Contents lists available at SciVerse ScienceDirect
Palaeogeography, Palaeoclimatology, Palaeoecology
j ourna l homepage: www.e lsev ie r .com/ locate /pa laeo
Gradual onset of anoxia across the Permian–Triassic Boundary in Svalbard, Norway
Anna M. Dustira a,⁎, Paul B. Wignall b, Michael Joachimski c, Dierk Blomeier d,Christoph Hartkopf-Fröder e, David P.G. Bond b
a Department of Geology, University of Tromsø, Dramsveien 201, 9037 Tromsø, Norwayb School of Environment, University of Leeds, LS2 9JT Leeds, United Kingdomc GeoZentrum Nordbayern, Friedrich-Alexander University of Erlangen-Nürnberg, Schlossgarten 5, 91054 Erlangen, Germanyd Norwegian Polar Institute, Fram Centre, Hjalmar Johansens gt. 14, 9296 Tromsø, Norwaye Geological Survey North Rhine-Westphalia, De-Greiff-Str. 195, 47803 Krefeld, Germany
The Permian–Triassic extinction event is considered to be the most devastating environmental crisis of thePhanerozoic. With many coinciding factors involved, the role of marine anoxia during the extinction is stillpoorly understood. In this study a boreal Permian–Triassic Boundary (PTB) section from Svalbard, Norwayhas been investigated with the aim of better understanding the timing and nature of local marine anoxiaonset and extinction events across the PTB. The section comprises the Kapp Starostin and Vikinghøgdaformations; δ13Corg values indicate the PTB is located within the lower part of the Vikinghøgda Formation.Lag deposits at the top of the Kapp Starostin Formation indicate a marine hiatus and ensuing transgressionduring the Late Permian shortly before the extinction event, implying concurrence of major ecologicalchanges and sea-level rise. Pyrite framboid size distributions and total organic carbon (TOC) were used toevaluate bottom water oxygen conditions, and show that changes in bottom water redox conditions andextinction are clearly linked. Oxic to dysoxic bottom water conditions prevailed during deposition ofKapp Starostin Formation sediments and changed to anoxic to euxinic conditions above the formationboundary. The onset of anoxia is not abrupt but rather shows a gradual increase within the Kapp StarostinFormation during the Late Permian. The tipping point where bottom waters reach a long-term state of anoxiato euxinia coincides with the final extinction event, though changes in biotic assemblages at the top of theKapp Starostin Formation indicate a marine ecosystem crisis prior to this. Oxygen depletion in the boreal region,as seen in our study section and correlating to Greenland and Arctic Canada, seems to be consistently moresevere than in lower latitude settings.
Of the five great mass extinctions during the Phanerozoic, thePermian–Triassic Boundary (PTB) environmental crisis was the mostdevastating; approximately 90% of all marine species and 70% of ter-restrial vertebrate families were eliminated (Erwin, 2006). Despitedecades of research, a general consensus on the actual cause(s) ofthe extinction event remains elusive. A wide range of contributoryfactors has been suggested, such as (but by no means exclusively) hy-percapnia, ocean acidification, nutrient input fluxes or increased sed-imentation rates (Knoll et al., 1996; Payne et al., 2007; Algeo et al.,2010; Winguth and Winguth, 2012). Many researchers attribute thePermian–Triassic mass extinction events ultimately to greenhouseand toxic gas emissions from the Siberian trap eruptions, based onextensive and wide-ranging evidence linking volcanic activity to
rights reserved.
mass extinctions both on land and in the oceans (Vogt, 1972; Renneet al., 1995; Wignall, 2001; Beerling et al., 2007; Ganino and Arndt,2009; Algeo et al., 2011). Oceanic anoxia is proposed to be the keyfactor in the marine PTB extinction, as evidence of widespread anoxicto dysoxic marine conditions has been established coinciding withthis event (Wignall and Twitchett, 1996, 2002a; Bond and Wignall,2010). The intensity and duration of anoxia in marine sediments atthe PTB varies greatly on a global scale, making study and comparisonof boundary sections from a range of different palaeolatitudes and en-vironments important. While many of the lower-latitude sectionshave been well investigated in this respect, less information is avail-able from Boreal and high latitude settings, which we address here.
The aim of this study is to better understand the timing and natureof Permian–Triassicmarine anoxia onset in boundary sections in centralSpitsbergen (Fig. 1). Previous work in the region (e.g. Wignall et al.,1998; Nabbefeld et al., 2009; Bond and Wignall, 2010) has been ham-pered by a lack of clarity concerning the position of the PTB, whichcan now hopefully be resolved. Wignall et al. (1998) provided modestorganic carbon isotope data from the Festningen section in western
Fig. 1.Map showing the study area (arrow) at the contact between Permian and Triassic deposits on central Spitsbergen, Svalbard, Norway. Undifferentiated Carboniferous/Permianstrata are marked in green, and Triassic deposits in purple. White indicates ice- and snow-covered areas.Modified map courtesy of Norwegian Polar Institute.
Spitsbergen in an attempt to define the PTB, supplemented with a fewdatapoints from Tschermakfjellet in central Spitsbergen. Here the lattersection is targeted with more detailed and high-resolution sampling ofthe δ13Corg record that allows high-resolution correlation with otherPTB sections. The presence of anoxic to euxinic bottomwater conditionsis established using the size distribution (diameter) of framboidal pyriteas the main proxy, supported by total organic carbon (TOC) and totalsulphur (TS) data, and facies and palynological analyses.
Pyrite framboids occur in sedimentary deposits as raspberry-shaped collections of microcrysts, usually spherical or near-spherical,and are distinct from pyrite crystals and lumps formed during laterdiagenesis (Wilkin et al., 1996). Anaerobic sulphur-reducing bacteriaproduce H2S (hydrogen sulphide); this reacts with iron to form aniron sulphide precipitate, and through a chain of transitions, the ironsulphide alters to pyrrhotite, greigite and then pyrite in the form offramboids (Sweeney and Kaplan, 1973; Wilkin et al., 1996). Becausethis is a redox-dependent process involving weakly oxidizing steps(Wilkin et al., 1996), pyrite framboids only form at the redox boundarywhere oxygen-bearing and hydrogen sulphide-bearing waters come incontact. Typically, in oxygenated bottomwater settings, this is encoun-tered in the upper sediment column, but in euxinic settings the redoxboundary is within the water column. Framboids formed in the water
column have a limited diameter range, as once they have grown to acertain extent (diameters of 5.0±1.7 μm) they will sink out of the nar-row redox interface zone and into oxygen-free bottom water and sedi-ment, where further growth is not possible (Wilkin et al., 1996). Thesmallest mean diameters (3–5 μm) with very limited size range areindicative of euxinic conditions (sulphidic lower water column) (Bondand Wignall, 2010). Framboids forming within the sediment underoxic bottom water conditions attain greater diameters on average(7.7±4.1 μm) as they grow within the sediment where they reside atthe redox interface zone for a longer time (Wilkin et al., 1996).Framboid size distributions have been successfully applied as anoxiaand euxinia indicators in ancient marine sediments, including PTB sec-tions (Wignall et al., 2005; Shen et al., 2007; Bond and Wignall, 2010;Liao et al., 2010).
