Terrestrial and marine extinction at the Triassic-Jurassic boundary synchronized with major carbon-cycle perturbation: A link to initiation of massive volcanism?

Abstract

Mass extinction at the Triassic-Jurassic (Tr-J) boundary occurred about the same time (200 Ma) as one of the largest volcanic eruptive events known, that which characterized the Central Atlantic magmatic province. Organic carbon isotope data from the UK and Greenland demonstrate that changes in flora and fauna from terrestrial and marine environments occurred synchronously with a light carbon isotope excursion, and that this happened earlier than the Tr-J boundary marked by ammonites in the UK. The results also point toward synchronicity between extinctions and eruption of the first Central Atlantic magmatic province lavas, suggesting a causal link between loss of taxa and the very earliest eruptive phases. The initial isotopic excursion potentially provides a widely correlatable marker for the base of the Jurassic. A temporary return to heavier values followed, but relatively light carbon dominated the shallow oceanic and atmospheric reservoirs for at least 600 k.y.

Full text

q2002 Geological Society of America. For permission to copy, contact Copyright Permissions, GSA, or editing@geosociety.org.
Geology; March 2002; v. 30; no. 3; p. 251–254; 4 figures; Data Repository item 2002021. 251
Terrestrial and marine extinction at the Triassic-Jurassic boundary
synchronized with major carbon-cycle perturbation: A link to
initiation of massive volcanism?
Stephen P. Hesselbo*
Stuart A. Robinson 

Department of Earth Sciences, University of Oxford, Parks Road, Oxford OX1 3PR, UK
Finn Surlyk Geological Institute, University of Copenhagen, Øster Voldgade, DK-1350 Copenhagen K, Denmark
Stefan Piasecki Geological Survey of Denmark and Greenland, Thoravej 8, DK-2400 Copenhagen NV, Denmark
Figure 1. Early Jurassic reconstruc-
tion of northern Pangaea (from Zie-
gler, 1990) showing Central Atlantic
magmatic province (CAMP) as white
area enveloped by dashed lines (from
McHone, 2000). 1—St. Audrie’s Bay,
southwest England (Fig. 2); 2—Astar-
tekløft, East Greenland (Fig. 3); 3—ap-
proximate position of Queen Charlotte
Islands, western Canada; 4—Cso
99 va
9r,
Hungary (Fig. 4). Dark gray shading is
oceanic crust; medium gray shading
is continental crust; light gray shad-
ing is orogenic belt.
ABSTRACT
Mass extinction at the Triassic-Jurassic (Tr-J) boundary oc-
curred about the same time (200 Ma) as one of the largest volcanic
eruptive events known, that which characterized the Central At-
lantic magmatic province. Organic carbon isotope data from the
UK and Greenland demonstrate that changes in flora and fauna
from terrestrial and marine environments occurred synchronously
with a light carbon isotope excursion, and that this happened ear-
lier than the Tr-J boundary marked by ammonites in the UK. The
results also point toward synchronicity between extinctions and
eruption of the first Central Atlantic magmatic province lavas, sug-
gesting a causal link between loss of taxa and the very earliest
eruptive phases. The initial isotopic excursion potentially provides
a widely correlatable marker for the base of the Jurassic. A tem-
porary return to heavier values followed, but relatively light car-
bon dominated the shallow oceanic and atmospheric reservoirs for
at least 600 k.y.
Keywords: mass extinction, volcanism, carbon isotopes, Triassic,
Jurassic.
INTRODUCTION
Chronology of biotic and geologic events relative to disturbances
in biogeochemical cycles, such as that of carbon, is of prime impor-
tance to understanding the cause and character of mass extinction at
the Triassic-Jurassic (Tr-J) boundary. Anomalies in the ratios of
13
Cto
12
C in organic and inorganic materials (i.e., carbon isotope excursions)
are commonly associated with extinction events and other environ-
mental catastrophes (e.g., Holser et al., 1989; Dickens et al., 1995;
Gro¨cke et al., 1999; Hesselbo et al., 2000). Although published carbon
isotope data from across the Tr-J boundary have low-resolution sample
spacing, possible diagenetic modification, and limited stratigraphic
coverage, existing data indicate a degree of disturbance in the global
carbon cycle during the extinction event. In particular there is a neg-
ative carbon isotope excursion of apparently short duration (McElwain
et al., 1999; Ward et al., 2001; Pa´lfy et al., 2001).
