulsed O
2
purge of 12 ml into a
catalytic combustion furnace operating at 1030
o
C. The frosted quartz reaction tube
21
is packed with granular tungstic oxide on alumina (WO
3
+ Al
2
O
3
) and high purity
reduced copper wire for quantitative oxidation and O
2
resorption. Water is removed
from the combustion products with a 10-cm magnesium perchlorate column, and the
SO
2
is separated from other gases with a 0.8-m PTFE GC column packed with
Porapak 50-80 mesh heated to 90
o
C. The cycle time for these analyses was 210
seconds with reference gas injection as a 30-s pulse beginning at 20 seconds.
Sample SO
2
pulses begin at 110 seconds and return to baseline values between 150
and 180 seconds, depending on sample size and column conditions. Isotope ratios
are determined by comparing integrated peak areas of m/z 66 and 64 for the
reference and sample SO
2
pulses, relative to the baseline of ~1 x 10
-11
A. Isotopic
results are expressed in the δ notation as per mil (‰) deviations from the Vienna
Canyon Diablo Troilite (VCDT) international standard. One sigma uncertainties of
these measurements (± 0.3‰ or better) were determined by multiple analysis of a
standard barite (NBS 127) interspersed with the samples.
Carbon and Oygen: Carbonate powders were reacted for 10 minutes at 90
o
C
with anhydrous H
3
PO
4
with a Multiprep inlet system in-line with a water trap and
dual inlet GV Isoprime gas source mass spectrometer in the Stable Isotope Facility of
the University of Maryland Geochemical Laboratories. Isotopic results are
expressed in the standard δ notation as permil (‰) deviations from the Vienna
Peedee Belemnite (VPDB) international standard. Uncertainties determined by
multiple measurements of a laboratory standard carbonate (calibrated to NBS-19)
during each run of samples were better than 0.05‰ for both C and O isotopes.
22
Strontium: Micro-drilled carbonate and evaporite powders were weighed and
placed in cleaned micro-centrifuge tubes, then pre-treated for 2 hours with 0.5M
ammonium hydroxide (~8.2 pH) to leach loosely bound Sr from clay and other
silicate surfaces (Montanez et al., 1996). Samples were centrifuged and aspirated
between each of three leaches and the final addition of 0.5M acetic acid, which was
allowed to sit for 12 hours. Strontium was isolated and purified by passing the
solution through a small Sr-spec© loaded column with successive washes of 3 and
7M HNO
3
(primarily to remove Ca from the solution), and then a wash of 0.05M
HNO
3
(to elute purified Sr), which was collected in small Teflon © beakers. This
aliquot was dried to a spot, re-dissolved in ~2 µl of 3M HNO
3
., and then ~1 µl was
loaded carefully on the middle of a pre-baked Re filament. The Sr spot was mixed
with ~0.8 µl of TaO in water and then dried, allowing the filament to briefly glow
slightly in air after about 5 minutes of progressive heating.
Samples were analyzed on a VG 54 multi-collector thermal ionization mass
spectrometer through the measurements of masses 84 through 88 in dynamic mode.
Strontium isotope ratios were collected for instrumental mass fractionation based on
an exponential law by monitoring systematic changes in
86
Sr/
84
Sr. Multiple NBS-
987 Sr isotope standard solutions measured over the period of sample analyses
yielded an average of 0.710245 ± 0.000011 (2σ).

23
Trace Element Analyses
I performed the trace elemental analyses at the University of Southern
California. Approximately 1.5 mg of micro-drilled powdered sample were acidified
with 1ml of 10% nitric acid and analyzed with a JY Ultima-C ICP-AES with
Polychronometer. Samples were analyzed with interspersed standards of known
concentrations to correct for analytical drift. Replicate analyses of standards
produced results within 0.7%, 1.5%, 1.4%, 2.0%, and 11% of known values for Mn,
Sr, Mg, Ca, and Fe respectively.

24
CHAPTER 2: ENVIRONMENTAL AND DIAGENETIC VARIATIONS IN
CARBONATE ASSOCIATED SULFATE: AN INVESTIGATION OF CAS IN
THE LOWER TRIASSIC OF THE WESTERN UNITED STATES

Abstract
An integrated stable isotope, trace element and petrographic analysis of Early
Triassic (Spathian) carbonates and evaporites along a proximal to deep
paleoenvironmental transect reveals significant variations in δ
34
S composition of
carbonate associated sulfate (CAS), but shows close agreement between the δ
34
S of
dolostone CAS and interbedded evaporites. The variations in the δ
34
S of CAS are
strongly correlated with the Ca/Mg composition of carbonates, suggesting that the
variations are driven by the degree of dolomitization. Both dolostones and
evaporites exhibit lower δ
34
S values than limestones from all localities. Three
hypotheses may explain the similarity in δ
34
S between interbedded dolostones and
evaporites: 1) proximity to continentally-derived sources of
34
S-depleted sulfur
phases may produce lower δ
34
S values in both evaporites and dolostones, 2) the
dolostones are formed from brines during a late stage of evaporation and record a
34
S-depleted Rayleigh-type distillation signal, 3) a δ
34
S depth gradient existed during
the Early Triassic such that limestones formed in distal waters record higher δ
34
S
values than evaporites and dolostones formed in proximal settings. A lack of
25
correlation between δ
34
S
CAS
in limestone samples and indicators of diagenesis (such
as Mn/Sr and
87
Sr/
86
Sr) suggest that the δ
34
S of CAS is robust against most
diagenetic processes, with the exception of dolomitization. Results from subtidal,
well-preserved limestones suggest that the δ
34
S of Spathian seawater sulfate may
have been heavier than previously suggested from analyses of evaporite deposits.

Geologic Setting
The Moenkopi Formation is an Early to Middle Triassic succession of marine
and non-marine sedimentary rocks that are exposed in the Nevada, Arizona, New
Mexico and Utah areas. Early Triassic marine deposition is represented in the
Moenkopi by the Black Dragon Member, the Sinbad Limestone Member, the Virgin
Limestone Member and the Shnabkaib Member.
The Virgin Limestone Member is a mixed carbonate-siliciclastic succession
that is dated as Spathian by the occurrence of the Tirolites ammonoid fauna. At Blue
Diamond, the Moenkopi Formation is mapped as the Virgin Limestone Member
(Carr et al., 2000), even though there is a considerable interval of bedded evaporites,
characteristic of the Shnabkaib Member. However, because there is a considerable
amount of carbonate deposited above the evaporite interval, it is likely that the two
members interfinger at Blue Diamond, and so samples from that locality are treated
here as Virgin Limestone Member.
26
At Rainbow Gardens, the Moenkopi Formation consists primarily of bedded-
evaporites and gypsum-rich shale, with lesser carbonate. For this reason, the Virgin
Limestone Member at Rainbow Gardens is considered to be represented only by the
basal fossiliferous limestone unit, and is overlain by the Shnabkaib Member, which
comprises the bulk of the section. The Middle Red Member that overlies the Virgin
Limestone Member at other localities is absent at Rainbow Gardens. Strontium
isotope chemostratigraphy presented in Chapter 2 suggest that the evaporites and
carbonates of the Shnabkaib Member at Rainbow Gardens are Spathian to possibly
earliest Anisian in age. In this work, samples from the Virgin Limestone and
Shnabkaib Members are treated as Spathian.
At Beyond Lost Cabin, the measured section begins at the first calcareous
siltstone above the Timpoweap Conglomerate Member. The Lower Red Member is
likely represented by a brown-weathering massive sandstone unit twenty meters
from the base of the measured section. However, because of lithologic similarities
between limestones below the sandstone and those above, the Virgin Limestone and
Lower Red Members can be interpreted to interfinger, and so the entire measured
section is here considered the Virgin Limestone Member. Strontium isotope
chemostratigraphy presented in Chapter 2 corroborate a Spathian to possible earliest
Anisian (Middle Triassic) age for the entire measured section at Beyond Lost Cabin.
The Timpoweap Conglomerate Member represents the initial transgression of
the Early Triassic ocean onto underlying Permian strata in the Beyond Lost Cabin
area. The Lower Red Member records the early Spathian regression suggested by
27
Carr and Paull (1983) and is followed by another transgressive episode that
deposited the Virgin Limestone Member, and the Shnabkaib Member in more
proximal settings. Because of its distal paleogeographic setting, the Beyond Lost
Cabin section most likely records both the oldest and youngest Lower Triassic rocks
of the three southern Nevada localities.
In order to investigate the abundance and δ
34
S composition of CAS with
changing environment and diagenetic grade, I systematically sampled carbonates and
bedded evaporites along an environmental transect within three Moenkopi Formation
localities—Beyond Lost Cabin (BLC), Blue Diamond (BD), and Rainbow Garden
(RG) (Figure 2-1). For detailed stratigraphic and locality information, see Appendix
A. The three localities are approximately equally spaced when shortening along the
Wilson Cliffs and Keystone thrusts (greater than 20km displacement; Burchfiel et al.,
1997) is restored to the original configuration.

Previous Results
The sulfur isotopic composition of Moenkopi Formation bedded sulfate-
bearing evaporites has been previously reported by other researchers (Claypool et al.,
1980; Wilgus, 1981). Previous sulfur isotopic studies of Lower Triassic evaporites
reveal an anomalous enrichment in the
34
S composition of Early Triassic ocean
basins from around the world (Holser and Magaritz, 1987). Although there is a
broad range of values (Figure 2-2 ), most Spathian δ
34
S values fall between +25 and
28

Figure 2-1: Map showing the localities used in this study: Beyond Lost Cabin
(BLC), Blue Diamond (BD), Rainbow Garden (RG). Modified from Bissel
(1970).
29

Figure 2-2: Spathian δ
34
S values. Modified from Holser (1984).
30
+30‰ VCDT (e.g., Cortecci et al., 1981; Holser, 1984; Holser and Magaritz, 1987;
Holser et al., 1988; Strauss, 1997) with a mean of +27‰ (Holser et al., 1988). Early
Triassic δ
34
S values are distinctly more positive than the rest of the Phanerozoic,
with the exception of the Cambrian and Devonian periods (e.g., Cortecci et al., 1981;
Holser, 1984; Holser and Magaritz, 1987; Strauss, 1997).
The Early Triassic strontium isotope record published by Korte et al. (2003)
reveals a nearly monotonic rise in the
87
Sr/
86
Sr composition of seawater during the
Early Triassic; during the Spathian, seawater
87
Sr/
86
Sr was between 0.7080 and
0.7082 (Figure 2-3). The δ
13
C
carbonate
profile for the Early Triassic demonstrates that
the Early Triassic was characterized by large fluctuations in the δ
13
C composition of
seawater carbonate (Baud et al., 1996; Payne et al., 2004; Corsetti et al., 2005).
Spathian δ
13
C
carb
values appear to fall between -1‰ and +3‰ PDB—a range of
~4‰ (Figure 2-4). In this study, relative differences in δ
18
O between samples are
used to infer different diagenetic pathways, but not the δ
18
O composition of seawater
carbonate (e.g. Veizer et al., 1997; 1999).

Results
The results of the geochemical and petrographic analyses are given in
Appendix B, as well as in Figure 2-5, Figure 2-6 and Figure 2-7, and are discussed in
detail below.
31

Figure 2-3: Early Triassic strontium isotopes. Modified from Korte et al. (2003)
32

Figure 2-4: Early Triassic carbon isotopes. Modified from Payne et al. (2004).
33

Figure 2-5: Chemostratigraphy from Beyond Lost Cabin.
34

Figure 2-6: Chemostratigraphy from Blue Diamond. The lower 100 meters were
inaccessible during this study.
35

Figure 2-7: Chemostratigraphy from Rainbow Gardens.

36
Petrographic Analysis
The samples from Beyond Lost Cabin (BLC) are predominantly bioclastic
wackestones. Petrographic analysis of the stained thin sections demonstrates the
presence of iron rich phases and the general absence of dolomite in the samples. In
terms of relative cathodoluminescence, the BLC samples had medium to high levels
of brightness (orange, levels 5-7, see Methods in Chapter 1).
Carbonate samples from Blue Diamond (BD) include micrites, bioclastic
micrites and oolites. The staining of sample BD-0 indicated the predominance of Fe-
poor calcite in the bioclasts and much of the micritic matrix. The rest of the samples
from this locality were predominantly dolomite. Some of the dolostones exhibit
primary fabric retention (mimetic dolomitization). Compared to the other samples in
this study, the Blue Diamond thin sections exhibit low to moderate levels of
cathodoluminescence (dark red to orange, levels 1-6.)
The Rainbow Garden (RG) carbonate samples are either oolite grainstones or
oolitic and bioclastic micrites. The oolites were largely grain supported with 50-70%
porosity. Staining indicated that both Fe-poor calcite and dolomite are present in the
section. The samples from the lower 25 m of the section are predominantly calcitic,
with some limestone samples exhibiting significant dolomitization. Most of the rest
of the samples are predominantly dolomite with minor Fe-rich calcite. Much of this
section has clearly experienced pervasive dolomitization, although some of the
samples retain their depositional fabrics. Compared to the other samples in this
study, the thin sections from Rainbow Garden exhibit anywhere from low to very
37
high cathodoluminescence (from dark red to bright orange, levels 2-9). The two
oolite samples found within the exposure interval (RG-38 and RG-45) are the most
cathodoluminescent samples of this study (bright orange, levels 8 and 9).

Trace Element Analyses
The Ca/Mg ratios based on quantitative chemical analysis are in agreement
with the results of the thin section staining. Samples from the Beyond Lost Cabin
locality are completely limestone, whereas samples from Blue Diamond and
Rainbow Garden include limestone, partially dolomitized limestone and dolostone.
For the purposes of this study, thin section staining and trace element analysis
suggest that carbonates can be divided into limestones, dolostones and partially
dolomitized limestones using Ca/Mg ratios. Samples with Ca/Mg ratios greater than
10 are considered to be limestone; those with lower Ca/Mg ratios are considered to
be dolostone. Samples with Ca/Mg ratios less than 50 but greater than 10 are
considered to be partially dolomitized limestone.
Fe abundance results indicate that the limestone samples from Beyond Lost
Cabin are the most enriched of all limestone samples in this study (Figure 2-8a). The
limestones from Rainbow Garden contain the least Fe, with the one limestone sample
from Blue Diamond exhibiting an intermediate Fe content. However, as a whole,
dolostone samples exhibit the highest Fe concentrations, especially those from the
Blue Diamond locality. Samples with the greatest luminescence are also those that
contain the most Mn (Appendix A), a know luminescence activator, which are
38

Figure 2-8: Stable isotope and trace element cross plots versus Ca/Mg. Locality
abbreviations are the same as in Figure 2-1.

39
highlighted by samples from the two exposure intervals at Rainbow Garden. The
least luminescent samples (those from Blue Diamond) exhibit Mn concentrations
comparable with other samples, but exhibit higher Fe concentrations, a known
luminescence inhibitor.
Most of the samples exhibit Mn/Sr values close to or less than 1 (Figures 2-5
through 2-7). Samples with higher Mn/Sr values (e.g., the two exposure interval
dolostone samples from Rainbow Garden) exhibit comparable Sr concentrations to
other samples but contain much higher concentrations of Mn.
The [SO
4
]
CAS
results show a variation with Ca/Mg and with locality (Figure
2-8b). Samples with the highest CAS concentrations exhibit the lowest Ca/Mg
values. Limestone samples from the three localities exhibit the lowest CAS
concentrations (~20 to ~700 ppm with only one sample from Rainbow Garden
exhibiting a higher value of ~2900 ppm). Partially dolomitized limestones from
Rainbow Garden exhibit intermediate CAS concentrations (~700 to ~900 ppm).
Dolostone samples from Blue Diamond and Rainbow Garden exhibit the highest
CAS concentrations (~700 to ~6800 ppm). As a whole, samples from Blue Diamond
and Rainbow Garden contain more CAS than samples from Beyond Lost Cabin.

Isotopic Analyses
Strontium: About half of the carbonate samples in this study fall within the
range of expected values for the Spathian (between 0.7080 and 0.7082; Figures 2-5
through 2-7, Figure 2-9). The remainder of the samples exhibit only slightly more
40

Figure 2-9: δ
34
S of evaporites and carbonates vs.
87
Sr/
86
Sr.

41
radiogenic strontium isotope values (the highest being 0.70832). Notably, the most
radiogenic
87
Sr/
86
Sr values are from dolostone samples taken from the Blue Diamond
locality; however, there is no discernable trend between strontium isotopes and
Ca/Mg. The evaporite samples from Blue Diamond exhibit
87
Sr/
86
Sr values in close
agreement to closely associated dolostone samples (~0.7082) whereas evaporite
samples from Rainbow Garden exhibit values that are more radiogenic (~0.7084)
than closely associated dolostones.
Oxygen: The carbonates in this study exhibit a strong relationship between
δ
18
O and Ca/Mg compositions (Figure 2-8c). Limestones from Beyond Lost Cabin
and Blue Diamond exhibit the lowest oxygen isotope values (approximately -8.6 to -
7.3‰ VPDB). Partially dolomitized limestones from Rainbow Garden exhibit
intermediate oxygen isotope values (approximately -8 to -6.5‰). Dolostones from
Blue Diamond and Rainbow Gardens exhibit the highest oxygen isotope values
(approximately -5.4 to -2.9‰).
Carbon: The majority of samples in this study exhibit δ
13
C
carb
values within
the range -3 to +1‰ VPDB (Figures 2-5 through 2-7). Overall, the range of carbon
isotope values in this study is about 2‰ more negative than results from Spathian-
aged carbonates reported by Payne et al. (2004), but the range of values is
approximately the same (~4‰). The carbon isotopic results from this study do not
vary systematically with any of the other isotopic or trace elemental results.
Sulfur: The δ
34
S values of both evaporites and CAS in this study fall
between +25 and +38‰ VCDT. Interbedded evaporite and dolostone at Blue
42
Diamond and Rainbow Garden show very close agreement (Figures 2-5 through 2-7,
Figure 2-9). Differences between the δ
34
S of interbedded evaporite and dolostone
are between -0.4 and +1.6‰, which is comparable to the results from interbedded
evaporite and dolostone reported for a Mesoproterozoic succession by Kah et al.
(2004).
The results indicate a strong relationship between δ
34
S
CAS
and Ca/Mg
compositions (Figure 2-8d). Limestones from all three localities exhibit higher
sulfur isotope values (approximately +31 to +38‰) than either evaporite or
dolostone. Partially dolomitized limestone samples from Rainbow Garden exhibit
intermediate δ
34
S
CAS
values (~ +30‰), whereas dolostone samples from Blue
Diamond and Rainbow Garden exhibit the lowest sulfur isotope values
(approximately +25 to +29‰). The range of evaporite δ
34
S values from Blue
Diamond and Rainbow Garden is similar to the range of dolostone δ
34
S values (+25
to +30‰; Figure 2-9).
There is also an apparent trend between δ
34
S
CAS
compositions and CAS
concentrations (Appendix B). The dolostone samples from Blue Diamond and
Rainbow Garden with the lowest δ
34
S values tend to have the highest CAS
concentrations whereas the limestones from Beyond Lost Cabin with the highest δ
34
S
values tend to have the lowest CAS concentrations. Likewise, there is an apparent
trend between δ
34
S
CAS
and δ
18
O (Figures 2-5 through 2-7). Limestones with the
highest δ
34
S
CAS
values exhibit the lowest δ
18
O values whereas dolostones with the
43
lowest δ
34
S
CAS
values exhibit the highest δ
18
O
carb
values. However, there are no
significant trends between δ
34
S
CAS
and Mn/Sr, δ
34
S
CAS
and [Fe], δ
34
S
CAS
and δ
13
C
carb
,
or δ
34
S and
87
Sr/
86
Sr.
Interpretations
The interpretations that follow are based on the assumption that the samples
from the three localities are broadly time equivalent. This assumption is strongly
supported by the strontium isotope results that imply a Spathian age for the three
studied sections (Figure 2-3, Figures 2-5 through 2-7). The results of the combined
analyses reported here reveal significant variation in the δ
34
S of CAS both within a
given locality and between the three localities. A lack of correlation between the
variations in δ
34
S and indicators of diagenesis such as Mn/Sr and
87
Sr/
86
Sr suggest
that the CAS results cannot be simply ruled out as being “diagenetic” in origin.
Rather, the strong correlation between δ
34
S
CAS
and Ca/Mg (Figure 2-8d) suggests
that dolomitization played a prominent role in the observed δ
34
S
CAS
values. Trends
could also be discerned between δ
34
S
CAS
and [SO
4
]
CAS
, and δ
34
S
CAS
and δ
18
O
carb
;
however, these trends are similarly driven by the degree of dolomitization (Figure 2-
8 b and c).
The
34
S enrichment of the distal limestone samples relative to the more
proximal dolostone samples may be explained by anaerobic sulfate reduction during
burial diagenesis. However, three lines of evidence argue against that hypothesis.
Although the distal limestones contain high amounts of Fe (a possible indicator for
44
burial diagenesis; Tucker et al., 1990), the more proximal dolostones as a whole
contain much more Fe (Figure 2-8a). The distal limestones exhibit strontium
isotopic values within or close to expected values for the Spathian, and most have
Mn/Sr values less than one (Figures 2-5 and2-9). The strontium isotopic and trace
element results indicate that the distal limestones are recording a marine strontium
signal. Lastly, petrographic analyses failed to reveal significant amounts of pyrite or
pyrite pseudomorphs that would be expected if sulfate reduction were a prominent
reaction and if iron was widely available in the rocks during burial.
Limestone samples exhibiting Ca/Mg values greater than 50 are interpreted
here to be recording the least altered δ
34
S
sulfate
values, which range between +31 to
+38‰ VCDT, a variation of 8‰; this entire range is represented by the samples
from Beyond Lost Cabin. Therefore, we interpret the chemostratigraphic plot of
δ
34
S
CAS
from Beyond Lost Cabin (Figure 2-5) to best represent the variation of
δ
34
S
sulfate
during the Spathian. Explanations for the observed variation in δ
34
S
sulfate

during the Early Triassic are beyond the scope of this chapter, but have been
addressed elsewhere (Chapter 5; Marenco et al., 2006). The following discussion
introduces three hypotheses to explain why the dolostone and evaporite samples in
this study consistently exhibit lower δ
34
S values than limestones.

