Strontium Isotope Stratigraphy and Carbonate Sedimentology of the Latest Permian to Early Triassic in the Western United States, Northern Iran and Southern China

Home , Jinogondolella, Phosphoria Formation, Union Wash Formation

Strontium isotope stratigraphy and carbonate sedimentology of the latest Permian to Early Triassic in the western United States, northern Iran and southern China

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

Alexa Ruth Clements Sedlacek

Graduate Program in Geological Sciences

The Ohio State University

2013

Dissertation Committee:

Matthew R. Saltzman, Advisor

William I. Ausich

Stig M. Bergström

David H. Elliot

Alexa Ruth Clements Sedlacek

2013

Abstract

Environmental perturbations associated with the late-Permian mass extinction and

Early Triassic recovery are widely studied, and our understanding of global events during this biotic crisis continues to improve. Marine carbonates preserve geochemical proxy records of global change, including a rapid rise in seawater 87Sr/86Sr that began the near

Guadalupian-Lopingian (Middle-Late Permian) boundary and continued through the end of the Early Triassic. Linking the increase in 87Sr/86Sr to enhanced weathering rates that resulted from environmental disturbances and terrestrial extinction during the latest

Permian is inconsistent with the onset of the 87Sr/86Sr rise approximately 10 million years earlier at the end of the Middle Permian. However, several recent studies indicate that the rate of 87Sr/86Sr rise increased during the late Permian, and this rise may be the result of enhanced weathering rates during the extinction and recovery interval.

This study improves the resolution of the seawater 87Sr/86Sr record of the late

Permian through Early Triassic through analyses of whole rock samples collected from

Zal, Iran and Dawen, South China. The carbonate succession at Zal, Iran preserves the entire Early Triassic and was previously analyzed for δ13C. Analyses of Sr isotopes at a resolution similar to the δ13C record enables direct comparison between these two

ii records, and also allows for determinations in the changes in rate of strontium rise. The

Permian-Triassic boundary interval was sampled at higher resolution by analyzing rocks from a biostratigraphically constrained sections deposited at Dawen on the Great Bank of

Guizhou, South China. These results indicate that the rate of strontium rise increased during the latest Permian, possibly as a result of elevated weathering rates during the post-extinction interval.

The final chapter of this dissertation examines marine carbonates from the

Confusion Range, Utah, recently shown to represent relatively continuous deposition across the Permian-Triassic boundary. The proposed boundary interval consists of several distinctive lithologies which are similar in many regards to some Tethyan sections that contain microbially mediated carbonates of earliest Triassic age. Although microbial characteristics are found within the boundary interval of the Confusion Range section, future petrographic and chronostratigraphic studies are needed to determine if this microbial carbonate was deposited as part of a global event or simply reflect the local depositional environment.

iii

This dissertation is dedicated to Marlene E. Weaver

Acknowledgments

I am sincerely grateful to my advisor, Matthew Saltzman, for his intellectual guidance and patience. I thank him and my committee members, Bill Ausich, Stig

Bergström and David Elliot for their support through hours of helpful discussion. I am grateful to Jim Collinson for his guidance in the field and to Ken Foland for discussing the various merits and limitations of strontium isotope stratigraphy. I am deeply grateful to Paul Garvin and Rhawn Denniston, who inspired my interest in geology and encouraged me to pursue graduate school.

The Geological Society of America’s student research grant program and the

Friends of Orton Hall Fund supported my research. The laboratory work that it required would not have been possible without the support and mentorship of Jeff Linder, who educated me in all things related to the clean lab and mass spectrometer. I am also grateful to Angie Rogers and Susie Shipley, who moved administrative mountains to accommodate my needs.

I owe sincere gratitude to my peers at OSU, particularly Amanda Howard, Soo

Yeun Ahn, and Christina O’Malley for their constant support and encouragement. My parents, Janet Weaver and Laurence Clements, and my grandmother Marlene Weaver enabled me to finish what I started. Finally I owe my deepest appreciation to Patrick, who unfailingly supports and encourages me to pursue my goals, and to our daughter, for

v being such a good sleeper. Her recent transformation from baby to toddler provided the final incentive for me to complete this work.

Vita

July 4, 1984 ...... Born-Nottingham, U.K.

2005...... B.A. Geology, Religion, Cornell College

2007-2008 ...... Graduate Fellow, School of Earth Sciences,

The Ohio State University

2009-2010 ...... Graduate Research Assistant, School of

Earth Sciences, The Ohio State University

2008-2013 ...... Graduate Teaching Assistant, School of

Earth Sciences, The Ohio State University

Publications

Saltzman, M.R., and Sedlacek, A.R.C, 2013, Chemostratigraphy indicates a relatively complete Late Permian to Early Triassic sequence in the western United States: Geology, v. 41, p. 399-402.

Abstracts

Sedlacek, A.R.C., Saltzman, M.R., Algeo, T.J., Horacek, M., Richoz, S., Brandner, R., and Foland, K., 2012, Coupled C and Sr isotope stratigraphy of the Early Triassic of Zal,

Iran: A record of increased weathering. Geological Society of America Abstracts with vii

Programs, vol., 44, p. 62.

Trigg, C., Saltzman, M.R., Carlucci, J.R., Westrop, S.R., Leslie, S.A., Young, S.A.,

Bergstrom, S.M., Sedlacek, A.R.C., and Edwards, C.T., 2012, Positive carbon isotope shift in the Middle to Late Ordovician: No link with sea level? Geological Society of

America Abstracts with Programs, vol., 44, p. 237.

Sedlacek, A.R.C., Saltzman, M.R., Algeo, T. J., Horacek, M., Foland, K., Linder, J.S.,

Howard, A., Sedlak, C., and Walters, A.P., 2011, Strontium isotope stratigraphy from the

Late Permian to Early Triassic of Zal, Iran. Geological Society of America Abstracts with

Programs, vol. 43, p. 603.

Sedlacek, A.R.C. and Saltzman, M.R., 2010, Evidence for an Earliest Triassic

Microbialite from the Confusion Range, UT. Geological Society of America Abstracts with Programs, vol. 42, p. 132.

Howard, A., Saltzman, M.R., Sedlacek, A.R.C., Sedlak, C., Foland, K., Linder, J.S.,

Leslie, S.A., and Young, S.A., 2010, Strontium and neodymium isotope stratigraphy of the middle Ordovician and weathering of the Appalachian Mountains. Geological Society of America Abstracts with Programs, v. 42, p. 466. viii

Sedlacek, A.R.C. and Saltzman, M.R., 2009, The Permian-Triassic transition in the Great

Basin (Confusion Range-Spruce Mountain composite): Age constraints based on a continuous carbon isotope record, strontium isotopes, and sequence stratigraphy.

Geological Society of America Abstracts with Programs, vol. 41, p. 360.

Sedlacek, A.R., Saltzman, M.R., and Linder, J., 2008, Strontium isotope stratigraphy of the Permian-Triassic Boundary interval in the Great Basin, USA: How much of a record is preserved? Geological Society of America Abstract with Programs, v. 40, p. 265.

Fields of Study

Major Field: Geological Sciences

Table of Contents

Abstract ...... ii

Acknowledgments...... v

Vita ...... vii

List of Tables ...... xv

List of Figures ...... xvii

Chapter 1: Coupled carbon and strontium isotope stratigraphy from the late Permian to the Early Triassic of Zal, Iran: A record of increased weathering ...... 1

Abstract ...... 1

Introduction ...... 2

Background ...... 3

Seawater 87Sr/86Sr ...... 3

Paleogeography ...... 4

Methods and Results ...... 5

Discussion ...... 6

Diagenetic Alteration ...... 6

Early Triassic Time Scale ...... 7

Coupled 87Sr/86Sr and δ13C Variation ...... 8

Chapter 2: Strontium isotope stratigraphy across the latest Permian extinction horizon and Permian-Triassic boundary at Dawen, Great Bank of Guizhou, South China ...... 13

Abstract ...... 13

Introduction ...... 14

Background ...... 17

Geologic Setting ...... 17

87Sr/86Sr of marine carbonates ...... 18

Methods ...... 19

Results ...... 20

Discussion ...... 20

Sr isotope trends at Dawen ...... 21

A composite curve of seawater 87Sr/86Sr through the latest Permian and Early

Triassic...... 22

Whole rock versus brachiopod calcite and preservation of seawater 87Sr/86Sr ...... 24

Calculation of changes in the rate of 87Sr/86Sr rise in the latest Permian ...... 26

Chapter 3: A unique facies succession through the Permian-Triassic boundary transition interval (Gerster and Thaynes formations) from the Confusion Range, Utah ...... 41

Abstract ...... 41

Introduction ...... 42

Background ...... 45

Lithostratigraphy and general age of Gerster and Thaynes formations, Confusion

Range, Utah ...... 46

Conodont biostratigraphy and biozones ...... 47

Chemostratigraphy...... 48

Methods ...... 50

Lithologic study ...... 50

Whole rock 87Sr/86Sr ...... 50

Results ...... 51

Lithologic study ...... 51

Strontium isotope analysis ...... 53

Discussion ...... 54

xii

Depositional environment of the Gerster-Thaynes transition beds and sea level

change ...... 55

Early Triassic Microbialites and Confusion Range lithologies C and D ...... 57

87Sr/86Sr of transitional lithologies ...... 59

Implications and Conclusions ...... 60

Acknowledgments ...... 61

References ...... 85

Chapter 1 ...... 85

Chapter 2 ...... 91

Chapter 3 ...... 97

Appendix A: Supplemental Material for Chapter 1 ...... 105

Locality Map ...... 105

Paleogeographic Map ...... 106

Stratigraphy at Zal ...... 107

Diagenesis ...... 110

Early Triassic Timescale ...... 112

Data Tables ...... 115

xiii

Appendix B: Supplemental Material for Chapter 2 ...... 121

Appendix C: Supplemental Material for Chapter 3 ...... 127

xiv

List of Tables

Table 1. 87Sr/86Sr results from Zal, Iran ...... 115

Table 2. Differences in rates of 87Sr/86Sr rise (∂Sr/∂t) when calculated using ages of

Algeo et al. (2012). ∂Sr/∂t was calculated by fitting a line through the Sr isotopic data for each (sub)stage so as to minimize the sum of squared deviations...... 117

Table 3. Differences in linear sedimentation rates (LSRs) when calculated using ages of

Algeo et al. (2012). LSRs (in units of meters per million years) were assumed to be constant for each (sub)stage and were calculated by dividing the thickness of a time- stratigraphic unit by the duration of that (sub)stage...... 118

Table 4. Differences in rates of 87Sr/86Sr rise (∂Sr/∂t) when calculated using ages of

Gradstien et al. (2012). ∂Sr/∂t was calculated by fitting a line through the Sr isotopic data for each (sub)stage so as to minimize the sum of squared deviations...... 119

Table 5. Differences in linear sedimentation rates (LSRs) when calculated using ages of

Gradstein et al. (2012). LSRs (in units of meters per million years) were assumed to be constant for each (sub)stage and were calculated by dividing the thickness of a time- stratigraphic unit by the duration of that (sub)stage...... 120

Table 6. 87Sr/86Sr results from Dawen, Guizhou, China ...... 122

Table 7. Brachiopod LMC from Brand et al., 2012 ...... 125

Table 8. Results of 87Sr/86Sr from the Confusion Range, Utah, USA ...... 128

xvi

List of Figures

Figure 1. δ13C profile for Zal, Iran, from Horacek et al. (2007a) and 87Sr/86Sr data for the same sample suite. Samples associated with volcanic sills (open symbols) are considered altered. 87Sr/86Sr values rise most rapidly during the Dienerian substage, highlighted in gray. The onset of the positive δ13C excursion corresponds with the increase in the rate of 87Sr/86Sr rise. Dien. = Dienerian. Early Triassic substage ages from Algeo et al.

(2012)...... 10

Figure 2. Linear sedimentation rates (LSR) calculated for the Early Triassic substages.

LSRs were calculated for Zal following the methods of Algeo and Twitchett (2010) by dividing thickness by time (m.y.) for each substage. LSRs are given for each substage to the left of the curve, and corresponding ∂Sr/∂t (in units of 10-4 m.y.-1) are shown in boxes.

Maximum sedimentation rates are observed during the Dienerian substage. Chang. =

Changhsingian, Griesbach., Gr. = Griesbachian, D. = Dienerian, Sm. = Smithian. LSRs and ∂Sr/∂t are given in DR Table S2...... 11

Figure 3. Hypothesis linking Siberian Traps volcanism to increases in Early Triassic seawater 87Sr/86Sr and other marine perturbations. Climatic warming led to faster chemical weathering rates, and destruction of terrestrial ecosystems resulted in greater exposure of bedrock to physical weathering. Enhanced weathering increased the flux of xvii nutrients to the oceans, stimulating marine productivity and organic carbon burial (which was also favored by high sedimentation rates), and causing positive δ13C excursions.

Increased fluxes and/or higher 87Sr/86Sr values of riverine Sr resulted in rising seawater

87Sr/86Sr. Modified from Algeo et al. (2011)...... 12

Figure 4. Seawater 87Sr/86Sr of the Phanerozoic. The interval investigated in this study, spanning the late Permian the Early-Middle Triassic boundary, is highlighted in grey.

This represents the most rapid rise in seawater 87Sr/86Sr of the Phanerozoic. (After Kani et al., 2008)...... 29

Figure 5. 87Sr/86Sr of late Permian through Triassic after McArthur et al. (2012). A rise in seawater 87Sr/86Sr occurred from near the end-Capitanian through the Early-Middle

Triassic boundary (shaded interval). This interpretation of the seawater strontium isotope curve has a Permian Triassic boundary value near 0.7072. A significant slowing in the rate of rise occurs during the Late Permian (Wuchiapingian), but may be an artifact of timescale uncertainties. A slowing of the rate of rise (or even falling values) has also been proposed for the latest Permian by Twitchett (2007) and Brand et al. (2012), although this does not show up in the McArthur et al. (2012) curve. Gray dashed lines are drawn at 87Sr/86Sr= 0.7072, 0.7071 and 0.7070...... 30

Figure 6. Global correlation charts for the Late Permian and Early Triassic taken from the individual Permian and Triassic chapters of the Geologic Timescale 2012 (Gradstein

xviii et al., 2012). Note that there is some disagreement between these two chapters in terms of the conodont zonation in the latest Permian...... 31

Figure 7. Permian-Triassic boundary paleogeographic reconstruction after Algeo and

Twitchett (2010). Permian-Triassic boundary sections analyzed for strontium isotopes considered in this study. South China Block: 1. Dawen, Great Bank of Guizhou, China

(this study), 2. Meishan, South China (Cao et al., 2009); Iranian microcontinent: 3. Zal, northwestern Iran (Sedlacek et al., in preparation), 4. Abadeh, central Iran (Korte et al.,

2003, 2004); Eastern Pangean margin: 5. Sass de Putia, Dolomites, Italy (Brand et al.,

2012), 6. Val Brutta, Dolomites, Italy (Brand et al., 2012). Gray shaded area indicates shallow shelf environments...... 33

Figure 8. A) The Nanpanjiang basin, part of the South China Block during the late

Permian, is located southwest of Meishan, China. B) Dawen section is located on the

Great Bank of Guizhou (GBG), one of several isolated carbonate platforms deposited in the Nanpanjiang Basin during the Permian-Triassic boundary interval. After Lehrmann et al. (2003)...... 34

Figure 9. 87Sr/86Sr of the Permian-Triassic boundary interval Dawen, GBG, South China.

Samples were collected from the upper Wuchiaping Formation and the calcimicrobial framestone facies of the lower Daye Formation. Lowermost sample analyzed is from 400 cm below the dissolution surface, from within the Clarkina chanxingensis-deflecta zone of Chen et al. (2009). The base of the Hindeodus parvus sensu lato zone is defined by xix the first occurrence datum (FOD) of H. parvus (Chen et al., 2009). Filled symbols indicate Sr concentrations greater than 400 ppm, which are considered least altered samples (Fig. 10)...... 35

Filled symbols, with concentrations higher than 300 ppm, correspond to filled symbols in

Figure 9. Dashed lines are drawn at 0.7071 and 0.7072. Shaded interval corresponds to samples with Sr concentration below 400 ppm that are considered altered...... 37

Clarkina meishanensis- Hindeodus eurypyge biozone. Dawen data are placed in correct stratigraphic order. However their ages may be relatively older or younger than Brand et al. (2012) data within the biostratigraphic age constraints. Grey dashed lines are drawn at

87Sr/86Sr = 0.7071 and 0.7072...... 38

87Sr/86Sr = 0.7071 and 0.7072. Each section indicates seawater 87Sr/86Sr greater than

0.7072 in the basal Triassic (H. parvus zone). Abbreviations for Meishan: Wuchiaping. =

Wuchiapingian, changxingen. = Clarkina chanxingensis, I. isarcica = Isarcicella isarcica, ch.-pr. = C. chanxingensis yini - Hindeodus praeparvus, m. -e. = Clarkina meishanensis meishanensis- H. eurypyge, p. = Hindeodus parvus, s. = Isarcicella staeschei. Abadeh: trans. = C. transcaucasica, orientalis- medicons. = C. orientalis-C. mediconstricta, subc. = C. subcarinta, ch.d. = C. changxingensis- C. deflecta, i = C. iranica, h. = C. hauschkei, m.p. = C. meishanensis- C. praeparvus, p. = Hindeodus parvus. Zal: Ps. = Pseudotoceras, Ph. = Phisonites-Dhzulfites, Pa. = Paratirolites beds, m.-p. = C. meishanensis- H. praeparvus, H. parv = H. parvus ...... 39

Earliest Triassic Microbialites (ETMs). U= upwelling, S= sinking. Shaded area represents shallow shelf environment. Map after Algeo and Twitchett (2010)...... 63

Figure 14. Map showing location of the Confusion Range study area in west-central

Utah, in the Great Basin region, United States ...... 64

Figure 15. 87Sr/86Sr and d13C of the Gerster and Thaynes formations of the Confusion

Range. Regionally, the thickest Gerster-Thaynes succession is preserved at the xxi

Confusion Range. The lower Gerster Formation is a cherty fossiliferous limestone, which is overlain by a series of transitional lithologies highlighted in grey. This interval is expanded in Figure 16...... 65

Figure 16. Detailed measured section of the Gerster-Thaynes transitional lithologies A-F.

Strontium isotope analyses reveal a shift to more radiogenic values in lithologies C-E, and the rising trend of seawater 87Sr/86Sr is indicated by the dashed line through the least radiogenic values. Filled symbols indicate Sr concentrations greater than 100 ppm, however, these samples with highest concentrations occur within samples that deviate from the seawater trend. The interval of cover within lithology E occurs between the top of the gully section and the base of the next exposed limestone ledge...... 66

Figure 17. The negative carbon isotope excursion present in most Permian-Triassic boundary sections is used to correlate the Confusion Range section to the Guryul Ravine,

Pakistan and the GSSP for the Permian-Triassic boundary at Meishan, China (Saltzman and Sedlacek, 2013)...... 67

Figure 18. A rise in seawater 87Sr/86Sr occurred from the late Middle Permian through the Early Triassic. Least radiogenic values from the Confusion Range are plotted here to constrain a minimum age estimate for the negative carbon isotope excursion (see text for discussion) (Saltzman and Sedlacek. 2013)...... 69

Figure 19. Field photograph of red calcareous sandstone (lithology A). Rock hammer for scale...... 70 xxii

Figure 20. Field photographs of chert beds (lithology B). A) Smaller chert nodules within tan carbonate cement are overlain by a bedded chert interval. Pen for scale is 15 cm long. B) Largest chert nodules occur at the base of lithology B. Chert nodules occur within grey carbonate cement, and tan colored cement is observed near the top of the photograph. Pen for scale is 15 cm long...... 71

Figure 21. Field photograph of the gully interval containing laminated limestones

(lithology C), fenestral limestone bed (lithology D) and microgastropod packstone with potential digitate fabric (lithology E). The base of this section is a brecciated limestone containing angular chert fragments, which represents the top of lithology B. Hammer for scale...... 72

Figure 22. Field photograph of laminated limestone (lithology C). A) Teepee structures of Collinson et al. (1976) are desiccation structures, field notebook is 19 cm wide. B)

Laminar beds with secondary calcite infilling between micritic layers. Beds show minor offset, possibly as part of a desiccation structure (US dime for scale). C) Sub millimeter scale wavy bedding within the laminar limestone is potentially microbial in origin (US dime for scale) occur at the base of lithology C immediately above the brecciated limestone that characterizes the upper portion of lithology B...... 73

Figure 23. Field photograph of fenestral limestone unit (lithology D). A) Lower portion of fenestral unit overlying Bed C. Pen for scale is 15 cm long. B) Contact between

xxiii lithology D and lithology E. Ruler for scale (16 cm) is placed at near the base of potential digitate structures...... 75

Figure 24. Field photographs of gully exposure with lithologic contacts outlined in yellow. A) The base of the gully exposure consists of the highest chert bed (lithology B).

