The Role of the Siberian Traps in the Permian-Triassic Boundary Crisis: Analysis Through

Home , Claraia, Cordaites, Ninety East Ridge

The Role of the Siberian Traps in the Permian-Triassic Boundary Crisis: Analysis through

Chemical Fingerprinting of Marine Sediments using Rare Earth Elements

A thesis submitted to the

Graduate School

of the University of Cincinnati

in partial fulfillment of the

requirement for the degree of

Master of Science

Alan Santistevan

B.A. University of California, Berkeley

December 2018

Committee Chair: T.A. Algeo, Ph.D.

Department of Geology, University of Cincinnati, Cincinnati OH 45221-0013 USA

Abstract

In this study, the question of microenvironmental change related to the overall eruption pattern from the Siberian Traps was examined. Using the unique rare earth element geochemistry that was produced from the successive eruptions, a geological trace that links the

Siberian Traps to the environmental collapse from the effects of windblown ash was found in using over 400 marine sediment samples taken from 7 different Permian-Triassic Boundary sections. The following results show values likely indicative of material derived from the lower crust, upper mantle with the Eu/Eu* anomalies from Zal, a microcontinent in the Tethys Ocean,

Gujo-Hachiman, an open ocean site in the eastern Panthalassic Ocean, along with Guryul Ravine and Spiti on the northwestern margin of Gondwana. These results represent a global signal, and could reflect a deleterious effect upon marine and terrestrial ecosystems from the possible volcanic ashfall that was produced from the Siberian Traps.

Table of Contents

1. Introduction

2. Background

2.1. The Permian paleoenvironment

2.2. The Permian-Triassic Boundary crisis

2.3. Potential causes of the Permian-Triassic Boundary crisis

2.4. Rare earth element geochemistry

3. Methods

4. Study Sections

4.1. Gujo Hachiman, Japan

4.2. Ubara, Japan

4.3. Black Ridge West, Greenland

4.4. Spiti, India

4.5. Guryul Ravine, India

4.6. Zal, Iran

4.7. Chaohu, China

5. Results

5.1. Gujo Hachiman, Japan

5.1.1. Al, TOC, and TIC values

5.1.2. REE ratios

5.1.3. Ce/Ce* and Eu/Eu* anomalies

5.2. Ubara, Japan

5.2.1. Al, TOC, and TIC values

5.2.2. REE ratios

5.2.3. Ce/Ce* and Eu/Eu* anomalies

5.3. Black Ridge West, Greenland

5.3.1. Al, TOC, and TIC values

5.3.2. REE ratios

5.3.3. Ce/Ce* and Eu/Eu* anomalies

5.4. Spiti, India

5.4.1. Al, TOC, and TIC values

5.4.2. REE ratios

5.4.3. Ce/Ce* and Eu/Eu* anomalies

5.5. Guryul Ravine, India

5.5.1. REE ratios

5.5.2. Ce/Ce* and Eu/Eu* anomalies

5.6. Zal, Iran

5.6.1. TOC, and TIC values

5.6.2. REE ratios

5.6.3. Ce/Ce* and Eu/Eu* anomalies

5.7. Chaohu, China

5.7.1. Al, TOC, and TIC values

5.7.2. REE ratios

5.7.3. Ce/Ce* and Eu/Eu* anomalies

6. Discussion

6.1. Significance of Eu/Eu* anomalies for understanding sediment provenance

6.2. Global Patterns of REE variation at the Permian-Triassic Boundary

6.3. Significance of wind blown material emanating from the Siberian Traps

7. Conclusions

References

Appendices

1. INTRODUCTION

The Permian-Triassic boundary (PTB) 251.9 million years ago (Burgess et al. 2014) marked the most severe biological cataclysm in the history of life, with up to 95% of all marine species, and 70% of all terrestrial species going extinct (Benton 2005; Sabney and Benton 2008).

It was also the only event in Earth’s history when both plants and insects suffered heavy extinction (Labandiera and Sepkowski 1993; Looy et al. 1999). It further took life a full 10 million years through a dynamic multi-step recovery phase to even begin returning the Earth to its pre-existing conditions from the bleak environment that was produced from the near annihilation (Chen and Benton 2012). Dating of the Siberian Traps, the largest aerial flood basalt eruptions of the last 500 million years, has shown that they were effectively coeval with the PTB mass extinction and the respective biozonation (Renne et al. 1995; Kamo et al. 2003, see Figure 1) and thus a potential cause of that event.

Climatic effects related to volcanic eruptions have long been documented in relation to environmental disturbances (Rampino et al. 1985) but none to the extent as the Siberian Traps and the PTB Mass Extinction. Contemporary models of volcanic induced climate perturbations are calculated by the total mass of ash output and the height the eruption column rises in the atmosphere in relation to the volcanoes’ latitude, and by how efficiently atmospheric air currents disperse the aerosols worldwide (Courtillot 2002). The rate at which gases are emitted is also a dependent factor and it has been inferred that the gas output from the repeated eruptions (Black et al. 2011, Black et al. 2015) would have been enough to lead to the complete collapse of all normal interactions between the physical world and life (Bond and Wignall 2014).

The connections between the flood basalt gas eruptions and the extinction lies within the chemistry of the Cambrian and Ordovician crustal rocks that were laid down before. These

Cambrian and Ordovician units contained what would become the greenhouse gases after the magmas extruded through and melted them, thereby mobilizing the gases to release into the atmosphere (Black et al. 2013). Ordovician age gypsum sediments that were melted through produced sulfur while Cambrian age coal deposits produced the carbon dioxide. Therefore, the consecutive link is the magma, the crustal rocks, and the changes within the atmospheric chemistry, which caused a global environmental change bringing about the extinction (Elkins-

Tanton 2010, Black et al. 2015).

Volcanic units are commonly characterized by unique rare earth element (REE) signatures (Wyman 1996) and published studies have shown that the Siberian Traps had an unusual REE chemistry (Lightfoot et al. 1990, 1993; Arndt et al. 1993, 1995, 1998; Federenko et al. 1997, 2000). With the unique signature that is produced from the REE chemistry, PTB sections from a wide global distribution were analyzed in order to determine the characteristic signature of the Siberian Traps in these successions, and thus, test the relationship between the eruptions of the Siberian Traps to the environmental deterioration of the PTB mass extinction.

The signature of the volcanic ash fall provides information regarding the geographic distribution of the windblown ash and the relationship between the regional environmental change and its effects upon marine and terrestrial ecosystems. Furthermore, the importance of the project is that its results will allow Earth Scientists to calculate with a greater precision the rate at which regional environmental degradation happens and the specific effects it has on biological species systems.

Figure 1. Geologic Timescale of the Late Permian into the Early Triassic. Includes geomagnetic polarity and associated bio zones from the Late Permian-Early Triassic. Figure created from Time Scale Creator 2014 https://engineering.purdue.edu/Stratigraphy/tscreator/index/index.php

2. BACKGROUND

2.1. The Permian paleoenvironment

The Permian was characterized by a major global climate shift, and the changes were witnessed within the compositional and dominance patterns, along with the biogeographical distribution of Permian flora. The Permian was ushered in during ice-house conditions followed by warming patterns that increased global temperatures, likely resulting from the build-up of atmospheric CO2 with interrupting periods of abrupt cooling (Saunders and Reichow 2009).

Climate modeling suggests that atmospheric pCO2 rose from values similar to pre-industrial levels in the Early Permian to as much as 10x the pre-industrial levels by the End-Permian, a

time when glaciation was completely gone and the Siberian Traps began erupting (Kiehl and

Shields 2005). The rise in CO2 is speculated to have resulted from the lack of continental land mass weathering from the absence of extensive mountain belts and increasing aridity (Kiehl and

Shields 2005). The buildup of CO2 led to both atmospheric and ocean warming, contributing to the loss of Gondwana polar ice sheets, resulting in decreased latitudinal oceanic circulation and mixing (Kidder and Worsley 2004). In addition to a Permian atmosphere that was already beginning to be characterized by low oxygen conditions, increased oceanic stratification and the spread of oceanic stagnation exacerbated the issue (Kidder and Worsley 2004).

Terrestrially, the Permian hosted a rich faunal and floral diversity and as time within the Permian progressed, (from the Early to Middle for example) the proportion of seed plants increased, which reduced the swamp dominant lycopsids and sphenosids (Ziegler 1990). Biome level biogeographical analyses have yielded seven distinct biomes within the Permian, the first being a cold temperate biome between a paleolatitude of 60o and 90o where the Siberian Traps were located (Willis and McElwain 2002). While in the area surrounding the Siberian Craton, the pervasive presence of abundant external water sources from the lagoons, swamps and shallow basins found by Czamanske et al. 1998 would have created intense water-magma interactions leading to phreatomagmatic eruptions (Jerram et al. 2015, Black et al. 2015), see

Figure 7. The remaining six biomes include a cool temperate biome that projected relatively greater diversity than the higher colder temperate biome, where the palynological and paleosol data from the latest Permian indicated an abundance of deciduous forests within the higher latitudes (Taylor et al. 1992). The northern hemisphere was composed mainly of cordaites,

sphenopsids and ferns, while the southern hemisphere was marked by a high proportion of glossopterids and lycopsids (Willis and McElwain 2002).

A warm temperate biome that occupied areas of Greenland, Scandinavia and North

America was characterized in the northern hemisphere by a high diversity of cordaites, pteridosperms, gingkoales, sphenopsids and ferns, along with the presence of glossopterids in the south (Willis and McElwain 2002). The mid-latitude desert biome gives evidence of aridity from the evaporate deposits in the southern hemisphere with no fossil floras preserved. The subtropical desert biome also shows strong evidence of evaporate deposits like the mid- latitude deserts, but support cycadales, gingkoales, and conifers on the coastal margins (Rees et al. 1999). The summer wet or tropical biome held ginkgoales, conifers and cordiates but was prone to seasonality in precipitation patterns (Willis and McElwain 2002), while the tropical ever wet biome that straddled the equatorial latitudes was composed of gigantopterids, pteridosperms, sphenopsids, ferns, and ginkgoales (Ziegler 1990; Rees et al. 1999). These floral changes coupled with the evidence of increased provinciality provide evidence for increased aridity and greater pole to equator temperature gradients (Ziegler 1990).

The Permian environment was also affected by the Pangaean supercontinent, which had a profound influence on the atmospheric circulation system. During the Permian, the main continental landmasses were orientated in a north-south direction enveloped by the

Panthalassic Ocean to the east, north, and south with the Tethys Ocean along the eastern edge

(see Figure 2). The single continent stretched latitudinally across every part of the zonal atmospheric circulation thereby producing an extraordinary effect on the Permian global climate (Dubiel et al. 1991). The transition from the Permian to the Triassic was unique in that

the extreme paleoclimate observed was due to the global occurrence of red beds and evaporate deposits within numerous study sites. This was highlighted by the sites within the

Colorado Plateau where there are over 2500 m of aeolian-deposited sandstones (Blakey et al.

1988; Dubiel et al. 1991). Paleowind data from the cross-strata within these aeolian units was compiled to create a dataset on the dune migration direction for the Permian inferring the paleowind directions (Peterson 1988).

Figure 2. Paleogeography world map of the Late Permian. Includes white outlining of contemporary continental extent along with the major subduction zones of the north and east of the Paleo-Tethys Ocean basin. Inferred wind direction patterns from Peterson 1988, Miller 2001. Figure taken from The Paleontology Portal paleoportal.org/kiosk/sample_site/period_11.html, modified by Santistevan 2016.

Peterson (1988) used this data set with reconstructed Pangaean paleogeography to model atmospheric circulation during the Permian and described the climate as monsoonal

with the presence of subtropical high-pressure cells off the northwest coast of Pangaea and monsoonal low-pressure cells off the east coast of Pangaea (Robinson 1973; Peterson 1988;

Parrish and Peterson 1988, Miller 2001). The subtropical high-pressure cells are a stable component of atmospheric circulation that forms from the temperature differentials between the land and sea (Parrish and Peterson 1988). The monsoonal circulation patterns likely allowed the development of semi-arid deserts at low latitudes while the large mountain chain; the

Central Pangean Mountains formed on the equator, probably enhanced monsoonal circulation

(like what is seen today with the Tibetan Plateau), while dominant western cross-equatorial

Pangaean winds flowed (Parrish and Peterson 1988; Dubiel et al. 1991, Miller 2001). The strong cross-equatorial winds blowing west thus would have influenced the distribution of the volcanic tephra produced from the Siberian Traps.

The closest historical analog to the Siberian Traps were the Miocene Columbian flood basalts and the Jurassic Purana basalts although the closest contemporary analogs to the

Siberian Traps in predictive eruption style and gas output was the Iceland volcanic eruption of

Laki in 1783. The closest analog in predictive ash dispersal was the Iceland volcanic eruption of

Eyjafjallajökull in 2010. The comparison between the two events to the Siberian Traps can be drawn from the following similarities: Laki was a fissure volcano that erupted for a duration of 8 months emitting as much as 14 cubic km of basaltic lava, some tephra and an estimated 80 Mt of sulfuric acid aerosol, 4 times more than El Chichon in 1982, and 80 times more than Mount

St. Helens in 1982 (Sigurdsson 1982). Laki also displayed intermittent pulses of explosive episodes and voluminous ash fall production in between lava emplacement pulses (Thordarsen et al. 2003). These explosive phases were found to generate immense fire fountains producing

eruption columns that breached the tropopause (Thoradsen et al. 2003). Eyjafjallajökull is a stratovolcano located at 63o37’12” N latitude and during April 2010 erupted and deposited a tephra distribution large enough that it disrupted and suspended all major airport traffic for 5 days (Hendry 2010). Based on the latitudinal position of Eyjafjallajökull, a composite map shows the amount of time it took ash to travel half way around the world (see Figure 3). The Siberian

Traps were erupted at a high northern paleolatitude, like the Eyjafjallajökull eruption, thus likely facilitated the injection of material into the stratosphere. The unusually high proportions of tephra derived from the Siberian Traps found its eventual deposition in many of sample sites used in this study, as will be discussed in Section 6.1. and Section 6.2.

Figure 3. Composite map of the ash travel from the 2010 Eyjafjallajökull eruption. Image spanning 14-25 April 2010 and based on data by the London Volcanic Ash Advisory Centre. The stratovolcano designated by the red dot, image density representing ash concentration and

extent. Figure taken from http://www.metoffic.gov.uk/aviation/vaac/vaacuk_vag.html

2.2. The Permian-Triassic Boundary crisis

The Permian-Triassic mass extinction event 251.9 million years ago (Burgess et al. 2014) has long been recognized as the greatest biological catastrophe ever in the history of the

Phanerozoic Eon where 95% of the marine species along with 75% of the terrestrial species went extinct (Erwin 2008). Marine invertebrates underwent extraordinary decimation during the End-Permian, and the extinction event not only reset the evolutionary clock for most of the major groups, but also marked one of the single most important large-scale events in the history of evolution, coming second only to the Cambrian Explosion. In the marine realm, trilobites, rugose corals, goniatites, strophomenate brachiopods, blastoids, and rostroconchs, along with spiriferid and orthid brachiopods, and fenestrate bryozoans all appeared for the last time (Payne and Clapham 2012).

The extinction was especially severe for crinoids where four of the five Permian clades went extinct, along with foraminifera where there was a 91% genus loss in total (Payne and

Clapham 2012). Reef systems were also hit hard and this redirected the entire landscape of the oceanic environment with the rugose and tabulate coral extinction giving rise to the scleractinian corals within the Triassic, along with the collapse of the hyper-calcified sponge microbe reef systems and the replacement of Triassic microbialite reefs (Payne and Clapham

2012). Jin et al. (2000) undertook an exhaustive study of the Late Changhsingian formations and recorded the occurrences of approximately 333 species from 162 genera of 15 different major

marine groups over a span of 64 fossil horizons and found that the Meishan section of South

China yields an extinction rate of 94% above Bed 25 while the extinction rate does not exceed

33% below Bed 24 (Jin et al. 2000), see Figure 4.

Figure 4. Stratigraphic ranges of fossil species (indicated by vertical gray lines). The Meishan section in China, with a time range spanning from the Late Permian into the Early Triassic shows species numbers on the x axes. 333 species from 162 genera from 15 different marine fossil groups were tested. Fossil ranges are scaled to rock thickness, with an abrupt faunal change near the base of bed 25, approximately 42 meters up, figure and caption taken from Jin et al 2000.

Terrestrially the decimation was just as bad, and although the resolution of the terrestrial fossil record doesn’t provide the same clarity that the marine record shows, extinction estimates range up to 75% at the Family level (Benton et al. 2004; Ward et al. 2005).

Of the 48 families that were present within the last 5 million years of the Permian documented within Russia and the Karoo basin, 36 had died out, with losses that included 10 families of basal tetrapods and 17 families of therapsids (Benton et al. 2004; Ward et al. 2005) thus

collapsing the complex multi-tier Late Permian ecosystems that had existed on land. Vertebrate extinction patterns were likely induced by major floral transitions taking place at the End-

Permian as well. Major plant ecosystems were being restricted in the midst of the environmental change of the Permian, however, despite evidence of trauma at the Permian-

Triassic boundary, much of the change occurred over a 25 million year period spanning the

Middle to Late Permian. During this interval the total Family diversity dropped by 50%, and while the environmental shifts had devastating impacts on ecosystems at the regional level, communities usually rebounded with re-vegetation patterns of conifer dominated forests, reorganization of existing planet species, and the evolution of entirely new species (Looy et al.

1999).

Further statistical analyses have shown that physiological susceptibility to hypercapnia from the elevated pCO2 played a pivotal factor in the selective survival amongst marine invertebrates (Knoll et al. 2007). Hypercapnia is the product from an increased level of carbon dioxide in the blood. With the increase of pCO2, it proves difficult for organisms to exchange oxygen for carbon dioxide within the lungs resulting in an inverse relationship between the amounts of oxygen and carbon dioxide in the bloodstream. In a reexamination of Sepkoski’s genus compendium, Knoll et al. (2007) separated the marine animals into three different groups based on their skeletal physiology, and found that the grouping of animals with a heavy calcium carbonate skeleton and a large ratio of calcium carbonate compared to their organic tissues suffered severe extinction, for example, corals and calcite brachiopods. The second order grouping was that of animals that had an intermediate ratio of a calcium carbonate skeleton to living organic tissue, while the third grouping was composed of animals whose

skeletons were comprised of any material other than calcium carbonate, such as cartilaginous fish etc. (Knoll et al. 2007). The results were overwhelmingly supportive of hypercapnic stress with over 85% of the genera in the first group going extinct at the End-Permian, while the extinction rates for the second and third grouping was only 54% and 5% respectively (Knoll et al. 2007). Other trigger mechanisms such as anoxia and hydrogen sulfide poisoning were certainly in operation as well with inducing the marine extinction numbers, however, Knoll’s results seem to implicate hypercapnia as the dominant mechanism, and that, because the

Siberian Traps released large amounts of pCO2, it is possible that the eruptions were the dominant cause of the extinctions (Payne and Clapham 2012), see Figure 5.

Figure 5. Physiological selectivity of the End-Permian extinction. Clades with poorly buffered respiratory physiology are shaded dark brown, moderately buffered clades are in brown, and well-buffered clades are in light yellow. Invertebrate extinctions are calculated as a raw percentage on the basis of all Changhsingian genera from Paleobiology Database data; foraminifera extinction is also a raw percentage from Groves & Altiner (2005). Figure taken from Payne and Clapham 2012.

Recovery from the End-Permian biotic crisis did not begin to unfold until about 5 million years after the extinction horizon and full recovery with ecosystem numbers matching those of pre-existing values took 10 million years (Chen and Benton 2012). Sun et al. (2012) undertook a

18 study of d Oapatite of conodonts from sections within the Nanpangjiang Basin of South China to reconstruct Late Permian to Middle Triassic equatorial sea surface temperatures and concluded that lethally hot ocean temperatures exerted a direct control on the biotic marine recovery, and that equatorial seaways were depauperate of fish diversity and calcareous algae. The temperatures of which would have exceeded 40o C at times and were characterized by widespread oxygen minimum zones that stretched from the deep ocean zones to shallow environments (Bottjer 2012). Furthermore, these waters would have been rich in hydrogen sulfide along with widespread anoxia and sulfide toxicity (Bottjer 2012). The ocean conditions created a pattern dictating taxa to emigrate from the tropics or succumb to extinction. Sun et al. (2012) suggest that low oxygen-dependent thermal tolerant organisms, such as the vertebrates were the first to vacate the equatorial latitudes thereby stalling ecosystem recovery.

