Triassic mass extinction: facies evidence from northern Italy and the western United States

Triassic mass extinction: facies evidence from northern Italy and the western United States

Palaeogeography. Palaeoclimatology, Palaeoecology, 93 (1992): 21-46 21 Elsevier Science Publishers B.V., ,Amsterdam Anoxia as a cause of the Permia...

5MB Sizes 0 Downloads 6 Views

Palaeogeography. Palaeoclimatology, Palaeoecology, 93 (1992): 21-46

21

Elsevier Science Publishers B.V., ,Amsterdam

Anoxia as a cause of the Permian/Triassic mass extinction: facies evidence from northern Italy and the western United States P. B. W i g n a l l a a n d A. H a l l a m b "Department of Earth Sciences, University of Leeds, Leeds, LS2 9JT, Great Britain bSchool of Earth Sciences, University of Birmingham, Birmingham, B15 2TT, Great Britain (Received June 10, 1991 ; revised and accepted November 5, 1991 )

ABSTRACT Wignall, P. B. and Hallam, A., 1992. Anoxia as a cause of the Permian/Triassic mass extinction: facies evidence from northern Italy and the western United States. Palaeogeogr., Palaeoclimatol., Palaeoecol., 93: 21-46. Facies, faunal and geochemicalevidencefrom the Permian/Triassic boundary sediments of the Dolomites and Idaho indicates a major anoxic event in the earliest Triassic. In both regions, the basal beds consist of finely laminated micrites with common syngenetic pyrite, The only fauna consists of occasional bedding plane assemblages of Lingula or Claraia, a typical lower dysaerobic assemblage. This is a level where previous studies have shown a major negative carbon isotope excursion and a cerium anomaly. In the Dolomites, the pyritic micrite directly overlies strata containing a diverse and typically Permian marine fauna of algae, foraminifers (including fusulinids) and articulate brachiopods, implying an abrupt extinction in contradiction to many previous views. Sequence stratigraphic analysis of the Dolomite boundary sediments reveals a minor sequence boundary in the late Permian followed by extremely rapid transgression leading to the development of the relatively deep water pyritic micrite - - a maximum flooding surface at the Permo-Triassic boundary. A further pulsed deepening in the lower Griesbachian, recorded in both the Dolomites and Idaho, lead to the widespread establishment of dysaerobic facies. It is clear that most of the extinctions occurred at the erathem boundary although the subsequent failure of the marine benthos to fill the empty ecospace in the ensuing Griesbachian may have been due to the widespread development of dysaerobic conditions.

Introduction The mass extinction event at the P e r m i a n / T r i a s sic b o u n d a r y was the most severe crisis in the history of life ( R a u p , 1979) b u t there is no consensus as to its cause. The late P e r m i a n was generally a time of low sea-level (Holser et al., 1986; Holser a n d Magaritz, 1987) a n d complete m a r i n e sections of this age are c o n s e q u e n t l y rare (Baud et al., 1989). As species diversity is related to the habitable area this has led to the suggestion that the diversity m i n i m u m for shallow m a r i n e taxa at least, is due to a global l o w s t a n d ( J o h n s o n , 1974; Schopf, 1974; Maxwell, 1989). However, J a b l o n s k i

Correspondence to: P. B. Wignall, Department of Earth Sciences, University of Leeds, Leeds, LS2 9JT, Great Britain. 0031-0182/92/$05.00

(1985) has d e m o n s t r a t e d that the retreat of the sea from c o n t i n e n t a l shelves is linked with an increase in h a b i t a t area a r o u n d oceanic islands. Such islands a p p e a r to provide a refuge for a diverse m a r i n e f a u n a at times of eustatic lowstand. J a b l o n s k i ' s (1985) data is derived from the Pleistocene record b u t the rapid glacioeustatic fluctuations of this time m a y n o t be directly c o m p a r a b l e in their effects to the longer term sea level variations elsewhere in the Phanerozoic (Hallam, 1989). Regressions can also p r o d u c e a p p a r e n t extinctions by reducing the v o l u m e of m a r i n e strata in which taxa can be preserved a n d collected - - this back smearing effect (cf. Signor a n d Lipps, 1982) m a y have been substantial in the Late P e r m i a n due to the paucity of strata of this age. Stanley (1984, 1988) has suggested that the

~ 1992 - - Elsevier Science Publishers B.V.

22

extinctions were due to global cooling. However, this is inconsistent with palaeoclimatic evidence which indicates that major glaciation, particularly of Gondwana, ceased in the mid-Permian long before the mass extinction (Caputo and Crowell, 1985). In addition, Fliigel and Reinhardt (1989) have demonstrated an increase in diversity of reefbuilding organisms up to the latest Permian - - a pattern that is unlikely to occur during global cooling when reef habitat area declines. Fischer (1964), and several subsequent authors, have suggested that lowered salinity in the world's oceans led to marine mass extinctions, presumably of stenohaline taxa. Precipitation of large amounts of evaporites in the mid-late Permian is held responsible for the salinity reduction. However, Benson (1984) considers that salinity reduction was insufficient to account for such a major mass extinction and Erwin (1990) has pointed out that the interval of evaporite formation considerably predates the extinction. Anoxic events and their associated black shales have been implicated in a number of mass extinction events (e.g. Hallam, 1989) but the Permian/ Triassic mass extinction has not in the past been attributed to such a cause. Indeed, several studies have suggested that there was an increase in ventilation in the basal Triassic seas and oceans (Brandner, 1988; Bottjer et al., 1988; Holser et al., 1991). However, Hallam (1991) recently proposed on the basis of facies and geochemical evidence that anoxia in the earliest Triassic was at least a major contributory factor to the extinction event. This idea needs to be tested by thorough examination of key sections. The diversity decline in the latest Permian is generally considered to have been gradual (e.g. Nakazawa and Runnegar, 1973; Teichert, 1990) reaching a minimum in the topmost Permian Dorashamian stage and the basal Triassic Griesbachian stage (Newell, 1988). Such a protracted crisis is implicit in both the sea-level fall and global cooling mechanisms. The extinction event thus contrasts with many other mass extinctions which appear to have been much more abrupt, particularly those associated with anoxic events (e.g. Elder, 1987; Hallam, 1987). However, few detailed bed-by-bed palaeontological studies have been undertaken

P. B. WIGNALL AND A. HALLAM

across the boundary and those which have demonstrate an apparent increase in diversity in the latest Permian, terminated by a drastic decline in the earliest Triassic (Fliigel and Reinhardt, 1988; No6, 1988). In this study we present palaeoecological and sedimentological analyses of Permian/Triassic boundary sections in northern Italy and the western United States which reveals the intimate association of dysaerobic facies and the demise of Permian taxa. Geochemical evidence for anoxia

Cerium anomaly The relative proportion of rare earth elements (REE) in the ancient world's oceans is recorded in the biogenic apatite of fish and conodonts. Ce is relatively depleted in oxic oceans such as those of the present day as it is rapidly precipitated from solution with metal oxides (Wright et al., 1987; Elderfield, 1988). However, at times of anoxia Ce remains in solution and this negative Ce anomaly is not developed. Thus, evidence from REEs indicates that oceanic anoxia was typical of the early Palaeozoic and that a transition to more oxygenated conditions occurred in the later Palaeozoic. However, the earliest Triassic is marked by an abrupt return to anoxic conditions followed by oxic conditions for the remainder of the Phanerozoic (Holser et al., 1986; Wright et al., 1987; Wright, 1989).

Sulphur isotopes The marine sulphur isotopic data show a rapid positive swing in the early Triassic which exceeds in amplitude all other events in the Phanerozoic (Claypool et al., 1980; Hoiser and Magaritz, 1987). This is most easily explained by the removal of large amounts of isotopically light sulphur in pyrite from the world's oceans (Garrels and Lerman, 1981). Pyrite formation is associated with anoxic and particularly euxinic conditions where pyrite forms in the water column (Raiswell and Berner, 1985). The magnitude of the swing implies that the early Triassic pyrite-forming anoxic event was

23

ANOXIA ~S A CAUSE OF FHE PERMIAN/TRIASSIC MASS EXTINCTION

one of the largest in the Phanerozoic although Holser and Magaritz (1987) have suggested that if the late Permian oceans were relatively depleted in sulphate, due to extensive evaporite formation (cf. Fischer, 1964), then the significance of this event may be less.