Hypothesis 1: Influence of Continentally-derived Sulfur
The δ
34
S
evaporite
values (+25 to +30‰ VCDT) were distinctly lower than the
δ
34
S
limestone
values (+31 to +38‰). One possible interpretation for the offset in
45
values is that the evaporites formed in an evaporative setting where seawater was
mixed with continentally derived fluids; such fluids might contain
34
S-depleted
sulfur phases resulting from the oxidative weathering of shales and organic matter
(Kaufman et al., 2007).
The stratigraphic co-occurrence of dolostone and evaporite suggests that, for
the samples in this study, the dolomitization process was strongly facies-dependent;
this interpretation is supported by the lack of dolomitization in the most distal
locality (Beyond Lost Cabin). The facies dependency of dolomitization might
explain why the range of δ
34
S
evaporite
values (+25 to +30‰ VCDT) and δ
34
S
dolostone

values (+25 to +29‰) show such strong overlap. The seawater involved in the
dolomitization process likely experienced evaporative concentration of Mg. The
trend of increasing [SO
4
]
CAS
with decreasing Ca/Mg (Figure 2-8b) supports the
interpretation that dolomitization occurred in seawater that had experienced
evaporative concentration. The sulfate concentrations observed within dolomites
(735 to 6732 ppm) is remarkably similar to that observed within evaporative
dolomites from the Ordovician to Miocene rock record (~1000 to ~7000 ppm ;Staudt
and Schoonen, 1995). Likewise, the trend of increasing δ
18
O
carb
with decreasing
Ca/Mg (Figure 2-8c) suggests that the dolomitizing fluid had experienced
evaporation insofar as
18
O becomes preferentially enriched in water as evaporation
progresses (e.g., Hoefs, 1997).
Dolomitization most likely occurred early in the study section. In thin
section, some dolomitized samples show remarkable primary fabric retention of
46
ooids (mimetic dolomitization). Although there are no distinct trends between
87
Sr/
86
Sr and Ca/Mg, the fact that some dolostones record marine strontium isotope
values (Figure 2-9) makes it unlikely that dolomitization was a late diagenetic
process.

Hypothesis 2: Evaporitic
34
S Enrichment
The isotopic fractionation of oxygen during evaporation and condensation
has been well documented and is known as the Rayleigh distillation effect(e.g.,
Hoefs, 1997). A depletion of
34
S in late-stage evaporite deposits has long been noted
from ancient evaporite basins and has led previous workers to speculate that a
Rayleigh-type distillation effect occurs as evaporite minerals are deposited (Nielsen
and Ricke, 1964; Holser and Kaplan, 1966; Raab and Spiro, 1991). Assuming a
fractionation between seawater and evaporite of 2‰ (e.g., Claypool et al., 1980), the
δ
34
S evolution of a hypothetical brine (starting at +35‰) and precipitate is shown as
a function of fraction of fluid remaining in Figure 2-10a. According to this ideal
model, towards the end of the evaporation process, the δ
34
S composition of the
remaining fluid approaches 25‰, or 10‰ less than the starting fluid. Figure 2-10b
shows the isotopic difference between a brine and its initial isotopic composition as a
function of evaporation at different fractionations between seawater and evaporite
(∆
evaporite-seawater
of 1.6‰ to 2.4‰); it shows that regardless of the fractionation factor
used, towards the end of evaporation, the remaining brine is depleted by more than
8‰ than the starting brine.
47

Figure 2-10: Model results of Rayleigh-type
34
S depletion in a hypothetical brine
and precipitate as evaporation progresses. A.) graph shows
1 −
=
α
f R R
initial fluid

and
fluid rock
R R α = where R
fluid
is the ratio
34
S:
32
S in the fluid, R
initial
is the
starting ratio
34
S:
32
S in the fluid, R
rock
is the ratio
34
S:
32
S in the rock, f is the
fraction of fluid remaining, and α is the fractionation factor between fluid
and the precipitate, taken to be 1.002. Results are shown in the standard delta
notation. B.) Graph shows
fluid initial
R R − as a function of f at different
values of α (1.0014, 1.002, and 1.0024). α values are shown as ∆
evaporite-fluid

(1.6‰, 2.0‰, and 2.4‰).

48
If the dolomitizing fluid formed from seawater towards the end of
evaporation (without recharge from the ocean), then the ideal model presented here
might explain why δ
34
S is approximately 10‰ lower in dolostones than in
limestones. Unfortunately, the results of this study do not conclusively support or
refute this hypothesis. For example, it is not readily apparent why none of the
evaporite samples would have recorded δ
34
S values comparable to the most
34
S-
enriched limestones. Presumably some of the evaporite samples would have formed
during the early stages of evaporation, when the Rayleigh effect would have been
minimal. In order to further explore this issue, a finer-scale sampling within the
evaporite units is needed.
The question of Rayleigh-type fractionation of sulfur during evaporation has
been previously explored experimentally by Raab and Spiro (1991). They found that
an isotopic depletion occurred in the brine during the gypsum and halite stability
fields (down to about 30% of fluid remaining), but this depletion was only about 2‰
(Raab and Spiro, 1991). During more highly evaporated stages, the experimental
brines actually showed a
34
S enrichment, although this data was described as
“tentative” (Raab and Spiro, 1991). Nonetheless, the results of Raab and Spiro
suggest that the fractionation between seawater and gypsum changes as evaporation
progresses, and that the ideal model for evaporation presented above may not always
apply.
However, for the present study, the ideal model does seem to fit the data.
The high δ
18
O values of dolostones relative to those of limestones would suggest that
49
the dolomitizing fluid had experienced some degree of evaporative concentration.
Likewise, the [SO
4
] results also suggest that highly evaporated seawater was the
source of the dolomitizing fluid. The question of Rayleigh-type fractionation of
sulfur during the precipitation of evaporites warrants further study, in particular with
brines of different compositions from modern seawater.

Hypothesis 3: δ
34
S Gradient with Depth
An alternative interpretation of the δ
34
S versus Ca/Mg results is that,
assuming dolomitization was facies controlled and occurred in shallower settings, the
CAS results are recording a primary seawater δ
34
S
sulfate
gradient with depth. The
presence of a δ
34
S
sulfate
gradient in Early Triassic seas has been previously suggested
(Newton et al., 2004; Marenco et al., 2006). The δ
34
S equivalency between
evaporite and dolostone, coupled by marine strontium isotopic compositions in some
dolostones (and only slightly more radiogenic values in others) supports this
hypothesis. Causes for a sulfur isotopic gradient with depth during the Early Triassic
are beyond the scope of this chapter, but are discussed elsewhere (Chapters 4 and 5;
Isozaki, 1997; Newton et al., 2004; Grice et al., 2005; Kump et al., 2005; Marenco et
al., 2006) and are likely the result of the non-conservative behavior of this major ion
in Early Triassic seawater.

50
Implications for future CAS studies
The results of this study have a number of important implications for the
study of carbonate associated sulfate:
1) The δ
34
S of CAS in ancient rocks is robust to most diagenetic processes
that would affect alteration indicators such as Mn/Sr and
87
Sr/
86
Sr. Lyons et al.
(2004) concluded that the δ
34
S
CAS
of recent non-dolomitized sediments was
independent of diagenetic processes; our results extend their conclusions to ancient
rocks.
2) The δ
34
S of CAS in some basins may be altered during dolomitization, as
indicated by a strong correlation between δ
34
S
CAS
and Ca/Mg. Although there may
be multiple pathways for dolomitization to occur, the results presented here suggest
that at least some dolostones in the rock record may contain altered δ
34
S
CAS

compositions. There may be evidence to suggest that dolomitization mechanisms
may have been different prior to the Phanerozoic (Corsetti et al., 2006); therefore,
the effect of dolomitization on CAS in Precambrian rocks deserves further
investigation.
3) The mineral phase of choice for sulfur isotope chemostratigraphy is calcite
(or aragonite, if available) from rocks formed in open, subtidal shelf settings. To
avoid a late diagenetic
34
S enrichment from sulfate reduction, organic-rich samples
with abundant pyrite or pyrite pseudomorphs should be avoided. Because of the
possibility of contamination with continentally-derived sources of
34
S-depleted sulfur
51
and the potential for Rayleigh
34
S depletion during evaporation, evaporite deposits
are less ideal for sulfur chemostratigraphy than subtidal shelf limestones.

52
CHAPTER 3: OXIDATION OF PYRITE DURING EXTRACTION OF
CARBONATE ASSOCIATED SULFATE

Abstract
The sulfur isotopic composition of carbonate associated sulfate (CAS) has
been used to investigate the geochemistry of ancient seawater sulfate. However, few
studies have quantified the reliability of δ
34
S of CAS as a seawater sulfate proxy,
especially with respect to later diagenetic overprinting. Pyrite, which typically has
quite depleted δ
34
S values, can be a common constituent of sedimentary rocks. The
oxidation of pyrite, whether during diagenesis or sample preparation, could
adversely influence the sulfur isotopic composition of CAS. Here, I report the
results of CAS extractions using HCl and acetic acid with samples spiked with
varying amounts of pyrite. The results show a very strong negative linear
relationship between the amount of pyrite added to the sample and the resultant δ
34
S
value. Likewise, a very strong positive linear relationship is present between the
amount of pyrite added and the observed concentration of sulfate from the extraction
process. This data represents the first unequivocal evidence that pyrite is oxidized
during the CAS extraction process. Mixing models indicate that in samples with
much less than 1 weight % pyrite and relatively high δ
34
S
pyrite
values, the isotopic
offset imparted by oxidation of pyrite should be much less than -4‰. The diagenetic
oxidation of pyrite and subsequent incorporation of
34
S-depleted sulfur into CAS is
53
unavoidable and difficult to quantify, so samples with even moderate amounts of
pyrite or pseudomorphs after pyrite should be considered suspect for CAS analysis.

Methods and Material Studied
The limestone sample used in this study was collected from the Spathian
Virgin Limestone Member of the Moenkopi Formation at the Beyond Lost Cabin
locality in southern Nevada, at a height of 126 meters from the base. Detailed
stratigraphic and locality descriptions are given in Chapter 2 and Appendix A.
Isotopic analyses of Spathian evaporites from more onshore facies of the Moenkopi
Formation have yielded δ
34
S values approaching +30 ‰ CDT (Wilgus, 1981;
Marenco et al., In Review) whereas CAS studies from the more distal Beyond Lost
Cabin locality have reported values as high as +38 ‰ VCDT (Chapters 2 and 5;
Marenco et al., In Review). Consequently, I argue in Chapter 2 that seawater sulfate
δ
34
S was likely around +35 ‰ CDT during the Spathian.

Results
The pyrite used in this study has a δ
34
S value of +9.39 ‰ VCDT (this value
will be referred to as δ
34
S
pyrite
). For the following discussion, sulfate resulting from
the CAS extraction process will be referred to as ‘apparent CAS’; its isotopic
composition will be called δ
34
S
app
and its concentration will be [CAS]
app
. The
average apparent CAS concentration of the two unspiked samples are taken to
54
represent the actual sulfate concentration in the original limestone sample ([SO
4
]
lime

= 552 ppm). The two unspiked samples exhibit a standard deviation of 18 ppm, thus
we regard the uncertainty in the [CAS]
app
measurements to be ~3%. Although the
samples extracted using acetic acid resulted in more insolubles (18% of total starting
mass vs. 12% for HCL), the [CAS]
app
values were similar to those extracted using
HCl (Table 1), implying that the extra insolubles were un-reacted limestone. The
isotopic compositions of the two unspiked samples are assumed to be that of the
original limestone sample (δ
34
S
lime
). The average δ
34
S value of the unspiked samples
is +34.7 ‰ VCDT (standard deviation = 0.4‰, n = 4 including one replicate for
each).
For the following discussion, the mass of pyrite (M
pyrite
) relative to the
starting sample mass will be referred to as ‘weight percent pyrite’ whereas the mass
of pyrite relative to the mass of limestone dissolved (M
lime
, the starting sample mass
minus insolubles) will be referred to as ‘normalized fraction pyrite’. This distinction
is made to facilitate the discussion of two-component mixing (see below), which
assumes that only pyrite and limestone (not insolubles) contributed to the amount of
sulfate extracted from the starting sample. The data reveal a positive linear
correlation between apparent CAS concentration and weight percent pyrite (Figure
3-1; HCl R
2
= 0.9716; acetic R
2
= 0.9783) for both CAS methods. Likewise, there is
a distinct negative linear relationship between weight percent pyrite and δ
34
S
app

(Figure 3-2; HCl R
2
= 0.9736; acetic R
2
= 0.9727).
55

Figure 3-1: Plot of sulfate concentration from CAS extraction versus weight percent
pyrite. Long-dashed line is a linear regression through the HCl data with an
R
2
of 0.9716. The short-dashed line is a linear regression through the Acetic
data with an R
2
of 0.9783.
56

Figure 3-2: Plot of δ
34
S results from CAS extraction versus weight percent pyrite.
The long-dashed line is a linear regression through the HCl data with an R
2
of
0.9736. The short-dashed line is a linear regression through the Acetic data
with an R
2
of 0.9727.

57
The apparent CAS concentration can be modeled according to the linear
equation:
pyrite 4 lime 4 app
] [SO ] [SO [CAS] + =
[SO
4
]
pyrite
is the concentration of sulfate resulting from the oxidation of pyrite
and is related to the mass of pyrite according to:
lime
pyrite
pyrite 4
M
M
w ] [SO =
where w represents the ppm sulfate formed per gram pyrite, assuming that the
amount of sulfate oxidized from pyrite is significant enough to affect [CAS]
app
, but is
insignificant compared to the mass of insolubles. The linear equation then becomes:
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
+ =
lime
pyrite
lime 4 app
M
M
w ] [SO [CAS]
so that on a plot of [CAS]
app
vs normalized mass pyrite, w can be determined
from the slope of the line (Figure 3-3). A linear regression through the entire data set
gives a slope of 1679 ± 205 ppm (R
2
= 0.9129).

Likewise, the isotopic composition of apparent CAS can be modeled by
treating it as a mixture of two components:
pyrite
34
app
pyrite 4
lime
app
pyrite 4
app
34
S
[CAS]
] [SO
S
[CAS]
] [SO
S δ δ δ
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
+
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
− =
34
1
58

Figure 3-3: Plot of modeled [CAS]
app
values versus the M
pyrite
/M
lime
values from this
study. The line is fit through the entire data set (both HCl and acetic, R
2
=
0.9129, standard error = 205.1 ppm).

59
Which can be simplified to:
lime
34
app
pyrite 4
lime
34
pyrite
34
app
34
S
[CAS]
] [SO
S S S δ δ δ δ +
⎟
⎟
⎠
⎞
⎜
⎜
⎝
⎛
− = )(
Figure 3-4 shows the plot of δ
34
S
app
versus [SO
4
]
pyrite
/[CAS]
app
for the
samples in this study. A linear regression fit through the entire data set has an
intercept that yields δ
34
S
pyrite
= 8.9 ‰, quite similar to the measured δ
34
S composition
of the pyrite used in this study (+9.4 ‰), suggesting that little to no fractionation of
sulfur occurred during the oxidation of pyrite. The slope of the regression should
yield δ
34
S
pyrite
– δ
34
S
lime
. The observed result is -25.7 ± 2.4 ‰ (R
2
= 0.9494), quite
similar to the result expected from the endmember compositions, equal to 34.7-9.4 =
25.3 ‰.

Discussion
The results reported here provide conclusive evidence that pyrite is oxidized
by either hydrochloric or acetic acid during the extraction of CAS. Our findings
have significant implications for the future use of CAS to study ancient seawater
sulfate. As a minimum, studies of CAS should include some attempt to quantify
pyrite abundance in samples used. With only 1 % pyrite with a δ
34
S composition of
+9 ‰ VCDT, an isotopic depletion of over 1 ‰ was observed. Because pyrites in
sedimentary rocks can exhibit much lower δ
34
S values, as low as -50 ‰ VCDT
(Hoefs, 1997), the isotopic influence of pyrite in this study are small compared to the
60

Figure 3-4: Plot of modeled δ
34
S
app
values versus [SO
4
]
pyrite
/[CAS]
app
values from
this study. The line is fit through the entire data set and has a slope of -25.7
‰ (R
2
= 0.9494, standard error = 2.4 ‰).

61
range of possible values. Using the slope of the line in Figure 3-3, we can predict the
effects of pyrites with different δ
34
S compositions on [CAS]
app
and δ
34
S
app
. Figure 3-
5 shows a plot of [CAS]
app
and δ
34
S
app
-δ
34
S
lime
at various compositions of δ
34
S
pyrite

(shown as ∆δ = δ
34
S
lime
-δ
34
S
pyrite
) versus mass fraction pyrite. Even at low pyrite to
limestone ratios, the oxidation of pyrite can have significant effects on the δ
34
S of
apparent CAS if the δ
34
S composition of the pyrite is much less than that of the
limestone. Realistically, limestones are unlikely to have more than 1 weight %
pyrite (e.g., Riccardi et al., 2006). Therefore it should be noted that with a large ∆δ
(e.g., ∆δ = 70 in Figure 3-5), a sample with a normalized fraction pyrite of 1 % or
less exhibits a maximum isotopic offset of about -4 ‰. Consequently, in samples
with much less than 1 weight % pyrite and ∆δ values much lower than 70, it can be
argued that the isotopic effect of any pyrite oxidized would be much less than -4 ‰.
Conversely, the pyrite used in this analysis is coarser-grained than most naturally
occurring syngenetic pyrite. The increased surface area to volume ratio in naturally
occurring pyrite may allow it to be more reactive than that used in this study.
We have limited our experiments to the effects of dissolution during the
acidification stage of the CAS extraction procedure. Presumably, the oxygen used to
oxidize pyrite during dissolution comes from oxygen in the atmosphere. However,
replicate CAS extractions under aerobic and anaerobic conditions have yielded
similar results (Dr. Timothy Lyons, personal communication, 2007). Regardless,
even an oxygen free analysis might not be free of contamination if pyrite forms,
62

Figure 3-5: Results of modeling using the slope of the line in Figure 3-3 with a
hypothetical [SO
4
]
lime
of 300 ppm and different δ
34
S
pyrite
values. The left-
hand axis shows the difference between the δ
34
S of apparent CAS and
limestone, assuming that the difference between the δ
34
S of limestone and
pyrite (∆δ) are either 10, 30, 50, or 70 ‰. The right-hand axis shows the
concentration of apparent CAS. Both plots are given relative to M
pyrite
/M
lime
.

63
oxidizes, and is subsequently incorporated as CAS during diagenesis. If the
oxidation and subsequent incorporation occurred in nature before the sample was
collected, there is no corrective action that can be taken. Therefore, samples with
abundant pyrite or pyrite pseudomorphs should be avoided for CAS analysis.