The stratigraphy of the Permian-Triassic boundary crisis is critical in understanding the temporal and spatial distribution of the extinction, therefore, a quick review is essential. The

Global Boundary Stratotype Section and Point (GSSP) to the boundary lies in the Meishan section of South China and the boundary between the Permian and the Triassic is at the base of

Bed 27c, the Hindeodus parvus conodont horizon. The successive layers of rock that stretch throughout the Upper Permian are referred to as “beds” and are numbered 19 – 27 spanning

the Changhsingian formation, with Bed 25, being a volcanic ash layer designating the largest magnitude of the extinction (Hongfu et al. 2001). Some beds include a secondary method of rate and timing which is designated by a sequence of “a-d”. Running up the boundary, the rock composition of the Permian is a limestone characterized with layers of clay and chert. Bed 24 is dominated by a bioclastic micrite and argillaceous micrite composition, while Bed’s 25 and 26 constitute a white clay and black clay respectively, see Figure 6. The intermittent layers of ash recognized at the PTB and throughout the Lower Triassic of the Meishan GSSP and other sections in South China confirm that there was a significant explosive volcanism event at time that may have been related to the Siberian Traps.

Figure 6. The Global Stratotype Section and Point (GSSP) of the Permian-Triassic Boundary. The Meishan section outcrop that spans the entire Late Permian Extinction into the Early Triassic Recovery. Figure and photo taken from Hongfu et al. 2001.

2.3. Potential causes of the Permian-Triassic boundary crisis

The cause of the event has long been debated, and various hypotheses have been proposed, including a meteorite impact, flood basalt volcanism, global oceanic anoxia, and long-term climate change (Hallam and Wignall 1997; Wignall 2007). However, dating of the

Siberian Traps has shown that they were effectively coeval with the PTB mass extinction (Renne et al. 1995; Kamo et al. 2003; Shen et al. 2011) and, thus, a likely cause of that event. Albeit hard evidence linking the Siberian Traps to the mass extinction has been lacking. Burgess and

Bowring, using Chemical Abrasion-Thermal Ionization Mass Spectrometry (CA-TIMS) to improve upon existing 206PB/238U dating of the extinction and the Siberian Traps, isolated 17 zircon crystals from the provinces (two of which crosscut their respective lava sequences). They also isolated zircon crystals from intercalated welded tuffs within the successive lava units, along with pervoskite crystals located at the base of the Noril’sk and Meymecha-Kotuy to constrain the dates of the emplacement of the Siberian Traps. In addition to the CA-TIMS, they also utilized LA-ICMPS (Laser Ablation Inductively Coupled Plasma Mass Spectrometry) to age date over 800 isolated zircon crystals collected from the various pyroclastic rock deposits of the southern region of the Siberian Traps province using one laboratory, and the identical method parameters they employed in their 2014 paper (Burgess et al. 2014, Burgess et al. 2015). Their results display a synchrony confidence at the ~0.04% level or better, and project two major episodes of extrusive magmatic behavior in the Noril’sk and Meymecha-Kotuy regions. The first episode overlapped and actually preceded the extinction event by 300 ± 126 ky, followed by a

temporary queue in the magmatism/deposition just before 251.483±0.089 Ma, continuing on for at least 500 ky into the Early Triassic (Burgess et al. 2015). The entire eruptive event was emplaced/erupted in a maximum of ~2 Ma, and these new results calculate an extinguishing time frame to a maximum of 61 ± 48 ky (Burgess et al. 2015).

The Siberian Traps represent the largest volume of flood basalt provinces on Earth (see

Figure 7 below), accounting for ~2-3 million cubic kilometers of volcanic ejecta spewing out starting around 252 Ma (Renne et al. 1995; Reichow et al. 2002, 2009). The Siberian Traps also displayed a complex plumbing system where elaborate pipe structures fed from different dikes and sills complexes led to the development of effusive and explosive lava flows. Estimating the exact number of pipes has proved burdensome due to the repeated glaciations and resultant decompression of relief along with the expansive Taiga coverage of the area, however, prospecting surveys from iron ore deposit mapping projects gives a number of more than 300 magnetite-bearing pipes from the Tunguska basin alone (Polozov et al. 2015). These pipes are rooted at depths of 2-4 km within high temperature magma-sediment mixing zones and contain, while rare, basement xenoliths indicative of deep-seated metamorphic and magmatic rocks. Conduits filled with sediment breccias also characterize the pipes and volcanic rocks cemented by halite, calcite, anhydrite and craters up to a kilometer wide filled with altered volcaniclastic material (Svensen et al. 2009, Jerram et al. 2016). The magma-brine (evaporite) interaction has been theoretically and experimentally described and confirmed as to creating phreatomagmatic explosions by the volcanic molten-fuel-coolant interaction, or MFCI

(Zimanowski et al. 1997, Polozov et al. 2016).

This type of explosion is not only highly violent, but also transports fine volcanic tephra to high levels within the atmosphere and is able to traverse wide areas as well (Polozov et al.

2016). The explosive volcanism of the Siberian Traps could thus be broken down into three distinct groups: 1. Deeply rooted sediment-magma interactions and pipe eruption where sills were emplaced within evaporates, 2. Shallow magma-water interactions in areas inundated with water and 3. Lava flows and fire fountaining during the main stage of effusive volcanism

(Jerram et al. 2016), all leading to extensive degassing. Black et al. (2012) found that this degassing from the Siberian Traps would have resulted in approximately ~6300-7800 Gt of S,

~3400-8700 Gt of Cl, and ~7100-13600 Gt of F being discharged within the atmosphere. The location of the Siberian Traps at a high paleolatitude within an aqueous environment meant a lower troposphere, and water-magmatic interactions producing phreatomagmatic events and an easy means for the large volatile loads to be injected into the stratosphere from fire fountaining. As a result, this then led to the delivery of gases and tephra within the stratosphere and thereby causing the environmental deterioration (Black et al. 2012, 2013,

2015), see Figure 7.

Figure 7. Present day map of western Siberia. A. Corresponding colors reflect the outcropping of basaltic lavas and various volcanoclastic rocks, along with the distribution of intrusive igneous rock. B. The Late Permian paleogeographical setting of the Siberian Craton, highlighting the geographic setting and importance of the water-magma interactions. Figure taken from Black et al. 2015, modified from Czamanske et al. 1998.

It is now generally agreed upon that the Siberian Traps were the most likely mechanism in acting as the catalyst for the massive species loss, however it is unclear whether the thermal anomaly of the Siberian Traps originated from a mantle plume or was the result of a lithosphere controlled process. What is clear however, is that the large igneous province covered an expansive area and the quantity of gases emitted was substantial. The estimation of the original extent that the Siberian Traps covered has proved difficult owing to the erosional processes over the last 252 million years, but the most accepted estimates tend to fall between a low of 3 x 106 km3 with a high as much as 5 x 106 km3 (Saunders and Reichow 2009). Although the

lithology of the Siberian Traps is often broadly defined as being universal, i.e. basalts or tuffs, the exact qualifiers of the material can change dramatically and have important defining characteristics on the contextual setting from where the rock was derived. The Siberian Craton to the east is comprised mainly of basaltic lavas and tuffs in a thick outcropping, while west of the craton in the West Siberian Basin and along the Taimyr Peninsula extensive subcropping of basalts is found, and it is speculated that the hidden basalts may actually be greater in quantity than the overlying main outcrops of the east and north (Reichow et al. 2009). Unique REE signatures characterize the volcanic units (Wyman 1996) and published studies have shown that each successive suite or magma type that was emitted from the Siberian Traps produced a distinct REE chemical fingerprint (Lightfoot et al. 1990, 1993; Arndt et al. 1993, 1995, 1998;

Federenko et al. 1997, 2000), see Figure 8.

Figure 8. The Noril’sk region with REE ratios of La/Sm and Gd/Yb plotted against the individual chemical suites from the Upper Permian through the Lower Triassic. Stratigraphic figure taken and modified from Kamo et al. 2003, with REE data and figure taken from Lightfoot et al. 1993.

Much of the published work has been undertaken from the Noril’sk region because it is the thickest section and most complete sequence. It is also characterized by a wide variety of volcanic rocks including tuffs, picrites, and basalts (Saunders and Reichow 2009). The entire sequence is divided into 11 different suites dependent on geology and geochemistry and further subdivided into a Lower Sequence and an Upper Sequence. The observed 3 different magma suites of the Lower Sequence are the Ivakinsky, the Sverminksy, and the Gudchikhinsky while the observed 8 different magma suites of the Upper Sequence are the Khakanchansky, the Tuklonsky, the Nadezhdinsky, the Morongovsky, the Mokulaevsky, the Kharaelakhshy, the

Kumginsky, and the Samoedsky (Saunders and Reichow 2009). The Mokulaevsky, Lower

Nadezhdinsky and Tuklonsky suites of the Upper Sequence all represent sources of shallow melting with low titanium values, and little crustal contamination whereas the Ivakinsky, the

Sverminsky, and the Gudchikhinsky suites of the Lower Sequence all represent material deeply sourced, high in titanium and contains significant crustal contamination (Saunders and Reichow

2009), see Figure 9.

Figure 9. Correlated stratigraphic columns from the Noril’sk and Meymecha-Kotuy areas. Geochemical suites from the two major study sections of the Lower Triassic, suites are correlated and begin with the Tungusskaya Series of the Upper Permian and are matched radiometrically. Figure taken from Kamo et al. 2003.

Paleomagnetic, geochemical and paleontological correlations with the Upper Sequence of the Noril’sk region have also been established with the Putorana, the Tunguska and the

Meymecha-Kotuy regions, all of which have established correlations that match temporally by radiometric dating. (Kamo et al. 2003; Saunders and Reichow 2009). The Putorana Sequence comprises 3 suites of low Ti tholeiitic rocks, and running up from the Putorana base are the

Ayansky series, followed by the Honnamakitsky series (Kamo et al. 2003). The Honnamakitsky series is capped by the Nerakarsky series, which has been age correlated with 40Ar/39Ar techniques using Zircons to fall between the Kharaelakhshy and Kumginsky suites of the Upper

Sequence of the Noril’sk (Kamo et al. 2003), see Figure 9. The Meymecha-Kotuy sequence, which is located 500 km east-northeast of Noril’sk is correlated paloemagnetically and geochemically with a sequence as thick as 3000 m and represents an age younger than the

~3500 m of the Noril’sk Lower Sequence (Lind et al. 1994). The Meymecha-Kotuy sequence begins atop the Noril’sk section, above the last suite, the Samoedsky (Kamo et al. 2003). From the base of Sverminksy suite of the Noril’sk region to the top of the Tyvankitsky suite of the

Meymecha-Kotuy sequence is an age difference of 251.7 ± 0.4 and 251.1 ± 0.3 Ma respectively, synchronously reflecting the temporal framework in which the Siberian Traps erupted.

The latitude and placement of the Siberian Traps was also very critical to the extinction event. Placement wise, the Siberian Craton is surrounded by thick successions of rich coals and organic carbon dating back to the Proterozoic and Paleozoic. The extent of the basalts flooded over and extruded through these successions leading to the extreme decarbonation, which added to the already deadly preexisting degassing and volatile load (Knoll et al. 2007). It is inferred that these volcanogenic sulfate aerosols were the likely cause of the environmental collapse that led to the mass extinctions. Moreover, sulfate aerosols would have emitted enough dust and soot into the atmosphere to temporarily block incoming UV, which would

have led to cooling patterns that would have temporally impaired photosynthesis and disrupted plant communities. While CO2 would have led to long term warming with the loss of polar ice and reduction of ocean circulation, further disrupting ecosystems leading to the biotic crisis

(Saunders and Reichow 2009). This was also compounded by factors already ongoing in the

Permian like reduced orogenesis, reduced exposure of rock and weathering, reduction of CO2 drawdown, and steady accumulation of CO2 within the atmosphere (Saunders and Reichow

2009).

2.4. Rare Earth Element Geochemistry

The rare earth elements (REE) consist of the 15 chemical elements of the lanthanide series of the Periodic Table; scandium and yttrium show similar behavior and are sometimes included in the REE group. They are located in Group 3 of the 6th (5d electron configuration) and 7th (5f electron configuration) periods of the periodic table. They range from lanthanum (Z

= 57) to lutetium (Z = 71) and can be further divided according to their atomic weight as being light rare earth elements (LREE), middle rare earth elements (MREE), and heavy rare earth elements (HREE). The LREEs range from lanthanum to neodymium, the MREEs span from samarium to holmium, while the HREEs span from erbium to lutetium. Because the REEs form such a consistent group whose chemical properties change systematically and gradually from lanthanum to lutetium, their effectiveness in understanding reaction processes within marine environments has become invaluable in oceanographic studies (Elderfield 1988). Of these

chemical properties, REEs show little variation in solubility, and their unique electron shell structure results in similar ionic radii (Albarede 2003).

REE distributions can be “summarized” in the form of ratios for LREEs, MREEs, and

HREEs. By convention, LREEs are represented by La, MREEs by Sm, and HREEs by Yb, and the relative enrichments of LREEs to MREEs is given by Lan/Smn, that of LREEs to HREEs by Lan/Ybn, and that of MREEs to HREEs by Smn/Ybn (Shen at el. 2013). Some older studies make use of

Lan/Lun as a proxy for LREE/HREE, however, Lan/Ybn is preferred because it represents a ratio of two REEs with odd atomic numbers, thus, more similar concentration patterns in odd/even distributions (Sholkovitz et al. 1994; Lecuyer et al. 2004; Shen et al. 2013) because of the irregular odd/even REE abundance patterns that come up (Elderfield 1988). For the MREEs, Sm is used in place of Gd because of the potential anomalies associated with Gd in seawater (de

Baar et al. 1985; Alibo and Nozaki 1999; Shen et al. 2013). The ratios between LREEs, MREEs, and HREEs are often used to identify the differences in detrital and hydrogenous sources of

REEs (Musashino 1990; Lecuyer et al. 2004) along with interelemental fractionation processes

(Sholkovitz 1992; Sholkovitz et al. 1994, 1999). Complete REE distributions are usually plotted on spidergrams; histograms that plot the abundances of a given set of elements in an analyzed sample set relative to their abundance to a certain standard. Conventionally, the y-axis is elemental abundance in percentage and the x-axis is the range of elements, in this case La through Lu.

A spidergram depicting a negative slope will be relatively enriched with LREE, while a positive slope will depict relative HREE enrichment. The slope patterns are used to characterize the different types of material present in the sample along with defining the source of the

material. An enrichment of LREE is common in material derived from the upper continental crust, while HREE enrichment commonly reflects a lower-crustal or mantle source. Samples that are heavily influenced by seawater (called the “hydrogenous” fraction) will also show a pattern of relative HREE enrichment. Seawater is depleted in LREEs because of their preferential scavenging onto sinking particulates. Marine sediments (especially Mn nodules and

Fe-Mn crusts) are locally enriched in HREEs and seawater displays vertical gradients in REE patterns. Total REE concentrations increase with depth, and HREEs become more strongly enriched relative to LREEs with depth. However, these trends have seen to be reversed in water bodies that undergo extensive biogeochemical cycling because the REEs undergo dissolution, returning the preferentially removed LREE from the falling particles (Elderfield 1988).

Another important chemical property of the REEs is their valence state. REEs occur exclusively in the 3+ oxidation state with two exceptions; europium can take on a valence of 2+ while cerium can also have a valence of 4+, leading to chemical reaction differences in how Eu and Ce will partition versus the other REEs. Such reaction differences can lead to what are called Eu and Ce anomalies. These anomalies are defined as “negative” or “positive” by their comparisons with the value interpolated from the values of the adjacent REEs. Eu and Ce anomalies are calculated as follows:

0.5 Ce/Ce* = CeN / ((LaN *PrN) ) (1)

0.5 Eu/Eu* = EuN/((SmN *GdN) ) (2) where the Post-Archean Average Shale (PAAS) normalized concentration of each given element is represented by the subscript “N” (Taylor and McLennan 1985; McLennan 2001; Shen et al.

2013). The REE distributions from the PAAS normalized concentrations show very even

patterns worldwide and seem to have an enrichment of LREEs. The consistency suggests that the REE patterns of the shales reflect the values of the upper-continental crust (Taylor and

McLennan 1985), whereas an enrichment of HREE would suggest a source of mantle material.

Equations 1 and 2 are geometric approximations that generally yield similar values for the Ce and Eu anomaly. REE values normalized to chondritic abundances have been plotted to show total REE abundance for the upper, lower and total continental crust, see Figure 10 below.

Figure 10. REE patterns for the upper, lower and total continental crust. Normalized to chondritic abundances for Post Archaean Average Shale (P.A.A.S.), Average Archaean Shale (A.A.S.), and K-granite. Granite patterns produce a discernible Eu depletion, thus indicating K- rich granites are not a significant source for average Archaean sedimentary rocks. Figure and caption taken from Taylor and McLennan 1981, 1985).

Eu anomalies can occur in response to aeolian and certain hydrothermal inputs

(Elderfield 1988). Europium in a reduced divalent state, Eu2+ cations are preferentially incorporated into plagioclase, an abundant mineral in the Earth’s crust, because Eu2+ is similar in size and holds the same charge as Ca2+, a major constituent of plagioclase and other mafic minerals (Elderfield 1988). For this reason, enrichment or depletion of Eu in silicate minerals depends largely on the tendency to be incorporated into plagioclase. Since Eu in a trivalent

state (Eu3+) is an incompatible element in oxidizing magmas, it substitutes for Ca2+ in reducing magmas (Sinha 1983). In magmas that crystallize plagioclase, the plagioclase mineral will project a positive Eu anomaly and the remaining magma will show a negative Eu anomaly

(Sinha 1983). Therefore, this process of chemical differentiation can yield magmatic products that show either a positive or a negative Eu anomaly (Sinha 1983). The REE pattern of upper- crustal material has a marked depletion of Eu (Eu/Eu* = 0.64) due to the retention of Eu within plagioclase, however plagioclase is not stable below 40 km, so the lower crustal, upper mantle material is thus enriched in Eu (Eu/Eu* = 1.17).

The Ce anomaly is linked to the oxidation of Ce3+ to Ce4+. Ce anomalies can occur as a response to redox conditions in the ocean or in response to oxidization from atmospheric oxygen (O2) under alkaline conditions (Elderfield 1988). Ce anomalies are related to the solubility of Ce in aqueous systems. Under reducing conditions, Ce3+ is soluble, whereas under

4+ oxidizing conditions Ce precipitates as a solid phase, Ce02 (Shields and Stille 1998). The geochemical signature of Ce anomalies can also be preserved long term in the sediment provided that later diagenetic processes do not alter them. This is especially useful for sediments that were deposited under oxic or anoxic conditions (Shields and Stille 1998).

However the application of the Ce anomaly as a proxy for ocean anoxia through the measurement of REE concentrations of authigenic minerals can sometimes be misleading if early diagenetic alterations have taken place, and the recycling of REEs and reduced forms of

Fe, Mn and Ce related to organic decay will reset the Ce signal seen in the anomalies regardless if late diagenetic alteration of REEs is rare or not (Shields and Stille 1998).

3. METHODS

Over 400 samples of marine sedimentary rock collected from 7 Permian-Triassic

Boundary (PTB) sections throughout the world formed the foundation of this study. The specimens were analyzed for rare earth elements (REEs) in the Department of Environmental,

Earth and Ocean Sciences at the University of Massachusetts, Boston. All REE ratios were normalized against the “average upper continental crust” as reported by McLennan (2001; G-3).

Samples were prepared for the ICP-MS instrument, a Perkin Elmer ELAN 9000 (see Figure 11) by first predigesting 20 mg of the samples in 1 mL of nitric acid under the fume hood for 24 hours.

After which the samples were then transferred into Teflon bombs, placed on a carousel and run in a MARSxpress microwave from CEM industries capable of analyzing 32 samples at a time. The samples were then run at a setting of 15 minutes to ramp to a temperature of 200 degrees

Celsius, where the microwave would then hold at 200 degrees Celsius for another 15 minutes, then cool down for 15 minutes. After this, 0.5 mL of HNO3 and HCl were added, and the samples were again placed in the MARSxpress microwave and ran for a second round with microwave settings of a higher temp and a longer time. Upon completion, the samples were then diluted with 50 mL Milli-Q water and ready to be run in the ICP-MS instrument. Liquid sampling was conducted through an auto dilution auto sampler with Scott and Cyclonic spray chambers, Meinhard and cross-flow nebulizers that included microvolume and an Aridus desolvating nebulizer.

Figure 11. Instrumentation. The ICP-MS, a Perkin ELAN 9000, housed within the Environmental Analytical Facility at the University of Massachusetts, Boston. Photo courtesy of Jeremy Williams, Assistant Professor, Kent State University.

