Triassic–Jurassic boundary events: Problems, progress, possibilities

Full text

Editorial
Triassic–Jurassic boundary events: Problems, progress, possibilities
1. Problems
As for most geological period boundaries, the
Triassic–Jurassic (T–J) transition, ∼200 million years
ago, was a critical juncture in Earth history during which
profound biotic and environmental changes took place.
Early comparisons with the end-Cretaceous extinction
and the involvement of extraterrestrial impact have now
largely, although not entirely, given way to more Earth-
bound explanations of events. At the T–J boundary the
supercontinent Pangaea, which had dominated the
palaeogeographic face of the Earth for the previous
∼100 million years, began a fragmentation that has
lasted through to the present day. The most obvious
manifestation of this process was the production of an
estimated two and a half million cubic kilometres of
magma with a focus at the centre of Pangaea, and now
known as the Central Atlantic Magmatic Province, or
CAMP. At more-or-less the same time profound
changes took place in the key elements of the biosphere,
most notably and obviously in the marine carbonate
producing organisms, including those upon which we
rely for precise stratigraphic correlation such as
ammonites. The case for a dominant volcanic deus ex
machina now looks incontestable, even if the origin of
the volcanism and the precise mechanisms by which
environmental changes were driven require much
further explanation.
Details of timing are crucial for understanding cause
and effect relationships in Earth history, and the lack of a
reliable and widely applicable biostratigraphic frame-
work has greatly hampered our understanding of T–J
events. It is also plainly the case that in order to
reconstruct past events, a physical record of their
passing is essential. Here again the Triassic–Jurassic
boundary has proved problematic because complete
marine sedimentary successions are both few and not
very far apart, an observation that has strongly
suggested unusually low global sea levels. The relative
lack of good marine successions has also delayed the
definition of the boundary and the selection of a global
stratotype section and point (GSSP); at the time of
compilation of this collection of papers decisions had
not been made.
In order to facilitate advances in these major issues,
IGCP Project 458 was set up in 2001 under the
leadership of the editors of this special issue. The
project was conceived as multi-disciplinary with the aim
of integrating palaeontological, stratigraphical, sedi-
mentological, geochemical, geochronological, palaeo-
magnetic and mineralogical data from T–J boundary
sections globally. Amongst the principal activities we
anticipated were: field studies directed towards previ-
ously known localities as well as recently or newly
discovered ones; compilation of global databases with
improved and revised taxonomy, biochronology and
palaeobiogeography of major fossil groups, and analysis
of patterns of the end-Triassic extinction and Early
Jurassic recovery; new radiometric ages and high reso-
lution biostratigraphic correlation to establish a reliable
temporal framework; assessment of environmental
perturbations and their role in different extinction
scenarios using geochemical proxy methods; further
studies of the Central Atlantic Magmatic Province and
the search for a hypothetical end-Triassic impact to
provide clues to the trigger of global environmental
change. The overarching view was that reconstruction
of the end-Triassic events would use an Earth systems
approach to integrate all new findings into the most
plausible models.
The papers collected in the present volume individ-
ually touch upon many of the areas of study anticipated
for IGCP project 458. For convenience we have grouped
the papers into four main thematic sections, whilst
recognizing that many of them span several of these
topics. Some of the most important results in terms of
relative timing of events around at boundary are
summarized in Fig. 1.
Palaeogeography, Palaeoclimatology, Palaeoecology 244 (2007) 1–10
www.elsevier.com/locate/palaeo
0031-0182/$ - see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.palaeo.2006.06.020

Fig. 1. Relative timing of major events around the T–J boundary, as observed in key sections discussed in this volume. Successions are correlated on the basis of: 1) carbon isotope stratigraphy; 2)
ammonite biostratigraphy; 3) radiolarian biostratigraphy, and; 4) magnetostratigraphy. QCI= Queen Charlotte Islands; OM = Orange Mountain; LU = Lower Unit. Comments on P. tilmanni from von
Hillebrandt (pers comm.).
2Editorial

2. Progress
2.1. The stratigraphic record
The first set of six papers present syntheses of the
record of T–J boundary events, most in broadly Tethyan
locations. Relatively deep-water settings provide the best
opportunities for determination of the sequence of
stratigraphic events across the T–J boundary. The
carbonate succession at Csővár, Hungary, was deposited
in an intra-platform basin that exhibits relatively constant
sedimentation through the boundary interval. The
section has previously yielded evidence of a negative
carbon-isotope anomaly in both bulk carbonate and
organic matter co-incident with the palaeontologically
defined boundary —as far as this could be identified on
the basis of scarce ammonites. In a new study, Pálfy et al.
present a truly integrated stratigraphic dataset. The new
data lead to important negative and positive findings.
Included amongst the most important observations is the
rare occurrence of the conodont “Neohindeodella”detrei
3 m above the first definitively Jurassic psiloceratid
ammonite. Although it is difficult to categorically rule
out reworking (redeposited beds definitely occur in the
succession), the evidence supports the idea that the last
conodonts finally went extinct in the earliest Jurassic.
Other sedimentary and palaeontological parameters
show little change —for example the late Rhaetian
clay mineral and foraminiferal assemblages are very
similar to those in the early Hettangian. New stable
isotope data, carefully screened for diagenetic effects,
are also used to suggest that the principal isotope
excursion recognized in the succession contains hitherto
unrecognized high frequency structure and also pre-
serves a record of significant water mass warming.