This paper presents pyrite framboid size distribution data coupledwith stable carbon isotopes (δ13C org), total organic carbon (TOC) andtotal sulphur (TS), facies analysis and palynological data from ahigh-resolution dataset from Svalbard that allows a detailed pictureof the timing and development of the marine depositional environ-ment during the PTB. In principle, this information should also enablethe correlation to findings from comparable depositional basins with-in the region, e.g. East Greenland (Wignall and Twitchett, 2002b;
Nielsen and Shen, 2004; Nielsen et al., 2010), and Arctic Canada(Grasby and Beauchamp, 2008, 2009; Algeo et al., 2012).
1.1. Geological setting
The study area encompasses Upper Permian to Lower Triassic ma-rine sediments from central Spitsbergen in Svalbard, Arctic Norway(Fig. 1). The exposure investigated near Tschermakfjellet on DicksonLand comprises a little over 5 m of the Upper Permian to Lower Triassictransition, and was previously investigated by Wignall et al. (1998).These transitional Upper Palaeozoic to Mesozoic sedimentary depositsfrom Spitsbergen reflect deposition on a broad epicontinental shelfat the northeastern margin of Pangaea that comprised Svalbard,parts of eastern North Greenland (Wandel Sea Basin), the Barents Sea(Finnmark Platform, Stappen High), Arctic Canada (Sverdrup Basin)and Russia (Timan–Pechora Basin) (Stemmerik and Worsley, 2005).The shelf was located at approximately 45° N in the Late Permian(Stemmerik and Worsley, 2005) and was characterised by pronouncedpalaeo-reliefwith both local basins and platforms developed. The inves-tigation area is located in a deeper shelf section of central Spitsbergenwhere the Upper Permian sediments of the Kapp Starostin Formationreach their thickest (>400 m thick), generally thinning towards theSouth (pinching out at Sørkapp–Hornsund High (Dallmann, 1999))and Northeast (Nordaustlandet).
1.2. Lithostratigraphy
Outcrops of Permian and Triassic successions are exposed at multi-ple localities on Spitsbergen. Upper Permian sediments on centralSpitsbergen are included in theKapp Starostin Formation (TempelfjordenGroup) (Fig. 2), which mainly consist of light and dark spiculitic chert,siliceous shale to siltstone and partly glauconitic sandstone withminor fossiliferous limestone (Dallmann, 1999). Lower Triassic sedi-ments comprise the Vikinghøgda Formation (Sassendalen Group),which are dominated by shale and siltstone (Dallmann, 1999).
Previous authors have defined the base of the Vikinghøgda Forma-tion as a bioturbated green sandstone bed (Mørk et al., 1999), howev-er glauconitic sandstones are more typical lithologies of the KappStarostin Formation. On the basis of Vikinghøgda and Kapp StarostinFormation descriptions from Dallmann (1999) and outcrop datafrom Tschermakfjellet, it can be contended to the contrary that theformational boundary is best placed above the glauconitic sand-stones, at the sharp lithological change from a thin phosphatic pebblebed to mudstone and siltstone at the base of the VikinghøgdaFormation. There has been debate as to whether the Kapp Starostin/
Tem
pelfjorden Gp
Sassendalen G
p
Vøringen Member
KapSvenskeegga member
Revtanna member Hovtinden member
Deltada
LusitaniadaleVendomdale
Tokrossøya Formation
Gp West SpitsbergenSouth Spitsbergen
Vardebukta Formation
Tvillingodden Formation
Bravaisberget Formation
Hiatus
Fig. 2. Basic lithostratigraphy of Spitsbergen from Per
Vikinghøgda Formation boundary is continuous or not: some regardit as unconformable (Steel and Worsley, 1984; Mørk et al., 1989;Stemmerik and Worsley, 1989; Stemmerik, 1997; Mørk et al., 1999),whereas others suggest at least some sections are conformable(Mangerud and Konieczny, 1993; Wignall et al., 1998). Likewise onthe Finnmark Platform, there is a lack of clarity whether the correlativeuppermost Permian deposits exhibit continuous deposition or whethera hiatus is present (Ehrenberg et al., 1998). Evidence at Tschermakfjellet(addressed further on in Sections 3.1. and 4.2.) points clearly to a hiatusduring the latest Permian, though not extending into the Triassic.
1.3. Biostratigraphy
Permian age determinations have been undertaken bymeans of bryo-zoans, brachiopods, corals, conodonts and palynomorphs (Szaniawskiand Malkowski, 1979; Biernat and Birkenmajer, 1981; Nakamura et al.,1987; Nakrem, 1988; Stemmerik, 1988; Nakrem, 1991; Nakrem et al.,1992; Mangerud and Konieczny, 1993; Buggisch et al., 2001; Chwieduk,2007). However, due to the paucity of good age-diagnostic taxa, preciseage constraints are generally poor for the Permian strata of Spitsbergen,and especially within the uppermost part of the Kapp Starostin Forma-tion, hence the uncertainty regarding the status of the PTB boundaryin the region (Wignall et al., 1998). One clue may come from thecommon occurrence of the fungal spore Reduviasporonites stoschianus(= Tympanicysta stoschiana; for taxonomy see Visscher et al. (2011))in the basal Vikinghøgda Formation (Mørk et al., 1999), which maycorrelate with a fungal spike in the latest Permian of many Tethyansections (Ouyang and Utting, 1990; Afonin et al., 2001; Visscheret al., 2011). It should be noted, however, that interpretation ofReduviasporonites as a fungus is not universally supported, as someauthors believe it to actually be of algal origins, in which case it wouldlose significance with respect to the mass extinction (Foster et al.,2002).
2. Materials and methods
2.1. Facies analysis
Microfacies studies were undertaken in order to investigate thelithology, facies, grain sizes and compositional variation of the sedi-ments from the section. A total of 10 thin sections were prepared atthe University of Tromsø, and analysed using plain- and polarisedlight.
Tatarian
Induan
Olenekian
AnisianLadinian
Kazanian
UfimianKungurian
Artinskian
Perm
ianT
riassicLow
erLow
erU
pper
Vøringen Member
p Starostin Formation
len member
n membern member
Stensiofjellet member
ChronostratigraphyCentral Spitsbergen NE Spitsbergen
Botneheia Formation
Vikinghøgda Formation
Mid.
mian to Middle Triassic, after Dallmann (1999).