Conclusions
The results of this study support the continued use of CAS to study the δ
34
S
composition of ancient seawater sulfate. However, care must be taken to quantify
the amount and isotopic composition of pyrite in CAS samples. If pyrite δ
34
S and
abundance is known, then the mixing models presented here can be used as a first
approximation of the offset on δ
34
S of CAS imparted by the oxidation of pyrite.
However, because sedimentary pyrite is likely to be much finer-grained than that
used in this study, the isotopic influence exerted by its presence might be even
greater.

64
CHAPTER 4: COMBINED PERMO-TRIASSIC SULFUR AND CARBON
ISOTOPE VARIATIONS AND THEIR IMPLICATIONS FOR THE END-
PERMIAN MASS EXTINCTION

Abstract
δ
34
S from carbonate-associated sulfate coupled with δ
13
C from the Permo-
Triassic (P-T) boundary at Çürük Da ğ, Turkey, reflect a combination of long-lived
oceanographic factors and intense volcanism from the Siberian Traps that led to the
largest biotic crisis of the Phanerozoic. A pronounced increase in average δ
34
S
values (from ~+20‰ to greater than +30‰ VCDT) and a decrease in average δ
13
C
values (from +5‰ to +1‰ VPDB) across the P-T boundary support a carbon cycle
shift from organic carbon burial on land during the Permian to pyrite sulfur burial
under euxinic conditions during the Early Triassic (Berner, 2005), as do
corroborating data (Isozaki, 1997). A detailed examination of the boundary interval
reveals large negative fluctuations (~10‰) in δ
34
S that occur without any variation in
δ
13
C before the mass extinction horizon, indicating that they were probably not
caused by changing redox conditions in the upper ocean, but rather resulted from
large inputs of
34
S-depleted sulfur to the oceans, possibly from multiple eruptions of
the Siberian Traps. At the mass extinction horizon, δ
34
S increases dramatically, and
δ
34
S and δ
13
C exhibit negatively correlated fluctuations through a short stratigraphic
65
interval. Here, the negative co-variation between δ
34
S and δ
13
C indicates rapidly
changing redox conditions in the upper ocean, perhaps due to an oscillating
chemocline separating surface oxic from deep anoxic/sulfidic (euxinic) water
masses. The direct cause of the mass extinction was likely the euxinification of the
shallow ocean, which may have introduced toxic levels of CO
2
and H
2
S to the
atmosphere. The release of reductants and nutrients from Siberian Trap volcanism
before the mass extinction may have caused the already-present deep-water euxinia
to expand into the shallow ocean.

Geologic Setting
The analyses presented in this chapter were performed on samples from a P-T
boundary section in Turkey that is 1) the only section discussed in this work that was
not measured and sampled by me (sampling was peformed by David J. Bottjer,
Aymon Baud and Sylvain Richoz), and 2) the only representative of the Tethys
Ocean discussed in this work.
The P-T boundary section at Çürük Da ğ near Antalya, Turkey (Figures 1-3
and 4-1), consists of well-preserved limestone rocks that formed in a shallow, normal
marine platform setting in the ancient Neo-tethys ocean (Marcoux and Baud, 1986).
The Permian limestones belong to the Pamucak Formation and are represented by a
thick (400 to 600 m) cyclic succession of inner to outer platform facies. The upper
100 m of the Formation studied here is Wuchiapingian through Changhsingian
66

Figure 4-1: Map showing the location of Çürük Da ğ.

67
(Upper Permian) in age and is made up of black nodular limestones with localized
chert. The Upper Permian limestones are rich in calcareous algae and small
foraminifera, with intervals containing brachiopods, echinoderms, crinoids and
bryozoans. Most of the Upper Permian limestones are classified as bioclastic
wackestones and are interpreted to have been deposited under low energy conditions
below wave base. Two meters below the base of the Kokarkuyu Formation marks a
facies change into high-energy bioclastic grainstones, and then into oolitic
grainstones with echinoderms, bivalves and foraminifera of latest Changhsingian
age. The Wuchiapingian-Changhsingian boundary at this locality is not certain and
is considered an upper limit, based on the first occurrence of characteristic
Changhsingian benthic foraminifers (Richoz, 2004).
The base of the Triassic is indicated by the presence of the conodont
Hindeodus parvus (Richoz, 2004). The extinction level is placed at 30cm below the
P-T boundary and is marked by the disappearance of the Permian fauna and the onset
of a fossil-poor calcimicrobial unit (Baud et al., 1997) that continues upward with
interbedded oolitic grainstones for 40 m.

Sample Preservation
It is important to note that there is no evidence for condensed intervals in the
Çürük Da ğ section. Condensed intervals are problematic for CAS studies because of
the amount of rock needed for each analysis (~150g). Rock samples for CAS can
68
represent more than 5-10 cm of stratigraphic thickness and thus result in a time-
averaged δ
34
S value. In highly condensed sections, such as the P-T GSSP at
Meishan, China(Bowring et al., 1998), CAS data may average as much as 10-20 k.y.
per sample (using the dates from (Bowring et al., 1998) to calculate a sedimentation
rate for Meishan, see below). Consequently, high-frequency δ
34
S variations such as
those reported here and in (Newton et al., 2004) would be obscured at Meishan. The
preservation of the section is considered excellent. Organic carbon content is low,
which is also considered a benefit, insofar as post-depositional alteration of buried
organic matter by sulfate reducing bacteria may affect the CAS of the carbonates via
the later oxidation of their by-product, pyrite. A recent study of the P-T boundary at
Meishan and the nearby Shangsi sections reveals that both sections contain abundant
pyrite associated with the P-T boundary (Riccardi et al., 2006). An apparent
correlation between pyrite abundance and negative δ
34
S values (-15 ‰ VCDT) of
CAS suggests that pyrite was likely oxidized during the CAS extraction process
during that study (Fig. 2 in Riccardi et al., 2006).
As discussed in Chapter 2, trace elements preserved in limestones are
commonly used to access post-depositional alteration. The results of a trace element
analysis for Mn, Sr, Ca, and Mg are given in Appendix B. Only two of the samples
exhibit Mn/Sr ratios of ~1, while rest of the samples exhibit Mn/Sr ratios of less than
0.5. There is no correlation between Mn/Sr and δ
34
S or Mn/Sr and δ
13
C. Low Mn/Sr
ratios are characteristic of mineral phases formed in seawater because of the
oxidative strength of the ocean. Minerals formed during diagenesis tend to
69
incorporate more Mn than those formed in seawater, and result in higher Mn/Sr
ratios (Tucker et al., 1990). The low Mn/Sr ratios of all of the samples suggest that
the samples have not experienced much meteoric alteration, if any. The complete
lack of correlation between Mn/Sr and either δ
34
S or δ
13
C suggests that meteoric
diagenesis has not affected the sulfur or carbon isotopic composition of the samples.
It has been suggested that degree of dolomitization affects d34SCAS values (Chapter
2; Marenco et al., In Review). The only dolomite sample in this study is from 20.5
meters above the Permo-Triassic boundary and did not result in a low δ
34
S value
compared to the rest of the samples in this study. Thus, the Çürük Da ğ section
represents an excellent candidate for CAS investigation.

Carbon Isotope Chemostratigraphy, Correlation and Timing
The high resolution δ
13
C profile from Çürük Da ğ (Richoz, 2004) can be
readily compared to that from the P-T GSSP section at Meishan (Figure 4-2D). At
Meishan, two ash beds with an age difference of 900 k.y. are 4.5m apart, giving an
accumulation rate for the interval of 200 k.y. per meter. Two prominent negative
δ
13
C shifts below the P-T boundary at Meishan are spaced 0.6 m apart, giving a
temporal separation of 120 k.y. using the calculated accumulation rate (Figure 4-2D).
The two negative shifts can be correlated to corresponding shifts separated by
approximately 12 m at Çürük Da ğ, giving an accumulation rate of 10 k.y. per meter
(Figure 4-2CD), and an estimated age range of 1.4 m.y. for the entire study section.
70
Figure 4-2: Chemostratigraphy of the Çürük Da ğ P-T boundary section. The mass
extinction level is marked by a black dashed line (small dashes). The date for
the Permo-Triassic mass extinction is from Mundil et al. (Mundil et al.,
2004). A. δ
34
S of carbonate associated sulfate reported as ‰ VCDT. B. A
five-point moving average plot through the δ
34
S data. C. δ
13
C of carbonate
reported as ‰ PDB. Black dots are micro-drilled analyses from Richoz
(Richoz, 2004). Black open triangles are bulk analyses from this study. D. A
reproduction of the δ
13
C profile at Meishan, China, as reported by Bowring et
al. (Bowring et al., 1998). Red dashed lines indicate tie-points between the
δ
13
C profiles from Çürük Da ğ and Meishan, China. Black dashed lines (large
dashes) mark horizons with radiometric dates from Bowring et al. (Bowring
et al., 1998) with a time difference of 900 k.y. Ch. = Changxingian. Griesb.
= Griesbachian.
71

Figure 4-2: Chemostratigraphy of the Çürük Da ğ P-T boundary section
.
72
An accumulation rate of 10 k.y. per meter is in excellent agreement with average
accumulation rates of Phanerozoic carbonate platforms(Bosscher and Schlager,
1993). In the absence of datable ash beds at Çürük Da ğ, the estimated accumulation
rate is considered to be a first approximation, as it has not been determined whether
the sections at Meishan and Çürük Da ğ were deposited at a constant rate.

Results
Carbon Isotopic Fluctuations
Carbon isotopic analysis was performed on the bulk powders from which
CAS was extracted, in addition to the previous δ
13
C analyses performed by (Richoz,
2004) from the same section in order to directly compare the δ
34
S results with δ
13
C
from the same samples. The bulk analyses are shown on Figure 4-2C as open
triangles. Although δ
13
C analyses on bulk samples can be problematic, Figure 4-2C
shows that the bulk analyses and the micro-drilled analyses of Richoz (2004) exhibit
strong agreement. The one exception is the bulk sample at 15m above the P-T
boundary that resulted in a δ
13
C value considerably lower than the corresponding
micro-drilled data. Consequently, the result at +15m will not be included in the
interpretations below.
The δ
13
C profile shows very little variation before the mass extinction level
(Figure 4-2C). δ
13
C values average around +5‰ VPDB for much of the studied Late
Permian section. At 12 meters below the P-T boundary, δ
13
C begins to drop to
73
values between +3 and +4‰ VPDB. At the mass extinction horizon (30cm below
the P-T boundary), δ
13
C values drop drastically to ~+1‰ VPDB. Immediately
above the mass extinction level, δ
13
C is considerably more variable than in the pre-
extinction interval. Fluctuations with magnitudes between 1 and 2‰ characterize
the 15m above the mass extinction level (Figures 4-2 and 4-3).

Sulfur Isotopic Fluctuations
The sulfur isotopic profile through the Çürük Da ğ section is shown in Figure
4-2A. A 5-point moving average plot through the sulfur isotopic results is shown in
Figure 4-2B. Sampling resolution is highest (~1m or less) in the 9m preceding the
mass extinction level through 20m above the P-T boundary. Consequently, most of
the interpretation that follows will be restricted to the immediate pre- and post-
extinction intervals.
δ
34
S is highly variable through the entire section studied. Throughout most
of the Upper Permian interval, δ
34
S values vary from about +20 to + 25‰ VCDT as
can be seen in the moving average plot (Figure 4-2B). At 9m below the P-T
boundary, three large (~10‰/m) negative δ
34
S excursions are recorded. Following
each excursion, δ
34
S returns to values approaching +25‰ VCDT. The estimated
accumulation rate suggests that the first complete excursion (from high values to low
values and back) lasted 50 k.y. and the second two lasted 20 k.y. Beginning at the
mass extinction horizon, the δ
34
S fluctuations become smaller in magnitude and
74

Figure 4-3: δ
34
S of carbonate associated sulfate (blue dots) and δ
13
C
bulk
of carbonate
(open triangles) versus height at Çürük Da ğ. Only the interval where
sampling resolution was one meter or less is shown.

75
frequency (~3-5‰/2m) and comprise a prominent δ
34
S rise to values greater than
+30‰ VCDT by the top of the studied section (Figure 4-2AB).
Although δ
34
S appears to exhibit variability in the lower-resolution Upper
Permian interval (-100 to -9m), values as low as those in the immediate pre-
extinction interval are not recorded. The moving average plot (Figure 4-2B) shows
an overall increase in δ
34
S through the study section. Superimposed on the overall
δ
34
S rise is a prominent drop in δ
34
S in the ~9m below the extinction horizon
followed by an equally prominent rise in δ
34
S in the immediate post-extinction
interval.

Sulfur and Carbon Isotopic Covariation.
Although δ
34
S shows considerable variation in both the pre and post-
extinction intervals, δ
13
C shows little to no variation until after the mass extinction
level. Figure 4-3 shows δ
34
S and δ
13
C analyzed from the same bulk powders in the
immediate pre and post-extinction intervals. δ
34
S and δ
13
C do not co-vary before the
mass extinction interval, but immediately following the extinction level, they exhibit
a strong negative covariance (Figure 4-3). Likewise, a cross plot of δ
34
S and δ
13
C
(Figure 4-4) shows no covariation in the pre-extinction interval, and a negative
covariation (slope = -2, R
2
= .306) in the post-extinction interval.
76

Figure 4-4: Cross plot of δ
34
S of carbonate associated sulfate and δ
13
C
bulk
of
carbonate. Only results from the interval where sampling resolution was one
meter or less are shown.

77
Discussion
The magnitude of the δ
34
S fluctuations reported here are comparable to those
reported by Newton et al. (Newton et al., 2004) from Italy. Overall, δ
34
S from Çürük
Da ğ is somewhat higher than that from Italy(Newton et al., 2004), attesting to the
better preservation of the Çürük Da ğ samples. It is highly unlikely that altered
samples from geographically distant localities would result in similar isotopic
fluctuations. Studies of the P-T GSSP at Meishan, China have not revealed high-
frequency fluctuations in δ
34
S (Kaiho et al., 2001; Kaiho et al., 2006). The time-
averaged results from Meishan are testimony to the highly condensed nature of the
section.

Long-term Trends
During the 1.4 m.y. represented in the study section, average δ
34
S increases
from +20 to greater than +30‰ VCDT while average δ
13
C decreases from +5 to
+1‰ VPDB. Global compilations of evaporite and CAS δ
34
S and carbonate δ
13
C
data have demonstrated a switch from low to high δ
34
S and high to low δ
13
C across
the P-T boundary (Claypool et al., 1980; Holser, 1984; Kampschulte and Strauss,
2004). The causes of long term carbon and sulfur cycle changes associated with the
Permo-Triassic transition have been extensively studied via numerical modeling
(Berner, 2002; Berner, 2005). The long term (10
6
-10
7
year) changes in the carbon
and sulfur cycles have been interpreted to have resulted from a decrease in the rate of
78
burial of organic matter on land and an increase in the rate of pyrite burial under
euxinic conditions (Berner, 2005). Although the post-extinction δ
13
C profile exhibits
multiple short term (10
4
year) fluctuations (see below), the bulk of the drop to
average values of ~1‰ occurs at the mass extinction level. Berner (2002) explored
multiple hypotheses that attempt to explain the drop in δ
13
C at the mass extinction
level and concluded that degassing and oxidation of methane (Krull and Retallack,
2000) could have readily produced the drop. The source of the methane has been
attributed to destabilization of methane clathrates (Bowring et al., 1998) as well as to
emplacement of the Siberian Traps onto the world’s largest coal basin (Kamo et al.,
2003). However, it is also possible that a combination of mechanisms produced the
δ
13
C shift (Berner, 2002), such as methane release, an increase in the influx of
organic matter due to extinctions on land and in the oceans (Erwin, 1994), and a
reorganization of the global carbon cycle due to a drop in terrestrial and marine
production (Broecker and Peacock, 1999).
On a 100 k.y. time-scale, there are two distinct trends in δ
34
S and δ
13
C
(Figure 4-2BC). At approximately 12m below the mass extinction level, both δ
34
S
and δ
13
C begin to decrease (Figure 4-2BC) from values around +25 to +12‰ VCDT
and from +5 to +3‰ VPDB respectively. Because the pre-extinction δ
34
S and δ
13
C
values are positively correlated, such changes probably do not reflect changing redox
conditions in the shallow ocean, but rather may have resulted from a secular input of
isotopically depleted sulfur and carbon. Beginning at the mass extinction level, δ
34
S
79
increases dramatically while δ
13
C decreases sharply (Figure 4-2BC). In contrast to
the pre-extinction trends, the negatively correlated post-extinction δ
34
S and δ
13
C (10
5

year) trends may have been caused by a change to more reducing conditions, an
interpretation supported by the low Early Triassic O
2
levels modeled by Berner
(2005).

Pre-extinction Short-term Trends
On a 10 k.y. time-scale (Figure 4-2A), there are three negative δ
34
S
excursions below the mass extinction level. Because there are no corresponding
δ
13
C shifts, the ~10‰ δ
34
S anomalies are not likely caused fluctuating redox
conditions in the shallow ocean, but instead may have resulted from an input of
34
S-
depleted sulfur to the surface oceans. A possible source of anomalous
34
S-depleted
sulfur may have been volcanogenic gasses from the Siberian Traps, the emplacement
of which have been shown to coincide with the End Permian mass extinction
(Mundil et al., 2004). The error range for the date of the main pulse of Siberian Trap
volcanism is ± 300 k.y. (Renne et al., 1995), suggesting that the δ
34
S excursions that
begin approximately ~100 k.y. below the mass extinction level are in agreement with
the age of the Siberian Traps.
Modern oceans are very resistant to sulfur isotopic change because of the
long residence time of sulfate (~10 m.y.). However, if a sufficiently large fraction of
the ocean were anoxic during the P-T transition, then continued burial of pyrite may
have resulted in lower sulfate concentrations, as would the massive burial of sulfate
80
evaporites well-known from the Permian Period (e.g., Fischer, 1964). With
decreased oceanic sulfur residence times and on short time-scales, massive inputs of
sulfur to the surface ocean would strongly affect the δ
34
S composition of surface
ocean sulfate (which would be recorded in shallow water carbonates as CAS), with
minimal effect on the ocean as a whole. To test this hypothesis, a two-box model
was created to simulate the δ
34
S response of the surface and deep ocean to secular
inputs of sulfur. The model is defined by the following four equations:
() ()
[]
()() []
surface surface deep deep ocean mix
surface P surface surface P E surface E V V W W
surface
R C R C A
R O R O R I R I
dt
dM
− +
− − + = ⎟
⎠
⎞
⎜
⎝
⎛
υ
α α
34

(1)
()
[]()() []
surface deep ocean mix surface P E V W
surface
C C A O O I I
dt
dM
− + − − + = ⎟
⎠
⎞
⎜
⎝
⎛
υ
32

(2)
()() []
() ()
()
deep P deep deep P deep deep surface surface ocean mix
deep
R O R C R C A
dt
dM
α υ − − = ⎟
⎠
⎞
⎜
⎝
⎛
34

(3)
()() []
() deep P deep surface ocean mix
deep
O C C A
dt
dM
− − = ⎟
⎠
⎞
⎜
⎝
⎛
υ
32

(4)
where surface is the surface reservoir, deep is the deep reservoir, M is the total mass
of either
34
S or
32
S in a given reservoir, I is an input of sulfate to either reservoir, O is
an output of sulfur from either reservoir, R is the ratio of
34
S to
32
S in a given
81
reservoir or flux, W refers to weathering, V refers to volcanism, E refers to evaporite,
P refers to pyrite, α is the isotopic fractionation factor of a given reaction, υ
mix
is the
mixing velocity of the ocean, A
ocean
is the surface area of the ocean, and C is the
concentration of sulfate in either reservoir, as given by the following equation:
()
ocean reservoir
reservoir
reservoir
A H
M M
C
32 34
+
=
(5)
where H is the height of a given reservoir. Because the natural abundance of
32
S
(95.02%) is significantly greater than
34
S (4.21%), changes in the fluxes of the less
abundant isotope have a greater effect on the δ
34
S of seawater sulfate; thus, in the
model the isotopic composition of reservoirs and fluxes are not factors in equations 2
and 4. Model parameters were initialized to present-day values (Table 4-1), and
model δ
34
S
reservoir
(CDT) sensitivity to H
reservoir
, υ
mix
, I
W
, I
V
, and α
P
(converted from ∆
notation) were determined by varying C
surface
and C
deep
. H
surface
was set to 200m or
less. A time step of 1 year was used. The amount of sulfur estimated by (Kamo et
al., 2003) to have erupted during the Siberian Traps emplacement (1.6 x 10
17
moles)
was divided into three eruptive periods of 100 years each and entered as I
V
at 5, 25,
and 45 k.y.
For all model runs, δ
34
S
surface
shifts of 10‰ or more were only possible with
C
surface
less than 5 mM (Figure 4-5). Volcanically induced negative δ
34
S
surface
shifts
were made larger by decreasing H
surface
, increasing I
V
, decreasing υmix, and
increasing the fractionation between sulfate and pyrite α
P
. Isotopic rebounds
82

units initial notes

t
step
years 1 See text for explanation of time step
A
ocean
m2 3.61E+14
H
ocean
m 3711
H
surface
m 200
H
deep
m 3511
V
mix
m3/year 4
C
surface
mols/m3 28.4
C
deep
mols/m3 28.4
R
surface
ratio 0.0459 converted from δ
34
S
surface
= 20‰ CDT
R
deep
ratio 0.0459 converted from δ
34
S
deep
= 20‰ CDT
I
V
mols/year 0 See text for explanation of volcanic inputs.
I
W
mols/year 3.12E+12
O
E
mols/year 1.87E+12
Initial value set to 1.87X10
12
. Subsequent values are a
function of sulfate concentration: O
E
=
[(1.87X10
12
)/(V
surrface
C
surface
)
initial
]X(V
surface
C
surface
)
O
P(surface)
mols/year 9.36E+11
Initial value set to 9.36X10
11
. Subsequent values are a
function of sulfate concentration: O
P(surface)
=
[(9.36X10
11
)/(V
surface
C
surface
)
initial
]X(V
surface
C
surface
)
O
P(deep)
mols/year 3E+11
R
V
0.045 converted from δ
34
S
V
= 0‰ CDT
R
W
0.04527 converted from δ
34
S
W
= 6‰ CDT
∆
E
1.002 converted from ∆
E(evaporite-seawater)
= 1.5 ‰
α
P(surface)
0.955 converted from ∆P
(pyrite-seawater)
= -35 ‰
α
P(deep)
0.955 converted from ∆P
(pyrite-seawater)
= -35 ‰

Table 4-1: Model parameters and initial values.