Attention has focused recently on relationships between massive
igneous eruptions and major environmental perturbations, including
those associated with extinction (Wignall, 2000, and references there-
in). In the case of the Tr-J boundary, the Central Atlantic Magmatic
Province has been implicated (McHone, 1996; Marzoli et al., 1999;
Pa´lfy et al., 2000; Fig. 1), but timing of extrusions relative to geo-
chemical and biotic change on land and in the sea is unclear. Interpre-
tations of relative timing based on radiometric data place considerable
emphasis on small analytical uncertainties (Pa´lfy et al., 2000), and are
thus suspect. Here we present carbon isotope data from two key Tr-J
boundary sections from north-central Pangaea, one marine and the oth-
er nonmarine (Fig. 1), which we then use to resolve the chronology of
extinction and environmental change.
*E-mail: stephen.hesselbo@earth.ox.ac.uk.
MARINE RECORD
The mainly marine sequence at St. Audrie’s Bay, southwest Eng-
land, is a candidate global stratotype for the base of the Hettangian
Stage, and thus also for the base of the Jurassic System (Warrington
et al., 1994). This location is therefore suitable for construction of a
high-resolution carbon isotope curve calibrated against fossil content
and sedimentary facies (Fig. 2). The highest horizon regarded as mark-
ing the Tr-J boundary is the lowest occurrence of the ammonite Psil-
oceras planorbis, a supposedly widely correlatable biostratigraphic
marker of definite Jurassic character (Warrington et al., 1994). How-
ever, alternative and lower horizons have been proposed as the bound-
ary, based on lithology and nonammonite faunas (Poole, 1979; Hallam,
1990) or palynology (Orbell, 1973). Thus, the Tr-J boundary has been
placed by different authors between ;12.6 and 19.7 m height in the
section shown in Figure 2.
Carbon isotope values from bulk organic matter (d
13
C
org
5per
mil deviation in
13
C/
12
C with respect to a standard; e.g., method of
Hesselbo et al., 2000)
1
show fluctuations through the succession. Of
note is a negative excursion within the Cotham Member (;24‰ at
12.7 m height in Fig. 2, labeled initial excursion), followed by a second
1
GSA Data Repository item 2002021, Carbon isotope data from St. Au-
drie’s Bay, England, and Astartekløft, East Greenland, is available on request
from Documents Secretary, GSA, P.O. Box 9140, Boulder, CO 80301-9140,
editing@geosociety.org, or at www.geosociety.org/pubs/ft2002.htm.

252 GEOLOGY, March 2002
Figure 2. Measured section and carbon isotope data (PDB is Pee-
dee belemnite) from St. Audrie’s Bay, southwest England.Biostra-
tigraphy and lithostratigraphy follow Orbell (1973), Mayall (1981),
Woollam and Riding (1983), and Warrington et al. (1994). Strata re-
cord gradual transition from evaporitic lacustrine conditions in
Late Triassic (Norian), through to normal marine conditionsin Early
Jurassic (Hettangian). Although marine conditions developed in
terminal Triassic (Rhaetian), progression to deeper water was in-
terrupted by shallowing and subaerial exposure. This event is evi-
dent in Cotham Member, which comprisesshallowing-upward low-
er portion, topped by prominent mud-cracked erosion surface in
middle, overlain by beds showing flat-topped wave ripples and des-
iccation cracks (Mayall, 1983). Initial isotope excursion occurs at
important marine-flooding surface that heralds trend toward deeper
water facies. Precise boundary positions for spore and pollen
zones (Orbell, 1973) and dinoflagellate cyst zones (definition of
Woollam and Riding, 1983) are based on examination of same sam-
ples used for isotopic analysis. Shaded areas inchronostratigraphy
and biostratigraphy columns indicate limits of uncertainty.
negative shift in the lower Blue Lias Formation (above 17.5 m height
in Fig. 2, labeled main excursion). Considered in isolation, these ex-
cursions might result from changing proportions of organic compo-
nents of varied isotopic composition. However, in this case we reject
such an interpretation because the same excursions are recognized in
coeval successions. Relatively light carbon values that characterize the
middle part of the Westbury Formation (;7.5 m in Fig. 2) may also
represent an excursion, but there are too few supporting data. The main
negative excursion persists through the Psiloceras planorbis biozone,
representing a time interval of .300 k.y. based on recognition of or-
bital obliquity cycles (Weedon et al., 1999).