50 µm
AB
Fig. 3. ScanningElectronMicroscope (SEM) image of pyrite framboids occurring in a clusterfrom the study section. Framboids (A) which reflect brightly and have indistinctmicrocrysts display secondary diagenetic growth of pyrite within the framboid, howeverthe original diameter is unchanged. Framboids which reflect more weakly (B) have beenaffected by weathering, though the original framboid diameter can still be discerned aswell.
The analysis of the carbon isotopic composition of organic carbonwascarried out at the Stable Isotope Laboratory, GeoZentrum Nordbayern,Friedrich-Alexander University of Erlangen-Nürnberg. Samples weredecarbonised using 10% HCl. Residues were washed intensively withde-ionised water, dried at 60 °C and homogenised using an agatemortar. Carbon isotope analyses of organic carbon were performedwith an elemental analyser (Carlo-Erba 1110) connected online toThermoFisher Delta Plus mass spectrometer. All carbon isotope valuesare reported in the conventional δ-notation in per mille relative toV-PDB (Vienna-PDB). Accuracy and reproducibility of the analyseswere checked by replicate analyses of international standards (USGS40). Reproducibility was ±0.05‰ (1σ).
2.3. TOC and TS
Total organic carbon (TOC) and total sulphur (TS) of bulk sampleswere analysed using a Leco CS-200 at the University of Tromsø.Weathered surfaces were removed from samples, which were thencrushed using a swing mill. For TS analysis, between 0.02 and 0.4 gof sample material was mixed with accelerator material (to improvecombustion) and combusted in a pure-oxygen environment. Thecarbon and sulphur were measured in the form of SO2, CO2 and COusing infrared cells. To measure TOC, samples weighing 0.2–0.8 gwere first reacted with 2 M HCl at 65 °C, in order to remove calciteand dolomite. The method and standard used for each sample werechosen to match the amount of C and S in each sample as closely aspossible. Reproducibility (1σ) was ±0.5% (or 2 ppm) for carbon and±1.5% (or 2 ppm) for sulphur.
Fig. 4. Lithological log showing formations, samples (used for framboid analysis), lithology,populations, including mean, standard deviation and standard error from each respectivecounted per sample) is b30, i.e. the sample only contains rare framboids, size distributionsfor dysoxic/oxic versus anoxic bottom water conditions as interpreted byWilkin et al. (1996between the two categories. Total organic carbon (TOC) %, and total sulphur (TS) % of bulk
2.4. Pyrite framboid analysis
Pyrite framboids across the PTB were identified and measuredfrom 38 polished blocks, taken from different locations along thetransition, and examined using a CamScan Series 4 Scanning ElectronMicroscope (SEM) set in backscatter mode at the University of Leeds.Framboids were measured to the nearest 0.5 μm. Where possible, atleast 100 framboids were recorded per sample; in samples withmore sparsely occurring framboids, this was not possible due totime- and sample constraints. Unweathered framboids appear muchsharper and brighter in the SEM images, but even weathered speci-mens, as well as those appearing to be diagenetically overprintedwith secondary pyrite overgrowth, were measured because the orig-inal shape and outline of the individual microcrysts were still visible(Fig. 3).
2.5. Palynology
Palynological samples were processed applying standard palyno-logical techniques, i.e. demineralisation with HCl–HF–HCl followingsieving of the residue with a 10 μm polyester fabric mesh (Ashrafand Hartkopf-Fröder, 1996; Batten, 1999). As the thermal maturityof the samples is low, no oxidation was necessary. Polyvinyl alcoholwas used as mounting medium and Elvacite 2044™ epoxy resin wasused as embedding medium to prepare permanent strew mounts.All slides are housed in the palynological collection of the GeologicalSurvey of North Rhine-Westphalia.
3. Results
3.1. Facies
A little over 5 m of sediments are exposed at the Tschermakfjelletsection (Fig. 4). Intensely bioturbated chert and glauconitic sandstonecomprise the lowermost 3.7 m of the section, at which point a phos-phatic lag deposit marks the top of the Kapp Starostin Formation;above this is a transition into Vikinghøgda Formation sedimentswhich are dominated by finely laminated, dark brown mud- to silt-stone (Fig. 4).
Kapp Starostin Formation strata consist of three facies types:spiculitic chert, glauconitic sandstone, and glauconitic fossiliferousmuddy sandstone (Fig. 5). Bioturbated, wavy medium- to thin-bedded, grey-coloured spiculitic chert is the dominant facies. This ismainly composed ofmonaxonmegascleric sponge spicules,megascleresdefined as having diameters greater than 3 μm after Scholle andUlmer-Scholle (2003), or with lengths greater than 63 μm afterGammon and James (2001). The abundant needles are commonlydensely packed, forming a component-supported fabric (packstone).In some areas the original spicules may be fully intact, though spiculeswill usually exhibit partial dissolution. No in situ sponges were foundwithin the facies. Additional components include common to abundant,rounded glauconite grains, phosphatic and/or poorly preserved frag-mented calcitic skeletal material, and rare to common subangular torounded quartz sand grains. During diagenesis, biogenic silica wasremobilised and original skeletal and matrix material was replaced todifferent extents by multigenerational chalcedony or microcrystallinequartz. Where silicification has not completely obliterated originalfeatures, spiculites typically have a clotty mud matrix. The random
colour and bedding features of the section studied at Tschermakfjellet. Pyrite framboidsample are shown to the right of the lithological log; where N (number of framboidsfor those samples are not depicted in the plot. Pyrite framboid mean diameter zones
) are coloured in red (anoxic) and blue (dysoxic/oxic). Note the large amount of overlapsamples are shown to the right.
orientation of sponge spicules, andmud-filled burrow cross-sections in-dicate this facies is intensely bioturbated.
Glauconitic sandstone facies occurs near the upper boundary of theKapp Starostin Formation; there is also a single thin-bedded horizonaround 1.6 m from base (Fig. 4). Abundant, rounded to well-rounded,very fine- to fine sand-size, glauconite grains characterise this faciestogether with common to abundant sub-angular to rounded, veryfine- to fine sand-size quartz grains. Generally rare phosphatic skeletalfragments (lingulids?) are more common towards the top of the KappStarostin Formation within this facies type; calcitic skeletal fragmentsare poorly preserved. The topmost glauconitic sandstone beds featurean extreme abundance of glauconite, in the form of both individualgrains as well as extensive cement.
Glauconitic fossiliferous muddy sandstone occurs at the top ofthe formation 3.7 m above base of section, in one 10 cm-thick bed(phosphatic lag, Fig. 4). This facies consists of subangular-subroundedquartz and glauconite grainswith phosphatic shell material (lingulids?)set in amudstonematrix. Above the sandstone, the Vikinghøgda Forma-tion is characterised by laminated mudstone and siltstone completelydevoid of fossils (except palynomorphs) and bioturbation.