83

Figure 4-5: Model results with conditions that are explained in the text. A. Modern
rate of sulfur input from weathering. B. One quarter of modern rate of sulfur
input from weathering.
84
following a volcanic eruption only returned to pre-eruption δ
34
S
surface
values or
higher when I
W
was lowered to one quarter of the modern flux, otherwise the isotopic
rebound would be large but of a lesser magnitude than the corresponding negative
shift. δ
34
S
deep
(the sulfur isotopic composition of the bulk of the ocean) was
extremely resistant to isotopic change. In order to produce a shift in δ
34
S
deep
of more
than 1 ‰ over the course of an entire model run, C
deep
would have to be less than 2
mM.
Figure 4-5 shows the δ
34
S
reservoir
response to volcanism when H
surface
is set to
200m, C
deep
and C
surface
are equal and allowed to vary, I
W
is set to modern values
(Figure 4-5a) or one quarter of modern values (Figure 4-5b), and ∆
sulfate-pyrite
is set to
-45 ‰. Research in recent marine and non-marine environments has suggested that
∆
sulfate-pyrite
decreases as sulfate concentration decreases (Habicht et al., 2002).
However, at concentrations around 1mM, the data from (Habicht et al., 2002) exhibit
∆
sulfate-pyrite
values that range from -15 to less than -30 ‰. Therefore, -45 ‰ is a
feasible ∆
sulfate-pyrite
value at concentrations under 5 mM. All other parameters are set
to modern values. The results shown in Figure 4-5b best approximate the measured
isotopic data reported here. Because of the configuration of the continents at the end
of the Paleozoic Era, it is difficult to estimate what the weathering flux of sulfur
would have been. It is possible that much of the terrestrial hydrologic cycle was
sequestered in the interior of the supercontinent Pangea; this could have lead to a
decrease in the weathering flux to the oceans. However, a decrease in weathering
85
flux is not necessarily in agreement with a rise in oceanic
87
Sr/
86
Sr reported to have
started in the latest Permian (Korte et al., 2003), but this increase could also have
been caused by decreasing rates of seafloor spreading. It can be said with certainty
that there was less continental margin area exposed to the oceans than there are
today. Therefore, with geologically feasible parameters, the model results support
the hypothesis that 10‰ δ
34
S shifts can be recorded in shallow water carbonates
following massive inputs of sulfur when sulfate concentrations are extremely low.
Fluid inclusion studies from the Upper Permian Delaware Basin suggest that oceanic
sulfate concentration was 18mM (Horita et al., 2002), which is 10 mM lower than
the modern ocean. However, the exact stratigraphic position of the evaporites from
the Delaware Basin is unknown, and their proximity to the P-T boundary remains
uncertain. It is possible that continued deposition of marine evaporites lowered
seawater sulfate concentrations even further by the end of the Permian.

Post-extinction Short-term Trends
Superimposed on the post-extinction δ
34
S rise are 3 to 5‰ oscillations every
2m (20 k.y.) In the 8m above the boundary, each δ
34
S shift is accompanied by an
opposite shift in δ
13
C. The 20 k.y. shifts appear to be too rapid to result from
changes in the whole ocean inventories of carbon and sulfur. A possible explanation
for the negatively correlated shifts is a fluctuating chemocline between oxic and
euxinic water masses. Below the chemocline, euxinic waters would be characterized
by
34
S-depleted sulfide,
34
S-enriched sulfate, and
13
C-depleted carbon. If the
86
chemocline shoaled, carbonates forming under more euxinic conditions should
acquire sulfate with high δ
34
S and carbonate with low δ
13
C. Upon a return to oxic
conditions, remnant sulfide from the euxinic period would oxidize to
34
S-depleted
sulfate, which would be recorded in shallow water carbonates as a decrease in δ
34
S,
while an increase in photosynthesis would drive δ
13
C to higher values.
Geochemical modeling to explore the feasibility of rapid δ
34
S shifts resulting
from a fluctuating chemocline has been performed by Kump et al. (2005). With
decreased atmospheric oxygen, as has been suggested for the P-T transition (Berner,
2005; Huey and Ward, 2005), and increased phosphorous in the oceans, areas of
upwelling could experience “chemocline upward excursions” that could potentially
introduce H
2
S to the shallow ocean and atmosphere, with disastrous consequences
for life (Kump et al., 2005). Biomarkers of anoxygenic sulfur oxidizing bacteria
associated with the P-T boundary support the presence of euxinic conditions in the
photic zone as these bacteria are strict anaerobes that require light and sulfide (Grice
et al., 2005).
The slope of the post-extinction sulfur vs. carbon plot (Figure 4-4) can be
shown to represent the ratio between TCO
2
in the deep ocean to SO
4
in the surface
ocean in a two-layered system:
Let:
(TCO
2
)
i
= (TCO
2
) of surface ocean (i = surface) or deep ocean (i = deep)
(TSO
4
)
i
= (SO
4
) of surface ocean (i = surface) or deep ocean (i = deep)
C
ox
= carbon oxidized in the deep ocean
87
δ
13
C
ox
= δ
13
C of oxidized organic carbon
δ
34
S
red
= δ
34
S of sulfur reduced during organic carbon oxidation
δ
13
C
i
= δ
13
C of surface ocean TCO
2
(i = surface) or deep ocean TCO
2
(i =
deep)
δ
34
S
i
= δ
34
S of surface ocean SO
4
(i = surface) or deep ocean SO
4
(i = deep)
R
ox
= The ratio of sulfur to carbon utilized during oxidation of organic
matter. If little oxygen is present in sinking surface water relative to the amount of
carbon oxidized in the deep ocean, then R
ox
is 0.5 according to the following
reaction:
2 2 4 2
2 2 CO S H SO H C
org
+ ⎯→ ⎯ +
The following mass balance equations are defined:
(1)
ox surface deep
C TCO TCO + = ) ( ) (
2 2

(2)
ox ox deep surface
C R TSO TSO + = ) ( ) (
4 4

(3) ) ( ) ( ) (
13 13
2
13
2 ox ox surface surface deep deep
C C C TCO C TCO δ δ δ + =
(4)
red ox ox deep deep surface surface
S C R S TSO S TSO
34 34
4
34
4
) ( ) ( δ δ δ + =
The following equations are defined:
(5)
surface ox
C C C
13 13 13
δ δ − = ∆ assumed to be -20‰
(6)
deep red
S S S
34 34 34
δ δ − = ∆ assumed to be -40‰
Combining equations (1), (3) and (5):
) ( ) ) (( ) (
13 13 13
2
13
2
C C C C C TCO C TCO
surface ox surface ox deep deep deep
∆ + + − = δ δ δ
88
(7) C
TCO
C
C C
deep
ox
surface deep
13
2
13 13
) (
∆ = − δ δ
Similarly, combining equations (2), (4) and (6):
( ) ( ) S S C R S TSO S C R TSO
deep ox ox deep deep surface ox ox deep
34 34 34
4
34
4
) ( ) ( ∆ + + = + δ δ δ
(8) S
TSO
C R
S S
surface
ox ox
surface deep
34
4
34 34
) (
∆ − = − δ δ
If isotopic changes in sulfur and carbon reflect shoaling of a chemocline, then
gradients in δ
13
C and δ
34
S should be correlated. In a two-layered system, the higher
δ
34
S values and lower δ
13
C values in Figure 4-4 would represent deep ocean δ
34
S and
δ
13
C respectively. Similarly, the lower δ
34
S values and higher δ
13
C values would
represent shallow ocean δ
34
S and δ
13
C respectively. Therefore, the slope of a line fit
through the data would represent the ratio:
surface deep
surface deep
C C
S S
13 13
34 34
δ δ
δ δ
−
−

Which can be re-written using equations (7) and (8) as:
C
S
TSO
TCO
R
surface
deep
ox
13
34
4
2
) (
) (
∆
∆
−
Assume:
2
20‰ -
‰ 40
13
34
=
−
=
∆
∆
C
S
and 5 . 0 =
ox
R (an upper limit)
Then the slope of a line fit through the data in Figure 4-4 can be written as:
surface
deep
TSO
TCO
) (
) (
4
2
−
89
Sulfate in modern oceans is approximately 13 times more abundant than
TCO
2
resulting in TCO
2
:SO
4
values of about 0.08 (Brown et al., 1989). However,
the slope of the post-extinction sulfur vs. carbon plot is -2, suggesting that the P-T
oceans contained twice as much carbon than sulfur. The abundance of carbon
relative to sulfur could explain why δ
34
S of seawater sulfate is much more variable
than the δ
13
C of TCO
2
across the P-T boundary. The TCO
2
:SO
4
ratio could have
been driven to such a high value by decreasing the concentration of SO
4
in the
oceans, increasing TCO
2
in the oceans, or by a combination of the two.

Implications for the End Permian Mass Extinction
Because the three ~10‰ δ
34
S excursions begin ~9m below the mass
extinction level, whatever caused the excursions most likely did not immediately
cause the mass extinction. Intense volcanism from the Siberian Traps may explain
the input of
34
S-depleted sulfur to the oceans as well as the gradual decline of δ
13
C
that is recorded below the mass extinction level. Emplacement of the Siberian Traps
through an extensive coal deposit (Kamo et al., 2003) may have provided even more
34
S-depleted sulfur in the from the oxidation of abundant pyrite contained in the
coals. It is possible that along with
34
S-depleted sulfur, large quantities of
volcanogenic reductants were introduced into the oceans. A large influx of
reductants may have acted to expand euxinia already present in the oceans, and tip
the balance in favor of widespread euxinia. An increase in nutrient supply to the
oceans from volcanic eruptions may have also increased organic matter production
90
and oxygen removal in the deep ocean. The expansion of euxinia into the shallow
ocean and atmosphere is a likely cause of the mass extinction. The link between
euxinia and the mass extinction is strongly supported by the exact correlation
between the sharp increase in δ
34
S, the onset of negatively correlated δ
34
S and δ
13
C
shifts, and the disappearance of a diverse Permian fauna.
The δ
34
S results presented here are similar to those reported by Newton et al.
(Newton et al., 2004) from a P-T boundary section in Italy. The δ
34
S data from Italy
combined with those reported here suggest that the oceanographic conditions
postulated here were at least widespread throughout the Tethys Ocean. However,
biomarker evidence from around the globe suggest that shallow ocean euxinia was a
global phenomenon during the P-T transition(Summons et al., 2006), and not just
restricted to the Tethys Ocean.

Conclusions
Sulfur and carbon isotopic results from Çürük Da ğ suggest that End Permian
mass extinction was the climactic result of a number of geologic factors including
extensive removal of oxygen by burial of oxidized sulfur in marine evaporites during
the Permian and subsequent prolonged deep-ocean anoxia, extensive volcanism from
the Siberian Traps, and the eventual euxinification of the shallow oceans and the
introduction of H
2
S and high levels of CO
2
to the atmosphere. Because δ
34
S values
remain elevated into the Lower Triassic at Çürük Da ğ, it is likely that the
91
environmental conditions that led to the End Permian mass extinction persisted for
some time into the Early Triassic, possibly delaying the biotic recovery.

92
CHAPTER 5: EARLY TRIASSIC SULFUR ISOTOPES FROM THE
WESTERN UNITED STATES
Abstract
Sulfur isotopic results from CAS and evaporites from the western U.S. show
that following the end Permian mass extinction, seawater δ
34
S remained elevated
(between +30 and +38 ‰ VCDT) for about five million years. The prolonged
interval of elevated δ
34
S is indicative of a long period of increased pyrite burial
under anoxic conditions. The presence in basinal rocks of lithologic features
characteristic of deposition under anoxic conditions, coupled with a lack of those
features in more proximal shelf sediments, suggests that the euxinia that likely
caused the end Permian mass extinction retreated to the deep ocean for most of the
Early Triassic.
Geologic Setting
Dinwoody Formation
The Dinwoody Formation represents the earliest Triassic transgression in
North America, and spans the Griesbachian into the Dienerian (Carr and Paull, 1983)
and is here considered to be Induan. At the localities studied here, the Dinwoody
Formation is overlain by the Woodside Formation, although at localities in Idaho and
Nevada not studied here, the Dinwoody Formation spans the entire Dienerian and is
overlain by the Thaynes Formation (Carr and Paull, 1983). The Dinwoody
93
Formation is a mixed carbonate-siliciclastic succession, with characteristic brown-
weathering limestones deposited in a shallow marine setting (Paull et al., 1989).
However, at the Black Tail Creek, Montana locality, the brown limestones of the
Dinwoody are nearly indistinguishable from those of the underlying Permian Gerster
Formation, and have to be differentiated based on fossil content.

Woodside Formation
The Woodside Formation is a predominantly siliciclastic marginal marine
succession representing marine regression toward the end of the Induan (Carr and
Paull, 1983). The Woodside Formation is overlain by the Thaynes Formation. At
the localities studied, the base of the Woodside Formation was placed at the start of
the first largely siliciclastic interval following the more-mixed limestone-siliciclastic
units of the Dinwoody Formation.

Thaynes Formation
The Thaynes Formation is a mixed carbonate-siliciclastic succession
representing a marine transgression during the Smithian to Spathian (Carr and Paull,
1983). In this study, the base of the Thaynes Formation was taken to be the first
prominent limestone unit following the Woodside, which contained the Meekoceras
ammonoid fauna at both localities. Thus, the base of the Thaynes is distinctly
Smithian in age. The start of Spathian deposition within the Thaynes is less
precisely known in the absence of biostratigraphic data. The early Spathian was a
94
time of marine regression on the North American coast, and it is likely that the
Smithian-Spathian boundary is located within the large siliciclastic intervals between
the Meekoceras-bearing limestone units and the darker-weathering, prominent
limestone units of the upper Thaynes Formation. The limestones of the upper
Thaynes Formation include a subsequent marine transgression during mid-Spathian
time (Carr and Paull, 1983). Therefore, for the purposes of this study, the Smithian-
Spathian boundary was placed at the base of the upper, dark-weathering limestone
interval at Hidden Pasture, Montana and Cascade Springs, Utah. Future work with
strontium isotope chemostratigraphy will help refine this placement.

Union Wash Formation
The Union Wash Formation is a predominantly-shale succession with
prominent limestone intervals and represents an Early Triassic transgression onto
underlying Permian strata and a subsequent regression (Stone et al., 1991). The base
of the Union Wash Formation is marked by the Smithian Meekoceras ammonoid
fauna. Based on conodont biostratigraphy, the Union Wash is believed to range up
through the Spathian, possibly even the Early Anisian (Stone et al., 1991).
The calcareous shales and limestones of the Union Wash Formation contain
abundant pyrite. Following the conclusions presented in Chapter 3, the analyses
from the Union Wash Formation are excluded from the sulfur isotope
chemostratigraphy of the western United Sates presented in this chapter. However, a
number of important sedimentological characteristics of the Union Wash Formation
95
warrant the formation’s inclusion in this text. The Union Wash Formation, in
particular at Union Wash proper, consists largely of black, organic-rich, pyritic shale,
the type often interpreted to represent deposition under anoxic conditions (Woods
and Bottjer, 2000). The Union Wash Formation at Darwin Hills contains hundreds
of meters of limestone that contain what have been interpreted to be sea-floor
precipitated, formerly-aragonite fans, that may also represent deposition under
anoxic conditions (Woods et al., 1999). Strontium isotopic results presented in
Appendix B fall well-within the expected range of values for the Spathian, and thus
argue against a late diagenetic origin for the precipitates, although an early, marine
diagenetic origin cannot be ruled out. Lastly, the Union Wash Formation records an
interval of time spanning the Smithian through the Spathian, and contains both pyrite
and CAS that can be used for a detailed sulfur isotopic investigation. I plan to use
the mixing models presented in Chapter 3 during future investigations of the Union
Wash in order to determine the actual CAS δ
34
S signal recorded in the pyritic
carbonates of the Union Wash.

Moenkopi Formation
The Moenkopi Formation and its Timpoweap Conglomerate, Lower Red,
Virgin Limestone, Middle Red and Shnabkaib Members were introduced in Chapter
2. Here I discuss the Smithian marine deposition within the Moenkopi formation
that is recorded by the Black Dragon and Sinbad Limestone Members in Utah.
96
The two Sinbad Limestone Member localities studied here are located in the
San Rafael Swell area of Utah. In the San Rafael Swell area, the Sinbad Limestone
Member is predominantly oolitic and silty dolomite with interbedded evaporite
deposits, and thus represents a marginal marine environment. At the Road Cut, San
Rafael Swell locality, the measured section includes the upper part of the underlying
Black Dragon Member, which consists of siltstone and shale with minor carbonate.
Together, the Black Dragon and Sinbad Limestone Members represent a marine
transgression and are followed by a regression recorded by the deltaic Torrey
Member (Blakey, 1974).
The occurrence of the Meekoceras ammonoid fauna dates the Sinbad
Limestone Member as Smithian (Blakey, 1974). The age of the underlying Black
Dragon Member is much less certain (Blakey, 1974), but because the samples from
the Black Dragon Member were taken only a few meters from the base of the Sinbad
Limestone Member, they are treated here as Smithian in age as well; this correlation
can be tested by future strontium isotopic chemostratigraphy.

Results
Sulfur chemostratigraphic results from the Western US localities are given in
Figures 5-1 through 5-9 and Appendix A. Results from the Union Wash and Darwin
Hills localities are not given because of the abundance of pyrite in samples from
97

Figure 5-1: Sulfur isotope chemostratigraphy and generalized lithology from the
Batten and Stokes, San Rafael Swell, Utah locality.
98

Figure 5-2: Sulfur isotope chemostratigraphy and generalized lithology from the
Beyond Lost Cabin, Nevada locality.
99

Figure 5-3: Sulfur isotope chemostratigraphy and generalized lithology from the
Blacktail Creek, Montana locality.
100

Figure 5-4: Sulfur isotope chemostratigraphy and generalized lithology from the
Blue Diamond, Nevada locality.
101

Figure 5-5: Sulfur isotope chemostratigraphy and generalized lithology from the
Cascade Springs, Utah locality.
102

Figure 5-6: Sulfur isotope chemostratigraphy and generalized lithology from the
Grasshopper Creek, Montana locality.
103

Figure 5-7: Sulfur isotope chemostratigraphy and generalized lithology from the
Hidden Pasture, Montana locality.
104

Figure 5-8: Sulfur isotope chemostratigraphy and generalized lithology from the
Rainbow Gardens, Nevada locality.
105

Figure 5-9: Sulfur isotope chemostratigraphy and generalized lithology from the
Road Cut, San Rafael Swell, Utah locality.