NONMARINE RECORD
A purely terrestrial record of carbon isotope composition and en-
vironmental change across the Tr-J boundary can be gained from flu-
vio-lacustrine sediments of the Kap Stewart Formation in Jameson
Land, East Greenland (Dam and Surlyk, 1993). On the margins of the
basin, the Kap Stewart Formation is predominantly in delta-plain fa-
cies. We studied one such section, Astartekløft, because this is prin-
cipally where Harris (1937) recognized a paleobotanical transition zone
containing leaf fossils from the Triassic Lepidopteris and the Jurassic
Thaumatopteris biozones (Fig. 3).
In addition to abundant leaves, the Kap Stewart Formation con-
tains plentiful coalified wood. The carbon isotopic composition of fos-
sil wood has previously been used to correlate between marine and
terrestrial strata (Gro¨cke et al., 1999; Hesselbo et al., 2000; seefootnote
one). A plot of d
13
C
wood
from Astartekløft exhibits a distinct negative
excursion of ;23.5‰ coincident with the macroplant zonal boundary
(Fig. 3; method of Gro¨cke et al., 1999). Strongly negative d
13
C
wood
values persist through at least 15 m of section before returning to more
normal levels.
GLOBAL CORRELATION
Using a combination of biostratigraphy and isotope stratigraphy,
we can establish a robust relative chronology. Previous palynological
work in East Greenland (Pedersen and Lund, 1980) equated the Kap
Stewart Formation up to and including the transition zone with strata
up to and including the lower Cotham Member of southwest England
(Figs. 2 and 3). The same study indicated a likely hiatus in KapStewart
sedimentation equivalent to the Langport Member and some of the
adjacent strata in southwest England, although precise definitions of
the missing section cannot be determined using the available pollen
and spore data. Nevertheless, these palynological correlations indicate
that the negative carbon isotope excursion recognized from East Green-
land must correlate to both the initial excursion and the main excursion
at St. Audrie’s Bay (Fig. 4), implying a significant hiatus immediately
above Harris’s plant bed D (45 m height in Fig. 3). This level is at the
base of a fluvial channel sandstone, and the hiatus may represent only
local erosion.
Recent studies show a negative d
13
C excursion below the first
appearance of Psiloceras in marine sections in western Canada (Ward

GEOLOGY, March 2002 253
Figure 3. Measured section and carbon isotope data from Astar-
tekløft, East Greenland. Biostratigraphy and lithostratigraphy
from Harris (1937), Pedersen and Lund (1980), and Dam and Sur-
lyk (1993). Plant beds are from Harris (1937), and beds A–D were
relocated through collection and identification of leaf fossils; 2L,
‘‘two lowest’’; T-P,
Todites princeps
bed; P-A,
Phlebopteris an-
gustiloba
bed; T-B,
Thaumatopteris brauniana
bed. Shaded areas
in lithostratigraphy, chronostratigraphy, and macrofloral zona-
tion columns indicate uncertainty due to lack of data. Possible
minor hiatus is identified on basis of carbon isotopestratigraphy:
others are also likely. PDB is Peedee belemnite.
et al., 2001) and Hungary (Pa´lfy et al., 2001; Fig. 4). We correlate this
anomaly with our initial negative excursion at St. Audrie’s Bay (Fig.