! Carbon~sulphur

ratios

Detailed modelling of the carbon and sulphur cycles indicates that organic carbon to pyrite sulphur ratios were lower than present-day values during both the early Palaeozoic and earliest Triassic (Berner and Raiswell, 1983). This is interpreted to be due to extensive, if not ocean-wide, euxinic conditions and deposition of large amounts of pyrite. Thus, like the Ce anomaly data, C/S values indicate an earliest Triassic anoxic event and the return of conditions similar to those of the early Palaeozoic. Northern Italy

The Permian and Triassic strata of northern Italy records one of the few complete marine successions across the erathem boundary according to the isotopic evidence of Magaritz et al. (1988) and Baud et al. (1989). In most regions of the world a hiatus occurs at this boundary due to an inferred major eustatic regression, but a rapid subsidence rate in northern Italy created accommodation space for marine sedimentation (Broglio Loriga et al., 1986a). Deposition occurred in the western-most region of the Palaeotethyan Ocean (Fig. l), in two depocentres separated by an area of slower subsidence centred on Comelico (Bosellini and Hardie, 1973; No6, 1987; Fig.2). Latest Permian deposition is represented by the Bellerophon Formation, which consists of up to 350 m of limestones, dolomites and minor evaporites (Bosellini, 1964; Broglio Loriga et al., 1986: No6, 1987). In the more open marine eastern basin, centred on Carnia, black, organic-rich limestones (the Badiota facies) are dominant, particularly in the upper part of the Formation (Assereto et al., 1973; Carulli et al., 1988). On a smaller scale, punctuated aggradational cycles (PACs of Good-

.... .i..i".5.. ~ , ^

:i

' : i .;

pataeotomys

" ""

Fig. l. Palaeogeographic map showing the two study rgions. l= Northern Italy. 2-Idaho, western United States. Reconstruction based on Yin (1985).

win and Anderson, 1985) are the dominant depositional motif. More restricted Bellerophon facies occur in the western basin centred on Cadore with marginal facies prograding from the area southwest of the Adige (Fig.2). Thus open marine limestones interdigitate with gypsiferous and sandy dolomites (Fig.3). The topmost metre or so of the Bellerophon Formation is an algal/foram grainstone of remarkable lateral extent for it occurs throughout the region and rests on a variety of facies (Fig.3). This passes gradationally upwards into the Tesero Ooolite Horizon (TOH) which is also widespread (Bosellini, 1964). The grainstone, "records the spreading of a shallow marine facies on a very wide area" (Broglio Loriga et al., 1986a, p.27). It probably rests on an unconformity for it is associated with a major facies change and a change in isotopic ratios as shown in the data of Magaritz et al. (1988), although these authors consider the boundary to be comformable. Assereto et al. (1973) appear to have previously recognised this unconformity which they considered to occur at the Permian/Triassic boundary. However, they incorrectly interpreted the TOH to rest directly on this surface when in fact this only occurs on the southwest margin of the Cadore Basin. Also, the erathem boundary occurs within, not beneath, the TOH as diverse Permian taxa occur in the basal

24

P.B. WIGNALL AND A. HALLAM

N

7.8

3

20kin /

South #

"SASS DE

PUTIA

Tyrol~ ' ~

QN" -

-

C o m e l i c o

"UOMO

"BLETTERBACH

f

O} l IP

(3

High

~_~.__/~~/~Aust ria

6

50kin

area of m.ain alJ~_._.,,,---

-

I

i, f

Trento

";-'..

F- 2 J r

~'-U -"

(

_,~----"-,--~

'

/

c7

oGo-

Fig.2. Location map of localities in the Cadore Basin. Inset, location of study area in northern Italy.

~

~~

Siusi Membe

Fig.3. Simplified cross section through the latest Permian and earliest Triassic strata of the Cadore Basin. Based on authors' data, Broglio Loriga et al. (1986a), No6 (1987) and Brandner (1988). few metres (Broglio Loriga et al., 1986a; Neri et al., 1986; No6, 1987). The T O H is the basal unit of the Mazzin Member which is itself the lowest unit of the Werfen Formation. The Mazzin Member above the T O H is a remarkably uniform micritic unit, poor in fauna, developed throughout the region. The absence of lateral facies variation in the centre of the Member (Fig.3) is in marked contrast to the

variability of the underlying Bellerophon Formation. The progradation of the gypsiferous and dolomitic Andraz Horizon terminated Mazzin micrite deposition in the Cadore Basin. A further rapid marine flooding event established the more silty and fossiliferous Siusi Member in the late Griesbachian (Fig.3). In the following sections detailed lithological and palaeontological evidence is presented from

ANOXIA AS A ('AIJSE OF THE PERMIAN/TRIASSIC MASS EXTINCTION

25

four localities in the Cadore Basin that record depositional conditions from close to the basin margin to the basin centre. Carbonate petrography and foraminifer and algal determinations were assessed in 100 thin sections and in acetate peels.

This, in turn, is overlain by an oolitic horizon (Fig.5) which appears to mark the top of the TOH because the remainder of the accessible section consists of micrites, finely laminated and pyritic in part. The only fauna in the micrites consists of scattered, rare Lingula and Unionites. Higher in the Mazzin Member, sharp-based, centimetre-thick ostracod packstones occur interbedded with the micrites.

Bletterbach section

The entire Permian and early Triassic succession of the South Tyrol is exposed in the splendid sections of the Butterloch canyon. The Bellerophon/Werfen contact is seen at the head of the canyon at Bletterbach (Figs.2 and 4). Layers of nodular gypsum in marly dolomite characterise the highest Bellerophon and these are sharply overlain by the TOH (Conti et al., 1986, Fig.45). Detailed logging at the top of the TOH (Fig.5) revealed a dolomitised sparite with poorly preserved shells sharply overlain by a sparry micrite.

Tesero section

The roadside section at Tesero is one of the most easily accessible and therefore most intensively studied Permian Triassic sections in the region (Broglio Loriga et al, 1986a; Neri et al., 1986; No+, 1987; Magaritz et al., 1988, Figs.6 and 7). The locality is slightly more distant from the basin margin than the Bletterbach section (Fig.2)

Siusi Member Andraz Horizon Mazzin Mbr

Betterophon Formafion

Fig.4. Bletterbach section, showing sharp contact of the Werfen on the BellerophonFormation.

26

I'. B. W I G N A L L A N D A. H A L L A M

BLETTERBACH

SECTION

K~f ~Lingula --*==~Eoe/o~ ~Unionites

]

oosparite

]

bioclastic[irnestone

.~j intrac[astic limestone approximately4rn gap in section

ii •

I

laminated,pyriticrnlcrite

~

laminated,marlymicrife

]

laminatedrnicrite

~

sparite

~

silty,marlydolostone

TOH TeseroOolite Horizon PACs 12

~~ ~

Fig.5. Log of the upper part of the Tesero Oolite Horizon (TOH) and part of the Mazzin Member at Bletterbach. Key also refersto Figs.6 and 13-l 6. Numberedpunctuated aggradational cycles(PACs 5 and 6) are based on a tentativecorrelation with those of the Tesero section (Fig.6). Columns 1 and 2 as for Fig. 6.

and this is reflected in the more distal facies developments. The topmost 10 m of the Bellerophon Formation consists of 0.5 m-scale alternations of hard and soft weathering dolomicrites from which No6 (1987) records ostracods. Acre-thick marly bioclastic wackestone with abundant echinoid debris rests sharply on the topmost dolomicrite. This is the base of a thin aggradational unit which passes upwards through a packstone into a grainstone. A diverse and prolific assemblage of Permian forams and algae (discussed below) occurs throughout the cycle whilst complete specimens of Miocidaris are found in the base of the packstone. The unit culminates in a m-thick bed of oolitic grainstone (the precise thickness varies according to different authors, Fig.7), recrystallised in its upper part to sparry calcite - - a diagnostic feature of freshwater phreatic diagenesis (No& 1987). This unit is a typical punctuated aggradational cycle or PAC in the terminology of Goodwin and Anderson (1985). The sparry oolites are sharply overlain by flaggy micrites, with a fauna of Permian brachiopods and bivalves, which pass upwards into recrystallised sparites. This second PAC is capped by a packstone

displaying drusy calcite, some ferroan calcite crystals and scattered large dolomite rhombs. This freshwater phreatic diagenesis has destroyed much of the original texture although occasional badly preserved "ghosts" of algae and forams can be discerned. A total of six PACs occur in the Tesero section (Fig.6) although they vary in their range of facies developments. PACs 1, 3, 4, and 5 are capped by oolites whilst PACs 1-4 show evidence of freshwater diagenesis in their upper portion. PAC 6 shows weakly developed aggradation as it only culminates in a wackestone containing the common microgastropod Coelostylina werfensis. The basal beds of the PACs are also variable. PACs 1 and 2 contain the diverse fauna noted above. The basal bed of PAC 3 is finely laminated, pyritic and devoid of fauna whilst the laminated silty, marly micrites at the Base of PAC 4 also contain pyrite as well as small conical tubes of uncertain affinity (Fig.8). Pyrite crystals in the marly micrites overlying PAC 6 show evidence for reworking for they are concentrated as basal lags of sharp-based laminae (Fig.9), implying weak benthic current activity and pyrite formation in surficial sediments or the water column. Polished thin sections reveal that the majority of the pyrite has weathered to goethite in the Tesero outcrop. However, three pyritic levels were also detected in the Gartnerkofel core from southern Austria (Fig.2) at the same stratigraphic horizons; these give light sulphur isotopic values implying syngenetic formation (Holser et al., 1989, 1991). Higher beds of PACs 3 and 4 and the basal beds of PACs 5 and 6 consist of finely laminated micrite with sharp-based beds several millimetres thick consisting of flattened intraclasts (Fig. 10). Dissolution of this fabric-type is common and a range of stylonodular fabrics is developed (Figs. 11 and 12), mainly towards the upper portions of the PACs. At its most extreme development (Fig.12), the original fine laminae are destroyed and an intraclastic or burrow mottled appearance is produced, particularly in hand specimen. The remainder of the Mazzin Member above PAC 6 reveals a more uniform facies development in which cyclicity can no longer be discerned (Fig.6). Finely laminated micrites are initially

ANOXIA AS A CAUSE OF THE PERMIAN/TRIASSIC MASS EXTINCTION

2"7

c

I

1

TESERO

nn storm beds : osfracods and ~crogastrepods

SECTION

]

k

".:

~ef"layer

: :..'.- -:::- .,.i

!