106
those localities. The combined results are plotted versus stage in Figure 5-10, with
evaporite data from Holser (1984) for comparison.
δ
34
S
CAS
results from the Induan stage show a range of values from 20-36‰,
averaging (28‰), whereas the published evaporite-based data average ~10‰. The
one dolostone CAS result is 20‰, while the limestones average 29‰.
Smithian δ
34
S
CAS
ranges from 14 to 35‰, while the published evaporite-
based data ranges from 7 to 27‰, with average values ~13‰. CAS results from this
study exhibit an average of 21‰, with a limestone δ
34
S average of 32‰, and a
dolostone average of 18‰.
Spathian δ
34
S results include analyses from limestones, dolostones and
evaporites and range from 25 to 38‰, with an average value of 30‰. Limestone
CAS results average 33‰, while dolostones and evaporites separately average 28‰.
Previously-published evaporite-based data range from 12 to 35‰, and average 27‰
(Holser et al., 1988).

Discussion
Evaporites, Dolostones and Limestones
A number of important observations can be made from the Western U.S.
sulfur isotopic data. The observations regarding Spathian limestones, dolostones and
evaporites reported in Chapter 2 apply also to the rest of the Early Triassic. With
little exception, limestones are always
34
S-enriched relative to coeval dolostones and
107

Figure 5-10: Early Triassic sulfur isotopic results separated by stage. Data points
are rounded to the nearest whole permil value.

108
evaporites. Possible explanations for the observed relationship between dolostones
and evaporites were discussed in Chapter 2 and in Marenco et al. (In Review).
Regardless of the mechanism, the results from Chapter 2 and those from the rest of
the Early Triassic would suggest that limestone CAS is the best proxy for the
isotopic composition of seawater sulfate.
Previous authors have discussed the sometimes-large range of δ
34
S values
exhibited by coeval evaporites, and have concluded that average values should be
used to describe the actual δ
34
S value of seawater sulfate at any one time (e.g.,
Claypool et al., 1980; Ayora et al., 1994). A large range of δ
34
S values is observed
in limestones as well. However, because of a lack of fine-scale (sub-stage)
correlation between multiple localities, it is difficult to ascribe the observed range of
δ
34
S values to actual sub-stage fluctuations in seawater sulfate δ
34
S. For the reasons
discussed in Chapter 4 and in Marenco et al. (In Prep.), it is reasonable to suggest
that the Early Triassic may have been a time when seawater sulfate exhibited
extreme fluctuations in δ
34
S on the order of 10
4
years. However, in the absence of
sub-stage resolution, average δ
34
S values from limestones will be used to interpret
long-term changes in seawater sulfate δ
34
S.

The Early Triassic δ
34
S rise
As previous authors have noted (e.g., Claypool et al., 1980; Wilgus, 1981;
Holser and Magaritz, 1987; Holser et al., 1988; Kampschulte and Strauss, 2004),
109
δ
34
S increases from lower values in the Induan to its highest Early Triassic values in
the Spathian. However, the CAS data indicate that the increase to high δ
34
S values
occurred much earlier than previously thought. Whereas the evaporite-based data
suggest that seawater sulfate had an isotopic composition of about 10‰ during the
Induan, the limestone CAS data reveal that seawater was closer to 29‰ during the
Induan.
The evaporite-based data also show that δ
34
S changed little between the
Induan and the Smithian, but experienced a large increase from the Smithian to the
Spathian. The limestone CAS data show an opposite trend, where the largest
increase in δ
34
S occurs between the Induan and the Smithian, while the Smithian and
Spathian exhibit similar average δ
34
S values. Because of the considerable duration
of the Spathian stage (~3 m.y., Lehrmann et al., 2006; Ovtcharova et al., 2006), the
limestone CAS results suggest that oceanic sulfate δ
34
S was elevated for millions of
years after the End Permian mass extinction.

Biotic recovery from the End Permian mass extinction
Elevated seawater sulfate δ
34
S during the bulk of the Early Triassic suggests
an extended interval of increased pyrite burial. Prolonged oceanic anoxia and
possible resultant intervals of H
2
S and CO
2
toxicity (Chapter 4; Knoll et al., 1996;
Grice et al., 2005; Kump et al., 2005; Knoll et al., 2007; Marenco et al., In Prep.)
may have acted to delay the biotic recovery from the End Permian mass extinction.
110
However, while the highest δ
34
S values are found during the Smithian and
Spathian, most of the sedimentological evidence for shallow-ocean anoxia has been
reported from the Induan (Wignall and Hallam, 1992; Wignall and Twitchett, 1996).
This apparent paradox can be explained by the long residence time of sulfate in the
oceans. The geochemical modeling in Chapter 4 showed that even with low oceanic
sulfate levels, the bulk of the ocean is extremely resistant to isotopic change, and
following any perturbation of the sulfur cycle, the surface ocean rapidly re-
equilibrates with the deep ocean. Therefore, if the shallow ocean were well-
oxygenated while the deep ocean remained anoxic, shallow marine sediments would
record the high δ
34
S values of the deep ocean. Consequently, during the Smithian
and Spathian, an absence of diagnostic sedimentary features for anoxia means that
the shallow ocean may have been well-oxygenated, while the high δ
34
S values
suggest that the deep ocean remained anoxic. It is likely that oceanic anoxia
expanded onto the continental shelves during the immediate Permo-Triassic
transition, but receded back into the deep ocean following the Induan; this
interpretation is supported by the sedimentological evidence for shallow-water
anoxia being found only near the P-T boundary (Wignall and Hallam, 1992; Wignall
and Twitchett, 1996), while such evidence in deep-water sediments is found
throughout the Early Triassic (Isozaki, 1997). The occurrence of pyritic, organic-
rich shales and calcium-carbonate fans within the basinal units of the Union Wash
Formation, coupled with the non-pyritic, non-organic-rich, and more fossiliferous
limestones and siliciclastics of the coeval Moenkopi Formation, argue in favor of the
111
relegation of anoxia to the deep ocean. If such a scenario were true, then incursions
of H
2
S and CO
2
onto the continental shelves would have decreased through time
following the earliest Triassic. Furthermore, repeated incursions of anoxic deep-
ocean CO
2
need not be invoked to explain the multiple Early Triassic carbon isotopic
excursions of Payne et al. (2004), but can better be explained by a prolonged interval
of volcanism and resultant volcanogenic CO
2
, as suggested by Payne and Kump
(2007).

Conclusions
The Early Triassic was a five million year period of elevated seawater sulfate
δ
34
S. The high δ
34
S composition of seawater sulfate during the Early Triassic was
likely caused by a prolonged interval of heightened pyrite burial in anoxic deep
water. The Early Triassic sedimentary rocks of the Western U.S. include basinal
shales and slope carbonates with sedimentary structures indicative of burial under
anoxic conditions, as well as strata deposited in shallower water settings that lack
abundant lithologic evidence for anoxia. Therefore, it is likely that the euxinia that
led to the End Permian mass extinction in shallow-water and terrestrial settings
(Chapter 4; Grice et al., 2005; Kump et al., 2005; Riccardi et al., 2006; Marenco et
al., In Prep.) was relegated to the deep ocean during the bulk of the Early Triassic.

112
CHAPTER 6: CONCLUSIONS
The Fidelity of the CAS Signal
Summary
The results presented in this work support the continued use of CAS as a
proxy for geobiological, chemostratigraphic and paleoenvironmental study. The
correlation between δ
34
S and depositional setting has a number of important
implications for multiple geoscience disciplines. For the sulfur chemostratigrapher,
the results presented in Chapter 2 suggest that the field should move away from
relying on evaporites as the rock phase of choice for studying ancient seawater
sulfate δ
34
S. Instead, the much more abundant limestones seem to be more robust
indicators of the isotopic composition of seawater sulfate.
For the sedimentary petrologist, the difference in the δ
34
S range between
limestones and dolostones presents issues pertaining to the timing and nature of
dolomitization. If the patterns uncovered in Chapter 2 are true for most ancient
sedimentary basins, then perhaps lower δ
34
S in dolostones compared to coeval distal
limestones could be an indicator of early dolomitization. If dolostones and
limestones exhibit strong overlap in δ
34
S, then perhaps the dolostones were
originally limestones that underwent dolomitization much later in their diagenetic
history.
The positive relationship between Ca/Mg and [CAS] raises an important
question about our understanding of dolomitization. It is generally accepted that the
113
reason that dolomite does not precipitate from seawater is the high abundance of
sulfate in seawater, which is believed to be a dolomite inhibitor based on laboratory
experiments (e.g., Tucker et al., 1990). However, the results from Chapter 2 suggest
that dolomites can contain on the order of 10
3
ppm sulfate. One interpretation is that
our understanding of sulfate as a dolomite inhibitor is wrong. However, another
interpretation could be that the dolomitization occurred after all of the sulfate had
been precipitated as gypsum or calcite CAS. The remaining low-sulfate fluid may
have been Mg-rich enough for dolomitization to occur, in a process that did not
remobilize the CAS sulfate. However, because of the order of mineral formation in
the Usiglio sequence (carbonate minerals form before sulfate minerals; Usiglio,
1849), it is unclear why the dolomites would exhibit a Rayleigh-type distillation
signal if sulfate were not remobilized during dolomitization.
For the paleo-chemical oceanographer, the results presented in Chapter 2 may
suggest the presence of a shallow to deep δ
34
S gradient during the Early Triassic. If
this were true, then the assumption that Earth’s oceans have had high sulfate
abundance since the oxygenation of the atmosphere should be questioned. However,
regardless of the results I presented in Chapter 4, I believe that a Rayleigh-type
distillation effect following the precipitation of gypsum is the most parsimonious
explanation, considering that some of evaporite-based δ
34
S values from the global
compilation are as high as those from limestones (Figure 5-10).

114
Future Work
There are still many questions pertaining to the formation and preservation of
CAS. In particular, I plan to perform a quantitative study of how sulfate becomes
incorporated into various limestone phases. CAS workers have operated under the
assumption that CAS abundance is proportional to seawater sulfate abundance.
However, nobody has yet quantified how CAS abundance relates to seawater sulfate
abundance. Staudt and Schoonen (1995) showed that modern bulk marine micrite
had CAS concentrations on the order of 10
4
ppm while biogenic agragonite phases
had up to 10
5
ppm. I plan to conduct experiments growing carbonate-precipitating
organisms in fluids of varying sulfate concentration in an effort to quantify the
incorporation of sulfate into the carbonate lattice.

Permo-Triassic Sulfur Isotopes
Summary
The Permo-Triassic transition was characterized by an extremely anomalous
sulfur cycle, as evidenced by the δ
34
S fluctuations observed at Permo-Triassic
boundary sections (Chapter 4; Newton et al., 2004; Riccardi et al., 2006) and the
elevated δ
34
S values throughout the entire Early Triassic. The results from Çürük
Da ğ suggest that the deep oceans were anoxic for much of the Late Paleozoic, and
that by the end of the Paleozoic, they had become euxinic and spread onto the
continental shelves.
115
The elevated δ
34
S values during the Early Triassic suggest that deep ocean
anoxia persisted for millions of years following the mass extinction. However, a
lack of anoxia-diagnostic sedimentary features in shallow settings suggest that the
shallow ocean was not anoxic for much of the Early Triassic.

Future Work
Much more work is needed to understand the dynamics of the Permo-Triassic
sulfur cycle. I plan to further refine the Early Triassic δ
34
S curve presented in this
work with higher-resolution sampling and strontium stratigraphic and
biostratigraphic controls. In particular, I plan to revisit the Union Wash Formation
samples to perform a combined pyrite-sulfide and CAS study, using the mixing
models presented in Chapter 3 to better approximate the true seawater sulfate
isotopic composition.
Another goal of mine is to visit other Permo-Triassic boundary sections to
see if the combined sulfur and carbon variations observed at Çürük Da ğ are an
isolated phenomenon. There have now been a number of Permo-Triassic boundary
CAS studies (Newton et al., 2004; Kaiho et al., 2006; Riccardi et al., 2006), but the
Çürük Da ğ study is the only one to investigate δ
34
S and δ
13
C from the same bulk
samples.
116
BIBLIOGRAPHY

Ayora, C., Taberner, C., Pierre, C. and Pueyo, J.J., 1994. Refining the δ
34
S and δ
18
O
values of sulphate in ancient oceans. Mineralogical Magazine, 58A: 32-33.

Baud, A., Atudorei, V. and Sharp, Z., 1996. Late Permian and Early Triassic
evolution of the northern Indian margin; carbon isotope and sequence
stratigraphy. Geodinamica Acta, 9(2-3): 57-77.

Baud, A., Cirilli, S. and Marcoux, J., 1997. Biotic response to mass extinction: The
lowermost Triassic microbialites. Facies, 36: 238-242.

Becker, L., Poreda, R.J., Hunt, A.G., Bunch, T.E. and Rampino, M., 2001. Impact
event at the Permian-Triassic boundary; evidence from extraterrestrial noble
gases in fullerenes. Science, 291(5508): 1530-1533.

Becker, L., Poreda, R.J.B.A.R., Pope, K.O.H.T.M. and Nicholson, C.I.R., 2004.
Bedout; a possible end-Permian impact crater offshore of northwestern
Australia. Science, 304(5676): 1469-1476.

Berner, R.A., 2002. Examination of hypotheses for the Permo-Triassic boundary
extinction by carbon cycle modeling. Proceedings of the National Academy
of Sciences (USA), 99(7): 4172-4177.

Berner, R.A., 2005. The carbon and sulfur cycles and atmospheric oxygen from
Middle Permian to Middle Triassic. Geochimica et Cosmochimica Acta,
69(13): 3211-3217.

Berner, R.A. and Raiswell, R., 1983. Burial of organic carbon and pyrite sulfur in
sediments over Phanerozoic time; a new theory. Geochimica et
Cosmochimica Acta, 47(5): 855-862.

Bissell, H.J., 1970. Petrology and petrography of lower Triassic marine carbonates of
southern Nevada (U.S.A.). In: H.M.E. Schuermann (Editor), International
sedimentary petrographical series, pp. 27.

Blakey, R.C., 1974. Stratigraphic and Depositional Analysis of the Moenkopi
Formation, Southeastern Utah. Bulletin - Utah Geological and Mineral
Survey, 104: 1-81.

117
Bosscher, H. and Schlager, W., 1993. Accumulation rates of carbonate platforms.
Journal of Geology, 101(3): 345-355.

Bowring, S.A. et al., 1998. U/Pb zircon geochronology and tempo of the end-
Permian mass extinction. Science, 280(5366): 1039-1045.

Broecker, W.S. and Peacock, S., 1999. An ecological explanation for the Permo-
Triassic carbon and sulfur isotope shifts. Global Biogeochemical Cycles,
13(4): 1167-1172.

Brown, J. et al., 1989. Ocean chemistry and deep-sea sediments. Pergamon Press,
134 pp.

Burchfiel, B.C., Cameron, C.S. and Royden, L.H., 1997. Geology of the Wilson
Cliffs-Potosi Mountain area, southern Nevada. International Geology
Review, 39(9): 830-854.

Burdett, J.W., Arthur, M.A. and Richardson, M., 1989. A Neogene seawater sulfur
isotope age curve from calcareous pelagic microfossils. Earth and Planetary
Science Letters, 94(3-4): 189-198.

Canfield, D. and Thamdrup, B., 1994. The Production of 34S-depleted sulfide during
bacterial disproportionation of elemental sulfur. Science, 266: 1973-1975.

Canfield, D.E., Habicht, K.S. and Thamdrup, B., 2000. The Archean sulfur cycle and
the early history of atmospheric oxygen. Science, 288(5466): 658-661.

Carr, M.D., McDonnell-Canan, C. and Weide, D.L., 2000. Geologic map of the Blue
Diamond SE Quadrangle, Nevada. Nevada Bureau of Mines and Geology,
Reno, NVLas Vegas, NV.

Carr, T.R. and Paull, R.K., 1983. Early Triassic stratigraphy and paleogeography of
the Cordilleran Miogeocline. In: M.W. Reynolds and E.D. Dolly (Editors),
Rocky Mountain paleogeography symposium 2; Mesozoic paleogeography of
the West-Central United States. Society of Economic Paleontologists and
Mineralogists Rocky Mountain Section, pp. 39-55.

Clark, D.L., 1979. Permian-Triassic boundary; Great Basin conodont perspective. In:
D.L. Clark (Editor), Great Basin stratigraphy and paleontology. Brigham
Young University Department of Geology, pp. 85-90.

118
Claypool, G.E., Holser, W.T., Kaplan, I.R., Sakai, H. and Zak, I., 1980. The age
curves of sulfur and oxygen isotopes in marine sulfate and their mutual
interpretation. Chemical Geology, 28(3-4): 199-260.

Corsetti, F.A., Baud, A., Marenco, P.J. and Richoz, S., 2005. Summary of Early
Triassic carbon isotope records. Comptes Rendus Palevol, 4(6-7): 405-418.

Corsetti, F.A., Kidder, D.L. and Marenco, P.J., 2006. Trends in oolite dolomitization
across the Neproterozoic-Cambrian boundary: A case study from Death
Valley, California. Sedimentary Geology, 191: 135-150.

Cortecci, G., Reyes, E., Berti, G. and Casati, P., 1981. Sulfur and oxygen isotopes in
Italian marine sulfates of Permian and Triassic ages. Chemical Geology,
34(1-2): 65-79.

Dalton, R., 2004. Comet impact theory faces repeat analysis. Nature, 431: 1027.

Dickson, J.A.D., 1966. Carbonate identification and genesis as revealed by staining.
Journal of Sedimentary Petrology, 36(2): 491-505.

Erwin, D.H., 1993. The Great Paleozoic Crisis: Life and Death in the Permian.
Critical Episodes in Earth History Series. Columbia University Press, New
York, 327 pages pp.

Erwin, D.H., 1994. The Permo-Triassic extinction. Nature, 367(6460): 231-236.

Farley, K.A. and Mukhopadhyay, S., 2001. An extraterrestrial impact at the Permian-
Triassic boundary? Science, 293: 2343a.

Fischer, A.G., 1964. Brackish oceans as the cause of the Permo-Triassic marine
faunal crisis. In: A.E.M. Nairn (Editor), Problems in Palaeoclimatology.
Interscience, London, pp. 566-574.

Glickson, A., 2004. Comment on "Bedout: A Possible End-Permian Impact Crater
Offshore of Northwestern Australia". Science, 306(5696): 613.

Grassineau, N.V., Mattey, D.P. and Lowry, D., 2001. Sulfur isotope analysis of
sulfide and sulfate minerals by continuous flow-isotope ratio mass
spectrometry. Analytical Chemistry, 73: 220-225.

Grice, K. et al., 2005. Photic zone euxinia during the Permian-Triassic superanoxic
event. Science, 307(5710): 706-709.

119
Habicht, K.S., Gade, M., Thamdrup, B., Berg, P. and Canfield, D., 2002. Calibration
of Sulfate Levels in the Archean Ocean. Science, 298(5602): 2372-2374.

Hoefs, J., 1997. Stable Isotope Geochemistry. Springer-Verlag, 201 pp.

Holser, W.T., 1984. Gradual and abrupt shifts in ocean chemistry during
Phanerozoic time. In: H.D. Holland and A.F. Trendall (Editors), Patterns of
change in Earth evolution. Springer-Verlag, Berlin, Heidelberg, New York,
Tokyo, pp. 123-143.

Holser, W.T. and Kaplan, I.R., 1966. Isotope geochemistry of sedimentary sulfates.
Chemical Geology, 1(2): 93-135.

Holser, W.T. and Magaritz, M., 1984. Chemical events near the Permian-Triassic
boundary, AGU 1984 spring meeting. American Geophysical Union, pp. 297.

Holser, W.T. and Magaritz, M., 1987. Events near the Permian-Triassic boundary.
Modern Geology, 11(2): 155-180.

Holser, W.T., Schidlowski, M., Mackenzie, F.T. and Maynard, J.B., 1988.
Geochemical cycles of carbon and sulfur. In: C.B. Gregor, R.M. Garrels, F.T.
Mackenzie and J.B. Maynard (Editors), Chemical cycles in the evolution of
the Earth. John Wiley and Sons, Inc, pp. 276.

Horita, J., Zimmermann, H. and Holland, H.D., 2002. Chemical evolution of
seawater during the Phanerozoic; implications from the record of marine
evaporites. Geochimica acta Cosmochemica, 66(21): 3733-3756.

Huey, R.B. and Ward, P.D., 2005. Hypoxia, global warming and terrestrial Late
Permian extinctions. Science, 308: 398-401.