Sedimentary archives of three basins from the
northern, central and southern Apennines, the La Spezia,
the Mt. Camicia, and Lagonegro basins, provide a rich
source of information for reconstructing Late Triassic
palaeoenvironments and the palaeogeographic evolution
of western Tethyan areas. Ciarapica argues that these
continuous successions of basinal facies are of particular
value, as they also reflect coeval evolution of adjacent
platforms through occurrence of platform-derived com-
ponents. Evidence is inferred for a Late Norian platform
drowning, climate change from arid to humid conditions,
a spread of dysaerobic facies and increasing eutrophiza-
tion. The establishment of oligotypic benthic communi-
ties, for example suggested by foraminiferan
associations dominated by Triasina hantkeni, is inter-
preted as a biotic response and a first step in the end-
Triassic extinction. Somewhat at odds with observations
from elsewhere, the T–J boundary appears to mark only
a second, lesser step of the end-Triassic events. At the T–
J boundary the disappearance of the stress-tolerant asso-
ciations, return of hot and arid climate, and a final, short
anoxic episode, followed by rapid resumption of carbon-
ate platform building are observed in the Apennines.
By way of contrast, the Southern Alps of Lombardy,
Italy, preserve a T–J transition recorded in a predom-
inantly carbonate shelf and ramp setting. Galli et al.
analyze the sedimentary history, faunal and microfloral
assemblages, and stable isotope evolution of the
boundary interval. The focus of their attention is the
newly proposed Malanotte Formation, a conspicuous,
thin-bedded, micritic limestone unit that occurs between
the fossilifererous, more shallow-water carbonates of
the Rhaetian Zu Limestone, and the Hettangian Con-
chodon Dolomite (cf. Galli et al., 2005). The T–J
boundary is drawn near the base of the transgressive
Malanotte Formation, on the basis of a gradual change
in the pollen assemblages. More pronounced is the
slightly earlier abrupt extinction of diverse micro- and
macrofaunal associations at the top of the Zu Limestone,
a level that is also inferred to represent platform
drowning. The T–J boundary is closely correlated
with lithological change, accentuated by a gap inferred
from an Fe-crusted hardground or a thin layer rich in
siliciclastic components. The Malanotte Formation is
largely devoid of micro- or macrofauna and lacks the
recognizable lithological cycles that characterize the
underlying Zu Formation, but it does reveal a three-
stage evolution of the carbon isotope ratio in sea water.
In the base, i.e. at the T–J boundary, a moderate
negative excursion is recorded in bulk organic carbon,
followed by a rebound to positive values, which is in
turn followed by a more modest negative shift.
Another area with previously less well-known
Tethyan T–J boundary sections is the High Tatra Mts in
the Western Carpathians, at the Slovak–Polish border.
There, the intra-shelf Zliechov Basin, broadly similar to
the more familiar Alpine Kössen basins, preservea record
of T–J transition studied in several sections. Michalík et
al. present new data from four key sections that were
subject to multidisciplinary investigation. Carbonate
deposition of the Fatra Formation, consisting of numer-
ous shallowing-upward cycles, was abruptly terminated
at the erosional T–J boundary. The conspicuous
‘boundary shale' of the overlying Kopienec Formation
is suggested to reflect both a carbonate production crisis
and a sudden increase of riverine influx of terrigenous
fine siliciclastic sediment, conceivably related to global
changes in climate and ocean chemistry. The moderately
diverse Rhaetian biota of the Fatra Formation, largely
3Editorial

inferred from skeletal components in the microfacies,
disappears at the boundary. A turnoveris observed among
the foraminifera, which here provide the primary
biostratigraphic framework. The earliest Hettangian
associations in the Kopienec Formation are dominated
by stress tolerant ostracods, which also suggest eutrophic
conditions. Immediately below the boundary, a negative
carbon isotope anomaly is recorded from bulk carbonate,
although it is definitely modest in comparison to some
other reported T–J boundary anomalies. Even more
ambiguous is the presence of microspheres in limestone
beds not far below the boundary. Although they
approximately correlate with the level of extinction and
isotope anomaly, their origin has not been convincingly
demonstrated. However tempting it is to infer impact-
related spherules, a more mundane explanation is that
they are diagenetic or hydrothermal alteration products of
spherical primary sedimentary particles, e.g. ooids.
Depositional environments in the northeastern part of
the Iberian Peninsula were quite different from those in
the fully Tethyan areas around the T–J boundary. Gómez
et al. summarize a wealth of stratigraphic, sedimentolog-
ical, and palynological information and report new
geochemical data. Eustatic sea-level changes variably
covered the low-relief area with extensive, extremely
shallow carbonate platforms and coastal playa and sabkha
flats, leaving a stratigraphic record of mixed carbonates
and evaporites arranged in sedimentary cycles. Contrary
to earlier opinions, Gómez et al. find no evidence for any
major sea-level change or unconformity near or at the T–J
boundary, and emphasize the remarkable lateral continu-
ity of the latest Triassic and earliest Jurassic strata.
Asturias is the only area of Iberia with a slightly different
record: the T–J boundary is placed in a carbonate
sequence there, whereas elsewhere it falls within an
evaporitic unit. The carbonates in Asturias yield an
organic carbon isotope record and add a new entry to the
growing list of locations where the T–J boundary negative
δ
13
C anomaly is recognized. The boundary is drawn with
varying precision on the basis of palynology, and the
distribution of palynomorphs is also used to infer a
climate and vegetation history. A moderate latest Rhaetian
plant extinction is followed by considerable diversifica-
tion in the earliest Hettangian, which appears related to a
shift from arid to warmer and more humid climate
conditions, as reflected by a change in dominance of
xerophytic to hygrophytic pollen producers.