4). The difference in position of the negative excursion at the three
locations with respect to the first appearance of the ammonite Psilo-
ceras probably reflects global diachroneity of the ammonite faunas. In
particular, our isotopic correlations imply a relatively late appearance
of Psiloceras in the nearshore UK sections (cf. Hodges, 1994; Bloos
and Page, 2000). The results thus underline inadequacies in the use of
ammonite markers for the base of the system. Furthermore, our data
suggest that radiolarian extinctions that are well documented from Ca-
nadian sections (Carter et al., 1998, and references therein) are tem-
porally indistinguishable from turnover in terrestrial floras. It appears
that the last conodonts became extinct at some level between the major
initial extinctions (e.g., of land plants and radiolarians) and the first
occurrence of Psiloceras in the sections investigated.
IMPLICATIONS
A perturbation represented by more than one isotopic excursion
occurred in the global carbon cycle at the Tr-J boundary. Disturbance
was sustained through a significant interval of the earliest Jurassic,
perhaps .600 k.y. Isotopically light carbon characterized both surface-
ocean and atmospheric reservoirs. All applicable mechanisms for gen-
erating major light carbon isotopic anomalies in exchangeable surface
reservoirs link to increased atmospheric pCO
2
(e.g., Kump and Arthur,
1999; Dickens, 2000), and a recent suggestion that atmospheric pCO
2
levels were stable across the boundary (Tanner et al., 2001) is incom-
patible with our data. The initial excursion in western Canada was
interpreted to represent productivity collapse following some environ-
mental (extraterrestrial) catastrophe (Ward et al., 2001). However, this
is only one of a number of possible explanations, principal alternatives
being dissociation of gas hydrates or efflux of volcanogenic CO
2
(e.g.,
Dickens et al., 1995; Kump and Arthur, 1999; Dickens, 2000; Hesselbo
et al., 2000; Pa´lfy et al., 2001). Moreover, productivity collapse cannot
explain the entire pattern of isotopic anomalies that characterizes in-
tervals of organic carbon burial.
Timing of the carbon isotope excursions relative to Central At-
lantic magmatic province eruptions is significant to interpretation of
the Tr-J boundary event. Magmatic activity on the Atlantic margin
lasted ;2 m.y. and studies of palynology and vertebrate footprints
indicate abrupt terrestrial extinctions only ;20 k.y. before eruption of
the first basalts in eastern North America (Fowell et al., 1994; Olsen
et al., 2002). The Tr-J boundary in those sections has been recognized
by palynological assemblages dominated by the pollen Corollina sp.
(5Classopollis), a thermophyllic taxon (cf. McElwain et al., 1999;
Hesselbo et al., 2000). Flood abundance of Corollina sp. in our sam-
ples (52%–90% of the terrestrial assemblage) occurs exactly at the
level of the initial negative excursion at St. Audrie’s Bay. Thus, we
establish a correlation between the isotope excursion and initiation of
major basaltic eruptions, at least where we have a good knowledge of
their ages in Central Atlantic marginal areas.
CONCLUSIONS
A major perturbation in the global carbon cycle occurred at the
end of the Triassic, affecting atmospheric and shallow-oceanic isotopic
reservoirs. Data are consistent with an input of isotopically light car-
bon, most likely from CO
2
outgassing associated with the Central At-
lantic magmatic province, and also possibly with a component derived
from gas hydrate. We suggest that the initial excursion provides a suit-
able marker for the Tr-J boundary (cf. Paleocene-Eocene boundary).
The inferred initial pulse of CO
2
(and other gases) coincides with floral
and faunal turnover in marine and terrestrial settings. Overall, increased
atmospheric CO
2
persisted for .600 k.y. after the initial excursion,
and the flora and fauna adapted to the new environmental conditions.
Knowledge of the chemistry and physics of volatile and aerosol pro-
duction during initiation of large igneous provinces will be crucial to
understanding the ultimate cause of extinctions such as that at the Tr-
J boundary.

254 GEOLOGY, March 2002
Figure 4. Correlations between key Triassic-Jurassic boundary sections. Western Canada is from Ward et al. (2001); Hungary is from
Pa´lfy et al. (2001). PDB is Peedee belemnite.
ACKNOWLEDGMENTS
We acknowledge support from the Natural Environment Research Council, UK (stu-
dentship to Robinson), the Royal Society (London), and the Burdett-Coutts Fund (Univer-
sity of Oxford). We thank also two anonymous referees, M. Becker, L. Stemmerik, O.