Hurtgen, M.T., Arthur, M.A., Suits, N.S. and Kaufman, A.J., 2002. The sulfur
isotopic composition of Neoproterozoic seawater sulfate; implications for a
snowball Earth? Earth and Planetary Science Letters, 203(1): 413-429.

Isozaki, Y., 1997. Permo-Triassic boundary superanoxia and stratified superocean;
records from lost deep sea. Science, 276(5310): 235-238.

Isozaki, Y., 2001. An extraterrestrial impact at the Permian-Triassic boundary?
Science, 293: 2343a.

Kah, L.C., Lyons, T.W. and Frank, T.D., 2004. Low marine sulphate and protracted
oxygenation of the Proterozoic biosphere. Nature, 431: 834-837.
120
Kaiho, K., Chen, Z.-Q., Kawahata, H., Kajiwara, Y. and Sato, H., 2006. Close-up of
the end-Permian mass extinction horizon recorded in the Meishan section,
South China: Sedimentary, elemental, and biotic characterization and a
negative shift of sulfate sulfur isotope ratio. Palaeogeography,
Palaeoclimatology, Palaeoecology, 239: 396-405.

Kaiho, K. et al., 2001. End-Permian catastrophe by a bolide impact; evidence of a
gigantic release of sulfur from the mantle. Geology, 29(9): 815-818.

Kamo, S.L., Czamanske, G.K.A.Y., Fedorenko, V.A.D.D.W. and Trofimov, V.R.,
2003. Rapid eruption of Siberian flood-volcanic rocks and evidence for
coincidence with the Permian-Triassic boundary and mass extinction at 251
Ma. Earth and Planetary Science Letters, 214(1-2): 75-91.

Kampschulte, A., Bruckschen, P. and Strauss, H., 2001. The sulphur isotopic
composition of trace sulphates in Carboniferous brachiopods: implications
for coeval seawater, correlation with other geochemical cycles and isotope
stratigraphy. Chemical Geology, 175: 165– 189.

Kampschulte, A. and Strauss, H., 2004. The sulfur isotopic evolution of Phanerozoic
seawater based on the analysis of structurally substituted sulfate in
carbonates. Chemical Geology, 204: 255-286.

Kaplan, I.R., Emery, K.O. and Rittenberg, S.C., 1963. The distribution and isotopic
abundance of sulphur in recent marine sediments off southern California.
Geochimica et Cosmochimica Acta, 27(4): 297-331.

Kaufman, A.J., Corsetti, F.A. and Varni, M.A., 2007. The effect of rising
atmospheric oxygen on carbon and sulfur isotope anomalies in the
Neoproterozoic Johnnie Formation, Death Valley, USA. Chemical Geology,
237: 47-63.

Knoll, A.H., Bambach, R.K., Canfield, D.E. and Grotzinger, J.P., 1996. Comparative
Earth history and Late Permian mass extinction. Science, 273(5274): 452-
457.

Knoll, A.H., Bambach, R.K., Payne, J.L., Pruss, S. and Fischer, W.W., 2007.
Paleophysiology and end-Permian mass extinction. Earth and Planetary
Science Letters, 256: 295-313.

Korte, C., Kozur, H.W., Bruckschen, P. and Veizer, J., 2003. Strontium isotope
evolution of Late Permian and Triassic seawater. Geochimica et
Cosmochimica Acta, 67(1): 47-62.
121
Krull, E.S. and Retallack, G.J., 2000. d13C depth profiles from Paleosols across the
Permian-Triassic boundary; evidence for methane release. Geological Society
of America Bulletin, 112(9): 1459-1472.

Kump, L.R., Pavlov, A.A. and Arthur, M.A., 2005. Massive release of hydrogen
sulfide to the surface ocean and atmosphere during intervals of oceanic
anoxia. Geology, 33(5): 397-400.

Lehrmann, D.J. et al., 2006. Timing of recovery from the end-Permian extinction;
geochronologic and biostratigraphic constraints from south China. Geology,
34(12): 1053-1056.

Lyons, T.W., Walter, L.M., Gellatly, A.M. and Marini, A.M., 2004. Sites of
anomalous organic remineralization in the carbonate sediments of South
Florida, U.S.A.: The sulfur cycle and carbonate-associated sulfate. In: J.
Amend, K. Edwards and L. T. (Editors), Microbial Sulfur Transformations
Throughout Earth's History: Development, Changes, and Future of the
Biogeochemical Sulfur Cycle: Geological Society of America Special Paper
379. Geological Society of America, Boulder, Colorado, pp. 161-176.

Marcoux, J. and Baud, A., 1986. The Permo-Triassic boundary in the Antalya
Nappes (western Taurides, Turkey). In: G. Cassinis (Editor), Permian and
Permian-Triassic boundary in the South-Alpine segment of the western
Tethys, and additional regional reports. Societa Geologica Italiana, pp. 243-
252.

Marenco, P.J., Corsetti, F.A., Bottjer, D.J., Baud, A. and Kaufman, A.J., 2006. Sulfur
isotope anomalies and the biotic recovery from the End-Permian mass
extinction. In: Q. Yang, Y. Wang and E.A. Weldon (Editors), Second
International Palaeontological Congress. University of Science and
Technology of China Press, Beijing, China, pp. 406.

Marenco, P.J., Corsetti, F.A., Kaufman, A.J. and Bottjer, D.J., In Review.
Environmental and diagenetic variations in carbonate associated sulfate: An
investigation of CAS in the Lower Triassic of the Western U.S.A.
Geochimica et Cosmochimica Acta.

Marenco, P.J. et al., In Prep. Combined Permo-Triassic sulfur and carbon isotope
variations and their implications for the End-Permian mass extinction. Earth
and Planetary Science Letters.

122
Mekhtiyeva, V.L., 1974. Sulfur isotopic composition of fossil molluscan shells as an
indicator of hydrochemical conditions in ancient basins. Geochemistry
International, 11(6): 1188-1192.

Montanez, I.P., Banner, J.L., Osleger, D.A., Borg, L.E. and Bosserman, P.J., 1996.
Integrated Sr isotope variations and sea-level history of Middle to Upper
Cambrian platform carbonates; implications for the evolution of Cambrian
seawater (super 87) Sr/ (super 86) Sr. Geology, 24(10): 917-920.

Mundil, R., Ludwig, K.R., Metcalfe, I. and Renne, P.R., 2004. Age and timing of the
Permian mass extinctions: U/Pb dating of closed-system zircons. Science,
305: 1760-1763.

Newton, R., Pevitt, P., Wignall, P.B. and Bottrell, S., 2004. Large shifts in the
isotopic composition of seawater sulphate across the Permo-Triassic
boundary in northern Italy. Earth and Planetary Science Letters, 218: 331-
345.

Nielsen, H. and Ricke, W., 1964. Schwefel-isotopen verhältnisse von evaporiten aus
deutschland; Ein beitrag zur kenntnis von δ34S im meerwasser-sulfat.
Geochimica et Cosmochimica Acta, 28: 577-591.

Nielsen, J. and Shen, Y., 2004. Evidence for sulfidic deep water during the Late
Permian in the East Greenland Basin. Geology, 32(12): 1037-1040.

Ovtcharova, M. et al., 2006. New Early to Middle Triassic U–Pb ages from South
China: Calibration with ammonoid biochronozones and implications for the
timing of the Triassic biotic recovery. Earth and Planetary Science Letters,
243(3-4): 463-475.

Palmer, A.R. and Geissman, J.W., 1999. 1999 Geologic time scale, Geological
Society of America.

Paull, R.A., Paull, R.K. and Kramer, B.R., 1989. Depositional history of Lower
Triassic rocks in southwestern Montana and adjacent parts of Wyoming and
Idaho. In: D.E. French and R.F. Grabb (Editors), Montana Centennial
Edition, Field Conference Guidebook. Montana Geological Society, pp. 69-
90.

Payne, J.L. and Kump, L.R., 2007. Evidence for recurrent Early Triassic massive
volcanism from quantitative interpretation of carbon isotope fluctuations.
Earth and Planetary Science Letters, 256: 264-277.

123
Payne, J.L., Lehrmann, D.J.W.J., Orchard, M.J.S.D.P. and Knoll, A.H., 2004. Large
perturbations of the carbon cycle during recovery from the end-Permian
extinction. Science, 305(5683): 506-509.

Paytan, A., Kastner, M., Campbell, D. and Thiemens, M.H., 1998. Sulfur isotopic
composition of Cenozoic seawater sulfate. Science, 282(5393): 1459-1462.

Raab, M. and Spiro, B., 1991. Sulfur isotopic variations during seawater evaporation
with fractional crystallization. Chemical Geology; Isotope Geoscience
Section, 86(4): 323-333.

Raup, D.M., 1979. Size of the Permo-Triassic bottleneck and its evolutionary
implications. Science, 206(4415): 217-218.

Renne, P.R. et al., 2004. Is Bedout an impact crater? Take 2. Science, 306: 610-611.

Renne, P.R., Zhang, Z., Richards, M.A., Black, M.T. and Basu, A.R., 1995.
Synchrony and causal relations between Permian-Triassic boundary crises
and Siberian flood volcanism. Science, 269(5229): 1413-1416.

Retallack, G.J., 1995. Permian-Triassic life crisis on land. Science, 267: 77-80.

Riccardi, A.L., A., A.M. and R., K.L., 2006. Sulfur isotopic evidence for chemocline
upward excursions during the end-Permian mass extinction. Geochimica et
Cosmochimica Acta, 70: 5740-5752.

Richoz, S., 2004. Stratigraphie et variations isotopiques du carbone dans le Permien
supérieur et le Trias inférieur de la Néotéthys (Turquie, Oman et Iran),
University of Lausanne, Lausanne, 251 pp.

Stanley, S.M. and Yang, X., 1994. A double mass extinction at the end of the
Paleozoic Era. Science, 266(5189): 1340-1344.

Staudt, W.J. and Schoonen, M.A.A., 1995. Sulfate incorporation into sedimentary
carbonates. In: M.A. Vairavamurthy and M.A.A. Schoonen (Editors),
Geochemical transformations of sedimentary sulfur: American Chemical
Society Symposium Series. American Chemical Society, pp. 332-345.

Stone, P., Stevens, C.H. and Orchard, M.J., 1991. Stratigraphy of the Lower and
Middle(?) Triassic Union Wash Formation, east-central California. USGS
Bulletin: 26.

124
Strauss, H., 1997. The isotopic composition of sedimentary sulfur through time.
Palaeogeography, Palaeoclimatology, Palaeoecology, 132(1-4): 97-118.

Summons, R. et al., 2006. Molecular evidence for radical changes in ocean
chemistry, globally, across the Permian Triassic boundary. In: Q. Yang, Y.
Wang and E.A. Weldon (Editors), The Second International Palaeontological
Congress, Beijing, China, pp. 4-6.

Tucker, M.E., Wright, V.P. and Dickson, J.A.D., 1990. Carbonate sedimentology.
Blackwell Sci. Publ., Oxford, 482 pp.

Usiglio, J., 1849. Analyse de I'eau de la Méditerranée sur les côtes de France.
Annalen der Chemie, 27: 172-191.

Wignall, P.B. and Hallam, A., 1992. Anoxia as a cause of the Permian/Triassic mass
extinction; facies evidence from northern Italy and the Western United States.
Palaeogeography, Palaeoclimatology, Palaeoecology, 93(1-2): 21-46.

Wignall, P.B. and Newton, R., 2003. Contrasting deep-water records from the Upper
Permian and Lower Triassic of South Tibet and British Columbia: evidence
for a diachronous mass extinction. Palaios, 18(2): 153-167.

Wignall, P.B., Thomas, B., Willink, R. and Watling, J., 2004. Is Bedout an impact
crater? Take 1. Science, 306: 609.

Wignall, P.B. and Twitchett, R.J., 1996. Oceanic anoxia and the end Permian mass
extinction. Science, 272(5265): 1155-1158.

Wilgus, C.K., 1981. A stable isotope study of Permian and Triassic marine evaporite
and carbonate rocks, Western Interior, U.S.A. Doctoral Thesis, University of
Oregon, Eugene, OR, 109 pp.

Woods, A.D. and Bottjer, D.J., 2000. Distribution of ammonoids in the Lower
Triassic Union Wash Formation (eastern California); evidence for
paleoceanographic conditions during recovery from the end-Permian mass
extinction. Palaios, 15(6): 535-545.

Woods, A.D., Bottjer, D.J., Mutti, M. and Morrison, J., 1999. Lower Triassic large
sea-floor carbonate cements; their origin and a mechanism for the prolonged
biotic recovery from the end-Permian mass extinction. Geology, 27(7): 645-
648.

125
Yin, H., Zhang, K., Tong, J., Yang, Z. and Wu, S., 2001. The Global Stratotype
Section and Point (GSSP) of the Permian-Triassic boundary. Episodes, 24(2):
102-114.

Zharkov, M.A., 1981. History of Paleozoic salt accumulation. Springer Verlag,
Berlin, 316 pp.

126
APPENDIX: LOCALITIES, MEASURED SECTIONS AND DATA

The index to lithologic symbols used in the stratigraphic columns presented
in this text is given in Figure A-1.

Batten and Stokes Locality, San Rafael Swell, Utah (Figure A-2)
From I70 westbound, take exit 129 and head east/north to Buckhorn Draw
Road. Locality is directly east of road near 38°58’27.11”N, 110°39’53”W.

0-0.5 meters
Silty dolostone, light-tan to yellow-weathering with small gastropods and
small bivalves (scarce), oolitic in places with local intraformational conglomerates.

0.5-1 meters
Largely-covered siltstone.

1-3.2 meters
Light tan-weathering dolostone with local crossbedding, intraformational
conglomerate and oolite. Common bivalve molds, some filled with spar cement.
Small (<5cm) interbeds of siltstone. At 2 meters, abundant small bivalves and
microgastropods, mostly filled with spar.
127

Figure A-1: Index of lithologic symbols used in this text. Resistance of beds were
scored on a scale of 1 to 4, where 1 represents a bed that is poorly-exposed
and partially covered, and 4 represents a bed that is ledge-forming.
128

Figure A-2: Stratigraphic column for the Batten and Stokes, San Rafael Swell, Utah
locality (left), and the Road Cut, San Rafael Swell, Utah locality (right).

129
3.2-5 meters
Largely-covered siltstone.

5-10.5 meters
Large ledge-forming carbonate with a somewhat less resistant and more silty
base (from 5 to 5.5 meters). Unit is dolomitic towards base but transitions into
limestone by the top (limestone at 9 meters). Common spar-filled microgastropods
and bivalves. Abundant intraformational conglomerates at 6.5 meters. At 6.75, large
(~60 cm) spar-filled voids. Large bivalves at 7.8 meters. At 8 meters, large bivalves
oriented vertically and in clusters, possibly in life position.

10.5-11 meters
Largely covered siltstone.

11-13 meters
Very platy-weathering calcareous siltstone.

13-14.5 meters
Silty limestone oolite with thin (< 5cm) gypstone interbeds.

14.5-19 meters
Brown-weathering silty dolostone that weathers with large pock marks.
130
Beyond Lost Cabin, Nevada (Figure A-3)
The Virgin Limestone Member of the Moenkopi Formation was measured
about two miles east of Lost Cabin Springs, near 36° 4'55.73"N, 115°37'5.69"W. To
get to the section, take the unimproved dirt road off of westbound 160 just past the
bridge over Lovell Wash, near 36° 0'16.41"N, 115°38'44.88"W. Take the left fork
near 36° 1'43.31"N, 115°37'45.94"W. Take the left fork near 36° 2'31.82"N,
115°38'4.85"W. Take the right fork near 36° 3'1.47"N, 115°38'50.33"W. Take right
fork near 36° 4'19.25"N, 115°38'27.21"W. Base of section is the first carbonate
outcrop above unconformity with Permian Strata. The unconformity is marked by a
conglomerate

0-20 meters
Limestone and shale, with some siltstone and sandstone. From 0 to 3.5
meters, calcareous shale, weathers gray, with thin (few millimeters in thickness)
siltstone interlaminae. From 3.5 to 4.5 meters, a shelly limestone packstone to
grainstone with bivalve shell fragments. From 5 to 8.5 meters, a silty limestone,
weathers gray with horizontal trace fossils on bedding surfaces. From 11 to 12.5
meters, a coarse-grained skeletal packstone with bivalve shell fragments, interbedded
with micritic limestone; hummocky cross stratification. From 13 to 13.5 meters,
sandstone and limey sandstone. From 15.5 to 17, coarse, sandy limestone with zones
131

Figure A-3: Stratigraphic column for the Beyond Lost Cabin, Nevada locality.
132

Figure A-3 (continued): Stratigraphic column for the Beyond Lost Cabin, Nevada
locality.
133
of crinoidal packstone to grainstone and lenses of coarse sand with mud rip-up clasts.
From 18.5 to 20 meters, gray, sandy limestone, massive.

20-33.5 meters
Brown sandstone, massive.

33.5-45.5
Largely-covered shale.

45.5-62.5 meters
Silty limestone, micritic limestone, calcareous mudstone, and mudstone, with
some sandstone. From 47 to 48 meters, sandy limestone, somewhat poorly exposed
with significant horizontal bioturbation on bedding surfaces. From 54.5 to 56
meters, coarse-silty limestone with significant bedding plane bioturbation. From
59.5 to 60 meters, sandy limestone with abundant small bivalve fragments; in this
interval, there is float containing ammonoids.

62.5-70 meters
Poorly exposed calcareous shale and calcareous siltstone.

134
70-88.2 meters
Skeletal limestone and calcareous shale, with minor silty limestone and shale.
From 72.3 to 77.3, calcareous siltstone and calcareous mudstone with two-meter tall
stromatolite bioherms. Bioherms thin laterally to about one meter in hight. From 80
to 83 meters, a coarse-grained limestone with zones of crinoidal grainstone and
zones of mud rip-up clasts. From 87.2 to 88.2, coarse-grained limestone with bivalve
fragments and crinoids.

88.2-104.1 meters
Siltstone with one prominent limestone bed. From 88.2-97.5 meters, poorly
exposed siltstone, with horizontal bioturbation visible on exposed bedding plane at
97.5 meters. At 97.5 meters, a thin (10cm) coarse-grained limestone with bivalves,
microgastropods and crinoid debris.

104.1-118.2 meters
Bioclastic grainstones separated by shale intervals. At 104.1, 20 centimeters
of crinoidal grainstone followed by 30 centimeters of silty limestone. From 104.6 to
105.6, coarse-grained limestone with bivalve fragments and crinoid bits. From 106.1
to 107.5 meters, brown-weathering silty to sandy limestone, excellent marker bed;
contains abundant bivalve fragments and crinoid stars. From 107.5 to 108 meters,
gray-weathering bioclastic (bivalve and star-shaped crinoid debris) grainstone with
cross-bedding, lenticular bedding, and herring-bone cross bedding in places. From
135
109 to 109.5, bioclastic (bivalves and crinoids) grainstone, followed by 1 meter of
finely-laminated micrite and silty limestone, with visible vertical and horizontal
bioturbation. From 112.5 to 113, bioclastic (bivalves and both star-shaped and round
crinoid debris) grainstone. From 116 to 118.2, 20 cm of siltstone and shale followed
by three 50 cm intervals of coarse-bioclastic grainstones (weathers both gray and
brown) separated by thin (~15cm) mud intervals.

118.2-126 meters
Largely-covered shale.

126-146 meters
Bioclastic grainstones and wackestones (at 130, 135.5, 138, 142, and 145.7
meters, pentagonal and round crinoid debris and bivalve fragments), siltstones and
shale. From 126 to 128, bioclastic (pentagonal crinoid debris) wackestone
containing multiple one-meter tall stromatolite bioherms. From 141 to 142 meters,
siltstone and calcareous shale, upper 20 centimeters weathers white and contains
bivalves; bedding planes are heavily bioturbated. From 142 to 144.5, bioclastic
wackestones with clotted thrombolite in middle part; large bivalves and large
horizontal trace fossils on bedding planes. From 145.5 to 145.7, micrite with
bedding obscured due to high degree of bioturbation; large horizontal trace fossils on
bedding surface. At 145.7, thirty centimeters of bivalve grainstone including large
(~10cm) bivalves.
136
146-161.5 meters
Poorly-exposed shale with one prominent bivalve grainstone at 151 meters

161.5-187 meters
Mostly poorly-exposed calcareous shale and siltstone, finely-laminated
micrite and more resistant bioclastic limestones (161.5, 168, 174, 178, and 186.5,
with bivalves and crinoids). Calcareous shale intervals exhibit both vertical and
horizontal bioturbation.