}i/(

6-[ o

o

~_!:iiiiii!~!iil

Z_-_-Z_Z-Z-Z

y.

~z~z-_~_-avicul0pecfinids

o

oo°

\5:;:::!:i proximo"

" '" d~stal

PACs

o./ (33

nL-7-

Fig.& Log of the Tesero section from the uppermost Bellerophon to a level high in the Mazzin Member. Column I shows the occurrence of dolomite (shaded black) and column 2 indicates the presence of evidence for freshwater phreatic diagenesis. Such evidence, combined with petrographic facies analysis is used to interpret the PACs.

dominant. Centimetre-thick, sharp-based shelly packstones, containing ostracods and microgastropods, interbedded with more massive micrites containing scattered ostracods, become common towards the top of the outcrop. The packstones can be laterally persistent although some are more lens-like.

Interpretation of" carbonate j'acies at Tesero The PAC motif was not recognised by Neri et al. (1986) and No6 (1987) although Brandner (1988, Fig.2) recognised parasequences analogous to the cycles described here. However, he appears to identify more parasequences than the five and

28

P. B, WIGNALL AND A. HALLAM

NERI et a[.(1986)

THIS STUDY

NOE (1987)

infrasparite

L_55 ase

~sfic limestone ,_~

Tidal flat facies microspar with cryptalgal laminae

limestone mestone d mud flats cracks

4

~ ' ~ I d brachiopod horizon

~od horizon ==-~-fidal flat facies

[ga[ timestone Jurrows in ,[omife

• ~

Marine euxinic facies finely (aminafed pyritic micrife Peritidal facies - • ~ do[omifised and recrysf• ~ •1 at[ised infrabiomicrite sparite

~'o

,~

oosparite

tlm

Fig.7. Comparison of logs of the TOH at Tesero. weak sixth PAC we record. The basal PAC shows a spectrum of depositional environments from a mid-shelf wackestone to peritidal oolites. The three pyritic levels appear to record deeper conditions for such finely laminated, poorly fossiliferous facies are typical of deep shelf dysaerobic to anaerobic environments, Our interpretation of the depositional environments (Fig.7) differs considerably from that of previous authors. No6 (1987) considered the TOH to represent solely peritidal and supratidal deposition. The laminated micrite facies were interpreted as tidal flat facies with algal mats. The thin intraclastic layers were considered to indicate mudflat erosion by major storms. Supporting evidence for a peritidal origin comes from the extensive freshwater phreatic diagenesis but this is only found at the tops of the PACs (Fig.6). The pyrite from the micrites should have a heavy isotopic signature if it formed in tidal flats but the light values recorded from pyrite in the Gartnerkofel

core indicates syngenetic formation (Holser et al., 1989). In fact the pyritic micrites and recrystallised sparry micrites have a mutually exclusive distribution testifying to their different depositional environments. The finely laminated micrites of the Mazzin Member indicate an absence of a burrowing fauna presumably indicating that dysaerobic-anaerobic conditions were important. Above PAC 6, the upward transition from finely laminated micrite to micrites with sharp-based intraclastic horizons and ostracod/microgastropod packstones reflects an increase in the frequency and energy of current events possibly reflecting a slight shallowing of the depositional environment. The packstones have been previously interpreted as distal tempestites (Neri et al., 1986; Brandner, 1988). The succession above the TOH at Bletterbach (Fig.5) records a similar increase in the current energy. Neri et al. (1986, p. 116) consider that the Mazzin Member, "is affected by strong bioturbation which

ANOXIA AS A CAUSE OF THE PERMIANiTRIASSICMASS EXTINC'['ION

29

Fig.8. Photomicrograph of silty micrite from the base of PAC 4, Tesero, showing scattered opaques (goethite replacing pyrite), ostracods and problematical conical tubes. Similar tubes have been illustrated from the earliest Triassic of Yugoslavia (Ramovg, I986). Scale bars is 0.5 ram.

Fig.9. Photomicrograph of a sharp-based lag of pyrite (now mostly weathered to goethite) from the pyritic horizon immediately above PAC 6 at Tesero. This is interpreted to have formed by a weak storm-induced erosion and redeposition evenl. Scale bar is 0.5 mm.

30

P. B. W I G N A L L A N D A. H A L L A M

Fig.10. Direct print from acetate peel of finely laminated micrite with a sharp-based horizon of intraclasts (distal tempestite). Same horizon as Fig.9. Scale bar is 5 mm.

Fig.11. Acetate peel print from close to the base of PAC 5, Tesero, showing alternation of finely laminated micrites and stylonodular layers (sensu No6, 1987). A single specimen of Unionites is seen, preserved in life position. Some of the smaller, circular "nodules" (example arrowed) may be burrows. Scale bar is 5 mm. m a y o b l i t e r a t e the original b e d d i n g " , i m p l y i n g a normally oxygenated depositional environment. Similarly, B r a n d n e r (1988) considers the Werfen seas to have been better ventilated t h a n the Bellerop h o n Seas due to an increase in h y d r o d y n a m i c

energy r e c o r d e d by the thin s t o r m beds. H o l s e r et al. (1991) also refer to high energy tempestites in the M a z z i n M e m b e r . H o w e v e r , there is no convincing evidence for " s t r o n g b i o t u r b a t i o n " in the M a z z i n a l t h o u g h well d e v e l o p e d s t y l o n o d u l a r

31

A N O X I A AS A C A U S E O F T H E P E R M I A N / T R I A S S I C M A S S E X T I N C T I O N

Fig. 12. Acetate peel print of a well developed stylonodular fabric from the upper portion of PAC 6, Tesero, probably derived from an originally finely laminated precursor. Scale bar is 5 mm.

fabrics (developed from finely laminated precursors) do have a burrowed appearance (Fig.12). The thin storm beds, c o m m o n in the upper parts of the Mazzin, require an absence of bioturbation not an increase in energy. Thin beds are readily destroyed by bioturbation unless burrowers are inhibited by low oxygen levels (e.g. Ekdale and Mason, 1988).

Boundaries of" the Tesero Oolite Horizon The lowest oolite is taken as the base of the T O H (Fig.7) but the sharpest lithological break, and therefore the most obvious choice for the formational boundary, occurs at the base of the wackestone half a metre below this level. Indeed, the change of carbon and oxygen isotopic values across this lower boundary led Magaritz et al. (1988) to infer the presence of a minor disconformity at this level. The top of the T O H at Tesero appears to have been arbitrarily chosen as the top of our PAC 3 by Broglio Loriga et al. (1986a). No6 (1987) logically suggested that the top should record the highest occurrence of peritidal conditions within

the Mazzin and she chose the same horizon (Fig.7). In fact peritidal facies are developed above PAC 3 at the top of PAC 4 (Fig.8) and a half metre thick channelled oolite, of probable peritidal facies caps PAC 5. Uomo section

The U o m o section in the central Dolomites occurs in a considerably more distant location from the basin margin than Tesero (Fig.2). This is reflected in the more distal facies development of both the Bellerophon and Werfen Formations (Broglio Loriga et al., 1986b). The uppermost Bellerophon limestones are poorly-exposed, dark, bioclastic wackestones with common ostracods and forams. The topmost 0.3 m of the Formation is a f o r a m - a l g a l grainstone (Fig. 13) dominated by species of the miliolid Hemigordius (Table 1). This rests erosively on the underlying strata and a hiatus is indicated by the presence of rounded clasts of wackestone showing truncated styolites, in the base of the grainstone. A 2.0 m thick oolitic grainstone, altered by freshwater phreatic diagenesis, caps the foram-algal

32

P.

B.

WIGNALL

AND

A. HALLAM

UOMO SECTION

PF ~f ts

{,

thin iumachelles of gastropods and bivalves

31 v..:. ..... i . . - ..=.:

3:

,.+

3( N N fO

6.5m of micrite with a ¢*"l"""'4"lar fabric

Iostylina osl'racods PAEs

Fig.13. Log of the topmost Bellerophon Formation and lower Mazzin Member at Uomo. PACs 1-5 are interpreted to correlate with those at Tesero. Column 1 indicates evidence for freshwater phreatic diagenesis where black.

grainstone (Fig.13). This is clearly PAC 1 as seen at Tesero for the same sequence of lithologies is developed. A second aggradational cycle occurs above PAC 1 and, again like the equivalent level at Tesero, it aggrades into an extensively recrystallised bioclastic grainstone at the top. A further three PACs occur above this level although they do not culmi-

nate in such distinctly peritidal facies (Fig.13). Finely laminated pyritic micrites occur at the bases of PACs 3 and 4. The weakly developed PAC 6 at Tesero is even more poorly defined at U o m o because a 6.5 m thick succession of micrites overlying the stromatolitic horizon at the top of PAC 5 culminates only in a Coelostylina wackestone. This passes sharply upwards into finely laminated

A N O X I A AS A C A U S E O F T H E P E R M I A N , T R I A S S I C M A S S E X T I N C T I O N

-~

%

,,q

m. o e

~o

~,~

~o

o II

E

~m II r~ 0

o~

oo

~

~

'~~

i~.