This is overlain by lithology C, along an irregular contact. The fenestral fabric of lithology D directly underlies the microbial grainstone (lithology E) with potential digitate structure. The ruler for scale placed on laminations limestone (lithology C) is 16 cm long. B) Flat truncation of potential digitate structures occurs just below field notebook (19 cm long)...... 76

Figure 25 Photomicrographs of the typical Gerster Formation from the Confusion Range

(photo width 3 mm for all images). A) Bryozoan from typical Gerster Formation at the

Confusion Range collected approximately 19 m below the base of the red sandstone. B)

Brachiopod, bryozoan and echinoderm fossil fragments from the Gerster Formation 0.5 m below the base of the red sandstone. C) Echinoderm fragment from 0.5 m below the base of the red sandstone...... 77

Figure 26. Photomicrographs of carbonate cement from lithology B, texture indicates extensive recrystallization. Photo width is 3 mm for A and B. A) Sample collected from nodular chert interval (lithology B) 2.6 m above the top of the red sandstone. B) Sample collected from bedded chert interval with tan carbonate cement 4.4 m above the top of the red sandstone...... 78 xxiv

Figure 27. Polished section and photomicrographs from lithology C. A) Polished slab of lithology C shows sub-millimeter light-dark alternating bands of micritic limestone. B)

Photomicrograph of lithology C with arrow indicating chambered structure similar to

Renalcis, but this structure may be caused by carbonate neomorphism. Photo is 3 mm wide. C) Lobate margins and clotted texture indicate microbial origin. Photo width is 3 mm...... 79

Figure 28. A) Hand sample from lithology D show large coated grains. B)

Photomicrograph of lithology D shows abundant ooids and pellets fabric consistent with a microbial origin. Photo is 3 mm wide...... 80

Figure 29. Polished hand sample of lithology E (microgastropod packstone) reveals that apparent digitate structures observed in the field occur along fractures in the rock, indicating this is a diagenetic feature. Microgastropods are clearly visible within and above the secondary ‘digitate’ fabric...... 81

Figure 30. Photomicrographs from lithology E and lithology F reveal similarities between these two lithologies. Photos are both 3 mm wide. A) Thin section from lithology E contains recrystallized molluscan shells. B) Thin section from lithology F at the top of the transitional lithologies also contains recrystallized molluscan shells not visible in handsample ...... 82

xxv

Figure 31. Cross plot of 87Sr/86Sr of the Confusion Range shows a weak negative correlation, and is indicative of diagenetic alteration. Sr concentrations are low (below

250) for all samples...... 83

Figure 32. Cross plot of 87Sr/86Sr for the lithologies of the Gerster-Thaynes transition in the Confusion Range. Concentrations are very low for all samples in this interval, making them more susceptible to diagenetic alteration...... 84

Figure 33. Map of Iran showing the location of the study section near the village of Zal in northwestern Iran (after Kozur, 2007)...... 105

Figure 34. Paleogeographic map of Pangea during the Early Triassic. The Zal section was located in the equatorial Tethys on the northern margin of the Iranian microcontinent, which formed part of the Cimmeria terranes. After Algeo and Twitchett

(2010)...... 106

Figure 35. Carbon and strontium isotope data plotted against stratigraphic thickness at

Zal. The Dienerian is represented by approximately 300 meters of section, allowing for relatively high resolution sampling through this substage. The two Griesbachian data points associated with a volcanic sill are considered altered and are shown with unfilled symbols...... 108

Figure 36. Crossplot of 87Sr/86Sr and Sr concentration (ppm) shows no clear covariation in Changhsingian, Griesbachian, Dienerian and Smithian age samples, indicating that the isotopic values of these rocks may be relatively unaltered by diagenesis. Covariation xxvi present in Spathian samples that are from dolomitized lithologies and these values are likely altered...... 111

Figure 37. Carbon and strontium isotope data plotted against dates from the Geologic

Time Scale 2012 (Gradstein et al., 2012). Here, the duration of the Dienerian is longer by 0.58 m.y. relative to its duration in the Algeo et al. (2012) timescale (0.27 m.y.), lowering its ∂Sr/∂t. However, even after accounting for differences in the Early Triassic timescale, ∂Sr/∂t remains higher during the Dienerian than for other Early Triassic substages (Tables 2, 4)...... 113

Figure 38. Linear sedimentation rates (LSR) calculated for the Late Permian-Early

Triassic based on the Geologic Time Scale 2012 (Gradstein et al., 2012) rather than on the Algeo et al. (2012) timescale, as in Figure 2. LSRs were calculated for Zal following the methods of Algeo and Twitchett (2010) by dividing thickness by time (m.y.) for each stage or substage. LSRs are given for each (sub)stage to the left of the curve, and corresponding ∂Sr/∂t (in units of 10-4 m.y.-1) are shown in boxes. Maximum sedimentation rates occurred during the Dienerian, although Dienerian rates are reduced relative to Figure 1.2 owing to the longer duration of this substage in the GTS 2012 (0.85 m.y.). Chang. = Changhsingian, Griesbach., Gr. = Griesbachian, D. = Dienerian, Sm. =

Smithian, Spath. = Spathian. LSR and ∂Sr/∂t are given in DR Table S2...... 114

xxvii

Chapter 1: Coupled carbon and strontium isotope stratigraphy from the late Permian to

the Early Triassic of Zal, Iran: A record of increased weathering

Abstract

Recovery from the late Permian mass extinction was slowed by continued environmental perturbations during the Early Triassic. Rapid fluctuations of the Early

Triassic carbon isotope record indicate instability in the global carbon cycle. Positive carbon isotope excursions were potentially the result of enhanced continental weathering due to episodes of volcanism and warming, but linking weathering rates to carbon cycling has proven difficult. One proxy for weathering is the 87Sr/86Sr of marine carbonate, and we present here a 87Sr/86Sr record from an uppermost Permian-Lower Triassic succession

13 near Zal, Iran that is coupled to a δ Ccarb record. At the base of the Dienerian, an increase in the rate of 87Sr/86Sr rise occurred concurrently with the onset of the first large positive δ13C excursion of the Early Triassic. The rapid rise in 87Sr/86Sr is consistent with a previous study of sedimentation rates in the Tethys region that provided evidence of increased bedrock weathering. Elevated weathering rates may have increased both the sediment and nutrient flux into the oceans, which enhanced organic carbon burial rates and led to heavier carbonate δ13C values. Subsequent decoupling of the C and Sr isotope records in the remainder of the Early Triassic may reflect the long residence time of Sr compared to C in the oceans. 1

Introduction

The Early Triassic was a period of protracted recovery from the late Permian mass extinction, which devastated marine and terrestrial ecosystems (Retallack, 1995;

Erwin et al., 2002; Payne et al., 2006; Algeo et al., 2011). Low diversity levels characterized the Early Triassic Griesbachian (early Induan) substage, with robust recovery of many clades occurring in the Spathian (late Olenekian substage) (Payne et al., 2006; Algeo et al., 2011), although the tempo of recovery in benthic and pelagic communities remains the subject of discussion (Stanley, 2009; Song et al., 2011). The

Early Triassic record is also marked by instability in the carbon cycle and global temperature fluctuations, which suggest that biotic recovery may have been inhibited by persistent environmental disturbances (Payne et al., 2004; Horacek et al., 2007a, Sun et al., 2012). Payne and Kump (2007) proposed that the oscillatory nature of the Early

Triassic marine carbonate δ13C record (e.g. Payne et al., 2004; Horacek et al., 2007b) was linked to warming caused by episodic pulses of volcanism from the Siberian Traps and related increases in weathering rates, nutrient flux into the world oceans, and organic carbon burial. Evidence for increased sedimentation rates has been documented and attributed to increased weathering resulting from higher temperatures, lower pH of precipitation, and increased bedrock exposure due to the destruction of terrestrial ecosystems during the end-Permian crisis (Algeo and Twitchett, 2010). However,

13 directly linking elevated weathering rates to δ Ccarb fluctuations is difficult because sedimentation rates are typically calculated on timescales coarser than the carbon isotope curve and may reflect local tectonic influences. One way to examine proxy evidence for

2 weathering at the same resolution as variations in Early Triassic δ13C profiles is by coupling C and Sr isotope records, which represent global conditions. Enhanced continental weathering may have contributed to the large and rapid global increase of marine carbonate 87Sr/86Sr from approximately 0.7070 to 0.7082 during the late Permian-

Early Triassic (Martin and Macdougall, 1995; Korte et al. 2003, 2004, 2006; Huang et al.,

2008), but linkages to contemporaneous δ13C variations are poorly understood.

Here we present a 87Sr/86Sr time series generated from bulk carbonate samples from an uppermost Permian-Lower Triassic section at Zal, Iran. This section has been biostratigraphically constrained and was previously analyzed for carbon isotopes

(Horacek et al., 2007a). Thus, the present study is the first high-resolution analysis of

13 strontium isotopic variation in the Early Triassic that is coupled with a δ Ccarb record from the same field site (Fig. 1).

Background

Seawater 87Sr/86Sr

Seawater 87Sr/86Sr ratios are globally homogeneous at any given time because the residence time of Sr in the ocean vastly exceeds the oceanic mixing time (Veizer, 1989).

Marine carbonates record systematic secular variation in seawater 87Sr/86Sr through geologic time, providing a chemostratigraphic tool (Veizer et al., 1999; Martin and

Macdougall, 1995; Korte et al., 2003, 2004, 2006; Huang et al., 2008). The strontium isotopic composition of seawater is predominately controlled by inputs at mid-ocean ridges through hydrothermal exchange (87Sr/86Sr = ~ 0.703) and from rivers through

3 continental erosion (87Sr/86Sr = 0.705 to > 0.800; mean ~0.712; Palmer and Edmond,

1989). Variations in seawater 87Sr/86Sr reflect changes in the concentration of Sr and the

87Sr/86Sr of dominant sources (Veizer et al., 1999), with the hydrothermal source being less radiogenic (~0.703) and generally accounting for less of the total Sr input in the ocean system (Palmer and Edmond, 1989). The concentration and isotopic value of the riverine flux is highly variable due to heterogeneities in the isotopic composition of continental crust, the concentration of Sr in source rocks, and weathering rates. Thus, much of the secular variation in seawater 87Sr/86Sr is thought to result from changes in the

87Sr/86Sr and total Sr concentration of the riverine flux (Berner, 2004).

The rapid increase in seawater 87Sr/86Sr during the late Permian and Early Triassic has been well documented by analyzing brachiopod low magnesium calcite, conodont apatite (Martin and Macdougall, 1995; Korte et al., 2003, 2004, 2006) and well-preserved micrite (Huang et al., 2008). The rate of the Permo-Triassic rise is greater than that of the late Cenozoic (post-8 Ma) rise that has been attributed to increased continental weathering rates linked to rapid orogenic uplift (Hodell et al., 1990; Martin and

Macdougall, 1995; Korte et al., 2003).

Paleogeography

The study section is located in northwestern Iran, 30 km south of Julfa, near the village of Zal (see Appendix A Fig. 33). During the Early Triassic, the Iranian microcontinent formed part of the Cimmeria terranes, separating the Neo- and Paleo-

Tethys oceans (Segnör, 1984; Fig. 34). The section represents relatively continuous

4 deposition on a tropical carbonate platform from the late Permian through the Early

Triassic (Horacek et al., 2007a). At Zal, Early Triassic sedimentation rates were greatest during the Dienerian (Fig. 2), enabling high-resolution sampling through the relatively short duration of this Early Triassic substage.

Methods and Results

Forty-seven bulk rock powders corresponding to those analyzed for δ13C by

Horacek et al. (2007a) were obtained and processed for 87Sr/86Sr following the methods of Montañez et al. (1996). Approximately 40 mg of powders were cleaned with ultra pure 1M ammonium acetate buffered to a pH of 8, then dissolved in 4% acetic acid. The resulting supernatant was spiked with an 84Sr tracer, and Sr was isolated using chromatography methods calibrated to 2N HCl. Samples were loaded with HCl on a Re double filament configuration with a Ta2O5 activator, and isotopic ratios were measured using a Finnegan MAT 261A thermal ionization mass spectrometer in the Ohio State

University Radiogenic Isotope Lab. The laboratory reproducibility of the standard SRM

987 is 0.710242 ± 0.000010 (1σ reproducibility). Reported 87Sr/86Sr values are corrected for instrumental fractionation using normal 86Sr/88Sr of 0.119400.

Chemical analyses reveal a continuous increase in 87Sr/86Sr (0.7071) to the top

(0.7081) of the section (Fig. 1, Table 1). Nearly 300 meters of Dienerian-age strata account for the majority of isotope values reported, and these samples reveal the most rapid changes in both the strontium and carbon isotope records (Fig. 2, 35). The

Dienerian is characterized by a large positive excursion in δ13C and a rapid increase in

87Sr/86Sr. The overlying Smithian exhibits a large negative shift in δ13C and a slight slowing of the rate of 87Sr/86Sr change (∂Sr/∂t). The base of the Spathian is marked by a second large positive excursion in δ13C that coincides with a sharp decline in the rate of

87Sr/86Sr increase, while the bulk of the Spathian exhibits slow trends toward lower δ13C and higher 87Sr/86Sr values.

Discussion

Diagenetic Alteration

Our data record the ~0.001 rise in seawater Sr-isotope trends recognized for the late Permian-Early Triassic interval in earlier studies (Korte et al., 2003, 2004, 2006;

Huang et al., 2008). However, in order to determine whether the rate of rise in 87Sr/86Sr varies for individual stages or zones, increased sampling density near stage boundaries is needed. Bulk rock methods, where applied to thick stratigraphic sections such as Zal, can provide this level of resolution, but the role of diagenesis must be addressed (Veizer et al., 1999; Waltham and Gröcke, 2006). Sr concentrations are considered one of the more sensitive and reliable indicators of diagenetic alteration, which generally proceeds with loss of Sr (Veizer, 1989). At Zal, the majority of samples yielded Sr concentrations >200 ppm, with >1000 ppm for some Dienerian-age samples (Fig. 36). These concentrations are consistent with those reported in other studies of marine carbonates that were interpreted as yielding primary seawater 87Sr/86Sr values (e.g., Montañez et al., 1996;

Huang et al., 2008). Only two Griesbachian samples associated with igneous sills exhibit

87Sr enrichment relative to stratigraphically proximal samples, and provide clear evidence

6 of diagenetic alteration (Fig. 1). The remaining 45 samples yield 87Sr/86Sr values that define a coherent pattern of secular variation that we consider to represent primary late

Permian-Early Triassic seawater values (Veizer et al., 1999). The preservation of seawater trends at Zal is also supported by comparisons with the latest Permian

(Changhsingian) brachiopod data of Brand et al. (2012) that have values near ~0.7071, and comparisons with the Early Triassic conodont data of Korte et al. (2004) that indicate values of ~0.7074 to 0.7077 for the N. dieneri conodont zone (late Dienerian).

Early Triassic Time Scale

In addition to diagenesis, the Early Triassic time scale has a source of uncertainty in the calculation of linear sedimentation rates (LSR) and rates of change in seawater Sr isotopes (∂Sr/∂t) for individual stages or substages. At Zal, the Permian-Triassic boundary and Griesbachian substage are biostratigraphically constrained using conodonts, and other substage assignments are based on a combination of index fossils and carbon isotope correlations with well-dated sections elsewhere in the Tethys

(Horacek et al., 2007a).

Owing to uncertainties in the age assignments for the Early Triassic substage boundaries, we used the timescales of both Algeo et al. (2012) and the Geologic Time

Scale (GTS) 2012 in calculating linear sedimentation rates and ∂Sr/∂t. For both timescales, the highest rates of ∂Sr/∂t were during the Dienerian (Tables 2-5). However,

Dienerian ∂Sr/∂t calculated using the Algeo et al. (2012) timescale is approximately 3 times greater than that of the GTS 2012. Although 87Sr/86Sr continued to increase during

7 the Olenekian, the rate of rise diminished through the end of the Spathian as sedimentation rates decreased (Fig. 2).

Coupled 87Sr/86Sr and δ13C Variation

Based on these data, the most likely explanation for the peak rate of Sr rise during the Dienerian is high continental weathering rates. The combination of warming, linked to episodic eruptions of the Siberian Traps (Payne and Kump, 2007; Sun et al., 2012), and the increased area of exposed bedrock due to delayed recovery of terrestrial ecosystems (Algeo and Twitchett, 2010) intensified weathering reactions and increased the surface area available for chemical and physical weathering (Fig. 3). Changes from meandering to braided streams (Ward et al., 2000) and elevated sedimentation rates

(Algeo and Twitchett, 2010) indicate that continental physical weathering rates increased during the Induan. High sedimentation rates correspond to increased organic carbon burial rates in nearshore environments, and elevated weathering rates increased the amount of nutrients delivered to the oceans, bolstering primary productivity, resulting in further organic carbon burial (Berner, 2004). These elevated productivity rates are consistent with evidence for eutrophication and episodic expansion of the oxygen minimum zone during the Early Triassic (Algeo et al., 2011, 2012; Horacek et al.,

13 2007a). Therefore, the positive δ Ccarb excursion through the Denierian is linked to elevated organic carbon burial as proposed by Payne and Kump (2007) (Fig. 3).

Silicate weathering removes atmospheric CO2 (Berner, 2004), and a period of rapid weathering is consistent with a reported moderate cooling trend during the late Dienerian

(Sun et al., 2012), with minor recovery in some clades during the Dienerian possibly

87 86 related to CO2 drawdown (Stanley, 2009; Sun et al., 2012). Although Sr/ Sr continues to rise until the end of the Spathian, the rate of rise was sharply diminished around the

Smithian-Spathian boundary (SSB), which is consistent with decreased sedimentation rates during the Olenekian (Algeo et al., 2011; Fig. 2). Decreased global temperatures and incipient recovery of terrestrial ecosystems at the SSB contributed to reduced continental weathering rates (Looy et al., 1999; Algeo and Twitchett, 2010; Sun et al.,

2012). At Zal, the δ13C and 87Sr/86Sr profile decouples at the base of the Olenekian, probably as a consequence of the greater residence time of strontium in the ocean relative to carbon. No similar link between the two isotope records occurs with later positive carbon isotope excursions or high temperatures of the late Smithian (Sun et al., 2012).

This may be due to minor recovery of terrestrial ecosystems during the Olenekian or less fresh rock surfaces for available weathering.

248

249

250

251

252

253

254 Age -4 -2 0 2 4 6 0.7070 0.7074 0.7078 0.7082 (Ma) 13C -VPDB 87Sr/86Sr (Horacek et al., 2007a) (This Study)

52 m m.y.-1 560 1.0 480 149.3 m m.y.-1 4.0 400

-1 1111.1 m m.y. 8.1 240

160

80 -1 169.6 m m.y. 2.9 -1 0 11.1 m m.y. Chang. 1.4 (sub)stage Chang. Gr. D.Sm. Spathian

Age (Ma) 254 252 250 248

Figure 2. Linear sedimentation rates (LSR) calculated for the Early Triassic substages. LSRs were calculated for Zal following the methods of Algeo and Twitchett (2010) by dividing thickness by time (m.y.) for each substage. LSRs are given for each substage to the left of the curve, and corresponding ∂Sr/∂t (in units of 10-4 m.y.-1) are shown in boxes. Maximum sedimentation rates are observed during the Dienerian substage. Chang. = Changhsingian, Griesbach., Gr. = Griesbachian, D. = Dienerian, Sm. = Smithian. LSRs and ∂Sr/∂t are given in DR Table S2.

Siberian Traps Volcanism

Destruction of Climate Terrestrial Ecosystems Warming

Bedrock Weathering

Increased nutrient flux Increased clastic sedimentation rate in Primary marine systems productivity

Organic carbon burial

87 86 Increased Sr/ Sr Positive 13C excursion

Figure 3. Hypothesis linking Siberian Traps volcanism to increases in Early Triassic seawater 87Sr/86Sr and other marine perturbations. Climatic warming led to faster chemical weathering rates, and destruction of terrestrial ecosystems resulted in greater exposure of bedrock to physical weathering. Enhanced weathering increased the flux of nutrients to the oceans, stimulating marine productivity and organic carbon burial (which was also favored by high sedimentation rates), and causing positive δ13C excursions. Increased fluxes and/or higher 87Sr/86Sr values of riverine Sr resulted in rising seawater 87Sr/86Sr. Modified from Algeo et al. (2011).

Chapter 2: Strontium isotope stratigraphy across the latest Permian extinction horizon

and Permian-Triassic boundary at Dawen, Great Bank of Guizhou, South China

Abstract

The most rapid rise in seawater 87Sr/86Sr during the Phanerozoic occurred from the Paleozoic minimum near the Guadalupian-Lopingian (Middle-Late Permian) boundary through the end of the Early Triassic. Linking the rising trend in 87Sr/86Sr to enhanced weathering rates that resulted from environmental disturbances and terrestrial extinction during the latest Permian is inconsistent with the onset of the 87Sr/86Sr rise approximately 10 million years earlier at the end of the Middle Permian. However, potential changes in the rate of 87Sr/86Sr rise across the latest Permian extinction event could have implications for paleoenvironmental change. In the present study, bulk carbonate samples from the Permian-Triassic boundary interval of Dawen, Great Bank of

Guizhou, South China were analyzed for 87Sr/86Sr. These data are compared with other

87Sr/86Sr records from biostratigraphically constrained Permian-Triassic boundary sections. At Dawen, least radiogenic 87Sr/ 86Sr increased from 0.70710 in the late

Changhsingian Clarkina chanxingensis-deflecta conodont biozone to 0.70719 in the

Hindeodus parvus biozone. This Early Triassic value is consistent with other studies and indicates that the rate of seawater 87Sr/86Sr rise increased significantly during the Early

Triassic relative to the Late Permian (Lopingian) rate of change. Elevated weathering 13 rates during the latest Permian extinction and Early Triassic recovery interval, and a corresponding increase in the flux of radiogenic Sr to the marine system likely contributed to the rapid rise in the strontium isotopic signature of seawater.