All samples were analyzed for whole rock major and trace element concentrations using a wavelength-dispersive Rigaku 3070 XRF spectrometer at the University of Cincinnati. Results were calibrated using USGS, NIST, and internal laboratory standards. A suite of carbonate and marl laboratory standards were used in this study, the compositions of which were determined by XRAL incorporated through XRF and INAA. Analytical precision based on replicate analyses was better than ±2% for major and minor elements and ±5% for trace elements. Detection limits were ~5 ppm for most trace elements. Carbon and sulfur elemental concentrations were measured using an Eltra 2000 C-S analyzer. Data quality was monitored via multiple analyses of the USGS SDO-1 standard (TC = 9.68 wt%; TS = 5.35 wt%), yielding an analytical precision (2σ) of

±2.5% of reported values for carbon and ±5% for sulfur. An aliquot of each sample was digested in 2N HCl at 50°C for 12 hours to dissolve carbonate minerals, and the residue was

analyzed for total organic carbon (TOC) and non-acid-volatile sulfur (NAVS); total inorganic carbon (TIC) and acid-volatile sulfur (AVS) were obtained by difference. Additional auxiliary analyses were performed on some samples in other laboratories including magnetic susceptibility, stable isotope analyses, and biomarker analyses.

4. STUDY SECTIONS

In the approach to looking at the PTB mass extinction, REE distributions in 7 different

PTB sections were analyzed using over >400 sedimentary samples. The wide geographic distribution from which the samples were undertaken was done in order to investigate the origin of the REE anomalies that have been observed at the boundary level in India and China

(Algeo et al. 2007; Shen et al. 2012). The study sections include outcrops in Japan (Gujo-

Hachiman and Ubara), Greenland (Black Ridge West), India (Spiti and Guryul Ravine), Iran (Zal),

and China (Chaohu), see Figure 12.

Figure 12. Global paleogeography of the Late Permian. Study sites identified by red stars; Siberian Traps shown by red field at top of map. Base map courtesy of Ron Blakey, Colorado Plateau Geosystems (http://jan.ucc.nau.edu/~rcb7/); modified figure provided by Santistevan 2016.

4.1. Gujo Hachiman, Japan

The Gujo-Hachiman section is now located within the Mino and Tanba Belts of central

Japan but was originally deposited in the Panthalassic Ocean. At the time of deposition, the section was located in the middle of the Panthalassic Ocean at a paleolatitude within ~10o of the paleoequator as shown by paleomagnetic studies (Shibuya and Sasajima 1986; Ando et al.

2001). The Gujo-Hachiman section was deposited in a moderate-productivity setting, a small distance away from the paleoequator (Algeo et al. 2011). The entire Gujo-Hachiman section consists of chert and a black shale unit, where the lowermost 7 meters is a thick section of chert

with the uppermost 65 cm the black shale (Algeo et al. 2010). Within the basal 40 cm of the chert unit there are thin fossil bearing siliceous laminae found to be Middle to Late

Changhsingian age based on radiolarian biostratigraphy, above the shale unit the section is devoid of fossils and the boundary placement is somewhere within that unit however the exact placement is cryptic (Algeo et al. 2010), see Figure 13.

Previous work from Algeo et al. (2010) demonstrated that the transition from a decrease in chert to the increase of black shales represented 4-5x decrease shifts in the sinking flux of biogenic silica from the eradication of the radiolarian silica, projecting massive changes to the food chain either through a reduction in productivity or a change in primary producers no longer favored from the radiolarians. Wignall and Twitchett (1996) also found that the lithologic shift was evidence that the Japanese deep-water section underwent a large transformation from oxic to euxinic conditions. This is reflected through the 30x increase of framboidal pyrite from syngenetic origin (Algeo et al. 2010). The Gujo-Hachiman pattern therefore showed that the ultimate indicator for the redox conditions was instead the nutrient fluxes and high primary productivity rates, opposed to the stagnant Panthalassic circulation patterns and the regions of the Panthalassic that experienced limited redox changes (Algeo et al. 2010).

Figure 13. Gujo-Hachiman outcrop. The extinction horizon of Gujo-Hachiman, chert represented by the numeral I in the Upper Permian transitioning into siliceous claystones. The boundary is represented by the dashed line. Photo taken from Algeo et al. 2010.

4.2. Ubara, Japan

The Ubara section was deposited within the Panthalassic Ocean in a deep abyssal, high productivity area close to the paleoequator, and it is now found, along with the Gujo-Hachiman section, within the Mino-Tanba Belts of central Japan (Algeo et al. 2010). At the time of deposition, the section was located in the middle of the Panthalassic Ocean at a paleolatitude within ~10o of the equator as shown by paleomagnetic studies (Shibuya and Sasajima 1986;

Ando et al. 2001). Ubara constitutes 3 lithostratigraphic units; the basal unit is more than 3 meters thick and is made up of gray bedded cherts, the middle unit is around a 1 meter thick and is a composite of alternating gray chert and black shale beds with the shale beds containing an abundance of pyrite nodules, while the tertiary unit is over 1 meter thick and is made up entirely of anoxic black shales (Kakuwa et al. 2008; Algeo et al. 2011), see Figure 14.

The sedimentation rate for Ubara was low, <10 m Myr-1, which is consistent with the abyssal depositional settings (Algeo et al. 2011). The Permian-Triassic boundary falls between

the second and third lithostratigraphic unit and was constrained using radiolarian (Kuwahara et al. 1991, 1998) and conodont biostratigraphy (Yamakita et al. 1999). Neoalbaillella ornithoformis, a Mid-Late Permian age radiolarian was collected from Unit I, and the radiolarian

A. triangularis, a radiolarian assigned to the Neoalbaillella optima zone from the latest Permian, was found in Unit II (Algeo et al. 2011). In contrast, the conodont H. parvus was found 10 cm above Unit II placing the Lower Triassic within the Black Shales of Unit III (Algeo et al. 2011). The section is measured within 25 beds, designated by a UB prefix starting at UB-25 in the Late

Permian running up to UB-01 in the Early Triassic.

Previous studies by Kakuwa et al. (2008) documented that within the Ubara section radiolarian chert gradually shifts to an argillaceous chert and that the last radiolarian chert bed occurs at 10 cm below the boundary, and that the decrease in radiolarian productivity started from 80 cm below the boundary. By the onset of the Griesbachian, from the stratigraphic distribution of the 56 Late Permian radiolarians from the 37 genera that were preserved in the chert, all had almost completely disappeared (Sano et al. 2010). The kill mechanism was derived from the anoxic condition of the Panthalassic surface layer (Sano et al. 2012). The Ubara decease in radiolarian productivity coincides with TOC increases from the sudden onset of anoxia laying cause for the linkage between the diversity loss of the Permian radiolarians and the rapid environmental collapse (Kakuwa et al. 2008).

Figure 14. Ubara outcrop. The extinction horizon represented by the short dashed line, with a transition from chert to black shales below the boundary. A 2mm pyrite layer is also identified. Photo taken from Algeo et al. 2010.

4.3. Black Ridge West, Greenland

The Black Ridge West section is from the East Greenland Basin and belongs to the

Wordie Creek Formation, comprising >700 m of siliciclastic sediments deposited in a marine setting spanning the Upper Permian. The East Greenland Basin comprised a narrow, elongate, fault-controlled basin during the Early Triassic at a time of active rifting and rapid subsidence, resulting in one of the most expanded Permian-Triassic boundary sections in the world

(Twitchett et al. 2001). At the time of deposition, the section was located along the northern boundary of the Euramerican plate at ~30 degrees north paleolatitude. Immediately underlying the dark gray, micaceous, pyritic and laminated muddy siltstones of the Wordie Creek is the

13 Schuchert Dal Formation, and it is within this transition that d Ccarb values show a negative

13 excursion of 8%-9% combined with a negative excursion in the d Corg values of 10%-11%. These

shifts document the total collapse of the marine ecosystem, and the eventual collapse of the terrestrial ecosystem, all within a time period of only about 10 to 60 ky (Twitchett et al. 2001).

13 The expanded character of this section, the presence of Hindeodus parvus, the detailed d Ccarb

13 and d Corg profiles, and the potential for correlation with coeval terrestrial sections in the same region make this one of the most important Permian-Triassic boundary sections in the world, see Figure 15.

The section records an abundant marine record in conjunction with a terrestrial palynomorph record. Extensive petroleum work within the sedimentary basin undertook by

Stemmerik et al. (1992; Pattison and Stemmerik 1996) combined with earlier palynostratigraphic work from Piasecki (1984) laid much of the biostratigraphic information down. Permian sediments were constrained with the documentation of the ammonoid

Cyclolobus kullingi, however the section has major sampling gaps. Hindeodus parvus first appears 23.5 m above the top of the Schuchert Dal and at bottom of the Wordie Creek, however the latter is 10 m before Claraia, the bivalve disaster taxa of the Lower Triassic appears, suggesting the boundary could be reassigned (Twitchett et al. 2001). Tetrad spore assemblages, fungal remains, collapse of gymnosperms and pollen deficiency all point towards evidence for a shift in floral patterns terrestrially, and this shift coincides with the same stratigraphic interval as the marine realm (Looy et al. 1999).

Figure 15. Black Ridge West outcrop. Lithostratigraphic and biostratigraphic data from Black 13 Ridge West constrained to the δ Corganic curve. Figure taken from Twitchett et al. 2001.

4.4. Spiti, India

In the Spiti section of the Indian Himalayan region, the Upper Permian is composed of black shale with subordinate layers of brachiopod-bearing siltstones that belong to the Kungri

Formation (Krystyn and Orchard 1996), see Figure 16. The Kungri is capped by an unconformity, above which are found brownish highly fossiliferous, silty limestone beds of the Lower Triassic, where a stark contrast between the boundaries is clearly recognizable (Krystyn and Orchard

1996). At the time of deposition Spiti was located in the southern end of the Tethys Ocean along the northern margin of Gondwana at a paleolatitude of ~30 degrees south.

Biostratigraphic control of the Lower through Middle Triassic was undertaken by Krystyn et al.

(1996) using ammonoid zonation and revised again in 2004. Orchard and Krystyn (1998) constrained the Lower Triassic even further with conodont zonations. The samples utilized in this study were collected from the Muth site in particular, located off the Pin River, south of the

Guling, Lingti and Lalung sites. Muth is a Hindeodus typicalis zone and precedes the H. parvis datum, however, because of the younger date of the sediments, Muth contains the Clarkina species and implies H. parvis was extant (Krystyn et al. 1996). The section is 110-120 m thick, and takes on a rich fossiliferous assembly from a distinct gray, brownish weathered limestone band of the Orthoceras bed (Krystyn et al. 1998).

Figure 16. Spiti outcrop sections. Lithostratigraphic data from the Spiti Valley cross correlating the section sites of Muth, Guling and Lalung. Figure taken from Bhargava et al. 2004.

4.5. Guryul Ravine, India

At the time of deposition, Guryul Ravine was located in the southern Tethys Ocean along the northern margin of Gondwana at a paleolatitude of ~30 degrees south. The Guryul

Ravine section in Kashmir is relatively expanded, with faunal changes taking place over a

stratigraphic interval of ~100 m, of witch the Upper Permian is composed of calcareous sandstones and sandy limestones interbedded with reworked-graded-hummocky mudstones

(Brookfield et al. 2003). The section is thought to be continuous and complete through the

Permian-Triassic boundary interval, see Figure 17. The depositional settings of the Upper

Permian, Zewan Formation show evidence of a storm influenced inner shelf environment that underwent effects of salinity changes resulting in depauperate fossil assemblages (Nakazawa et al. 1975; Brookfield et al. 2003).

A sedimentary gap separates the transition from the Zewan sandstones into the

Khunamuh Formation shales and is marked by increases in the benthic diversity, quartz reductions and concentrations of mud input (Brookfield et al. 2003). Silty organic shales further define the Khunamuh Formation, however the lack of dark organic rich laminae and pyritic precipitates is evidence for no episodes of anoxic conditions during the time of deposition

(Brookfield et al. 2003). The boundary beds show evidence of reworking and breakage as well as a high level of bioactivity but little fossil appearances, while at the top of the Zewan only a few foraminifera species are found along with one brachiopod (Nakazawa and Kapoor 1981).

Hindeodus parvus does not appear until 3.5 m into the Khunamuh coinciding with the dominant bivalve Claraia and six other Triassic ammonoid species of Orthoceras, Lytophiceras, and

Glytophiceras (Nakazawa and Kapoor 1981; Kapoor 1996).

Chemostratigraphic studies utilizing major and minor trace elements: TOC, TIC, REEs, and d13C-d15N have been undertaken with the intent to explore the environmental changes spanning a 20 m section across the boundary (Algeo et al. 2007). C-isotopes depict two negative events before the extinction horizon within the Upper Permian constrained by the

Changxingensis zone correlating with eustatic rise which continues into the Lower Triassic suggesting the Siberian Traps as the driver (Algeo et al. 2007). The consistent negative C shift is correlative with the other Permian-Triassic boundaries that experience C excursions of -3% to -

8% percent, however, Guryul Ravine is unique in that the C signal is gradual opposed to abrupt as seen in other sections, so the gradualness speaks to the stratigraphic completeness of the section (Algeo et al. 2007).

Figure 17. Guryul Ravine outcrop. Lithostratigraphic data from the Muth section of Guryul Ravine. Figure taken from Brookfield et al. 2003.

4.6. Zal, Iran

The Zal section of Iran was deposited on a microcraton in the middle of the Tethys

Ocean, at a paleolatitude of ~30 degrees south (Horacek et al. 2007). Because of rapid

syndepositional subsidence, the Upper Permian and Lower Triassic succession is >800 m thick at

Zal (Horacek et al. 2007). The Upper Permian part of the section comprises gray nodular limestones and intercalated black shales that accumulated in shallowing-upward megasequences (Horacek et al. 2007). Uppermost Permian beds are marked by the presence of the ammonoid Paratirolites while the gray, yellow to red clays designate the Permian-Triassic boundary, constrained with the appearance of Hindeodus parvus (Horacek et al. 2007).

Deposited over this are yellow-gray, laminated platy limestones of the Elikah Formation of the

Lower Triassic with evidence of oncoid and stromatolitic structures. Furthermore, the boundary is dominated by the presence of the bivalve Claraia throughout the Lower Triassic, which makes a first appearance 11 m up from the boundary (Horacek et al. 2007), see Figure 18.

From the boundary 20 m upsection are thick-bedded dark gray semi-bioturbated laminated limestones, overlain 23 m up by a 30 m volcanic sill superseded by black nodular limestones rich in Claraia (Horacek et al. 2007). From there, gray laminated limestones dominant until 130 m with the second volcanic sill at 178 m, followed by dark gray weakly bioturbated, abundant rich pyrite nodular limestones until 190 m upsection (Horacek et al.

2007). 230 m upsection are thin-bedded, fine-grained limestones with oncoids, 40 m above that are intercalated layers of gray limestones and red oolitic/oncolitic layers followed by burrowed pale grayish yellow finely laminated marls (Horacek et al. 2007). 410 m up the section is identified with alternating red, yellow, gray oncolites, and at 429 m up the section shifts to exposed limestones with oolitic grainstones (Horacek et al. 2007). Marls characterized with tempestite cycles rich in microfossils take over 474 m upsection until 670 m up, during which the marls undergo carbonate content mixing with dolomitized brownish gray limestones

(Horacek et al. 2007). The section is capped 665 m up with thick Middle Triassic dolomites

(Horacek et al. 2007).

Previous Isotope studies undertaken by Horacek et al. (2007) recorded high d13C values within the Upper Permian, after which isotope values rebound to more positive values within the Griesbachian after the boundary is crossed (Horacek et al. 2007). Ascending upsection values continue to rise through the Dienerian, reaching as high as 8% into the Smithian boundary, where values then level off to below 0% followed by a second positive excursion of

3% only to drop again to negative values in the Spathian ending with a positive rise by the

Anisian boundary. The isotope results are consistent with L’Uomo, the Italian section, evidence that the western end of the Tethys was recording the same signal from the eastern end of the

Tethys. This pattern represents a consistent trend in oceanographic changes with global significance and a possible trigger pulled from the Siberian Traps.

Figure 18. Zal outcrop. An extensive lithologic column from the Zal outcrop cross compared to 13 two other Iranian PTB sections, plotted with δ C curves. Figure taken from Horacek et al. 2007.

4.7. Chaohu, China

The Chaohu section was deposited on the South China craton, at a paleolatitude of

~30 degrees north during the Early Triassic, in the eastern Tethys Ocean region. It consists of strata from the Upper Permian Dalong Formation and Lower Triassic Yinkeng, Helongshan, and

Nanlinghu Formations. It is over 250 m thick and is composed of micritic limestones, nodular limestones and shale, along with argillaceous mudstones (Shuangying et al. 2007). The Lower

Triassic units record consistent and successive patterns of regressive/transgressive cyclicity in the form of mudstone/limestone cycles (Gang et al. 2008). The section represents a deep open-

water basin to lower slope setting, as inferred from the lack of pervasive sedimentary structures and bioclastic particles along with the presence of pyrite and high V/(V+Ni) ratios

(Shuangying et al. 2007). The sedimentary cycles are laterally extensive and may record fluctuating sea levels, possibly indicative of a Milankovitch orbital control (Shuangying et al.

2007). However, tectonic and volcanic activity between the North China Block and the South

China Block could have been the catalyst for the relative sea-level changes and the development of cyclicity through repeated uplift, subsidence and deformation of the basin

(Shuangying et al. 2007), see Figure 19.

Cyclostratigraphic studies from Gang et al. (2008) also mapped the biozonation of the

Upper Permian and Lower Triassic. Calculating precession cycles and depositional rates they come up with an age of 252.6 Ma for the Permian-Triassic boundary constrained to Mundil et al. (2004). Permian through Triassic sediments are marked by the presence of 8 conodont zones beginning in ascending order with Hindeodus typicalis, Clarkina krystni, Neospathodus kummeli,

Neospathodus dieneri, Neospathodus waageni (the appearance defining the Induan-Olenekian boundary), Neospathodus pingdingshanesis, Neospathodus homeri, and Neospathodus anhuinensis (Zhao et al. 2008). Ammonoid zones correspond as the following; Ophiceras-

Lytophiceras through the Neospathodus krystni zone, Prionolobus-Gyromites through the

Neospathodus dieneri zone and Flemingites-Euflemingites aligned with Neospathodus waageni zonation (Gang et al. 2008). The bivalve Claraia is prevalent throughout the entire Griesbachian

(Gang et al. 2008; Zhao et al. 2008).

Figure 19. Chaohu outcrop. Chaohu lithologic column plotted against sedimentary cycles and regressive sequences. Figure taken from Shuangying et al. 2006.

5. RESULTS

5.1. Gujo-Hachiman, Japan

5.1.1. Whole rock, Al, and TOC values

Ninety-nine samples were analyzed for Gujo-Hachiman with data sets for whole rock percent, percent Al and TOC, REE ratios, along with Ce/Ce* and Eu/Eu* anomalies, see Table 1.

The Gujo-Hachiman stratigraphic section spans a depth of over 750 cm, and the PTB is located at sample site ITJ-96. All stratigraphic positions will be cited relative to “zero” at the base of the section. Al values range from 0.23% to 6.64% with an overall average of 1.61%. Six hundred and ninety-five cm upsection, the section is dominated by 80% - 95% chert with a transition to an illite influence in the next 58 cm. The percent increase in illite from sample ITJ-14 at 695 cm to sample ITJ-13 at 705 cm records a 24.3% reduction in chert. Samples ITJ-13 through the top of the section at sample ITJ-1 record an average of 51% chert, 49% illite through 58 cm. The increase in illite reflects an increase in Al, and Al values average rise by 4.13% from sample ITJ-

14 at 695 cm through the top of the section. TOC values reflect Al values through the illite to chert transition. The 705 cm mark of the section records TOC values consistently below 1% while TOC values from sample ITJ-13 through the top of the section average 1.24% see Figure

20.

Figure 20. Gujo-Hachiman whole rock, Al, and TOC values. From left to right, sedimentary values for Al and TOC, lined up against the lithostratigraphic column of Gujo-Hachiman. Figure taken from Algeo et al. 2003.

5.1.2. REE ratios

La/Sm ratios range from 0.11 at sample ITJ-24 at a depth of 647 cm, to 1.17 at sample

ITJ-6 at a depth of 737 cm, while average ratios value at 0.90. La/Yb ratios range from 0.20 at sample ITJ-24 at a depth of 647 cm, to a value of 4.22 with sample ITJ-11 at a depth of 713 cm, with an overall section average of 2.52. Sm/Yb ratios range from 1.20 with sample ITJ-14 at a depth of 695 cm to 4.00 with sample ITJ-41 at a depth of 473 cm, with an overall section average of 2.72. The overall signal for the ratios project an increase in values toward the top of the section with a steady increase for all three ratios around sample ITJ-38 at a depth of 497 cm, with an overall enrichment of middle rare earth elements, see Figure 21.

Figure 21. Gujo-Hachiman REE profiles. From left to right, sedimentary values of La/Sm, La/Yb, Sm/YB, and the combined REE ratio plot, lined up against the lithostratigraphic column of Gujo- Hachiman. Figure taken from Algeo et al. 2003.