o

p

e~

o ¢-

o~ N N

e~

~,

~

E ~

E ~

34

P.B. WIGNALL AND A. HALLAM

micrites, pyritic at their base (Fig. 13). As at Tesero, much of the fine lamination has been altered to a stylonodular fabric. The upper Mazzin is also familiar, thin packstones with bivalves (mostly Unionites), Coelostylina and rare intraclasts occur interbedded with micrites containing rare specimens of Lingula and Unionites. The contact of the Andraz Horizon with the overlying Siusi Member is well seen in the slopes of Uomo. This reveals a sequence of beds, similar to those seen at the Bellerophon/Werfen contact (Fig. 14a). The topmost Andraz consists of unfossiliferous, centimetre-laminated silty dolomicrites. Rounded clasts of this facies occur in the wackestones at the base of the erosive-based Siusi Member. The wackestone passes upwards into a grainstone (Fig.14a) containing abundant bivalve fragments, gastropods and rarer ooids and echinoderm debris (ophiuroids and possibly crinoids). Much of the overlying 60 m of the Siusi consists of flaggy, silty, marly micrites with dense bedding plane assemblages of Unionites and Claraia. C. claraia of latest Griesbachian age (Nakazawa, 1981), occurs in the lower 20 m. This is abruptly succeeded by C. aurita characteristic of the succeeding Nammalian Stage, indicating that the Stage boundary occurs within the Siusi Member (e.g. Broglio Loriga et al., 1986a). The level of turnover coincides with a remarkable, half metre thick, red, gastropod grainstone in which extensive haematisation of Coelostylina shells has occurred (Fig. 14b). The bed contains tabular siltstone clasts up to 30 cm in diameter in its upper half and is

t

~,~ding

planes covered in Claraia

claraia

I , ' , ' , L ~ fragmented gastropods, ~ Claraiaand echinoderms g N "U_ o

erosive-based, suggesting an hiatus of unknown duration at the stage boundary. Sass de Putia section

The Sass de Putia section lies close to the centre of the Cadore Basin (Fig.2) where a relatively thin, distal Permian-Triassic succession is developed (Fig.15; Broglio Loriga et al., 1986a; Broglio Loriga et al., 1986b). The upper Bellerophon consists of wackestones and thin marley partings and a diverse marine fauna dominated by forams (Table 1). The topmost marly layer is sharply overlain by a bioclastic grainstone containing abundant forams (including fusulinid species, of which Nankinella cf. quasihunanensis is the most abundant). A moderately diverse brachiopod fauna occurs in the same bed (Broglio Loriga et al., 1986c). This appears to be the basal bed of PAC 1 for it is succeeded by 0.65 m of oolitic grainstone (Fig.15). PAC 2 consists of marly micrites with aviculopectinids overlain by a second oolitic grainstone. Higher PACs cannot be discerned because the succeeding 3.5 m of strata consists of finely laminated micrites with ostracods whilst higher levels in the Mazzin are poorly exposed. Scattered outcrops reveal only micrites with a few intraclastic layers. As at Uomo, the Andraz/Siusi contact is well exposed (Fig. 16). The uppermost Andraz Horizon consists of thin bedded, marly micrites and rippled fine sandstones with rare clay-lined burrows. The contact with the Siusi Member is sharp and the

,_ 2"~--C. aurifa E:~,',~-horizan

of siltstone clasts

I EI ~ L"-..~"--.1)tabular cross bedded

~-~ ~gastropod grainstone •~ E I~;i::] flaggy siltstones with .2 u~

I~i!-::] 0 ~ .....

1

Unionifes ,

,

mwpg

,

bright red, sitty dolomicrite

I-

(._

....

A.

B.

mwpg

Fig.14a. Log of the Andraz Horizon:Siusi Member Contact at Uomo. b. Log of the lower intra-Siusi horizon around the Griesbachian/Nammalianstage boundary, Uomo.

ANOXIA AS A CAUSE OF THE PERMIAN/TRIASSIC MASS EXTINCTION

SASS DE PUTIA SECTION b rare

7

ostracods and Unionifes

E

flaggy, marly micrites with scattered Unionites and packstones/grainstones up to 0.4 m thick with abundant shells of gastropods and bivalves including Claraia.

Faunal and floral changes across the Permian• Triassic boundary

7

~

PACs i~

-Ombomo

mwpg Fig. 15. Log of the Bellerophon/Werfen contact at the Sass de Putia section.

~

35

. . . . ta becomes common higher in the section bed of oo[ite micrite sive-based ool.ite with "ac[asfs ,raia c/araJa and Planolites

she{[y packsfones inferbedded marly micrifes

Sass de Putia Section and small Planolites

oderm debris common ;

common

crife grades up into rippled sandstone

mwpg c <

Fig. 16. Log of tile Andraz/Siusi boundary beds at the Sass de Putia.

basal bed is an echinoderm-mollusc packstone containing rounded aggregates of dolomite rhombs presumably reworked from the underlying Andraz. The succeeding strata consists of alternations of

The distribution of the microfauna and microflora at Tesero has been analysed in a series of thin sections sampled every 0.25 m in the Bellerophon and TOH and around every 1.0 m in the remainder of the Mazzin Member. More widely spaced samples were also taken at the Uomo and Sass de Putia sections. The macrofauna was identified by more closely spaced sampling in the field. The packstones and grainstones of the lowest PAC contain a prolific, diverse well preserved fauna and flora (Fig.17) including taxa indicative of the Dorashamian stage (No6, 1987). Permian taxa also occur in the overlying PAC although there is a decline in the species richness, and preservation is poor due to extensive recrystallisation. This is the highest horizon at which Permian species occur and No6 (1987) chose the base of the overlying micrite as the Permian-Triassic boundary (this was also one of the alternatives discussed in Broglio Loriga et al., 1986a). This is an eminently suitable choice for it corresponds to the base of the first pyritic horizon and correlates with the loss of the Ce anomaly in the REE profile and a major negative carbon isotope shift as detected in the Gartnerkofel core (Holser et al., 1989, 1991) - - the latter event occurs globally (Baud et al., 1989). Unfortunately ammonoids, the favoured fossils for defining the erathem boundary (Teichert, 1990), are not found at this level possibly reflecting slightly lowered salinity within the Cadore Basin at this time. However, conodonts are also useful for defining the boundary and they are known sporadically from the region (Perri and Andraghetti, 1987). Hindeodus parvus has been recorded ranging from the first pyritic horizon to a level in the mid Mazzin at Gartnerkofel (Sch6nlaub, 1991). From the Tesero section itself, H. latidentatus occurs in the base ofPAC 2 (Broglio Loriga et al., 1988). Both suggest an horizon around the earliest part of the Otoceras woodwardi

36

P. B. W I G N A L L A N D A. H A L L A M

E IE

' !

Izll

Ill ;

,

%!

I llI! tn

II!

I l l 1I

.~

{~ Z ~ ~t j

.~

tJ . ~.~ e.~ ~.~

'n E v, ,J ~ ~.~

~ ~,~,m

,a ~ : ~

. ~ *... ~ ~

Fig.17. Floral and faunal ranges in the Tesero section. Additional Permian taxa are recorded in Broglio Loriga et al. (1986a), Neri et al. (1986) and No6 (1987).

Zone, the basal ammonoid zone of the Triassic (Broglio Loriga et al., 1986a). H. typicalis, which ranges across the Permian/Triassic boundary, has also been recorded from the upper Bellerophon and the Mazzin (Perri and Andraghetti, 1987), indicating that the top Bellerophon unconformity is unlikely to mark a major break in deposition. Only three of the twenty-six taxa recorded from the latest Permian at Tesero passed into the Triassic (Fig. 17). This appears to represent a true extinction event for these are the highest occurrences for many major late Palaeozoic groups such as the fusulinids and brachiopod taxa. Similarly, this is the last occurrence of Gymnocodium bellerophontis, the last representative of a successful group of late Permian red algae. Gymnocodiaceans reappeared in the late Cretaceous (Wray, 1977) but it is unlikely that they were directly related to the Palaeozoic family of this name. The successful late Permian green alga Mizzia also disappeared at this level (Herak et al., 1977).