The continental record of the T–J events is no less
important than the marine record. Tanner and Lucas
provide a significant re-interpretation of the facies and
stratigraphic relationships among the upper part of the
Chinle Group and the lower part of the Glen Canyon
Group of Utah and Arizona, which were situated near
the western margin of Pangaea at the time. Their
discussion centres on the development of erg deposits
within the Wingate Formation, which initiated in the
latest Triassic, and which were perhaps fuelled by
exposed shoreline sands during a Rhaetian lowstand in
sea-level. Their main conclusion is that the mosaic of
continental facies is best deciphered in the context of a
north–south palaeogeographic transition from domi-
nantly fluvial–lacustrine environments in the north, to
dominantly aeolian palaeoenvironments in the south.
Reinterpretation of formation boundaries results in the
recognition of more than one regional unconformity
between demonstrable Triassic strata of the Chinle
Group and the Jurassic Glen Canyon Group. This has
significant implications regarding a precise placement of
aT–J boundary in these often poorly fossiliferous rocks.
2.2. Biotic change
There are many examples of major environmental
change events in the Phanerozoic that are characterized
by the flood abundances of opportunistic, or ‘disaster’
taxa, but their presence has not hitherto been highlighted
for the T–J boundary event. Van de Schootbrugge et al.
examine the stratigraphic micropalaeontology of the
candidate GSSP section at St Audrie's Bay, England, and
quantify changes in both organic walled and calcareous
microfossils at the start of the ‘main’negative isotope
excursion (i.e. the long duration shift to light carbon
isotope values that occurs at St Audrie's Bay about 2 m
below the lowest examples of the Jurassic ammonite
Psiloceras). In the St Audrie's Bay section it is shown
that members of the green algae –prasinophytes and
acritarchs –become particularly abundant at the onset of
the ‘main’negative excursion; at the same time, red algae
and calcareous nannoplankton are minor constituents of
the microflora. The observations are interpreted by Van
de Schootbrugge et al. to represent an ecosystem
response to raised atmospheric CO
2
. Isotope and
elemental data from the oyster Liostrea hisingeri,
collected through the same interval, provide valuable
indications of parallel changes in major environmental
variables. These data incidentally provide the first
convincing evidence that the ‘main’T–J carbon isotopic
curve based on bulk organic matter is present in marine
carbonate as well, albeit with half the amplitude (in
common with other Mesozoic excursions). Oxygen
isotopes and Mg/Ca ratios from the oysters are used to
argue for a 4 °C sea-floor temperature increase, and a
parallel decrease in salinity by at least 3 PSU at the start
of the ‘main’negative isotope excursion.
4Editorial

The Queen Charlotte Islands of northwestern Canada
continue to provide rich palaeontological data, central
to our understanding of the T–J extinction. Longridge
et al. add new data on the ammonoid and radiolarian
diversity trends and biochronology of two important
T–J boundary sections in the Queen Charlotte Islands.
Of these sections, the one on Kunga Island, has pre-
viously provided the sole radiometric age estimate for
the T–J boundary in marine rocks, and the other, at
Kennecott Point, has yielded one of the best carbon
isotope records spanning the extinction interval. Long-
ridge et al. document a moderately diverse ammonoid
succession across the boundary, including new discov-
eries that significantly reduce the ammonoid ‘gap’of
the boundary interval and now permit correlation to
early Hettangian ammonoid zones recognized at New
York Canyon, Nevada. Not to be overlooked is the
excellent radiolarian record from this interval in which
Longridge et al. describe a profound decrease in not only
radiolarian diversity, but also morphologic complexity
amongst earliest Jurassic spumellarian and entactiniid
taxa.
To measure a mass extinction only by the proportion
of lost taxa and change in diversity is an oversimplifi-
cation. The ecologic impact may be equally significant
and can be estimated by assessing the reorganization of
communities. The compositional change of brachiopod
communities across the T–J boundary in the Northern
Calcareous Alps is investigated by Tomašových and
Siblík. Using an array of multivariate analytical
techniques, they demonstrate the profound effects
among brachiopods during the T–J boundary extinction.
The turnover at the boundary is an order of magnitude
higher than within the Rhaetian and the Hettangian.
Contrary to some earlier suggestions, the T–J brachio-
pod extinction is abrupt, with no indication of any
protracted decline during the Rhaetian. Removal of the
incumbents, i.e. extinction of superfamilies with dom-
inant members in latest Triassic communities, led to a
fundamental reorganization of community structure.
Testing for two competing hypotheses, Tomašových
and Siblík find more support for true compositional
change across the T–J boundary than they do for a
previous proposal of changing habitat preference within
major brachiopod groups. Brachiopods are rare, but not
absent, in the earliest Hettangian survival phase. Their
recovery was underway by late early Hettangian to mid
Hettangian times, as indicated by newly established
communities with an increasing degree of between-
habitat differentiation.
An alternative way to analyze extinction character-
istics is to interrogate a global database. This approach
has the advantage of providing an overview, with the
disadvantage of reduced stratigraphic resolution. Kies-
sling et al. use the Paleobiology Database to analyze
abundance and diversity patterns of marine benthic
organisms (sponges, corals, bivalves, gastropods and
brachiopods) from the Middle Triassic (∼240 Ma ago)
to the Middle Jurassic (∼160 Ma ago), paying particular
attention to possible biases in the dataset. Their analysis
confirms the reality of the T–J mass extinction, but it
also throws up some evidence for selectivity for certain
groups. Taxa that were reef-dwelling, with an inshore
habitat preference, preferring carbonate substrates, and
confined to low latitudes, exhibit higher extinction risk
than other groups. Intriguingly, the same characteristics
seem also to apply to background extinctions, lending
weight to the idea that the T–J extinction represents an
intensification of background processes with, perhaps,
an emphasis on extinctions in reefs and inshore
environments during (or at the end of) the Rhaetian.