5.1.3. Ce/Ce* and Eu/Eu* anomalies

Ce/Ce* values range from a section low of 0.79 with sample ITJ-24 at a depth of 647 cm to a section high of 1.39, 6 cm above with sample ITJ-22. Ce/Ce* values average 0.90 throughout the entire section. Eu/Eu* values range from a section low of 0.82 with sample ITJ-

11 at a section depth of 713 cm to a section high of 1.61 with sample ITJ-65 at a section depth of 308 cm. Eu/Eu* values average at 1.05 throughout the entire section, see Figure 22.

Figure 22. Gujo-Hachiman cerium and europium anomaly profiles. From left to right, sedimentary values of Ce/Ce* and Eu/Eu* lined up against the lithostratigraphic column of Gujo-Hachiman. Figure taken from Algeo et al. 2003.

5.2. Ubara, Japan

5.2.1. Al, TOC, and TIC values

Twenty-five samples were analyzed for Ubara with data sets for whole rock percentage, percent Al, percent TOC and TIC, REE ratios, along with Ce/Ce* and Eu/Eu* anomalies, see

Table 2. The Ubara stratigraphic section spans a depth of 128.5 cm and the PTB is located at sample site UB-17. All stratigraphic positions are cited relative to “zero” at the base of the section. Al values range from a section low of 0.67% with sample UB-03 near the base of the section at a depth of 28 cm, to a section high of 5.72% with sample UB-18, one sample interval directly above the LPME horizon at a depth of 91.8 cm upsection. Al values average at 3.01% overall throughout the section. Whole rock percentage of Ubara mimics the whole rock transition at Gujo-Hachiman with a shift from a chert domination of 80-90% near the base of the section, to 52% at the top of the section. The increase of illite subsequently leads to an increase of Al throughout the section. TOC acts in the same manner and records a 300% increase after the LPME horizon from sample UB-17 to the top of the section. Before the LPME horizon, TOC values average 0.13% while the average climbs to 0.52% beyond the LPME horizon, while TIC values average 0.12% throughout the entire section, see Figure 23.

Figure 23. Ubara aluminum, total organic carbon, and total inorganic carbon profiles. From left to right, whole rock percentage for Ubara, with light gray total Chert percentage and dark gray total Illite percentage, next 3 plots are sedimentary values for Al, TOC and TIC, lined up against the lithostratigraphic column of Ubara. Figure taken from Algeo et al. 2003.

5.2.2. REE ratios

La/Sm ratios range from a section low of 0.11 with sample UB-22 at a depth of 118.8 cm, 5 sample sites above the LPME horizon, to a section high of 0.77 with sample UB-09 at a

depth of 58 cm. La/Sm values average 0.3 throughout the entire section. La/Yb ratios range from a section low of 0.14 at sample site UB-22, 27 cm above the LPME horizon, to a section high of 0.86 with sample UB-09 at a depth of 58 cm. La/Yb values average 0.5 throughout the entire section. Sm/Yb ratios range from a section low of 0.9 at two different sample, the first with UB-16 immediately below the LPME horizon at a depth of 90.8 cm and again with sample

UB-13, exactly 30 cm below UB-16. The section high of 1.26 is recorded with sample UB-22, at a depth of 118.8 cm. Sm/Yb values average 1.38 throughout the entire section with an overall enrichment of MREE, see Figure 24.

Figure 24. Ubara REE profiles. From left to right, sedimentary values for La/Sm, La/Yb, Sm/Yb, and the combined REE ratio plot lined up against the lithostratigraphic column of Ubara. Figure taken from Algeo et al. 2003.

5.2.3. Ce/Ce* and Eu/Eu* anomalies

Ce/Ce* values range from a section low of 0.80 with sample UB-22 at a depth of 118.8 cm, to a section high of 1.12 at the very top of the section at 133.8 cm with sample UB-25.

Ce/Ce* values average 1.0 throughout the entire section. Eu/Eu* values range from a section low of 0.96 with sample UB-09 at a depth of 67.5 cm to a section high of 2.34 with sample UB-

11 at a depth of 118.8 cm. Eu/Eu* values average 1.2 throughout the entire section, see Figure

25.

Figure 25. Ubara cerium and europium anomaly profiles. From left to right, sedimentary values of Ce/Ce* and Eu/Eu* lined up the lithostratigraphic column of Ubara. Figure taken from Algeo et al. 2003.

5.3. Black Ridge West, Greenland

5.3.1. Al, TOC, and TIC values

One hundred twenty-six samples were analyzed for Black Ridge West with data sets for percent Al, percent TOC and TIC, REE ratios, along with Ce/Ce* and Eu/Eu* anomalies, see

Table 3. The Black Ridge West stratigraphic section spans a depth of 93.10 m, and the PTB is located at sample site BRW-10. All stratigraphic positions are cited relative to “zero” at the base of the section. Al values range from a section low of 4% with sample BRW-P-19, 2.20 m from the base of the section, to a section high of numerous samples weighting in at over 10%. The section takes on a transition in total Al percentage, increasing in values as the section progresses, where values weight in under >10% until sample BRW-P-34, 5.65 m from the base, when values above then begin to fluctuate between 9-10%. TOC values range from section lows of 0.08 % to section highs of 0.63% with an average of 0.18% throughout the entire section. TIC values range from section lows of 0% to a section high of 5.08% near the base of the section with sample BRW-P-19, 2.20 m from the base of the section. TIC values average 0.59% throughout the entire section, and graphic representatives depict higher values near the base with a decrease in values averaging in at >1% after sample BRW-P-4, 3.70 m from the base, see

Figure 26.

Figure 26. Black Ridge West aluminum, total organic carbon, and total inorganic carbon profiles. From left to right, sedimentary values of Al, TOC, and TIC, lined up the lithostratigraphic column of Black Ridge West. Figure taken from Twitchett et al. 2003.

5.3.2. REE ratios

Fifteen samples for LREE/HREE REE ratio activity were analyzed and begin with sample

BRW-P-40, 8 m from the base of the section. La/Sm ratios range from a section low of 0.5 with sample BRW-P-126 at the top of the section, to a section high of 1.04 with sample BRW-P-67,

14.45 m upsection. La/Sm values average 0.66 overall throughout the entire section. Fourteen

samples for LREE/MREE REE ratio activity were analyzed and begin with sample BRW-P-40, 8 m from the base of the section. La/Yb ratios range from a section low of 0.83 with sample BRW-P-

65, 9.80 m from the base of the section, to a section high of 2.96 with sample BRW-P-49, 19.35 m from the base of the section. La/Yb values average 2.17 overall throughout the entire section. Fourteen samples for MREE/HREE REE ratio activity were analyzed and begin with sample BRW-P-40, 8 m from the base of the section. Sm/Yb ratios range from a section low of

1.1 with sample BRW-P-65, 9.80 m from the base, to a section high of 4.43 with sample BRW-P-

81, 25.85 m from the base of the section. Sm/Yb values average 3.45 overall throughout the entire section, see Figure 27.

Figure 27. Black Ridge West REE profiles. From left to right, sedimentary values of La/Sm, La/Yb, Sm/Yb, and the combined REE ratio plot, lined up the lithostratigraphic column of Black Ridge West. Figure taken from Twitchett et al. 2003.

5.3.3. Ce/Ce* and Eu/Eu* anomalies

Fifteen samples for Ce/Ce* and Eu/Eu* anomalies were analyzed and begin with sample

BRW-P-40, 8 m from the base of the section. Ce/Ce* values range from a section low of 0.9 with sample BRW-P-65, 9.80 m from the base of the section, to a section high of 1.25 with sample

BRW-P-40 at the base of the section. Ce/Ce* values average 1.04 overall throughout the entire

section. Eu/Eu* values range from a section low of 0.96 with sample BRW-P-65, 9.80 m from the base of the section, to a section high of 1.12 with sample BRW-P-126 at the top of the section. Eu/Eu* values average 1.02 overall throughout the entire section, see Figure 28.

Figure 28. Black Ridge West cerium and europium anomaly profiles. From left to right, sedimentary values of Ce/Ce* and Eu/Eu*, lined up the lithostratigraphic column of Black Ridge West. Figure taken from Twitchett et al. 2003.

5.4. Spiti, India

5.4.1. Al, TOC, and TIC values

One hundred thirty-five samples were analyzed for Spiti with data sets for whole rock percentage, percent Al, percent TOC and TIC, REE ratios, along with Ce/Ce* and Eu/Eu* anomalies, see Table 4. The Spiti stratigraphic section spans a depth of 10.5 m of the

Changhsingian, overlain by 16.02 m of the Lower Triassic, a succession that spans up until the

Spathian. All Triassic stratigraphic positions are cited relative to “zero” at the base of the LPME horizon located at sample site AL2-9. The Upper Permian depth ascending from -10.50 m. Al values range from a section low of 0.04% with sample AL2-74 located within the Smithian, 9.42 m above the LPME, to a section high of 11.35% with sample AL2-10 at a depth of 0.48 m above the LPME horizon deposited within the Griesbachian. Al values average 4.42% overall throughout the entire section. TOC values range from a section low of 0.08% with sample AL2-

086, located within the Smithian, at a depth of 11.49 m above the LPME horizon, to a section high of 2.71% with sample AL2-110 near the top of the section, 15.43 m up from the LPME horizon deposited within the Spathian. TOC values average 0.49% overall throughout the entire section. TIC values range from a section low of 0% to a section high of 12.03% with sample AL2-

097, located within the Smithian, 13.09 m above the LPME horizon. TIC values average 6.12% overall throughout the section, see Figure 29.

Figure 29. Spiti aluminum, total organic carbon and total inorganic carbon profiles. From left to right, sedimentary values for Al, TOC, and TIC, lined up against the lithostratigraphic column of Muth. Figure taken from Bhargava et al. 2004.

5.4.2. REE ratios

One hundred thirty five samples were analyzed for LREE/HREE ratio activity. La/Sm ratios range from a section low of .38 with sample AL2-79, to a section high of 1.86 with sample

AL2-085. La/Sm values average 1.04 throughout the entire section. La/Yb values range from a

section low of .71 with sample AL1-12, to a section high of 3.16 with sample AL2-56. La/Yb values average 1.41 throughout the entire section. Sm/Yb values range from a section low of

.57 with sample AL2-14, to a section high of 3.20 with sample AL3-12A. Sm/Yb values average

1.40 throughout the entire section, see Figure 30.

Figure 30. Spiti REE profiles. From left to right, sedimentary values of La/Sm, La/Yb, Sm/Yb, and the combined REE ratio plot, lined up against the lithostratigraphic column of Muth. Figure taken from Bhargava et al. 2004.

5.4.3. Ce/Ce* and Eu/Eu* anomalies

One hundred thirty five samples were analyzed for Ce/Ce* and Eu/Eu* anomalies.

Ce/Ce* values range from a section low of .27 with sample AL2-113, located at the ceiling of the section, to a section high of 1.41 with sample AL2-77. Ce/Ce* values average at .96 throughout the entire section. Eu/Eu* values range from a section low of .64 with sample AL2-19, to a section high of 2.08 with sample AL2-54. Eu/Eu* values average 1.02 throughout the entire section, see Figure 31.

Figure 31. Spiti cerium and europium anomaly profiles. From left to right, sedimentary values of Ce/Ce* and Eu/Eu*, lined up against the lithostratigraphic column of Muth. Figure taken from Bhargava et al. 2004.

5.5. Guryul Ravine, India

5.5.1. REE ratios

Sixty samples were analyzed for Guryul Ravine with data sets for REE ratios along with

Ce/Ce* and Eu/Eu* anomalies, see Table 5. The Guryul Ravine stratigraphic section spans a depth of 69.5 m of the Changhsingian, overlain by 20.3 m of the Griesbachian. All lowermost

Triassic stratigraphic positions are cited relative to “zero” at the base of the LPME horizon, with the uppermost Permian depth ascending from -69.5 m, see Figure 32. La/Sm ratios range from a section low of 0.26 at stratigraphic depth of 6.2 m to a section high of 1.63 at a stratigraphic depth of 7.9 m. La/Sm values average 0.74 overall throughout the entire section. La/Yb ratios range from a section low of 0.38 at a stratigraphic depth of 6.2 m to a section high of 15.13 at a stratigraphic depth of 12.8 m. La/Yb values average 2.88 overall throughout the entire section.

Sm/Yb values range from a section low of 1.02 at a stratigraphic depth of 16.7 to a section high of 12.94 at a stratigraphic depth of 12.8 m. Sm/Yb values average 3.34 overall throughout the entire section, see Figure 32.

Figure 32. Guryul Ravine REE profiles. From left to right, sedimentary values of La/Sm, La/Yb, Sm/Yb, and the combined REE ratio plot, lined up against the lithostratigraphic column from Guryul Ravine. Figure taken from Brookfield et al. 2003.

5.5.2. Ce/Ce* and Eu/Eu* anomalies

Ce/Ce* values range from a section low of 0.67 at a stratigraphic depth of 10.4 m, to a section high of 1.14 at a stratigraphic depth of 7 m. Ce/Ce* values average 0.97 overall throughout the entire section. Eu/Eu* values range from a section low of 0.8 at a stratigraphic

depth of 10.1 m, to a section high of 2.37 at a stratigraphic depth of 11.6 m. Eu/Eu* values average 1.19 overall throughout the entire section, see Figure 33.

Figure 33. Guryul Ravine cerium and europium anomaly profiles. From left to right, sedimentary values of Ce/Ce* and Eu/Eu*, lined up against the lithostratigraphic column from Guryul Ravine. Figure taken from Brookfield et al. 2003.

5.6. Zal, Iran

5.6.1. TOC, and TIC values

Fifteen samples were analyzed for Zal with data sets for whole rock percent, percent

TOC and TIC, REE ratios, along with Ce/Ce* and Eu/Eu* anomalies, see Table 6. TOC values range from a section low of 0.07% at the stratigraphic site of IZ 3, two sample sites away from beginning the sequence and the PTB, to a section high of 0.4% at the stratigraphic site of IZ 8, 6 sample sites above the section low, see Figure 34. TOC values average at 0.11%, and the binary nature of the TIC values is attributed to the carbonate percentage of the entire section. TIC values range from a section low of 0% to a section high of 26.04% at the stratigraphic site of IZ

90, three quarters upsection from where the sequence starts. The section high is aberrant, as values consistently waver between percentages of 10 to 12%, with an average of 11.1%, see

Figure 34.

Figure 34. Zal total organic carbon and total inorganic carbon profiles. From left to right, sedimentary values of TOC and TIC, lined up against the lithostratigraphic column of Zal. Figure taken from Horacek et al. 2007.

5.6.2. REE ratios

La/Sm ratios range from a section low of 0.65 with sample IZ 132 at the top of the section, to a section high of 1.34 with sample IZ 3, located two sample sites above the base of the section. La/Sm ratios average 0.91 overall throughout the entire section. La/Yb ratios range from a section low of 0.84 with sample IZ 14, to a section high of 2.53 with sample IZ 32. La/Yb

ratios average 1.41 overall throughout the entire section. Sm/Yb ratios range from a section low of 1.06 with sample IZ 14, to a section high of 2.15 located at sample IZ 23. Sm/Yb ratios average 1.53 overall throughout the entire section, see Figure 35.

Figure 35. Zal REE profiles. From left to right, sedimentary values of La/Sm, La/Yb, and Sm/Yb with the combined REE ratio plot, lined up against the lithostratigraphic column of Zal. Figure taken from Horacek et al. 2007.

5.6.3. Ce/Ce* and Eu/Eu* anomalies

Ce/Ce* values range from a section low of 0.61 with IZ 3, two sample sites above the base of the section, to a section high of 1.01 located at the top of the section with sample IZ

132. Ce/Ce* values consistently ascend in value in relation to the section and values average

0.91 overall throughout the entire section. Eu/Eu* values range from a section low of 0.99 at the base of the boundary with the first sample of the section, IZ 2, to a section high of 1.25, 6 sample sites above with sample IZ 8. Eu/Eu* values average 1.09 overall throughout the entire section, see Figure 36.

Figure 36. Zal cerium and europium anomaly profiles. From left to right, sedimentary values of Ce/Ce* and Eu/Eu*, lined up against the lithostratigraphic column of Zal. Figure taken from Horacek et al. 2007.

5.7. Chaohu, China

5.7.1. Al, TOC, and TIC values

Forty-two samples were analyzed for Chaohu with data sets for whole rock percentage, percent Al, percent TOC and TIC, REE ratios, along with Ce/Ce* and Eu/Eu* anomalies, see

Table 7. The Chaohu stratigraphic section spans a depth 800 cm below the boundary, overlain by 785 cm of Griesbachian sediment. All lowermost Triassic stratigraphic positions are cited relative to “zero” at the base of the LPME horizon, with the uppermost Permian depth ascending from a depth of -800 cm. Al values range from a section low of 0.4% at a stratigraphic depth of -222.5 cm below the boundary, to a section high of 12.21% at a stratigraphic depth of

0, demarcating the boundary. Al values average 6.46% overall throughout the entire section.

TOC values range from a section low of 0.7% at a stratigraphic depth of 49 cm above the boundary, to a section high of 2.84%, 7.5 cm above the boundary. TOC values average 0.35% overall throughout the entire section. TIC values range from a multiple section lows of 0% to a section high of 11.01% at a stratigraphic depth of -422.5 below the boundary. TIC values average 3.19% overall throughout the entire section, see Figure 37.

Figure 37. Chaohu aluminum, total organic carbon, and total inorganic carbon profiles. From left to right, sedimentary values of Al, TOC, and TIC, lined up against the lithostratigraphic column of Chaohu. Figure taken from Shuangying et al. 2006

5.7.2. REE ratios

La/Sm samples range from a section low of 0.1 at the base of the section -800 cm below the boundary to a section high of 2.98, located 217.5 cm above the boundary. La/Sm values average 1.19 overall throughout the entire section. La/Yb samples range from a section low of

0.4 at the base of the section to a high of 3.33 at the boundary. La/Yb values average 1.04

overall throughout the entire section. Sm/Yb values range from a section low of 0.14 at a stratigraphic depth of 80 cm above the boundary, to a section high of 4.28 at a stratigraphic depth of 0, demarcating the boundary. Sm/Yb values average 1.10 overall throughout the entire section, see Figure 38.

Figure 38. Chaohu REE profiles. From left to right, sedimentary values for La/Sm, La/Yb, and Sm/Yb, lined up against the lithostratigraphic column of Chaohu. Figure taken from Shuangying et al. 2006.

5.7.3. Ce/Ce* and Eu/Eu* anomalies

Ce/Ce* values range from a section low of 0.59 at the very base of sequence at a stratigraphic depth 8 m below the boundary, to a section high of 2.03 at a stratigraphic depth of

8 – 18 m above the boundary. Values average 1.12 and the overall plot shows consistency overall with two pulses of 2.03 and 1.93 at respective stratigraphic depths of 8 – 18 m and 235

– 240 m above the boundary. Eu/Eu* values range from consecutive section lows of 0.76, 0.77, and 0.77 at stratigraphic depths of 120 – 160 m to a section high of 1.32 at a stratigraphic of 1.5 m below the boundary, with values averaging 0.99, see Figure 39.

Figure 39. Chaohu cerium and europium anomaly profiles. From left to right, sedimentary values of Ce/Ce* and Eu/Eu*, lined up against the lithostratigraphic column of Chaohu. Figure taken from Shuangying et al. 2006.

6. DISCUSSION

6.1. Significance of Eu/Eu* anomalies for understanding sediment provenance

The REE pattern of upper-crustal material has a marked depletion of Eu (Eu/Eu* =

0.64) due to the retention of Eu within plagioclase, however plagioclase is not stable below 40 km. As a result, lower crustal, upper mantle material is enriched in Eu (Eu/Eu* = 1.17), therefore, any Eu anomaly is lower crustal or mantle derived in origin (Taylor et al. 1981). The composition of upper crustal material was derived from the uniformity, and the variation of REE patterns as discussed in Section 2.4 was found to be close to granodiorite (Taylor et al. 1981).

Of the 7 stratigraphic sites sampled in this study, all yielded REE data and these 7 sites represent a global wide distribution of the environmental conditions at the Permian-Triassic boundary.

Gujo-Hachiman, an open ocean site close to the equator from the middle of the

Panthalassic Sea has a Eu/Eu* value of 1.04, a 63% increase from Eu crustal values of 0.64, see

Figure 22. Ubara, another equatorial Panthalassic Ocean site in close proximity to Gujo-

Hachiman, has a similar Eu/Eu* value of 1.09, a 70% increase from Eu crustal values, see Figure

25. Black Ridge West, from the northern margin of the Euramerican plate at 30o north paleolatitude has a Eu/Eu* values of 1.02, a 59% increase from crustal values, see Figure 28.