Buggisch and No6 (1986) and No6 (1987) noted that the topmost Bellerophon faunas are the most diverse of the entire Formation and that several lineages, such as that of the miliolid Hemigordius, were diversifying rapidly at this time. A similar diversity increase has been noted for Tethyan reef faunas (FliJgel et al., 1988; Fliigel and Reinhardt, 1989). Such observations do not accord with the widely held view that the Permian/Triassic crisis was a protracted event with a gradual diversity decline (e.g. Newell, 1988; Teichert, 1990). However, these latter studies equate diversity with species richness but this value is most directly controlled by the sample size (Saunders, 1968). As the area of late Permian sediments is of very limited extent (e.g. Schopf, 1974), the global sample size is of necessity small and the gradual decline could simply be an artefact. Dominance-diversity, which considers both the number of species (species richness) and the proportions in which they occur (equitability), is a more appropriate measure of

A N O X I A AS A C A U S E O F T H E P E R M I A N / T R I A S S I C M A S S E X T I N C FION

extinction events. Thus the trend of dominance diversity in the upper Bellerophon Formation has been investigated using the Shannon-Wiener dominance-diversity index, H, where:

H = ~ piln Pi i--1

p is the proportion of species i in a sample with s species. This widely used index has the advantage of being virtually sample-size independent for all but the smallest samples (Wignall, 1990a). For consistency, samples were chosen from bioclastic packstone/grainstones and the taxa identified in 3.5 cm 2 of thin section. It is likely that taxa with multi-element skeletons, such as echinoids, contribute more skeletal elements to the sediment than single element taxa such as foraminifers for example. No attempt was made to quantify this bias as the relative proportions of higher taxa remained roughly constant in the samples. The results (Table 1) confirm the qualitative assessments of No6 (1987) and other workers, that there is an upward increase in dominance-diversity in the Bellerophon. Ostracods dominate the thin section assemblages of the Mazzin Member above PAC 2 and H values do not exceed 0.3. The earliest Triassic macrofaunal assemblages are dominated by the shallow infaunal bivalve Unionites (Fig. ll) and Lingula which can be locally abundant on some bedding planes. Higher in the Tesero section, moderate numbers of the small, poorly ornamented Claraia wangi-griesbachi appear (Fig.17) in association with the Unionites bedding plane assemblages. The conodont Isarcicella isarcica appears at around the same level as Claraia (Sch6nlaub, 1991) and indicates the Ophiceras Zone of Upper Griesbachian age (Broglio Loriga et al., 1986a). Discrete bedding plane assemblages o f bivalves containing "paper pectens" such as Claraia are typical, if not diagnostic, of oxygen-restricted (dysaerobic) deposition (Wignall, 1990b). Thus the combination of lithological and palaeontological evidence indicates dysaerobic/anaerobic depositional conditions for much of the Mazzin Member.

Sequence stratigraphic interpretation The four sections record marked lateral facies variations from the margin to the centre of the

37

Cadore Basin but correlation can still be achieved using faunal data and the identification of sequence stratigraphic-style bounding surfaces. Undoubtedly the most distinctive surface is the sharp, erosional contact in the uppermost Bellerophon which separates a wide range of facies beneath from remarkably uniform strata immediately above. Facies changes in the succeeding strata are related to aggradational cycles which become progressively more poorly defined towards the basin centre (Fig. 18). However, the basal two cycles can be correlated across the basin along with the pyritic micrite which caps them. The second major pyritic level above the sixth PAC is also widespread and both horizons were probably also recorded in the northern Carnia Basin by Holser et al. (1989, 1991). Less widespread, but also of use in correlation, is the aviculopectinid horizon in the basal beds of PAC 2. As Brandner (1988) recognised, the PACs are equivalent to parasequences, the smallest scale aggradational sediment packages in the sequence stratigraphic model of Exxon Production Research (EPR, cf. Posamentier and Vail, 1988). The stacking of the parasequences reveals other aspects of the EPR model. Thus PAC 1 rests on an unconformity or sequence boundary and the more limited extent of the oolites in PAC 2 suggests backstepping, a characteristic of transgressive systems tracts. Such tracts are overlain by maximum flooding surfaces consisting of slowly deposited, deep water marine facies (condensed sections, sensu Loutit et al., 1988) - - the finely laminated, pyritic micrite at the base of PAC 3 is probably a condensed section. Maximum flooding surfaces are overlain by progressively more progradational parasequences of the highstand systems tract - - PACs 3 5 appear to form an example. Thus by PAC 5, peritidal facies, represented by stromatolites in the Uomo section, are developed far into the basin (Fig.13). The succeeding PAC 6 and the central part of the Mazzin Member indicate the establishment of a uniformly deep shelf environment across the basin. The pyritic micrites overlying PAC 6 may represent another maximum flooding surface. Shallowing towards the top of the Mazzin culminated in the development of erosive surfaces associated with reworked clasts, probably sequence

38

P. B. W I G N A L L A N D A. H A L L A M

Butterloch ...........

~-_----

Uomo

Tesero

Sa$s

de Pufia

t

_1 ...........................

....

~¢m

FACIES

TYPES

[ ~ Supratida[, gypsiferous []~Supratidat, dolomitic []Perthdat g shallow marine[~ Oohfic shoal [ ] D i s t a l fempeshte

[]

Shelf micrife (inctudes [aminated dysaerobic facies) •

Anaerobic (pyrd'lc) m,cr~te

Fig.18. Correlation of the four sections measured in the Cadore Basin.

boundaries, in the base and lower part of the Siusi Member. Although somewhat controversial (e.g. Hallam, 1988), the EPR model allows the detailed documentation of eustatic sea-level changes (Fig. 19). A relatively minor late Dorashamian sea-level fall produced a sequence boundary and minor hiatus in the topmost Bellerophon. This was succeeded by an extremely rapid, stepped sea-level rise in the latest Dorashamian and early Griesbachian. The erathem boundary and mass extinction occurs at the level of maximum rate of sea-level rise which caused basinal anaerobic deposits to extend far onto the basin margin. Minor pulses of sea-level fall occurred in the later Griesbachian producing two minor sequence boundaries (Fig. 19). Permian/Triassic sea-level changes have also

been reported by Haq et al. (1988) but they do not accord with our interpretation (Fig.19). Although they consider that sea-level was rising across the boundary, the depicted rate of rise is amongst the slowest they record for any interval in their post-Palaeozoic chart (Haq et al., 1988). This is curious because the rapidity of the earliest Griesbachian transgression Is well known (e.g. Kummel and Teichert, 1970; Schopf, 1974; Holser and Magaritz, 1987) and the quantitative analysis of Paull and Paull (1986) indicates that it might have been spectacularly rapid. Western United States

A thick Permian and Lower Triassic succession is preserved in the Utah-Idaho-Wyoming portion

ANOXIA AS A CAUSE OF THE PERMIAN,TRIASSIC MASS EXTINCTION

39

~ I [ , - ~ T R AN SfRESSION "

z

~c

THIS

HAQ et at (1988)

STUDY

._J

~c

z

j

z

~c

~-r L]3 .~ <£. oO ¢.q

I.-L~

maximum ~.

flooding surfaces _...__d . . . . . . . . . .....

~, \ ~

o~

sequence boundary

Fig. 19. Interpreted eustatic sea level changes across the Permian/Triassic boundary based on evidence from Italy, compared with the propose curve of Haq et al. (1988).

of the Rockies. This records deposition on the western margin of the Pangaea supercontinent (Fig.l), in a similar palaeolatitude to northern Italy. The youngest preserved Permian strata belong to the celebrated Phosphoria Formation although there is an hiatus of considerable duration spanning much of the latest Permian (Boyd and Maughan, 1973; Collinson and Hasenmueller, 1978). Collinson et al. (1976) attributed the hiatus to a collision and uplift event. However, the basinal configurations of both the Phosphoria and Lower Triassic Formations are remarkably coincident with their depocentres lying in southeast Idaho and the basin margin lying to the east in Wyoming (Maughan, 1979; Paull and Paull, 1986; Fig.20). There is also no evidence of an angular unconformity as the Lower Triassic formations gently onlap onto a flat, peneplained surface of the Phosphoria (Newell and Kummel, 1942; Paull et al., 1985). It is more probable that the hiatus is due to the global late Permian regression. The lower Lower Triassic is divided into two major formations. The Woodside Formation of the eastern outcrops (principally in Wyoming) is an unfossiliferous red siltstone (Newell and Kummel, 1942) with minor gypsum beds (Thomas,

1934) - - clearly a marginal facies. The Woodside interdigitates with the Dinwoody Formation in western Wyoming and eastern Idaho although in the basal Griesbachian a shaly leaf of the Dinwoody extended eastwards across the entire Triassic outcrop (Paull and Paull, 1987). The Dinwoody Formation is dominated by grey-green calcareous siltstones with minor limestones and sandstones (Kummel, 1957; Paull and Paull, 1987). Boyd and Maughan (1973) recorded finely laminated, organic-rich silty shales near the base - - probably a dysaerobic facies. Subsequent studies have failed to confirm the organic enrichment although dysaerobic to anaerobic conditions undoubtedly occur in the basin-centre Dinwoody of southeast Idaho (Kummel, 1957; Paull et al., 1985). Fieldwork was concentrated on the lower Dinwoody Formation in southeast Idaho to investigate the evidence for dysaerobic facies in the Lower Griesbachian. The region has been subject to late Cretaceous, eastward-directed thrusting (Collinson and Hasenmueller, 1978). However, the displacement of the thrust blocks does not appear to have been great (less than 100 kin) and the relative palaeogeography is still preserved; thus basinal facies still occur westwards of marginal facies.

40

P. B, W I G N A L L A N D A. H A L L A M

Ro.."'C,Biackfo'o't /

,,

_

~" ~ ,totalities studied

Dinwoody

Basin

0 ~ ~ ~ ? ?

IDAH

e

IDAHO

0

/

.

{z:

.--

~

Paris • . Can =Paris _.Y on.--- =Paris

\

V

&Hot

Fig.20. Location of the Idaho study sections and (inset) area of the Dinwoody Basin.

Sections of SE Idaho

Hot Springs, Bear Lake Valley The inverted succession at Hot Springs (Fig.21) shows a virtually complete transect from the upper Phosphoria Formation to a level high in the Lower Triassic (Kummel, 1957; Ciriacks, 1963; Carr and Paull, 1983). The basal bed of the Dinwoody is an upward fining sandstone containing abundant phosphatic ooids and rounded clasts of phosphatic mudstone reworked from the underlying Phosphoria (Paull and Paull, 1986, Fig.7). The succeeding 5 m is unexposed although Kummel (1957) reports a shale facies at this level. This is overlain by a thick succession of silty micrites exhibiting platey fissility (Fig.21). In thin section, thin discontinuous laminae of clay minerals and minor organic matter occur which maybe a product of pressure solution. The only fauna consists of scattered

specimens of Lingula borealis. This rather equivocal facies is succeeded by current ripple-laminated sandstones interbedded with minor sandy lumachelles. The fauna is more common and includes Myalina sp. and Claraia stachei which is restricted to the lumachelles. Unionites spp. appear a short distance above the section shown in Fig.21.