Introduction

The largest mass extinction of the Phanerozoic occurred during the latest Permian, affecting both marine and terrestrial ecosystems (Retallack, 1995; Erwin et al., 2002).

Recovery during the Early Triassic was hindered by continued environmental disturbances and extreme warming (Payne et al., 2006; Payne et al., 2007; Algeo et al.,

2011; Sun et al., 2012; Brand et al., 2012). One of the most prominent geochemical proxy records of global change during the Late Permian to Early Triassic is the rise in seawater 87Sr/86Sr that has been documented in numerous studies during the past four decades (Veizer and Compston, 1974; Burke et al., 1982, Veizer, 1989; Martin and

Macdougall, 1995; Veizer et al. 1999; Korte et al., 2003, 2004; Huang et al., 2008;

McArthur et al., 2012; Figs. 4, 5). The Late Permian increase in seawater 87Sr/86Sr began near the Guadalupian-Lopingian (Middle-Late Permian; Figs. 4, 5) boundary and continued through the Early Triassic (Martin and Macdougall, 1995; Korte et al., 2003,

2004; Kani et al., 2008; Huang et al., 2008; McArthur et al., 2012; Fig. 5). The rise from the Paleozoic minimum of approximately 0.7069 (Korte et al., 2006; Kani et al., 2008;

McArthur et al., 2012) to 0.7082 at the end of the Early Triassic (Korte et al., 2003) represents the most rapid strontium isotope rise (approximately 1 ×10-4 /myr) during the

Phanerozoic and is greater than the well-documented Cenozoic rise (approximately 0.3

×10-4/myr) (Hodell et al., 1990; Martin and Macdougall, 1995; Korte et al., 2003, 2004;

McArthur et al., 2012; Fig. 4). However, the causes of the Late Permian to Early Triassic rise and the timing of changes in the rate of rise through this interval remain the subject of discussion (Twitchett, 2007; Brand et al., 2012; McArthur et al., 2012).

The overall rising 87Sr/86Sr trend has been attributed to a reduction of seafloor spreading rates in the Capitanian (latest Middle Permian; Fig. 5, 6), rifting of Pangea, and enhanced continental weathering rates during a warming climate (Martin and

Macdougall, 1995; Korte et al., 2003, 2004, 2006; Kidder and Worsley, 2004; Twitchett,

2007; Kani et al., 2008; Brand et al., 2012). Changes in the rate of rise during the latest

Permian extinction interval and a possible brief transition to a flat or falling 87Sr/86Sr trend may also be associated with variations in mid-ocean ridge activity or continental weathering (Twitchett, 2007; Brand et al., 2012), which may have occurred in response to destruction of terrestrial ecosystems or Siberian Traps volcanism. However, resolving whether changes in the rate of seawater 87Sr/86Sr rise or a brief transition to falling values occurred during the latest Permian in association with the extinction event is problematic because it requires both the ability to establish time lines that can be assigned absolute ages and to account for any secondary overprinting of primary seawater values.

Studies indicate that well-preserved low magnesium calcite (LMC) brachiopod shells have the highest potential to reliably maintain the primary seawater isotopic values

(e.g. Kearsey et al., 2009; Brand et al. 2012). However, during the post-extinction interval, the availability of well-preserved brachiopods is limited (Korte et al., 2003).

Although well-preserved brachiopod LMC is not always available, strontium isotopic

15 values from other materials can be used to establish seawater 87Sr/86Sr trends. Conodont elements (Martin and Macdougall, 1995; Korte et al. 2003, 2004; Twitchett, 2007) and whole rock samples (Huang et al., 2008; Cao et al., 2009; Saltzman and Sedlacek, 2013) have been used to improve the resolution of strontium isotope data through the Permian-

Triassic boundary (PTB) interval. Although whole rock and conodont samples are more susceptible to diagenetic alteration, which generally increases 87Sr/86Sr (Brand and

Veizer, 1980; Brand et al., 2012), least radiogenic values can preserve seawater 87Sr/86Sr trends.

Here we present new whole rock strontium isotope data from the PTB interval of the Dawen section, Great Bank of Guizhou, Guizhou, South China (Figs. 4, 5). The

Dawen section contains the key index fossil for the base of the Triassic Hindeodus parvus and is also chemostratigraphically correlated to the Global Stratotype Section and Point

(GSSP) of the basal Triassic at Meishan, Zhejiang, China using 13C (Chen et al., 2009).

In addition, the Dawen section contains microbialite deposits that mark the latest Permian extinction horizon in various parts of the Tethys (Payne et al., 2007; Chen et al., 2009).

These bio-, chemo- and litho-stratigraphic constraints at Dawen enable correlation to other strontium isotope records produced from biostratigraphically constrained sections elsewhere in the world. Our results from Dawen indicate that during the late

Changhsingian seawater 87Sr/86Sr was ~ ≤ 0.70710, consistent with Brand et al. (2012).

Results from Dawen also indicate that the strontium isotopic composition of seawater was ≤ 0.70719 in the basal Induan, which is less than recent work that placed 87Sr/86Sr of the Permian-Triassic boundary at ~0.7072-0.7073 (Fig. 5; McArthur et al., 2012) but

16 consistent with least radiogenic data reported by other authors (Korte et al., 2006; Huang et al., 2008). Our results suggest that the rate of 87Sr/86Sr rise was low (or possibly falling slightly) during the latest Permian and increased significantly near the latest Permian extinction horizon with a continued high rate of rise through the Early Triassic.

Furthermore, calibrating the 87Sr/86Sr for the PTB interval in sections with key index fossils such as Hindeodus parvus at Dawen improves the potential usefulness of strontium chemostratigraphic methods to determine ages or minimum ages for sections where biostratigraphic constraints are limited.

Background

Geologic Setting

The Dawen section, located in Guizhou Province, South China (Fig. 8A), is one of several continuous PTB carbonate successions deposited on the Great Bank of

Guizhou (GBG) (Lehrmann et al., 2003; Chen et al., 2009). During the PTB interval, the

GBG was one of many isolated carbonate platforms located in the Nanpanjiang Basin

(Figure 8B) on the South China Block in the equatorial eastern Tethys Ocean (Fig. 7)

(Tong and Yin, 2002, Lehrmann et al., 2003). The Upper Permian Wujiaping Formation, a shallow marine skeletal packstone, is overlain by the Daye Formation along a sharp and irregular contact, which is interpreted as a submarine dissolution surface (cf. Payne et al.,

2007; Fig. 9). However, the continuity of conodont biozones across the dissolution surface indicates that this contact represents only a minor hiatus, or diastem (Chen et al.,

2009).

At Dawen, Chen et al. (2009) used bio- and chemostratigraphic methods to correlate with the PTB interval at Meishan. The upper Wujiaping Formation contains an approximately 2‰ negative carbon isotope excursion (Fig. 9) which was used to correlate this portion of the section to Bed 24e at Meishan (Clarkina changxingensis deflecta zone) (Chen et al., 2009; Gradstein et al 2012). Bed 24e is just below the peak of the mass extinction (Bed 25 at Meishan). The base of the calcimicrobial framestone was assigned to the C. zhejiangensis-H. eurypyge zone and is considered latest

Changhsingian in age (based on correlation with Meishan). The H. parvus zone contains

H. parvus parvus and H. parvus erectus, and its base is defined by the First Occurrence

Datum (FOD) of H. parvus parvus. This FOD is considered to closely approximate the

First Appearance Datum (FAD) of H. parvus at Meishan (Chen et al., 2009). The present study analyzed samples collected from the Wujiaping Formation and calcimicrobial framestone unit of the Lower Daye Formation (Fig. 9). The dissolution surface present here is recognized throughout the GBG region and thought to be coincident with the main extinction event (Payne et al., 2007).

87Sr/86Sr of marine carbonates

Seawater 87Sr/86Sr is homogenous at any given time because the residence time of

Sr in the ocean greatly exceeds oceanic mixing time (Veizer, 1989). The 87Sr/86Sr of marine carbonates is used as a proxy for Phanerozoic seawater values, because no fractionation occurs when Sr replaces Ca in CaCO3 minerals (Veizer, 1989; Veizer et al.,

1999; Martin and Macdougall, 1995; Korte et al., 2003, 2004, 2006; Huang et al., 2008).

The major sources of strontium into the marine system are at mid-ocean ridges through hydrothermal exchange and from rivers through continental erosion (Palmer and

Edmond, 1989). Variations in the 87Sr/86Sr seawater curve reflect changes in the riverine and hydrothermal sources of strontium (Veizer and Compston, 1974; Palmer and

Edmonds, 1989, Veizer, 1989; Martin and Macdougall, 1995; Veizer et al., 1999).

Heterogeneity of the strontium isotopic composition of continental source rocks makes the isotopic signature of the riverine flux highly variable and changes in source rocks and weathering rates are the dominant control on seawater 87Sr/86Sr (Berner, 2004).

Methods

Approximately 60 mg of bulk rock powders were pretreated with ultrapure 1M ammonium acetate buffered to a pH of 8 (Montañez et al., 1996; Bailey et al., 2000).

Samples were then leached in 4% acetic acid, and the resulting supernatant was collected and spiked with an 84Sr tracer. Sr was isolated using Biorad AG 50×8 cation exchange resin and a 2N HCl based ion exchange following the laboratory methods of Foland and

Allen (1991). Samples were loaded with HCl on a Re double filament configuration with a Ta2O5 activator. Isotopic compositions were measured with a Finnegan MAT 261A thermal ionization mass spectrometer at The Ohio State University. The laboratory reproducibility of the standard SRM 987 is 0.710242 ± 0.000010 (1σ reproducibility).

Reported 87Sr/86Sr values are corrected for instrumental fractionation using normal

86Sr/88Sr of 0.119400.

Results

87Sr/86Sr of Dawen samples range from 0.707103 to 0.707407, but least radiogenic values increased in 87Sr/86Sr from 0.707103 to 0.707193 though the PTB interval (Fig. 5; Table 1). Sr concentrations are below 500 ppm in all samples, and samples with the highest Sr concentrations yielded the lowest 87Sr/86Sr values (Fig. 10).

We consider the least radiogenic samples with concentrations higher than 400 ppm most likely represent primary seawater strontium isotopic values (Figs. 9, 10). This is because secondary alteration typically results in decreases in Sr concentrations in limestones and increases in 87Sr/86Sr to ratios more radiogenic than the initial primary seawater value

(due largely to contamination by radiogenic strontium from siliciclastic components;

Veizer, 1989). Using the least radiogenic data with highest Sr concentrations (Fig.10) to constrain the 87Sr/86Sr curve at Dawen, values increase from approximately 0.707103 in the upper Wujiaping Formation (Clarkina changxingensis-deflecta conodont zone) to near 0.7072 in the Hindeodus parvus zone (Fig. 9). An increase to values ranging from

0.707238 to 0.707407 occurs near the dissolution surface and mass extinction interval, but Sr concentrations of these samples are less than 200 ppm. Nonetheless, although the values associated with the dissolution surface are likely altered, it is possible that this short-term rise is significant because other samples above and below this surface have similar Sr concentrations but do not have elevated 87Sr/86Sr.

Discussion

Sr isotope trends at Dawen

At Dawen, the strong negative correlation between Sr concentration and 87Sr/86Sr is consistent with diagenetic alteration. Therefore we used the least radiogenic values with the highest Sr concentrations (> 400 ppm Sr) to define primary seawater 87Sr/86Sr estimates (Figs. 9, 10). While the cutoff of 400 ppm Sr is necessarily arbitrary and the designation of altered versus unaltered samples is defined by a line through a continuum of Sr concentrations, this concentration is similar to other whole rock studies interpreted to preserve seawater 87Sr/86Sr values and trends (e.g., Denison et al., 1994; Montañez et al., 1996). The 87Sr/86Sr increase at Dawen from 0.707103 near the base of the section to

0.707193 within the H. parvus zone is consistent with values reported for the latest

Permian and PTB interval by some authors (Huang et al., 2008), and overall less radiogenic than PTB 87Sr/86Sr values presented by others (Martin and Macdougall, 1995;

Korte et al., 2003, 2004; McArthur et al., 2012). Although the 87Sr/86Sr trend from latest

Permian to Early Triassic was rising overall, some studies indicate a brief interval of falling or flat values occurred prior to the latest Permian extinction horizon (Twitchett,

2007; Brand et al., 2012). At Dawen, although based only on three samples, a falling trend from 0.70717 to 0.707103 in the lowest part of the sampled section beneath the extinction horizon is not inconsistent with other studies (Twitchett, 2007; Brand et al.,

2012) (Figs. 9, 11). These three samples contain relatively high Sr concentrations, but additional study is needed to confirm this trend. Potential causes of a falling trend during an interval of general seawater 87Sr/86Sr rise may be the result of a decrease in riverine flux, a decrease in the 87Sr/86Sr of the riverine flux (possibly as a result of weathering less

21 radiogenic basalts from the Siberian Traps), an increase in spreading rates, or a combination of these factors.

Also of potential significance to the study of the latest Permian extinction, we note that 87Sr/86Sr at Dawen increased near the dissolution surface and mass extinction interval (Fig. 9). Because these samples contain relatively low Sr concentrations, we consider the absolute values to be altered, but the spike in 87Sr/86Sr could be real. A similar 87Sr/86Sr spike through the latest Permian occurred in other South China sections

(e.g. Zhongliang Mountain, Chongqing (Huang et al., 2008) and Meishan core, Meishan,

China (Fig. 12) (Cao et al., 2009). Furthermore, at Dawen, other samples with similarly low Sr concentrations (~ 200 ppm) do not show 87Sr/86Sr values that are as elevated as those at the dissolution surface and latest Permian extinction horizon. The possibility of a spike to more radiogenic values also requires further investigation and higher resolution sampling across the boundary interval in other PTB sections.

A composite curve of seawater 87Sr/86Sr through the latest Permian and Early Triassic

Here we attempt to define the change in seawater 87Sr/86Sr by using data reported from latest Permian, pre-extinction brachiopods of the Italian Dolomites (Brand et al.,

2012) and the post-extinction latest Permian to earliest Triassic using our new data at

Dawen, China. This requires construction of a composite data set merging the sampled sections at Dawen and the Italian Alps. Brand et al. (2012) sampled the upper

Bellerophon and lower Werfen Formations of the Sass de Putia and Val Brutta sections of the Italian Dolomites (Fig. 7, 11), and correlated the Bulla Member of the upper

Bellerophon Formation with Bed 24e at Meishan (C. changxingensis-deflecta zone), which corresponds to samples from the upper Wujiaping Formation at Dawen (Figs. 9,

11). The lowermost Tesero Member of the Werfen Formation is correlated with Bed 25-

27b at Meishan, consistent with a latest Changhsingian age and coincident with the main pulse of extinction (C. meishanensis meishanensis- H. eurypyge zone), and likely correlates to the lowermost samples of the Daye Formation at Dawen (Brand et al., 2012;

Gradstein et al., 2012). The least radiogenic 87Sr/86Sr from unaltered brachiopods of the

Bulla Member range from 0.707051-0.707100 at Val Brutta and 0.707060-0.707078 at

Sass de Putia. Unaltered brachiopod samples from the lower Tesero Member yielded

0.707108 at Val Brutta and 0.707092 at Sass de Putia. These results indicate that

87Sr/86Sr remained relatively invariant until the extinction horizon (lower Tesero

Member), where the rate of 87Sr/86Sr increase rises (Brand et al., 2012). However, limited sampling from the Tesero Member and no lower Triassic data points limit the ability to assess changes in the rate of rise to the latest Permian, and we use the Dawen data set beginning at the latest Permian extinction horizon and continuing through the H. parvus zone of the earliest Triassic to determine changes in the rate of 87Sr/86Sr increase.

At Dawen, the least radiogenic data from the H. parvus zone at Dawen, which contains Sr concentrations of approximately 425 ppm and suggests a small offset from unaltered brachiopods of approximately 0.4×10-4, indicates that seawater 87Sr/86Sr was 0.70715

(based on our measured whole rock value of 0.70719) through the base of the Early

Triassic (Fig. 9).

Other well-dated carbonate successions deposited during the PTB interval in the

Tethys (Iran and South China) have been studied for 87Sr/86Sr trends but likely were altered to more radiogenic values (Fig. 12). However, if we assume that the least radiogenic samples within a short stratigraphic interval have been shifted to more radiogenic values by approximately the same amount, these key sections may still potentially be used for assessing general seawater trends through the late Permian and

Early Triassic in comparison to the Italian Alps and Dawen composite. Data from the

Meishan Core show that 87Sr/86Sr generally increases from 0.7071 in the Changhsingian

C. subcarinata zone to around 0.7072 in the C. changxingensis zone and may be offset from primary seawater trends by approximately 1 ×10-4 (Fig. 11). In the Meishan Core, one sample from the H. parvus zone is approximately 0.7076, and samples are as high as

0.7080 in the Isarcicella isarcica zone. At the PTB section of Abadeh, Iran, all

Changhsingian samples have strontium isotopic values greater than 0.7072, and a strontium isotope sample from the I. isarcica zone is 0.707266 (Korte et al., 2003, 2004).

The I. isarcica zone at the PTB section of Zal, Iran produced less radiogenic values

(0.707217-0.707220), but this data set contains no data from the H. parvus zone. At Zal,

Iran (Sedlacek et al., in preparation; Chapter 1), Changhsingian values increase from

0.707072 to 0.707212, but direct correlation to Abadeh and Meishan is difficult due to the use of regional ammonoid biostratigraphy (Horacek et al., 2007; Richoz et al., 2010).

Whole rock versus brachiopod calcite and preservation of seawater 87Sr/86Sr

Well-preserved brachiopod LMC is the marine carbonate phase that is considered least susceptible to diagenetic alteration and arguably the best material for producing reliable Paleozoic seawater isotope curves (Korte et al., 2003, 2004, 2006; Kearsey et al.,

2009; Brand et al., 2012). However, suitable brachiopod LMC is not always available or present at high enough resolution to assess variations in the rate of 87Sr/86Sr change.

Indeed, the post-extinction latest Permian and Early Triassic recovery interval contains few brachiopods and a suitable alternative must be investigated. In the Permian-Triassic boundary interval, conodont elements (Martin and Macdougall, 1995; Korte et al., 2003,

2004, 2006; Twitchett, 2007) and whole rock (Huang et al., 2008; Saltzman and

Sedlacek, 2013) have been used and appear to preserve seawater 87Sr/86Sr trends despite the fact that some, or even most, of these samples may be altered to more radiogenic absolute values. Nonetheless, the degree of alteration is a controversial issue because few studies have directly compared brachiopods, conodonts and whole rock trends at sufficient stratigraphic resolution to address the offsets among the phases. Although

Brand et al. (2012) compared these phases and demonstrated that brachiopod calcite was the most reliable material, the sampling density of conodont and whole rock was limited in comparison to brachiopods. Our data from Dawen represent the highest resolution whole rock data known from the latest Permian to Early Triassic, and allows an updated assessment of the reliability of whole rock data for preservation of primary seawater

87Sr/86Sr trends.

The least radiogenic 87Sr/86Sr value from whole rock samples at Dawen is from the latest Permian Clarkina changxingensis-deflecta zone and is 0.707103 with a

25 relatively high Sr concentration of 427.3 ppm. This 87Sr/86Sr value is 0.38 ×10-4 and

0.43×10-4 higher than the least radiogenic, unaltered brachiopod samples from the

Clarkina changxingensis-deflecta zone in the Bulla Member at Val Brutta and at Sass de

Putia, respectively, in northern Italy (Fig. 11; Brand et al., 2012). This offset in 87Sr/86Sr of < 0.5 ×10-4 is significantly smaller than that based on the other whole rock data plotted by Brand et al. (2012), which were more radiogenic than brachiopods by an amount that varied between ~1×10-4 and 4×10-4. We conclude that although whole rock samples are more susceptible to diagenesis, careful analysis of samples with the highest Sr concentrations and least radiogenic values within a section may be used as a close approximation (< 0.5 ×10-4) of seawater 87Sr/86Sr when brachiopod LMC is unavailable, such as in much of the Early Triassic.

Calculation of changes in the rate of 87Sr/86Sr rise in the latest Permian

Least radiogenic data from Dawen and Zhongliang Mountain indicate that seawater 87Sr/86Sr during the Permian-Triassic boundary interval was between 0.7071 to

0.7072, and somewhat less than the value proposed by McArthur et al. (2012) (Fig 5).

Therefore, seawater 87Sr/86Sr value rose by 2 ×10-4 during the Lopingian and 11 ×10-4 during the Early Triassic. Using these absolute changes in 87Sr/86Sr values and applying the Permian and Triassic stage dates of Gradstein et al. (2012) (Fig. 6), the rate of seawater 87Sr/86Sr rise during the Lopingian was approximately 0.3 ×10-4/myr and of a similar magnitude to that of the Cenozoic. The rate of 87Sr/86Sr rise increased to 2.2 ×10-

4/myr during Early Triassic. This increase in the rate of rise occurred at the end of the

Permian and was likely due in part to elevated weathering rates during the extinction and recovery interval (Korte et al., 2007; Algeo and Twitchett, 2010). Errors in age assignments of the Early Triassic and late Permian stage boundaries may account for part of the accelerated rate of rise calculated for the post-extinction interval. However, maintaining the Lopingian rate of rise (0.3 ×10-4/myr) would require an expansion of this portion of the timescale by almost 30 million years.