3.2. Carbon isotopes (δ13Corg)
The carbon isotope record across the Permian–Triassic Boundary(Fig. 6) is characterised by a major negative excursion to −31.3‰interrupted by two short-term positive events (e.g. Algeo et al.,
2007, 2008; Korte and Kozur, 2010; Algeo et al., 2012). The firstminimum in δ13Corg is generally observed near the PTB which isbiostratigraphically defined by the first appearance datum (FAD)of the conodont species Hindeodus parvus. On Spitsbergen, theTschermakfjellet carbon isotope record reveals a negative shift of4‰ (this study) at the transition from the Kapp Starostin into theVikinghøgda Formation (Fig. 6). After the carbon isotope minimum,values increase towards the top of the studied interval by 1.5‰. Incomparison, carbon isotope ratios of organic carbon decrease by 8‰at the transition from the Kapp Starostin into the Vardebukta Forma-tion in the Festningen section (Wignall et al., 1998).
3.3. Total organic carbon (TOC) and total sulphur (TS)
TOC ranges up to 0.6% in the Kapp Starostin Formation, with aslight yet significant logarithmic positive trend (TOC (%)~log(Height(0–3.8 m)), β=4.29×10−3, adjusted R2=0.38, F(1,39)=25.1,pb0.001) towards the upper boundary of the formation. This occursfrom the base to about 3.5 m above base of section (Fig. 4). At thetransition from the Kapp Starostin Formation to the VikinghøgdaFormation, starting with sample 6-1, TOC values begin to rise suddenlyto over 1.0%, and mostly remain above 1.2% in the VikinghøgdaFormation from 3.8 m above base to the top of the section (Fig. 4).
TS values experience approximately 8 peaks from 0.6 m abovebase to the upper boundary of the Kapp Starostin Formation. TSvalues in Kapp Starostin Formation range up to 2.4%, but mostly
Perm
ian
Kap
p S
taro
stin
For
mat
ion
Vik
ingh
øgd
a Fo
rmat
ion
Tria
ssic
5
4
3
2
1
0(‰)-32 -30 -28
Carbon Isotopes (δ13Corg)
phosphatic lag
(m)
Fig. 6. Simplified lithological profile at Tschermakfjellet with corresponding resultsfrom δ13Corg analysis.
remain below 1.0%. Within the Vikinghøgda Formation, TS rangesfrom 0.4% to 2.1%. No significant trends in TS values are evident with-in either the Kapp Starostin or Vikinghøgda formations, in compari-son with e.g. Algeo et al. (2007) and Algeo et al. (2008) wheresulphur was found to show significant chemostratigraphic variation.
TS appears to be present in greater concentrations in the VikinghøgdaFormation, and excluding the anomalous peaks in the lower part ofthe section appears to trend to higher minimum values from 3.4 mabove base (Fig. 4). The general increase in TS concentrations fromthe top of the Kapp Starostin Formation through the VikinghøgdaFormation coincides with a decrease in δ13Corg values (Fig. 6).
3.4. Pyrite framboid analysis
Pyrite framboids were identified in all samples that were analysedunder SEM. The Kapp Starostin Formation contains mainly rare tooccasional or common framboids (Table 1), while the VikinghøgdaFormation samples contain predominantly abundant framboids(Fig. 4). Framboids were found to occur either individually or in clus-ters (Fig. 3), and commonly occurred with diagenetic pyrite crystals.Both mean framboid diameters as well as standard deviation aregreater in Kapp Starostin Formation samples (Fig. 7). Standard erroris lowest overall in the Vikinghøgda Formation, due to the tightersize distribution and larger sample sizes (>100 framboids countedper sample). Mean framboid diameters in Kapp Starostin Formationstrata range from 4.32 to 9.08 μm; those in Vikinghøgda Formationrange from 3.48 to 5.63 μm. Within the Vikinghøgda Formation,mean diameters generally remain on the lower end of the scale, al-though there are some abrupt negative and positive fluctuations(samples 8-1, and 8-5, Fig. 4).
3.5. Palynology
Most of the samples yielded a palynomorph assemblage dominat-ed by marine, short spined acanthomorph acritarchs of low diversity,and few bisaccate taeniate pollen grains. Trilete miospores are veryrare. The preservation of bisaccate pollen is poor; they are distortedor the sacci are disrupted from the central body. Therefore a detailedpalynological analysis is not attempted and instead we have con-centrated on an overall characterisation of the assemblage and theacritarch/pollen ratio.
Pollen grains and amorphous organic matter of the samples (exceptsample 6-1) from the Kapp Starostin Formation are orange-brown todark brown. Bisaccates constitute about 10–20% of the palynomorphs.The assemblages from the Vikinghøgda Formation are characterisedby extremely abundant acanthomorph acritarchs (up to 98%), fewyellow pollen grains and light coloured amorphous organic matter.Under incident blue light fluorescence acritarchs show strong yellowfluorescence indicating low thermal alteration. The samples from nearthe boundary between the two formations (samples 6-1 and 7-1)show a transitional character.
4. Discussion
4.1. Permian–Triassic Boundary
The PTB is tentatively placed near the first δ13Corg minimum, justabove the base of the Vikinghøgda Formation (Fig. 6). The increasein δ13Corg above this level may correspond to the first short-termpositive event of Korte and Kozur (2010), the base of interval F ofHermann et al. (2010) and interval D of Algeo et al. (2012). TheFestningen section (Wignall et al., 1998) as well as several othersections (Hermann et al., 2010; Korte and Kozur, 2010) displaytwo major negative shifts in δ13Corg around the PTB, while atTschermakfjellet only one is visible, located above the hiatal horizonat the top of the Kapp Starostin Formation. Considering correlationsof both δ13Corg and lithology of the Festningen and Tschermakfjelletsections, it is most likely that the depositional hiatus ensures thatthe first negative shift is absent from Tschermakfjellet section, thusit is the second major negative shift which is present in the rock re-cord (Fig. 6). The minor negative δ13Corg shift in the Tschermakfjellet
Table 1Results from SEM pyrite framboid analysis and accompanying statistics. Rare=b30 (framboids counted per sample); occasional=30–49; common 50–98; abundant 99 and above.
section, in comparison to the major shift at Festningen (Wignall et al.,1998), is attributed to thin development of the topmost KappStarostin Formation at Tschermakfjellet and the likelihood that partof the excursion is missing. The gradual transition to heavier δ13Corgvalues after the minimum at Tschermakfjellet is comparable to thatat Festningen (Wignall et al., 1998).