Paris Canyon An inverted succession of the upper Phosphoria and lower Dinwoody Formations is well exposed in a hillside approximately 100 m northwest of the small hydroelectric power station in Paris Canyon (Fig.20). The contact between the formations appears to be marked by a small fault (Fig.21) although Ciriacks (1963) considered that a complete section occurs here. The lowest Dinwoody consists of fissile, silty micrites with well-developed laminae. This lithol-

41

A N O X I A AS A C A U S E O F T H E P E R M I A N ; T R I A S S I C M A S S I ~ X T I N C I I O N

HOT SPRINGS

LITHOLOGY

B

Sandstone, fine, ripple laminated Limestone, shelly, sandy Hittite, silty, laminated

@ Hicrife, silty, massive @ Hittite, marly, pyritic ii 10m section

Hittite, phosphatic with chert nodules FAUNA

~ Lingula

(~

ophicerafid

Claraia stachei ( ~ hlyatina

Oiplocraterion

PARIS CANYON 3metres

)

fi'ed contact

Fig.21. Correlation of the Permian/Triassic boundary sections of Hot Springs and Paris Canyon, SE Idaho.

ogy grades into highly fissile marly micrite with occasional bedding planes covered by large specimens of C. stachei and rarer ophiceratids. In thin section this lithology is laminated on a 0.5-3 m m scale. The thinner laminae consist of carbonaterich and clay-rich alternations. The thicker laminae are carbonate-rich, sharp-based horizons commonly showing small flame structures at their base. Pyrite crystals appear to have been initially common although they are now weathered to goethite.

Total organic carbon values are around 0.25% in this lithology. Nearby exposures of the higher Dinwoody Formation reveal heterolithic lithologies, including peritidal oolites and rippled sandstones. Bear Creek A short section at a temporary trackside exposure 4 km east of Blackfoot Reservoir, in the valley

42

of Bear Creek (Fig.20), revealed 2.5 m of strata at a level low in the Dinwoody. This mostly consisted of silt laminated calcareous shale with common pyrite forming up to 5% in thin section, although it has now mostly weathered to goethite. Discrete bedding plane assemblages consist of small species of Claraia lacking radial ornament, rarer Myalina, ophiceratids and orthocone nautiloids. This is the first record of orthocone nautiloids from the Griesbachian, although they are unfortunately poorly preserved.

Discussion and comparison with northern Italy The depositional environment of the Lower Dinwoody at Hot Springs is rather enigmatic although the very low diversity and presence of wispy lamination could possibly suggest oxygen-restricted deposition. More unequivocal dysaerobic deposition is recorded in the more westerly localities of Paris Canyon and Bear Creek which lie closer to the centre of the Dinwoody Basin (Fig.20). Fine lamination, common pyrite and bedding planes covered in thin-shelled bivalves such as Claraia are typical of lower dysaerobic facies (e.g. Savrda and Bottjer, 1987; Wignall, 1990a). The thicker, sharp-based laminae appear to have been rapidly deposited from short-lived current events and maybe the distal product of storms. The preservation of such fine-scale sedimentary structures again testifies to low oxygen values inhibiting a bioturbating infauna. Although not studied in detail, the higher levels of the Dinwoody appear to be a shallower water facies (Kummel, 1957). The Dinwoody thus records a rapid marine flooding event across a flat-lying surface of the Phosphoria Formation. This led to the establishment of relatively deep water oxygen restricted conditions across the entire region for dysaerobic facies also appear to occur in marginal locations in Wyoming (Paull and Paull, 1986). The age of this basal Dinwoody event can be determined from several lines of biostratigraphic evidence. The conodonts Hindeodus typicalis and Isarcicella isarcica are recorded from the region (Paull et al., 1985) and indicate a level in the mid-Griesbachian. As noted, ophiceratids also occur but they are of little precise biostratigraphic value as they range

P. B. W I G N A L L A N D A. H A L L A M

throughout the Griesbachian (Yin et al., 1988). C. stachei first appears in the mid Lower Griesbachian according to Yin (1985). Therefore the Dinwoody flooding event is an intra-Griesbachian event and it probably correlates with the second flooding event in northern Italy (Fig. 19). The benthic faunas recorded from the Griesbachian of Italy and Idaho are virtually identical at the generic level, in accord with the previously recognised lack of endemicity in Lower Triassic faunas (e.g. Kummel, 1973; Kristan-Tollman, 1987). Equally remarkable is the close similarity of the contemporaneous lithologies of the midMazzin Member and lower Dinwoody Formation. We attribute this to rapid deepening, due to eustatic sea level rise, leading to the establishment of similar deep water dysaerobic conditions in these widely separated regions.

Conclusions Geochemical data based on Ce values, carbon/ sulphur ratios and probably sulphur isotopes suggest an abrupt transition to widespread anoxic conditions in low latitude marine waters of the earliest Triassic. It seems clear that this anoxic event is implicated in the mass extinction of shallow marine taxa. Kozur's (199l) studies of the deep water paleopsychospheric ostracod faunas shows that they are little affected by the event. Kozur's conclusion is that deep oceanic waters remained fully oxygenated throughout the Permian and Triassic, in contrast to the modelling studies of Hoffman et al. (1991). This shelf anoxic event hypothesis has been tested with reference to stratigraphic sections in the Alps of northern Italy and the Rocky Mountains of the United States. Conodont and other evidence suggests that sections across the Permain/ Triassic boundary in the Italian Dolomites are amongst the most complete known, with the Upper Permian Bellerophon Formation passing into the Lower Triassic Werfen Formation without any significant hiatus. The Tesero Oolite Horizon at the base of the Werfen, underlain by a minor unconformity, represents a major spread of marine facies following a short-lived regression. A diverse Permian fauna and flora occurs within the Tesero

ANOXIA AS A CAUSE OF "IHE PERMIAN/TRIASSIC MASS EXTINC'I ION

but it disappears at the base of a pyritic micrite marking a major negative shift in carbon isotope values. On the basis of detailed examination of four sections across the Permian/Triassic boundary in the Dolomites, representing both proximal and distal facies with respect to the margin of the Cadore Basin, a number of punctuated aggradational cycles can be recognised. Laminated micrites formerly interpreted as tidal fiat facies in fact represent dysaerobic to anaerobic conditions at the sea bed, and claims of extensive bioturbation in the basal Triassic Mazzin Member of the Werfen Formation are not supported. Trustworthy evidence for a peritidal origin comes only from the cycle tops. The highest Permian is represented by the Bellerophon Formation, and it has been claimed that the foraminifers, like the reef faunas over a wider Tethyan zone, are most diverse at the top of the Formation. The more general claim of a decline in marine faunal diversity towards the end of the Permian is questionable, because it could be an artifact of small sample size due to the limited amount of marine strata of latest Permian age. The trend of dominance-diversity up the Bellerophon succession, using the Shannon-Wien index, is a better measure of extinction because it considers both species richness and equitability. The results presented here support the claim that there is indeed an upward increase in dominance diversity up the Bellerophon succession, and the disappearance of Permian faunas at the end of the period is abrupt, taking place not at the level of regression but shortly above, associated with a major transgression, within the Tesero Horizon. A combination of lithological and palaeontological evidence indicates dysaerobic to anaerobic conditions for much of the overlying Mazzin. Sections in the western United States reveal a considerable hiatus at the Permian/Triassic boundary, and the Dinwoody Formation records a rapid marine flooding event in the early Triassic, with the establishment of relatively deep-water oxygenrestricted conditions across the entire region. The broad similarity in lithofacies and biofacies between the Dinwoody and Mazzin suggests that the events studied in detail in the Alps are likely

43

to reflect a regionally much more extensive situation.

Acknowledgements We thank the NERC for funding all the fieldwork. P. B. W. thanks George Grabowski for arranging analysis of Idaho samples and Simon Dean for seeking pyrite in the Dolomite samples.