Where body fossils are absent, trace fossils might
provide crucial additional information about extinction
patterns. An analysis of the T–J boundary trace fossil
record is provided by Barras and Twitchett for three sites
in southern England, including the candidate GSSP at St
Audrie's Bay. This contribution provides a detailed
account of changing ichnofauna of an interval from the
upper Langport Member of the Lilstock Formation
through five Jurassic ammonoid zones of the Blue Lias
Formation (culminating in the semicostatum Zone).
Their data reveal how eight ichnogenera show signifi-
cant patterns of infaunal changes through the interval.
Above a moderately diverse ichnofossil assemblage in
the Langport Member is a notable gap in trace fossils in
the ‘Pre-Planorbis Beds’. The authors do not relate this
absence of ichnotaxa directly to CAMP effects, because
of perceived differences in timing, but instead point up a
role for marine anoxia. The focus of their study is the
Early Jurassic recovery interval, rather than the lead up to
the extinction. The recovery amongst ichnotaxa above
the Pre-Planorbis Beds documents a significant increase
in ichnotaxic diversity and an increase in the depth of
burrowing.
Complementary to the well-known continental
sequences in eastern North America are those of the
western United States: the vast outcrops of fluvial,
aeolian, and lacustrine sedimentary rocks of the Chinle
and Glen Canyon groups. In a companion paper to their
stratigraphic account, Lucas and Tanner document what
is perhaps the best known terrestrial vertebrate record
spanning the T–J boundary, including reptilian skeletal
remains as well as their traces. They provide a revised
biochronology for the interval and subdivide the Late
5Editorial

Triassic and Early Jurassic strata into five biochrons
based upon the first appearance of reptile taxa. A
significant finding is an increase in both the abundance
and size of dinosaurian ichnotaxa leading up to the T–J
boundary. This event corresponds to the loss of
cruotarsan and phytosaur reptiles and the footprint
ichnogenus Brachychirotherium.
2.3. Carbon-isotope stratigraphy
The precise stratigraphic relationship between bios-
tratigraphically important fossil groups and carbon–iso-
tope compositions of carbonate and organic sedimentary
matter has become critical to understanding T–J events,
as emphasized in Fig. 1. In an integrated palynological
and isotopic study of the classic boundary sections of the
Salzkammergut, Austria, Kürschner et al. provide an-
swers to several outstanding questions of correlation. By
constructing a composite carbon isotope curve of bulk
organic matter from two nearby sections, they find the
now increasingly replicated pattern of an abrupt ‘initial’
negative isotope excursion, closely followed by an
extended ‘main’isotope excursion (Hesselbo et al.,
2002, 2004). The initial isotope excursion occurs
immediately above the top of the hemipelagic carbonate
Kössen Formation, in the lowest few centimetres of the
‘Grenzmergel’(or boundary marl), and it had been
missed in a previous isotopic study of an adjacent section
due to relatively wide sample spacing. The negative
excursion is coincident with the highest occurrence of
conodonts, and the succeeding 1–2 m sees the highest
occurrences of typically Triassic palynomorphs. The
start of the ‘main’isotope excursion occurs at the same
level as the lowest occurrence of Cerebropollenites
thiergartii, a pollen grain that has previously been
suggested as a base-Jurassic marker. Whatever taxon is
adopted as a definitive guide for the T–J boundary, it
becomes clear that the principal period of environmental
change takes place within the Grenzmergel and is
bracketed by the two negative isotope excursions.
Interestingly, like Van de Schootbrugge et al., Kuersch-
ner et al. also recognize the occurrence of a green-algal
bloom, but in this case at the same time as the initial
negative excursion.
The candidate GSSP at Muller Canyon, Nevada,
USA, is another crucial section that reveals the
relationship between the organic carbon–isotope curve
and biostratigraphically important taxa —in this case
ammonites and bivalves. In a re-sampling and re-
measuring exercise, Ward et al. reproduce the broad
characteristics of a previously published carbon–isotope
curve based on bulk marine organic matter (Guex et al.,
2004). However, they also find important contrasts with
the previous work. Most notably, Ward et al. recognize
that the lowest occurrence of the typically Jurassic
pectinacean bivalve Agerchlamys boellingi, and the
lowest find of the ammonite Psiloceras sp., occur
immediately above an ‘initial’negative isotope excur-
sion as defined by multiple data points. If the carbon–
isotope curve can be relied upon for correlation, which
looks increasingly likely, then the implication is that
base of the Jurassic as defined in North American
sections on any faunal criterion correlates to horizons
many believe to be Triassic in European sections.
In addition to yielding an important record of biotic
change across the T–J interval boundary the Queen
Charlotte Islands' succession in Canada was one of the
first to show evidence for an abrupt negative carbon–
isotope excursion coincident with biotic change, in this
case radiolarians. Williford et al. here present an extended
record of carbon isotope data from bulk organic matter
from the Hettangian succession at Kennecott Point in the
Queen Charlotte Islands (cf. Ward et al., 2004). A really
striking feature of their new data is the magnitude of a
positive excursion lying between an initial negative
excursion (corresponding closely to the level of radiolar-
ian turnover) and what they interpret as the main
(Hettangian) negative excursion. Explanations of the T–
J boundary record now have to include both a potential
source of isotopically light carbon to produce the negative
excursion and an explanation for where all the light
carbon subsequently goes. Williford et al. prefer a
scenario that involves principally a switch of carbon
burial flux from carbonate to organic matter.