4.2. Facies and redox
In comparison with previous facies studies of the Kapp StarostinFormation, the spiculitic chert facies at Tschermakfjellet correspondsbest with the light-coloured spiculite facies of Ehrenberg et al. (2001),who have interpreted the depositional environment as below storm-base in an oxygenated setting. Glauconite and phosphatic horizonsare generally indicative of extremely low depositional rates in marinesettings; glauconite requires exposure at the sediment–water inter-face for durations up to 10,000 years in order to form (Odin andMatter, 1981). With deposition rates so low, the upper beds of theKapp Starostin Formation are considered to record a major slow-down in sedimentation within an agitated environment, insinuatingthat the depositional environment would not likely have beenbelow storm-base as suggested by Ehrenberg et al. (2001). Winnowingduring the depositional hiatus is likely responsible for concentrating thesand and removing finer grain sizes. The sharp decline of δ13Corg values(Fig. 6) starting in the upper Kapp Starostin Formation provides addi-tional evidence for the condensation horizon at the formation bound-ary. The sharp formational contact may then record a hiatal surface
capped by a phosphatic facies (the glauconitic fossiliferous muddysandstone at top of the Kapp Starostin Formation, Fig. 4) interpretedto be a basal transgressive lag, and implies a transgression just beforethe final marine ecosystem collapse. Evidence for amajor marine trans-gression appears to also be supported by land–plant biomarkers from anear-by boundary section on Spitsbergen (Nabbefeld et al., 2010) aswell as East Greenland (Fenton et al., 2007), although land–plant bio-marker enrichment within shallow marine sediments can alternativelybe interpreted as the product of terrestrial vegetation demise andsubsequent increased soil erosion (Sephton et al., 2005; Wang andVisscher, 2007). Above this horizon, fine lamination and lack of bio-turbation within the siltstone- to mudstones facies indicate depositionin a deeper, anoxic marine environment (distal shelf) below stormwave base.
The framboid diameter data broadly support the redox interpreta-tion based on facies analysis (Fig. 4). The lower parts of the study sec-tion contained few framboids suggesting well oxygenated conditionsin accord with the thoroughly bioturbated nature of the spiculites atthis level. In the lower 1.5 m of the section framboids are close tothe average mean diameter of 7.7 μm seen in modern oxic/dysoxicenvironments (Wilkin et al., 1996). Above this point the values grad-ually decrease overall, indicating a gradual transition from weaklydysoxic to lower dysoxic conditions (Bond and Wignall, 2010).Thus, there is a long-term trend of decreasing oxygenation towardsthe top of the formation. This is followed by the onset of anoxicand possibly euxinic conditions in the Vikinghøgda Formation thatare only alleviated by improved oxygenation (albeit to dysoxic
1.5 2.0 2.5 3.0 3.5 4.0
45
67
89
Standard deviation (µm)
Mea
n (µ
m)
8
48
110
140160
163
184
194204
224 239
266279
289299
309
316
363
376
386
394
406
416
421
422
431
441
456
471
491
Kapp Starostin FmVikinghøgda Fm
oxic-dysoxic
anoxic-euxinic
Fig. 7. Mean pyrite framboid diameters against standard deviation, differentiatingbetween the Kapp Starostin Formation (circle) and Vikinghøgda Formation (triangle)samples. Samples with fewer than 30 framboids per sample were excluded. The solidline across the plot after Wilkin et al. (1996) marks the divide between bottomwater conditions with at least some oxygen levels (dysoxic to oxic), and completelyun-oxygenated bottomwater conditions (anoxic). These differentiating environmentalclassifications are approximate, as the transition from one category to the next is gradualrather than abrupt.
conditions) within a cross-bedded siltstone unit (Fig. 4), as one possi-bility the result of a storm event where the agitation of the watercolumn might oxygenate the bottom waters.
Low TOC in the Kapp Starostin Formation is clearly linked to a lowpreservational potential of organic carbon in oxic/weakly dysoxicconditions. The subtle trend of increasing TOC towards the top ofthe Kapp Starostin Formation corresponds closely with declining ben-thic oxygen levels; higher organic carbon content in the overlyingVikinghøgda Formation coincides with the facies change into lami-nated anoxic/euxinic strata (Sun and Wakeham, 1994). This trend inTOC concentrations is consistent with other sections from northernand western Pangaean margins (Wignall and Newton, 2003; Bondand Wignall, 2010; Algeo et al., in press). The erratic TS peaks in theKapp Starostin Formation probably result mainly from variable con-centrations of diagenetic pyrite, because syngenetic pyrite framboidsare rare within the Kapp Starostin Formation. TS in the VikinghøgdaFormation can be attributed both to late diagenetic pyrite, commonlyseen in SEM, as well as common to abundant pyrite framboids.
The relationship between redox changes and extinction can bebroadly examined at Tschermakfjellet. The loss of sponge spiculesand bioturbation both occur at the formational boundary indicatingthe loss of marine benthos was at this level. This coincides with theshift to anoxic/euxinic deposition recorded by the pyrite framboiddata (and supported by TOC data) and thus clearly shows the link be-tween bottom water anoxia/euxinia and extinction at this site: aphenomenon seen at many boundary sites throughout the world(e.g. Wignall and Twitchett, 2002a). However there is evidence forsubtle redox changes prior to this main extinction event. TOC valuesincrease in the 2 m below the formation boundary and framboidpopulations suggest a shift to dysoxic bottom waters at the samelevel. Furthermore the dominance of phosphatic shell material inthe topmost sandstone, if derived from lingulids suggests benthiccommunities had already suffered substantial extinction losses beforethe onset of the main phase of anoxia/euxinia. Lingulids are commonpost-mass extinction disaster taxa that dominate benthic communi-ties for a brief interval after the end-Permian mass extinction
(Rodland and Bottjer, 2001). This scenario is identical to the oneseen from Boreal sections in NW Canada, where the initial changesin benthic communities also coincide with anoxic pulses beforethe long-term establishment of anoxia sees the loss of all benthos(Wignall and Newton, 2003). A similar scenario is seen in the WestBlind Fiord section of the Canadian Arctic where a similar facieschange, from the cherty Lindström Formation to the dark shales ofthe Blind Fiord Formation, occurs in the Permo-Triassic boundary in-terval. The uppermost 2 m of the Lindström Formation shows a tran-sition from oxygenated to “weakly euxinic” redox levels (Algeo et al.,2012, p. 1440) followed by improved ventilation in the basal metresof the Blind Fiord Formation and then onset of prolonged and intenseeuxinia. This redox history is identical to that seen at Kapp Starostin,where the basal 2 m of the Vikinghøgda Formation records a well-oxygenated interval before the onset of euxinia (Bond and Wignall,2010, Fig. 6). This interval may be missing at a short hiatus developedat the formational boundary at Tschermakfjellet. A similar situation isseen in the Buchanan Lake section of Arctic Canada (Grasby andBeauchamp, 2009).