References Assereto, R., Bosellini, A., Sestini, N. F. and Sweet, W. C., 1973. The Permian Triassic boundary in the southern Alps (Italy). In: A. Logan and L. V. Hills (Editors), The Permian and Triassic Systems and their Mutual Boundary. Mem. Can. Soc. Pet. Geol., 2:176 199. Baud, A., Magaritz, M. and Holser, W. T., 1989. Permian Triassic of the Tethys: carbon isotope studies, Geol. Rundsch., 78:649 677. Benson, R. H., 1984. The Phanerozoic "'crisis" as viewed from the Miocene. In: W. A. Berggren and J. A. Van Couvering (Editors), Catastrophes and Earth History. The New Uniformitarianism. Princeton Univ. Press, pp,437 446. Berner, R. A. and Raiswell, R., 1983. Burial of organic carbon and pyrite sulfur in sediments over Phanerozoic time: a new theory. Geochim. Cosmochim. Acta, 47:855 862. Bosellini. A.. 1964. Stratigrafia. petrografia e sedimentologia delle facies carbonatiche al limite Permo-Trias nelle Dolomiti Occidentali. Mem. Mus. St. Nat. Ven. Trident., 15:1 106. Bosellini, A. and Hardie, A. L., 1973. Depositional theme of a marginal marine evaporite. Sedimentology, 20:5 27. Bottjer, D. J.. Droser, M. L. and Wang. C.-Y., 1988. Fine scale resolution of mass extinction events: trace fl)ssil evidence from the Permian-Triassic boundary in South China. Geol. Soc. Am. Abstr. with Prog., 20: AI06. Boyd, D. W. and Maughan, E. K., 1973. P e r m i a n Triassic boundary in the middle Rocky Mountains. 111: A. Logan and L. V. Hills (Editors), The Permian and Triassic Systems and their Mutual Boundary. Mem. Can. Soc. Pet. Geol., 2: 294 317. Brandner, R., 1988. The Permian-Triassic boundary in the Dolomites (Southern Alps, Italy), San Antonio section. In: Rare Events in Geology, IGCP Project I99, Vienna, Abstr. Vol., pp.49- 56. Broglio Loriga, C.. Conti, M. A., Farabegoli, E., Fontana, D., Mariotti, N., Massari, F., Neri, C., Nicosia, I3., Pasini, M., Perri, M. C., Pittau, P., Posenato, R., Venturini, C. and Viel, G., 1986a. Upper Permian and P:T boundary in the arca between Carnia and the Adige Valley. In: Permian and Permian Triassic Boundary in the South-Alpine Segment of the Western Tethys. IGCP Project 203, Brescia, Excursion Guidebook, pp.23 28. Broglio Loriga. C., Fontana, D., Massari, F., Neri, C., Pasini, M. and Posenato, R., 1986b. The Upper Permian sequence and the P:T boundary in the Sass de Putia Mr. (Dolomites). In: Permian and Permian Triassic boundary in the South-

44 Alpine Segment of the Western Tethys. IGCP Project 203, Brescia, Excursion Guidebook, pp.73-90. Broglio Loriga, C., Neri, C. and Posenato, R., 1986c. The early macrofaunas of the Werfen Formation and the PermianTriassic boundary in the Dolomites (Southern Alps, Italy). Stud. Trent. Sci. Nat., Acta Geol., 62: 3-18. Broglio Loriga, C., Neri, C. and Posenato, R., 1986d. The Werfen Formation (Lower Triassic) in the Costabella Mt., Uomo Section. In: Permian and Permian-Triassic Boundary in the South-Alpine Segment of the Western Tethys. IGCP Project 203, Brescia, Excursion Guidebook, pp. 116-133. Broglio Loriga, C., Neri, C., Pasini, M. and Posenato, R., 1988. Marine fossil assemblages from Upper Permian to Lowermost Triassic in the western Dolomites (Italy). Mem. Soc. Geol. Ital., 34:5 44. Buggisch, W. and No~, S., 1988. Upper Permian and PermianTriassic boundary of the Carnia (Bellerophon Formation, Tesero Horizon, northern Italy). Mem. Soc. Geol. ltal., 34: 91-106. Caputo, M. V. and Crowell, J. C., 1985. Migration of glacial centres across Gondwana during the Paleozoic era. Bull. Geol. Soc. Am., 96: 1020-1036. Carr, T. R. and Paull, R. K., 1983. Early Triassic stratigraphy and paleogeography of the Cordilleran miogeocline. In: E. D. Dolly, M. W. Renolds and D. R. Spearing (Editors), Mesozoic pateogeography of the west-central United States. In: Soc. Econ. Paleontol. Mineral. Rocky Mountain Sect., Paleogeogr. Symp., 2: 39-55. Carulli, G. B., Radrizzani, C. P. and Ponton, M., 1988. The Permian Triassic boundary in the Paularo area (Carnia). Mere. Soc. Geol. Ital., 34: 107-120. Ciriacks, K. W., 1963. Permian and Eotriassic bivalves of the Middle Rockies. Bull. Am. Mus. Nat. Hist., 125: I 100. Claypool, G. E., Holser, W. T., Kaplan, I. R., Sakai, H. and Zak, I., 1980. The age curves of sulfur and oxygen isotopes in marine sulfate, and their mutual interpretation. Chem. Geol., 28: 199-259. Collinson, J. W. and Hasenmueller, W. A., 1978. Early Triassic paleogeography and biostratigraphy of the Cordilleran miogeosyncline. In: D. G. Howell and K. A. McDougall (Editors), Mesozoic Paleogeography of the Western United States. Soc. Econ. Paleontol. Mineral. Pacific Sect., Pacific Coast Paleogeogr. Symp., 2:175 187. Collinson, J. W., Kendall, C. G. and Marcantel, J. B., 1976. Permian-Triassic boundary in eastern Nevada and westcentral Utah. Bull. Geol. Soc. Am., 87: 821-824. Conti, M. A., Fontana, D., Mariotti, N., Massari, F., Neri, C., Nicosia, U., Pasini, M. and Pittau, D., 1986. The Bletterbach-Butterloch Section (Val Gardena Sandstone and Bellerophon Formation). In: Permian and Permian Triassic Boundary in the South-Alpine Segment of the Western Tethys. IGCP 203, Brescia, Excursion Guidebook, pp.91-110. Ekdale, A. A. and Mason, T. R., 1988. Characteristic tracefossil associations in oxygen-poor sedimentary environments. Geology, 16: 720-723. Elder, W. P., 1987. The paleoecology of the Cenomanian Turonian (Cretaceous) stage boundary extinctions at Black Mesa, Arizona. Palaios, 2: 24-40.

P. B. WIGNALL AND A. HALLAM

Elderfield, H., 1988. The oceanic chemistry of the rare-earth elements. Philos. Trans. R. Soc. Lond., A325: 105-126. Erwin, D. H., 1990. The end-Permian mass extinction. Ann. Rev. Ecol. Syst., 21: 69-91. Fischer, A. G., 1964. Brackish oceans as a cause of the PermoTriassic marine faunal crisis. In: A. E. M. Nairn (Editor), Problems in Palaeoclimatology. Interscience, London, pp.566-574. Flfigel, E. and Reinhardt, J., 1989. Uppermost Permian reefs in Skyros (Greece) and Sichuan (China): Implications for the Late Permian extinction event. Palaios, 4:502 518. Fliigel, E., Reinhardt, J. and Senowbari-Daryan, B., 1988. Reef biotopes near the Permian/Triassic boundary (Skyros, Greece and Sichuan, China) - - environmental trends in extinction? In: Rare Events in Geology, IGCP Project 199, Vienna, Abstr. Vol., p.6. Garrels, R. M. and Lerman, A., 1981. Phanerozoic cycles of sedimentary carbon and sulfur. Proc. Natl. Acad. Sci. USA, 78: 4652-4655. Goodwin, P. W. and Anderson, E .J., 1985. Punctuated aggradational cycles: General hypothesis of episodic stratigraphic accumulation. J. Geol., 93: 515-533, Hallam, A., 1987. Radiations and extinctions in relation to environmental change in the marine Lower Jurassic of north west Europe. Paleobiology, 13: 152-168. Hallam, A., 1988. A reevaluation of Jurassic eustasy in the light of new data and the revised Exxon curve. In: C. K. Wilgus et al. (Editors), Sea-level Changes: an Integrated Approach. Soc. Econ. Paleontol. Mineral. Spec. Publ., 42: 261 273. Hallam, A., 1989. The case for sea-level change as a dominant causal factor in mass extinction of marine invertebrates. Philos. Trans. R. Soc. Lond., B325: 437-455. Hallam, A., 1991. Why was there a delayed radiation after the end-Palaeozoic extinction? Hist. Biol., 5:257 262. Haq, B. U., Hardenbol, J. and Vail, P. R., 1988. Mesozoic and Cenozoic chronostratigraphy and cycles of sea-level change. In: C. K. Wilgus et al. (Editors), Sea-level Changes: an Integrated Approach. Soc. Econ. Paleontol. Mineral. Spec. PUN., 42: 71-108. Herak, M., Kochansky-Devid+, V. and Gugid, I., 1977. The development of the dascyclad algae through the ages. In: E. Flfigel (Editor), Fossil Algae: Recent Results and Developments. Springer, Berlin, pp.143-154. Hoffman, A., Gruszczynski, M. and Malkowski, K., 1991. On the interrelationship between temporal trends in 613C, 6180, and 634S in the world ocean. J. Geol., 99: 355-370. Holser, W. T. and Magaritz, M., 1987. Events near the Permian Triassic boundary. Mod. Geol., 11: 155-180. Holser, W. T., Magaritz, M. and Wright, J., 1986. Chemical and isotopic variations in the world ocean during Phanerozoic time. In: O. Walliser (Editor), Global Bio-events. Lecture Notes Earth Sci., 30: 63-74. Holser, W. T., Schonlaub, H.-P., Attrep, M., Boeckelmann, K., Klein, P., Magaritz, M., Orth, C. J., Fenninger, A., Jenny, C., Kralik, M., Mauritsch, H., Pak, E., Schramm, J.M., Stattgegger, K. and Schmoller, R., 1989. A unique geochemical record at the Permian/Triassic boundary. Nature, 337: 39-44.