2.4. Causes and consequences
Plate motions incessantly operate in the background
of all other Earth phenomena. The changing palaeogeo-
graphy around the T–J boundary is analyzed by
Golonka, on the basis of two global palaeogeographic
maps constructed for the Late Triassic and Early Jurassic,
respectively. More detailed lithofacies maps for the two
intervals are provided for crucial areas where the T–J
transition proved eventful, including the western Tethys,
eastern Tethys, Palaeotethys and eastern Asia, north-
western Laurasia, and western Gondwana. The closure
of Palaeotethys was expressed in the main convergent
event, the Indosinian orogeny, which completed the
assembly of eastern Pangaea. In the same time, rifting in
the future Central Atlantic area heralded the break-up of
the supercontinent. The changing palaeogeography is an
important backdrop to the T–J boundary events but most
tectonic phenomena operate at longer time scales. A
6Editorial

notable exception is the magmatism of the Central
Atlantic Magmatic Province (CAMP).
Indeed, flood basalt volcanism of the CAMP is
implicated in the currently most favoured scenario
explaining environmental changes and biotic extinctions
at the T–J boundary. Clearly, relative timing of the
boundary events and the eruptions, and the duration of
the latter, is of paramount importance in refining or
refuting the purported causal link. Two sister papers in
this volume contribute new radio-isotopic ages for
CAMP basalts and interpret their significance.
Vérati et al. present a suite of 12 new
40
Ar/
39
Ar ages
from Moroccan CAMP basalts complemented by
another two ages from correlative lava flows from
Portugal. In Morocco, the CAMP flows are grouped into
four units on the basis of their stratigraphy and
geochemical characteristics. The first three flow units
account for 90% of the total lava volume. Significantly,
their ages overlap within error, suggesting that the bulk
of volcanic activity occurred within a short time span, in
less (perhaps much less) than the 2 Ma resolution
afforded by the analytical uncertainty of the dating
method. The Moroccan ages are centered around a mean
of 199.1 ± 1 Ma ago. Only the fourth and volumetrically
minor flow unit has a resolvably youngest mean age of
196.6 Ma ago. The flow ages and chemical composi-
tions suggest that this unit is a product of late-stage
asthenospheric upwelling, representing a milestone in
the magmatic evolution of the Atlantic rifting process.
The Portuguese lava flows are demonstrably coeval with
their Moroccan counterparts and unquestionably can be
assigned to the CAMP. Significantly, the new suite of
ages presented here confirm the earlier suggestion that
CAMP volcanism is synchronous with the T–J
boundary. The caveat is a recognition of problems
associated with both the
40
Ar/
39
Ar method applied here
and the U–Pb method used to date the boundary from an
ash bed in a marine section.
Nomade et al. set out to address the same problems:
what is the chronology (i.e. age and duration) of CAMP
volcanism and, on the basis of the temporal relation-
ships, how is it related to the T–J boundary events? The
team also reports a set of new
40
Ar/
39
Ar ages from their
17 samples, split among three of the four continents
where CAMP occurs. This brings the total number of
published dates to over 100, making CAMP the
temporally best constrained large igneous province.
Despite chronologic reviews published as recently as in
2003 and 2004, a new effort is justified as some 50
new
40
Ar/
39
Ar dates were obtained in the last 3 years
alone. The ‘quality control’applied by Nomade et al. is
also more stringent than in previous studies. After
filtering out less reliable dates and those exhibiting
disturbed isotopic systems, only the most robust
plateau ages are considered further and 58 dates are
accepted as valid. It is reassuring that this much larger
dataset principally confirms and refines the conclusions
of earlier studies. The new synoptic chronology of
CAMP reveals that intrusive magmatism commenced
∼201 Ma ago, extrusions occurring about 1 Ma later in
the African margin, and followed soon after in North
America, before spreading to South America. Peak
activity, represented by ∼80% of the dates, is restricted
to a short period between 199 and 197.5 Ma ago. Small-
volume eruptions form a protracted tail-end of activity
to as late as ∼190 Ma ago. A pattern of north-to-south
migration of volcanism emerges, although geographic
distribution of the data is uneven with the strongest
representation of African (mostly Moroccan) samples.
The difference in timing of CAMP volcanism in
North America and in North Africa is a matter of some
considerable debate (e.g. Knight et al., 2004; Marzoli et
al., 2004). Whiteside et al. frame the questions in terms
of synchronism between Moroccan and North America
activity, and the age relationship to the major pulse of
extinction in continental settings, and they attempt to
answer these questions using a variety of stratigraphic
arguments. Additionally, they provide new cyclostrati-
graphic, lithostratigraphic, and biostratigraphic data
from several continental basins in eastern North
America and Morocco. Significant are the new data
from Partridge Island (Fundy Basin, Nova Scotia) and
the Argana Basin (Morocco), and revised sections
elsewhere in North America (e.g. Newark and Hartford
basins). On both continents, the authors define an end-
Triassic extinction event based primarily on palynology
and, to a lesser extent, on tetrapod footprint data. The
loss of pollen species and tetrapod ichnotaxa coincides
more or less with the onset of Corollina (i.e. Classo-
pollis) dominated pollen assemblages.
As previously reported, based on astrochronology,
the extinction event is proposed to predate the earliest
CAMP flow by ∼20 ka (e.g. Olsen et al., 2002).