Guryul Ravine, from the northern margin of the Gondwana plate with a paleolatitude of 30o south has Eu/Eu* values of 1.19, an 86% increase from Eu crustal values, see Figure 33. Zal, at

30o south on a microcraton in the middle of the Tethys Ocean has a Eu/Eu* value of 1.09, another 70% increase from Eu crustal values, see Figure 36. Chaohu, also on a microcraton in the eastern Tethys Ocean region at 30o north, has a Eu/Eu* value of 0.99, a 55% increase from

Eu crustal values, see Figure 39. The 55% to 86% increase in Eu from the average crustal Eu values is significant for sediment provenance. The anomalies suggest that the synchronous

sediments within the 7 PTB and Lower Triassic study sites analyzed had a derivation from a lower crustal, upper mantle source, the patterns of which will be discussed next. However, in order to expand on this, a prelude, qualitative discussion regarding the averages and patterning of the 7 different stratigraphic sites used in this study is needed before.

The general pattering of the combined sections show average values close to 1.0 through most of their thickness if the Eu/Eu* anomalies from each section are subtracted. In the absence of the sporadic spikes, these values would tend to represent crustal values therefore simply displaying evidence of erosional factors and a mixing from a variety of source materials in the shale throughout the depositional history. Albeit, if there were ash inputs derived from the Siberian Traps, it would be expected they would be sporadic from the different temporal rates of the explosive events produced from the large igneous province. This indeed is evident in the data collected, as the signals show up in the depositional history as relatively thin layers, consistent with the sporadic nature of the events. Furthermore, the pattern of just a few samples recorded through the section with anomalously high Eu/Eu* values is consistent with the interpretation that the Siberian Traps were producing episodic events of explosive eruptions, and that the inputs of lower crustal derived volcanic material was recorded in the sections utilized in this study. The kinematics of which will be discussed in section 6.3., but first a discussion on the REE variation at the PTB is needed.

6.2. Global Patterns of REE variation at the Permian-Triassic Boundary

The Noril’sk region of the Siberian Traps extends over 3500 m in a continuous section and records the entirety of the eruption phase. The Noril’sk region is the most complete and studied section, therefore providing the greatest clarity on the distinct geochemical magma patterns. Two suites within the Noril’sk section, the Gudchikhinsky and Tuklonsky formations, containing both picrites and tholeiitic lavas, were sampled using drill cores and surface exposures for major and trace element data (Lightfoot et al. 1993). The Gudchikhinsky from the

Lower Sequence of the Noril’sk region is characterized by high TiO2 wt% and high Sm/Yb ratios, values that are more comparable to the deep asthenospheric mantle, while the Tuklonsky is characterized by magmas influenced by material from the continental lithosphere, material that was recycled within the crust containing moderate Sm/Yb ratios and low total TiO2 wt%

(Lightfoot et al. 1993). Gudchichinsky Sm/Yb ratios range from 2.3 – 3.1 with TiO2 values ranging from 1.2 – 2.3 wt% (Lightfoot et al. 1993). Tuklonsky Sm/Yb ratios range from 1.6 – 1.8 with TiO2 values ranging from 0.45 – 0.95 wt% (Lightfoot et al. 1993).

Given the unique geochemistry of the individual suites of the Siberian Traps, the global patterns of REE variation at the Permian-Triassic boundary can be correlated to the magma patterns. This creates a direct linkage between the eruptions of the Siberian Traps and the microenvironmental change associated with the outgassing and volcanic ash input to the atmosphere. The high Sm/Yb ratios values of the Gudchichinsky suite correlate with sites on the northern margin of Euramerica, the northern margin of Gondwana within the Tethys Ocean, and open ocean site in the middle of the Panthalassic Ocean. Black Ridge West from Euramerica has a Sm/Yb ratio of 3.45, see Figure 27. Guryul Ravine has a Sm/Yb ratio of 3.34, see Figure 32, while the open Panthalassic ocean site of Gujo-Hachiman has a Sm/Yb ratio of 2.72, see Figure

21. While the moderate Sm/Yb ratios of Tuklonsky also correlate with sites from Gondwana, microcratons within the Tehtys Ocean, and the open ocean of Panthalassic. Chaohu has a

Sm/Yb ratio of 1.10, see Figure 38. Zal, the section that was deposited upon the microcraton within Tethys Ocean, has a Sm/Yb ratio of 1.52 from the carbonate entirety of the site, see

Figures 35. Ubara the equatorial open Panthalassic Ocean site has a Sm/Yb ratio of 1.37, see

Figure 24.

The dominant volume of these magmas was derived from a mantle plume within the lower crust or upper mantle, and the control for the geochemical compositions was the reprocessing of magma within crustal reservoirs from periodic replenishing, tapping and continuous fractionation (Wooden et al. 1993). With lower crustal or mantle-derived material, the correlations within the Sm/Yb ratios from the corresponding study sections provide a narrative for microenvironmental change. However, the global patterns of REE variation at the

Permian-Triassic boundary range go beyond the correlations with only the Sm/Yb ratios, and include the overall rare earth element enrichment for each of the 7 sample sites presented in this study. The overall chemical signals give insight into and reaffirm sediment provenance. The

Panthalassic Ocean equatorial site of Ubara projects an enrichment of MREEs with an overall increasing slope towards HREEs, see Figure 40. The neighboring site of Gujo-Hachiman, 10 degrees north and east of Ubara, projects enrichment in MREEs, see Figure 40. However, the overall enrichment slope was shown to be the inverse of Ubara, with higher values of LREEs around 1.90 ppm progressing towards values around 0.60 ppm for HREEs, see Figure 40.

Within the Tethys Ocean, on the far eastern end of the oceanic cul-de-sac, Chaohu REE values are relatively stable with slightly higher HREE values, see Figure 40, with a pronounced

Ce anomaly, showing an increase of over 800%. Zal, to the far west of the Tethys Ocean, projects an overall MREE enrichment, see Figure 40. Guryul Ravine, to the southern end of the

Tethys Ocean, projects an overall MREE enrichment with a pronounced Eu anomaly, see Figure

40. Black Ridge West, on the northern end of the Euramerica plate, projects an overall enrichment of MREEs with a pronounced Eu anomaly as well, see Figure 40, while the overall global REE variation tends to indicate sediments enriched overall in MREE and HREEs from a derived mantle plume within the asthenosphere, see Figure 40.

Figure 40. Combined REE spidergrams from each study section. From left to right, top to bottom; Gujo-Hachiman, Ubara, Black Ridge West, Spiti, Guryul Ravine, Zal, and Chaohu, with comparative REE spidergram.

6.3. Significance of wind blown material emanating from the Siberian Traps

The size and extent of the Siberian Traps was much as 5 x 106 km3 (Saunders and

Reichow 2009) and the amount of material that was degassed was approximately ~6300 – 7800

Gt S, ~3400 – 8700 Gt Cl, and ~7100 – 13,600 Gt S (Black et al. 2011). Given the high latitude placement of the Siberian Traps, the boundary between the troposphere and stratosphere was lower than it is at the equator, thus providing the conduit for material to be transported into the atmosphere easier via a shorter connection. The question then becomes: could the trajectories of material dispersed from the Siberian Traps be forecasted? The significance of the wind-blown material emanating from the Siberian Traps is pivotal in characterizing the chemical fingerprints within this study’s sites. Volcanic ash transport and dispersion models can calculate the paths of ash clouds over continents and hemispheres (Webley and Mastin 2008). Volcanic ash transport and dispersion (VATD) models require input parameters in forecasting tephra distribution such as plume height, mass eruption rate, tephra-size distribution and the duration of the eruption (Mastin et al. 2009).

While VATD models for the Siberian Traps haven’t ever been tested or even attempted, inferences correlated with petrographic, geochemical and climate model analyses have been drawn in this study to start the conversation on predicting the trajectory of tephra distribution from the Siberian Traps. The entire Siberian Traps complex is composed of 20-30% total volume in volcaniclastic rocks (Ross et al. 2005), and the presence of tuffs, tuff breccias and volcaniclastic sequences running hundreds of meters thick near the base of the sequence within the Meymecha-Kotuy region, suggest explosive episodes (Naumov and Ankudimova

1995; Black et al. 2012) that likely induced the volatile load within the stratosphere.

Sedimentary clasts within the volcanic units are consistent with a phreatomagmatic origin and may represent a significant proportion of the explosive activity produced from the Siberian

Traps, and the origin was most likely from the interaction with the hydrologic setting (Black et al. 2012). Contact heating and metamorphism related to sill intrusion within the Tunguska sequence created explosive events of gas release from the invasion of magma upon the

Tunguska sediments (Svenson et al. 2009).

Plume height, the elevation at which most of the ash spreads laterally from the plume into the ash cloud, for the Siberian Traps has been suggested to exceed 10km in height from models on large fissure eruptions that are associated with flood basalt provinces (Stothers et al.

1986). Recent studies undertaken from Glaze et al. (2011) used modeling to show the relationship between vent geometry and eruption rate as the primary controllers on plume height. Mass eruption rate is calculated by total erupted mass divided by the eruption duration

(Langmann et al. 2011). The mass eruption for longer lasting flood basalt eruptions like the

Siberian Traps has been estimated to be on the order of 103 -104 m3/s (Bryan et al. 2010). The tephra size distribution for basaltic eruptions is less than 30 µm, a size capable of being carried for hundreds of miles before settling onto land or within the ocean (Rose and Durant 2009). The duration of the Siberian Traps magmatism, the time period over which a significant amount of ash was being continuously emitted into the atmosphere, was less than 2 Ma, dated using

40Ar/39Ar data (Reichow et al. 2008).

Given the size and extent, the VATM source parameters, the explosive episodes and the high latitude placement of the Siberian Traps, the discussion then shifts to the transport

method: the wind. Early studies conducted on wind-blown volcanic ash and their effects within forests, agricultural locations, and human health showed strong statistical correlations between ash weights and high wind speeds as the two greatest factors on the effects of atmospheric travel, ecosystem disruption, soil formation and overall human health (Fowler and Lopushinsky

1985). In accordance with the scale of the Siberian Traps, the wind would have shaped and shifted the entire plume trajectory, due to the high latitude placement of the Siberian Traps.

The plume trajectory would have likely entered the jet stream that spans the globe from 30o to

60o in latitude, centered at an altitude of 10 km, where suspended tephra could have traveled to meet maximum core wind speeds at up 130 m/s (Bursik et al. 2009). However, the jet stream varies with season, albeit, if the Permian monsoonal pattern circulated the tephra within the jet stream during the summer months when monsoonal patterns were reverse and wind directions traveled northeast from the equator (Loope et al. 2004), see Figure 2, then there could have been a tephra deposition worldwide within 50 days:

40,074,274.944 m / 10 m/s = 1113 hours (3)

The implications for this tephra distribution are significant for microenvironmental change during the Permian Triassic boundary interval and indicative of sediment provenance.

The overall Eu/Eu* percent increases from the global wide distribution of Permian-Triassic sediments, which suggests that the material was derived from a lower crustal or mantle derived source. While the necessary steps to disperse the volcanic materials over large areas were certainly in effect during the times of the eruptions, a variety of processes conjointly hindered

this. The high latitude placement of the Siberian Traps, the explosive nature in which to transport the material within the atmosphere, the wind patterns and tephra size distribution and the 50-86% percent increases of Eu within the sediments of the sites that comprise the global distribution during the End-Permian, collectively show that all of these processes likely culminated in the most severe biological catastrophe in the history of Earth.

The future directions of this study include further research within petrography to illuminate the clarity at the boundary better. In fact, there has been a limited amount of work that has been done on the claystone beds of the Permian-Triassic boundary. The beds have shown the presence of phenocrysts, which document large and pervasive volcanic eruptions.

Within the Permian-Triassic boundary sections of South China, has been shown to have abundant ash beds, and the beds below the PT boundary are of local origin, including the claystones of Bed 25. In which, there are large shifts in Eu anomalies creating doubt as to whether or not the ash layers are local or not. If the layers at Bed 25 and those stratigraphically higher are not local of origin, then they must represent wind-blown material from the Siberian

Traps eruptions. A comparative cross-examination of the changes in phenocrysts between the earlier ash beds and the beds above Bed 25 would give definite answers. The cross examination would have to be rigorous and include a systematic study of the volcanic phenocryst types, sizes and frequencies within the claystones located above Bed 25 ash layers. This would then be followed by characterization work with separating the phenocrysts by macerating the samples, filtering them out by washing out the residues, and then determining the chemical composition of the phenocrysts using scanning electron microscope. The advantage of this procedure is that the microprobe analysis can yield concentrations for all elements, not just the majors.

7. CONCLUSIONS

The noxious gas output, the volcanic ash fall and dust that the Siberian Traps produced were catastrophic and local communities were undoubtedly decimated. The larger question however being, was it enough to induce the Permian Triassic biodiversity calamity and if so, could evidence for a global signal be found? Data sets using REE geochemistry of marine sediments as tracers for diagnosing geochemical fingerprints of the microenvironmental change related to the Siberian Traps, which took place at the PTB, suggests reducing conditions in the ocean and extreme terrestrial fluctuations. The HREE enrichment patterns, and Eu/Eu* anomalies picked up at Ubara and Zal are indicative of material that was derived from the lower crust or upper mantle. This is contrasted by the high Sm/Yb ratios of Black Ridge West, Guryul

Ravine, and Gujo Hachiman which match the values of the Gudchichinksy suite of the Lower

Noril’sk region, values that are comparable to the asthenosphere. The size and extent of the

Siberian Traps, in addition to the high latitudinal placement, the Permian monsoonal climate and strong equatorial winds, along with VATM values of the Siberian Traps and the contemporary comparisons to Icelandic volcanoes, all amounts to data that seems to suggest that the erupted material from the Siberian Traps was transported into the atmosphere and subsequently windblown and deposited.

The End-Permian was a wasteland, environmental conditions were some of the most extreme conditions the world had ever seen and biodiversity was violently affected. Massive

carbon release in the atmosphere and oxygen depleted oceans created the largest biological crisis ever. Fast forward to present day, and we are facing some of the same problems, although it is not a large igneous province that we face but rather anthropogenic causes.

Human induced changes to the atmosphere and environment are operating on a level unparalleled by any ancient analog and while climatic disasters of the past may not provide the exact temporal clarity to forecast what the current climate landscape will look like in the near future, it does provide a sobering reminder of what the Earth could end up looking like if we don’t take the necessary preemptive steps to slow down the current rate of warming.

ACKNOWLEDGMENTS

Special and sincere thanks to Tom Algeo for advising, editing, and providing technical insight throughout the duration of this thesis. Warren Huff and Barry Maynard for advising and editing the continuous drafts. Thanks to Craig Dietsch, Krista Smilek and the entire Department of

Geology at the University of Cincinnati. Thanks to Robyn Hannigan, and Jeremy Williams from the Environmental Analytical Facility Department from the University of Massachusetts, Boston for allowing the use of all lab facilities.

REFERENCES

Albarede, F., 2009. Geochemistry: an introduction. Cambridge University Press, 2nd Edition, July

2009, 355 pages

Algeo, T.J., Hannigan, R., Rowe, H., Brookfield, M., Baud, A., Krystyn, L., and B. Ellwood, 2007.

Sequencing events across the Permian-Triassic boundary, Guryul Ravine (Kashmir, India).

Palaeogeography, Palaeoclimatology, Palaeoecology 252 (2007) 328-346

Algeo, T. J., Hinnov, L., Moser, J., Maynard, J., Elswick, E., Kuwahara, K., and H. Sano, 2010.

Changes in productivity and redov conditions in the Panthalassic Ocean during the Latest

Permian. Geology, February 2010.

Algeo, T.J., Kuwahara, K., Hiroyoshi, S., Bates, S., Lyons, T., Elswick, E., Hinnov, L., Ellwood, B.,

Moser, J., and B.J. Maynard, 2011. Spatial variation in sediment fluxes, redox conditions,

and productivity in the Permian-Triassic Panthalassic Ocean. Paleogeography,

Paleoclimatology, Paleoecology 308 (2011) 65-83

Algeo, T.J., Henderson, C.M., Ellwood, B., Rowe, H., Elswick, E., Bates, S., Lyons, T., Hower, J.C.,

Smith, C., Maynard, B.J., Hays, L.E., Summons, R.E., Fulton, J., and K.H. Freeman, 2012.

Evidence for a diachronous Late Permian marine crisis from the Canadian Artic region.

Geological Society of America Bulletin, 6 February 2012.

Alibo, D.S., and Y. Nozaki, 1999. Rare earth elements in seawater: particle association, shale

normalization and Ce oxidation. Geochemica et Cosmochemica Acta 63, 363-372

Ando, A., Kodama, K., and S. Kojima, 2001. Low-latitude and Southern Hemisphere origin of

Anisian (Triassic) bedded chert in the Inuyama area, Mino terrane, central Japan. Journal

of Geophysical Research 106 (B2), 1973–1986.

Arndt, N.T., Czamanske, G.K., Wooden, J.L., and V.A., Fedorenko, 1993. Mantle and crustal

contributions to continental flood volcanism. Tectonophysics 223, 39-52.

Arndt, N., Lehnert, K., and Y. Vasil’ev, 1995. Meimechites: highly magnesian lithosphere-

contaminated alkaline magmas from deep subcontinental mantle. Lithos 34, 41-59.

Arndt, N., Chauvel, C., Czamanske, G., and V. Fedorenko, 1998. Two mantle sources, two

plumbing systems: tholeiitic and alkaline magmatism of the Maymecha River basin,

Siberian flood volcanic province. Contrib. Mineral. Petrol. 133, 297-313.

Benton, M.J., Tverdokhlebov, V.P., and M.V. Surkov, 2004. Ecosystem remodeling among

vertebrates at the Permian-Triassic boundary in Russia. Nature 432:97-100.

Benton, M.J., 2005. When life nearly died: the greatest mass extinction of all time. London:

Thames and Hudson, ISBN 0-500-28573

Bhargava, O.N., Krystyn, L., Balini, M., Lein, R., and A. Nicora, 2004. Revised litho- and sequence

stratigraphy of the Spiti Triassic. Albertiana 30, International Meeting and Field

Workshop on Triassic Stratigraphy of the Himalayas, June 25 – July 6, 2004.

Black, B.A., Elkins-Tanton, L.T., Rowe, M. C., and I. U. Peate, 2012. Magnitude and consequences

of volatile release from the Siberian Traps. Earth and Planetary Science Letters 317-318

(2012) 363-373

Black, B.A., Hauri, E.H., Elkins-Tanton, L.T., and S. M. Brown, 2013. Sulfur isotopic evidence for

sources of volatiles in Siberian Traps magmas. Earth and Planetary Science Letters 394

(2013) 58-69

Black, B.A., Lamarque, J.F., Shields, C.A., Elkins-Tanton, L.T., and J.T. Kiehl, 2013. Acid rain and

ozone depletion from pulsed Siberian Traps magmatism. Geology 42(1), 67-70.

Black, B.A., Weiss, B.P., Elkins-Tanton, L.T., Veselovskiy, R.V., and A. Latyshev. 2015. Siberian

Traps volcaniclastic rocks and the role of magma-water interactions. GSA Bulletin;

September/October 2015; v. 127; no. 9/10; p. 1437-1452.

Blakey, R.C., Peterson, F., and G. Kocurek, 1988. Synthesis of late Paleozoic and Mesozoic eolian

deposits of the Western Interior. Sedimentary Geology 56: 3-126.

Bond, P.G., and P.B. Wignall, 2014. Large igneous provinces and mass extinctions: an update.

The Geological Society of America, Special Paper 505. 2014.

Bottjer, D.J., 2012. Life in the Early Triassic Ocean. Science, 338, 336 (2012). DOI:

10.1126/science.1228998

Brookfield, M., Twitchett, R.J., and C. Goodings, 2003. Paleoenvironments of the Permian-

Triassic transition sections in Kashmir, India. Paleogeography, Paleoclimatology,

Paleoecology 198 (2003) 353-371.

Bryan, S.E., Peate, I.U., Peate, D.W., Self, S., Mawby, M.R., Marsh, J.S., and J.A. Miller, 2010. The

largest volcanic eruptions on Earth. Earth Sci. Rev. 102, 207-229.

Burgess, S.D., Bowring, S., and Shen, S., 2014. High-Precision timeline for Earth’s most severe

extinction: Proceedings of the National Academy of Sciences of the United States of

America, v. 111, no. 9, p. 3316.

Burgess, S.D., and Bowring, S.A. 2015. High-precision geochronology confirms voluminous

magmatism before, during, and after Earth’s most severe extinction. Science Advances

1, e1500470 (2015).

Bursik, M.I., Kobs, S.E., Burns, A., Braitseva, O.A., Bazanova, L.I., Melekestev, I.V., Kurbatov, A.,

and D.C. Pieri, 2009. Volcanic plumes and wind: Jetstream interaction examples and

implications for air traffic. Journal of Volcanology and Geothermal Research. 186 (2009)

60-67

Chen, Z.Q., and M.J. Benton, 2012. The Timing and Pattern of Biotic Recovery following the End-

Permian Mass Extinction. Nature Geoscience, 2012; published online 27 May 2012.