ANOXIA AS A CAUSE OF THE PERMIAN/TRIASSIC MASS EXTINCTION

Holser, W. T., Schonlaub, H. P., Boeckelmann, K., Magaritz, M. and Orth, C. J., 1991. The Permian-Triassic of the Gartnerkofel-1 core (Carnic Alps, Austria): Synthesis and conclusions. Abh. Geol. Bundesanst., 45: 213-232. Jablonski, D., 1985. Marine regressions and mass extinctions: A test using modern biota. In: J. W. Valentine (Editor), Phanerozoic Diversity Patterns. Princeton Univ. Press, Princeton, pp.335-354. Johnson, J. G., 1974. Extinction of perched faunas. Geology, 2:479 482. Kozur, It., 1991. Permian deep-water ostracods from Sicily (Italy) Part 2: Biofacial evolution and remarks on the Silurian to Triassic paleopsychospheric ostracods. Geol. Palfiontol. Mitt. Innsbruck, 3:25 38. Kristan-Tollmam E., 1987. Triassic of Tethys and its relations with the Triassic of the Pacific Realm. In: K. G. McKenzie (Editor), Shallow Tethys 2. Proc. Int. Symp. Shallow Tethys 2, Wagga Wagga. Balkema, Rotterdam, pp.I69-186. Kummel, B., 1957. Paleoecology of Lower Triassic formations of southeastern Idaho and adjacent areas. Mere. Geol. Soc. Am., 67(3): 437 468. Kummel. B., 19713. Lower Triassic (Scythian) molluscs. In: A. Hallam (Editor), Atlas of Palaeobiogeography. Elsevier, Amsterdam, pp.225 233. Kummel, B. and Tcichert, C., 1970. Stratigraphy and paleontology of the Permian Triassic boundary beds, Salt Range and Trans-lndus Ranges, West Pakistan. Spec. PuN. Univ. Kans. Dep. Geol., 4:1 110. Loutit. T. S., Hardenbol, J., Vail, P. R. and Baum, G. R., 1988. Condensed sections: The key to age dating and correlation of continental margin sequences. In: C. K. Wilgus et al. (Editors), Sea-Level Changes: an Integrated Approach. Soc. Econ. Paleontol. Mineral. Spec. PUN., 42:183 213. Magaritz, M., Bar, R., Baud, A. and Holser, W. T., 1988. The carbon-isotope shift at the Permian/Triassic boundary in the Southern Alps is gradual. Nature, 331:337 339. Maughan, E. K., 1979. Petroleum source rock evaluation of the Permian Park City Group in the north-eastern Great Basin, Utah, Nevada, and Idaho. In: G. W. Newman and H. D. Goode (Editors), Basin and Range Symp., Rocky Mountain Assoc. Geol. and Utah Geol. Assoc., pp.523 530. Maxwell, W. D., i989. The end Permian mass extinction. In: S. K. Donovan (Editor), Mass Extinctions: Processes and Evidence. Belhaven Press, London, pp. 152-173. Nakazawa, K., 1981. Analysis of late Permian and early Triassic faunas of Kashmir. In: K. Nakazawa and H. M. Kapoor (Editors), The Upper Permian and Lower Triassic Faunas of Kashmir. Mem. Geol. Surv. India, pp.193-204. Nakazawa, K. and Runnegar, B., 1973. The Permian-Triassic boundary: a crisis for bivalves? In: A. Logan and L. V. Hills (Editors), The Permian and Triassic systems and their mutual boundary. Mere. Can. Soc. Pet. Geol., 2: 608-621. Neri, C.. Pasini, M. and Posenato, R., 1986. The PermianTriassic boundary and the early Scythian sequence - - Tesero Section, Dolomites. In: Permian and Permian Triassic boundary in the south-Alpine segment of the Western Tethys. 1GCP 203, Brescia, Excursion Guidebook, pp. 111-116.

45

Newell, N. D., 1988. The Paleozoic/Mesozoic erathem boundary.' Mem. Soc. Geol. Ital., 34:303 311. Newetl; N. D. and Kummel, B., 1942. Lower Eo-Triassic stra'tigraphy, western Wyoming and southeast Idaho. Bull. Geol. Soc. Am., 53: 937-996. No6, S. U., 1987. Facies and paleogeography of the marine Upper Permian and of the Permian-Triassic boundary in the southern Alps (Bellerophon Formation, Tesero Horizon). Facies, 16:89 142. Noe, S. U., 1988. The Permian-Triassic boundary in the Southern Alps - - A study of foraminiferal evolution. In: Rare Events in Geology, IGCP Project 199, Vienna, Abstr. Vol., p.19. Paull, R. A. and Paull, R. K., 1987. Lower Triassic rocks within and adjacent to the thrust belt of Wyoming, Idaho, and Utah. In: Wyo. Geol. Assoc. Guidebook, 38th Field Conf.,pp.149 161. Paull, R K. and Paull, R. A., 1986. Epilogue for the Permian a retrospective view from the Triassic in western Cordillera. Univ. Wyo. Contrib. Geol., 24: 243- 252. Paull, R. K., Paull, R. A. and Anderson, A. L., 1985. Conodont biostratigraphy and depositional history of the Lower Triassic Dinwoody Formation in the Meade Plate, southeastern Idaho. Utah Geol. Assoc. Publ., I4:55 65. Pcrri, M. C. and Andraghetti, M., 1987. Permian Triassic and Early Triassic conodonts form the southern Alps, Italy. Riv. ltal. Paleontol. Stratigr., 93:291 ~328. Posamentier, H. W. and Vail, P. R., 1988. Eustatic controls on elastic deposition II Sequence and systems tract models. In: C. K. Wilgus et al. (Editors), Sea-level Changes: an Integrated Approach. Soc. Econ. Paleontol. Mineral. Spec. Publ., 42:109 124. RamovL A., 1986. Marine development of the Upper Permian in the type Karavanke Mountains and the Scythian. In: Permian and Permian Triassic Boundary in the SouthAlpine Segment of the Western Tethys. IGCP 203, Brescia, Excursion Guidebook, pp.43 51. Raup, D. M., 1979. Size of the Permo-Triassic bottleneck and its evolutionary implications. Science, 206:217 218. Sanders, H. L., 1968. Marine benthic diversity: a comparative study. Am. Nat., 102: 243-282. Savrda, C. E. and Bottjer, D. J., 1987. The exaerobic zone, a new oxygen-deficient marine biofacies. Nature, 327: 54-56. Sch6nlaub, H. P., I99 I. The Permian:Triassic of the Gartnerkofel-1 core (Carnie Alps, Austria): Conodont biostratigraphy. Abh. Geol. Bundesanst., 45:79 98. SchopL T. J. M., 1974. Permo-Triassic extinctions: relation to sea floor spreading. J. Geol., 82:129 143. Signor, P. W. and Lipps, J. H., 1982. Sampling bias, gradual extinction patterns, and catastrophes in the fossil record. In: L. T. Silver and P. H. Schulz (Editors), Geological implications of large asteroids and comets on the Earth. Geol. Soc. Am. Spec. Publ., 190: 291-296. Stanley, S. M., 1984. Marine mass extinctions: a dominant role for temperature. In: M. H. Nitecki (Editor), Extinctions. Chicago Univ. Press, pp.69 117. Stanley, S. M., 1988. Climatic cooling and mass extinctions of Paleozoic reef communities. Palaios, 3:228 232.

46 Teichert, C., 1990. The Permian-Triassic boundary revisited. In: E. G. Kauffman and O. H. Walliser (Editors), Extinction Events in Earth History. Lect. Notes Earth Sci., 30: 199-238. Thomas, H. D., 1934. Phosphoria and Dinwoody tongue s in lower Chugwater of central and south-eastern Wyoming. Bull. Am. Assoc. Pet. Geol., 18: 1655-1697. Wignall, P. B., 1990a. Benthic palaeoecology of the Late Jurassic Kimmeridge Clay of England. Spec. Pap. Palaeontol., 43. Wignall, P. B., 1990b. Observations on the evolution and classification of dysaerobic communities. In: W. Miller (Editor), Paleocommunity temporal dynamics: the long-term development of multispecies assemblies. Paleontol. Soc. Spec. Publ., 5: 99-111. Wray, W. D., 1977. Late Paleozoic calcareous red algae. In: E. Flugel (Editor), Fossil Algae: Recent Results and Developments. Springer, Berlin, pp.167-176.

P. B. WIGNALLAND A. HALLAM Wright, J., 1989. REE in fossil apatite record Phanerozoic OAE. Int. Geol. Congr. Abstr. Vol., Washington, DC, pp.383-384. Wright, J., Schrader, H. and Holser, W. T., 1987. Paleoredox variations in ancient oceans recorded by rare earth elements in fossil apatite. Geochim. Cosmochim. Acta, 51: 631-644. Wright, J., Seymour, R. S. and Shaw, H. F., 1984. REE and Nd isotopes in conodont apatite: variations with geological age and depositional environment. In: D. L. Clark (Editor), Conodont Biofacies and Provincialism. Geol. Soc. Am. Spec. Pap., 196: 325-340. Yin, H., 1985. Bivalves near the Permian-Triassic boundary in South China. J. Paleontol., 59: 572-600. Yin, H., Yang, F. Q., Zhang, K. and Yang, W. P., 1988. A proposal to the biostratigraphic criterion of Permian/Triassic boundary. Mere. Soc. Geol. Ital., 34: 329-344.