Elsewhere in the world, anoxia is also developed during thePermo-Triassic transition although the details inevitably vary from lo-cation to location and particularly with water depth — deeper watersections invariably show better anoxic development (e.g. Wignalland Twitchett, 2002a; Algeo et al., 2008; Bond and Wignall, 2010;Liao et al., 2010; Zhou et al., 2012). Curiously, it is the higher souther-ly palaeolatitudes that show the weakest development of this anoxicevent. For example, on the Perigondwanan margin anoxia is welldeveloped in deeper shelf locations (Wignall et al., 2005) but onlytransient dysoxia is displayed in shallow shelf seas in the region(Wignall and Hallam, 1993; Wignall and Newton, 2003).
4.3. Palynology
The sediments of the Kapp Starostin Formation are characterised by ahigher amount of terrestrial-derived organic matter than those from theVikinghøgda Formation. However the palynomorph assemblage fromboth formations mainly consists of marine, short spined acanthomorphacritarchs. The low diversity especially in the Vikinghøgda Forma-tion may indicate a stressed environment (as supported by faciesand geochemical data), but the distribution of acritarchs in UpperPermian/Lower Triassic sediment and their ecological patterns hasnot been explored in detail. In general, Palaeozoic acritarch assemblagesof low diversity and high dominance and composed of sphaeromorphsand short-spined acanthomorph genera are typical of inshore-shallowwater facies (Tyson, 1995). However, in the deep water basinal faciesthe diversity is also reduced. The similar composition of inshore andbasinal facies assemblages and the limited knowledge of Late Permian/Early Triassic acritarch distribution in general indicate that caution isnecessary when using the Tschermakfjellet assemblage for proximal–distal facies analysis.
Acritarchs are regarded to be important primary producers andessential in the marine food web. The abundance of marine planktonin all samples, especially from the Vikinghøgda Formation, indicatesthat following the Late Permian–Triassic extinction event, the acritarchswere not severely affected by the crisis. Sections in eastern CentralSpitsbergen (Reymyrefjellet, Sveltihel, Stensiöfjellet) likewise show nodecrease in acritarch abundance at the Kapp Starostin Formation/Vikinghøgda Formation boundary (Mangerud and Konieczny, 1991).
5. Conclusions
Detailed study of the boundary section at Tschermakfjellet onSpitsbergen provides insight into the onset of anoxia/euxinia andtiming of other events at the PTB in northeastern shelf seas of Pangaea.On the basis of δ13Corg values it is suggested that the Permian–TriassicBoundary is located within the Vikinghøgda Formation in the
Tschermakfjellet section. Evidence for a marine transgression dur-ing the Late Permian shortly before the extinction event is present,implying that major ecological shifts coincided with sea level rise.Moreover, changes in bottom water redox conditions and extinc-tion are clearly linked, as the tipping point where bottom watersreach a long-term state of anoxia to euxinia coincides with thefinal extinction event, though changes in biotic assemblages atthe top of the Kapp Starostin Formation indicate a marine ecosystemcrisis prior to this. The onset of anoxia is not abrupt; this is evidentparticularly from TOC values, which show a gradual increase withinthe uppermost Kapp Starostin Formation. And although palynologicaldata suggest that acritarchs remained untouched by anoxia in theSvalbard region, benthic organisms were so severely affected by thelong-term environmental changes that an ecological crisis in the ben-thic realm is evident even before the onset of persistent anoxia. Evi-dence from Tschermakfjellet, Spitsbergen corresponds to that fromGreenland and Arctic Canada, indicating anoxic to euxinic conditionsand severe loss of biodiversity were widespread in the boreal realm.
Acknowledgments
Fieldwork for this study was supported by the Norwegian PolarInstitute. The authors would like to thank Eric Condliffe at the LeedsElectron Microscopy and Spectroscopy Centre, University of Leeds,andWerner Buggisch, Ralf Groen, and Tatjana Grunt who participatedin the fieldwork in Svalbard. We are additionally indebted to ThomasAlgeo, David Kidder and an anonymous reviewer for providing usefulcomments on an earlier version of this manuscript, and MikkoVihtakari for statistical assistance.
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.palaeo.2013.02.004.
References
Afonin, S.A., Barinova, S.S., Krassilov, V.A., 2001. A bloom of Tympanicysta Balme (greenalgae of zygnematalean affinities) at the Permian–Triassic boundary. Geodiversitas23, 481–487.
Algeo, T.J., Ellwood, B., Nguyen, T.K.T., Rowe, H., Maynard, J.B., 2007. The Permian–Triassic boundary at Nhi Tao, Vietnam: evidence for recurrent influx of sulfidic water-masses to a shallow-marine carbonate platform. Palaeogeography PalaeoclimatologyPalaeoecology 252, 304–327.
Algeo, T., Shen, Y.A., Zhang, T.G., Lyons, T., Bates, S., Rowe, H., Nguyen, T.K.T., 2008.Association of 34S-depleted pyrite layers with negative carbonate δ13C excursionsat the Permian–Triassic boundary: evidence for upwelling of sulfidic deep-oceanwater masses. Geochemistry, Geophysics, Geosystems 9.
Algeo, T.J., Hinnov, L., Moser, J., Maynard, J.B., Elswick, E., Kuwahara, K., Sano, H., 2010.Changes in productivity and redox conditions in the Panthalassic Ocean during thelatest Permian. Geology 38, 187–190.
Algeo, T.J., Chen, Z.Q., Fraiser, M.L., Twitchett, R.J., 2011. Terrestrial–marineteleconnections in the collapse and rebuilding of Early Triassic marine ecosystems.Palaeogeography Palaeoclimatology Palaeoecology 308, 1–11.
Algeo, T., Henderson, C.M., Ellwood, B., Rowe, H., Elswick, E., Bates, S., Lyons, T., Hower,J.C., Smith, C., Maynard, B., Hays, L.E., Summons, R.E., Fulton, J., Freeman, K.H., 2012.Evidence for a diachronous Late Permian marine crisis from the Canadian Arcticregion. Geological Society of America Bulletin 124, 1424–1448.
Algeo, T.J., Henderson, C.M., Tong, J., Feng, Q., Yin, H., Tyson, R.V., in press. Plankton and pro-ductivity during the Permian–Triassic boundary crisis: an analysis of organic carbonfluxes. Global and Planetary Change. http://dx.doi.org/10.1016/j.gloplacha.2012.02.008.
Ashraf, A.R., Hartkopf-Fröder, C., 1996. Die Siebverfahren bei der Aufbereitungpalynologischer Proben. Neues Jahrbuch für Geologie und Paläontologie,Abhandlungen 200, 221–235.
Batten, D.J., 1999. Small palynomorphs. In: Jones, P.T., Rowe, N.P. (Eds.), Fossil plantsand spores: modern techniques: Geological Society of London, pp. 15–19.