Conclusions

Based on comparison of Sr isotope data from Dawen, China to other latest

Permian and PTB age 87Sr/86Sr values, we conclude that primary seawater 87Sr/86Sr at the

PTB interval was between 0.7071 and 0.7072, which is lower than that proposed by

McArthur et al. (2012) (Fig. 5). The rate of 87Sr/86Sr rise may have increased by an order of magnitude during the latest Permian interval. This increase in the rate of 87Sr/86Sr rise was likely related to environmental disturbances associated with the end-Permian mass extinction including enhanced weathering caused by a loss of vegetative cover resulting from the mass extinction on land (Algeo and Twitchett, 2010). Although future changes to the Early Triassic timescale may diminish these calculated rates, it is likely that this will remain the most rapid rate of change in seawater 87Sr/86Sr during the Phanerozoic.

Whole rock samples from the C. changxingensis-deflecta zone at Dawen are within 0.4 ×10-4 of contemporaneous well-preserved brachiopod LMC from the Italian

Alps published by Brand et al. (2012). A minor falling trend from the whole rock samples is also present in samples with least radiogenic 87Sr/86Sr and highest Sr

27 concentrations. Therefore, whole rock methods can record seawater trends, and well preserved whole rock samples may potentially retain primary seawater strontium isotopic values. The use of whole rock strontium isotope stratigraphy can be used to improve the resolution of the seawater 87Sr/86Sr curve in the Paleozoic and Mesozoic in sections where suitable brachiopod LMC is absent. The high rate of 87Sr/86Sr rise through the

Lopingian (Late Permian) and Early Triassic increases the utility of whole rock strontium chemostratigraphic methods to provide minimum age constraints using least radiogenic values.

Acknowledgements

Tom Algeo provided the samples from Dawen, and Richard Twitchett provided helpful discussion on correlation between PTB sections. Jeff Linder and Amanda

Howard assisted with strontium isotopic analyses.

0.7095

0.7090

0.7085

0.7080

0.7075

0.7070

Camb. Ord. Sil. Devonian Carb. Permian Triassic Jurassic Cretaceous Cenozoic Paleozoic Mesozoic

500 400 300 200 100 0 Age (Ma)

Figure 4. Seawater 87Sr/86Sr of the Phanerozoic. The interval investigated in this study, spanning the late Permian the Early-Middle Triassic boundary, is highlighted in grey. This represents the most rapid rise in seawater 87Sr/86Sr of the Phanerozoic. (After Kani et al., 2008).

0.7080

0.7078

0.7076

0.7074

0.7072

0.7070

0.7068 Rhaet. Norian Carn. Lad. Anis. Ol In C Wu. Cap. Late Middle Early Lopin. Gu. Triassic Permian

210 220 230 240 250 260 Age (Ma)

Figure 5. 87Sr/86Sr of late Permian through Triassic after McArthur et al. (2012). A rise in seawater 87Sr/86Sr occurred from near the end-Capitanian through the Early-Middle Triassic boundary (shaded interval). This interpretation of the seawater strontium isotope curve has a Permian Triassic boundary value near 0.7072. A significant slowing in the rate of rise occurs during the Late Permian (Wuchiapingian), but may be an artifact of timescale uncertainties. A slowing of the rate of rise (or even falling values) has also been proposed for the latest Permian by Twitchett (2007) and Brand et al. (2012), although this does not show up in the McArthur et al. (2012) curve. Gray dashed lines are drawn at 87Sr/86Sr= 0.7072, 0.7071 and 0.7070.

Figure 6. Global correlation charts for the Late Permian and Early Triassic taken from the individual Permian and Triassic chapters of the Geologic Timescale 2012 (Gradstein et al., 2012). Note that there is some disagreement between these two chapters in terms of the conodont zonation in the latest Permian.

Early Triassic Time Scale AGE Sub- Geo- Tethyan Boreal Age Conodonts (Ma) Age Mag Ammonoids Ammonoids 250 Rohillites Neospathodus rohilla Vavilovites dieneri (Morph 3) sverdrupi Gyronites Sweetospatho- frequens dus Proptychites kummeli “Pleurogyronites” candidus planidorsatus- Discophiceras Proptychites 251 rosenkrantzi Neogondollela strigatus Orphiceras krystyni Ophiceras tibeticum commune Isarcicella Otoceras isarcica woodwardi Upper Otoceras Otoceras boreale Hindeodus 252 fissisellatum parvus

Changh- Hypophiceras lower Clark. meishanensis H. praeparvus singian changxingense O. boreale Clarkina hauschkei

Permian Time Scale AGE Epoch/Age Polarity (Ma) (Stage) Chron Conodont Zonation Triassic Clarkina zhejiangensis- 252.2 GN1/ C. meishanensis sns Clarkina deflecta-C. yini Changh- Clarkina changxingensis singian Clarkina subcarinata 254.2 Clarkina wangi Clarkina longicuspidata 255 Clarkina orientalis

Clarkina transcaucasica

Wuchia- Clarkina guangyuanensis pingian Clarkina leveni Clarkina asymmetrica Clarkina dukouensis 259.8 Clarkina postbitteri postbitteri 260 Clarkina postbitteri hongshuiensis Jinogondolella granti Jinogondolella xuanhanensis

Jinogondolella prexuanhanensis Capitanian Jinogondolella altudaensis

Jinogondolella shannoni 265 265.1 Jinogondolella postserrata

Figure 6 32

Paleotethys 1 5 6 Panthalassa 3 Panthalassa 4

PANGEA Neotethys

Figure 7. Permian-Triassic boundary paleogeographic reconstruction after Algeo and Twitchett (2010). Permian-Triassic boundary sections analyzed for strontium isotopes considered in this study. South China Block: 1. Dawen, Great Bank of Guizhou, China (this study), 2. Meishan, South China (Cao et al., 2009); Iranian microcontinent: 3. Zal, northwestern Iran (Sedlacek et al., in preparation), 4. Abadeh, central Iran (Korte et al., 2003, 2004); Eastern Pangean margin: 5. Sass de Putia, Dolomites, Italy (Brand et al., 2012), 6. Val Brutta, Dolomites, Italy (Brand et al., 2012). Gray shaded area indicates shallow shelf environments.

A B Guiyang Yangtze Nanjing Platform

Wuhan Meishan

Guiyang Yangtze Block GBG Kunming Dawen Nanning Nanpanjiang Basin 0 350 km 0 100 km

Figure 8. A) The Nanpanjiang basin, part of the South China Block during the late Permian, is located southwest of Meishan, China. B) Dawen section is located on the Great Bank of Guizhou (GBG), one of several isolated carbonate platforms deposited in the Nanpanjiang Basin during the Permian-Triassic boundary interval. After Lehrmann et al. (2003).

Figure 9. 87Sr/86Sr of the Permian-Triassic boundary interval Dawen, GBG, South China. Samples were collected from the upper Wuchiaping Formation and the calcimicrobial framestone facies of the lower Daye Formation. Lowermost sample analyzed is from 400 cm below the dissolution surface, from within the Clarkina chanxingensis-deflecta zone of Chen et al. (2009). The base of the Hindeodus parvus sensu lato zone is defined by the first occurrence datum (FOD) of H. parvus (Chen et al., 2009). Filled symbols indicate Sr concentrations greater than 400 ppm, which are considered least altered samples (Fig. 10).

bio- Fm system stage Lith. meters zone 10

calcimicrobial framestone 8 fossiliferous packstone

Dissolution 0 Surface

-2

Falling trend?

-4

0.7070 0.7072 0.7074 -2 -1 0 1 2 3 4 87 86 13 Figure 9 Sr/ Sr  C

0.7075

Altered Least altered

0.7074

0.7073

0.7072

0.7071

0.7070 150 200 250 300 350 400 450 500 Sr ppm

Figure 10. Cross plot of 87Sr/86Sr and Sr ppm for the Dawen section. There is a negative correlation between Sr concentration and 87Sr/86Sr; therefore we consider samples with highest Sr concentrations the most likely to preserve seawater strontium isotopic trends. Filled symbols, with concentrations higher than 300 ppm, correspond to filled symbols in Figure 9. Dashed lines are drawn at 0.7071 and 0.7072. Shaded interval corresponds to samples with Sr concentration below 400 ppm that are considered altered.

Correlation to bio- Sass de Putia Val Brutta Meishan zone Bed 25- C. m.- 27b Unaltered bLMC H.eur. (Brand et al., 2012) Altered bLMC (Brand et al., 2012) Whole Rock (Dawen)

Bed 24e

20cm 5 cm

0.7070 0.7072 0.7074 0.7070 0.7072 0.7074 87Sr/86Sr 87Sr/86Sr

Figure 11. 87Sr/86Sr of Sass de Putia and Val Brutta brachiopod LMC analyzed by Brand et al. (2012). Unfiled circles indicate altered brachiopods; filled circles indicate well preserved brachiopods. Triangles represent data from Dawen, three data points from within the Clarkina changxingensis-deflecta conodont biozone, and one from the Clarkina meishanensis- Hindeodus eurypyge biozone. Dawen data are placed in correct stratigraphic order. However their ages may be relatively older or younger than Brand et al. (2012) data within the biostratigraphic age constraints. Grey dashed lines are drawn at 87Sr/86Sr = 0.7071 and 0.7072.

Figure 12. 87Sr/86Sr from Meishan, China (Cao et al., 2009), Abadeh, central Iran (Korte et al., 2003, 2004), and Zal, northwestern Iran (Sedlacek et al., in prep). Fields indicate correlation based on conodont biozones for each section. Gray lines are drawn at 87Sr/86Sr = 0.7071 and 0.7072. Each section indicates seawater 87Sr/86Sr greater than 0.7072 in the basal Triassic (H. parvus zone). Abbreviations for Meishan: Wuchiaping. = Wuchiapingian, changxingen. = Clarkina chanxingensis, I. isarcica = Isarcicella isarcica, ch.-pr. = C. chanxingensis yini - Hindeodus praeparvus, m. -e. = Clarkina meishanensis meishanensis- H. eurypyge, p. = Hindeodus parvus, s. = Isarcicella staeschei. Abadeh: trans. = C. transcaucasica, orientalis- medicons. = C. orientalis-C. mediconstricta, subc. = C. subcarinta, ch.d. = C. changxingensis- C. deflecta, i = C. iranica, h. = C. hauschkei, m.p. = C. meishanensis- C. praeparvus, p. = Hindeodus parvus. Zal: Ps. = Pseudotoceras, Ph. = Phisonites-Dhzulfites, Pa. = Paratirolites beds, m.-p. = C. meishanensis- H. praeparvus, H. parv = H. parvus

p. m.p. h. i. ? ? s.

40 p. m.-e. ch.-pr.

0.7070 0.7072 0.7074

Figure 12 40

Chapter 3: A unique facies succession through the Permian-Triassic boundary transition

interval (Gerster and Thaynes formations) from the Confusion Range, Utah

Abstract

An understanding of global events during the Permian-Triassic boundary interval is limited by the paucity of complete boundary sections from Panthalassa and western

Pangaea. A recent chemostratigraphic study indicated that a relatively continuous

Permian-Triassic boundary carbonate succession is present in the upper Gerster and lower Thaynes formations of the Confusion Range, Utah. The proposed boundary interval consists of several distinctive lithologies which are similar in many regards to some Tethyan sections that contain stromatolitic carbonate, clotted fabrics, and digitate structures within microbially mediated carbonates of the basal Triassic.

In the Confusion Range section, the Gerster Formation includes several hundred meters of cherty, fossiliferous packstone of Permian age. However, the uppermost 20 meters of the Gerster Formation in the Confusion Range that is thought to preserve the

Permian-Triassic boundary interval is unique in the Great Basin region. Lithologies within this interval include a cross bedded red sandstone overlain by a limestone unit containing large chert nodules in a recrystallized carbonate cement. Overlying the chert beds, the basal Thaynes is made up of a thin interval of laminated limestones with 41 upturned margins that may represent desiccation features known as tepee structures. This laminated unit also contains interbedded stromatolitic fabric, characterized by mm-scale wavy bedding. Thin sections of the laminated limestone reveal clotted textures and lobate structures consistent with a microbial origin. Above the tepees is a fenestral limestone containing coated grains. Thin sections from the fenestral limestone reveal pellets and ooids. The fenestral unit is capped by a resistant microgastropod-rich packstone and grainstone, which in turn is overlain by more typical green-gray, ammonite-bearing shaly limestones that are characteristic of the Thaynes Formation throughout the Great Basin region.

Although microbial characteristics are found within the Gerster-Thaynes transition of the Confusion Range section, future petrographic and chronostratigraphic studies are needed to determine if this microbial carbonate was deposited as part of a global microbialite event, or if these strata reflect the local depositional environment.

87Sr/86Sr analysis of the Gerster and Thaynes in the Confusion Range reveals a correspondence between the laminated and fenestral facies and highly radiogenic strontium isotope ratios, which indicates that primary seawater 87Sr/86Sr was not preserved due to the original mineralogy of the sediments and stabilization during diagenesis.

Introduction

The largest mass extinction of the Phanerozoic occurred during the latest Permian, resulting in the destruction of marine and terrestrial ecosystems (Raup and Sepkoski,

1982; Retallack, 1995; Erwin et al., 2002). Major changes in the carbon and strontium

42 isotopic composition of seawater are well documented from rocks of late Permian-Early

Triassic age, further attesting to the severity of the environmental disturbances and changes in the global carbon cycle that were associated with the biological mass extinction (Baud et al., 1989; Veizer, 1989; Martin and Macdougall, 1995; Korte et al.,

2003, 2004, 2006; Payne et al., 2004, Korte and Kozur, 2010; Saltzman and Sedlacek,

2013). In addition to these distinctive biological changes and isotopic shifts in the

Permian-Triassic boundary (PTB) interval, abrupt changes in sedimentary deposition have also been described. The observed changes in sedimentary deposition reveal important evidence of the timing and causes of mass extinction and global environmental changes on land and in the oceans.

On land, the most prominent sedimentary changes associated with mass extinction include the cessation of coal deposition (Retallack et al., 1996) and a shift from meandering to braided fluvial deposition systems (Ward et al., 2000). The destruction of terrestrial ecosystems and a warmer climate apparently limited plants that had formed coal deposits, and this plant loss also reduced thick cohesive soils that held river banks and meanders in place. Marine sedimentary depositional systems are also marked by changes in the post-extinction interval including an Early Triassic chert gap that may reflect the loss of planktonic taxa that produced siliceous microfossils (Henderson, 1997;

Isozaki, 1997; Wignall and Newton, 2003). In addition, much attention has focused on the significance of an observed reduction in the skeletal content of marine carbonates and coeval increase in the abundance of microbially mediated carbonate (microbialites) and oolite (Sano and Nakashima, 1997; Baud et al., 1997, 2005, 2007; Kershaw et al., 2002,

2007; Payne et al., 2006; Groves et al., 2007). The shift to microbialite deposition, which includes laminated stromatolites and clotted thrombolites (Riding, 2000), has been attributed to a reduction of metazoan grazers and opening of niche space to opportunistic microbes in post-extinction ecosystems (Fischer and Arthur, 1977; Schubert and Bottjer,

1992; Lehrmann, 1999), diminished skeletal carbonate deposition due to loss of shell- producing organisms (Baud et al., 1997), and an increase in the carbonate saturation of seawater due to the reduction of a skeletal carbonate sink (Kershaw et al., 2002; Heydari et al., 2003, Lehrmann et al., 2003).

Determining the global nature and significance of these shifts in carbonate depositional patterns and reduction of skeletal carbonate production is difficult because most studies are based on Tethyan PTB sections (e.g. China, Turkey, Iran, and Italy;

Lehrmann, 1999; Baud et al., 2007; Payne et al., 2007; Kershaw et al., 2002, 2007; Ezaki et al., 2008), and few continuous sections from western Pangea and Panthalassa are known (Fig. 13). Kershaw et al. (2007) noted the apparently synchronous deposition of microbialite deposits from the Tethys region in the basal Triassic Hindeodus parvus conodont biozone and called these deposits Earliest Triassic Microbialites (ETMs) to distinguish them from later Early Triassic microbialite deposits that may not be regionally synchronous. Although Early Triassic microbialites have been reported from the western United States (western Pangea) (e.g. Schubert and Bottjer, 1992, 1995; Pruss et al., 2006; Woods and Baud, 2008), these examples are Olenekian in age and are significantly younger than ETMs deposited immediately after the latest Permian mass extinction. ETMs in the Tethys formed in areas of upwelling where oxygen poor,

44 bicarbonate rich water stimulated microbial activity (Kershaw et al., 2007, 2009; Fig. 13), and although no ETMs are described from western Pangea, their presence was predicted due to the likely paleogeographic position within a zone of upwelling (Fig. 13; Kershaw et al., 2007).

Here, we examine a unique Late Permian to Early Triassic facies succession from western Pangea preserved in the Confusion Range of west-central Utah, which includes stromatolites, fenestral limestone, and ooids that may be time equivalent with the ETMs deposited in the Tethys. Although the PTB was long thought to be represented by a major unconformity in the Great Basin region (Newell, 1947; Hose and Repenning, 1958;

Collinson et al., 1976; Wardlaw and Collinson, 1978), a recent chemostratigraphic study indicated relatively complete deposition of a Late Permian to Early Triassic age shallow marine carbonate sequence in the Confusion Range (Saltzman and Sedlacek, 2013; Fig.

14). We present detailed descriptions of the Permian-Triassic boundary interval lithologies for the Gerster and Thaynes formations of the Confusion Range (Figs. 15, 16), including use of 87Sr/86Sr to address questions regarding original depositional environment, mineralogy and diagenesis of the preserved carbonate succession. Our results indicate a change in carbonate depositional styles in the Permian-Triassic boundary interval, similar to those deposited in the Tethys Ocean during the post extinction interval.

Background

Lithostratigraphy and general age of Gerster and Thaynes formations, Confusion Range,

Utah

In the Great Basin region, the latest Permian through earliest Triassic was long thought to be absent in marine carbonate successions (Newell, 1948; Hose and

Repenning, 1959; Collinson et al., 1967). However, some studies questioned the presence of a hiatus based on a lack of both physical evidence of an unconformity and preservation of diagnostic index fossils (Bissel, 1973; Alvarez and O’Connor, 2002). In the region, the thickest succession of the Permian Gerster Formation and Triassic

Thaynes Formation is located in the Confusion Range (Fig 14), which contains a significant interval of undated carbonate of possible latest Permian to earliest Triassic age

(Wardlaw and Collinson, 1978) (Figs. 15, 16).

The lower Gerster Formation in the Confusion Range section is typical of this formation throughout the Great Basin region, consisting of cherty limestone with brachiopod, bryozoan, and echinoderm fragments (Hose and Repenning, 1959).

However, the uppermost undated Gerster contains an expanded sequence of lithologies unique for the region (Figs. 15, 16). These unique lithologies begin with a fine-grained red sandstone that is a persistent marker bed within the region (Collinson et al., 1976).

Above the red sandstone is a limestone bed containing unusually large and abundant chert nodules. At the top of the cherty beds is a brecciated horizon that is overlain by fenestral limestone and laminated limestones (Collinson et al., 1976). These transitional lithologies (which are unlike any described from the typical Gerster or Thaynes formations in the surrounding Great Basin) are capped by a microgastropod wackestone

(Hose and Repenning, 1959), which is overlain by more typical green-gray, ammonite- bearing shaly limestones that are characteristic of the Thaynes Formation throughout the

Great Basin region. Although this general succession in the Confusion Range section was described by Collinson et al. (1976), these authors did not include detailed documentation of the unique lithofacies and the inferred depositional environments.

Conodont biostratigraphy and biozones

Collinson et al. (1976) also provided a general discussion of the Permian and

Triassic age assignments of the transition between the Gerster and Thaynes formations in the Confusion Range. Wardlaw and Collinson (1978) provided a more detailed analysis and used brachiopod and conodont biostratigraphy to assign ages to the Gerster

Formation. In the Confusion Range, the youngest conodont faunas reported from the

Gerster Formation contain Neogondolella bitteri and Neospathodus divergens that were assigned to the Wordian stage (Guadalupian) (Wardlaw and Collinson, 1978, 1986), although more recent work indicates that these faunas may be as young as Wuchiapingian

(Lopingian) (Henderson et al., 1997). However, at the Confusion Range section, the upper 79 meters of Gerster contain few fossils and could not be assigned to a biozone

(Fig. 15) (Wardlaw and Collinson, 1978).

Part of the confusion over the age of the transitional lithologies between the

Gerster and Thaynes in the Confusion Range is due to the changes in formational nomenclature assigned to these beds. Collinson et al. (1976) assigned the transitional beds above the red sandstone and cherty beds to the basal “Thaynes” but noted that the

47 contact with the underlying Gerster was difficult to discern and lacked relief. The age of the laminated, fenestral and microgastropod-rich beds is controversial. Collinson et al.