Green, S. Wyatt, the Danish Natural Science Research Council, the Carlsberg Foundation,
the Geological Survey of Denmark and Greenland, and the University of Copenhagen.
Carbon isotope analyses were carried out at the Radiocarbon Accelerator Unit at the Uni-
versity of Oxford. We also thank T. O’Connell and M. Humm. This is a contribution to
International Geological Correlation Programme Project 458.
REFERENCES CITED
Bloos, G., and Page, K.N., 2000, The basal Jurassic Ammonite Succession in the North-
West European Province—Review and new results, in Hall, R.L., and Smith, P.L.,
eds., Advances in Jurassic Research 2000: Zu¨ rich, TransTech, p. 27–40.
Carter, E.S., Whalen, P.A., and Guex, J., 1998, Biochronology and palaeontology of Lower
Jurassic (Hettangian and Sinemurian) radiolarians, Queen Charlotte Islands, British
Columbia: Geological Survey of Canada Bulletin 496, 162 p.
Dam, G., and Surlyk, F., 1993, Cyclic sedimentation in a large wave- and storm-domi-
nated anoxic lake; Kap Stewart Formation (Rhaetian–Sinemurian), Jameson Land,
East Greenland, in Posamentier, H.W., et al., eds., Sequence stratigraphy and facies
associations: International Association of Sedimentologists Special Publication 18,
p. 419–448.
Dickens, G.R., 2000, Methane oxidation during the late Palaeocene thermal maximum:
Bulletin de la Socie´te´Ge´ologique de France, v. 171, p. 37–49.
Dickens, G.R., O’Neil, J.R., Rea, D.K., and Owen, R.M., 1995, Dissociation of oceanic
methane hydrate as a cause of the carbon isotope excursion at the end of the Paleo-
cene: Paleoceanography, v. 10, p. 965–971.
Fowell, S.J., Cornet, B., and Olsen, P.E., 1994, Geologically rapid Late Triassic extinctions:
Palynological evidence from the Newark Supergroup, in Klein, G.D., ed., Pangea:
Paleoclimate, tectonics, and sedimentation during accretion, zenith, and breakup of
a supercontinent: Geological Society of America Special Paper 288, p. 197–206.
Gro¨cke, D.R., Hesselbo, S.P., and Jenkyns, H.C., 1999, Carbon-isotope composition of
Lower Cretaceous fossil wood: Ocean-atmosphere chemistry and relation to sea-level
change: Geology, v. 27, p. 155–158.
Hallam, A., 1990, Correlation of the Triassic-Jurassic boundary in England and Austria:
Geological Society [London] Journal, v. 147, p. 421–424.
Harris, T.M., 1937, The fossil flora of Scoresby Sound East Greenland. Part5: Stratigraphic
relations of the plant beds: Meddelelser om Grønland, v. 112, no. 2, 112 p.
Hesselbo, S.P., Gro¨cke, D.R., Jenkyns, H.C., Bjerrum, C.J., Farrimond, P., Morgans Bell,
H.S., and Green, O.R., 2000, Massive dissociation of gas hydrate during a Jurassic
oceanic anoxic event: Nature, v. 406, p. 392–395.
Hodges, P., 1994, The base of the Jurassic System: New data on the first appearance of
Psiloceras planorbis in southwest Britain: Geological Magazine, v. 131, p. 841–844.
Holser, W.T., Scho¨nlaub, H.-P., Attrep, M., Jr., Boeckelmann, K., Klein, P., Magaritz, M.,
Pak, E., Schramm, J.-M., Stattgeger, K., and Schmo¨ller, R., 1989, A unique geo-
chemical record at the Permian/Triassic boundary: Nature, v. 337, p. 39–44.
Kump, L.R., and Arthur, M.A., 1999, Interpreting carbon-isotope excursions: Carbonates
and organic matter: Chemical Geology, v. 161, p. 181–198.
Marzoli, A., Renne, P.R., Piccirillo, E.M., Ernesto, M., Bellieni, G., and De Min, A., 1999,
Extensive 200-million-year-old continental flood basalts of the Central Atlantic mag-
matic province: Science, v. 284, p. 616–618.