Existing basalt geochemical data are used to support this
correlation, and Whiteside et al. note that the strati-
graphically lowest flows from North America are
geochemically High Titanium Quartz normative
(HTQ) basalts that are most similar to the HTQ-type
flows from the Argana Basalt in Morocco. However,
correlation of the North American basalts to the Central
High Atlas Basin in Morocco is problematic as these are
High Iron High Titanium Quartz normative (HFTQ)
basalts for which there are no real correlatives in North
America. Whiteside et al. propose that the HFTQ flows
7Editorial

of the central High Atlas Basin are part of a magmatic
sequence in which the HFTQ evolved from earlier HTQ
magmas. Additionally, they specifically dispute a
previous correlation of the short reverse magnetochron
recognized in Morocco which had implied that North
American flood basalts are younger than those found in
Morocco. Instead, they suggest that a short reverse
magnetochron may yet be found in poorly sampled
North America basalts above the palynologically
defined T–J boundary, and propose that an independent
test of their hypothesis would be recognition of the
‘initial’carbon–isotope negative excursion in strata
below the oldest basalts in these continental settings.
Ocean acidification, through the build up of dissolved
carbon dioxide in the oceans, has been an important
putative mechanism behind degradation of marine car-
bonate ecosystems for several past events (as well as at the
present day). This is particularly relevant for times when
carbonate platform drowning appears to have accelerated,
when extinctions take place preferentially within shallow
marine carbonate communities, and when carbonate
skeletal mineralogy seems to undergo significant change.
Berner and Beerling apply a numerical carbon cycle model
to investigate whether volcanic gases of direct magmatic
origin were sufficient in quantity to account for these
phenomena via oceanic carbonate undersaturation. In
addition to the role of carbon dioxide, they also examine
the part played by sulphur dioxide, and the possible relative
amounts of these two gases during basaltic volcanism,
together with feedback mechanisms that potentially
include release of methane from gas hydrates. Their
conclusions are simple; gasses directly produced from
CAMP volcanism can explain oceanic carbonate under-
saturation phenomena, but only just. It is necessary to have
starting conditions close to undersaturation (i.e. very high
atmospheric carbon dioxide) and release of amounts
volcanic gas at the very upper limits of plausibility.
CAMP is implicated not only in the generation of
excess atmospheric and oceanic carbon dioxide, but also
in its drawdown via carbonation reactions during
weathering. The seawater record of ‘signature’isotopes
such as strontium and osmium, which are biased towards
unradiogenic values in juvenile basalts, may give a clue
as to how CAMP affected weathering processes. Cohen
and Coe compile parallel Sr and Os isotope datasets from
across the T–J boundary and carry out a semi-
quantitative analysis of the results. They find that close
similarities exist between the Sr and Os isotope records
of the T–J boundary and those of the Toarcian Oceanic
Anoxic Event, some 17 million years later, which also
coincided with eruption of a continental flood basalt
Large Igneous Province (LIP), the Karoo–Ferrar.
Perturbations to the seawater Sr-isotope record coinci-
dent with LIP emplacement take the form of sudden
increases in the proportion of radiogenic strontium,
interpreted as increases in continental weathering rates
superimposed on an overall trend brought about by long-
term decreasing in
87
Sr/
86
Sr ratios, presumably reflect-
ing long-term decreasing continental weathering rates.
The seawater osmium isotope records for both events
also show abrupt changes to more radiogenic values,
albeit much more transient than for strontium. Thus, it
appears from the T–J boundary record that CAMP
eruptions initially promoted a large increase in conti-
nental weathering, without the lavas themselves being
strongly weathered and contributing a significant
unradiogenic flux (the same is also true for Karoo–
Ferrar). In the case of CAMP, the subsequent Os-isotope
record suggests that this situation was short lived: a rapid
return to unradiogenic Os values in the earliest
Hettangian indicates input of Os directly from the
intense weathering of CAMP lavas that lasted for the
next ∼3 Ma. By the end of the Hettangian it was all over,
with both Sr and Os isotope values returning to their
long-term trajectories.
Some of the best clues to the end-Triassic events may
have been buried deep in Panthalassa. Hori et al. made an
attempt to read the palaeontological and geochemical
archives preserved in a slowly accumulated deep-sea
chert sequence in Japan. The radiolarian extinction is one
of the most promising palaeontological markers of the
T–J boundary. In the Kurusu section of the Inuyama
area, the rapid radiolarian turnover is subdivided into
three events (E1 to E3). First go some of the taxa of a
diverse Triassic assemblage (E1). No more than 0.5 Ma
later there is a wholesale extinction of the remaining
species and the origination of a few new Jurassic forms.
This E2 event is recorded in a single bed that is estimated
to have accumulated in less than 10 ka and is taken as the
T–J boundary. Significantly, the E2 event also corre-
sponds to the last occurrence of conodonts (Misikella
posthersteini). Then E3 is a post-extinction interval
characterized by a low diversity fauna of small, spherical
spumellarians. The level of E1 coincides with tantalizing
geochemical signals. Among the Rare Earth Elements, a
distinct Ce anomaly is interpreted to signal a brief
acidification of sea water. The next higher chert bed
records an anomalously high abundance of Platinum
Group Elements (PGEs). The Ce anomaly is compatible
with CO
2
and SO
2
emissions from either a volcanic or an
impact source. However, the PGE peak can be best
accounted for by a calculated 2.5% admixing of impact
melt-derived material. If this is correct, the putative
impact may have played a role in the plankton extinction
8Editorial

at E1, but it cannot be directly implicated in the T–J
boundary extinction, half a million years after. Signif-
icantly, in between lies another chert layer that contains
basaltic glass and lithic fragments. If derived from a
CAMP source, this may be the first direct evidence that
some CAMP eruptions were violent enough to spread
airborne volcanic particles around the globe. The Kurusu
section clearly yields important pieces of the T–J puzzle,
yet fitting them together is not straightforward.