(1976) and Carr (1981) reported the Triassic (Smithian) conodont Parachirognathus ethingtoni from these beds in the Confusion Range. However, Wardlaw and Collinson

(1986) later assigned these same basal “Thaynes” beds to the uppermost “Gerster”, consistent with conodonts they identified as the Permian genus Merrillina in the microgastropod-rich wackestone facies. Because Merrillina is likely ancestral to

Parachirognathus (Orchard, 2007), taxonomic uncertainty regarding the timing and morphological definition of this transition can explain the discrepancy in age assignments of these key beds. Above the microgastropod-rich beds, the lower Thaynes consists of siltstone, mudstone and interbedded brownish-gray limestone beds typical of the region and containing abundant Meekoceras ammonites of Smithian age (Carr, 1981; Collinson et al., 1976). Because of the uncertainties in age estimates using fossils, chemostratigraphic stratigraphic methods were utilized by Saltzman and Sedlacek (2013) to establish minimum age estimates of the undated interval.

Chemostratigraphy

Marine carbonates preserve a negative carbon isotope excursion and a rapid rise in seawater 87Sr/86Sr during the PTB interval that have paleoceanographic significance but are also useful for global correlation independent of biostratigraphy. The negative

13 δ Ccarb excursion is widely used as a chemostratigraphic tool to correlate PTB sections throughout the Tethys (e.g., Korte et al., 2004) (Fig 17). Saltzman and Sedlacek (2013)

48 analyzed the Gerster and Thaynes formations of the Confusion Range for δ13C (Fig. 15,

16). At the Confusion Range section the negative carbon isotope excursion begins in the upper chert beds, and δ13C values reach their minimum within the interval containing laminated, fenestral, and oolitic facies (Saltzman and Sedlacek, 2013).

Because multiple negative carbon isotope excursions occurred during the Early

Triassic, strontium isotope chemostratigraphic methods were applied to provide minimum age estimates (Saltzman and Sedlacek, 2013; Fig. 18). Because diagenesis typically alters 87Sr/86Sr to more radiogenic values (Brand and Veizer, 1980), the least radiogenic 87Sr/86Sr from samples corresponding to the negative carbon isotope excursion were used to constrain minimum age estimates for the upper Gerster and lower Thaynes formations (Fig. 18). In the Confusion Range, the least radiogenic 87Sr/86Sr near the

13 onset of the negative δ Ccarb excursion is 0.70730. Because 0.70730 falls in between best estimates of seawater Sr for the base Triassic (0.7071–0.7072) and middle Induan

13 (~0.7074) (Korte et al., 2004), this indicates that the onset of the negative δ Ccarb excursion can be no younger than early Induan (the first stage of the Early Triassic).

Further support for this conclusion is that 87Sr/86Sr of 0.70766 from a younger sample within the δ13C minimum is lower than the 0.70774 from the latest Induan N. pakistani conodont zone (Fig. 18; Martin and MacDougall, 1995). Thus, the negative δ13C excursion is most likely latest Permian to earliest Triassic in age because the other Early

Triassic negative carbon isotope excursions are younger than Induan (Saltzman and

Sedlacek, 2013). Although Sr isotopes provide minimum age estimates, in this paper we

49 the use of Sr isotopes in the Confusion Range section as a sensitive indicator of depositional and diagenetic environments.

Methods

Lithologic study

The 19.5 m of transitional beds from the Gerster to the Thaynes formations were logged at cm scale from the base of the calcareous red sandstone through the top of the exposed section. The GPS coordinates of the section are 12 S 268459 4366086, which is the same outcrop described by Collinson et al. (1976) and was confirmed by J.W.

Collinson in the field. Samples were collected from within single beds, and contacts were determined by changes in lithologic fabric. Samples of representative lithologies were collected, photographed, and cut using a rock saw. Cut sides were then polished to reveal detailed lithologic fabrics. Petrographic thin sections were made from hand samples to further investigate the lithologic fabric.

Whole rock 87Sr/86Sr

Carbonate rock hand samples were prepared for isotopic analyses by removing all weathered surfaces. Then samples were cleaned with a sonicator using deionized water to remove loose sediment. Rock powders were generated by drilling with a preference for micritic limestone and avoidance of bioclasts, secondary veins, and cements.

Approximately 60 mg of rock powders were pretreated with ultrapure 1M ammonium acetate buffered to a pH of 8 (Montañez et al., 1996; Bailey et al., 2000). Samples were

50 then leached in 4% acetic acid and the resulting supernatant was collected and spiked with an 84Sr tracer. Sr was isolated using Biorad AG 50×8 cation exchange resin and a

2N HCl based ion exchange following the laboratory methods of Foland and Allen

(1991). Samples were loaded with HCl on a Re double filament configuration with a

Ta2O5 activator. Isotopic compositions were measured with a Finnegan MAT 261A thermal ionization mass spectrometer at The Ohio State University. The laboratory reproducibility of the standard SRM 987 is 0.710242 ± 0.000010 (1σ reproducibility).

Reported 87Sr/86Sr values are corrected for instrumental fractionation using normal

86Sr/88Sr of 0.119400.

Results

Lithologic study

Field relationships

In the Confusion Range, the lower 200 meters of Gerster Limestone contains abundant chert, and typical late Paleozoic skeletal grains including echinoderms, brachiopods, and bryozoans (Fig. 15). Rock types are mainly wackestones and packstones and similar to that in the Gerster Formation in the surrounding Great Basin region. The overlying Gerster-Thaynes transitional beds of the Confusion Range section contain 6 distinctive lithologies (A-F) that do not occur in the surrounding region (Fig.

16). At the base of the distinctive sequence is a prominent 3.5 m thick bed of fine- grained red calcareous sandstone containing high angle crossbeds (lithology A, Figs. 16,

19). Above the red sandstone is an approximately 6.5 m thick chert interval (lithology B)

51 that contains large chert nodules (Fig. 20). The carbonate cement of lithology B is grey in lower beds and tan in the upper beds (Fig 20 A). The size of the chert nodules decrease near the top of the bed, and the uppermost chert beds contain smaller bedded chert (Fig. 20 A). In a gully exposure overlying the chert beds are three distinctive lithofacies that correspond to lithologies C-E in Figure 16 (Fig. 21). The base of the gully section contains a cherty brecciated limestone bed (uppermost lithology B) overlain by a 0.2 m thick laminated limestone with stromatolitic textures and tepee structures (Fig.

22). Although these textures exhibit heterogeneity within the bed (Fig. 22A-C), the laminated limestones are grouped together as lithology C. The thinly bedded limestones are overlain by a 0.7 m thick fenestral limestone bed containing coated grains (lithology

D; Fig. 23 A). The top of the section is characterized by a microgastropod-rich packstone with possible digitate structures that is ~ 0.7 m thick (lithology E; Figs. 23B; 24). Above a minor covered interval, a recrystallized grainstone with pink-grey calcite crystals continues to the top of the section (lithology F). Above F are the more typical gray-green shaly, ammonite-bearing limestones of the Thaynes Formation as described throughout the Great Basin region.

Petrographic analysis and polished slabs

Thin sections of the typical Gerster limestone facies contain echinoderm, bryozoans and brachiopods (Fig. 25). Thin sections of the carbonate cement from the nodular chert interval of the uppermost Gerster (lithology B) have extensive recrystallization of primary fabrics (Fig. 26). Polished hand samples of the laminated

52 limestone (lithology C; Fig. 27A) reveal sub-millimeter scale light-dark banding with calcite cement infilling, and thin sections have recrystallization of primary micrite, with clotted fabrics, lobate margins, and a chambered structure (Fig. 27B-C). Hand samples of the fenestral unit (lithology D) reveal abundant coated grains up to 1 cm in diameter

(Fig. 28A), and thin sections show pellets, ooids and some clotted texture (Fig. 28B).

Polished hand samples of the digitate bed from lithology E reveals discoloration along fractures and abundant microgastropods throughout the sample (Fig. 29). Thin sections of lithology E collected above the digitate beds have recrystallization of primary micrite and shell fragments (Fig 30A). Samples from the recrystallized grainstone at the top of the transitional lithologies also contains recrystallized shell fragments, although shells are not visible in hand sample (Fig. 30B). All thin sections of lithologies C-E have extensive recrystallization and carbonate cements.

Strontium isotope analysis

Samples analyzed for carbon isotopes by Saltzman and Sedlacek (2013) were analyzed here for Sr isotopes and Sr concentrations to further characterize depositional environments and diagenetic stabilization of carbonate rock types in the Confusion Range section (Figs. 15, 16). Samples from the lower Gerster Formation yielded 87Sr/86Sr values as low as 0.70699. In the lower portion of the Gerster Limestone, 87Sr/86Sr increases to approximately 0.7071. In the upper 50 meters, 87Sr/86Sr increases more rapidly through the top of the section to approximately 0.7080 (Fig. 15). Sr

53 concentrations are low (less than 250 ppm) in all Confusion Range samples, and 87Sr/86Sr weakly covaries with Sr concentrations (Fig. 31).

Within the transitional lithologies A-F described above, 87Sr/86Sr rises generally from 0.7072 to 0.7080. However, superimposed on this rise, 87Sr/86Sr peaks at the most radiogenic values in the laminated limestone, fenestral, and lower microgastropod packstone beds (lithologies C-E) before falling again in lithology F. Sr concentrations are less than 150 ppm in the majority of samples from this interval. 87Sr/86Sr and strontium concentrations through this interval have no covariation (Fig. 32).

Discussion

The lithologic descriptions and associated Sr isotope values in the Confusion

Range section are assumed herein to include the latest Permian extinction horizon and

PTB interval based on the chemostratigraphic study by Saltzman and Sedlacek (2013).

First, the evidence for sea level changes based on depositional environments is discussed and correlated with the better dated sections in the Tethys. Secondly, the possibility of microbialite deposition in the Confusion Range section beds C and D is discussed in the context of the causes of the latest Permian extinction event. Thirdly, the implications of new Sr isotope data for carbonate environments and diagenesis are discussed. Although subsequent studies may ultimately provide new age estimates for some of the lithologies discussed below in the Confusion Range section, the uniqueness of these lithologies in the context of the Late Permian to Early Triassic of western North America (western

Pangea) attest to their significance for this general time period.

Depositional environment of the Gerster-Thaynes transition beds and sea level change

The lower Gerster Formation is a cherty fossiliferous packstone containing abundant brachiopods, bryozoans, and echinoderms and interpreted to represent open marine conditions (Fig. 25). The overlying red sandstone bed (lithology A) represents a clear break in the depositional pattern of open marine limestones that proved useful for mapping purposes in the Confusion Range (Hose and Repenning, 1959), and likely represents a sea level lowstand. This prominent lowstand may correlate with a latest

Permian sequence boundary recognized in the Tethys prior to the mass extinction horizon

(Collinson et al., 1976; Payne et al., 2007; Saltzman and Sedlacek, 2013). Although this red sandstone bed may represent a hiatus in deposition between the typical Gerster

Formation and transitional lithologies, continuity of the δ13C and 87Sr/86Sr records (Figs.

15, 16) across this unit indicate that relatively little time is missing in this section.

The presence of chert (lithology B) above the red sandstone likely represents an increase in sea level and a return to open, possibly deep marine conditions, which is consistent with a latest Permian transgression recognized in the Tethys region (Payne et al., 2007). However, extensive recrystallization of the micrite within the chert interval indicates diagenetic alteration, and thin sections contain no recognizable fossil content that may indicate water depth (Fig. 26). Chert nodules decrease in size toward the top of this bed (Fig. 20), and the uppermost portion of the chert bed contains small angular bedded chert clasts and brecciated cherty limestone that are interpreted as reworked from lower chert beds (Fig. 20A). The cessation of chert deposition during the Early Triassic

55 is well documented in many parts of the world (Henderson, 1997; Isozaki, 1997; Wignall and Newton, 2003; Sperling and Ingle, 2006); therefore these chert beds are interpreted as having been deposited during the global transgression of latest Permian age. However, we also note that Mata and Woods (2008) described similar nodular features within micritic limestone from the Union Wash Formation of Olenekian age, but attributed them to bioturbation and differential compaction of silty layers. Future investigation of lithology B is required to determine probable origins of the large nodules present in the

Confusion Range.

The gully exposure above the last chert bed in the upper Gerster Formation contains three distinct lithologies (C-E), which were likely deposited in shallow to transitional marine environments and indicate a relative fall in sea level in this area. This interpretation is supported by the presence of a brecciated horizon, tepee structure desiccation features (lithology C), and the fenestral fabric and coated grains of lithology

D. Alternatively, these features, here interpreted as deposition during a lowstand in sea level, may indicate a submarine dissolution surface (Payne et al., 2007; Collin et al.,

2009). The overlying microgastropod packstone (lithology E) likely represents a return to more open marine conditions during sea level rise. Thin sections from the uppermost limestone beds (lithology F) reveal abundant recrystallized fossil content not visible in hand sample, and this bed is interpreted to be an extensively altered portion of the microgastropod packstone (Fig. 30) possibly consistent with an exposure surface during a subsequent relative sea level fall. Although these relative sea level changes in the

Confusion Range section may reflect global sea level, the role of tectonics and sediment

56 supply are important to constrain (e.g., Collinson et al., 1976) in future studies and comparisons with Tethyan sections.

Early Triassic Microbialites and Confusion Range lithologies C and D

The rocks assigned to lithologies C and D in the Confusion Range section are unique in containing evidence of microbially influenced carbonate deposition, including stromatolitic textures and coated grains. These strata may represent the latest Permian mass extinction horizon that is marked in the Tethys region by the loss of skeletal carbonate producing taxa and a shift to non-skeletal marine carbonate deposition (Baud,

1997; Payne et al., 2006), as evidenced by multiple microbialite deposits in well-dated

PTB successions in the South China block (Kershaw et al., 1999, 2002; Lehrmann, 1999;

Lehrmann et al., 2003; Ezaki et al., 2003; Payne et al., 2006), Iran (Heydari et al., 2003), southern Turkey (Baud et al., 2005), and Hungary (Hips and Haas, 2006). Although samples collected from lithologies C-E exhibit recrystallization in thin section, potential microbial structures are present (Figs. 27B-C, Fig. 28B) and their textures are comparable to those of microbial origin from the Tethys region. Lithology C contains irregular boundaries between light and dark layers visible in the polished hand sample (27A), which is similar to planar stromatolites from Early Triassic successions in Hungary

(Hips and Haas, 2006). Lobate margins and clotted textures in thin sections of lithology

C (Fig. 27B, C) are similar to textures described from microbial units in Tethyan sections

(Baud et al., 2005; Hips and Haas, 2006; Kershaw et al., 2007). A chambered structure in thin section from lithology C (Fig. 27C) resembles microbial Renalcis structures from

57 sections deposited on the Great Bank of Guizhou in South China (Lehrmann, 1999).

However, this may also represent an inorganic structure formed by neomorphism of calcite crystals, and further analysis is needed to confirm a microbial origin (B. Pratt, personal communication). The fenestral and ooidal fabric of lithology D is consistent with those in Tethyan Early Triassic sections including the Tesero Oolite Member from

Italy (Groves et al., 2007) and oolitic units from Turkey (Baud et al., 2005) and Iran

(Heydari et al., 2003). In the Bulla section, Italy, the Tesero Oolite Member contains oolitic packstone and grainstone of latest Permian age, and stromatolites are reported from the H. parvus zone (Groves et al., 2007). Sections from the Taurus Mountains,

Turkey, have an abrupt facies change from fossiliferous cherty limestone to oolitic units, which are overlain by calcimicrobial caprocks containing stromatolitic, thrombolitic, and fenestral textures (Baud et al., 2005). Similar sequences from the Great Bank of Guizhou

(GBG), South China contain cherty fossiliferous packstones of late Permian age overlain by calcimicrobial framestones. However, Nützel and Schulbert (2005) provided evidence of an apparent fenestral texture in the Sinbad Limestone on Olenekian age in Utah but attribute this texture to microstromatolites with sparitic cavities. Therefore, it is possible that what is described here as fenestral texture in lithology D is also of stromatolitic origin.

Earliest Triassic Microbialites (ETMs), including stromatolitic, thrombolitic, and clotted textures and abundant coated grains (Sano and Nakashima, 1997; Lehrmann,

2003; Baud et al., 2005, 2007; Groves et al., 2007; Ezaki et al., 2008), are known to occur within the H. parvus zone of the basal Triassic and were deposited during rising sea

58 levels after the late Permian lowstand (Kershaw et al., 2007). ETMs are thought to have formed at upwelling zones where dysoxic water supersaturated with carbonate promoted microbially mediated carbonate precipitation (Fig. 13; Kershaw et al., 2007). Warm temperatures in the late Permian and Early Triassic enhanced continental weathering rates increasing the riverine input of calcium and bicarbonate ions to the oceans which further promoted CaCO3 formation (Kershaw et al., 2007; Algeo and Twitchett, 2010;

Sun et al., 2012). The presence of ETMs on the western margin of Pangea was predicted by Kershaw et al. (2007) based on ocean circulation models (Fig. 13). The oolitic, fenestral bed and planar stromatolitic fabrics are similar to those documented from other

Permian-Triassic boundary age carbonates. However, although similarities between the

Confusion Range and Early Triassic Tethyan sections exist, it remains uncertain whether these microbialite deposits are part of a global microbial event, or simply reflect the local depositional environment. In addition, although younger Early Triassic microbialites have been described from the western United States (Schubert and Bottjer, 1992, 1995;

Pruss et al., 2006), the possible deposition of stromatolitic carbonate in the Confusion

Range would be the first reported occurrence of earliest Triassic age from the western

Pangean margin.

87Sr/86Sr of transitional lithologies

The 87Sr/86Sr of carbonate rocks is potentially a sensitive indicator of both primary depositional environment (marine versus non-marine) and secondary diagenetic stabilization of original carbonate mineralogy (aragonite versus calcite). 87Sr/86Sr values

59 of the lower Gerster limestone are internally coherent and the rising trend is similar to that of the global Late Permian seawater trend (Veizer et al., 1999; Korte et al., 2003,

2004), consistent with open marine deposition. However, low concentrations of strontium are indicative of diagenetic alteration (Brand and Veizer, 1980), possibly of original aragonitic carbonate mineralogy, which is unstable over million year time periods (Stanley and Hardie, 1999). Therefore, although the 87Sr/86Sr of these open marine successions may preserve seawater trends (Saltzman and Sedlacek, 2013), they do not preserve primary seawater 87Sr/86Sr values (Veizer, 1989).

The general rising seawater 87Sr/86Sr trend is interrupted in the upper portion of the chert interval (lithology B) and by a shift to more radiogenic values that likely represents a departure from seawater trends. These enriched 87Sr/86Sr values continue through the top of the microgastropod packstone. The observed deviation from primary seawater trends occurs within lithologies that were likely deposited in shallow to transitional marine environments (Fig. 16). This may indicate intermittent mixing with non-marine water masses with enriched 87Sr/86Sr values from weathered continental source rocks. Alternatively, the extreme enrichment in 87Sr/86Sr values for the succession beginning in the upper part of lithology B in the Confusion Range could indicate a larger percentage of original aragonite relative to beds above and below because aragonite is more easily altered compared to calcite. This is consistent with very low Sr concentrations throughout this succession.

Implications and Conclusions

This study represents the first detailed description of the Gerster-Thaynes transitional lithologies from the Confusion Range of Utah. The observed changes in lithologies from the typical (open marine) Gerster Formation to lithologies A-F represent a major change in depositional environment across the PTB from the western Pangean margin. These facies are similar to those from well-dated sections of the Tethys that formed after the late Permian mass extinction. The stromatolitic fabrics and clotted textures preserved in these transitional beds are considered microbial in origin based on comparison with Early Triassic microbialites from the Tethys. The presence of microbialite in the Confusion Range is consistent with the predictions of Kershaw et al.

(2007), and these are the oldest Early Triassic microbialites documented from the Great

Basin region. However, further investigation is needed to determine whether the deposition of microbially mediated carbonate in the Confusion Range was part of a global microbial event, or simply reflects the local depositional environment. Regardless, these unusual lithologies represent carbonate deposition in the Panthalassic Ocean during the late Permian extinction and Early Triassic recovery interval. Therefore, this succession provides material to test hypotheses regarding the global nature of events that occurred during this interval of extreme environmental disruption.

Acknowledgments

Steve Kershaw, Brian Pratt, Clint Cowan, and Soo Yeun Ahn provided helpful discussion of microbialite fabrics. Jim Collinson provided guidance in the field. Bill

Ausich, Soo Yeun Ahn, Cole Edwards, and Cristina Millan assisted with photographs and

61 photomicrographs. Jeff Linder and Amanda Howard helped with strontium isotopic analyses, and Ken Foland provided helpful discussion of strontium isotope records. Julie

Codispoti, Kevin Crawford, Cole Edwards, Jeremy Gouldey, Amanda Howard, Chris

Sedlak, Robert Swift, T.J. Woods assisted with field collections. This research was supported by a Geological Society of America student research grant.

S Paleo- S tethys Confusion Range U Panthalassa U U Panthalassa

PANGEA S Neotethys

Figure 13. Paleogeographic reconstruction of Early Triassic Earth and modeled ocean circulation patterns. The Confusion Range section was deposited on the western margin of Pangea. Red arrows indicating ocean circulation after Kershaw et al. (2007), are based on the model of Kidder and Worsley (2004). Areas of upwelling coincide with many Earliest Triassic Microbialites (ETMs). U= upwelling, S= sinking. Shaded area represents shallow shelf environment. Map after Algeo and Twitchett (2010).