Mayall, M.J., 1981, The Late Triassic Blue Anchor Formation and the initial Rhaetian
marine transgression in south-west Britain: Geological Magazine, v. 118, p. 377–384.
Mayall, M.J., 1983, An earthquake origin for synsedimentary deformation in a Late Triassic
lagoonal sequence, southwest Britain: Geological Magazine, v. 120, p. 613–622.
McElwain, J.C., Beerling, D.J., and Woodward, F.I., 1999, Fossil plants and global warming
at the Triassic-Jurassic boundary: Science, v. 285, p. 1386–1390.
McHone, J.G., 1996, Broad-terrane Jurassic flood basalts across northeastern North Amer-
ica: Geology, v. 24, p. 319–322.
McHone, J.G., 2000, Non-plume magmatism and rifting during the opening of the central
Atlantic Ocean: Tectonophysics, v. 316, p. 287–296.
Orbell, G., 1973, Palynology of the British Rhaeto-Liassic: Geological Survey of Great
Britain Bulletin, v. 44, p. 1–44.
Olsen, P.E., Koeberl, C., Huber, H., Montanari, A., Fowell, S.J., Et-Touhami, M., and Kent,
D.V., 2002, The continental Triassic-Jurassic boundary in Central Pangea: Recent
progress and preliminary report of an Ir anomaly, in Koeberl, C., and MacLeod, K.,
eds., Catastrophic events and mass extinctions: Impacts and beyond: Geological So-
ciety of America Special Paper 356 (in press).
Pa´lfy, J., Mortensen, J.K., Carter, E.S., Smith, P.L., Friedman, R.M., and Tipper, H.W.,
2000, Timing of the end-Triassic mass extinction: First on land, then in the sea?:
Geology, v. 28, p. 39–42.
Pa´lfy, J., Deme´ny, A., Hass, J., Hete´nyi, M., Orchard, M.J., and Veto, I., 2001, Carbon
isotope anomaly and other geochemical changes at the Triassic-Jurassic boundary
from a marine section in Hungary: Geology, v. 29, p. 1047–1050.
Pedersen, R.K., and Lund, J.J., 1980, Palynology of the plant-bearing Rhaetian to Hettan-
gian Kap Stewart Formation, Scoresby Sund, East Greenland: Reviews of Palaeo-
botany and Palynology, v. 31, p. 1–69.
Poole, E.G., 1979, The Triassic-Jurassic boundary in Great Britain: Geological Magazine,
v. 116, p. 303–311.
Tanner, L.H., Hubert, J.F., Coffey, B.P., and McInerney, D.P., 2001, Stability of atmospheric
CO
2
levels across the Triassic/Jurassic boundary: Nature, v. 411, p. 675–677.
Ward, P.D., Haggart, J.W., Carter, E.S., Wilbur, D., Tipper, H.W., and Evans, T., 2001,
Sudden productivity collapse associated with the Triassic-Jurassic boundary mass
extinction: Science, v. 292, p. 1148–1151.
Warrington, G., Cope, J.C.W., and Ivimey-Cook, H.C., 1994, St. Audrie’s Bay, Somerset,
England: A candidate Global Stratotype Section and Point for the base of the Jurassic
System: Geological Magazine, v. 133, p. 191–200.
Weedon, G.P., Jenkyns, H.C., Coe, A.L., and Hesselbo, S.P., 1999, Astronomical calibration
of the Jurassic time scale from cyclostratigraphy in British mudrock formations:
Royal Society of London Philosophical Transactions, ser. A, v. 357, p. 1787–1813.
Wignall, P.B., 2000, Large igneous provinces and mass extinctions: Earth-Science Reviews,
v. 53, p. 1–33.
Woollam, R., and Riding, J.B., 1983, Dinoflagellate cyst zonation of the English Jurassic:
Institute of Geological Sciences Report 83/2, 42 p.
Ziegler, P.A., 1990, Geological atlas of Western and Central Europe: The Hague, Shell
Internationale Petroleum Maatschappij B.V.
Manuscript received August 15, 2001
Revised manuscript received October 27, 2001
Manuscript accepted October 29, 2001
Printed in USA