Similarly puzzling is a uniquely extensive metre-
scale horizon of soft sediment deformation which occurs
immediately below the initial carbon isotope excursion
in eight discrete sedimentary basins in the UK region,
and covering an area of N250,000 km
2
. Simms reviews
published evidence for this ‘seismite’and concludes that
it represents only a single shock event, and is at least
locally overlain by sedimentary facies of plausible
tsunami origin. The facies successions are closely
comparable to those described from shallow marine
strata in proximity to the end-Cretaceous Chicxulub
impact crater. In view of the great distance from the T–J
boundary ‘seismite’to the nearest CAMP volcanic
rocks, Simms rejects the idea that these beds originated
in relation to violent CAMP eruptions. Instead he
suggests that the observed phenomena are compatible
with an impactor of relatively modest dimensions,
possibly some 2–3 km across, forming a so far
undiscovered crater of 40–50 km diameter, too small
to have had a significant effect on biotic change.
3. Possibilities
Despite much progress, a sufficiently high-resolution
geochronological framework is still lacking to firmly
establish the temporal link of CAMP's first and/or
largest eruptions, the environmental events, and the
extinction. The main unresolved issue is the comparison
of
40
Ar/
39
Ar and U–Pb dates. The first method is used
extensively in dating CAMP basalts but suffers from
uncertainty in the decay constant of
40
K. A current
revision of the constant (Villa and Renne, 2005) may
require recalculation of all
40
Ar/
39
Ar ages and their
upward adjustment by ∼1% (i.e. a published age of
200 Ma would be in fact be close to 202 Ma). Curiously,
the U–Pb method may also have produced ages that
systematically err on the young side. The going estimate
of the T–J boundary age hinges on a multi-grain zircon
U–Pb age (199.6 ± 0.4 Ma, Pálfy et al., 2000). Multi-
grain analyses are prone to leave slight Pb loss
undetected, hence producing marginally younger ages.
The remedy is now available, analysis of individual
crystals of zircon, also using improved methods to
eliminate the effects of Pb loss (e.g. Mundil et al.,
2004). Significantly, a single-crystal
206
Pb/
238
U age of
201.27 ± 0.27 Ma has been obtained for the North
Mountain basalt, a CAMP flow in Nova Scotia, Canada
(Schoene et al., 2006). Only the application of these
recent advances will help compare the timing of CAMP
and T–J boundary events with greater confidence.
There has been some attempt to use cyclostratigraphy
to calibrate the duration of events at the T–J boundary,
notably with respect to the CAMP volcanism in eastern
North America, but so far cyclostratigraphy has not been
used effectively to help understand the marine sections.
This is partly because most of the marine sections
investigated so far show major facies changes across the
boundary, and yet early attempts to use this method have
not been entirely unsuccessful (cf. Weedon et al., 1999).
With the development of high resolution lithological
and chemostratigraphic datasets much future progress
should be possible.
The proxy record for atmospheric carbon dioxide
change (e.g. McElwain et al., 1999; Tanner et al., 2001)
remains relatively weak, and there is much scope for
further work in this area, based on analyses of the well-
preserved plant fossils and soil carbonates that abound
in several basins around the world (e.g. Harris, 1937).
An improved terrestrial–marine correlation is essential.
One potentially powerful approach that has not yet been
harnessed for the T–J boundary, is the use of compound
specific carbon–isotopes as an alternative to analysis of
bulk organic matter, to provide a carbon–isotope
stratigraphy where the effects of mixing of different
organic components can be better controlled.
Whatever the quality of the present proxy record,
there does now seem to be widespread agreement that
carbon dioxide produced directly from CAMP, even with
a gas hydrate supplement brought about by greenhouse
warming, may not have been enough to cause all of the
evident environmental impacts. However, it has been
pointed out for other LIPs that baking of organic rich
rocks may generate massive additional amounts of
atmospheric and oceanic carbon (Svensen et al., 2004;
McElwain et al., 2005) and this mechanism remains an
unexplored possibility in the case of CAMP. Certainly
the huge extensional basins into which CAMP magmas
were intruded were at times enriched in organic matter
and might have provided a ready substrate for production
of thermogenic methane.
The debate about extraterrestrial versus volcanic
drivers for environmental change has not yet been
concluded, and it is noteworthy that all of the candidate
indicators of extraterrestrial impact –reports of PGE's
and soft sediment deformation –occur shortly prior to
9Editorial

CAMP volcanic activity. Pure coincidence aside, this
observation keeps alive the idea that there is an ‘impact
signal’–LIP connection, even if the mechanisms remain
highly controversial; for example, impact decompres-
sion melting, as recently articulated by Elkins-Tanton
and Hager (2005), or lithospheric gas explosion (Phipps
Morgan et al., 2005).
Acknowledgements
We wish to thank all the participants of IGCP 458,
and in particular all the referees whose hard work in
helping to produce this special issue is much appreci-
ated. Axel von Hillebrandt provided welcome critical
comment on this introduction.
References
Elkins-Tanton, L.T., Hager, B.H., 2005. Giant meteoroid impacts can
cause volcanism. Earth and Planetary Science Letters 239, 219–232.
Galli, M.T, Jadoul, F., Bernasconi, S.M., Weissert, H., 2005.
Anomalies in global carbon cycling and extinction at the
Triassic/Jurassic boundary: evidence from a marine C-isotope
record. Palaeogeography, Palaeoclimatology, Palaeoecology 216,
203–214.