115 114 113 W

Wells I-80 Salt Lake City

41 N Reno

93 I-15

I-80

30 mi

40 Nevada Utah

50 Confusion Range Ely 6 6 39

Figure 14. Map showing location of the Confusion Range study area in west-central Utah, in the Great Basin region, United States

silty limestone fenestral limestone, coated grains, laminae nodular chert 40 m sandstone

limestone

0.7070 0.7074 0.7078 -3 -2 -1 0 1 2 3 4 87 86 13 Sr/ Sr  Ccarb Saltzman and Sedlacek (2013)

Figure 15. 87Sr/86Sr and d13C of the Gerster and Thaynes formations of the Confusion Range. Regionally, the thickest Gerster-Thaynes succession is preserved at the Confusion Range. The lower Gerster Formation is a cherty fossiliferous limestone, which is overlain by a series of transitional lithologies highlighted in grey. This interval is expanded in Figure 16.

E PTB interval D C Grainstone Microgastropod rich wackestone Fenestral limestone Thinly bedded limestone B Brecciated cherty limestone Bedded chert 5 m Nodular chert-tan matrix Nodular chert-grey matrix A Red sandstone 0.7072 0.7076 0.7080 -4 -3 -2 -1 0 1 2 87Sr/86Sr 13C

Figure 16. Detailed measured section of the Gerster-Thaynes transitional lithologies A-F. Strontium isotope analyses reveal a shift to more radiogenic values in lithologies C-E, and the rising trend of seawater 87Sr/86Sr is indicated by the dashed line through the least radiogenic values. Filled symbols indicate Sr concentrations greater than 100 ppm, however, these samples with highest concentrations occur within samples that deviate from the seawater trend. The interval of cover within lithology E occurs between the top of the gully section and the base of the next exposed limestone ledge.

Figure 17. The negative carbon isotope excursion present in most Permian-Triassic boundary sections is used to correlate the Confusion Range section to the Guryul Ravine, Pakistan and the GSSP for the Permian-Triassic boundary at Meishan, China (Saltzman and Sedlacek, 2013).

Figure 17 Figure

0.7082

i 0.7080 h

0.7078 87Sr/86Sr g f e 0.7076 This study a (see Fig. 2) d 0.7074 Twitchett (2007) Dolomites c Twitchett (2007) b Meishan a 0.7072 Martin & Mac- dougall (1995) Korte et al. (2003) Korte et al. (2004) 0.7070 waagen. pakist. dieneri carina. isarci. parvus changx.-praep. Olenek. (Smith.) Induan Late Permian 250 Ma Early Triassic 252.2 Ma

Figure 19. Field photograph of red calcareous sandstone (lithology A). Rock hammer for scale.

Figure 21. Field photograph of the gully interval containing laminated limestones (lithology C), fenestral limestone bed (lithology D) and microgastropod packstone with potential digitate fabric (lithology E). The base of this section is a brecciated limestone containing angular chert fragments, which represents the top of lithology B. Hammer for scale.

Figure 22. Field photograph of laminated limestone (lithology C). A) Teepee structures of Collinson et al. (1976) are desiccation structures, field notebook is 19 cm wide. B) Laminar beds with secondary calcite infilling between micritic layers. Beds show minor offset, possibly as part of a desiccation structure (US dime for scale). C) Sub millimeter scale wavy bedding within the laminar limestone is potentially microbial in origin (US dime for scale) occur at the base of lithology C immediately above the brecciated limestone that characterizes the upper portion of lithology B.

Figure 22

Figure 23. Field photograph of fenestral limestone unit (lithology D). A) Lower portion of fenestral unit overlying Bed C. Pen for scale is 15 cm long. B) Contact between lithology D and lithology E. Ruler for scale (16 cm) is placed at near the base of potential digitate structures.

Figure 24. Field photographs of gully exposure with lithologic contacts outlined in yellow. A) The base of the gully exposure consists of the highest chert bed (lithology B). This is overlain by lithology C, along an irregular contact. The fenestral fabric of lithology D directly underlies the microbial grainstone (lithology E) with potential digitate structure. The ruler for scale placed on laminations limestone (lithology C) is 16 cm long. B) Flat truncation of potential digitate structures occurs just below field notebook (19 cm long).

Figure 25 Photomicrographs of the typical Gerster Formation from the Confusion Range (photo width 3 mm for all images). A) Bryozoan from typical Gerster Formation at the Confusion Range collected approximately 19 m below the base of the red sandstone. B) Brachiopod, bryozoan and echinoderm fossil fragments from the Gerster Formation 0.5 m below the base of the red sandstone. C) Echinoderm fragment from 0.5 m below the base of the red sandstone.

Figure 27. Polished section and photomicrographs from lithology C. A) Polished slab of lithology C shows sub-millimeter light-dark alternating bands of micritic limestone. B) Photomicrograph of lithology C with arrow indicating chambered structure similar to Renalcis, but this structure may be caused by carbonate neomorphism. Photo is 3 mm wide. C) Lobate margins and clotted texture indicate microbial origin. Photo width is 3 mm.

Figure 28. A) Hand sample from lithology D show large coated grains. B) Photomicrograph of lithology D shows abundant ooids and pellets fabric consistent with a microbial origin. Photo is 3 mm wide.

250

200

150

100

50 0.7068 0.7070 0.7072 0.7074 0.7076 0.7078 0.7080

87 86 Sr/ Sr

Figure 31. Cross plot of 87Sr/86Sr of the Confusion Range shows a weak negative correlation, and is indicative of diagenetic alteration. Sr concentrations are low (below 250) for all samples.

200

150

100

50 0.7072 0.7074 0.7076 0.7078 0.708 0.7082 87Sr/86Sr

References

Chapter 1

Algeo, T.J., and Twitchett, R.J., 2010, Anomalous Early Triassic sediment fluxes due to

elevated weathering rates and their biological consequences: Geology, v. 38, p.

1023-1026.

Algeo, T.J., Chen, Z.Q., Fraiser, M.L., and Twitchett, R.J., 2011, Terrestrial-marine

teleconnections in the collapse and rebuilding of Early Triassic marine ecosystems:

Palaeogeography, Palaeoclimatology, Palaeoecology, v. 308, p. 1-11.

Algeo, T.J., Henderson, C.M., Tong, J., Feng, Q., Yin, H., and Tyson, R.V., 2012,

Plankton and productivity during the Permian-Triassic boundary crisis: An analysis

of organic carbon fluxes: Global and Planetary Change, doi:

10.1016/j.globplacha.2012.02.008.

Berner, R.A., 2004, The Phanerozoic Carbon Cycle: CO2 and O2. Oxford, Oxford

University Press, 150 pp.

Brand, U., Posenato, R., Came, R., Affek, H., Angiolini, L., Amzy, K., and Farabegoli,

E., 2012, The end-Permian mass extinction: A rapid volcanic CO2 and CH4-

climatic catastrophe: Chemical Geology, v. 322-323, p. 121-144.

Brand, U., and Veizer, J., 1980, Chemical diagenesis of a multicomponent carbonate

system-1: trace elements: Chemical Geology, v. 50, p. 1219-1236.

Denison, R.E., Koepnick, R.B., Fletcher, A., Howell, M.W., and Callaway, W.S., 1994,

Criteria for the retention of original seawater 87Sr/86Sr in ancient shelf limestones:

Chemical Geology, v. 112, p. 131-143.

Erwin, D.H., Bowring, S.A., and Jin, Y.G., 2002, The end-Permian mass extinctions, in

Koeberl, C., and MacLeod, K.G., eds., Catastrophic Events and Mass Extinctions:

Impacts and Beyond: Boulder, Colorado, Geological Society of America Special

Papers 356, p. 363-383.

Galfetti, T., Bucher, H., Ovtcharova, M., Schaltegger, U., Brayard, A., Brühwiler, T.,

Goudemand, N., Weissert, H., Hochuli, P.A., Cordey, F., and Guodun, K., 2007,

Timing of the Early Triassic carbon cycle perturbations inferred from new U-Pb

ages and ammonoid biochronozones: Earth and Planetary Science Letters, v. 258, p.

593-604.

Gradstein, F.M., Ogg, J. G., Schmitz, M., and Ogg, G., eds., 2012, The Geologic Time

Scale 2012: Boston, MA, Elsevier, 1176 pp.

Hodell, D.A., Mead, G.A., and Mueller, P.A., 1990, Variation in the strontium isotopic

composition of seawater (8 Ma to present): Implications for chemical weathering

rates and dissolved fluxes to the oceans. Chemical Geology, v. 80, p. 291-307.

Horacek, M., Brandner, R., and Abart, R., 2007b, Carbon isotope record of the P/T

boundary and the Lower Triassic in the Southern Alps: evidence for rapid changes

in storage of organic carbon; correlation with magnetostratigraphy and

biostratigraphy: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 252, p.

347-354.

Horacek, M., Richoz, S., Brandner, R., Krystyn, L., and Spötl, C., 2007a, Evidence for

recurrent changes in Lower Triassic oceanic circulation of the Tethys: The δ13C

record from marine sections in Iran: Palaeogeography, Palaeoclimatology,

Palaeoecology, v. 252, p. 355-369.

Huang, S., Qing, H., Huang, P., Hu, Z., Wang, Q., Zou, M., and Liu, H., 2008, Evolution

of strontium isotopic composition of seawater from Late Permian to Early Triassic

based on study of marine carbonates, Zhongliang Mountain, Chongqing, China:

Science in China Series D: Earth Sciences, v. 51, p. 528-539.

Korte, C., Jasper, T., Kozur, H.W., and Veizer, J., 2006, 87Sr/86Sr record of Permian

seawater: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 240, p. 89-107.

Korte, C., Kozur, H.W., Bruckschen P., and Veizer, J., 2003, Strontium isotope evolution

of Late Permian and Triassic seawater: Geochimica et Cosmochimica Acta, v. 67, p.

47-62.

Korte, C., Kozur, H.W., Joachimski, M.M., Strauss, H., Veizer, J., and Schwark, L.,

2004, Carbon, sulfur, oxygen and strontium isotope records, organic geochemistry

and biostratigraphy across the Permian/Triassic boundary in Abadeh, Iran:

International Journal of Earth Science, v. 93, p. 565-581.

Kozur, H.W., 2007, Biostratigraphy and event stratigraphy in Iran around the Permian-

Triassic Boundary (PTB): Implications for the causes of the PTB biotic crisis:

Global and Planetary Change, v. 55, p. 155-176.

Looy, C.V., Brugman, W.A., Dilcher, D.L., and Visscher, H., 1999, The delayed

resurgence of equatorial forests after the Permian-Triassic ecological crisis:

Proceedings of the National Academy of Sciences (U.S.A.), v. 96, p. 13,857-13,862.

Martin, E.E., and Macdougall, J.D., 1995, Sr and Nd isotopes at the Permian/Triassic

boundary: A record of climate change: Chemical Geology, v. 125, p. 73-99.

Montañez, 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 87Sr/86Sr:

Geology, v. 24, p. 917-920.

Ovtcharova, M., Bucher, H., Schaltegger, U., Galfetti, T., Brayard, A., and Guex, J.,

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, v. 243, p. 463-475.

Palmer, M.R., and Edmond, J.M., 1989, The strontium isotope budget of the modern

ocean: Earth and Planetary Science Letters, v. 92, p. 11-26.

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, v. 256, p. 264-277.

Payne, J.L., Lehrmann, D.J., Wei, J., and Knoll, A., 2006, The pattern and timing of

biotic recovery from the end-Permian extinction on the Great Bank of Guizhou,

Guizhou Province, China: Palaios, v. 21, p. 63-85.

Payne, J.L., Lehrmann, D. J., Wei, J., Orchard, M.J., Schrag, D.P., and Knoll, A.H., 2004,

Large perturbations in the carbon cycle during recovery from the end-Permian

extinction: Science, v. 305, p. 506-509.

Retallack, G.J., 1995, Permian-Triassic extinction on land: Science, v. 267, p. 77-80.

Richoz, S., Leopold, K., Baud, A., Brandner, R., Horacek, M., and Mohtat-Aghai, P.,

2010, Permian-Triassic boundary interval in the Middle East (Iran and N. Oman):

Progressive environmental change from detailed carbonate carbon isotope marine

curve and sedimentary evolution: Journal of Asian Earth Sciences, v. 39, p. 236-253.

Sengör, A.M.C., 1984, The Cimmeride Orogenic System and the tectonics of Eurasia:

Geological Society of America, Special Paper 195, p. 1-82.

Song, H.J., Wignall, P.B., Chen, Z.Q., Tong, J.N., Bond, D.P.G., Lai, X.L., Zhao, X.M.,

Jian, H.S., Yan, C.B., Niu, Z.J., Chen, J., Yang, H., and Wang, Y.B., 2011,

Recovery tempo and pattern of marine ecosystems after the end-Permian mass

extinction: Geology, v. 39, p. 739-742.

Stanley, S.M., 2009, Evidence from ammonoids and conodonts for multiple Early

Triassic mass extinctions: Proceedings of the National Academy of Sciences

(U.S.A.), v. 106, p. 15264-15267.

Sun, Y.D., Joachimski, M.M., Wignall, P.B., Yan, C.B., Chen, Y.L., Jiang, H.S., Wang,

L.N., and Lai, X.L., 2012, Lethally hot temperatures during the Early Triassic

greenhouse: Science, v. 338, p. 366-370.

Veizer, J., 1989, Strontium isotopes in seawater through time: Annual Reviews of Earth

and Planetary Science, v. 17, p. 141-167.

Veizer, J., Ala, D., Azmy, K., Bruckshen, P., Buhl, D., Bruhn, F., Carden, G.A.F.,

Diener, A., Ebneth, S., Goddéris, Y., Jasper, T., Korte, C., Pawellek, F., Podlaha,

O.G., and Strauss, H., 1999, 87Sr/86Sr, δ13C and δ18O evolution of Phanerozoic

seawater: Chemical Geology, v. 161, p. 59-88.

Veizer, J., and Compston, W., 1974, 87Sr/86Sr composition of seawater during the

Phanerozoic: Geochimica et Cosmochimica Acta, v. 38, p. 1461-1484.

Waltham, D., and Gröcke, D.R., 2006, Non-uniqueness and interpretation of the seawater

87Sr/86Sr curve: Geochimica et Cosmochimica Acta, v. 70, p. 384-394.

Ward, P.D., Montgomery, D.R., and Smith, R., 2000, Altered river morphology in South

Africa related to the Permian-Triassic extinctions: Science, v. 289, p. 1740-1743.

Chapter 2

Algeo, T.J., Chen, Z.Q., Fraiser, M.L., and Twitchett, R.J., 2011, Terrestrial-marine

teleconnections in the collapse and rebuilding of Early Triassic marine ecosystems:

Palaeogeography, Palaeoclimatology, Palaeoecology, v. 308, p. 1-11.

Algeo, T.J., and Twitchett, R.J., 2010, Anomalous Early Triassic sediment fluxes due to

elevated weathering rates and their biological consequences: Geology, v. 38, p.

1023-1026.

Bailey, T.R., McArthur, J.M., Prince, H., and Thurwall, M.F., 2000, Dissolution methods

for strontium isotope stratigraphy: whole rock analysis: Chemical Geology, v. 167,

p. 313-319.

Berner, R.A., 2004, The Phanerozoic Carbon Cycle: CO2 and O2. Oxford, Oxford

University Press, 150 pp.

Brand, U., Posenato, R., Came, R., Affek, H., Angiolini, L., Amzy, K., and Farabegoli,

E., 2012, The end-Permian mass extinction: A rapid volcanic CO2 and CH4- climatic

catastrophe: Chemical Geology, v. 322-323, p. 121-144.

Brand, U., and Veizer, J., 1980, Chemical diagenesis of a multicomponent carbonate

system-1: trace elements: Chemical Geology, v. 50, p. 1219-1236.

Burke, W.H., Denison, R.E., Hetherington, E.A., Koepnick, R.B., Nelson, H.F., and Otto,

J.B., 1982, Variation of seawater 87Sr/86Sr throughout Phanerozoic time: Geology, v.

10, p. 516-519.

Cao, C., Love, G.D., Hays, L.E., Wang, W., Shen, S., and Summons, R.E., 2009,

Biogeochemical evidence for euxinic oceans and ecological disturbances presaging

the end-Permian mass extinction event: Earth and Planetary Science Letters, v. 281,

p. 188-201.

Chen, J., Beatty, T.W., Henderson, C.M., and Rowe, H., 2002, Conodont biostratigraphy

across the Permian-Triassic boundary at the Dawen section, Great Bank of Guizhou,

Guizhou Province, South China: Implication for the Late Permian extinction and

correlation with Meishan: Journal of Asian Earth Sciences, v. 36, p. 442-458.

Denison, R.E., Koepnick, R.B., Fletcher, A., Howell, M.W., and Callaway, W.S., 1994,

Criteria for the retention of original seawater 87Sr/86Sr in ancient shelf limestones:

Chemical Geology, v. 112, p. 131-143.

Erwin, D.H., Bowring, S.A., and Jin, Y.G., 2002, The end-Permian mass extinctions, in

Koeberl, C., and MacLeod, K.G., eds., Catastrophic Events and Mass Extinctions:

Impacts and Beyond: Boulder, Colorado, Geological Society of America Special

Papers 356, p. 363-383.

Foland, K.A., and Allen, J.C., 1991, Magma sources for Mesozoic anorogenic granites of

the White Mountain magma series, New England, USA: Contributions to

Mineralogy and Petrology, v. 109, p. 195-211.

Hodell, D.A., Mead, G.A., and Mueller, P.A., 1990, Variation in the strontium isotopic

composition of seawater (8 Ma to present): Implications for chemical weathering

rates and dissolved fluxes to the oceans: Chemical Geology, v. 80, p. 291-307

Horacek, M., Richoz, S., Brandner, R., Krystyn, L., and Spötl, C., 2007, Evidence for

recurrent changes in Lower Triassic oceanic circulation of the Tethys: The δ13C

record from marine sections in Iran: Palaeogeography, Palaeoclimatology,

Palaeoecology, v. 252, p. 355-369.

Huang, S., Qing, H., Huang, P., Hu, Z., Wang, Q., Zou, M., and Liu, H., 2008, Evolution

of strontium isotopic composition of seawater from Late Permian to Early Triassic

based on study of marine carbonates, Zhongliang Mountain, Chongqing, China:

Science in China Series D: Earth Sciences, v. 51, p. 528-539.

Kani, T., Fukui, M., Isozaki, Y., and Nohda, S., 2008, The Paleozoic minimum of

87Sr/86Sr ratio in Capitanian (Permian) mid-oceanic carbonates: A critical turning

point in the Late Paleozoic: Journal of Asian Earth Sciences, v. 32, p. 22-33.

Kearsey, T., Twitchett, R.J., Price, G.D., and Grimes, S.T., 2009, Isotope excursions and

palaeotemperature estimates from the Permian/Triassic boundary in the Southern

Alps (Italy): Palaeogeography, Palaeoclimatology, Palaeoecology, v. 279, p. 29-40.

Kidder, D.L., and Worsley, T.R., 2004, Causes and consequences of extreme Permo-

Triassic warming to globally equable climate and relation to the Permo-Triassic

extinction and recovery: Palaeogeography, Palaeoclimatology, Palaeoecology, v.

203, p. 207-237.

Korte, C., Jasper, T., Kozur, H.W., and Veizer, J., 2006, 87Sr/86Sr record of Permian

seawater: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 240, p. 89-107.

Korte, C., Kozur, H.W., Bruckshen, P., and Veizer, J., 2003, Strontium isotope evolution

of Late Permian and Triassic seawater: Geochimica et Cosmochimica Acta, v. 67, p.

47-62.

Korte, C., Kozur, H. W., Joachimski, M.M., Strauss, H., Veizer, J., and Schwark, L.,

2004, Carbon, sulfur, oxygen, and strontium isotope records, organic geochemistry,

and biostratigraphy across the Permian/Triassic boundary in Abadeh, Iran:

International Journal of Earth Sciences, v. 93, p. 565-581.

Lehrmann, D.J., Payne, J.L., Felix, S.V., Dillett, P.M., Wang, H.M., Yu, Y.Y., Xiao, J.F.,

and Orchard, M.J., 2003, Permian-Triassic boundary sections from shallow marine

carbonate platforms of the Nanpanjiang Basin, South China: implications for oceanic

conditions associated with the end-Permian extinction and its aftermath: Palaios, v.

18, p. 803-815.

Martin, E.E., and Macdougall, J.D., 1995, Sr and Nd isotopes at the Permian-Triassic

boundary: A record of climate change: Chemical Geology, v. 125, p. 73-99.

McArthur, J.M., Howarth, R.J., and Shields, G.A., 2012, Strontium Isotope Stratigraphy,

in Gradstein, F.M., Ogg, J.G., Schmitz, M., and Ogg, G., eds., The Geologic Time

Scale 2012: Boston, MA, Elsevier, p. 127-144.

Montañez, 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 87Sr/86Sr:

Geology, v. 24, p. 917-920.

Palmer, M.R., and Edmond, J.M., 1989, The strontium isotope budget of the modern

ocean: Earth and Planetary Science Letters, v. 92, p. 11-26.

Payne, J. L., Lehrmann, D. J., Follett, D., Seibel, M., Kump, L. R., Riccardi, A., Altiner,

D., Sano, H. and Wei, J., 2007, Erosional truncation of uppermost Permian shallow-

marine carbonates and implications for Permian-Triassic boundary events: GSA

Bulletin, v. 119, p. 771-784.