Courtillot, V., 2002. Evolutionary catastrophes: the science of mass extinctions. Cambridge

University Press. Cambridge, United Kingdom, 2002.

Czamanske. G.K., Gurevitch, A., Fedorenko, V., and Simonov, O. 1998. Demise of the Siberian

plume: Paleogeographic and paleotectonic reconstruction from the pre-volcanic and

volcanic record, north-central Siberia. International Geology Review, v. 40, no. 2, p. 95-

115. de Baar, H.J.W., Bacon, P.G., Brewer, P.G., and K.W. Bruland, 1985. Rare earth elements in the

Pacific and Atlantic Oceans. Geochim. Cosmochim. Acta 49, 1943-1959

Dubiel, R.F., Parrish, J.T., Parrish, J.M., and S.C. Good, 1991. The Pangean megamonsoon:

evidence from the Upper Triassic Chinle Formation. Palaios 6: 347-370.

Dubiel, R.F., Parrish, J.T., Parrish, J.M., and S.C. Good, 1991. The Pangean megamonsoon:

evidence from the Upper Triassic Chinle Formation. Palaios 6: 347-370.

100

Elderfield, H., 1998. The oceanic chemistry of the rare earth elements. Philos. Trans. R. Soc.

Lond., A 325, 105-126

Elkins-Tanton, L.T., 2010. Mass extinctions, volcanism, impacts and global environmental

change (press conference). European Geosciences Union. 7 May, 2010.

Erwin, D.H., 2008. Extinction: How life on Earth nearly ended 250 million years ago. Princeton

University Press. April 21, 2008, 320 pgs.

Fedorenko, V., and G. Czamanske, 1997. Results of new field and geochemical studies of the

volcanic and intrusive rocks of the Maymecha-Kotuy area, Siberian flood-basalt province,

Russia. International Geology Review 39, 479-531.

Fedorenko, V., Czamanske, G., Zen’ko, T., Budahn, J., and D. Siems, 2000. Field and geochemical

studies of the melilite-bearing Arydzhangsky Suite, and an overall perspective on the

Siberian alkaline-ultramafic flood-volcanic rocks. International Geology Review 42, 769-

804.

Folwer, W.B., and W. Lopushinsky, 1985. Wind-blown volcanic ash in forest and agricultural

locations as related to meteorological conditions. Atmospheric Environment, Vol. 20.

No. 3, pp. 421-426, 1986.

Gang, G., Jinnan, T., Shihong, Z., Jie, Z., and B. Lingyan, 2008. Cyclostratigraphy of the Induan

(Early Triassic) in West Pingdingshan Section, Choahu, Anhui Province. Science in China

Series D: Earth Sciences. Jan 2008, vol. 51, no. 1, 22-29

101

Glaze, L.S., Baloga, S.M., and J. Wimert, 2011. Explosive volcanic eruptions from linear vents on

Earth, Venus, and Mars: comparisons with circular vent eruptions. J. Geophys. Res. 116,

E01011.

Hallam, A., and P.B. Wignall, 1997. Mass extinctions and their aftermath. Oxford University

Press, U.K., Sept 11, 1997, 328 pgs.

Hendry, E.R., 2010. What we know from the Icelandic volcano Eyjafjallajökull. Smithsonian

Magazine, 22 April 2010.

Hongfu, Y., Kexin, Z., Jinnan, T., Zunyi, Y., and W. Shunbao, 2001. The Global Stratotype Section

and Point (GSSP) of the Permian-Triassic Boundary. Episodes 24.2 (2001):102-114.

Horacek, M., Brandner, R., and R., Abart, 2007. 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. Paleogeography, Paleoclimatology, Paleoecology 252 (347-354)

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

changes in Lower Triassic oceanic circulation of the Tethys: The δ13 C record from marine

sections in Iran. Paleogeography, Paleoclimatology, Palaeoecology 252 (2007) 355-369

Ivanov, A.V., 2007. Evaluation of different models for the origin of the Siberian Traps. Geological

Society of America Special Paper, 430 (2007), pp. 669-692.

Jerram, D.A., Svensen, H.H., Planke, S., Polozov, A., and Torsvik, T.H. 2015. The onset of flood

volcanism in the north-western part of the Siberian Traps: Explosive volcanism versus

102

effusive lava flows. Palaeogeography, Palaeoclimatology, Palaeoecology 441 (2016) 38-

50.

Jin, Y.G., Wang, Y., Wang, W., Sheng, Q., Cao, C.Q., and D.H. Erwin, 2000. Pattern of marine

mass extinction across the Permian-Triassic boundary in South China. Science, 289, 432-

436.

Kakuwa, Y., 2008. Evaluation of palaeo-oxygenation of the ocean bottom across the Permian-

Triassic boundary. Global and Planetary Change 63 (2008) 40-65.

Kamo, S.L., Czamanske, G.K., Amelin, Y., Fedorenko, V.A., Davis, D.W., and V.R. Trofimov, 2003.

Rapid eruption of Siberian flood-volcanic rocks and evidence for coincidence with the

Permian-Triassic boundary and mass extinction at 251 Ma: Earth and Planetary Science

Letters 214, 75-91.

Kapoor, H.M., 1996. The Guryul ravine section, candidate of the global stratotype and point

(GSSP) of the Permian-Triassic boundary (PTB). In: Yin, H. (Ed.), The Paleozoic-Mesozoic

Boundary. Candidates of the Globral Stratotype Section and Point of the Permian-

Triassic Boundary. China University of Geosciences Press, Wuhan, pp. 99-110.

Kidder, D.L., and T.R. Worsley, 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, 2004, 203: 207-237.

Kiehl, J.T., and C.A. Shield, 2005. Climate simulation of the Latest Permian: Implications for mass

extinction. Geology, 2005, 33: 757-760.

103

Knoll, A.H., Bmabach, R.K., Payne, J.L., Pruss, S., and W.W. Fischer, 2007. Paleophysiology and

end-Permian mass extinction. Earth and Planetary Science Letters. Vol. 256, Issues 3-4,

30 April 2007, Pgs. 295-313.

Krystyn, L and M.J. Ochard, 1996. Lowermost Triassic Ammonoid and Conodont Biostratigraphy

of Spiti, India. Albertiana 17, May 1996.

Kuwahara, K., Nakae, S., and A. Yao, 1991. Late Permian “Toishi-type” siliceous mudstone in the

Mino-Tamba Belt. Journal of the Geological Society of Japan 97, 1005-1008 (in

Japanese).

Kuwahara, K., Yao, A., and S. Yamakita, 1998. Reexamination of Upper Permian radiolarian

biostratigraphy. “Earth Science” (Chikyu Kagaku) 52, 391-404.

Labandeira CC, and J.J. Sepkoski, 1993. "Insect diversity in the fossil record". Science 261 (5119):

310–315. Bibcode:1993Sci...261..310L. doi:10.1126/science.11536548. PMID 11536548.

Langmann, B., Folch, A., Hensch, M., and V. Matthias, 2011. Volcanic ash over Europe during the

eruption of Eyjafjallajokull on Iceland, April – May 2010. Atmospheric Environment 48

(2012) 1-8.

Lecuyer, C., Picard, S., Garcia, J.P., Sheppard, S.M.F., Grandjean, P. and G. Dromart, 2004.

Thermal evolution of Tethyan surface waters during the Middle-Late Jurassic.

Palaeoceanography, Palaeoecology, 193. 457-471.

Lightfoot, P.C., Naldrett, A.J., Gorbachev, N.S., Doherty, W., and V.A. Fedorenko, 1990.

Geochemistry of the Siberian Trap of the Noril’sk with implications for the relative

contributions of crust and mantle to flood basalt magmatism area, USSR. Contrib.

Mineral. Petrol. 104, 631-644.

104

Lightfoot, P.C., Hawkesworth, C.J., Hergt, J., Naldrett, A.J., Gorbachev, N.S., Fedorenko, V.A.,

and W. Doherty, 1993. Remobilisation of the continental lithosphere by a mantle plume:

major-, trace-element, and Sr-, Nd-, and Pb-isotope evidence from picritic and tholeiitic

lavas of the Noril’sk District, Siberian Trap, Russia. Contrib. Mineral. Petrol. 114, 171-

188.

Lind, E.N., Kropotov, S.V., Czamanske, G.K., Gromme, S.C., and V.A. Fedorenko, 1994.

Paleomagnetism of the Siberian Flood basalts of the Noril’sk area: a constraint on

eruption duration. Int. Geol. Rev. 36 (1994) 1139-1150.

Loope, D.B., Steiner, M.B., and C.M. Rowe, 2004. Tropical westerlies over Pangean sand seas.

Sedimentology 51: 315-322.

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

equatorial forest after the Permian Triassic ecological crisis. Proceedings of the National

Academy of Sciences of the United States of America 96, 13 857-13 862.

Mastin, L.G., Guffanti, M., Sevranckx, R., Webley, P., Barsotti, S., Dean, K., Durant, A., Ewert,

J.W., Neri, A., Rose, W.I., Schiender, D., Siebert, L., Stunderi, B., Swanson, G., Tupper, A.,

Volentik, A., and C.F. Waythomas, 2009. A multidisciplinary effort to assign realistic

source parameters to models of volcanic ash-cloud transport and dispersion during

eruptions. Journal of Volcanology and Geothermal Research 186 (2009) 10-21.

McLennan, S.M., 2001. Relationships between the trace element composition of sedimentary

rocks and upper continental crust. Geochem. Geophys. Geosyst., vol. 2., Paper number

2000GC000109. Published April 20, 2001.

105

Miller, Keith B., 2011. Geology of the Kansas Flint Hills: Ancient ice ages, sea levels and climate

change. Fieldtrip Guidebook , 2001 Meeting of the American Scientific Affliliation.

Mundil, R., Ludwig, K.R., and I. Metcalfe, 2004. Age and timing of the Permian mass extinctions:

U/Pb dating of closed-system zircons. Science, 2004, 305: 1760-1763.

Musashino, M., 1990. The Panthalassa – a cerium rich Atlantic type ocean: sedimentary

environments of the Tamba Group, southwest Japan. Tectonophysics, v. 181, p. 165-177

Nabbefeld, B., Grice, K., Schimmelmann, A., Sauer, P.E., Bottcher, M.E., and R. Twitchett, 2010.

13 34 Significance of dDkerogen, d Ckerogen and d Spyrite from several Permian/Triassic (P/Tr)

sections. Earth and Planetary Science Letters. 295 (2010) 21-29

Nabbefeld, B., Grice, K., Twitchett, T.J., Summons, R.E., Hays, L., Bottcher, M.E., and M. Asif,

2010. An integrated biomarker, isotopic and paleoenvironmental study through the Late

Permian event at Lusitaniadalen, Spitsbergen. Earth and Planetary Science Letters 291

(2010) 84-96

Nakazawa, K., Kapoor, H.M., Ishii, K., Bando, Y., Okimura, Y., and T. Tokuoka, 1975. The Upper

Permian and the Lower Triassic in Kashmir, India. Mem. Fac. Sci. Kyoto Univ. Ser. Geol.

Mineral. 41, 1-106.

Nakazawa, K., and H.M. Kapoor, 1981. The Upper Permian and Lower Triassic Faunas of

Kashmir. Palaeontol. Indica New Ser. 46, 1-191.

106

Naumov, V.A., and L.A. Ankudimova, 1995. Palynological complexes and age of volcanic

sediments of the Angara-Katanga District, Middle Angara Region. Geology and

Geophysics. 36.

Orchard, M.J., and L. Krystyn, 1998. Conodonts of the Lowermost Triassic of Spiti, and new

zonation based on Neogondolella successions. Rev. Ital. Paleontol. Stratigr. 104, 341-368

Parrish, J.T., and F. Peterson, 1988. Wind directions predicted from global circulation models

and wind directions determined from eolian sandstones of the western United States.

Sedimentary Geology 56: 261-282.

Pattison, J., and L. Stemmerik, 1996. Upper Permian foraminifera from East Greenland: Bulletin

Gronlands Geologiske Undersogelse, v. 48, p. 39-59

Payne, J.L., and M.E. Clapham, 2012. End Permian Mass Extinction in the Oceans: An analog for

the Twenty-First Century? Annual Review of Earth and Planetary Sciences. Vol. 40:89-

111

Peterson, F., 1988. Pennsylvanian to Jurassic eolian transportation systems in the western

United States. Sedimentary Geology 56: 207- 260.

Piasecki, P., 1984. Preliminary palynostratigraphy of the Permian-Lower Triassic sediments in

Jameson Land and Scoresby Land, East Greenland: Geological Society of Denmark

Bulletin, v. 32, p. 139-144.

Polozov, A.G., Svensen, H.H., Planke, S., Grishina, S.N., Fristad, K.E., and D.A. Jerram. 2015. The

basalt pipes of the Tunguska Basin (Siberia, Russia): High temperature processes and

107

volatile degassing into the end-Permian atmosphere. Palaeogeography,

Palaeoclimatology, Palaeoecology 441 (2016) 51-64.

Rampino, M.R., Stothers, R.B., and S. Self, 1985. Climatic impact of volcanic eruptions. Nature,

313, 272. 24, January 1985.

Rees, P.M., Gibbs, M.T., Ziegler, A.M., Kutzbach, J.E., and P.J. Behling, 1999. Permian climates:

evaluating model predictions using global paleobotanical data. Geology 27, 891-894.

Reichow, M.K., Saunders, A.D., White, R.V., Pringle, M.S., Al’Mukhamedov, A.I., Medvedev, A.I.,

and N.P. Kirda, 2002. 40Ar/39Ar dates from the West Siberian Basin: Siberian Flood Basalt

Province doubled. Science 7 2002: Vol. 296, no. 5574, pp. 1846-1849.

Reichow, M.K., Pringle, M.S., Al’Mukhamedov, A.I., Allen, M.B., Andreichev, V.L., Buslov, M.M.,

Davies, C.E., Fedoseev, G.S., Fitton, J.G., Inger, S., Medvedev, A.Y., Mitchell, C., Puchkov,

V.N., Safanova, I.Y., Scott, R.A., and A.D. Saunders, 2009. The timing and extent of the

eruption of the Siberian Traps large igneous province: Implications for the end-Permian

environmental crisis. Earth and Planetary Science Letters 277 (2009) 9-20.

Renne, P.R., Zhang, Z., Richards, M.A., Black, M.T., and A.R. Basu, 1995. Synchrony and causal

relations between Permian-Triassic boundary crises and Siberian flood volcanism.

Science 269, 1413-1416.

Robinson, P.L., 1973. Paleoclimatology and continental drift: in Tarling, D.H., and Runcorn, S.K.

eds Implications of Continental Drift to the Earth Sciences, I: Academic Press, London, p.

449-476.

Rose, W.I., and A.J. Durant, 2009. Fine ash content of explosive eruptions. Journal of Vocanology

and Geothermal Research. 186, 32-39

108

Ross, W.I., Peate, I.U., McClintock, M.K., Xu, Y.G., Skilling, I.P., White, J.D.L., and B.F. Houghton,

2005. Mafic volcaniclastic deposits in flood basalt provinces: a review. Journal of

Volcanology and Geothermal Research 145 (2005) 281-314.

Sabney S and M.J. Benton, 2008. "Recovery from the most profound mass extinction of all

time". Proceedings of the Royal Society B 275 (1636): 759–765

Sano, H., Kukuwahara, K., Yao, A., and S. Agematsu, 2010. Panthalassan seamount associated

Permian-Triassic boundary siliceous rocks, Mino terrane, central Japan. Paleontol Res 14,

295-316.

Sano, H., Wada, T., and H. Naraoka, 2012. Late Permian to Early Triassic environmental changes

in the Panthalassic Ocean: Record from the seamount-associated deep marine siliceous

rocks, central Japan. Palaeogeography, Palaeoclimatology, Palaeology 363-364 (2012) 1-

10.

Saunders, A., and M. Reichow, 2009. The Siberian Traps and the End-Permian mass extinction: a

critical review. Chinese Science Bulletin. Vol. 54. No. 1 20-37.

Schoepfer, S.D., Henderson, C.M., Garrison, G.H., and P.D. Ward, 2011. Cessation of a

productive coastal upwelling system in the Panthalassic Ocean at the Permian-Triassic

Boundary. Paleogeography, Paleoclimatology, Paleoecology. 313-314 (2012) 181-188.

Schoepfer, S.D., Henderson, C.M., Garrison, G.H., Ward, P.D., Foriel, J., Selby, D., Shen, Y.,

Hower, J.C., and T.J. Algeo, 2013. Termination of a continent-margin upwelling system at

the Permian-Triassic boundary (Opal Creek, Alberta, Canada).

109

Shen, S. Z. Crowley, J.L., Wang, Y., Bowring, S.A., Erwin, D.H., Sadler, P.M., Cao, C., Rothman,

D.H., Henderson, C.M., Ramezani, J., Zhang, H., Shen, Y., Wang, X., Wang, W., Mu, L., Li,

W., Tang, Y., Liu, X., Liu, L., Jiang, Y., and Y. Jin, 2011. Calibrating the End-Permian Mass

Extinction. Science 334, 1367 (2011); DOI: 10.1126/science. 1213454

Shen, H., 2013. China’s Jiangxi updates heavy rare earth reserve and resource estimate.

http://proedgewire.com/

Sholkovitz, E.R., Shaw, T.J., and D.L. Schneider, 1992. The geochemistry of rare earth elements in

the seasonally anoxic water column and porewaters of Chesapeake Bay. Geochemica et

Cosmochemica Acta, v. 56, p. 3389-3402

Saunders, A., and M. Reichow, 2009. The Siberian Traps and the End-Permian mass extinction: a

critical review. Chinese Science Bulletin. January 2009. Vol. 54. No. 1, 20-37

Sholkovitz, E.R., Landing, W.M., and B.L. Lewis, 1994. Ocean particle chemistry: the

fractionation of rare earth elements between suspended particles and seawater.

Geochemica et Cosmochemica Acta, v. 58, p. 1567-1579

Shibuya, H., and S. Sasajima, 1986. Paleomagnetism of red cherts: a case study in the Inuyama

area, central Japan. Geochemical Journal 16, 213–223.

Shields, G., and P. Stille, 1998. Stratigraphic trends in cerium anomaly in authigenic marine

carbonates and phosphates: diagenetic alteration or seawater signals? Goldschmidt

Conference, Toulouse, Mineralogical Magazine. V. 62A, p. 1387-1388

110

Shuangying, L., Jinna, T., Kongyan, L., Fanjian, W., and H. Yangyang, 2007. The Lower Triassic

cyclic deposition in Choahu, Anhui Province, China. Paleogeography, Paleoclimatology,

Paleoecology 252 (2007) 188-199.

Sinha, S.P., 1983. The Europium Anomaly from “Systematics and properties of the lanthanides.”

Scientific Affairs Division, North Atlantic Treaty Organization, pp. 550-553.

Stein, R., 1991, Accumulation of organic carbon in marine sediments: Berlin, Springer-Verlag,

217 p.

Stemmerik, L. C., F.G., Piasecki, S., Jordt, B., Marcussen, C., and H. Nohr-Hansen, 1992.

Depositional history and petroleum geology of the Carboniferous sediments into the

northern part of East Greenland, in Vorren, T.O., et al., eds., Artic geology and

petroleum potential: Amsterdam, Elsevier, p. 67-87

Sun, Y., Joahimski, M.M., Wignall, P.B., Yan, C., Chen, Y., Jiang, H., Wang, L., and X. Lai, 2012.

Lethally hot temperatures during the Early Triassic greenhouse. Science 338, 336 (2012)

DOI: 10.1126/science.1224126

Svenson, H., Planke, A.G., Schmidbauer, N., Corfu, F., Podladchikov, Y.Y., and B. Jamtveit, 2009.

Siberian gas venting and the end-Permian environmental crisis. Earth Planetary Science

Letters. 277, 490-500.

Taylor, S.R., McLennan, S.M., Armstrong, R.L., and J. Tarney, 1981. The Composition and

Evolution of the Continental Crust: Rare Earth Element Evidence from Sedimentary

Rocks. Phil. Trans. R. Soc. Lond. A 301, 381-399 (1981).

111

Taylor, S.R., and S.M. McLennan, 1985. The Continental Crust: Its composition and evolution, an

examination of the geochemical record preserved in sedimentary rocks. Blackwell,

Oxford, 312.

Taylor, E.L., Taylor, T.N., and N.R. Cuneo, 1992. The present is not the key to the past – A polar

forest from the Permian of Antarctica. Science, 1992, 257: 1675 – 1677.