Beerling, D.J., Harfoot, M., Lomax, B., Pyle, J.A., 2007. The stability of the stratosphericozone layer during the end-Permian eruption of the Siberian Traps. PhilosophicalTransactions of the Royal Society A: Mathematical Physical and EngineeringSciences 365, 1843–1866.
Biernat, G., Birkenmajer, K., 1981. Permian brachiopods from the base of the KappStarostin Formation at Polakkfjellet, Spitsbergen. Studia Geologica Polonica 73, 7–24.
Bond, D.P.G., Wignall, P.B., 2010. Pyrite framboid study of marine Permian–Triassicboundary sections: a complex anoxic event and its relationship to contemporaneousmass extinction. Geological Society of America Bulletin 122, 1265–1279.
Buggisch, W., Joachimski, M., Lützner, H., Thiedig, F., Hüneke, H., 2001. Conodontstratigraphy of the Carboniferous and Permian strata from Brøggerhalvøya andthe Billefjorden Trough (Svalbard). Geologisches Jahrbuch B91, 637–689.
Chwieduk, E., 2007. Middle Permian rugose corals from the Kapp Starostin Formation,South Spitsbergen (Treskelen Peninsula). Acta Geologica Polonica 57, 281–304.
Dallmann, W.K. (Ed.), 1999. Lithostratigraphic Lexicon of Svalbard: Review and Recom-mendations for Nomenclature Use: Upper Palaeozoic to Quaternary Bedrock.Norsk polarinstitutt, Tromsø (318 pp.).
Ehrenberg, S.N., Pickard, N.A.H., Henriksen, L.B., Svana, T.A., Gutteridge, P., Macdonald,D., 2001. A depositional and sequence stratigraphic model for cold-water, spiculiticstrata based on the Kapp Starostin Formation (Permian) of Spitsbergen and equiv-alent deposits from the Barents Sea. AAPG Bulletin 85, 2061–2087.
Erwin, D.H., 2006. Extinction: How Life on Earth Nearly Ended 250 Million Years Ago.Princeton University Press, Princeton (296 pp.).
Fenton, S., Grice, K., Twitchett, R.J., Bottcher, M.E., Looy, C.V., Nabbefeld, B., 2007.Changes in biomarker abundances and sulfur isotopes of pyrite across the Permian–Triassic (P/Tr) Schuchert Dal section (East Greenland). Earth and Planetary ScienceLetters 262, 230–239.
Gammon, P.R., James, N.P., 2001. Palaeogeographical influence on Late Eocene biosiliceoussponge-rich sedimentation, southernWestern Australia. Sedimentology 48, 559–584.
Ganino, C., Arndt, N.T., 2009. Climate changes caused by degassing of sediments duringthe emplacement of large igneous provinces. Geology 37, 323–326.
Grasby, S.E., Beauchamp, B., 2008. Intrabasin variability of the carbon-isotope recordacross the Permian–Triassic transition, Sverdrup Basin, Arctic Canada. ChemicalGeology 253, 141–150.
Grasby, S.E., Beauchamp, B., 2009. Latest Permian to Early Triassic basin-to-shelf anoxiain the Sverdrup Basin, Arctic Canada. Chemical Geology 264, 232–246.
Hermann, E., Hochuli, P.A., Bucher, H., Vigran, J.O., Weissert, H., Bernasconi, S.M., 2010.A close-up view of the Permian–Triassic boundary based on expanded organiccarbon isotope records from Norway (Trondelag and Finnmark Platform). Globaland Planetary Change 74, 156–167.
Knoll, A.H., Bambach, R.K., Canfield, D.E., Grotzinger, J.P., 1996. Comparative earthhistory and Late Permian mass extinction. Science 273, 452–457.
Korte, C., Kozur, H.W., 2010. Carbon-isotope stratigraphy across the Permian–Triassicboundary: a review. Journal of Asian Earth Sciences 39, 215–235.
Liao, W., Wang, Y.B., Kershaw, S., Weng, Z.T., Yang, H., 2010. Shallow-marine dysoxiaacross the Permian–Triassic boundary: evidence from pyrite framboids in themicrobialite in South China. Sedimentary Geology 232, 77–83.
Mangerud, G., Konieczny, R.M., 1991. Palynological investigations of Permian rocksfrom Nordaustlandet, Svalbard. Polar Research 9, 155–167.
Mangerud, G., Konieczny, R.M., 1993. Palynology of the Permian succession of Spitsbergen,Svalbard. Polar Research 12, 65–93.
Mørk, A., Embry, A.F., Weitschat, W., 1989. Triassic transgressive– regressive cycles inthe Sverdrup Basin, Svalbard, and the Barents Shelf. In: Collinson, J.D. (Ed.),Correlation in Hydrocarbon Exploration. Graham and Trotman, for the NorwegianPetroleum Society, London, pp. 113–130.
Mørk, A., Elvebakk, G., Forsberg, A.W., Hounslow, M.W., Nakrem, H.A., Vigran, J.O.S.,Weitschat, W., 1999. The type section of the Vikinghøgda Formation: a newLower Triassic unit in central and eastern Svalbard. Polar Research 18, 51–82.
Nabbefeld, B., Grice, K., Twitchett, R., Summons, R., Hays, L., Böttcher, M., 2009.Biogeochemically cycling of C, S, H across the P/Tr boundary in Spitsbergen.Geochimica et Cosmochimica Acta 73, A925.
Nabbefeld, B., Grice, K., Twitchett, R.J., Summons, R.E., Hays, L., Böttcher, M.E., Asif, M.,2010. An integrated biomarker, isotopic and palaeoenvironmental study throughthe Late Permian event at Lusitaniadalen, Spitsbergen. Earth and Planetary ScienceLetters 291, 84–96.
Nakamura, K., Kimura, G., Winsnes, T., 1987. Brachiopod zonation and age of the PermianKapp Starostin Formation (Central Spitsbergen). Polar Research 5, 207–219.
Nakrem, H.A., 1988. Permian bryozoans from southern Spitsbergen and Bjørnøya. Areview of bryozoans described by J. Malecki (1968, 1977). Polar Research 6, 113–121.
Nakrem, H.A., 1991. Distribution of conodonts through the Permian succession ofSvalbard. Geonytt 1, 38–39.
Nakrem, H.A., Nilsson, I., Mangerud, G., 1992. Permian biostratigraphy of Svalbard(Arctic Norway)—a review. International Geology Review 34, 933–959.
Nielsen, J.K., Shen, Y., 2004. Evidence for sulfidic deep water during the Late Permian inthe East Greenland Basin. Geology 32, 1037–1040.
Nielsen, J.K., Shen, Y., Piasecki, S., Stemmerik, L., 2010. No abrupt change in redoxcondition caused the end-Permian marine ecosystem collapse in the East GreenlandBasin. Earth and Planetary Science Letters 291, 32–38.