Payne, J.L., Lehrmann, D.J., Wei, J., and Knoll, A., 2006, The pattern and timing of

biotic recovery from the end-Permian extinction on the Great Bank of Guizhou,

Guizhou Province, China: Palaios, v. 21, p. 63-85.

Retallack, G.J., 1995, Permian-Triassic extinction on land: Science, v. 267, p. 77-80.

Richoz, S., Krystyn, L., Baud, A., Brandner, R., Horacek, M., and Parvin, M-A., 2010,

Permian-Triassic boundary interval in the Middle East (Iran and N. Oman):

Progressive environmental change from detailed carbonate carbon isotope marine

curve and sedimentary evolution, Journal of Asian Earth Sciences, v. 39, p. 236-253.

Saltzman, M.R., and Sedlacek, A.R.C, 2013, Chemostratigraphy indicates a relatively

complete Late Permian to Early Triassic sequence in the western United States:

Geology, v. 41, p. 399-402.

Sedlacek, A.R.C., Saltzman, M.R., Algeo, T.J., Horacek, M., Brandner, R., Foland, K.,

and Denniston, R.F. (in preparation), Coupled carbon and strontium isotope

stratigraphy from the late Permian to Early Triassic of Zal, Iran: A record of

increased weathering.

Sun, Y.D., Joachimski, M.M., Wignall, P.B., Yan, C.B., Chen, Y.L., Jiang, H.S., Wang,

L.N., and Lai, X.L., 2012, Lethally hot temperatures during the Early Triassic

greenhouse: Science, v. 338, p. 366-370.

Tong, J.N., and Yin, H.F., 2002, The lower Triassic of South China: Journal of Asian

Earth Sciences, v. 20, p. 138-152.

Twitchett, R.J., 2007, Climate change across the Permian/Triassic boundary, in Williams,

M. et al., eds., Deep-time perspectives on climate change: Marrying the signal from

computer models and biological proxies: London, Geological Society of London,

The Micropalaeontological Society Special Publications, p. 191-200.

Veizer, J., and Compston W., 1974, 87Sr/86Sr composition of seawater during the

Phanerozoic: Geochimica et Cosmochimica Acta, v. 38, p 1461-1484.

Veizer, J., Ala, D., Azmy, K., Bruckshen, P., Buhl, D., Bruhn, F., Carden, G.A.F.,

Diener, A., Ebneth, S., Goddéris, Y., Jasper, T., Korte, C., Pawellek, F., Podlaha,

O.G., and Strauss, H., 1999, 87Sr/86Sr, δ13C and δ18O evolution of Phanerozoic

seawater: Chemical Geology, v. 161, p. 59-88.

Veizer, J., 1989, Strontium isotopes in seawater through time: Annual Review of Earth

and Planetary Sciences, v. 17, p. 141-167.

Chapter 3

Algeo, T.J., and Twitchett, R.J., 2010, Anomalous Early Triassic sediment fluxes due to

elevated weathering rates and their biological consequences: Geology, v. 38, p.

1023-1026.

Alvarez, W., and O’Connor, D., 2002, Permian-Triassic boundary in the southwestern

United States: Hiatus or continuity?, in Koeberl, C., and MacLeod, K.G. eds.,

Catastrophic events and mass extinctions: Impacts and beyond: Geological Society

of America Special Paper 356, p. 385-393

Bailey, T.R., McArthur, J.M., Prince, H., Thurwall, M.F., 2000, Dissolution methods for

strontium isotope stratigraphy: whole rock analysis: Chemical Geology, v. 167, p.

313-319.

Baud, A., Margaritz, M., and Holser, W.T., 1989, Permian-Triassic of the Tethys—

Carbon isotope studies: Geologische Rundschau, v. 78, p. 649-677.

Baud, A., Cirilli, S., and Marcoux, J., 1997, Biotic response to mass extinction: The

lowermost Triassic microbialites: Facies, v. 36, p. 238-242.

Baud, A., Richoz, S., and Marcoux, J., 2005, Calcimicrobial cap rocks from the basal

Triassic units: Western Taurus occurrences (SW Turkey): Comptes Rendus Palevol,

v. 4, p. 569-582.

Baud, A., Richoz, S., and Pruss, S., 2007, The Lower Triassic anachronistic facies in

space and time: Global and Planetary Change, v. 55, p. 81-89.

Bissell, H.J., 1973, Permian-Triassic boundary in the eastern Great Basin area: Canadian

Society of Petroleum Geologists Memoir, v. 2, p. 3118-344.

Brand, U., and Veizer, J., 1980, Chemical diagenesis of a multicomponent carbonate

system-1: trace elements: Chemical Geology, v. 50, p. 1219-1236.

Carr, T., 1981, Paleogeography, depositional history, and conodont paleoecology of the

Lower Triassic Thaynes Formation in the Cordilleran Miogeosyncline [unpublished

Ph.D. thesis]: University of Wisconsin-Madison.

Collin, P-Y., Kershaw, S., Crasquin-Soleau, and Feng, Q., 2009, Facies changes and

diagenetic processes across the Permian-Triassic boundary event horizon, Great

Bank of Guizhou, South China: a controversy of erosion and dissolution:

Sedimentology, v. 56, p 677-693.

Collinson, J.W., Kendall, C.G.St.C., and Marcantel, J.B., 1967, Permian-Triassic

boundary in eastern Nevada and west-central Utah: Geological Society of America

Bulletin, v. 87, p. 821-824.

Erwin, D.H., Bowring, S.A., Jin, Y., 2002, End-Permian mass extinctions; a review, in

Koeberl, C., and MacLeod, K.G. eds., Catastrophic events and mass extinctions:

Impacts and beyond: Geological Society of America Special Paper 356, p. 363-383.

Ezaki, Y., Liu, J., and Adachi, N., 2003, Earliest Triassic microbialite micro- to

megastructures in the Huaying area of Sichuan Province, south China: implications

for the nature of oceanic conditions after the end-Permian extinction: PALAIOS, v.

18, p. 388-402.

Foland, K.A. and Allen, J.C., 1991, Magma sources for Mesozoic anorogenic granites of

the White Mountain magma series, New England, USA: Contributions to

Mineralogy and Petrology, v. 109, p. 195-211.

Fischer, A.G., and Arthur, M.A., 1977, Secular variations in the pelagic realm, in Cook,

H.E., and Enos, P., eds., Deep-water carbonate environments: Society of Economic

Paleontologists and Mineralogists Special Publication 25, p. 19-51.

Groves, J.R., Rettori, R., Payne, J.L., Boyce, M.D., and Altiner, D., 2007, End-Permian

mass extinction of lagenide foraminifers in the Southern Alps (northern Italy):

Journal of Paleontology, v. 81, p. 415-434.

Henderson, C.M., 1997, Uppermost Permian conodonts and the Permian-Triassic

boundary in the Western Canada Sedimentary Basin: Bulletin of Canadian

Petroleum Geology, v. 45, p. 693-707.

Heydari, E., Hassanzadeh, J., Wade, W.J., and Ghazi, A.M., 2003, Permian-Triassic

boundary interval in the Abadeh section of Iran with implications for mass

extinction: Part 1-Sedimentology: Palaeogeography, Palaeoclimatology,

Palaeoecology, v. 193, p. 405-423.

Hips, K., and Haas, J., 2006, Calcimicrobial stromatolites at the Permian-Triassic

boundary ina western Tethyan section, Bükk Mountains, Hungary: Sedimentary

Geology, v. 185, p. 239-253.

Hose, R.K., and Repenning C.A., 1959, Stratigraphy of Pennsylvanian, Permian, and

Lower Triassic rocks of Confusion Range, west-central Utah. Bulletin of the

American Association of Petroleum Geologists, v. 43, p. 2167-2196.

Isozaki, Y., 1997, Permo-Triassic boundary superanoxia and stratified super-ocean:

Records from lost deep sea: Science, v. 276, p. 235-238.

Kershaw, S., Crasquin, S., Collin, P.Y., Li, Y., Feng, Q., and Forel, M.B., 2009,

Microbialites as disaster forms in anachronistic facies following the end-Permian

mass extinction, a discussion.

Kershaw, S., Li, Y., Crasquin-Soleau, S., Feng, Q., Mu, X., Collin, P.Y., Reynolds, A.,

and Guo, L., 2007, Earliest Triassic microbialites in the South China block and other

areas: controls on their growth and distribution: Facies, v. 53, p. 409-425.

Kershaw, S., Guo, L., Swift, A., and Fan, J.S., 2002, Microbialites in the Permian-

Triassic boundary interval in central China: Structure, age and distribution: Facies, v.

47, p. 83-89.

Kershaw, S., Zhang, T., and Lan, G., 1999, A microbialite carbonate crust at the Permian-

Triassic boundary in South China, and its palaeoenvironmental significance:

Palaeogeography, Palaeoclimatology, Palaeoecology, v. 146, p. 1-18.

Kidder, D.L., and Worsley, T.R., 2004, Causes and consequences of extreme Permo-

Triassic warming to globally equable climate and relation to the Permo-Triassic

extinction and recovery: Palaeogeography, Palaeoclimatology, Palaeoecology, v.

203, p. 207-237.

Korte, C., Jasper, T., Kozur, H.W., and Veizer, J., 2006, 87Sr/86Sr record of Permian

seawater: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 240, p. 89-107.

Korte, C., and Kozur, H.W., 2010, Carbon isotope stratigraphy across the Permian-

Triassic boundary: A review: Journal of Asian Earth Sciences, v. 39, p. 215-235.

100

Korte, C., Kozur, H.W., Bruckshen, P., and Veizer, J., 2003, Strontium isotope evolution

of Late Permian and Triassic seawater: Geochimica et Cosmochimica Acta, v. 67, p.

47-62.

Korte, C., Kozur, H. W., Joachimski, M.M., Strauss, H., Veizer, J., Schwark, L., 2004,

Carbon, sulfur, oxygen, and strontium isotope records, organic geochemistry, and

biostratigraphy across the Permian/Triassic boundary in Abadeh, Iran. International

Journal of Earth Sciences, v. 93, p. 565-581.

Lehrmann, D.J., 1999, Early Triassic calcimicrobial mounds and biostromes of the

Nanpanjiang Basin, South China: Geology, v. 27, p. 359-362.

Lehrmann, D.J., Payne, J.L., Felix, S.V., Dillett, P.M., Wang, H.M., Yu, Y.Y., Xiao, J.F.,

and Orchard, M.J., 2003, Permian-Triassic boundary sections from shallow marine

carbonate platforms of the Nanpanjiang Basin, South China: implications for oceanic

conditions associated with the end-Permian extinction and its aftermath: Palaios, v.

18, p. 803-815.

Martin, E.E., Macdougall, J.D., 1995, Sr and Nd isotopes at the Permian/Triassic

boundary: A record of climate change: Chemical Geology, v. 125, p. 73-99.

Mata, S.A., and Woods, A.D., 2008, Sedimentology and Paleoecology of the lower

member of the Lower Triassic (Smithian-Spathian) Union Wash Formation, east-

central California: PALAIOS, v. 23, p. 514-524.

Montañez, 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

101

platform carbonates: Implications for the evolution of Cambrian seawater 87Sr/86Sr:

Geology, v. 24, p. 917-920.

Newell, N., 1948, Key Permian section, Confusion Range, western Utah: Bulletin of the

Geological Society of America, v. 59, p. 1053-1058.

Nützel, A., and Schulbert, C., 2005, Facies of two important Early Triassic gastropod

lagerstätten: implications for diversity patterns in the aftermath of the end-Permian

mass extinction: Facies, v 51, p. 480-500.

Orchard, M.J., 2007, Conodont diversity and evolution through the latest Permian and

Early Triassic upheavals: Palaeogeography Palaeoclimatotology Palaeoecolology, v.

252, p. 93–117, doi:10.1016/j.palaeo.2006.11.037.

Payne, J.L., Lehrmann, D.J., Follett, D., Seibel, M., Kump, L.R., Riccardi, A., Altiner,

D., Sano, H., Wei, J., 2007, Erosional truncation of uppermost Permian shallow-

marine carbonates and implications for Permian-Triassic boundary events: GSA

Bulletin, v. 119, p. 771-784.

Payne, J.L., Lehrmann, D.J., Wei, J., and Knoll, A., 2006, The pattern and timing of

biotic recovery from the end-Permian extinction on the Great Bank of Guizhou,

Guizhou Province, China: Palaios, v. 21, p. 63-85.

Pruss, S.B., Bottjer, D.J., Corsetti, F.A., and Baud, A., 2006, A global marine

sedimentary response to the end-Permian mass extinction: Examples from southern

Turkey and the western United States: Earth-Science Reviews, v. 78, p. 193-206.

Raup, D., and Sepkoski, J., 1982, Mass extinctions in the marine fossil record: Science v.

215, p 1501-1503.

102

Retallack, G.J., 1995, Permian-Triassic extinction on land: Science, v. 267, p. 77-80.

Retallack, G.J., Veevers, J.J., and Morante, R., 1996, Global coal gap between Permian-

Triassic extinction and Middle Triassic recovery of peat-forming plants: Geological

Society of America Bulletin, v. 108, p. 195-207.

Riding, R., 2000, Microbial carbonates: the geologic record of calcified bacterial-algal

mats and biofilms: Sedimentology, v. 47, p. 179-214.

Saltzman, M.R., and Sedlacek, A.R.C, 2013, Chemostratigraphy indicates a relatively

complete Late Permian to Early Triassic sequence in the western United States:

Geology, v. 41, p. 399-402.

Sano, H., and Nakashima, K., 1997, Lowermost Triassic (Griesbachian) microbial

bindstone-cementstone facies, southwest Japan: Facies, v. 36, p. 1-24.

Schubert, J.K., and Bottjer, D.J., 1992, Early Triassic stromatolites as post mass

extinction disaster forms: Geology, v. 20, p. 883-886.

Schubert, J.K., and Bottjer, D.J., 1995, Aftermath of the Permian-Triassic extinction

event: Paleoecology of Lower Triassic carbonates in the western USA:

Palaeogeography, Palaeoclimatology, Palaeoecology, v. 116, p. 1-39.

Sperling, E.A., and Ingle, J.C., 2006, A Permian-Triassic boundary section at Quinn

River Crossing, northwestern Nevada, and implications for the cause of the Early

Triassic chert gap on the western Pangean margin: Geological Society of America

Bulletin, v. 118, p. 733-746.

Stanley, S.M., and Hardie, L.A., 1999, Hypercalcificaion: Paleontology links plate

tectonics and geochemistry to sedimentology: GSA Today, v. 9, p. 1-7.

103

Sun, Y.D., Joachimski, M.M., Wignall, P.B., Yan, C.B., Chen, Y.L., Jiang, H.S., Wang,

L.N., and Lai, X.L., 2012, Lethally hot temperatures during the Early Triassic

greenhouse: Science, v. 338, p. 366-370.

Veizer, J., Ala, D., Azmy, K., Bruckshen, P., Buhl, D., Bruhn, F., Carden, G.A.F.,

Diener, A., Ebneth, S., Goddéris, Y., Jasper, T., Korte, C., Pawellek, F., Podlaha,

O.G., and Strauss, H., 1999, 87Sr/86Sr, δ13C and δ18O evolution of Phanerozoic

seawater: Chemical Geology, v. 161, p. 59-88.

Veizer, J., 1989, Strontium isotopes in seawater through time: Annual Review of Earth

and Planetary Sciences, v. 17, p. 141-167.

Ward, P.D., Montgomery, D.R., and Smith, R., 2000, Altered river morphology in South

Africa related to the Permian-Triassic extinction: Science, v. 289, p. 1740-1743.

Wardlaw, B.R., and Collinson, J.W., 1978, Stratigraphic relations of Park City Group

(Permian) in eastern Nevada and western Utah: American Association of Petroleum

Geologists Bulletin, v. 62, p. 1171-1184.

Wardlaw, B.R., and Collinson, J.W., 1986, Paleontology and deposition of the

Phosphoria Formation: Contributions to Geology (Copenhagen), v. 24, p. 107-142.

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, v. 18, p. 153-167.

Woods, A.D., and Baud, A., 2008, Anachronistic facies from a drowned Lower Triassic

carbonate platform: Lower member of the Alwa Formation (Ba’id Exotic), Oman

Mountains: Sedimentary Geology, v. 209, p. 1-14.

104

Appendix A: Supplemental Material for Chapter 1

Locality Map

AZERBAIJAN TURKMENISTAN Caspian Sea Zal Section N

Tehran

IRAN IRAQ

SAUDI ARABIA

0 400 km Gulf of Oman

Figure 33. Map of Iran showing the location of the study section near the village of Zal in northwestern Iran (after Kozur, 2007).

105

Paleogeographic Map

Paleotethys

Panthalassa Iranian Panthalassa microcontinent PANGEA Neotethys Cimmeria

106

Stratigraphy at Zal

The stage and substage boundaries at Zal are assigned after Horacek et al.

(2007a). At Zal, the Permian-Triassic Boundary (PTB) and Griesbachian substage are biostratigraphically well constrained: Hindeodus parvus has been reported at 0.7 m, and

Isarcicella isarcica occurs at 3.5 m and 11 m (Horacek et al., 2007a, Richoz et al., 2010).

The bivalve Claraia dominates the upper Griesbachian beds, and the base of the

Dienerian has been placed near the top of the Claraia-rich interval. This placement is consistent with the Griesbachian-Dienerian boundary at Abadeh in central Iran, a section that contains more diagnostic fossils near the base of the Dienerian. The Dienerian-

13 Smithian boundary is marked by a pronounced positive δ Ccarb excursion. Below the peak of this excursion, Horacek et al. (2007a) reported Neogondolella dieneri, which confirms the Dienerian age of these strata. The presence of Pachycladina at 570 m indicates a Smithian age. Placement of the Smithian/Spathian boundary is based on correlation with existing C isotope curves from other biostratigraphically well- constrained sections. Its position is above the last Hadrodontina conodonts, which is consistent with other Iranian Early Triassic sections (Horacek et al., 2007).

107

Figure 35. Carbon and strontium isotope data plotted against stratigraphic thickness at Zal. The Dienerian is represented by approximately 300 meters of section, allowing for relatively high resolution sampling through this substage. The two Griesbachian data points associated with a volcanic sill are considered altered and are shown with unfilled symbols.

108

Figure 35

109

Diagenesis

Because diagenetic alteration typically depletes the Sr concentration of carbonate marine sediments, samples with higher Sr concentrations are likely to represent greater initial Sr concentrations, and the 87Sr/86Sr of these samples is less susceptible to alteration

(Brand and Veizer, 1980; Denison et al., 1994). The relatively high concentrations of Sr in most samples increase our confidence that the 87Sr/86Sr ratios from Zal represent primary Early Triassic seawater values (Fig. 36).

Diagenesis generally alters 87Sr/86Sr toward higher values due to exchange with clay minerals enriched in 87Sr, although Sr exchange with volcanic sills can potentially lower 87Sr/86Sr (Veizer and Compston, 1974). In the study section, two Griesbachian-age samples associated with sills exhibit 87Sr enrichment, and we consider these samples to be altered (Fig. 1, 35).

110

0.7082

Spathian Smithian Dienerian 0.7080 Griesbachian Changhsingian

0.7078

0.7076

0.7074

0.7072

0.7070 0 500 1000 1500 2000 Sr ppm

Figure 36. Crossplot of 87Sr/86Sr and Sr concentration (ppm) shows no clear covariation in Changhsingian, Griesbachian, Dienerian and Smithian age samples, indicating that the isotopic values of these rocks may be relatively unaltered by diagenesis. Covariation present in Spathian samples that are from dolomitized lithologies and these values are likely altered.

111

Early Triassic Timescale

The timescale used in Figures 1 and 2 differs from the ages in the newly released

Geologic Time Scale 2012 (Gradstein et al., 2012). We elected to follow the timescale presented by Algeo et al. (2012), which is in better agreement with radiometric age dates for the base of the Smithian and Spathian substages published by Galfetti et al. (2007) and Ovtcharova et al. (2006), respectively. When our data are plotted using the Geologic

Time Scale 2012 ages, the rate of 87Sr/86Sr change (∂Sr/∂t) during the Dienerian is reduced due to an increase in the duration of this substage, but the rate nonetheless remains higher than for any other Early Triassic substage (Figs. 37-38; Table 2). Using the Geologic Time Scale 2012, the rate of 87Sr/86Sr rise during the Dienerian increased by approximately 0.0004 m.y.-1 relative to that of the Griesbachian (Fig. 2). Although

87Sr/86Sr continued to increase during the Olenekian, the rate of rise diminished through the end of the Spathian as sedimentation rates decreased (Fig. 37).

112

248

249

250

251

252

253

254 Age -4 -2 0 2 4 6 0.7070 0.7074 0.7078 0.7082 (Ma) 13C -VPDB 87Sr/86Sr (Horacek et al., 2007a) (This Study)

Figure 37. Carbon and strontium isotope data plotted against dates from the Geologic Time Scale 2012 (Gradstein et al., 2012). Here, the duration of the Dienerian is longer by 0.58 m.y. relative to its duration in the Algeo et al. (2012) timescale (0.27 m.y.), lowering its ∂Sr/∂t. However, even after accounting for differences in the Early Triassic timescale, ∂Sr/∂t remains higher during the Dienerian than for other Early Triassic substages (Tables 2, 4).