Twitchett, R.J., Looy, C.V., Morante, R., Visscher, H., and P.B. Wignall, 2001. Rapid and

synchronous collapse of marine and terrestrial ecosystems during the end-Permian biotic

crisis. Geology. April 2001; v. 29; no. 4; p. 351-354

Ward, P.D., Botha, J., Buick, R., Kock, M.O., and D.H. Erwin, 2005. Abrupt and gradual extinction

among Late Permian land vertebrates in the Karoo Basin, South Africa. Science 307:709-

714.

Webley, P., and L. Mastin, 2008. Improved prediction and tracking of volcanic ash clouds.

Journal of Volcanology and Geothermal Research 186 (2009) 1- 9.

Wignall, P.B., and R.J. Twitchett, 1996, Oceanic anoxia and the end Permian mass extinction:

Science, v. 272, p. 1155–1158, doi: 10.1126/ science.272.5265.1155.

Wignall, P.B., and R. Newton, 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, 2003, V. 18, p. 153-167.

Wignall, P.B., 2007. The End-Permian mass extinction – how bad did it get? Geobiology, Vol. 5,

Issue 4, pgs. 303-309, December 2007

112

Willis, K.J., and J.C. McElwain, 2002. The evolution of plants. Oxford University Press Inc. New

York City, New York 2002.

Wooden, J.L., Czamanske, G.K., Fedorenko, V.A., Arndt, N.T., Chauvel, C., Bouse, R.M., King,

B.W., Knight, R.J., and D.F. Siems, 1993. Isotopic and trace-element constraints on

mantle and crustal contributions to Siberian continental flood basalts, Noril’sk area,

Siberia. Geochemistry and Cosmochemistry, Vol. 57, pp. 3677-3704, 1993.

Wyman, D.A., 1996. Trace element geochemistry of volcanic rocks. Geological Association of

Canada, Short Course Notes v. 12, 402 pp.

Yamakita, S., Kadota, N., Kato, T., Tada, R., Olighara, S., Tajika, E., and Y. Hamada, 1999.

Confirmation of the Permian/Triassic boundary in deep-sea sedimentary rocks; earliest

Triassic conodonts from black carbonaceous claystone of the Ubara section in the Tamba

Belt, Southwest Japan. Journal of the Geological Society of Japan 105, 895-898

Zhao, L., Tong, J., Sun, Z., and M.J. Orchard, 2008. A detailed Lower Triassic conodont

biostratigraphy and its implications for the GSSP candidate of the Induan-Olenekian

boundary in Chaohu, Anhui Province. Progress in Natural Science 18 (2008) 79-90.

Ziegler, A.M., 1990. Phytogeographic patterns and continental configurations during the

Permian period. IN: McKerrow, W.S. and Scotese, C.R. (Eds.) Palaeozoic,

Palaeogeography and Biogeography, pp. 363-379. London: Geological Society.

Zimanowski, B., Buttner, R., and V. Lorentz. 1997. Premixing of magma and water in MFCI

experiments. Bull. Volcanol. 58, 491-495.

113

Appendices

Table 1: Gujo-Hachiman sedimentary values, all data, 100 samples

Al Ti02 TOC La/Sm La/Yb Sm/Yb Ce/Ce* Eu/Eu*

ITJ-1 5.45 0.54 1.85 0.66 1.28 1.94 0.95 0.96 ITJ-2 6.64 0.64 1.08 ITJ-3 5.75 0.45 1.05 0.58 1.45 2.50 0.94 1.04 ITJ-4 6.59 0.63 1.30 0.67 1.25 1.88 0.94 1.41 ITJ-5 5.71 0.48 1.87 0.87 2.70 3.10 0.97 1.04 ITJ-6 5.58 0.49 0.57 1.17 3.87 3.32 0.98 0.86 ITJ-7 4.66 0.43 2.04 0.96 2.40 2.49 0.98 0.91 ITJ-8 5.69 0.53 0.78 1.09 2.93 2.70 0.96 0.86 ITJ-9 3.88 0.29 0.74 1.03 3.46 3.36 0.96 1.00 ITJ-10 4.80 0.44 1.43 1.12 3.77 3.38 0.95 0.84 ITJ-11 4.61 0.41 1.53 1.14 4.22 3.69 0.95 0.82 ITJ-12 4.50 0.40 1.32 1.09 3.46 3.18 0.97 0.86 ITJ-13 3.90 0.36 0.61 0.94 1.77 1.88 1.03 1.12 ITJ-14 1.33 0.13 0.21 0.65 0.78 1.20 1.07 1.06 ITJ-15 1.92 0.13 0.13 0.85 1.40 1.65 1.09 1.12 ITJ-16 2.41 0.25 0.21 0.66 0.95 1.44 0.92 1.06 ITJ-17 1.47 0.13 0.23 0.82 1.67 2.03 1.23 1.00 ITJ-18 2.37 0.19 0.21 0.96 2.58 2.68 1.23 0.96 ITJ-19 1.55 0.13 0.21 0.90 2.65 2.95 1.19 1.29 ITJ-20 0.93 0.13 0.28 1.00 3.38 3.38 1.25 0.94 ITJ-21 1.23 0.10 0.18 ITJ-22 1.71 0.14 0.14 0.82 2.12 2.57 1.39 0.98 ITJ-23 1.66 0.15 0.14 0.99 3.14 3.17 1.21 0.91

114

ITJ-24 0.96 0.14 0.22 0.11 0.20 1.71 0.79 1.19 ITJ-25 0.35 0.05 0.26 1.04 3.05 2.94 1.12 1.32 ITJ-26 1.85 0.23 0.19 0.98 3.26 3.32 1.20 0.96 ITJ-27 1.79 0.13 0.20 0.96 2.93 3.06 1.24 0.98 ITJ-28 0.35 0.03 0.31 0.92 1.98 2.15 1.12 1.08 ITJ-29 0.26 0.05 0.20 0.74 1.95 2.62 0.99 0.85 ITJ-30 1.07 0.12 0.21 0.78 2.18 2.81 1.21 1.00 ITJ-31 0.76 0.12 0.18

ITJ-32 0.26 0.04 0.24 ITJ-33 0.67 0.06 0.36 0.96 2.50 2.59 1.18 1.06 ITJ-34 0.80 0.11 0.24 ITJ-35 0.95 0.14 0.28 0.86 2.38 2.78 1.21 0.96 ITJ-36 2.06 0.16 0.24 ITJ-37 0.85 0.10 0.39 ITJ-38 0.57 0.06 0.24 0.88 2.24 2.54 1.16 1.16 ITJ-39 1.24 0.06 0.32 ITJ-40 0.41 0.10 0.30 0.91 2.68 2.96 1.16 1.02 ITJ-41 1.50 0.14 0.51 1.05 4.21 4.00 1.15 1.00 ITJ-42 1.42 0.09 0.21 0.99 3.92 3.97 1.14 0.94 ITJ-43 1.18 0.14 0.20

ITJ-44 1.34 0.09 0.28

ITJ-45 0.78 0.07 0.19

ITJ-46 0.36 0.06 0.18 ITJ-47 1.17 0.11 0.22 0.92 2.44 2.66 1.18 1.06 ITJ-48 0.76 0.12 0.16 ITJ-49 1.34 0.14 0.15 ITJ-50 2.26 0.19 0.16 0.91 2.61 2.87 1.19 0.94 ITJ-51 0.26 0.06 0.12 ITJ-53 0.24 0.12 0.15

ITJ-54 1.76 0.14 0.17

ITJ-55 1.50 0.10 0.14 ITJ-56 0.83 0.14 0.15 0.98 2.73 2.78 1.17 1.05 ITJ-57 0.91 0.12 0.15

ITJ-58 1.32 0.08 0.15 ITJ-59 1.80 0.16 0.12

115

ITJ-60 1.54 0.10 0.13

ITJ-61 1.63 0.15 0.16

ITJ-62 2.59 0.28 0.17

ITJ-63 0.34 0.10 0.20 ITJ-64 0.71 0.07 0.20 ITJ-65 0.93 0.12 0.15 0.90 2.38 2.64 1.14 1.61 ITJ-66 1.53 0.17 0.15 ITJ-67 1.11 0.13 0.17 ITJ-68 0.78 0.12 0.14

ITJ-69 0.90 0.13 0.14

ITJ-70 1.55 0.14 0.14 ITJ-71 1.21 0.14 0.15 0.82 2.00 2.44 1.22 1.08 ITJ-72 1.24 0.09 0.18

ITJ-73 1.96 0.13 0.11 ITJ-74 0.80 0.12 0.13 ITJ-75 0.34 0.09 0.18 ITJ-76 0.64 0.12 0.12 ITJ-77 0.85 0.11 0.14

ITJ-78 1.72 0.14 0.17

ITJ-79 1.12 0.11 0.15

ITJ-80 0.30 0.08 0.19

ITJ-81 2.87 0.38 0.20

ITJ-82 1.18 0.15 0.17 ITJ-83 0.86 0.15 0.18 1.00 3.40 3.39 1.17 1.26 ITJ-85 0.59 0.10 0.17 ITJ-86 0.33 0.11 0.15 ITJ-87 0.81 0.12 0.21

ITJ-88 0.95 0.13 0.19

ITJ-89 0.91 0.14 0.16 ITJ-90 0.41 0.09 0.19 1.03 3.33 3.24 1.15 1.27 ITJ-91 0.34 0.08 0.17

116

ITJ-92 0.23 0.04 0.18

ITJ-93 0.27 0.08 0.12

ITJ-94 0.67 0.09 0.21

ITJ-95 1.63 0.11 0.18 ITJ-96 1.16 0.10 0.15 ITJ-97 0.70 0.12 0.14 ITJ-98 1.27 0.14 0.18 ITJ-99 0.86 0.10 0.17 ITJ-100 0.58 0.09 0.21 0.97 2.44 2.50 1.06 1.17 ITJ-101 0.26 0.09 0.23

Table 2: Ubara sedimentary values, all data, 25 samples Al TOC TIC La/Sm La/Yb Sm/Yb Ce/Ce* Eu/Eu*

UB-25 5.226030904 0.51 0.1344209 0.21 0.15 0.69 1.12 1.10 UB-24 4.952484426 0.45 0 UB-23 4.147364287 0.76 0 UB-22 4.56664657 1.13 0.352098495 0.11 0.14 1.26 0.80 1.25 UB-21 4.399509873 0.72 0.030810692 UB-20 4.622393137 0.73 0

UB-19 4.802178518 0.10 0 UB-18 5.727249693 0.10 0 0.18 0.17 0.94 0.90 1.04 UB-17 1.936909406 0.19 0.083859489 0.35 0.37 1.06 0.97 1.09 UB-16 4.510631153 0.09 0 0.16 0.23 1.41 0.93 1.07 UB-15 3.718588779 0.10 0 0.28 0.41 1.46 0.93 1.13 UB-14 0.706264271 0.16 1.362994256 0.29 0.47 1.63 0.95 1.18 UB-13 3.021407425 0.09 0.186349601 0.22 0.32 1.50 0.94 1.22 UB-12 2.587657539 0.12 0 0.35 0.58 1.68 0.92 1.46 UB-11 2.690008529 0.10 0 0.66 0.58 0.87 0.62 2.34 UB-10 2.839830142 0.14 0 0.13 0.17 1.39 0.86 1.15 UB-09 1.974565893 0.11 0.001916401 0.77 0.86 1.13 0.95 0.96 UB-08 2.34202468 0.14 0 0.28 0.44 1.57 1.00 1.17 UB-07 1.108406199 0.09 0.007563379 0.20 0.31 1.56 0.95 1.18 UB-06 2.116922757 0.16 0 0.74 0.85 1.14 0.90 1.19 UB-05 1.687596667 0.10 0.031271305 0.32 0.74 2.34 1.06 1.14

117

UB-04 1.022367398 0.13 0.200684363 UB-03 0.670104938 0.11 0.628653907 0.70 0.81 1.16 0.96 0.99 UB-02 1.951659978 0.11 0 0.34 0.55 1.62 0.97 1.12 UB-01 2.0545266 0.32 0 0.33 0.57 1.73 0.95 1.11

Table 3: Black Ridge West sedimentary values, all data, 126 samples Al Ti02 TOC TIC La/Sm La/Yb Sm/Yb Ce/Ce* Eu/Eu*

BRW-P- 9.6 0.96 0.20 0.00 0.50 1.51 3.01 1.00 1.12 126 BRW-P- 9.7 0.95 0.20 0.00 125 BRW-P- 9.7 1.54 0.18 0.18 124 BRW-P- 9.4 0.86 0.24 0.30 123 BRW-P- 9.0 0.83 0.63 0.16 0.57 1.61 2.84 1.06 1.00 122 BRW-P- 9.5 0.94 0.14 0.24 121 BRW-P- 9.4 0.92 0.36 0.11 120 BRW-P- 9.5 0.81 0.20 0.27 119 BRW-P- 10.1 0.90 0.20 0.03 118 BRW-P- 10.7 0.95 0.23 0.00 0.67 2.78 4.17 1.06 1.03 117 BRW-P- 10.1 0.94 0.16 0.09 0.68 2.53 3.74 1.03 1.03 116 BRW-P- 9.0 0.85 0.39 0.27 115 BRW-P- 10.3 0.85 0.20 0.05 114 BRW-P- 10.6 0.94 0.20 0.00 113 BRW-P- 10.3 0.87 0.18 0.00 112 BRW-P- 9.6 0.84 0.10 0.14 111 BRW-P- 10.5 0.92 0.15 0.05 110 BRW-P- 9.9 0.86 0.08 0.48 109

118

BRW-P- 10.6 0.90 0.14 0.16 108 BRW-P- 9.6 0.86 0.15 0.24 107 BRW-P- 9.7 0.79 0.12 0.30 106 BRW-P- 10.0 0.90 0.17 0.09 105 BRW-P- 10.6 0.89 0.15 0.03 0.63 2.78 4.38 1.10 1.07 104 BRW-P- 9.2 0.82 0.37 0.04 103 BRW-P- 9.8 0.84 0.27 0.12 102 BRW-P- 9.6 0.84 0.13 0.29 101 BRW-P- 9.3 0.84 0.18 0.16 100 BRW-P- 9.6 0.88 0.19 0.22 99 BRW-P- 9.3 0.84 0.19 0.02 98 BRW-P- 10.0 0.86 0.15 0.10 97 BRW-P- 9.7 0.87 0.11 0.15 96 BRW-P- 9.5 0.87 0.14 0.11 95 BRW-P- 9.3 0.80 0.12 0.40 94 BRW-P- 9.6 0.85 0.18 0.01 93 BRW-P- 9.7 0.88 0.15 0.13 92 BRW-P- 10.1 0.91 0.14 0.02 91 BRW-P- 9.8 0.89 0.15 0.10 0.57 2.26 3.94 1.04 1.01 90 BRW-P- 9.7 0.85 0.12 0.16 89 BRW-P- 10.0 0.91 0.13 0.04 88 BRW-P- 10.3 0.92 0.12 0.03 87 BRW-P- 10.1 0.91 0.13 0.05 86

119

BRW-P- 10.5 0.91 0.14 0.03 85 BRW-P- 9.5 0.89 0.15 0.19 84 BRW-P- 9.2 0.82 0.11 0.38 61 BRW-P- 10.1 0.91 0.12 0.05 83 BRW-P- 10.5 0.87 0.13 0.01 60 BRW-P- 9.7 0.86 0.11 0.22 82 BRW-P- 9.4 0.83 0.15 0.13 59 BRW-P- 9.2 0.90 0.11 0.18 0.58 2.57 4.43 1.06 1.00 81 BRW-P- 9.5 0.87 0.12 0.25 58 BRW-P- 9.0 0.85 0.20 0.38 80 BRW-P- 9.0 0.79 0.12 0.61 0.53 1.60 3.00 1.03 1.01 57 BRW-P- 9.8 0.86 0.16 0.18 79 BRW-P- 9.7 0.83 0.19 0.24 0.66 2.91 4.39 1.09 1.05 56 BRW-P- 9.7 0.87 0.13 0.23 78 BRW-P- 9.5 0.83 0.14 0.34 0.56 2.33 4.16 1.04 1.03 55 BRW-P- 9.6 0.90 0.13 0.19 77 BRW-P- 10.1 0.83 0.12 0.31 54 BRW-P- 9.6 0.93 0.18 0.11 76 BRW-P- 9.4 0.90 0.13 0.28 53 BRW-P- 9.9 0.90 0.22 0.00 75 BRW-P- 9.8 0.88 0.18 0.17 52 BRW-P- 9.7 0.88 0.13 0.25 0.76 0.91 1.19 0.92 1.03 74 BRW-P- 9.8 0.85 0.14 0.39 51

120

BRW-P- 9.5 0.94 0.13 0.20 73 BRW-P- 9.9 0.86 0.17 0.19 50 BRW-P- 10.4 0.91 0.15 0.11 72 BRW-P- 10.9 0.90 0.15 0.07 0.69 2.96 4.29 1.04 1.03 49 BRW-P- 10.8 0.90 0.14 0.11 71 BRW-P- 10.2 0.91 0.14 0.10 48 BRW-P- 9.9 0.88 0.15 0.24 70 BRW-P- 10.0 0.88 0.16 0.16 47 BRW-P- 10.8 0.89 0.14 0.05 69 BRW-P- 10.3 0.92 0.16 0.00 46 BRW-P- 10.3 0.89 0.17 0.06 68 BRW-P- 10.3 0.92 0.19 0.00 45 BRW-P- 10.1 0.96 0.15 0.00 1.04 0.99 0.97 67 BRW-P- 10.3 0.93 0.17 0.00 44 BRW-P- 9.9 0.96 0.13 0.00 66 BRW-P- 10.0 0.91 0.12 0.01 43 BRW-P- 10.0 0.98 0.16 0.02 0.76 0.83 1.10 0.90 0.96 65 BRW-P- 9.9 0.95 0.15 0.04 42 BRW-P- 9.3 0.81 0.23 0.03 64 BRW-P- 9.5 0.82 0.25 0.02 41 BRW-P- 9.5 0.96 0.19 0.03 63 BRW-P- 10.1 0.92 0.18 0.00 0.76 2.89 3.79 1.25 0.99 40 BRW-P- 10.8 0.89 0.17 0.01 1.12 62

121

BRW-P- 10.5 0.86 0.33 0.00 39 BRW-P- 10.3 0.86 0.16 0.00 38 BRW-P- 10.6 0.86 0.17 0.00 37 BRW-P- 10.0 0.82 0.13 0.05 1.00 36 BRW-P- 10.0 0.86 0.19 0.02 35 BRW-P- 10.2 0.91 0.17 0.05 34 BRW-P- 9.3 1.01 0.34 0.02 33 BRW-P- 9.5 0.84 0.18 0.05 32 BRW-P- 9.5 0.79 0.15 0.30 1.03 31 BRW-P- 9.1 0.76 0.12 0.38 1.03 30 BRW-P- 9.2 0.78 0.22 0.51 29 BRW-P- 9.3 0.74 0.15 0.57 28 BRW-P- 8.3 0.73 0.15 0.78 27 BRW-P- 8.6 0.75 0.19 0.48 26 BRW-P-1 7.9 0.76 0.23 1.81 BRW-P-2 7.4 0.81 0.19 1.81 BRW-P-3 7.5 0.78 0.23 2.10 BRW-P-4 7.5 0.75 0.21 2.18 BRW-P-5 7.5 0.75 0.18 2.32 BRW-P-6 7.1 0.69 0.19 2.65 BRW-P-7 6.4 0.66 0.21 2.90

BRW-P-8 6.4 0.70 0.18 2.51 1.07 BRW-P-9 7.4 0.74 0.20 2.04 BRW-P- 7.6 0.75 0.21 2.10 10 BRW-P- 7.5 0.77 0.18 2.02 11

122

BRW-P- 7.8 0.76 0.27 1.99 12 BRW-P- 7.3 0.74 0.26 2.35 13 BRW-P- 5.7 0.60 0.38 4.27 14 BRW-P- 7.3 0.72 0.29 2.34 15 BRW-P- 7.2 0.77 0.29 2.00 16 BRW-P- 7.1 0.74 0.23 2.11 17 BRW-P- 7.4 0.75 0.41 2.19 18 BRW-P- 4.0 0.52 0.34 5.08 19 BRW-P- 6.6 0.65 0.31 2.68 20 BRW-P- 7.6 0.74 0.21 1.93 21 BRW-P- 7.1 0.73 0.22 2.11 1.01 22 BRW-P- 6.5 0.69 0.40 2.54 23 BRW-P- 7.1 0.69 0.32 2.30 24

Table 4: Spiti sedimentary values, all data, 135 samples Al TiO2 TOC TIC (%) (%) AL2-113 3.75 0.27 0.16 8.91 AL2-112 10.37 1.03 0.17 0.11 AL2-111 2.76 0.17 0.09 8.39 AL2-110 10.06 0.92 2.71 0.07 AL2-109 1.75 0.10 0.10 8.38 AL2-108 9.95 0.90 0.22 0.00 AL2-107 2.61 0.11 0.12 8.38 AL2-106 1.17 0.07 0.10 8.16 AL2-104 2.26 0.07 0.12 8.96 AL2-102 0.94 0.06 0.13 7.49 AL2-100 1.98 0.09 0.12 9.41 AL2-099 1.14 0.20 0.16 10.23 AL2-098 2.16 0.08 0.11 9.33