187-224 meters
Poorly-exposed calcareous shale and siltstone with lesser (0.1-1.5 meters
thick) bioclastic (bivalves, gastropods and crinoids) limestones at 205, 216, 218, and
223.5 meters. Section covered after last limestone outcrop.

Blacktail Creek, Montana (Figure A-4)
Near 44°45'10.00"N, 112°17'49.00"W. From I15 take Monida exit to Road
509, then Road 202. Exposure is along road, opposite creek. Because of the dip of
the section relative to the slope, the section is very poorly exposed.
137

Figure A-4: Stratigraphic column for the Blacktail Creek, Montana locality.
138

Figure A-4 (continued): Stratigraphic column for the Blacktail Creek, Montana
locality.
139

0-18.10 meters
Siltstone and shale capped by a thin (10 cm), brown fossiliferous limestone
unit. Fossils include large brachiopods, large scallops and large gastropods (> 1cm).
Fossils suggest a Permian age for this unit.

18.10-70 meters
Largely-covered siltstone and shale.

70-73 meters
Extremely fossiliferous brown-weathering limestone shell bed with high silt
content. Fossils include small bivalves, small gastropods and crinoids. Fossils
suggest that this is the first Triassic unit in the section.

73-91 meters
Largely-covered siltstone with limestone float.

91-100 meters
Poorly-exposed brown limestone with abundant small bivalves and crinoids.

100-137 meters
Covered interval.
140
137-180 meters
Poorly exposed brown limestone with abundant bivalves in dense
accumulations.

180-240 meters
Largely-covered siltstone interval with thin (10cm) calcareous siltstone beds

240-260 meters
Poorly exposed white siltstone.

Blue Diamond, Nevada (Figure A-5)
Take dirt road north off of highway 160 near 36° 2'2.27"N, 115°20'45.09"W
and take right at first fork (about 0.l miles down the road). Near 36° 2'19.10"N,
115°20'47.47"W turn right onto a much less-improved road that makes an immediate
steep but short climb. Because of construction in the area, only the top half of the
section at Blue Diamond was measured. Measured section begins about one meter
above a prominent white marker bed exposed above the road. Turn left at the fork
near 36° 2'23.98"N, 115°20'39.26"W and follow road to the base of the measured
section near 36° 3'0.72"N, 115°20'44.98"W.
141

Figure A-5: Stratigraphic column for the Blue Diamond, Nevada locality.

142
0-4 meters
Poorly-exposed sandy limestone (weathers brown) and limestone (weathers
gray).

4-9 meters
Largely-covered interval with poorly exposed shale and calcareous shale.

9-22.5 meters
Interbedded limestone, dolostone, siltstone and gypsum, poorly-exposed.
Some oolite present.

22.5-40 meters
Dolostone, mostly finely-laminated micrite and oolite, but some small
bivalves and rip-up clasts are present.

Cascade Springs, Utah (Figure A-6)
At Cascade Springs, take road to Hebor. Cross saddle with fork. The section
is exposed along the road near 40°28'4.05"N, 111°31'49.01"W. Base of section is
the lowermost limestone above alluvium.
143

Figure A-6: Stratigraphic column for the Cascade Springs, Utah locality.
144
0-1 meter
Brown bivalve-bearing limestone.

1-53 meters
Poorly exposed silt and calcareous siltstone.

53-72 meters
Calcareous shale and limestone. Limestone weathers light-brown. Abundant
bivalves and crinoid bits.

72-88 meters
Poorly exposed shale and siltstone.

88-108 meters
Siltstone and shale interval with prominent limestone beds. From 88 to 90
meters, a fossiliferous limestone bearing crinoids, bivalves and Meekoceras
ammonoids. From 90 to 104 meters, shale and bivalve-bearing calcareous shale.
From 104 to 106 meters, siltstone with large (~5cm in length) trace fossils on
bedding surface. From 106 to 108, crinoid and bivalve-bearing limestone.

145
108-120 meters
Limestone with lesser sandstone and siltstone. From 108 to 111 meters,
sandstone and siltstone with gutter casts. From 111 to 113 meters, limestone with
bivalve fragments and star-shaped crinoids. From 113 to 114 meters, limestone with
chert nodules and silicified, large bivalves. From 114 to 118 meters, very
fossiliferous limestone with bivalves and crinoids. From 118 to 119 meters, white
siltstone. From 119 to 120 meters, calcareous siltstone with bedding plane trace
fossils and silty limestone with bivalve fragments near top.

120-150 meters
Siltstone from 120 to 130 meters and from 135 to 150 meters, calcareous
siltstone from 130 to 135 meters.

Darwin Hills, California (Figure A-7)
The Union Wash was measured in three sections in the Darwin Hills area.
The Lower Member and the first ~180 meters of the Middle Member were measured
in a section starting near 36°18'44.17"N, 117°34'0.19"W. The remainder of the
Middle Member was measured in a nearby section (< 0.5km west from top of first
section) that starts near 36°19'2.99"N, 117°34'12.66"W. To get to both of these
sections, take the Darwin Road west out of Darwin, California and take unimproved
146

Figure A-7: Stratigraphic column for the Darwin Hills, California locality.
147

Figure A-7 (continued): Stratigraphic column for the Darwin Hills, California
locality.
148

Figure A-7 (continued): Stratigraphic column for the Darwin Hills, California
locality.
149

Figure A-7 (continued): Stratigraphic column for the Darwin Hills, California
locality.
150

Figure A-7 (continued): Stratigraphic column for the Darwin Hills, California
locality.

151

Figure A-7 (continued): Stratigraphic column for the Darwin Hills, California
locality.
152

road that forks off near 36°17'28.53"N, 117°37'7.90"W. Stay to the left of any forks
in the road and park at the end of the road near 36°19'13.90"N, 117°33'58.07"W.

0-22.5 meters
Light-grey, silty limestone and sandy limestone with brown siltstone
interbeds (~1cm) abundant. Irregularlay overlies Permian massive brown siltstone.
Microgastropods and small bivalves.

22.5-159 meters
Largely-covered shale and siltstone with minor limestone.

159-164 meters
Limestone.

164-182 meters
Poorly-exposed shale and siltstone.

182-189 meters
Ammonoid-bearing limestone

153
189-416 meters
Largely-covered shale and siltstone with minor limestone.

416-431 meters
Grey limestone with broken bedding and soft-sediment deformation.
Siltstone interbeds of about 3cm in thickness.

431-470 meters
Poorly-exposed shale with minor limestone.

470-474 meters
Limestone with interbedded shale (beds of a few centimeters each). Contains
ammonoids.

474-483 meters
Poorly-exposed shale.

483-498 meters
Limestone with small (2cm) brown shale interveds.

498-542 meters
Poorly exposed shale with minor limestone.
154
To get to the base of the measured section of the Upper Member of the Union
Wash Formation at Darwin Hills, take N. Main Street east out of Darwin, California
and turn left onto unimproved road at fork near 36°16'59.00"N, 117°34'28.41"W.
Park near 36°17'55.14"N, 117°34'15.46"W and hike down wash to base of section
near 36°18'3.70"N, 117°33'52.11"W. This section is referred to in this text as the
Precipitate Bearing Unit (PBU) because of the abundance of fan-shaped acicular
cement precipitates throughout the interval.

0-33 meters
Light brown-weatering to grey-weathering silty limestone. From 20 to 32
meters, abundant breccia with limestone clasts and silty limestone matrix. At 27
meters, abundant macroscopic pyrite cubes.

33-181 meters
Dark-grey to black-weathering limestone with abundant dissolution voids,
some with silt light-grey silt fill, possibly intraformational silt. Abundant acicular
fan-shaped cements within voids and without, nucleating from both the tops, bottoms
and sides of voids. Fossils extremely rare, except for rare bivalves at 112 meters and
more abundant bivalves at 153. Extremely thin (few millimeters) siltstone
laminations apparent in places. Bedding is chaotic with much soft-sediment
deformation within interval.

155
181-241 meters
Brown shale and light-green to brown-weathering siltstone, partially covered,
with platy limestone exposed from 234 to 234.5 meters.

241-308.5 meters
Dark-grey-weathering nodular limestone with siltstone partings. Abundant,
patchy rose-colored coloration on weathered surfaces. No bedding disruption.
Largely, micritic with some shale from 248.5 to 308.5 meters. Above this section is
the hinge area of a syncline.

Grasshopper Creek, Montana (Figure A-8)
Dinwoody Formation and part of the Woodside Formation are exposed at this
locality. The section is on private land and can be accessed with permission from the
owner. Section is accessed via the Grasshopper Creek Road exit from I15.
Exposure is to the west of I15. The section was measured up the southernmost nose
of the ridge, near 45° 5'45.01"N, 112°46'57.00"W. The lowermost part of the
Dinwoody is covered; the measured section begins at the first exposed bed.

0-12 meters
Limestone, weathers dark brown, but is grayish brown on fresh surface. Beds
are about 15 cm thick and are separated by thin shale beds of about 2-5cm in
156

Figure A-8: Stratigraphic column for the Grasshopper Creek, Montana locality.
157
thickness. Small bivalves (4-5cm) are abundant, as are small (< 1cm) lingulid
brachiopods.

12-25 meters
Covered interval containing two ~10cm limestone beds similar to those from
the base of the section.

25-32 meters
Sandy limestone unit with siltsone interbeds of varying thickness. Lingulid
brachiopods are more abundant than bivalves. Horizontal trace fossils were found on
sandstone float. At 27 meters, there is a 15cm interval of calcareous shale. The
interval between 28 and 29 meters is covered. At 30 meters there is a 75 cm-thick
bivalve-bearing shellbed.

32-35 meters
Covered interval.

35-51.75 meters
Poorly exposed fossiliferous silty limestone with significant siltstone
interbeds of varying thicknesses. Siltstone interbeds are more abundant than lower
in the section. A more resistant and less fossiliferous unit is found between 45 and
46.25 meters.
158
51.75-71 meters
Largely covered fossiliferous limestone beds (10-15cm in thickness) with
thin (<5cm) siltstone interbeds.

71-83.5 meters
Limestone beds (10-15cm in thickness) with thin (<5cm) siltstone interbeds.
At 71-72 meters, small scale cross-bedding, very few bivalves and no lingulid
brachiopods. Silty limestone from 74-75 meters. Between 77.5 and 78.5 meters, a
limestone shellbed, with some crossbeds. Between 78.5 and 83.5, a poorly exposed
siltstone unit.

83.5-91 meters
Sandy limestone with abundant large (10cm) spherical concretions. Bedding
plane shows abundant bivalves. From 85 to 88 meters, a poorly exposed, platy
weathering, white, silty limestone or calcareous siltstone.

91-106 meters
Poorly exposed calcareous siltstone and silstone with a 50cm silty limestone
at 102 meters.

159
Hidden Pasture, Montana (Figure A-9)
Dinwoody, Woodside and Thaynes Formations, near 44°40'39.99"N,
112°47'25.02"W. Take Dell exit off of I15 to the west of freeway. Take road to Big
Sheep Creek. Follow Big Sheep Creek Road past Deadwood Gulch turnoff and park
at Hidden Pasture trailhead. Hike ~2miles to base of Dinwoody near a signpost that
reads “82”.

0-12 meters
Brown-weathering limestone with bivalve and lingulid brachiopod shells,
with hummocky cross stratification. Covered intervals at 2 to 6.5 meters and 8 to 10
meters.

12-41 meters
Poorly exposed siltstone interval with multiple limestone beds. 19 to 22 is a
more resistant siltstone unit, thinly bedded with some calcareous siltstone. 26 to 28
is a thinly bedded (~5-10cm) limestone unit. Dense bivalve shellbeds revealed on
bedding planes. Wrinkle structures found in siltstone float in poorly exposed
siltstone interval from 28 to 30.5 meters. At 30.5 meters, a 10cm limestone bed with
abundant bivalves. 36 to 37 meters is a thinly bedded (5-10cm) limestone unit with
abundant bivalves.
160

Figure A-9: Stratigraphic column for the Hidden Pasture, Montana locality.
161

Figure A-9 (continued): Stratigraphic column for the Hidden Pasture, Montana
locality.
162

Figure A-9 (continued): Stratigraphic column for the Hidden Pasture, Montana
locality.
163

Figure A-9 (continued): Stratigraphic column for the Hidden Pasture, Montana
locality.
164
41-82 meters
Poorly exposed, thin-bedded (5-10cm) silty limestone and siltstone. More
resistant silty limestone with abundant lingulid brachiopods from 41 to 41.5 meters.
At 50 meters, there is a 20cm-thick silty-limestone with silicified shells and crinoid
fragments. Mostly covered siltstone interval with thin (~5cm), bivalve-rich silty
limestone beds from 50.2 to 57 meters. A more resistant silty limestone bed with
abundant bivalve shells outcrops from 57 to 61 meters. From 72 to 73 meters, a
variably bedded silty limestone (2-15cm) with siltstone interbeds, abundant bivalves
and lingulid brachiopods.

82-93 meters
Calcareous siltstone with thin (~5cm) bivalve-rich limestone interbeds, and
small cross bedding.

93-103.5 meters
Poorly exposed siltstone unit with minor limestone beds.

103.5-107 meters
Fossiliferous limestone with small (~5cm) siltstone interbeds.

115-118 meters
Bivalve-rich limestone with some silicification of fossils.
165
118-136 meters
Poorly-exposed siltstone unit with minor (~5cm) fossiliferous limestone
units. At 120 meters, large ripple marks visible on bedding surface. From 120.5 to
120.75 meters, a shell-rich limestone unit with bivalves and lingulid brachiopods. At
123 meters, 5cm of bivalve-rich limestone. From 124 to 125 meters, bivalve-rich
limestone.

136-155 meters
Poorly-exposed siltstone interval with 1-2 m-thick, bivalve-rich limestone
units from 136 to 137 meters, 144 to 145 meters, and 147 to 149 meters.

155-182 meters
Siltstone and calcareous siltstone with two limestone beds. From 155 to 158
meters, limestone bearing a few large bivalves. From 166 to 172.5 meters,
calcareous siltstone with gutter casts, and large bivalves on bedding surface (at 167
meters). From 172.5 to 173 meters, limestone with small, narrow bivalves. From
173 to 179 meters, siltstone with cross beds and gutter casts. From 179 to 182
meters, a light grey to white-weathering limestone with abundant microgastropods
and small bivalve fragments.

182-310 meters
Largely covered siltstone interval.
166
310-319 meters
Ledge-forming, large bivalve-bearing, light-brown to white weathering silty
limestone.

319-334 meters
Largely-covered siltstone and shale.

334-351 meters
Silty, well-laminated limestone, with abundant Meekoceras ammonoids,
marking the Smithian and base of Thaynes Formation.

351-456 meters
Mostly-covered siltstone and shale.

456-500 meters
Grey-weathering limestone, with abundant chert and silicified fossils. After
460 meters, the section was inaccessible because of a fence, but there is an estimated
50 meters of section beyond this point.

167
Rainbow Gardens, Nevada (Figure A-10)
Travel east on Lake Mead Boulevard out of Las Vegas (highway 147).
Follow signs to Rainbow Gardens. The section was measured near 36° 8'31.61"N,
114°57'21.53"W

0-4 meters
Bioclastic, ledge-forming limestone with small gastropods, small bivalves
and scaphopods. Overall massive with lenses of shell debris.

4-11 meters
Poorly-exposed interbedded gypsum and limestone, capped by a cross
bedded siltstone.

11-13 meters
Massive, ledge-forming limestone with small gastropods and bivalves.

13-25 meters
Largely-covered interval with thin limestone beds.

25-27 meters
Ledge-forming limestone with shell fragments and mud rip-up clasts.
168

Figure A-10: Stratigraphic column for the Rainbow Gardens, Nevada locality.
169
27-48 meters
Large exposure interval characterized by red, poorly-lithified siltstone,
punctuated by two thin (10cm) white-weathering oolite dolostones at 38 and 45
meters.

48-71 meters
Large gypsum interval, poorly-exposed and partially covered by road.

71-82.1 meters
Silty dolostone and oolitic dolostone with small bivalves. At 76 meters, a 20
centimeter-thick platy limestone. At 82, a ten-centimeter, prominent, tan-brown-
weathering oolite that makes a good marker bed.

82.1-100 meters
Limestone pebble-conglomerate with clasts 2-3mm in size; contains shell
fragments. Partly covered from 98-100 meters.

100-124 meters
Large gypsum interval with minor silty dolostone from 104.05 to 114.05.

124-127 meters
Silty limestone.
170
Road Cut, San Rafael Swell, Utah (Figure A-2)
From I70 eastbound, exit at the view area overlooking the San Rafael Swell.
Base of section is first unit that reacts significantly with HCl and is located ~.06
miles up the ramp from the road sign indicating oncoming traffic, and is near
38°56’4”N, 110°28’31’W. Underlying units of the Black Dragon Member of the
Moenkopi Formation are about 10 meters of crossbedded sandstone, siltstone, shale
and gypsum. At this locality the Sinbad Limestone Member begins at 13.25 meters
from the base of the measured section.

0-13.25 meters
Siltstone, shale and gypsum with thin (10-15cm) dolostone interbeds that are
light-tan to yellow-weathering, grey on fresh surface. Dolostones exhibit cm-scale
cross-bedding and are not fossiliferous. Dolostones at 0, 5, and 10 meters (the latter
two being dolomitic shales). From 9 to 10 meters, a red-weathering siltstone with
convolute bedding and abundant pyrite cubes.

13.25-15 meters
This unit is recognized here as the equivalent of the basal unit at the Batten
and Stokes locality, a light-tan to yellow-weathering herring-bone cross-bedded
oolite dolostone with spar-filled microgastropod molds and small bivalves.

171
15-16 meters
Calcareous siltstone with thin gypsum interbeds.

16-19.6 meters
Large dolostone unit with abundant bivalves and mirogastropods, and large
(15 cm) vertical trace fossils.

19.6-21.1 meters
Calcareous siltstone with large cross beds.

21.1-23.3 meters
Oolitic silty dolostone beds with thin (< 5cm) gypsum interbeds.

23.3-27 meters
Somewhat red-weathering silty dolostone with large crossbeds that weathers
with pock marks.

Union Wash, California (Figure A-11)
From Lone Pine, take the Owenyo-Lone Pine Road and eventually head
north. Near 36°41'41.86"N, 3'11.34"W, turn right onto two-track dirt road to go up
172

Figure A-11: Stratigraphic column for the Union Wash, California locality.
173

Figure A-11 (continued): Stratigraphic column for the Union Wash, California
locality.
174

Figure A-11 (continued): Stratigraphic column for the Union Wash, California
locality.
175

Figure A-11 (continued): Stratigraphic column for the Union Wash, California
locality.
176

Figure A-11 (continued): Stratigraphic column for the Union Wash, California
locality.
177

Figure A-11 (continued): Stratigraphic column for the Union Wash, California
locality.

178
the large alluvial fan into Union Wash. Base of section is Meekoceras bed near
36°42'23.15"N, 118° 0'53.16"W.

0-10.1 meters
Meekoceras-bearing dark grey-weathering limestone.

10.1-27 meters
Partially-covered siltstone and shale with some silty limestone.

27-52 meters
Limestone and calcareous siltstone (27 to 32 meters, 38 to 41 meters, 47 to
52 meters) with poorly exposed shale.

52-67 meters
Largely-covered shale interval

67-94 meters
Shale with more resistant beds (1-3 meters) of interbedded (~10cm)
limestone, calcareous siltstone and shale at 67 to 70 meters and 92 to 94 metes. This
unit is followed by a sincline-anticline pair but strat can still be traced around folds,
albeit roughly due to covered sections.

179
94-300 meters
Largely covered interval of shale with minor calcareous shale and calcareous
siltstone.

300-319 meters
Interbedded limestone and siltstone (~10cm beds) with lesser shale. Several
2cm intervals of bed-parallel acicular cements that seem to be filling bed-parallel
joints. Cements nucleate from the center of the joint and terminate at the underlying
and overlying beds. Limestones become more massive towards the top of the unit
and contain lenses of dark sparry cement. Overall geometry of the cement lenses is
similar to that of the cement lenses at Darwin Hill (see below), but cement texture is
highly fractured and recrystallized obscuring original cement fabric.

319-397 meters
Largely-covered siltstone and shale.

397-419 meters
Dark-grey to black micritic limestone that is cut by a sill (from 400-404
meters)

419-428 meters
Largely-covered shale interval.
180
428-435 meters
Dark limestone with small (5cm) siltstone and calcareous siltstone interbeds.