Odin, G.S., Matter, A., 1981. Origin of glauconites. Sedimentology 28, 611–641.Ouyang, S., Utting, J., 1990. Palynology of upper Permian and lower Triassic rocks,
Meishan, Changxing County, Zhejiang Province, China. Review of Palaeobotanyand Palynology 66, 65–103.
Payne, J.L., Lehrmann, D.J., Follett, D., Seibel, M., Kump, L.R., Riccardi, A., Altiner, D.,Sano, H., Wei, J., 2007. Erosional truncation of uppermost Permian shallow-marinecarbonates and implications for Permian–Triassic boundary events. Geological Societyof America Bulletin 119, 771–784.
Renne, P.R., Black, M.T., Zichao, Z., Richards, M.A., Basu, A.R., 1995. Synchrony and causalrelations between Permian–Triassic Boundary crises and Siberian flood volcanism.Science 269, 1413–1416.
Rodland, D.L., Bottjer, D.J., 2001. Biotic recovery from the End-Permian mass extinction:behavior of the inarticulate brachiopod Lingula as a disaster taxon. Palaios 16, 95–101.
Scholle, P.A., Ulmer-Scholle, D.S., 2003. A color guide to the petrography of carbonaterocks. Grains, Textures, Porosity, Diagenesis, vol. 77. American Association ofPetroleum Geologists Memoir, Tulsa, OK (474 pp.).
Sephton, M.A., Looy, C.V., Brinkhuis, H., Wignall, P.B., de Leeuw, J.W., Visscher, H., 2005.Catastrophic soil erosion during the end-Permian biotic crisis. Geology 33, 941–944.
Shen, W.J., Lin, Y.T., Xu, L., Li, J.F., Wu, Y.S., Sung, Y.G., 2007. Pyrite framboids in thePermian–Triassic boundary section at Meishan, China: evidence for dysoxicdeposition. Palaeogeography Palaeoclimatology Palaeoecology 253, 323–331.
Steel, R.J., Worsley, D., 1984. Svalbard's post-Caledonian strata — an atlas ofsedimentational patterns and palaeogeographic evolution. In: Spencer, A.M.(Ed.), Petroleum Geology of the North European Margin. Graham & Trotman Ltd.,London, pp. 109–135.
Stemmerik, L., 1988. Discussion. Brachiopod zonation and age of the Permian KappStarostin Formation (Central Spitsbergen). Polar Research 6, 179–180.
Stemmerik, L., 1997. Permian (Artinskian–Kazanian) cool-water carbonates in northGreenland, Svalbard and the western Barents Sea. In: James, N.P., Clarke, J.A.D.(Eds.), Cool-water carbonates: SEPM Special Publication, pp. 349–364.
Stemmerik, L., Worsley, D., 1989. Late Palaeozoic sequence correlations, North Greenland,Svalbard and the Barents Shelf. In: Collinson, J.D. (Ed.), Correlation in HydrocarbonExploration. Graham and Trotman, for the Norwegian Petroleum Society, London,pp. 99–111.
Stemmerik, L., Worsley, D., 2005. 30 years on — Arctic Upper Palaeozoic stratigraphy,depositional evolution and hydrocarbon prospectivity. Norwegian Journal of Geology85, 151–168.
Sun, M.-Y., Wakeham, S.G., 1994. Molecular evidence for degradation and preservationof organic matter in the anoxic Black Sea Basin. Geochimica et Cosmochimica Acta58, 3395–3406.
Szaniawski, H., Malkowski, K., 1979. Conodonts from the Kapp Starostin Formation(Permian) of Spitsbergen. Acta Geologica Polonica 24, 231–264.
Tyson, R.V., 1995. Sedimentary Organic Matter: Organic Facies and Palynofacies.Chapman & Hall, London (615 pp.).
Visscher, H., Sephton, M.A., Looy, C.V., 2011. Fungal virulence at the time of the end-Permian biosphere crisis? Geology 39, 883–886.
Vogt, P.R., 1972. Evidence for global synchronism in mantle plume convection, andpossible significance for geology. Nature 240, 338–342.
Wang, C., Visscher, H., 2007. Abundance anomalies of aromatic biomarkers in thePermian–Triassic boundary section at Meishan, China — evidence of end-Permianterrestrial ecosystem collapse. Palaeogeography Palaeoclimatology Palaeoecology252, 291–303.
Wignall, P.B., 2001. Large igneous provinces and mass extinctions. Earth-Science Reviews53, 1–33.
Wignall, P.B., Hallam, A., 1993. Griesbachian (Earliest Triassic) palaeoenvironmentalchanges in the Salt Range, Pakistan and southeast China and their bearing on thePermo-Triassic mass extinction. Palaeogeography Palaeoclimatology Palaeoecology102, 215–237.
Wignall, P.B., Newton, R., 2003. Contrasting deep-water records from the Upper Permianand Lower Triassic of South Tibet and British Columbia: evidence for a diachronousmass extinction. Palaios 18, 153–167.
Wignall, P.B., Twitchett, R.J., 1996. Oceanic anoxia and the end Permian mass extinction.Science 272, 1155–1158.
Wignall, P.B., Twitchett, R.J., 2002a. Extent, duration, and nature of the Permian–Triassicsuperanoxic event. Geological Society of America Special Paper 356, 395–413.
Wignall, P.B., Twitchett, R.J., 2002b. Permian–Triassic sedimentology of Jameson Land,East Greenland: incised submarine channels in an anoxic basin. Journal of theGeological Society 159, 691–703.
Wignall, P.B., Morante, R., Newton, R., 1998. The Permo-Triassic transition in Spitsbergen:δ13Corg chemostratigraphy, Fe and S geochemistry, facies, fauna and trace fossils.Geological Magazine 135, 47–62.
Wignall, P.B., Newton, R., Brookfield, M.E., 2005. Pyrite framboid evidence for oxygen-poor deposition during the Permian–Triassic crisis in Kashmir. PalaeogeographyPalaeoclimatology Palaeoecology 216, 183–188.
Wilkin, R.T., Barnes, H.L., Brantley, S.L., 1996. The size distribution of framboidal pyrite inmodern sediments: an indicator of redox conditions. Geochimica et CosmochimicaActa 60, 3897–3912.
Winguth, C., Winguth, A.M.E., 2012. Simulating Permian–Triassic oceanic anoxiadistribution: implications for species extinction and recovery. Geology 40, 127–130.
Zhou, L., Wignall, P.B., Su, J., Feng, Q.-L., Xie, S.-C., Zhao, L.-S., Huang, J.H., 2012. U/Moratios and δ98/95Mo as local and global redox proxies during mass extinctionevents. Chemical Geology 324–325, 99–107.