123

AL2-097 0.30 0.02 0.11 12.03 AL2-096 1.34 0.09 0.10 8.37 AL2-092 0.33 0.04 0.11 10.41 AL2-090 1.01 0.06 0.10 10.47 AL2-089 1.44 0.11 0.11 9.53 AL2-088 1.04 0.08 0.13 7.65 AL2-087 1.89 0.09 0.10 9.83 AL2-086 0.63 0.11 0.08 10.96 AL2-085 2.41 0.11 0.12 8.35 AL2-084 0.89 0.10 0.13 10.33 AL2-083 1.09 0.07 0.10 7.63 AL2-082 0.52 0.09 0.11 9.63 AL2-081 1.56 0.09 0.11 9.56 AL2-79 0.19 0.05 0.12 11.70 AL2-77 2.09 0.14 0.13 8.83 AL2-76 0.57 0.07 0.12 10.95 AL2-75 0.15 0.03 0.11 11.96 AL2-74 0.04 0.07 0.15 10.91 AL2-73 2.07 0.12 0.14 9.01 AL2-72 1.99 0.21 0.13 9.05 AL2-71 1.81 0.10 0.10 9.79 AL2-70 0.55 0.05 0.14 10.60 AL2-69 2.28 0.13 0.11 7.87 AL2-68 0.48 0.14 0.16 10.72 AL2-67 2.75 0.14 0.12 8.59 AL2-66 1.34 0.13 0.16 9.77 AL2-65 1.85 0.11 0.13 9.94 AL2-64 2.24 0.13 0.09 8.29 AL2-63 1.15 0.17 0.14 11.31 AL2-62 1.28 0.09 0.10 9.60 AL2-61 9.76 0.99 1.20 0.00 AL2-60 2.18 0.16 0.11 8.73 AL2-59 1.69 0.17 0.20 9.90 AL2-58 2.64 0.12 0.09 9.48 AL2-57 0.72 0.13 0.14 11.45 AL2-56 1.38 0.07 0.09 10.86 AL2-55 0.33 0.07 0.14 11.10 AL2-54 1.94 0.13 0.10 9.28 AL2-53 1.89 0.13 0.10 8.36 AL2-52 2.53 0.17 0.11 9.81 AL2-51 1.83 0.33 0.12 11.73 AL2-50 2.08 0.10 0.13 8.49

124

AL2-49 1.40 0.11 0.11 9.37 AL2-48 0.70 0.14 0.15 11.54 AL2-47 2.66 0.12 0.11 9.63 AL3-16C 3.46 0.24 0.11 8.02 AL3-16B 3.31 0.17 0.15 8.81 AL3-16A 2.80 0.15 0.13 9.08 AL3-15C 2.27 0.15 0.17 8.21 AL3-15B 2.55 0.14 0.11 9.38 AL3-15A 2.56 0.13 0.16 9.31 AL3-14C 2.55 0.14 0.13 9.19 AL3-14B 3.48 0.19 0.11 7.99 AL3-14A 3.77 0.18 0.14 8.12 AL3-13C 1.90 0.09 0.13 9.70 AL3-13B 1.61 0.09 0.11 9.62 AL3-13A 1.65 0.07 0.11 9.96 AL3-12C 2.79 0.05 0.18 10.50 AL3-12B 2.71 0.01 0.10 10.85 AL3-12A 2.51 0.03 0.13 11.16 AL2-46 10.72 1.08 0.42 0.21 AL2-45 9.03 0.16 0.27 5.01 AL2-44 1.09 0.13 0.13 8.92 AL2-43 10.37 1.09 0.52 0.00 AL2-42 9.17 0.90 1.77 0.16 AL2-41 2.23 0.06 0.57 0.42 AL2-40 10.16 1.14 1.80 0.00 AL2-39 9.12 0.88 1.52 0.07 AL2-38 9.21 0.96 0.26 0.00 AL2-37 0.17 0.17 0.23 9.51 AL2-36 0.95 0.07 0.57 1.17 AL2-35 9.40 0.99 0.29 0.00 AL2-34 9.36 0.94 1.84 0.00 AL2-33 8.85 0.90 0.51 0.00 AL2-32 3.30 0.13 0.40 7.55 AL2-31 8.34 0.77 2.44 0.10 AL2-30 9.58 1.00 2.06 0.00 AL2-29 9.56 0.93 2.33 0.00 AL2-28 9.67 1.01 1.88 0.10 AL2-27 2.11 0.17 0.12 9.94 AL2-26 8.27 0.74 0.92 0.41 AL2-25 0.39 0.09 0.26 8.74 AL2-24 9.39 0.94 0.59 0.06 AL2-23 9.93 1.07 2.03 0.15

125

AL2-22 9.59 0.94 1.11 0.01 AL2-21 0.69 0.09 0.13 9.73 AL2-20 9.36 0.92 0.86 0.00 AL2-19 1.00 0.12 0.22 7.08 AL2-18 8.64 0.79 0.87 0.00 AL2-17 8.42 0.74 1.69 0.33 AL2-16 1.04 0.09 0.10 9.33 AL2-15 0.98 0.17 0.35 6.47 AL2-14 4.53 0.12 0.12 9.91 AL2-13 7.88 0.76 0.88 1.40 AL2-12 0.42 0.10 0.15 8.11 AL2-11 3.75 0.08 0.14 11.08 AL2-10 11.35 1.24 0.30 0.00 AL2-9 0.07 0.15 0.15 11.76 AL2-8 1.74 0.02 0.13 11.14 AL2-7 0.33 0.16 0.23 8.90 AL2-6 0.39 0.20 0.13 9.10 AL2-5 6.88 0.64 0.50 0.00 AL1-16 2.18 0.00 0.12 8.19 AL1-15 10.76 1.02 0.49 0.00 AL2-4 10.11 1.07 0.42 0.00 AL2-3 11.13 1.10 0.49 0.00 AL1-14 7.87 0.74 0.66 0.00 AL1-13 10.71 1.07 0.37 0.00 AL2-2 10.87 1.13 0.96 0.00 AL1-12 10.03 0.97 1.37 0.19 AL2-1 10.28 1.00 1.71 0.51 AL1-11 10.27 0.95 1.31 0.00 AL1-10 9.82 0.97 0.74 0.00 AL1-09 9.90 0.96 0.56 0.00 AL1-08 9.57 0.95 1.92 0.00 AL1-07 9.16 0.84 1.65 0.00 AL1-06 9.77 0.92 1.98 0.00 AL1-05 9.46 0.91 1.49 0.00 AL1-04 6.36 0.76 0.86 0.00 AL1-03 9.34 0.94 0.51 0.00 AL1-02 10.33 0.99 2.13 0.01 AL1-01 9.79 0.91 1.90 0.04

Table 5: Guryul Ravine sedimentary values, all data, 84 samples La/Sm La/Yb Sm/Yb Ce/Ce* Eu/Eu*

126

-69.9 0.84 1.46 1.74 0.95 1.01 -64.9

-52.9

-48.4

-29.4

-22

-16.5 -10.2 0.79 0.85 1.09 0.95 1.07 -7.8

-0.1

0.1

0.4 0.7 0.78 1.21 1.56 0.93 1.09 1 1.3 0.75 0.83 1.10 0.94 1.03 1.6 0.69 1.47 2.12 0.97 1.12 1.9 2.2 0.58 1.18 2.03 0.90 1.19 2.5

2.8 3.1 0.64 1.06 1.66 1.10 0.98 3.4 3.7 0.91 4.87 5.37 0.96 1.01 4 1.00 4.26 4.27 0.86 1.22 4.3 0.92 4.45 4.82 0.97 1.02 4.6 0.77 4.47 5.84 1.01 1.17 4.9 0.67 2.45 3.64 0.91 1.31 5.2 0.74 3.92 5.29 0.94 1.08 5.5 0.77 3.53 4.61 0.95 1.09 5.8 0.83 4.77 5.76 0.92 1.12 6.1 0.95 4.20 4.41 0.95 1.05 6.2 0.26 0.38 1.47 0.83 1.20 6.4 0.90 7.19 8.02 0.97 1.05 6.5 0.78 2.08 2.66 0.94 1.25

127

6.7 0.86 2.22 2.58 0.94 1.27 6.9 0.48 0.90 1.87 1.02 1.23 7 0.69 1.84 2.64 1.14 0.95 7.3 0.89 3.58 4.03 0.99 0.94 7.6 0.56 1.13 2.01 1.07 2.20 7.8 7.9 1.63 4.99 3.06 0.93 1.07 8.1 0.49 0.84 1.74 1.04 1.11 8.4 0.49 0.96 1.97 1.08 1.23 8.7 0.63 1.09 1.73 1.00 1.15 9 0.69 1.56 2.26 1.03 0.94 9.3 0.73 0.87 1.18 0.93 1.08 9.5 0.54 0.80 1.47 0.97 1.25 9.6 1.21 9.46 7.79 0.98 0.85 9.8 0.70 1.28 1.81 1.00 2.16 10.1 0.91 3.61 3.95 0.88 0.80 10.4 0.61 1.18 1.92 0.67 1.25 10.7 0.61 1.41 2.32 0.97 1.06 11 0.75 3.04 4.04 1.06 0.96 11.3 0.52 1.04 1.99 0.97 1.27 11.6 0.95 4.18 4.41 0.96 2.37 11.9 0.81 5.06 6.25 0.96 1.07 12.2 0.61 1.57 2.58 0.98 1.09 12.5 12.8 1.17 15.13 12.94 1.02 1.02 13.1 13.4 0.91 9.47 10.43 0.98 1.01 13.7 0.66 1.46 2.22 1.01 1.33 14 0.97 5.90 6.10 1.03 0.96 14.3 0.86 6.25 7.27 1.02 1.02 14.6 0.79 2.22 2.80 0.98 1.00 14.9

15.2 15.5 0.68 1.45 2.13 1.03 1.05 15.8 0.61 0.67 1.10 0.99 1.46 16.1 0.62 0.65 1.05 1.03 1.29 16.4 0.50 0.90 1.78 1.08 1.18 16.7 0.60 0.61 1.02 0.78 1.34 17 0.67 0.72 1.06 0.92 1.40

128

17.3 0.60 0.82 1.37 0.90 1.47 17.6 1.33 10.70 8.06 1.00 1.02 17.9

18.2 18.5 0.74 0.92 1.25 1.00 1.75 18.8 0.82 0.86 1.05 0.95 1.15 19.1 0.95 4.52 4.76 1.02 0.93 19.4

19.7 20 0.61 0.63 1.03 0.95 1.60 20.3 0.83 1.85 2.22 1.01 1.06

Table 6: Zal sedimentary values, all data, 164 samples TiO2 TOC TIC La/Sm La/Yb Sm/Yb Ce/Ce* Eu/Eu*

IZ 132 0.10 0.11 9.66 0.65 0.95 1.47 1.01 1.23 IZ 131 IZ 130 0.08 12.73 IZ 129 0.00 IZ 128B 0.07 11.99 IZ 128A 0.03 0.10 11.74

IZ 128 0.10 12.04

IZ 127 0.16 0.10 11.90

IZ 126 0.02 11.11

IZ 125 0.10 11.26

IZ 124 0.01 0.12 12.63 IZ 123 0.00 IZ 122 0.10 10.81 IZ 121 0.00 IZ 120 0.00

IZ 119 11.10

IZ 118 0.11 11.33

129

IZ 117 0.14 0.00

IZ 116 0.00

IZ 115 0.00

IZ 114 0.03 0.14 11.38 IZ 113 0.09 12.30 0.67 1.13 1.69 0.99 1.05 IZ 112 0.09 11.94 IZ 11 0.00 IZ 110 0.03 0.09 11.77

IZ 109 0.03 0.09 11.41

IZ 108 0.00

IZ 107 0.06 0.09 11.00

IZ 106 0.00 IZ 105 0.05 0.11 11.44 IZ 104 0.00 IZ103 0.00 IZ102 0.11 12.00

IZ101 0.00

IZ 100 0.12 12.01

IZ 99 0.00

IZ 98.1 0.01 0.12 11.42

IZ 98 0.00 IZ 97 0.02 0.09 11.16 IZ 96A 0.10 11.85 IZ 96 0.00 IZ 95 0.00

IZ 94 0.00

IZ 93 0.01 0.10 12.35

IZ 92 0.00

IZ 91 0.00 IZ 90 0.03 0.15 26.04

130

IZ 86 0.12 11.21

IZ 85 0.00

IZ 84 0.07 11.17

IZ 83 0.11 11.20 IZ 83A 0.00 IZ 82 0.00 IZ 81 0.07 0.10 10.69 IZ 80 0.00

IZ 79 0.00

IZ 78 0.10 11.69

IZ 77 0.03 0.11 11.30

IZ 76 0.00 IZ 75 0.03 0.13 11.25 IZ 74 0.00 IZ 73 0.02 0.15 11.37 IZ 72 0.04 0.17 11.33

IZ 71 11.23

IZ 70B 0.03 0.12 10.92

IZ 70A 0.00

IZ 69 0.13 10.63

IZ 68 0.04 0.12 10.78 IZ 67 0.12 11.57 IZ 66 0.00 IZ 65 0.00 IZ 64 0.00

IZ 63 0.05 10.45

IZ 62 0.00 0.12 11.93

IZ 61 0.00 0.99 1.27 1.29 0.94 1.07

IZ 60 0.01 0.13 12.05 IZ 47 0.00

131

IZ 46B 0.00

IZ 46A 0.00 0.11 12.16

IZ 45 0.01 0.11 11.76

IZ 44 0.00 IZ 43 0.01 0.12 11.91 1.01 1.53 1.51 0.94 1.16 IZ 42 0.01 0.21 11.64 IZ 41 0.00 IZ 40 0.01 0.10 11.91 IZ 39 0.01 0.12 11.74

IZ 38 0.00 0.12 11.97

IZ 37 0.01 0.08 11.70

IZ 36 0.01 0.09 11.72

IZ 35 0.00 12.11 0.75 1.02 1.36 0.95 1.16 IZ 34 0.00 0.09 12.25 IZ 33 0.02 0.10 12.01 IZ 32 0.01 0.09 12.04 IZ 32A 0.01 0.08 11.74

IZ 31 0.00

IZ 30 0.02 0.10 11.49

IZ 29 0.01 0.08 11.48

IZ 28 0.01 0.10 11.71

IZ 27 0.01 0.09 12.06 IZ 26 0.02 0.09 11.56 IZ 25 0.01 0.17 11.83 IZ 24A 0.02 0.10 11.70 IZ 24 0.01 0.00 IZ 23 0.01 0.10 11.94 1.09 2.33 2.15 0.93 1.06 IZ 22 0.01 0.10 11.26 IZ 21 0.01 0.18 11.76 1.19 2.53 2.12 0.93 1.15 IZ 20 0.03 0.17 11.25 1.09 1.42 1.30 0.96 1.12 IZ 19 0.12 11.37

132

IZ 18 0.08 0.12 9.66

IZ 17 0.04 0.13 11.11

IZ 16 0.02 0.14 10.95

IZ 15 0.09 0.20 9.84 IZ 14 0.04 0.16 10.55 0.79 0.84 1.06 0.93 1.02 IZ 13 0.10 9.86 IZ 12 0.11 11.61 IZ 11 0.00 0.81 0.88 1.09 0.90 1.08 IZ 10 0.00 0.81 1.05 1.31 0.89 1.03

IZ 9 0.00 IZ 8 0.28 0.40 5.58 0.97 1.98 2.04 0.92 1.25 IZ 7 0.10 0.11 10.03

IZ 6 0.00 0.66 1.22 1.84 0.94 1.02

IZ 5 0.00 IZ 4 0.00 IZ 3 0.03 0.07 11.86 1.34 1.68 1.26 0.61 1.04 IZ 2 0.00 0.97 1.32 1.36 0.93 0.99 IZ 1 0.00 0.00

0.07 0.12 0.00

0.12 8.49

0.00

0.00 0.11 9.95 0.00 0.12 6.25 0.07 0.11 0.00

0.12 9.49

0.35 0.29 7.04

0.11 0.12 10.47

0.11 9.97

133

0.00

0.11 10.42

0.07 0.12 0.00

0.00 0.05 0.15 10.94 0.12 10.38 0.00 0.02 0.10 10.83

0.00

0.06 0.11 0.00 0.00 0.00 0.01 0.13 0.00 0.10 11.75

0.00

0.08 10.78

0.00

0.10 9.66

0.00 0.07 0.10 0.00 0.12 11.37 0.05 0.10 0.00 0.03 0.11 0.00

Table 7: Choahu sedimentary values, all data, 42 samples Al TiO2 TOC TIC La/Sm La/Yb Sm/Yb Ce/Ce* Eu/Eu*

755-760 1.54 0.36 0.09 8.46 1.26 0.79 0.63 1.20 1.21

134

700-705 8.06 0.88 0.10 2.37 2.77 1.71 0.62 1.26

600-605 6.85 0.63 0.09 4.36 0.69 1.34 1.95 0.98

500-505 5.54 0.59 0.08 4.29 0.65 1.31 2.03 1.00 430-435 6.89 0.59 0.07 4.36 1.90 1.03 0.54 1.26 1.00 365-370 5.86 0.61 0.10 4.36 1.32 0.69 0.52 1.27 1.15 350-355 6.63 0.64 0.09 4.22 0.96 1.10 1.15 0.93

335-340 4.90 0.42 0.07 5.60 0.89 1.64 1.83 0.93 315-320 5.87 0.59 0.08 4.87 0.72 1.26 1.76 1.01 280-285 6.33 0.59 0.08 3.96 1.63 0.83 0.51 1.33 0.87 265-270 9.33 0.88 0.09 0.02 1.87 1.00 0.53 1.20 1.13 255-260 5.12 0.59 0.08 5.61 0.18 0.07 0.41 0.77

235-240 7.66 0.64 0.08 2.73 0.28 1.93

215-220 8.62 1.05 0.08 0.12 2.98 2.14 0.72 1.24

200-205 4.85 0.54 0.08 6.43 0.62 1.15 1.85 1.04 185-190 6.42 0.52 0.12 4.33 1.40 0.72 0.51 1.32 1.19 165-170 4.54 0.46 0.08 6.00 0.66 1.26 1.91 1.02 155-160 4.38 0.41 0.09 6.13 1.95 1.07 0.55 1.27 0.77 135-140 4.51 0.48 0.09 6.10 1.81 0.67 0.37 1.28 0.77 120-125 9.49 0.79 0.13 0.00 1.10 1.78 1.62 1.00 0.76 105-110 4.04 0.39 0.10 6.15 0.47 0.50 1.08 1.01 95-100 5.37 0.66 0.12 4.15 1.78 0.88 0.50 1.21 1.08 80-85 9.10 0.88 0.08 0.03 0.93 2.87 3.10 0.93 70-75 6.38 0.53 0.09 4.12 1.81 0.26 0.15 1.58 1.09 50-55 9.03 0.85 0.13 0.05 1.87 0.26 0.14 1.60 1.06 33-38 4.57 0.46 0.10 5.51 0.59 0.75 1.26 1.05 18-25 8.94 0.80 0.07 5.82 2.28 1.27 0.56 1.18

8-18 8.99 0.68 0.24 0.07 1.84 1.00 0.54 2.03

5-8 8.94 0.06 0.13 0.00 1.76 1.06 0.61 0.96

0-5 12.04 0.32 0.14 0.08 0.32 0.17 0.54 0.84

0 to -5 7.98 0.47 0.32 0.01 0.67 1.71 2.57 1.01 -5 to -10 7.82 0.52 0.52 0.00 1.19 1.48 1.24 0.93 -10 to -15 7.35 0.70 1.33 0.00 0.63 0.87 1.39 0.88 -18 to -22 5.21 0.32 2.84 0.00 0.63 1.20 1.90 0.87

135

-22 to -25 7.17 0.42 0.10 0.00 1.70 1.75 1.03 1.45

-25 to -30 12.21 0.44 0.51 0.02 0.78 3.33 4.28 1.05 -45 to -50 4.79 0.25 0.24 0.00 0.41 0.62 1.49 0.97 0.84 -1.5 m 6.60 0.43 0.17 3.05 1.11 0.50 0.45 1.24 1.32 -2.5 m 0.40 0.41 1.91 6.34 0.57 0.76 1.33 0.91 0.94 -4.5 m 2.02 0.54 0.11 11.01 0.15 0.05 0.35 0.63 -7 m 4.45 0.21 1.88 0.08 2.15 0.64 0.30 1.24 0.78 -8 m 4.60 0.46 1.99 0.08 0.10 0.04 0.37 0.59

136