113

94.2 m m.y.-1 560 1.8

480 65.8 m m.y.-1

1.8 400

352.9 m m.y.-1 240 2.6

160

80 70.4 m m.y.-1 1.2 10.0 m m.y.-1 0 1.3 Chang. (sub)stage Chang. Gr. D. Sm. Spath. Age (Ma) 254 252 250 248

Figure 38. Linear sedimentation rates (LSR) calculated for the Late Permian-Early Triassic based on the Geologic Time Scale 2012 (Gradstein et al., 2012) rather than on the Algeo et al. (2012) timescale, as in Figure 2. LSRs were calculated for Zal following the methods of Algeo and Twitchett (2010) by dividing thickness by time (m.y.) for each stage or substage. LSRs are given for each (sub)stage to the left of the curve, and corresponding ∂Sr/∂t (in units of 10-4 m.y.-1) are shown in boxes. Maximum sedimentation rates occurred during the Dienerian, although Dienerian rates are reduced relative to Figure 1.2 owing to the longer duration of this substage in the GTS 2012 (0.85 m.y.). Chang. = Changhsingian, Griesbach., Gr. = Griesbachian, D. = Dienerian, Sm. = Smithian, Spath. = Spathian. LSR and ∂Sr/∂t are given in DR Table S2.

114

Data Tables

TABLE 1. 87Sr/86Sr ZAL, IRAN Sample Height Sr 87Sr/86Sr uncertainty (Sub)stage name (m) (ppm) IZ153a 728 31.91 0.708100 0.000012 Spathian IZ145 703.2 22.61 0.708128 0.000012 Spathian IZ132 645.2 154.12 0.708010 0.000010 Spathian IZ131 639.2 145.92 0.707982 0.000011 Spathian IZ122 600 263.85 0.707878 0.000010 Spathian IZ120 594 303.42 0.707871 0.000090 Smithian IZ113 565.7 486.19 0.707862 0.000012 Smithian IZ111 554.6 405.03 0.707724 0.000010 Smithian IZ108 544.35 517.68 0.707707 0.000009 Smithian IZ 104 525.7 563.64 0.707681 0.000010 Smithian IZ100 513.6 415.02 0.707737 0.000013 Smithian IZ99 512.5 357.66 0.707594 0.000011 Smithian IZ98 504.3 356.35 0.707662 0.000014 Smithian IZ96 491.5 314.81 0.707590 0.000006 Dienerian IZ95 487.5 287.16 0.707585 0.000007 Dienerian IS92 474.5 1222.40 0.707548 0.000007 Dienerian IS90 467.5 20.84 0.707605 0.000013 Dienerian IS86 462.5 559.23 0.707517 0.000009 Dienerian IS84 452.5 307.26 0.707524 0.000010 Dienerian IS83 447.5 278.13 0.707490 0.000006 Dienerian IS82 443.5 468.83 0.707514 0.000010 Dienerian IS76 406.5 738.54 0.707473 0.000007 Dienerian IS72 389.8 956.42 0.707503 0.000010 Dienerian IS69 374.3 578.81 0.707489 0.000008 Dienerian IS 67 361.6 1250.53 0.707468 0.000010 Dienerian IS 65 348 1189.30 0.707484 0.000008 Dienerian IS 64 346 637.78 0.707501 0.000009 Dienerian IS63 345.5 573.49 0.707513 0.000021 Dienerian IS62 343 1322.08 0.707498 0.000012 Dienerian IS61 335.5 831.36 0.707501 0.000009 Dienerian Continued Table 1. 87Sr/86Sr results from Zal, Iran

115

Table 1. continued

IZ60 333 636.85 0.707549 0.000013 Dienerian IZ 47 332 1071.98 0.707459 0.000009 Dienerian IZ44 315.2 560.58 0.707539 0.000007 Dienerian IZ43 314.4 1061.79 0.707467 0.000007 Dienerian IZ39 262.15 831.08 0.707427 0.000009 Dienerian IZ35 242.6 1905.06 0.707397 0.000008 Dienerian IZ34 237.6 592.75 0.707451 0.000009 Dienerian IZ32 227.6 947.20 0.707479 0.000013 Dienerian IZ29 178 730.00 0.707447 0.000014 Dienerian IZ28 170 1046.99 0.707396 0.000008 Dienerian IZ27 165 1191.02 0.707386 0.000007 Dienerian IZ26 160 1244.84 0.707358 0.000009 Dienerian IZ25 155 1387.48 0.707366 0.000016 Dienerian IZ24 145 1176.69 0.707401 0.000009 Dienerian IZ21 125 181.53 0.707413 0.000017 Dienerian IZ20 120 262.85 0.707594 0.000014 Griesbachian IZ18 110 585.17 0.707265 0.000008 Griesbachian IZ14 45 228.63 0.707645 0.000024 Griesbachian IZ11 30 562.11 0.707200 0.000011 Griesbachian IZ10 25 317.15 0.707153 0.000009 Griesbachian IZ8 15 418.00 0.707217 0.000010 Griesbachian IZ6 5 312.07 0.707220 0.000009 Griesbachian IZ4 -5 230.63 0.707212 0.000014 Changhsingian IZ3 -10 309.88 0.707166 0.000007 Changhsingian IZ2 -15 224.80 0.707099 0.000009 Changhsingian IZ1 -20 238.57 0.707072 0.000007 Changhsingian

116

TABLE 2. Rate of Change in Seawater 87Sr/86Sr (Algeo et al., 2012, timescale) Min (Sr) Uncertainty (age) ∂Sr/∂t Max ∂Sr/∂t (Sub)stage ∂Sr/∂t (×10-4) ± (kyr) (Myr-1) (Myr-1) (Myr-1) Spathian 2.5 0.6 2500 1.000 0.760 1.240 Smithian 2.7 0.6 670 4.030 3.134 4.925 Dienerian 2.2 0.4 270 8.148 6.667 9.630 Griesbachian 1.6 0.8 560 2.857 1.429 4.286 Changhsingian 2.5 0.2 1800 1.389 1.278 1.500

Table 2. Differences in rates of 87Sr/86Sr rise (∂Sr/∂t) when calculated using ages of Algeo et al. (2012). ∂Sr/∂t was calculated by fitting a line through the Sr isotopic data for each (sub)stage so as to minimize the sum of squared deviations.

117

TABLE 3. Linear Sedimentation Rates (Algeo et al., 2012, timescale)

Thickness Uncertainty (age) Rate Min rate Max Rate (Sub)stage (m) ± m (kyr) (Myr-1) (Myr-1) (Myr-1)

Spathian 130 10 2500 52.0 48.0 56.0 Smithian 100 10 670 149.3 134.3 164.2 Dienerian 300 20 270 1111.1 1037.0 1185.2 Griesbachian 95 20 560 169.6 133.9 205.4 Changhsingian 20 5 1800 11.1 8.3 13.9

Table 3. Differences in linear sedimentation rates (LSRs) when calculated using ages of Algeo et al. (2012). LSRs (in units of meters per million years) were assumed to be constant for each (sub)stage and were calculated by dividing the thickness of a time- stratigraphic unit by the duration of that (sub)stage.

118

TABLE 4. Rate of Change in Seawater 87Sr/86Sr (Geologic Time Scale 2012)

(Sr) Uncertainty (age) ∂Sr/∂t Min ∂Sr/∂t Max ∂Sr/∂t (Sub)stage (×10-4) ± (kyr) (Myr-1) (Myr-1) (Myr-1)

Spathian 2.5 0.6 1380 1.812 1.377 2.246 Smithian 2.7 0.6 1520 1.776 1.382 2.171 Dienerian 2.2 0.4 850 2.588 2.118 3.059 Griesbachian 1.6 0.8 1350 1.185 0.593 1.778 Changhsingian 2.5 0.2 2000 1.250 1.150 1.350

Table 4. Differences in rates of 87 Sr/86Sr rise (∂Sr/∂t) when calculated using ages of Gradstien et al. (2012). ∂Sr/∂t was calculated by fitting a line through the Sr isotopic data for each (sub)stage so as to minimize the sum of squared deviations.

119

TABLE 5. Linear Sedimentation Rates (Geologic Timescale 2012) Max Thickness Uncertainty (age) Rate Min rate (Sub)stage Rate (m) ± m (kyr) (Myr-1) (Myr-1) (Myr-1) Spathian 130 10 1380 94.2 87.0 101.4 Smithian 100 10 1520 65.8 59.2 72.4 Dienerian 300 20 850 352.9 329.4 376.5 Griesbachian 95 20 1350 70.4 55.6 85.2 Changhsingian 20 5 2000 10.0 7.5 12.5

Table 5. Differences in linear sedimentation rates (LSRs) when calculate d using ages of Gradstein et al. (2012). LSRs (in units of meters per million years) were assumed to be constant for each (sub)stage and were calculated by dividing the thickness of a time- stratigraphic unit by the duration of that (sub)stage.

120

Appendix B: Supplemental Material for Chapter 2

121

TABLE 6. 87Sr/86Sr RESULTS DAWEN, GUIZHOU, CHINA Height Sample Sr 87Sr/86Sr uncertainty (cm relative to δ13C name dissolution (ppm) surface) 1000 1000 434.9611 0.707193 0.000007 -0.16538 940 940 157.8553 0.707343 0.000025 0.71042 900 900 279.3352 0.707217 0.000011 0.21422 880 880 -0.82558 840-845 842.5 -0.33598 815-830 822.5 -0.24318 800-805 802.5 247.7405 0.707213 0.000016 -0.2056 760 760 202.9538 0.707259 0.000016 -0.00478 725-730 727.5 -0.25678 700-705 702.5 -0.7636 670-675 672.5 -1.07498 635 635 -0.39073 610 610 216.4038 0.707268 0.000018 -0.28538 585 585 -0.09098 555-565 560 -0.3893 520-525 522.5 319.4325 0.707181 0.00001 -0.19578 500 500 -1.22898 478-486 482 -0.80318 450-455 452.5 -0.27758 435 435 -0.34069 420 420 209.3788 0.707231 0.00001 0.28562 418-425 421.5 395 395 - 0.89598 390-400 395 -0.27638 368-372 370 -0.12218 360-365 362.5 197.2963 0.707208 0.000019 -0.05138 340-345 342.5 -0.61758 305-315 310 -0.09538 290-295 292.5 195.9648 0.707249 0.000009 -0.04298 248-253 250.5 0.22922 223-228 225.5 0.01242 Continued Table 6. 87Sr/86Sr results from Dawen, Guizhou, China

122

Table 6 continued

197 -203 200 - 0.41878 188-195 191.5 0.05842 128-135 131.5 1.01662 121-128 124.5 0.29762 119-124 121.5 0.24442 88-96 92 186.6003 0.707238 0.000012 0.27122 79-90 84.5 0.44142 72-80 76 0.6564 70-75 72.5 0.93762 47-55 51 -0.36578 40-45 42.5 1.152 30-36 33 185.1433 0.707267 0.000011 1.38001 20-26 23 176.1004 0.707258 0.000009 0.87642 13-17 15 1.67122 8-13 10.5 1.81522 7.5-10 8.2 0.73482 5-7.5 6.2 161.8007 0.707384 0.00001 0.78662 0-5 2.5 172.3416 0.707358 0.000013 0.05142 -2 to -6 -4 189.0523 0.707283 0.000012 0.56062 -10 to -13 -11.5 0.8538 -13 to -15 -14 0.818 -15 to -20 (1) -17.5 0.12772 -20 to -25 -22.5 191.1738 0.707407 0.000014 1.73561 -40 -40 255.4544 0.707292 0.000015 1.72421 -63 to -66 -64.5 0.51502 -75 to -82 -78.5 1.5436 -95 to - 100 -97.5 1.921 -115 to - 120 -117.5 1.81772 -135 to - 140 -137.5 1.72622 -160 -160 448.7349 0.707136 0.000011 2.36102 -185 -185 3.33362 -200 -200 2.51662 -220 -220 2.25542

Continued

123

Table 6 continued -240 to - 248 -244 2.87302 -265 to - 275 -270 424.2474 0.707103 0.000010 2.79522 -300 -300 1.96262 -315 -315 2.11682 -330 -330 1.14962 -345 -345 485.4233 0.707136 0.000011 1.50062 -370 -370 1.27662 -430 to - 435 (+43) -389.5 289.7215 0.70717 0.000011 2.57082 -450 to - 455 (+43) -409.5 2.94562 -480 to - 486 (+43) -440 1.52862 -500 to - 510 (+43) -462 1.94972

124

Table 7. Brachiopod LMC from Brand et al., 2012 Brachiopod LMC from Italian Alps (Brand et al., 2012 Table 7. Brachiopod LMC from Sass de Putia and Val Brutta (Brand et al., 2012) Section Sample Formation/ B-W Material 87Sr/86Sr Biozone Number Member cm Sass de 11A-59m Werfen/Tesero 3 Unaltered 0.707092 C. meishan. – H. eurypyge Putia Sass de 11A-56m Werfen/Tesero 3 Altered 0.707147 C. meishan. – H. eurypyge Puita Sass de 10 top-51 Bellerophon/Bulla -3 Altered 0.707164 C. chanxingensis- deflecta Puita

2 Sass de

5 10.67-43 Bellerophon/Bulla -20 Altered 0.707110 C. chanxingensis- deflecta

Puita Sass de 10.116- Bellerophon/Bulla -20 Unaltered 0.707067 C. chanxingensis- deflecta Puita 38m Sass de 6.X-33 Bellerophon/Bulla -92 Unaltered 0.707060 C. chanxingensis- deflecta Puita Sass de 6.60-23 Bellerophon/Bulla -103 Altered 0.707223 C. chanxingensis- deflecta Puita Sass de 6.60-21 Bellerophon/Bulla -103 Unaltered 0.707078 C. chanxingensis- deflecta Puita Sass de 5.55-10 Bellerophon/Bulla -104 Unaltered 0.707062 C. chanxingensis- deflecta Puita Sass de 3.51-4 Bellerophon -132 Unaltered 0.707075 Puita Sass de 3.51-2 Bellerophon -132 Altered 0.707201 Puita Continued 125

Table 7 Continued Val Brutta VB 10-119 Werfen/Tesero 7 Unaltered 0.707108 C. meishan. – H. eurypyge Val Brutta VB 9C-113 Bellerophon/Bulla -1 Altered 0.707103 C. chanxingensis- deflecta Val Brutta VB 9C-105 Bellerophon/Bulla -1 Unaltered 0.707100 C. chanxingensis- deflecta Val Brutta VB 9B-103 Bellerophon/Bulla -5 Unaltered 0.707088 C. chanxingensis- deflecta Val Brutta VB 9B-99 Bellerophon/Bulla -7 Unaltered 0.707089 C. chanxingensis- deflecta Val Brutta VB 9A-85 Bellerophon/Bulla -12 Unaltered 0.707077 C. chanxingensis- deflecta Val Brutta VB 9A-84 Bellerophon/Bulla -12 Unaltered 0.707051 C. chanxingensis- deflecta Val Brutta VB 8B-78 Bellerophon/Bulla -19 Unaltered 0.707065 C. chanxingensis- deflecta Val Brutta VB 8B-75a Bellerophon/Bulla -19 Altered 0.707328 C. chanxingensis- deflecta Val Brutta VB 8A-73 Bellerophon/Bulla -27 Unaltered 0.707099 C. chanxingensis- deflecta

6 Val Brutta VB 8A- 67 Bellerophon/Bulla -27 Unaltered 0.707069 C. chanxingensis- deflecta

126

Appendix C: Supplemental Material for Chapter 3

127

TABLE 8. 87Sr/86Sr Confusion Range, Great Basin, USA Sample Height Sr 87Sr/86Sr δ13C δ18O name (m) (ppm) 661 0.00 3.29 -8.01 664 9.00 122.06 0.707285 665 12.00 120.16 0.707031 2.93 -8.38 666 15.00 155.98 0.707007 3.47 -8.78 667 18.00 151.89 0.707050 3.10 -7.25 668 21.00 134.87 0.707067 2.79 -7.64 669 24.00 188.64 0.706994 2.67 -7.72 671 30.00 154.70 0.707010 2.85 -7.71 672 33.00 673 36.00 0.707047 2.65 -7.59 676 45.00 2.34 -6.84 678 51.00 2.90 -8.74 683 66.00 180.52 0.707034 2.79 -7.23 684 69.00 127.08 0.707192 2.97 -8.04 685 72.00 2.96 -7.58 688 81.00 235.93 0.707044 3.19 -7.32 692 93.00 186.00 0.707048 2.94 -7.88 693 96.00 2.70 694 99.00 238.53 0.706997 2.99 -6.85 699 114.00 94.69 0.707191 2.52 -8.20 706 135.00 182.17 0.707047 2.52 -6.09 707 138.00 2.71 -6.20 714 159.00 211.68 0.707144 2.40 -5.28 717 168.00 77.54 0.707103 718 171.00 171.08 0.707178 2.50 -8.30 720 177.00 154.53 0.707064 2.80 -5.50 6225 183.00 215.54 0.707116 2.32 -6.80 6226 184.00 1.60 6227 185.00 2.25 -6.89 Continued Table 8. Results of 87Sr/86Sr from the Confusion Range, Utah, USA

128

Table 8 continued

6228 186.00 6229 187.00 182.60 0.707141 2.14 -7.64 6230 188.00 137.3 0.707271 1.60 -8.440 6231 189.00 6232 190.00 124.81 0.707144 2.14 -7.96 6233 191.00 2.31 -7.20 6234 192.00 6235 193.00 6236 194.00 186.01 0.707 413 2.20 -8.317 6237 195.00 6238 196.00 1.72 -8.08 6239 197.00 2.17 -7.97 6240 198.00 158.46 0.707255 1.95 -7.922 6241 199.00 103.73 0.707573 1.81 -8.48 6242 200.00 133.43 0.707329 1.33 -8.30 6243 201.00 1.34 -8.77 6244 202.00 1.09 -9.95 6245 203.00 1.11 -8.60 6246 204.00 0.98 -9.05 6247 205.00 100.77 0.707514 1.14 -8.95 6248 206.00 1.749 -8.187 6249 207.00 1.52 -8.58 6250 208.00 93.36 0.707356 1.58 -7.95 6251 0.97 -8.87 6252 6253 1.36 -9.19 6254 128.47 0.707343 1.91 -7.78 6255 213.00 181.26 0.707314 1.78 -8.618 6256 6257 0.61 -10.03 6258 216.50 183.64 0.707237 1.54 -8.56 6259 217.00 179.87 0.707313 1.91 -8.27 6260 217.50 0.72 -8.00 Continued

129

Table 8 continued 937071 218.00 137.70 0.707314 1.55 -7.68 7072 218.25 108.74 0.707368 1.39 -7.34 7073 218.50 97.2 0.707480 1.43 7074 218.75 98.32 0.707300 1.26 -8.07 7075 219.10 52.96 0.707352 1.70 -7.149 7076 219.35 122.1 0.707356 1.43 -7.902 7077 219.60 143.3 0.707368 1.25 -7.729 7078 220.10 117.3 0.707349 1.13 -7.454 7079 220.35 117.1 0.707415 0.96 -7.301 7080 220.60 74.77 0.707789 0.22 -9.273 7081 221.35 0.22 -7.93 7082 221.60 130.8 0.707426 0.50 -7.550 7083 221.85 0.04 -10.069 7084 222.10 110.49 0.707456 0.38 -7.68 7085 222.35 99.03 0.707427 0.39 -7.989 7086 222.60 0.49 -7.71 7087 222.85 89.96 0.707509 0.15 -7.682 7088 223.10 -0.20 -8.10 7089 223.35 62.29 0.707644 -0.32 -7.95 7090 223.85 110.83 0.707614 7091 224.10 68.79 0.707942 -0.70 -6.411 7092 224.35 75.65 0.707964 -0.83 -6.471 7093 224.60 75.98 0.707917 -0.89 -6.283 7099 224.73 113.47 0.707979 -1.14 -7.76 7094 225.50 131.16 0.707896 -1.40 -8.54 7095 225.75 -1.51 -9.028 7096 226.00 133.03 0.707724 -2.07 -8.966 7097 226.25 73.13 0.707735 -2.06 -8.61 7098 226.50 107.46 0.707923 -2.12 -8.62 7100 226.75 74.93 0.707824 -2.19 -8.88 7101 228.25 83.9 0.707966 -2.12 -7.919 7102 228.50 -2.07 -8.04 7103 228.75 114.46 0.707780 -2.21 -8.04 Continued

130

Table 8 continued

7104 229.00 120.40 0.707816 7105 229.25 96.19 0.707842 -2.03 -8. 80 7106 229.50 117.01 0.707818 -2.27 -8.22 7107 230.00 191.70 0.707777 -2.07 -8.39 7108 230.25 94.51 0.707662 -2.21 -8.61 7109 230.50 110.17 0.707715 -1.98 -9.49 7110 230.75 -2.02 -8.93 7111 231.00 -1.81 -9.70 7112 231.25 117.05 0.707717 -1.83 -9.20 7113 231.50 -2.05 -9.21 7114 231.75 88.4 0.707681 -1.84 -9.309 7115 232.00 -2.12 -8.49 7116 232.25 -2.26 -8.21 7117 232.50 -2.23 -8.25 7118 232.75 69.7 0.707739 -2.30 -9.160 7119 233.00 172.23 0.707695 -2.47 -8.74 6261 235.00 142.62 0.708075 -3.33 -9.481

131