Guex, J., Bartolini, A., Atudorei, V., Taylor, D., 2004. High-resolution
ammonite and carbon isotope stratigraphy across the Triassic–Jurassic
boundary at New York Canyon (Nevada). Earth and Planetary Science
Letters 225, 29–41.
Harris, T.M., 1937. The fossil flora of Scoresby Sound East Greenland.
Part 5: stratigraphic relations of the plant beds. Meddelelser om
Grønland 112, 1–114.
Hesselbo, S.P., Robinson, S.A., Surlyk, F., Piasecki, S., 2002. Terrestrial
and marine extinction at the Triassic–Jurassic boundary synchronized
with major carbon-cycle perturbation:alinktoinitiationofmassive
volcanism? Geology 30, 251–254.
Hesselbo, S.P., Robinson, S.A., Surlyk, F., 2004. Sea-level change and
facies development across potential Triassic–Jurassic boundary
horizons, SW Britain. Journal of the Geological Society, London
161, 365–379.
Knight, K.B., Nomade, S., Renne, P.R., Marzoli, A., Bertrand, H.,
Youbi, N., 2004. The Central Atlantic Magmatic Province at the
Triassic–Jurassic boundary: paleomagnetic and Ar
40
/Ar
39
evi-
dence from Morocco for brief, episodic volcanism. Earth and
Planetary Science Letters 228, 143–160.
Marzoli, A., Bertrand, H., Knight, K.B., Cirilli, S., Buratti, N., Verati,
C., Nomade, S., Renne, P.R., Youbi, N., Martini, R., Allenbach, K.,
Neuwerth, R., Rapaille, C., Zaninetti, L., Bellieni, G., 2004.
Synchrony of the Central Atlantic magmatic province and the
Triassic–Jurassic boundary climatic and biotic crisis. Geology 32,
973–976.
Mundil, R., Ludwig, K.R., Metcalfe, I., Renne, P.R., 2004. Age and
timing of the Permian mass extinctions: U/Pb dating of closed-
system zircons. Science 305, 1760–1763.
McElwain, J.C., Beerling, D.J., Woodward, F.I., 1999. Fossil plants
and global warming at the Triassic–Jurassic boundary. Science
285, 1386–1390.
McElwain, J.C., Murphy, J.W., Hesselbo, S.P., 2005. Changes in
carbon dioxide during an oceanic anoxic event linked to intrusion of
Gondwana coals. Nature 435, 479–483, doi:10.1038/nature03618.
Olsen, P.E., Kent, D.V., Sues, H.-D., Koeberl, C., Huber, H.,
Montanari, A., Rainforth, E.C., Fowell, S.J., Szajna, M.J., Hartline,
B.W.,2002. Ascent of dinosaurs linked to an iridium anomaly at the
Triassic–Jurassic boundary. Science 296, 1305–1307.
Pálfy, J., Mortensen, J.K., Carter, E.S., Smith, P.L., Friedman, R.M.,
Tipper, H.W., 2000. Timing of the end-Triassic mass extinction:
first on land, then in the sea? Geology 28, 39–42.
Phipps Morgan, J., Reston, T.J., Ranero, C.R., 2005. Reply to A.
Glikson's comment on ‘Contemporaneous mass extinctions,
continental flood basalts, and ‘impact signals’: Are mantle
plume-induced lithospheric gas explosions the causal link?’
[EPSL 217 (2004) 263–285]. Earth and Planetary Science Letters
236, 938–941.
Schoene, B., Crowley, J.L., Condon, D.J., Schmitz, M.D., Bowring, S.A.,
2006. Reassessing the uranium decay constants for geochronology
using ID-TIMS U–Pb data. Geochimica et Cosmochimica Acta 70,
426–445.
Svensen, H., Planke, S., Malthe-Sorenssen, A., Jamtveit, B.,
Myklebust, R., Eidem, T.R., Rey, S.S., 2004. Release of methane
from a volcanic basin as a mechanism for initial Eocene global
warming. Nature 429, 542–545.
Tanner, L.H., Hubert, J.F, Coffey, B.P., McInerney, D.P., 2001.
Stability of atmospheric CO
2
levels across the Triassic/Jurassic
boundary. Nature 411, 675–677.
Villa, I.M., Renne, P.R., 2005. Decay constants in geochronology.
Episodes 28, 50–51.
Ward, P.D., Garrison, G.H., Haggart, J.W., Kring, D.A., Beattie, M.J.,
2004. Isotopic evidence bearing on Late Triassic extinction events,
Queen Charlotte Islands, British Columbia, and implications for
the duration and cause of the Triassic/Jurassic mass extinction.
Earth and Planetary Science Letters 224, 589–600.
Weedon, G.P., Jenkyns, H.C., Coe, A.L., Hesselbo, S.P., 1999.
Astronomical calibration of the Jurassic time scale from cyclos-
tratigraphy in British mudrock formations. Philosophical Transac-
tions of the Royal Society of London, A 357, 1787–1813.
Stephen P. Hesselbo
Department of Earth Sciences, University of Oxford,
Parks Road, Oxford, OX13P3, UK
E-mail address: stephen.hesselbo@earth.ox.ac.uk.
Corresponding author.
Christopher A. McRoberts
Department of Geology, State University of New York at
Cortland, P.O. Box 2000, Cortland, NY 13045, USA
József Pálfy
Research Group for Paleontology, Hungarian Academy
of Sciences-Hungarian Natural History Museum,
P.O. Box 137, Budapest, H-1431, Hungary
10 Editorial