435-462 meters
Poorly-exposed shale interval with two more-resistant limestone beds at 442
to 445.5 meters and 451 to 452 meters.

462-538 meters
Largely-covered shale and siltstone interval with thin (10-30 cm) limestone
and calcareous shale beds at 470, 479, 484, 493, 504, 509, and 519 meters, with
thicker limestone beds from 514 to 516 meters and 525 to 528 meters.

538-583 meters
Interbedded limestone and siltstone (~10cm beds) with shale in lower two
meters of unit. Bed-parallel joint-filling acicular cements like those seen
downsection occur near the base of this unit. Unit is cut by a dike from 581 to 583
meters.

583-602.1 meters
Shale and calcareous siltstone.

181
602.1-614 meters
Limestone unit cut by a dike from 611.5 to 614 meters.

614-764 meters
Largely-covered siltstone and shale interval. At 688 meters, a silt-supported
breccia, similar to that seen below the precipitate-bearing unit at Darwin Hills.

764-774 meters
Limestone bed with Parapopanoceras ammonoid fauna.

182
Table A-1: Stable isotopic analyses used in this work. BD=Blue Diamond;
BLC=Beyond Lost Cabin; BS=Batten and Stokes, San Rafael Swell;
BTC=Blacktail Creek; CD= Çürük Da ğ; DH2=Darwin Hills, Upper Member
section; GC=Grasshopper Creek; HP=Hidden Pasture; RC=Road Cut, San
Rafael Swell; RG=Rainbow Gardens. Stratigraphic height for the Çürük Da ğ
samples are given relative to the Permo-Triassic boundary at that locality.
Wu=Wuchiapingian; Ch=Changxingian; Ind=Induan; Smith=Smithian;
Spath=Spathian; Cal=calcite; Dol=dolomite; Eva=evaporite.
183
Loc Ht Min Stage δ
34
S [SO
4
]
87
Sr/
86
Sr δ
13
C δ
18
O

BD 40 Dol Spath 27.42 735 0.7082673 0.51 -5.05
BD 22.5 Dol Spath 28.55 3625 0.7083230 -0.51 -5.40
BD 21.5 Eva Spath 28.11 0.7082039 0.20 -5.43
BD 21 Dol Spath 28.67 6334 0.7082904 0.80 -4.54
BD 16.5 Dol Spath 28.98 6732 0.7081928 0.99 -4.20
BD 16 Eva Spath 30.26 1.32 -3.73
BD 14.5 Eva Spath 29.88 0.7081907 0.61 -3.70
BD 9 Dol Spath 27.78 6001 0.69 -2.93
BD 0 Cal Spath 31.68 725 0.7082206 -2.34 -7.48
BLC 224 Cal Spath 34.12 24 0.7081441 1.19 -7.35
BLC 218 Cal Spath 30.93 94 1.31 -7.20
BLC 163 Cal Spath 37.54 267 0.65 -7.73
BLC 128 Cal Spath 36.99 50 0.26 -7.38
BLC 127 Cal Spath 33.12 439 0.7082344 -0.02 -7.66
BLC 126 Cal Spath 34.39 564
BLC 97.5 Cal Spath 33.06 291 0.7082130 -2.69 -8.64
BLC 80 Cal Spath 31.04 371 -1.36 -8.10
BLC 74 Cal Spath 35.89 401 0.7081319 -0.59 -7.61
BLC 11.2 Cal Spath 31.51 303 -0.07 -7.25
BLC 3.5 Cal Spath 32.30 308 0.7081613 0.65 -7.60
BS 19 Dol Smith 16.35
BS 14 Dol Smith 20.84
BS 7 Dol Smith 18.67
BS 1 Dol Smith 21.26
BTC 182 Cal Ind 27.46
BTC 165 Cal Ind 27.94
BTC 137 Cal Ind 21.65
BTC 73 Dol Ind 19.62
CD 39.5 Cal Ind 34.84 74 1.09 -5.85
CD 39.2 Cal Ind 28.83 661 1.48 -6.10
CD 35.5 Cal Ind 22.67 233 1.54 -6.26
CD 30.6 Cal Ind 23.28 112 1.28 -5.89
CD 25.5 Cal Ind 30.82 34 1.09 -5.64
CD 20.5 Dol Ind 23.58 412 0.59 -7.30
CD 16.5 Cal Ind 23.55 286 1.26 -5.97
CD 15 Cal Ind 25.87 668 -0.79 -9.02
CD 14 Cal Ind 27.81 641 0.44 -5.55
CD 13 Cal Ind 26.42 450 1.58 -5.18
CD 12 Cal Ind 24.66 689 1.01 -5.77
CD 12 Cal Ind 24.60 440 1.13 -5.55
CD 11 Cal Ind 30.18 443 1.02 -4.98
CD 10 Cal Ind 28.87 667 0.74 -5.33
CD 9 Cal Ind 24.07 457 1.93 -6.24
CD 8 Cal Ind 25.36 94 2.30 -7.14
CD 7 Cal Ind 29.21 564 0.22 -5.48

Table A-1: Stable isotopic analyses used in this work
184
CD 6 Cal Ind 24.36 266 0.89 -6.53
CD 5 Cal Ind 23.92 1.17 -5.36
CD 4 Cal Ind 27.27 -0.05 -5.59
CD 3 Cal Ind 22.08 1.05 -5.61
CD 2 Cal Ind 25.47 741 0.50 -5.86
CD 1 Cal Ind 21.53 1.15 -5.41
CD 1 Cal Ind 22.47 1.10 -5.51
CD 0.56 Cal Ind 22.65 1.78 -6.17
CD 0.05 Cal Ind 23.16 1.70 -5.91
CD -0.05 Cal Ch 23.46 394 2.39 -6.64
CD -0.1 Cal Ch 21.31 2.35 -6.77
CD -0.15 Cal Ch 23.85 444 2.13 -6.40
CD -1 Cal Ch 16.14 3.28 -5.59
CD -2 Cal Ch 24.45 3.66 -4.87
CD -3 Cal Ch 12.23 3.37 -5.19
CD -4 Cal Ch 24.16 3.45 -4.11
CD -5 Cal Ch 14.15 3.33 -4.53
CD -6 Cal Ch 15.28 3.45 -4.97
CD -9 Cal Ch 28.50 1317 4.14 -7.75
CD -12 Cal Ch 20.40 445 4.70 -3.90
CD -22.5 Cal Wu 22.63 1113
CD -28 Cal Wu 25.62 913 4.82 -4.73
CD -39 Cal Wu 24.93 498 4.63 -5.11
CD -59.5 Cal Wu 18.04 710
CD -69 Cal Wu 20.51 598
CD -75.5 Cal Wu 18.17 541
CD -98 Cal Wu 22.99 609 4.43 -4.52
CS 120 Cal Spath 35.17
CS 108 Cal Spath 37.09
CS 65 Cal Ind 29.43
CS 0.5 Cal Ind 26.25
DH2 53 Cal Spath 20.99 0.7081556
DH2 43 Cal Spath 0.7081909
DH2 38 Cal Spath 25.05 0.7082103
DH2 33 Cal Spath 16.59 0.7082357
GC 102 Cal Ind 24.94
GC 52 Cal Ind 32.33
GC 30 Cal Ind 22.77
GC 0 Cal Ind 22.95
HP 351 Cal Smith 29.75
HP 311 Cal Ind 31.61
HP 179 Cal Ind 32.09
HP 144.5 Cal Ind 28.50
HP 103.5 Cal Ind 35.81
HP 82 Cal Ind 35.07
HP 60 Cal Ind 31.34

Table A-1 (continued): Stable isotopic analyses used in this
work.
185
HP 30.5 Cal Ind 35.67
HP 0.5 Cal Ind 23.00
RC 27 Dol Smith 14.10
RC 20 Dol Smith 15.55
RC 14 Dol Smith 22.17
RC 10 Dol Smith 17.07
RC 0 Cal Smith 34.87
RG 124 Cal Spath 29.62 721 0.7081993 -3.63 -6.50
RG 118 Eva Spath 30.31 -1.99 -6.45
RG 104.05 Dol Spath 26.34 6081 -1.64 -4.96
RG 104 Eva Spath 27.83
RG 102 Eva Spath 26.65 0.7084095 -1.21 -3.84
RG 82 Dol Spath 27.11 2186 0.7083651 -1.84 -5.40
RG 71 Dol Spath 29.10 1316 0.7082451 -0.55 -4.52
RG 62 Eva Spath 25.04 -0.68 -4.67
RG 48 Eva Spath 27.02 0.7083614 -0.01 -4.01
RG 45 Dol Spath 25.39 5378 0.7081196 -0.21 -4.15
RG 38 Dol Spath 26.51 5410 0.7081894 -0.04 -4.62
RG 25 Cal Spath 33.93 2864 0.7080556 -2.65 -8.48
RG 11 Cal Spath 29.55 875 -1.92 -7.70
RG 5 Eva Spath 26.95 0.7083770
RG 0 Cal Spath 30.22 789 0.7080959 -2.07 -7.99

Table A-1 (continued): Stable isotopic analyses used in this work.
186
Table A-2: Trace elemental and petrographic analyses used in this work.
Abbreviations are as in Table A. The heading CL refers the rank of relative
cathodoluminescence of samples discussed in Chapter 2. Stratigraphic height
for the Çürük Da ğ samples are given relative to the Permo-Triassic boundary
at that locality.
187
Loc Ht Min Stage [Mn] [Sr] [Mg] [Ca] [Fe] Ca/Mg Mn/Sr CL

BD 40 Dol Spath 279 197 101512 205017 3482 2.02 1.42 5
BD 22.5 Dol Spath 287 517 73543 263063 3446 3.58 0.55 5
BD 21.5 Eva Spath
BD 21 Dol Spath 229 550 108749 193818 3980 1.78 0.42 4
BD 16.5 Dol Spath 278 30575 101287 183838 4622 1.82 0.01 1
BD 16 Eva Spath
BD 14.5 Eva Spath
BD 9 Dol Spath 359 543 107064 201598 3783 1.88 0.66 6
BD 0 Cal Spath 318 356 4395 388979 1457 88.51 0.89 6
BLC 224 Cal Spath 190 572 2752 385039 1210 139.93 0.33 5
BLC 218 Cal Spath 257 333 2353 372202 1296 158.18 0.77
BLC 163 Cal Spath 248 638 3180 379146 1261 119.22 0.39
BLC 128 Cal Spath 214 1161 3696 353366 2079 95.61 0.18
BLC 127 Cal Spath 329 515 4780 375204 2078 78.50 0.64 7
BLC 126 Cal Spath
BLC 97.5 Cal Spath 1480 410 3352 347988 1959 103.83 3.61 7
BLC 80 Cal Spath 436 370 3053 370278 2039 121.28 1.18
BLC 74 Cal Spath 226 648 4371 370054 1636 84.67 0.35 7
BLC 11.2 Cal Spath 552 671 3619 343878 3172 95.02 0.82
BLC 3.5 Cal Spath 177 1832 3914 381105 1456 97.37 0.10 6
BS 19 Dol Smith
BS 14 Dol Smith
BS 7 Dol Smith
BS 1 Dol Smith
BTC 182 Cal Ind
BTC 165 Cal Ind
BTC 137 Cal Ind
BTC 73 Dol Ind
CD 39.5 Cal Ind 59 1386 2241 376528 1092 167.98 0.04
CD 39.2 Cal Ind 51 1232 1904 366418 696 192.41 0.04
CD 35.5 Cal Ind 83 219 21291 334961 1161 15.73 0.38
CD 30.6 Cal Ind 33 348 2682 363987 215 135.70 0.09
CD 25.5 Cal Ind 21 1515 1854 383949 324 207.07 0.01
CD 20.5 Dol Ind 223 193 66028 273171 807 4.14 1.16
CD 16.5 Cal Ind 66 390 3071 379485 631 123.57 0.17
CD 15 Cal Ind 25 1230 1969 366919 674 186.32 0.02
CD 14 Cal Ind 78 900 4469 342381 1156 76.62 0.09
CD 13 Cal Ind
CD 12 Cal Ind 60 518 3200 369920 1240 115.60 0.12
CD 12 Cal Ind 84 668 10780 360768 739 33.47 0.13
CD 11 Cal Ind 22 1643 1993 361364 512 181.33 0.01
CD 10 Cal Ind 40 1009 4295 367229 304 85.49 0.04
CD 9 Cal Ind 55 364 2123 378086 244 178.12 0.15
CD 8 Cal Ind 84 384 4753 363039 400 76.38 0.22

Table A-2: Trace elemental and petrographic analyses used in this work.
188
CD 7 Cal Ind 26 2020 2036 377131 194 185.21 0.01
CD 6 Cal Ind 82 378 4184 374310 423 89.47 0.22
CD 5 Cal Ind 67 727 2693 373900 182 138.82 0.09
CD 4 Cal Ind 25 1717 1926 359200 818 186.52 0.01
CD 3 Cal Ind 85 458 2715 376240 541 138.57 0.19
CD 2 Cal Ind 64 1185 3127 358887 822 114.76 0.05
CD 1 Cal Ind 42 930 2457 359719 672 146.41 0.05
CD 1 Cal Ind 30 1067 2428 371485 801 153.01 0.03
CD 0.56 Cal Ind 43 956 3309 362225 803 109.46 0.04
CD 0.05 Cal Ind 77 502 4368 361313 1004 82.72 0.15
CD -0.05 Cal Ch 72 334 3449 369417 881 107.10 0.22
CD -0.1 Cal Ch 79 334 4584 362349 1154 79.05 0.24
CD -0.15 Cal Ch 79 293 3413 363000 820 106.37 0.27
CD -1 Cal Ch 71 439 5449 362255 1932 66.48 0.16
CD -2 Cal Ch 449 461 3971 364639 383 91.84 0.97
CD -3 Cal Ch 114 402 3553 373292 873 105.06 0.28
CD -4 Cal Ch 54 529 4230 367035 427 86.77 0.10
CD -5 Cal Ch 98 413 6745 343352 1332 50.91 0.24
CD -6 Cal Ch 86 457 7924 346301 1001 43.70 0.19
CD -9 Cal Ch 74 449 3955 357032 1391 90.28 0.17
CD -12 Cal Ch 90 371 4953 381158 640 76.95 0.24
CD -22.5 Cal Wu 81 370 4259 364453 904 85.58 0.22
CD -28 Cal Wu 82 395 4851 365174 1628 75.28 0.21
CD -39 Cal Wu 59 285 3689 363323 597 98.48 0.21
CD -59.5 Cal Wu 92 324 4387 365763 698 83.38 0.28
CD -69 Cal Wu 89 404 4836 364056 1262 75.27 0.22
CD -75.5 Cal Wu 133 324 7078 374917 645 52.97 0.41
CD -98 Cal Wu 41 445 4920 358807 1216 72.93 0.09
CS 120 Cal Spath
CS 108 Cal Spath
CS 65 Cal Ind
CS 0.5 Cal Ind
DH2 53 Cal Spath
DH2 43 Cal Spath
DH2 38 Cal Spath
DH2 33 Cal Spath
GC 102 Cal Ind
GC 52 Cal Ind
GC 30 Cal Ind
GC 0 Cal Ind
HP 351 Cal Smith
HP 311 Cal Ind
HP 179 Cal Ind
HP 144.5 Cal Ind
HP 103.5 Cal Ind
HP 82 Cal Ind

Table A-2 (continued): Trace elemental and petrographic analyses used in this
work.
189
HP 60 Cal Ind
HP 30.5 Cal Ind
HP 0.5 Cal Ind
RC 27 Dol Smith
RC 20 Dol Smith
RC 14 Dol Smith
RC 10 Dol Smith
RC 0 Cal Smith
RG 124 Cal Spath 181 170 20122 377022 1006 18.74 1.07 5
RG 118 Eva Spath
RG 104.05 Dol Spath 178 289 104196 204963 2039 1.97 0.62 5
RG 104 Eva Spath
RG 102 Eva Spath
RG 82 Dol Spath 297 262 69439 247465 4357 3.56 1.13 7
RG 71 Dol Spath 211 274 109528 208539 2822 1.90 0.77 2
RG 62 Eva Spath
RG 48 Eva Spath
RG 45 Dol Spath 1650 232 103219 198712 2393 1.93 7.10 9
RG 38 Dol Spath 1316 182 101548 214969 1572 2.12 7.23 8
RG 25 Cal Spath 132 4691 4656 388427 1002 83.42 0.03 3
RG 11 Cal Spath 167 188 20831 311523 1156 14.95 0.89 3
RG 5 Eva Spath
RG 0 Cal Spath 133 167 8529 373688 565 43.81 0.79 7

Table A-2 (continued): Trace elemental and petrographic analyses used in this
work.

Abstract (if available)

Abstract The use of carbonate associated sulfate (CAS) to study sulfur isotope chemostratigraphy is investigated in detail. Results suggest that middle-shelf limestones are better suited for sulfur isotopic analysis than proximal evaporites or dolostones because of possible facies-related factors that preclude the latter phases from recording true seawater sulfate del34S values. Carbonate samples with pyrite should be avoided because of evidence of pyrite oxidation during the CAS extraction process.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The early Triassic recovery period: exploring ecology and evolution in marine benthic communities following the Permian-Triassic mass extinction

PDF

Paleoenvironmental and paleoecological trends leading up to the end-Triassic mass extinction event

PDF

Carbonate geochemistry in primary, diagenetic and biological systems

PDF

Unraveling mass extinctions: Permian to Early Jurassic onshore-offshore trends of marine stenolaemate bryozoans

PDF

Sedimentary geochemistry associated with the end-Triassic mass extinction: changes to the marine environment from an age constrained sedimentary section

PDF

Community paleoecology and global diversity patterns during the end-Guadalupian extinction (middle-late Permian) and the transition from the Paleozoic to modern evolutionary faunas

PDF

Characterization of geochemical and lithologic variations in Milankovitch cycles: Green River Formation, Wyoming

PDF

Germanium and silicon isotope geochemistry in terrestrial and marine low-temperature environments

PDF

Ecological recovery dynamics of the benthic and pelagic fauna in response to extreme temperature events and low oxygen environments developed during the early Triassic

PDF

Bioturbation in Cambrian siliciclastic shelf strata: paleoecological, paleoenvironmental, and temporal patterns

PDF

Lithium isotopes as a carbon cycle proxy: examining the effects of high-temperature weathering, understanding brachiopod archives, and validating Cenozoic records

PDF

Benthic paleoecology and macroevolution during the Norian Stage of the Late Triassic

PDF

The geobiology of fluvial, lacustrine, and marginal marine carbonate microbialites (Pleistocene, Miocene, and Late Triassic) and their environmental significance

PDF

Tracking fluctuations in the eastern tropical north Pacific oxygen minimum zone: a high-resolution geochemical evaluation of laminated sediments along western North America

PDF

Biotic recovery from the end-Permian mass extinction: Analysis of biofabric trends in the Lower Triassic Virgin Limestone, southern Nevada

$Isotopic insights to human-impacted hydrologic systems: from urban watersheds to hydraulically fractured basins$

PDF

Isotopic insights to human-impacted hydrologic systems: from urban watersheds to hydraulically fractured basins

PDF

Ecosystem export efficiency in an upwelling region: a two-year time series study of vertical transport, particle export and in-situ net and gross oxygen production

PDF

Evolutionary paleoecology and taphonomy of the earliest animals: Evidence from the Neoproterozoic and Cambrian of southwest China

PDF

Exploring seasonal climate variability in East African trees: High-resolution oxygen and carbon isotope records of cellulose of Vitex payos from Chyulu Hills, Kenya

PDF

Early Jurassic reef eclipse: Paleoecology and sclerochronology of the "Lithiotis" facies bivalves

Asset Metadata

Creator Marenco, Pedro Jose (author)

Core Title Sulfur isotope geochemistry and the end Permian mass extinction

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Geological Sciences

Publication Date 07/26/2007

Defense Date 04/26/2007

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag carbonate associated sulfate,early Triassic,End Permian,mass extinction,OAI-PMH Harvest,stable isotope,Sulfur

Place Name Turkey (countries)

Language English

Advisor Bottjer, David J. (committee chair), Capone, Douglas G. (committee member), Corsetti, Frank (committee member), Douglas, Robert (committee member), Fischer, Alfred G. (committee member)

Creator Email marenco@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m685

Unique identifier UC1216441

Identifier etd-Marenco-20070726 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-524467 (legacy record id),usctheses-m685 (legacy record id)

Legacy Identifier etd-Marenco-20070726.pdf

Dmrecord 524467

Document Type Dissertation

Rights Marenco, Pedro Jose

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu