Content uploaded by Charles E Mitchell
Author content
All content in this area was uploaded by Charles E Mitchell on Dec 01, 2015
Content may be subject to copyright.
Environmental changes in the Late Ordovician–
early Silurian: Review and new insights from
black shales and nitrogen isotopes
Michael J. Melchin1,†, Charles E. Mitchell2, Chris Holmden3, and Petr Štorch4
1Department of Earth Sciences, St. Francis Xavier University, Antigonish, Nova Scotia B2G 2W5, Canada
2Department of Geology, State University of New York at Buffalo, Buffalo, New York 14260, USA
3Saskatchewan Isotope Laboratory, Department of Geological Sciences, University of Saskatchewan, Saskatoon, Saskatchewan
S7N 5E2, Canada
4Institute of Geology AS CR, v.v.i., Rozvojova 269, 165 00 Praha 6, Czech Republic
ABSTRACT
The Late Ordovician (Katian-Hirnantian)
through earliest Silurian (Rhuddanian) in-
terval was a time of varying climate and sea
level, marked by a peak glacial episode in the
early-mid-Hirnantian. Synthesis of recently
published data permits global correlation of
at least two cycles of glacial advance and re-
treat with a distinct interglacial period that
is recognizable in sequence-stratigraphic and
chemostratigraphic records in many parts of
the world. A period of warming and sea-level
rise during the late Katian is marked by the
widespread occurrences of oceanic anoxia
in paleotropical and subtropical localities,
mostly confi ned to regions of inferred up-
welling and semirestricted marine basins.
Nitrogen isotope data show that the regions
of oceanic anoxia were marked by intense
water-column denitrifi cation in which cyano-
bacteria were the principal source of fi xed N.
In the overlying peak glacial interval of the
Hirnantian, sedimentary successions from
localities representing a wide range of water
depths and paleolatitudes indicate that anoxia
was restricted during the early-mid-Hirnan-
tian. The shift to more positive N isotope val-
ues also suggests less intense water-column
denitrifi cation. In the overlying late Hirnan-
tian and early Rhuddanian, the distribution
of black shales reaches its greatest extent in
the studied interval. Localities showing evi-
dence of anoxia are globally spread over all
paleolatitudes and water depths for which
data are available, indicating a Rhuddanian
ocean anoxic event comparable to examples
from the Mesozoic. It is accompanied by a
return to intensely denitrifying conditions
within the water column, as indicated by the
shift to negative N isotope values. The two
phases of Hirnantian mass extinction coin-
cide with rapid, climate-driven changes in
oceanic anoxia. The fi rst extinction occurred
at the onset of glaciation and with the loss of
anoxic conditions at the end of the Katian.
The second extinction occurred at the demise
of glaciation and coincided with the return
of anoxic conditions during the late Hirnan-
tian–early Rhuddanian. Integration of our N
isotope data with graptolite biodiversity rec-
ords suggests that the extinctions were pro-
foundly infl uenced by changes occurring at
the base of the marine food web, i.e., redox-
driven changes in nutrient cycling and pri-
mary producer communities.
INTRODUCTION
The Late Ordovician to early Silurian is now
recognized as a time interval of considerable
environmental and biodiversity changes. One
of Earth’s major glacial episodes took place in
the Late Ordovician. The climatic and oceano-
graphic changes associated with the Hirnantian
(latest Ordovician) glaciation are widely consid-
ered as key drivers for a major mass extinction
event (e.g., Brenchley et al., 2003). In addition,
the changes in climate and ocean state that span
this interval closely resemble the changes tak-
ing place in the late Cenozoic and modern world
(Armstrong, 2007), and thus they provide key
insights into the long-term relationship between
the global environment and biodiversity.
One of the most prominent patterns of the
stratigraphic record of the Late Ordovician–
early Silurian is alternation between oxygen-
ated and anoxic sediments in deep-shelf and
basinal environments on time scales of hun-
dreds of thousands of year to several million
years (Page et al., 2007). There is debate regard-
ing the extent and persistence of anoxia in the
deep oceans in the early Paleozoic (e.g., Strauss,
2006; Algeo and Ingall, 2007; Landing, 2011),
and the scarcity of ocean sediment records
for this time interval makes hypotheses about
oceanic anoxia diffi cult to test. Nevertheless,
the available record does provide insights into
relationships among climate change, sea-level
change, oceanic anoxia, and biodiversity.
In celebration of the 125th anniversary of
the Geological Society of America (GSA), this
paper has several primary objectives: to review
the state of understanding regarding environ-
mental and biodiversity changes through the
Late Ordovician–early Silurian time interval as
of the GSA centennial, 25 yr ago; review our
current state of understanding of the relation-
ships among climate, sea-level, oceanic anoxia,
and biodiversity in the Late Ordovician–early
Silurian; to provide a new synthesis and inter-
pretation of the available data on the distribution
of black shales in the latest Ordovician and ear-
liest Silurian, immediately before, during, and
after the Hirnantian glaciation; and, to present
and interpret new nitrogen isotope data from
two widely separated paleogeographic regions,
which link changes in the biosphere to changes
in ocean redox state.
Very few Ordovician–Silurian successions
around the world have been studied in suffi -
cient sedimentological, faunal, and geochemical
detail to precisely interpret the redox conditions
at the site of deposition. Therefore, as part of
this study, we will use a fi rst-order proxy for
oxygen-depleted depositional environments, the
occurrence of black shales, through three time
slices, before, during, and after the Hirnantian
For permission to copy, contact editing@geosociety.org
© 2013 Geological Society of America
1635
GSA Bulletin; November/December 2013; v. 125; no. 11/12; p. 1635–1670; doi: 10.1130/B30812.1; 8 fi gures; 2 tables; Data Repository item 2013352.
†E-mail: mmelchin@stfx.ca
Invited Review
CELEBRATING ADVANCES IN GEOSCI ENCE
1888 2013
Melchin et al.
1636 Geological Society of America Bulletin, November/December 2013
peak glacial event, to infer the changing global
distribution of dysoxic-anoxic ocean conditions
through this time interval.
STATE OF UNDERSTANDING IN 1988
The 10 yr ending in 1988 were a pivotal time
interval for our understanding of the biotic and
paleoenvironmental changes that took place in
the Late Ordovician–early Silurian. Although it
had been long recognized that the Late Ordo-
vician was a time of elevated extinction rates
(Newell, 1963, 1967), the global biodiversity
synthesis studies of Raup and Sepkoski (1982,
1986) demonstrated that the Late Ordovician
was one of the fi ve major extinction episodes
of the Phanerozoic. The work of Alvarez et al.
(1980), which suggested that a bolide impact
event may have triggered the end-Cretaceous
mass extinction, led to a surge of interest in mass
extinction studies. In addition, the global com-
pilation of diversity studies also suggested the
possibility that there may have been a periodic-
ity in global extinction patterns (e.g., Sepkoski,
1986). These observations lead to detailed stud-
ies of all of the major mass extinction episodes,
including paleoenvironmental studies of the
extinction intervals, with the objective of link-
ing biodiversity fl uctuations with paleoenviron-
mental changes.
In addition, the late 1970s to early 1980s
was the time interval in which the international
stratigraphic community completed the formal
defi nition of the base of the Silurian System
(Bassett, 1985; Cocks, 1985). This effort led to
a wealth of stratigraphic studies in many parts
of the world with the objective of selecting the
best global stratotype section and point (GSSP)
for the base of the Silurian, and also the criteria
for correlation of sections in other parts of the
world with the GSSP. Much of this global work
was summarized in a volume edited by Cocks
and Rickards (1988). Other important work
summarizing the state of our understanding of
the Late Ordovician world in 1988 includes a
paper synthesizing the state of understanding of
the Hirnantian glaciation and mass extinction
(Brenchley, 1989), and the proceedings of the
1988 International Symposium on the Ordovi-
cian System (abstracts in Williams and Barnes,
1988), which were published as a conference
volume of papers edited by Barnes and Wil-
liams (1991).
Late Ordovician–Early Silurian Glaciation
Compelling evidence for Late Ordovician gla-
ciation on Gondwana, centered in North Africa,
has been available since the work of Destombes
(1968), Beuf et al. (1971), and many others, as
summarized by Frakes (1979), and in a compre-
hensive volume edited by Hambrey and Harland
(1981). However, as Frakes noted, the timing
and duration and even mode of origin of many
of the reported glacial deposits were not well
constrained. Frakes et al. (1992) summarized
much of the work of the later 1970s and 1980s
and suggested that the glacial episode spanned
much of the Late Ordovician and early Silu-
rian, and that ice sheets covered more than just
the North African part of Gondwana. They did
note, however, that the origin and age of some of
the purported glacial deposits remained unclear,
and some, such as those in South America, were
early Silurian in age (e.g., Caputo and Crowell,
1985). Nevertheless, it was clear that the inter-
val with the best-documented and most wide-
spread evidence for glacial deposits was within
the Hirnantian Stage (Fig. 1) (Brenchley, 1988,
1989; Brenchley et al., 1991; Frakes et al., 1992).
The 1980s also marked a series of studies on
the sedimentological and biofacies distribution
record of sea-level changes associated with the
Hirnantian glaciation, which revealed that the
eustatic record of this glaciation could provide
estimates of the magnitude of the global sea-
level change, and therefore provide additional
insight into the timing and extent of glacial ice-
cap advance (e.g., Brenchley and Cullen, 1984;
Brenchley and Newall, 1984; Brenchley and
Štorch, 1989). Brenchley (1988) estimated that
sea level fell 50–100 m during the Hirnantian.
In comparison, sea level has risen ~110–140 m
since the Last Glacial Maximum (Siddall et al.,
2010). Frakes et al. (1992) also reviewed the
available evidence of the distribution of other
climate-sensitive sediments, such as evapo-
rates, during the Late Ordovician–early Silurian
interval.
The fi rst study reporting oxygen isotope data
from carbonates spanning the Late Ordovi-
cian–early Silurian was conducted by Jux and
Manze (1979), followed by the work of Orth
et al. (1986), Marshall and Middleton (1990),
and Middleton et al. (1991). Whereas Middle-
ton et al. suggested that the positive shift in δ18O
values (Fig. 1) was the result of a combination
of global ocean temperature fall and increasing
ice volume associated with the Gondwanan gla-
ciation, Orth et al. highlighted the possible role
of changes in seawater salinity in controlling the
δ18O values.
Orth et al. (1986), Marshall and Middleton
(1990), and Middleton et al. (1991) also pub-
Silurian
Lland.
Upper Ordovician
(Ashgill)
System
/Series
Katian (Rawtheyan) Rhud. Stage
Eustatic
Curve
Stable
Isotopes
Environmental
Changes
Biotic
Changes
Zones
Hirnantian
pacificus extraord. persc. acum.complex.
-5 -3 -1 +1 +3 +5 +7
δ13Ccarb‰δ18Ocarb‰
Ice caps melt
Climate warms
Deep flooding of shelves
Plankton extinction (conodonts, acritarchs)
Climate warms
Deep flooding of shelves
Growth of ice caps
Spread of floating ice
Climate cools
Sea level starts to fall
Possible oceanic overturn
Major extinction of deep shelf benthic
faunas (trilobites, cystoids)
Major extinction of graptolites
Extinction of conodonts and acritarchs
in the temperate regions
Glacial maximum
Shelves largely emergent
Possible extinction of some shelf biota
(brachiopods and corals)
Major Extinction
Low diversity biota starts to diversify
Figure 1. Series, stages, and
graptolite zones through the
late Katian to early Rhudda-
nian. Summary of the state of
understanding of the history of
sea-level change, stable O and
C isotope variations, and envi-
ronmental and biotic changes
as of 1988. Eustatic, environ-
mental, and biotic changes
are from Brenchley (1989);
isotopic data are from Mar-
shall and Middleton (1990).
Abbreviations are as follows:
Lland.—Llandovery; Rhud. —
Rhuddanian; complex.—com-
plexus; extraord.—extraordi-
narius; persc.—persculptus;
acum.—acuminatus.
Environmental changes in the Late Ordovician–early Silurian: Review and new insights from black shales and nitrogen isotopes
Geological Society of America Bulletin, November/December 2013 1637
lished δ13C data that showed a positive excursion
through the Hirnantian glacial interval (Fig. 1).
This change was interpreted to be the result of
the changing carbon isotopic composition of
seawater (rather than vital effects or diagenesis
of the samples), which Marshall and Middleton
suggested was the result of an increase in rate of
organic carbon burial.
In addition to the oxygen isotopic evidence
suggesting global cooling and the evidence of
the spread of glacial and glaciomarine sediments
in Hirnantian time, the paleogeographic distri-
bution of benthic faunas, particularly brachio-
pods and trilobites, indicated that cool-water
faunas may have migrated into lower paleo-
latitude regions during Hirnantian time (e.g.,
Brenchley, 1989; Fortey, 1989), supporting the
hypothesis of overall global cooling and con-
striction of tropical climate belts. Wilde (1991),
following from Berggren and Hollister (1977)
and Parrish (1982), proposed models of chang-
ing ocean circulation patterns for the Ordovician
based on available evidence of paleocontinental
reconstructions, and faunal and sediment distri-
bution patterns. He proposed that the onset of
glaciation in the Late Ordovician, together with
the changing positions of continents relative to
the ocean basins and surface current systems,
may have resulted in a change in patterns of
deep-ocean circulation, from sinking saline (but
warm) central ocean water in pre-Hirnantian
times to sinking of colder, high-latitude waters
during the Hirnantian. This would have resulted
in changes in patterns of ventilation of deep and
intermediate waters.
Anoxia and Black Shales
It has long been recognized that the Ordovi-
cian and Silurian periods were times of wide-
spread deposition of black shales, which are
interpreted to have been deposited under con-
ditions of anoxic bottom waters. In addition, it
has also been observed that the Ordovician and
Silurian were times of alternation between black
and gray shales, indicative of different levels of
organic matter preservation in deep-shelf, slope,
and basinal settings (e.g., Leggett, 1980). It is
not our intention here to undertake a thorough
review of research on modern and ancient envi-
ronments of deposition of black shales. Research
on this topic up to the late 1980s has been sum-
marized in a number of papers and volumes,
including Brooks and Fleet (1987), Klemme and
Ulmishek (1991), Arthur and Sageman (1994),
and Wignall (1994). Rather, we focus particu-
larly on the studies that specifi cally addressed
black shale formation during the Late Ordovi-
cian and earliest Silurian and their use as proxies
for changing paleoceanographic conditions.
Berry and Wilde (1978) and Wilde (1987)
proposed a model in which the deep oceans,
from the late Precambrian to the early-mid-
Paleozoic, became progressively better oxy-
genated in response to increased atmospheric
oxygen levels. They also suggested that oxy-
genation of the abyssal depths of the oceans
was episodic through the early-mid-Paleozoic
rather than steady and ongoing, and that the
Late Ordovician glaciation was a key episode in
the transition from a stratifi ed world ocean with
limited overturning circulation and permanently
anoxic bottom waters, to an ocean more remi-
niscent of the present day, i.e., one characterized
by strong vertical mixing and highly oxygen-
ated bottom waters. Although the global deep-
ocean anoxia model contains some key insights
in terms of the development of black shales,
particularly their spread into shelf settings dur-
ing times of globally high sea levels and warm
climates, the model failed to account for spatial
and temporal patterns observed by later workers
in sedimentary records of black shale deposition
through the Ordovician and Silurian. For exam-
ple, Leggett (1980) showed that black shales
occurred at a wide range of depths in the British
successions on both sides of the Iapetus Ocean
during some time intervals during the Cam-
brian–Ordovician, but in others, the deep-shelf,
slope, and basinal settings appear to have been
fully oxygenated, or at least dysoxic, rather than
anoxic. Thus, anoxia was apparently not persis-
tent at slope depths throughout the Ordovician
and early Silurian, as suggested by Berry and
Wilde (1978). Nevertheless, both the Berry and
Wilde and Leggett models highlight the impor-
tance of high sea level in the appearance of black
shale deposition on the continental shelves, and
proposed mechanisms for the observed increase
in organic carbon content of the shales through
increased productivity from nutrient upwell-
ing, and/or increased preservation of the settled
organic matter from the photic zone beneath
low-oxygen bottom waters. Other authors (e.g.,
Brenchley, 1988) also highlighted the possible
role of global warming associated with the end
of the glaciation in the late Hirnantian, which
they suggested may have resulted in more slug-
gish ocean circulation as a consequence of weak
latitudinal temperature gradients, and thus con-
tributed to the development of oceanic anoxia.
Timing and Cause of Extinction
Most of the available data compilations con-
cerning rates of species turnover of various
groups of fossil taxa in the late 1980s were only
available to a level of precision of stages at best
(summarized by Brenchley, 1989). Neverthe-
less, even those data showed that there were
clearly at least two phases of extinction through
the Late Ordovician. The fi rst one was near the
beginning of the Hirnantian Stage, approxi-
mately coincident with the onset of the Hirnan-
tian glaciation based on facies indicators of sea-
level change (Fig. 1). The second was associated
with the end of glaciation and the postglacial
sea-level rise and spread of black shales into
shelf regions in many parts of the world. Dif-
ferent groups of organisms were affected to dif-
ferent degrees by these two phases of extinction
(Brenchley, 1989).
Two investigations through the Hirnantian
extinction intervals at well-known Ordovician-
Silurian boundary successions, to test a bolide
impact hypothesis for the origin for the extinc-
tions (Wilde et al., 1986; Orth et al., 1986),
found no evidence of a signifi cant iridium
anomaly comparable to that of the Cretaceous-
Paleogene boundary. Although Wang and Chai
(1989) reported a weak iridium anomaly from
a level within the Hirnantian succession in the
eastern Yangtze Gorges area of China, Wang
et al. (1992) attributed this to variations in sedi-
mentation rate.
Most authors of the 1980s instead high-
lighted the importance of three environmental
infl uences that likely contributed to these two
phases of extinction. The fi rst of these was sea
level. Pre-Hirnantian Ordovician sea levels
were among the highest in Earth history, and
large areas of the interior of several of the major
cratons were occupied by shallow, epiconti-
nental seas with rich and often endemic ben-
thic faunas (e.g., Sheehan, 1973, 1975). Thus,
glacio-eustatic fall would have had the very
profound effect of draining these widespread
shallow-marine habitats, producing an unusu-
ally strong habitat-area effect (Schopf, 1974;
Simberloff, 1974), and shifting moderate- and
deep-shelf faunas to a relatively narrow region
of the continental shelf edges. The second peak
of extinction occurred during the sea-level rise
associated with the end of the glaciation, which
was also a period of widespread black shale
deposition on the shelves. Formerly oxygenated
areas of the shelves prior to and during the gla-
ciation were now unavailable to shallow-water
benthic organism, thus further limiting habitable
shelf area (Brenchley, 1989).
Global cooling and expansion of the polar to
cool water belts at the expense of the warmer
water realm at low latitudes may have also con-
tributed to Hirnantian extinction (Skevington,
1974; Sheehan, 1979). This hypothesis was
supported by patterns of replacement in ben-
thic assemblages, especially at mid-latitude sites
(e.g., Brenchley, 1989). In addition, Wilde and
Berry (1984) and Berger and Thierstein (1979)
suggested that the change from nonglacial
Melchin et al.
1638 Geological Society of America Bulletin, November/December 2013
to glacial conditions could have resulted in
oceanic overturning and/or upwelling of poten-
tially toxic waters into the photic zone, or both,
and, therefore, disruption of primary productiv-
ity. This could have led to a cascade of abrupt
extinctions during the onset and/or end of the
Hirnantian glaciation. At levels deeper in the
water column, Berry et al. (1987) proposed that
the denitrifi cation zone may have been a pre-
ferred habitat for diverse planktonic graptolite
communities. Both those authors and Melchin
and Mitchell (1991) suggested that ventilation
of the oceans during glaciation resulted in a
dramatic reduction in oxygen minimum zones
and, consequently, the proposed graptolite habi-
tat. This hypothesis was offered to explain the
particularly profound extinction of graptolites
that occurred at the beginning of the Hirnantian
glaciation. This process also may have affected
other planktonic organisms, including the
pelagic larvae of benthic animals.
SYNTHESIS OF CURRENT
UNDERSTANDING
There has been an enormous rate of growth
in research dealing with the changes in climate,
ocean conditions, and biodiversity through Late
Ordovician–early Silurian time in the past 25 yr.
We present here a synthesis of our current under-
standing and ongoing controversies surrounding
these events, particularly the environmental
changes associated with the Hirnantian glacia-
tion and the occurrence of marine anoxia prior
to, during, and after the peak glacial interval.
For this discussion, we employ the most recent
Ordovician and Silurian time scale (Cooper and
Sadler, 2012; Melchin et al., 2012). Based on
the stratigraphic information available to us,
we have replotted on that time scale a synthe-
sis of the isotopic data previously published
by Bergström et al. (2009, 2010), Young et al.
(2009), and Finnegan et al. (2011), as well as
data from several other studies focused more
narrowly on the latest Ordovician and earliest
Silurian (Fig. 2). Several features of our adapta-
tion of this chronology and the associated stable
isotopic record are important to emphasize here.
First, the international correlation of Hirnantian
strata and the associated stable isotope record is
somewhat controversial (see, for example, Ain-
saar et al., 2010; Delabroye and Vecoli, 2010;
Mitchell et al., 2011; Jones and Fike, 2013). A
review of these issues, together with a recent
analysis of the sea-level and Nd isotope data rel-
evant to Late Ordovician sea-level history, was
presented by Holmden et al. (2013). We follow
here the correlations advocated by Melchin and
Holmden (2006a), Melchin (2008), Achab et al.
(2011), Mitchell et al. (2011, 2012), and Holm-
den et al. (2013). In addition, we present an
extended version of Holmden et al.’s proposed
correlation in Figure 3 and a yet broader set of
correlations of Hirnantian strata from around the
globe in the GSA Data Repository (Fig. S11).
Second, present evidence suggests that the
effects of local carbon cycling had substantial
infl uence on the mid-Ordovician to early Silu-
rian carbon isotopic record (e.g., Melchin and
Holmden, 2006a, 2006b; Young et al., 2008;
LaPorte et al., 2009; Kaljo and Martma, 2011;
Kaljo et al., 2012), which suggests, in turn, that
the curve presented by Bergström et al. (2009)
probably does not record a purely oceanic
carbon reservoir signal. We show some of the
regional variation in the δ13Ccarb record for
the Hirnantian interval as a family of solid and
dashed curves in Figure 2 and in their original
detail along with δ13Corg and other data in Fig-
ure 3 (see also Fig. S1 [see footnote 1]).
Third, we regard the strong positive δ13Ccarb
excursion obtained from the Elkhorn Forma-
tion in the Cincinnati region by Bergström et al.
(2010) as basal Hirnantian rather than upper
Katian (Holmden et al., 2013; Fig. S1 [see
footnote 1]), consistent with the isotopic record
recently recognized in several other latest Ordo-
vician epicratonic sections in North America
(Bergström et al., 2006, 2011). The strata that
show the Elkhorn excursion exhibit only subtle
changes in faunal diversity and dominance (Hol-
land and Patskowski, 2007). However, due to a
break in sedimentation, only the rising limb of
the Elkhorn excursion was documented by Berg-
ström et al. (2010). In comparison, the rising
limb of the lower δ13Ccarb excursion on west-
ern Anticosti Island occurs within the upper-
most part of the Vauréal Formation, below the
main interval of faunal turnover in chitino zoans
(Achab et al., 2011), conodonts (McCracken
and Nowlan, 1988; Zhang and Barnes, 2002),
and brachiopods (Jin and Zhan, 2008).
Timing and Duration of Glaciation
There is now a wealth of evidence that gla-
cial ice packs may have been present through-
out some or all of the Ordovician and Silurian
periods and that advance-retreat cycles may
have infl uenced eustatic sea level during much
of this interval. The episode of continental-
scale glaciation that took place during early
to mid-Hirnantian time was a unique event
within the early Paleozoic, however. Evidence
of this event is recorded in the extensive record
of glacial deposits preserved on many parts of
the Gondwanan continent (e.g., Vaslet, 1990;
Buggisch and Astini, 1993; Caputo, 1998; Sut-
cliffe et al., 2000; Ghienne, 2003, 2011; Young
et al., 2004; Ghienne et al., 2007; Kumpu-
lainen, 2007; Schönian and Egenhoff, 2007; Le
Heron and Dowdeswell, 2009; Le Heron et al.,
2010; Loi et al., 2010; Moreau 2011), and in
well-dated glaciomarine deposits on the Gond-
wanan margin and peri-Gondwanan terranes
(e.g., Robardet and Doré, 1988; Štorch, 1990,
2006; Sačanski, 1994; Monod et al., 2003;
Gutiérrez-Marco et al., 2010; Schönlaub et al.,
2011; Mitchell et al., 2011). These studies also
demonstrated that the Hirnantian included two
major phases of ice advance and retreat, each
consisting of at least two smaller-scale phases
of ice-sheet growth and shrinkage. The Moroc-
can glaciogenic succession exhibits a strong ice
retreat and transgression within the lower part
of the Upper Second Bani Formation (Sutcliffe
et al., 2000; Loi et al., 2010; Ghienne, 2011).
This mid-Hirnantian interglacial event is also
recorded in other Gondwanan localities (e.g.,
Jordan and Saudi Arabia—Armstrong et al.,
2005; Turkey—Monod et al., 2003; Kozlu and
Ghienne, 2012; Niger, Mauritania, and Libya—
Moreau, 2011), as well as peri-Gondwanan
settings (e.g., Štorch, 2006). Although these
cycles manifest differently in different Gond-
wanan localities due to regional differences in
the sedimentary and isostatic response to glacial
advance and retreat, similarity in the overall
glacial-interglacial patterns over such a broad
region (e.g., Moreau, 2011) indicates that these
strata record a broadly shared pattern of major
cycles of ice-volume change.
Studies from more tropical, far-fi eld localities
also show clear isotopic and sequence-strati-
graphic evidence of the effects of the phase of
peak glaciation in the Hirnantian (e.g., Brench-
ley et al., 1994, 2003, 2006; Trotter et al., 2008;
Fan et al., 2009; LaPorte et al., 2009; Ainsaar
et al., 2010; Young et al., 2010; Finnegan et al.,
2011; Jones et al., 2011; Kaljo et al., 2012), and
some paleotropical regions also show sequence-
stratigraphic and chemostratigraphic evidence
for several cycles of glacial ice advance and
retreat within the Hirnantian (Desrochers et al.,
2010; Holmden et al., 2013), including two
widely recognized inter-regional unconformi-
ties referred to as HA and HB by Bergström
et al. (2006, 2011; see also Fig. 2 herein).
Although correlation between the tropical
realm in which the base of the Hirnantian Stage
has been defi ned and the clastic-dominated suc-
cession and cold-water faunas of the Gond wanan
margin remains problematic, the profound
eustatic changes produced by the Hirnantian
glaciation are recorded as lithofacies and geo-
1GSA Data Repository item 2013352, Extended Late
Ordovician-early Silurian correlation chart and iso-
tope data tables, is available at http://www.geosociety
.org/pubs/ft2013.htm or by request to editing@
geosociety.org.
Environmental changes in the Late Ordovician–early Silurian: Review and new insights from black shales and nitrogen isotopes
Geological Society of America Bulletin, November/December 2013 1639
chemical changes that provide an opportunity to
link together events across this range of settings
(Fig. 3). Recent work on these relationships has
helped to the constrain timing of the two main
Hirnantian glacial episodes.
Graptolites and characteristic elements of the
Hirnantia fauna are directly associated with one
another and are closely followed by Hirnan-
tian diamictites in the latest Katian–Hirnan-
tian part of the peri-Gondwanan succession of
the Prague region (Štorch, 2006; Mergl, 2011;
Mitchell et al., 2011). This region occurred on
the peri-Gondwanan Perunica terrane. Although
the paleogeographic position of this terrane in
the latest Ordovician has been a matter of some
debate (for a discussion of this debate, see Fatka
and Mergl, 2009), the evidence for glaciomarine
sediments in the Hirnantian part of the succes-
sion strongly suggests that it was close to the
Gondwanan margin at that time, ~50°S–60°S
according to Rothwell Group, L.P., (2011),
although the most recent paleomagnetic data
suggest mid-Silurian latitudes ~25°S (Tasáryová
et al., 2012). The Prague region succession
includes the basal Hirnantian δ13C excursion
recorded in sedimentary organic matter (δ13Corg;
Mitchell et al., 2011). Similar recent work by
Kaljo and Martma (2011) and Kaljo et al. (2012)
demonstrated the presence of a 3‰ shift in
δ13Ccarb values in northeastern Siberia precisely
at the base of the Hirnantian in association with
an abrupt shift to more shallow-water rocks
(Koren’ et al., 1983; Koren’ and Sobolevskaya,
HB
HA
Figure 2. Carbon, oxygen, and strontium chemostratigraphic features and temperature history of the late Mid Ordovician to early Silurian
time interval. Time scale is from Cooper and Sadler, (2012) and Melchin et al. (2012). Stage slices are from Bergström et al. (2009) and
Cramer et al. (2011). Carbon curve is modifi ed from Bergström et al. (2009) and Cramer et al. (2011). Oxygen and sea-surface temperature
(SST) data are from Finnegan et al. (2011). Strontium curve was derived from data presented in Young et al. (2009). HICE–Hirnantian
isotope carbon excursion; GICE–Guttenberg isotope carbon excursion; HA and HB are lowstand unconformity surfaces defi ned by Berg-
ström et al. (2006).
Melchin et al.
1640 Geological Society of America Bulletin, November/December 2013
2008). Positive shifts in sedimentary δ13Ccarb and
δ13Corg values associated with the base of the
Metabolograptus extraordinarius Zone have
also been reported from several sites in North
America, including Nevada and northern Can-
ada (Melchin and Holmden, 2006a; LaPorte
et al., 2009). New detailed sampling from the
Wangjiawan “riverside” section, very close to
the Hirnantian GSSP at the Wangjiawan-North
road cut, SE China, also reveals an abrupt ~1‰
shift in δ13Corg values immediately beneath the
fi rst appearance of M. extraordinarius (Gorjan
et al., 2012). Based on sedimentological evi-
dence at Dob’s Linn, Southern Uplands, Scot-
land, Armstrong and Coe (1997) placed the start
of the Hirnantian glacial maximum immediately
above Anceps Band D, which contains a diverse
Paraorthograptus pacifi cus Zone fauna. Since
Anceps Band E contains the lowest occurrence
M. extraordinarius, the base of the Hirnantian
must lie between these two graptolite-bearing
bands (Williams, 1982; Chen et al., 2000), near
the onset of the glacial maximum. Thus, it is
clear from these observations that the base of
the Hirnantian closely coincides with the onset
of the fi rst of the two major glacial advances.
Armstrong and Coe (1997) suggested that
Anceps Band E, which represents a return to
graptolite-bearing black shales, marked an
interglacial interval. This interval corresponds
to a decrease in δ13Corg values (Fig. 3) and is the
highest level of occurrence of the Diplograptina
(DDO graptolite fauna of Melchin and Mitchell ,
1991). In the Vinini Creek succession of
Nevada, Diplograptina become rare at the onset
of the Hirnantian but exhibit a strong recurrence
in the upper part of M. extraordinarius Zone
coincident with an ~2.5‰ negative excursion
in δ13Corg values, and additional geochemical
indicators of deepening (Fig. 3; LaPorte et al.,
2009; Štorch et al., 2011; Holmden et al., 2013).
The Anceps Band E and Vinini Creek Dip-
lograptina recurrence levels appear to represent
the mid-Hirnantian interglacial event. Likewise,
the highest occurrence of common and diverse
Diplograptina in the deep-shelf succession at
Wangjiawan, SE China, occurs in the uppermost
beds of the Wufeng Formation. Recent geo-
chemical studies suggest a brief return to more
anoxic-sulfi dic conditions in this same interval,
and in the corresponding Hirnantia fauna–bear-
ing calcareous mudstones of the lower part of the
Kuanyinqiao Formation in the more nearshore
Nanbazi section (Fig. 3; Yan et al., 2009; Gorjan
et al., 2012). On Anticosti Island, Desrochers
et al. (2010) placed the mid-Hirnantian intergla-
cial at the base of their transgressive sequence 3,
based on a particularly strong transgression in
the lower part of the Prinsta Member of the Ellis
Bay Formation. This interval again corresponds
Interglacial
Interglacial
Interglacial
Hirnantian
Hirnantian
HirnantianMid
Mid
Mid
?
?
?
black shale gray shale
siltstone cherty limestone
dolomitic mudstone argillaceous wackestone
calcareous mudstone packstone grainstone bioherms & grainstone limestone breccia-conglomerate
interbedded shale
& limestone
**
*
Diplograptina recurrence FAD of persculptus Zone
graptolites
Di
upper limit
common
Diplograptina
3
15
20 m
Mirny Creek, Siberia
δ
13Ccarb(‰)
shelly
shelly
beds
beds
shelly
beds
Q-70
Q-67
D
Di
*
-1
B
B
A
A
C
C
D
D
E
E
EB
EB
EB
2 m
-33 -31 -29
Birkhill Shale
Birkhill Shale
Birkhill Shale
Hartfell Shale
Hartfell Shale
Hartfell Shale
Dob’s Linn, Scotland
δ
13Corg (‰)
acuminatus
Zone
vesiculosus Zone
cyphus Zone
Di
*
TR1
TR2
TR4
TR5
Grindstone Mbr.
Lafram. *
20 m
Ellis Bay Fm.
Ellis Bay Fm.
Ellis Bay Fm. Fox Point Fm.
Fox Point Fm.
Fox Point Fm.
43210-1
δ13Ccarb(‰)
Velleda
Lousy
Cove
S.C.
Mill Bay Vauréal Fm.
Vauréal Fm.
Vauréal Fm.
-2 0 24
East End
West End δ13Ccarb(‰)
TR5
Lafram.
TR2
TR1
B. gamachiana
B. gam.
20 m
TR3
TR3
TR3
Prinsta
Prinsta
Prinsta
TR3
TR3
TR3
TR4
TR4
TR4
extra-
ordinarius persculptus
pacificus
typicus
mirus
lower
complexus
Katian Hirnantian Rhuddanian
Ordovician Silurian
ascensus
acuminatus
Anticosti Is., Que.Wangjiawan, Hubei
grap.
zones
Stages
Periods
Wufeng Fm.
Wufeng Fm.
Wufeng Fm. Lungmachi Fm.
Lungmachi Fm.
Lungmachi Fm.
Hirnantia
Hirnantia
beds
beds
Hirnantia
beds
443.8445.2446.5
1.0 m
2.0 m
Lungmachi Fm.
Lungmachi Fm.
Lungmachi Fm.
Wufeng Fm.
Wufeng Fm.
Wufeng Fm.
Kuanyinqiao Fm.
Kuanyinqiao Fm.
Kuanyinqiao Fm.
shelly
shelly
beds
beds
shelly
beds
δ13Corg (‰)
-28 -26-30
δ13Corg (‰) δ34S (‰)
-20 0 20
-30 -29 -28
vesiculosus
Zone
cyphus
Zone
Nanbazi,
Guizhou
pacificus
Zone
complexus
Zone
ascensus Zone
MA
Hirnantia
Hirnantia
beds
beds
Hirnantia
beds
Di
Di
GSSP
River-side
*
*
4.0 m
hiatus
Vinini Creek, Nevada
Vinini Formation
Vinini Formation
Vinini Formation
**
**
Di
**
*
δ
15NTN (‰)
-1 0 1 2
-32-31-30-29-28
δ13Corg (‰)
Figure 3. Correlation of key well-studied late Katian (Ordovician) to early Rhuddanian (Silurian) stratigraphic successions relative to the geochronological time scale and
graptolite zones commonly employed in this interval. Correlations are also based on carbon isotope chemostratigraphy and sequence stratigraphy, especially the stratigraphic
and paleontological expression of the mid-Hirnantian interglacial episode. TR—transgressive-regressive sequence. Data sources: Wangjiawan: Fan et al. (2009) and Gorjan
et al. (2012); Nanbazi: Yan et al. (2009, 2012); Vinini Creek: LaPorte et al. (2009), Štorch et al. (2011), Holmden et al. (2013); Anticosti Island: Desrochers et al. (2010), Achab
et al. (2011), Jones et al. (2011); Dob’s Linn: Williams (1982, 1983), Underwood et al. (1997), Armstrong and Coe (1997), Melchin et al. (2003); Mirny Creek: Koren’ et al.
(1983), Koren’ and Sobolevskaya (2008), Kaljo et al. (2012). FAD—fi rst appearance datum; GSSP—global stratotype section and point.
Environmental changes in the Late Ordovician–early Silurian: Review and new insights from black shales and nitrogen isotopes
Geological Society of America Bulletin, November/December 2013 1641
to a brief drop in δ13Corg values at both the east
and west ends of the island (Fig. 3). Relative to
the Anticosti chitino zoan biostratigraphy, the
interval of the mid-Hirnantian interglacial falls
within the Belonechitina gamachiana Zone
(Achab et al., 2011). The TR-3 sequence bound-
ary most likely corresponds to the HA uncon-
formity of Bergström et al. (2006, 2011). In the
succession at Mirny Creek, Siberia, the mid-
Hirnantian interglacial transgression appears to
be present in the upper part of Unit Q-68, where
it is marked by numerous carbonate debris-
fl ow beds, and is quickly followed by the low-
est occurrence of Metabolograptus persculptus
(Kaljo et al., 2012), as it is also at Vinini Creek
(Štorch et al., 2011).
Major regression of the second and largest of
the Hirnantian glacial episodes encompasses the
largest and most widespread positive shift of the
Hirnantian carbon isotope excursion. In every
well-dated graptolitic succession, the main
Hirnantian carbon isotope excursion peak falls
in the lower-middle part of the M. persculptus
Zone (e.g., Mirny Creek, Dob’s Linn, Vinini
Creek) or its correlative strata (Wangjiawan).
At Anticosti Island, this event is represented in
TR-4 and TR-5, in the upper part of the Lousy
Cove and Laframboise members of the Ellis
Bay Formation. The former includes an occur-
rence of Metabolograptus parvulus, which indi-
cates a M. persculptus Zone or slightly younger
age for the overlying Laframboise strata
(Melchin, 2008).
The last main phase of Hirnantian glaciation
ended in the middle part of the late Hirnan-
tian M. persculptus Zone (e.g., Brenchley
et al., 1994; Underwood et al., 1998; LaPorte
et al., 2009; Moreau, 2011; Kaljo et al., 2012).
According to the time scale of Cooper and
Sadler (2012), the duration of the Hirnantian
Stage was ~1.4 m.y. It is important to note,
however, the duration of the Hirnantian Stage
is not tightly constrained by high-resolution
radiometric dates, but is based on interpolation
between dated horizons in overlying and under-
lying intervals, and the estimated uncertainty on
the duration of the Hirnantian Stage is ±0.2 m.y.
(Cooper and Sadler, 2012, their Table 20.2).
Based on the combined biostratigraphic and
chemostratigraphic record, as described here,
the peak glacial interval occupies approximately
the lower 70%–80% the Hirnantian Stage and,
therefore, could be ~0.8–1.3 m.y. in duration.
As noted already, the glacial interval consisted
of two main phases of ice advance and retreat,
each of which could include at least two to three
smaller-scale cycles. If we accept that fi ve glacial
advance-retreat cycles can be recognized in both
Gondwanan (e.g., Sutcliffe et al., 2000; Moreau,
2011) and far-fi eld localities (Desrochers et al.,
2010; Holmden et al., 2013) over a duration of
~1 m.y., then the average cycle duration would
be ~200 k.y. This average suggests the possi-
bility that the cyclicity observed through this
interval was orbitally forced (e.g., Sutcliffe
et al., 2000; Armstrong et al., 2005; Armstrong,
2007; Desrochers et al., 2010). Boulila et al.
(2011) documented obliquity cycles of 200 k.y.
duration through long-period obliquity cycles
of 1.2 Ma in the Cenozoic, and 405 k.y. eccen-
tricity cycles are also well documented (Hin-
nov and Hilgen, 2012). These patterns match
well with the patterns seen in the Hirnantian.
In addition, Williams (1991), in an analysis
of cyclicity in Late Ordovician–early Silurian
evaporites in Australia, found spectral peaks
at ca. 206–233 k.y., as well as several shorter-
term cycles that were consistent with Ordovi-
cian–Silurian orbital parameters. Most recently,
Elrick et al. (2013) documented synchronous
changes in δ18O (derived from cono donts) and
relative water depth through 17 successive
meter-scale parasequences from early to late
Katian rocks at two Laurentian epicratonic
sites. This result is the most direct test to date
for the role of orbitally forced, eustatic driving
of depositional sequences in the run-up to the
Hirnantian glacial episode. Elrick et al. (2013)
suggested that these meter-scale cycles have
~100 k.y. durations, but at present, such esti-
mates of cycle durations are based on average
durations calculated from what are rather impre-
cisely dated more inclusive intervals. There is
currently no means of precisely constraining
the ages or durations of any of these individual
cycles within the Katian and Hirnantian glacial
intervals to determine that they are of similar
duration or to which of the orbital cycles they
truly correspond.
As noted previously, it now appears likely
that there were episodes of ice-sheet growth on
Gondwana during the Ordovician prior to the
large-scale Hirnantian glacial cycles. Frakes
et al. (1992) suggested that glaciation began
in the Middle Ordovician, signifi cantly predat-
ing the main Hirnantian phase of glaciation.
Brenchley et al. (1994) argued, however, on
the basis of available oxygen isotope evidence,
that Gondwanan glaciation was restricted to
the Hirnantian, advocating a very short-lived
deteriora tion of early Paleozoic climate in what
was dominantly a long period of greenhouse
conditions. Recent interpretations, based mainly
on indirect evidence from carbon isotopes or
the pattern of sedimentary cycles (or both),
indicated that the onset of Ordovician glacia-
tion may have begun in early Late Ordovician
time (e.g., Hamoumi, 1999; Pope and Steffen,
2003; Saltzman and Young, 2005; Loi et al.,
2010; Holland and Patzkowsky, 2012), with
fl uctuating, relatively small-scale Gondwanan
ice sheets until the Hirnantian (e.g., Loi et al.,
2010; Holmden et al., 2013). However, recently
published oxygen isotope data on conodonts
provides compelling evidence that signifi cant
global cooling dates back to at least the late
Middle Ordovician or early Late Ordovician
(Trotter et al., 2008; Rosenau et al., 2012; Elrick
et al. 2013). Those results are supported by
clumped-isotope analysis of carbonate fossils,
which suggest a coordinated reduction in sea-
surface temperatures and elevation of seawater
δ18O values, indicative of both climate cooling
and the presence of continental ice through the
Katian (Finnegan et al., 2011). High-frequency
fl uctuations in carbon isotope records in the
Darriwilian (upper Middle Ordovician) have
also led to suggestions of glacially infl uenced
climate changes in this interval (e.g., Ainsaar
et al., 2010; Zhang et al., 2011).
Although the latest Hirnantian–early Llando-
very is widely regarded as a time of signifi cant
glacial retreat, there is considerable evidence for
the persistence of smaller ice sheets both in the
North African (Moreau, 2011; Le Heron et al.,
2013) and South American parts of Gondwana
(see Page et al., 2007). There is sedimentologi-
cal evidence for at least three phases of ice-sheet
advance in the early Silurian of South America
(Caputo, 1998; Díaz-Martínez and Grahn, 2007;
Díaz-Martínez et al., 2011), which again is cor-
roborated by clumped-isotope analysis of car-
bonate fossils showing elevated seawater δ18O
values in the Llandovery (Finnegan et al., 2011).
Some authors have also suggested, based on C
and O isotopic evidence, that there were epi-
sodes of glacial ice advance in the Ludlow and
Pridoli (Kaljo et al., 1998; Lehnert et al., 2007,
2010; Calner, 2008; Žigaitė et al., 2010).
In summary, the available stratigraphic and
isotopic evidence now clearly indicates that con-
tinental ice sheets existed, at least intermittently,
through all or most of the Late Ordovician and
early Silurian, with peak ice volumes occurring
during the Hirnantian. Additionally, signifi cant
ice caps may have begun to form in the Early
Ordovician and may have persisted into later
Silurian times. Page et al. (2007) referred to
this interval as the early Paleozoic icehouse,
and it lasted at least 20 m.y. (from the Sand-
bian or early Katian through to the Llandovery),
although the peak glacial interval of the early-
mid-Hirnantian likely lasted just 0.85–1.2 m.y.
Cause of Glaciation
Many authors have proposed hypotheses con-
cerning the cause of the Late Ordovician glacia-
tion. One of the challenges in understanding this
event has been the evidence that atmospheric
Melchin et al.
1642 Geological Society of America Bulletin, November/December 2013
CO2 levels were some 8–20 times higher than
present atmospheric levels (PAL) during Late
Ordovician time, at least prior to the peak gla-
cial interval (Yapp and Poths 1992; Berner,
2006; Arvidson et al., 2006; Nardin et al., 2011).
The possibility that solar luminosity may have
been 4.5% lower (Herrmann et al., 2004) may
reduce this incongruity somewhat. Furthermore,
some very recent studies have suggested that
Late Ordovician pCO2 was less than 8 times
PAL (Vandenbroucke et al., 2010; Pancost et al.,
2013). It is clear that further work is required
to resolve the discrepancies implied by these
varied results.
Most hypotheses that seek to explain the Late
Ordovician glaciation have focused on mecha-
nisms for removing CO2 from the atmosphere
in advance of the glacial interval, believing
that global surface temperatures were closely
coupled with atmospheric CO2 concentra-
tions in the Paleozoic (e.g., Came et al., 2007).
Weathering of silicate crustal rocks is a sink
for atmospheric CO2 (Berner, 1991; Kump and
Arthur, 1999), and several authors have pro-
posed that this process led to cooling during
the Late Ordovician (e.g., Young et al., 2009).
Kump et al. (1999) suggested that the regional
crustal uplift produced by the Taconic orogeny
resulted in increased rates of silicate weathering
beginning in late Middle Ordovician time. In
addition, Finlay et al. (2010) provided Os iso-
tope data that they interpreted as indicating an
increase in the rate of silicate weathering dur-
ing the late Katian, which they attributed to the
effects of the Caledonian orogeny, although tec-
tonic uplift and erosion were taking place in the
British part of the Caledonides throughout much
of Middle and Late Ordovician (Armstrong and
Owen, 2001; Oliver, 2001). Indeed, signifi cant
mountain building was also taking place around
the margin of Gondwana in regions that pres-
ently occur in western Argentina (Astini and
Dávila, 2004; Vujovich et al., 2004; Astini et al.,
2007; van Staal et al., 2011), on the eastern mar-
gin of Australia (Foster and Gray, 2000; Fergus-
son, 2003), as well as in South China and the
Siberian Kip arc (Chen and Mitchell, 1996; de
Jong et al., 2006; Wang et al., 2006; van Staal
and Hatcher, 2010).
Young et al. (2009) addressed the signifi -
cance of a strong decrease in 87Sr/86Sr ratios
through late Darriwilian to late Sandbian time,
which they attributed to increased weathering of
basalts through the late Middle and Late Ordo-
vician. They suggested that although the rates
of CO2 outgassing during volcanism would
balance the increased rates of CO2 consump-
tion from weathering during the initial phases
of basalt emplacement, the weathering would
continue well after the volcanic activity ceased,
leading to lower CO2 levels in Katian time,
immediately before the Hirnantian glacial inter-
val. Lefebvre et al. (2010) proposed that forma-
tion of a large igneous province in late Katian
time could have led to an initial phase of global
warming followed by cooling through negative
feedback mechanisms, leading to the Hirnantian
glaciation. They noted, however, that compel-
ling evidence of a late Katian large igneous
province is lacking.
It has also been proposed that an initial
phase of global cooling and possibly glaciation
occurred in Sandbian time, associated with the
widely recognized Guttenberg carbon isotope
excursion (GICE) (e.g., Saltzman and Young,
2005; Page et al., 2007; Young et al., 2008; Pan-
cost et al., 2013). In addition, the possible asso-
ciation between this C isotope event, a positive
δ18O excursion, and the deposition of very thick
and widespread volcanic ash units led Buggisch
et al. (2010) to suggest that this cooling may
have been triggered by intense volcanism due
to production of high quantities of sun-blocking
volcanic ash and aerosols produced by the erup-
tions. More recently, high-resolution records
of paired δ13C and δ18O indicate that there is
no clear coincidence between a brief interval
of cooling, as suggested by the δ18O data, and
either the time of deposition of the ash beds or
the peak of the δ13C excursion (Herrmann et al.,
2010, 2011; Rosenau et al., 2012). Nevertheless,
as noted earlier, there is clear oxygen isotope
evidence of global cooling and the presence of
glacial ice through all or most of Late Ordovi-
cian time (Shields et al., 2003; Trotter et al.,
2008; Finnegan et al., 2011; see Fig. 2), and this
cooling was coincident with a prolonged phase
of explosive volcanism (Sell, 2011).
Young et al. (2010) used stratigraphic records
of Δ13C values reconstructed from paired δ13Ccarb
and δ13Corg profi les through the Hirnantian suc-
cession from Anticosti Island to suggest that
during the interval of peak glaciation within
the Hirnantian, atmospheric CO2 began to rise
in response to ice-sheet expansion and reduced
rates of silicate weathering, but may also have
been a result of relatively low rates of organic
carbon burial in the then-well-mixed oceans.
Their interpretation was in alignment with
numerical modeling results of Kump et al.
(1999), which were partly informed by a simi-
lar fi nding of a positive Δ13C excursion in the
Monitor Range section of Nevada. These inter-
pretations have been questioned, however, by
Melchin and Holmden (2006a) and LaPorte
et al. (2009), who reported large differences in
the individual records of Hirnantian δ13Ccarb and
δ13Corg excursions reconstructed from different
Hirnantian settings and successions and, thus,
concluded that Δ13C profi les are an unreliable
proxy for paleo-pCO2 trends through the Hirnan-
tian glaciation. Recent results by Metzger and
Fike (2013) demonstrated within-bed variations
of as much as 2‰ in δ13Ccarb and δ18Ocarb val-
ues obtained from even very carefully screened
samples of Late Ordovician carbon ates, which
may arise as a result of heterogeneity in allo-
chem sources within the sample. Such variation
also argues for considerable caution in the use
of Δ13C profi les as proxies for the composition
of the marine carbon reservoir (or for use of
δ13Ccarb values in detailed chemostratigraphic
correlations, among other implications).
Another potential sink for atmospheric CO2
is burial of organic matter. Page et al. (2007)
recently reviewed the potential for changing
rates of burial of organic matter as a mecha-
nism for modulating glacial-interglacial cycles
through the Late Ordovician–early Silurian gla-
cial interval. We return to a more detailed con-
sideration of carbon burial hypothesis and its
relationship to the Hirnantian glacial interval in
the Discussion section following presentation of
our results.
Some studies have pointed to the importance
of paleogeographic factors in controlling Late
Ordovician changes in climate. In particular,
changing positions of continents relative to
the South Pole and the paleotropics may have
affected the global distribution of heat and also
rates of continental weathering (Herrmann et al.,
2004; Armstrong, 2007; Nardin et al., 2011).
Nardin et al. (2011) provided quantitative esti-
mates of the relative importance of changing
paleogeography and silicate weathering rates as
factors that could infl uence the observed pattern
of global cooling through the Late Ordovician–
early Silurian. Armstrong (2007), in particular,
emphasized many close parallels between the
development of phases of glaciation through
the Ordovician and the late Cenozoic, in terms
of the interaction among declining atmospheric
CO2 and changing weathering rates, patterns
of nutrient cycling, and heat transport result-
ing from changing continental positions. He
also suggested that changes in solar insolation
through orbital cycles played a key role in con-
trolling the patterns of ice advance and retreat.
Lenton et al. (2012) provided both experi-
mental and modeling evidence to suggest that
successive phases of colonization of land, fi rst
by nonvascular plants and, later, by the earliest
vascular plants in Middle and Late Ordovician
time, had a signifi cant impact on patterns and
rates of silicate weathering, nutrient fl ux, carbon
burial, and changes in atmospheric pCO2. They
proposed that those changes contributed to the
onset of global cooling in Darriwilian-Sandbian
time, as well as the later pulse of cooling and
maximum glacial expansion in the Hirnantian.
Environmental changes in the Late Ordovician–early Silurian: Review and new insights from black shales and nitrogen isotopes
Geological Society of America Bulletin, November/December 2013 1643
It should be noted, however, that the available
record indicates that there was an important
phase of cryptospore diversifi cation during the
Hirnantian (Vecoli et al., 2011), rather than in
the late Katian, as required by this model (see
discussion later herein).
Boda Event—Preglacial Warming?
Several authors have noted that the Hirnan-
tian glacial event was preceded by an interval
marked by the spread of both shelf carbonates
and faunas interpreted to represent warm-water
assemblages into relatively high-latitude set-
tings in Katian time (e.g., Brenchley, 1989;
Villas et al., 2002; Boucot et al., 2003). Fortey
and Cocks (2005) named this interval the Boda
event. Subsequently, there has been consider-
able dispute over the timing and duration of
this event (e.g., Kaljo et al., 2011), as well as
whether the lithological and faunal signals rep-
resent global warming (e.g., Armstrong, 2007;
Armstrong et al., 2009a; Jimenéz-Sánchez and
Villas, 2010) or global cooling (Cherns and
Wheeley, 2007, 2009). Armstrong et al. (2009a)
published records of δ18O data from three widely
separated regions and argued that consistency
of the timing and patterns between the regions
indicated that the data represented a relatively
reliable primary signal despite the fact that they
were from whole-rock analyses of carbonates.
The warming event recognized by Armstrong
et al. (2009a, their Fig. 4) as the Boda event was
restricted to the late Katian P. pacifi cus Zone or
upper part of the Ka4 time slice of Bergström
et al. (2009). In contrast, other authors have sug-
gested that the Boda event may have spanned
much of the late Katian (e.g., Boucot et al., 2003;
Ainsaar et al., 2010; Fig. 2 herein). Finnegan
et al. (2011) also showed what appears to be a
brief warming event in the latest Katian based
on clumped oxygen isotope data. Indeed, the
combination of sea-surface temperature values
inferred by Trotter et al. (2008) and Finnegan
et al. (2011; replotted herein in Fig. 2) sug-
gest an extended period of warming in the late
Katian (Pleurograptus linearis Zone to P. pacifi -
cus Zone), interrupted by a brief, pre-Hirnantian
cooling event. Interestingly, detailed facies and
sequence-stratigraphic analysis of the Katian to
Hirnantian shelf succession in Morocco reveals
the presence to two large-scale glacio-eustatic
depositional sequences within the interval typi-
cally identifi ed as the Boda event (Videt et al.,
2010; Loi et al., 2010). The bryozoan-dominated
carbonate succession there is largely confi ned to
the earlier of these two sequences (Álvaro et al.,
2007; Loi et al., 2010) and thus is older than the
late Boda, P. pacifi cus Zone event examined by
Armstrong et al. (2009a).
Mass Extinction Timing and Mechanism
Some recent data compilations suggest that in
terms of magnitude of taxon loss, the two phases
of the Hirnantian extinction were the second
most severe of the Phanerozoic, after the Per-
mian-Triassic extinction (e.g., Bambach, 2006),
although a more recent, sampling-standardized
analysis by Alroy (2008) suggests, rather, that
the end-Triassic and end-Cretaceous events pro-
duced greater taxonomic loss. Relative to other
major mass extinction events of the Phanero-
zoic, those of the Hirnantian appear to have had
less impact on the basic structure of the marine
ecosystem (Droser et al., 2000; Brenchley
et al., 2001; McGhee et al., 2004, 2012; Bam-
bach, 2006) and long-term biodiversity levels
(Alroy, 2010).
The fact that the boundaries for the bases of
the Hirnantian and Rhuddanian are both defi ned
by GSSPs in graptolitic shales, with scarce or
absent conodont or benthic faunas, has made
it diffi cult to precisely correlate the two phases
of Hirnantian extinction between shallow-shelf
and basinal successions. Notwithstanding those
diffi culties, recent, high-resolution studies
have also shown that, at least for some groups,
diversities decline gradually through the latest
Katian to early-mid-Rhuddanian (e.g., Chen
et al., 2005; Rasmussen and Harper, 2011a;
Sadler et al., 2011), and that the intensity of
extinctions and relative magnitudes of the early
and later extinction peaks vary signifi cantly
geographically (e.g., Chen et al., 2003; Krug
and Patzkowski, 2007; Rasmussen and Harper,
2011a, 2011b; Finnegan et al., 2012). In addi-
tion, the extinction and subsequent diversifi -
cation have been shown to be selective, both from
a phylogenetic point of view (Chen et al., 2005;
Rong et al., 2006; Mitchell et al., 2007; Sadler
et al., 2011; Bapst et al., 2012), and based on
habitat and geographic range (Sheehan, 2001;
Cooper and Sadler, 2010; Goldman et al., 2011;
Finnegan et al., 2012), although in the case of
graptolites, at least, extinction was not selec-
tive with respect to colony morphology despite
signifi cant clade-level effects on extinction risk
(Bapst et al., 2012).
Recent studies investigating the cause of the
phases of mass extinction have focused mainly
on the connections between the trends in bio-
diversity change and the environmental changes
outlined earlier herein, particularly global cool-
ing in general (e.g., Stanley and Powell, 2003;
Stanley, 2010; McGhee et al., 2012), and the
associated shifts and constrictions in global
climate belts (e.g., Vandenbroucke et al., 2010;
Finnegan et al., 2012). Changes in ocean-water
chemistry, particularly levels of oxygenation and
nutrients, are widely regarded as being important
in relation to zooplankton assemblages, such as
graptolites (e.g., Chen et al., 2005; Finney et al.,
2007). In addition to the paleogeographic shift-
ing of global climate belts, changes in conti-
nental and microcontinental plate confi guration
have been suggested to be an important factor
in the long-term decline in diversity exhibited
by some taxa (e.g., Krug and Patzkowski, 2007;
Rasmussen and Harper, 2011a, 2011b). In par-
ticular, Rasmussen and Harper (2011a) showed
quantitatively that the progressive closure of
the Iapetus Ocean and amalgamation of Iapetan
microcontinents was a critical factor through the
Ordovician-Silurian transition in terms of loss
of γ-diversity in brachiopods. Most recently,
Hammarlund et al. (2012) suggested that the
mass extinction occurred in association with a
widespread anoxic event in the deep oceans dur-
ing the peak glacial interval, which resulted in
high rates of extinction among deep-water ben-
thic and nektonic organisms, at both the onset
and end of the event.
Melott et al. (2004) and Melott and Thomas
(2009) proposed that the Late Ordovician
extinction resulted from the effects of a cosmic
gamma-ray burst. Although the authors noted
that such an event would not leave a recog-
nizable physical or geochemical stratigraphic
record, they argued that their hypothesis makes
specifi c, testable predications about the paleo-
envir onmen tal and paleogeographic patterns of
extinction and their timing in relation to climate
change (e.g., Melott and Thomas, 2009), but
these predictions have not yet been tested.
Anoxia and Black Shales
Since 1988, there have been many papers and
volumes summarizing our general understand-
ing of black shales and marine anoxia (e.g.,
Arthur and Sageman, 1994; Wignall, 1994;
Strauss, 2006; Negri et al., 2009; Jenkyns, 2010;
Trabucho-Alexandre et al., 2012). Some have
focused specifi cally on euxinia (e.g., Meyer
and Kump, 2008), or the formation or petro-
leum source rocks (e.g., Harris, 2005; Bazhe-
nova, 2009). Still other studies have specifi cally
considered the role of sea-level change (e.g.,
Landing , 2011), large igneous provinces (Kidder
and Worsley, 2010; Lefebvre et al., 2010), nutri-
ent cycling (e.g., Van Cappellen and Ingall,
1994; Saltzman, 2005; Slomp and Van Cappel-
len, 2007; Tsandev et al., 2008, 2010; Tsandev
and Slomp, 2009; Palastanga et al., 2011; Ozaki
et al., 2011), or changes in eco system structure
(Butterfi eld, 2009, 2011). Some studies have
also highlighted the importance of paleogeo-
graphic confi guration on the development of
marine anoxia and black shales (e.g., Meyer
and Kump, 2008; Trabucho-Alexandre et al.,
Melchin et al.
1644 Geological Society of America Bulletin, November/December 2013
2012). Negri et al. (2009) particularly focused
on some important temporal differences among
the Paleozoic, Mesozoic, and Cenozoic in terms
of climate, biotic change, plate confi gurations,
temporal resolution, and distribution and dura-
tion of anoxic events that often make it diffi cult
to apply uniformitarian concepts to Paleozoic
black shales.
In concert with the development of our
understanding of the depositional conditions
necessary for formation of marine dysoxia to
anoxia and black shales, there have been major
advances in methods of determining and docu-
menting ancient redox conditions in deposi-
tional sequences (e.g., Allison et al., 1995;
Wilde et al., 2004; Boyer and Droser, 2009;
Lyons et al., 2009; Severmann and Anbar,
2009; Jenkyns, 2010; Boyer et al., 2011; Zhou
et al., 2011, 2012; Thompson and Kah, 2012).
Some of these studies have shown that although
many black shales represent deposition within
anoxic bottom waters, some represent dysoxic
depositional conditions. Space does not permit a
compre hensive review and synthesis of the con-
clusions of these studies. Here, we will consider
the specifi c depositional models that have been
proposed recently for some Late Ordovician and
earliest Silurian black shales. In addition, in the
Discussion, we will consider some of this recent
work in reference to our new data.
Studies of Mesozoic and Cenozoic organic-
rich marine sediments show that formation of
these sediments can be the result of a complex
interplay of many factors. Some of the factors
most relevant to the present study include:
(1) hydrographical setting of the marine basin;
(2) changes in sea level and climate;
(3) rate of nutrient input into oceans;
(4) rate of nutrient recycling within oceans;
(5) atmospheric oxygen levels; and
(6) changes in the biosphere.
We will consider each of these factors next as
they pertain to the Late Ordovician–early Silu-
rian interval.
Hydrographical Setting
Berry (2010) reviewed some of the well-
known occurrences of Ordovician black shales
in relation to modern analogue environments
and concluded that these ancient black shales
were formed primarily in three different depo-
sitional settings: shallow epeiric seas, silled
epicontinental basins, and shelf-edge upwell-
ing zones. The black shales that occur within
interior, relatively shallow epeiric seas, such as
those of the North American midcontinent (e.g.,
Mitchell and Bergström, 1991; Goldman and
Bergström, 1997) and the Baltic Platform (Lille,
2003), may have formed in stratifi ed basins
characterized by a fresh to brackish surface
layer and quasi-estuarine pattern of circulation
(Witzke, 1987). Nutrient input from continental
runoff promoted high productivity and anoxia
below the surface mixed layer.
Black shales that developed in silled epi-
conti nental basins appear to have arisen in part
as a consequence of intrabasinal circulation
that produced regional upwelling of nutrients
to drive productivity. The Katian succession
of the Yangtze Platform may be an example
of this type of setting (see, for example, Chen,
1984; Chen et al., 1987; Rong and Chen, 1987;
among others).
Marine upwelling zones overlying outer
continental shelf, slope, and ocean basin set-
tings are also well represented in the late Katian
black shales of the Vinini Formation in Nevada
(Finney and Berry, 1997; Finney et al., 2007;
LaPorte et al., 2009), the Phi Kappa Formation
of Idaho (Dover et al., 1980), and the Moffatt
Shales of Dob’s Linn, in the Southern Uplands
of Scotland (Armstrong and Coe, 1997). Both
the Vinini and Moffatt Shale successions are
interpreted to have been deposited in oceanic
settings beyond the edge of the continental shelf
(Emsbo, 1993; Finney et al., 2000, 2007; Need-
ham, 2004; Sawaki et al., 2010; Stone, 1995;
Stone et al., 2003; Strachan, 2012). In addition,
late Katian black shales intermittently deposited
on the slope of the Welsh Basin have been inter-
preted as being the result of deposition in a mar-
ginal upwelling system on the basis of several
lines of sedimentary, geochemical, and paleo-
geographic evidence (Challands et al., 2009).
Upwelling conditions and enclosed or semi-
enclosed marine basins with restricted circulation
are both circumstances commonly associated
with oxygen depletion below the surface mixed
layer in modern and ancient settings (e.g., Negri
et al., 2009). With the obvious exception of the
Black Sea, oxygen depletion in these settings
in the modern oceans seldom proceeds beyond
dysoxic conditions. That is, currently, these
hydro graphical conditions do not, by them-
selves, appear to routinely lead to the conditions
of intense oxygen depletion (anoxia or euxinia)
that exclude in situ benthos and result in lami-
nated black muds. There must be other factors
infl uencing this difference in outcome between
Ordovician and recent times, as we summarize
in the following.
Sea Level and Climate
Global sea levels were higher in the Late
Ordovician than at any time in the Cenozoic
(e.g., Haq and Schutter, 2008). The potential
for spread of oxygen-depleted slope waters into
shelf settings was, therefore, much greater than
at present. Transgressive to highstand inter-
vals are also times of warming of shelf waters
accompanied by the development of stable strat-
ifi cation and nutrient recycling in deep-shelf and
slope settings (e.g., Potter et al., 2005; Landing,
2011). This can lead to development of anoxia
and the spread of black shales. We discuss the
issues related to nutrient cycling separately
in the following, and focus here on the physi-
cal effects.
Many authors have focused on the important
differences between black shales of transgres-
sive systems versus highstand systems in the
interpretation of the processes of deposition and
organic matter production versus preservation
(e.g., Wignall, 1991, 1994; Harris, 2005). Many
of the studies of Ordovician–Silurian black
shales, however, have not been documented
within a well-resolved sequence-stratigraphic
framework. Further work will be required to
document the precise relationships between
black shale occurrence and sea-level change.
Indeed, many of the thicker black shale intervals
may span one or more transgressive-highstand
intervals. Therefore, for the following discus-
sion, we will refer to those black shales that
appear to have developed during intervals of ris-
ing to high sea level as transgressive/highstand.
The link between transgression/highstand and
anoxia is particularly clear in the widespread
black shales of the Sandbian–early Katian
(e.g., Leggett, 1980; Page et al., 2007; Land-
ing, 2011). Some regional sea-level curves also
indicate a transgressive to highstand phase in the
late Katian (Munnecke et al., 2010), which may
have been a factor in the development of black
shales in this interval. For instance, Page et al.
(2007) associated the late Katian pulse of black
shale deposition with a transgression/highstand.
Armstrong et al. (2009a) noted that this interval
of widespread black shale deposition is coinci-
dent with the Boda event, although as discussed
earlier herein, the interval they examined cor-
responds only to the second or younger phase
of that warming interval (Fig. 3). Therefore, the
late Katian black shale occurrences appear pri-
marily to have been deposited during an interval
of sea-level rise to highstand associated with a
temporary retreat of glacial ice in Gondwana.
Several authors have proposed that relatively
high sea levels and high sea-surface tempera-
tures during the Ordovician led to sluggish ocean
circulation, and that this played a role in the
enhanced levels of localized anoxia such as those
that developed during the early Katian (sum-
marized in Landing, 2011). Several modeling
studies of ocean circulation using both modern
and ancient continental confi gurations, however,
have suggested that changes in sea level and rates
of ocean circulation, by themselves, can only
result in development of oceanic anoxia under
conditions of extremely low equator-to-pole
Environmental changes in the Late Ordovician–early Silurian: Review and new insights from black shales and nitrogen isotopes
Geological Society of America Bulletin, November/December 2013 1645
temperature gradients (e.g., Bjerrum et al., 2006;
Ozaki et al., 2011; Montenegro et al., 2011).
The available evidence suggests that the late
Katian was a time when at least some glacial ice
existed at high latitudes (Finnegan et al., 2011),
and it was also a time of signifi cant paleolatitu-
dinal differences in planktonic faunas (Vanden-
broucke et al., 2010; Goldman et al., 2011). It
is therefore unlikely that the equator-pole tem-
perature gradient was low enough, by itself, to
induce global-scale ocean stagnation and anoxia,
although local anoxia in silled basins or epeiric
seas with limited exchange (e.g., Holm den et al.,
1998) may well have been enhanced by high
sea levels and sea-surface temperatures during
warmer interglacial phases, such as the Boda
event (Armstrong et al., 2009a).
Rates of Nutrient Input
Intervals of warmer, humid climate, which
we also expect to accompany high sea levels,
are likely to result in increased rates of weath-
ering and nutrient release from the continents,
as well as increased stratifi cation in shelf seas
and epicontinental seas due to freshwater input
(Armstrong et al., 2009b; Landing, 2011).
Saltzman (2005) discussed the role of nutrients
in Paleozoic ocean systems, particularly phos-
phorus and nitrogen, and considered their role
in the development of oceanic anoxia. Based
in part on the work of Pope and Steffen (2003)
and Pope (2004), who focused on upwelling-
related features in the Late Ordovician shelf set-
ting of western Laurentia, Saltzman suggested
that the Late Ordovician–early Silurian was a
time of glacial-interglacial cycles, with well-
ventilated oceans during the glacial periods.
These authors also suggested that intervals of
increased upwelling, continental weathering,
or both, could enhance the rate of P input into
surface waters, resulting in enhanced productiv-
ity and spread of anoxia in deeper waters, par-
ticularly during interglacial periods. Spread of
anoxia results in more effi cient remobilization
of P (see following), thereby enhancing the fur-
ther spread of anoxia (e.g., Van Cappellen and
Ingall, 1994). These suggestions are supported
by recent quantitative modeling studies (e.g.,
Tsandev and Slomp, 2009; Ozaki et al., 2011).
Rates of Nutrient Recycling
Changes in the rate of phosphorus remo-
bilization may also have contributed to the
enhanced development of localized anoxia in
Late Ordovician to early Silurian epeiric seas,
silled basins, and marginal upwelling zones.
Under well-oxygenated conditions, reactive P in
the water column (including particulate organic
phosphorus, and dissolved organic and inor-
ganic P) is readily buried in marine sediments,
where dissolved P or organically bound P can be
precipitated by bacterially mediated processes
as phosphatic minerals or be bound to iron oxy-
hydroxide minerals (Slomp and Van Cappellen,
2007). Under conditions of severely reduced
oxygen concentration in bottom waters, reactive
P is remobilized from sediment into the water
column, again primarily by bacterially mediated
processes (Ingall et al., 1993; Van Cappellen and
Ingall, 1994), where it fuels further productiv-
ity, permitting the further spread and persistence
of anoxia. Given that 70%–95% of the world’s
organic matter and its associated P are depos-
ited in shelf seas (e.g., Bjerrum et al., 2006),
episodes of sea-level rise that bring oxygen-
depleted waters onto the shelf would likely fuel
increased productivity.
Another factor related to nutrient recycling
is the rate of deposition of skeletal calcium
phosphate. As highlighted by Slomp and Van
Cappellen (2007), deposition of P in the form
of fi sh debris is a key sink for P in highly pro-
ductive waters, because although fi sh debris is
a form of reactive P, its increased rate of burial
under highly productive surface waters associ-
ated with anoxia outpaces its rate of dissolution
within anoxic sediment pore waters. Moreover,
they suggested that, for time intervals before the
diversifi cation of bony fi sh, the absence of abun-
dant fi sh “would result in higher steady state
concentrations of soluble P in the oceans than
today” as a consequence of the correspondingly
lower rates of export of skeletal P. Although
there is some debris from jawless fi sh as well
as conodonts preserved in Late Ordovician
sediments, it is likely that the mass of material
deposited was much lower than in more recent
times, after the “nekton revolution” of the Devo-
nian (Klug et al., 2010). This lower rate of skel-
etal P deposition means that oceans may have
been more prone to development of anoxia in
the Ordovician as a result of the larger pool of
reactive P in the oceans that was available for
recycling (Slomp and Van Cappellen, 2007).
Nitrogen availability also plays a critical role
in high-productivity systems but in a very differ-
ent manner than P. In this case, the rate-limiting
step involves generation of fi xed, or aerobically
bioavailable N. LaPorte et al. (2009) were the
fi rst to document stratigraphic changes in nitro-
gen isotope values of sedimentary organic mat-
ter in the Late Ordovician. They interpreted low
δ15N values in the late Katian black shales of the
Vinini Creek Formation, Nevada, as indicative
of a strongly expanded oxygen minimum zone
in which effi cient denitrifi cation left upwelling
waters defi cient in fi xed nitrogen. As a result,
algal productivity became strongly dependent
on supplies of fi xed nitrogen from cyanobac-
terial productivity, which is refl ected in pre-
served organic matter with low δ15N values.
These nitrogen isotope data support the gener-
ally held view that the late Katian Vinini shale
was deposited under strongly oxygen-depleted
conditions. By contrast, the stratigraphic tran-
sition into the Hirnantian is marked by a shift
to slightly higher δ15N values, which LaPorte
et al. (2009) attributed to increased ocean venti-
lation during glaciation, and a reduction in the
intensity of denitrifi cation in this setting. This,
they argued, led to higher inventories of fi xed
nitrogen recycled as part of the ocean N cycle,
and as a result, higher δ15N values in primary
producers.
In connection with the role of nitrogen cycling,
Logan et al. (1995) and Butterfi eld (2009,
2011) proposed that ecological drivers may
have played a role in controlling the develop-
ment and maintenance of anoxia in open-ocean
waters. Butterfi eld suggested that evidence from
organic biomarkers and nitrogen isotopes indi-
cates that phytoplankton communities during
some Phanerozoic oceanic anoxia events may
have returned to cyanobacterially dominated
communities similar to those present during the
Proterozoic. In particular, Butterfi eld drew upon
suggestions by Logan et al. (1995) that through
the Proterozoic, after the initial oxygenation of
the atmosphere, primary production in plank-
tonic oceanic ecosystems was dominated by
cyanobacteria. The small body size of these pro-
karyotic phytoplankton (0.2–2 μm), compared
to eukaryotic phytoplankton (2–200 μm; Jiang
et al., 2005), would have resulted in very low
settling rates through the water column. Because
such small cells essentially remain suspended in
the water column, the upper levels of the oceans
may have become turbid and oxygen depleted
as a result. A signifi cant component of organic
matter remineralization would also be moved
from the sediment-water interface to higher
levels in the water column, resulting in nutrient
retention within the surface mixed layer.
In this paper, we provide new δ15N data from
additional localities as a test of these hypotheses
related to nitrogen cycling and the changing role
of cyanobacteria in the primary producer com-
munities of the Late Ordovician–early Silurian
(see Results and Discussion sections).
Atmospheric O2 Concentration
Berry (2010), following Berry and Wilde
(1978), noted that in several marine environ-
ments that are prone to high oxygen demand,
development of anoxia and black shale deposi-
tion were more likely to occur in the Ordovi-
cian than in the Holocene. He attributed this
contrast to the lower atmospheric oxygen con-
centrations in the early Paleozoic, which pro-
vided lower rates of supply for the O2 needed
Melchin et al.
1646 Geological Society of America Bulletin, November/December 2013
to balance the demands of aerobic respiration
during organic matter (OM) decay and, there-
fore, permitted lower rates of productivity to
push a given hydrographic setting into anoxia.
Although there is a very wide range of model
estimates of atmospheric O2 for the Late Ordo-
vician (e.g., Bergman et al., 2004; Berner, 2006,
2009; Algeo and Ingall, 2007), all of those esti-
mates indicate lower than present atmospheric
levels in the Late Ordovician. We also note that
Zhou et al. (2012) suggested that the O2 concen-
tration of ocean waters was likely lower in the
Late Ordovician–early Silurian than in modern
times based on analysis of molybdenum isotope
data from South China.
Does the Distribution of Anoxia Change
through the Late Ordovician and
Early Silurian?
The positive reinforcing effects of greater P
recycling in epeiric seas above the fl ooded con-
tinents, combined with lower rates of P burial
in fi sh skeletal material, suggest that the style
of nutrient cycling in the Late Ordovician and
early Silurian oceans differed signifi cantly
from those present today. In particular, these
considerations suggest that relative to the pO2
available to meet respiratory demand, the Late
Ordovician to early Silurian may have been a
time of more sustained and higher productivity
than is commonly present in comparable hydro-
graphic settings in the recent past. These com-
bined factors suggest that regions of the oceans
in the late Cenozoic world that merely became
dysoxic under conditions of comparable (or per-
haps even higher) absolute rates of P input and
re cycling (e.g., Palastanga et al., 2011) would
have been more likely to become fully anoxic in
the Ordovician and early Silurian.
Although no systematic documentation of
the distribution of black shales has previously
been undertaken for either the late Katian or
Hirnantian (see Results and Discussion), it has
been widely recognized that black shales are
much less common and widespread in the early-
mid-Hirnantian than they are in the late Katian
(e.g., Melchin and Mitchell, 1991; Armstrong
and Coe, 1997; Finney et al., 2007; Page et al.,
2007). Furthermore, very few of the black shale
occurrences that do occur in the early-mid-
Hirnantian have been studied in detail in terms
of their mode of formation. One exception to
this generalization is the Hirnantian strata of the
Yangtze Platform region in South China.
The Yangtze Platform successions have been
the subject of considerable stratigraphic, paleo-
geographic, and geochemical study. This work
has shown that the restricted anoxic basin con-
ditions of the late Katian persisted in deeper
portions of the Yangtze Platform intermittently
through the early-mid-Hirnantian, while shal-
lower-water regions changed from anoxic to
oxic (Chen et al., 2004; Zhang et al., 2009; Yan
et al., 2012; Gorjan et al., 2012). Zhang et al.
(2009), however, suggested that the level of
anoxia in deep waters increased in the Hirnan-
tian glacial interval, based on their interpre-
tation of sulfur isotope values in pyrite (see
also Hammarlund et al., 2012). That interpreta-
tion of the S isotopes, however, is at odds with
Yan et al. (2009) and Zhang et al. (2011), who
documented a similar pattern of increasing δ34S
values in pyrite, which they interpreted to rep-
resent a diagenetic signal, rather than variation
in the size of the deep-ocean sulfur reservoir. In
addition, Yan et al. (2012) showed that the level
of anoxia decreased in both the deeper and shal-
lower portions of the Yangtze Platform in the
Hirnantian. Gorjan et al. (2012) suggested that
the δ34S record in South China cannot easily be
interpreted either in terms of changes in oceanic
anoxia or diagenetic processes. Further consid-
eration of the interpretation of the combined sig-
nals of C and S isotope data through this interval
is presented in the Discussion.
Molybdenum geochemistry is widely
regarded as potentially providing useful insights
into local and global oceanic redox conditions
(e.g., Lyons et al., 2009). Hammarlund et al.
(2012) documented Mo concentrations in two
sections around the margins of the Iapetus
Ocean, and Zhou et al. (2012) provided Mo/U
data from the Yangtze Platform region, South
China. In both cases, variations in these values
were interpreted to refl ect variations in local
bottom-water redox conditions, with indica-
tions of less strongly anoxic conditions during
the Hirnantian glacial interval in compari-
son to strata above and below, consistent with
other geochemical redox indicators from these
sections. Zhou et al. (2012) also examined
variations in Mo isotopic composition of the
Ordovician-Silurian boundary sediments in the
Yangtze region. They found strongly fl uctuating
δ98/95Mo values through the Hirnantian glacial
interval, which they attributed to a combination
of increased basin restriction due to sea-level
fall and crustal uplift in the early part of the
glacial interval (leading to strongly positive val-
ues), followed by more negative (although still
fl uctuating) values, which they suggested were
commonly linked to local expression of a global
oxygenation event.
Several authors have suggested that late
Hirnantian–Rhuddanian black shales are much
more widespread in their distribution than those
of the late Katian or early-mid-Hirnantian (e.g.,
Berry, 1998; Page et al., 2007). We test this
hypothesis via a new global compilation of
black shale occurrences (see Results and Dis-
cussion). As with the late Katian and Hirnan-
tian, no systematic study of the full global dis-
tribution has previously been undertaken for
the earliest Silurian, although Page et al. (2007)
showed that black shales were formed at a wide
range of paleolatitudes and water depths in the
latest Hirnantian and early Silurian.
At a global scale, the spread of anoxia in the
earliest Silurian is commonly attributed to post-
glacial sea-level rise and global warming, and a
slowed or even absent overturning circulation in
the oceans, leading to deep-ocean anoxia in low
paleolatitudes (e.g., Brenchley et al., 1994; Arm-
strong and Coe, 1997; Chen et al., 2004). Page
et al. (2007) suggested that periods of transgres-
sion/highstand and reduced rates of overturning
circulation may have also been associated with
increased rates of freshwater input from high lat-
itudes due to glacial melting, promoting oceanic
stratifi cation as well as increased rates of nutrient
input to surface waters in high latitudes.
Studies of the late Hirnantian–Rhuddanian
black shales of the Yangtze Platform support the
suggestion that the large, restricted deep basinal
setting of the late Katian was reestablished
by the earliest Silurian (e.g., Yan et al., 2012).
Paleogeographic studies show that the timing of
onset of black shale deposition in this basin was
diachronous throughout the South China region
(Su et al., 2009; Fan et al., 2011), due, at least in
part, to tectonic controls on basin development,
and partly to gradual transgression and facies
migration.
The most intensively studied black shales of
the late Hirnantian–Rhuddanian are those of the
North African–Arabian regions of Gondwana.
These black shales, often referred to as “hot
shales” because of their high uranium content
and distinctive signature on gamma logs, form
one of the world’s most important petroleum
source rocks (e.g., Klemme and Ulmishek,
1991). Along most of this high-latitude margin
of Gondwana, black shale deposition began
either in the latest Hirnantian (Underwood et al.,
1998; Loydell, 2007; Armstrong et al., 2009b)
or earliest Rhuddanian (Lüning et al., 2005),
and in some regions continued through all or
much of the rest of the Rhuddanian, sometimes
expanding in geographic extent through this
time interval (Le Heron et al., 2009; Moreau,
2011), likely spanning several short-term cycles
of sea-level change. In other regions, however,
organic-rich shales are restricted to the latest
Hirnantian–early Rhuddanian (e.g., Armstrong
et al., 2009b; Le Heron et al., 2009). Based
on these studies, several aspects of the depo-
sitional setting of these black shales are clear.
First, most recent papers have suggested that
they are mainly developed as highstand deposits
(e.g., Armstrong et al., 2009a; Le Heron et al.,
Environmental changes in the Late Ordovician–early Silurian: Review and new insights from black shales and nitrogen isotopes
Geological Society of America Bulletin, November/December 2013 1647
2009; although for an alternative interpretation,
see Lüning et al., 2000, 2005). Sediment starva-
tion likely contributed to their high total organic
carbon (TOC) values. Second, their distribution
was initially localized and strongly controlled
by preexisting topography of the Gondwanan
shelf margin. This topography was formed by
a combination of glacial erosive processes ,
late-glacial and postglacial deposition, ice-front
migration patterns, preexisting structural con-
trols, and glacio-isostatic rebound (Le Heron
et al., 2009; Moreau, 2011). Third, these stud-
ies also suggest that complete fl ooding of the
Gondwanan shelf did not occur until late Rhud-
danian or early Aeronian time. Thus, the indi-
vidual subbasins and paleovalleys remained
relatively isolated from one another and possi-
bly also from the open ocean by silled margins,
which was an important contributing factor in
the generation of anoxia in these basins. A case
in point, Loydell et al. (2009) suggested that
a mid-Rhuddanian hot shale formed during a
regressive interval, thus highlighting circulation
barriers as factors infl uencing the interpretation
of records of anoxia in this region.
Another controversy surrounding the origin
of these extremely organic-rich black shales is
the relative importance of the role of upwelling
(Lüning et al., 2000, 2006) versus freshwater-
induced stratifi cation (fi ord-like model; Arm-
strong et al., 2005, 2006, 2009b) as the primary
mechanisms for inducing enhanced primary
productivity and anoxia. It is noteworthy that
Armstrong et al. (2009b) provided organic geo-
chemical evidence for photic-zone euxinia in
this restricted setting during the deposition of
latest Hirnantian–early Rhuddanian hot shales
in Jordan.
Armstrong (2007), Page et al. (2007), Arm-
strong et al. (2009b), and Le Heron et al. (2009)
highlighted the signifi cant role that orbitally
induced cyclicity likely played in the timing and
extent of development of organic-rich sediments
in these successions. Le Heron et al. (2013) also
provided evidence that residual ice sheets per-
sisted after the main phase of early-mid-Hirnan-
tian glaciation, into the early Silurian, and that
these more localized ice sheets played an impor-
tant role in controlling the distribution of “hot
shales” in that particular region.
SOME REMAINING QUESTIONS
The past 25 yr has been a period of remark-
able growth in our knowledge of the environ-
mental and biotic changes that took place in
Late Ordovician to earliest Silurian time. New
insights are leading to a better understanding of
the long-term linkages between the processes
that control tectonics, climate, ocean circula-
tion, and changing biodiversity. Despite the
numerous advances that have been made in
recent years, however, several major questions
about the character of the late Katian to early
Rhuddanian world and causes of the Hirnantian
mass extinction remain to be answered. Among
these questions are the following:
(1) Was the Hirnantian a period of enhanced
carbon burial and intensifi ed deep-ocean anoxia,
or was it an interval of deep-ocean ventilation?
(2) Were there regional or global changes in
the community of primary producers through
the late Katian-Hirnantian, as suggested by
LaPorte et al. (2009), and did further changes
occur in the late Hirnantian–early Rhuddanian?
(3) How are the changes in oceanic anoxia
and primary producer communities related to
changes in the zooplankton assemblages, as
represented in the record of graptolites?
(4) Was there an oceanic anoxic event in the
late Hirnantian-Rhuddanian, as suggested by
Page et al. (2007)? If so, why did this event
occur in this particular time interval, in contrast,
say, to late Katian time?
In the following sections, we present new
data on the spatial and temporal distribution of
black shales in three time slices: late Katian,
Hirnantian, and early Rhuddanian, together with
new data on nitrogen isotopes in sedimentary
organic matter. We employ these data to test
several alternative hypotheses connected to the
questions listed here, with the goal of advancing
the current state-of-the-art in our understanding
of the role that climate-driven changes in ocean
anoxia may have played as a causative agent of
extinction.
METHODS
Black Shale Distribution
There are no previous studies that document,
at a high temporal resolution, the global distribu-
tion of black shales for the time intervals imme-
diately preceding, during, and following the
Hirnantian glaciation. Such an analysis is essen-
tial, however, for understanding the relationship
between changing temporal patterns of ocean
oxygenation and the climate and bio diversity
changes that took place through the glacial and
postglacial interval (Negri et al., 2009). Unfor-
tunately, there are very few regions for which
geochemical data are available to document
the changing redox conditions through the late
Katian–early Rhuddanian. As result, we have
chosen to map the distribution of sediments that
are commonly taken as indicative of deposition
under anoxic conditions, i.e., black shales and
other black or dark-gray mudrocks that lack evi-
dence of benthos. We emphasize the point that
not all black shales are indicative of bottom-
water anoxia and that not all sediments depos-
ited in anoxic bottom waters are black shales
(e.g., turbidites deposited in anoxic basins).
However, we believe that black shale distribu-
tion provides a fi rst-order proxy for the global
distribution of dysoxic to anoxic conditions at
the depositional site and a basis for the develop-
ment of hypotheses regarding temporal changes
in global dysoxia/anoxia for further testing. It is
clear that future studies will be required to more
rigorously document the paleoredox conditions
of deposition in each of these regions.
Our documentation of black shale distri-
bution is taken mainly from the literature, in
some cases supplemented by our own obser-
vations (see Table 1). Accordingly, the amount
of information that permits us to constrain the
depositional conditions varies substantially
from locality to locality. We use the term “black
shale” in the descriptive sense of Ferretti et al.
(2012), to refer to occurrences of strata reported
in the literature as shales (including calcareous,
dolomitic, or siliceous shales) that are black
or dark gray, although in some successions,
these strata may be interbedded with black,
laminated lime mudstones. This contrasts with
lighter-gray, brown, green, or red mudrocks or
those strata reported to contain fossils of in situ
benthic eukaryotes or evidence of bioturbation
(Boyer et al., 2011). Whereas a few regions are
very well documented in terms of the redox
state of the black shales based on geochemical
proxies (especially South China: Zhang et al.,
2009; Zhou et al., 2011, 2012; Hammarlund
et al., 2012; Yan et al., 2012), or have been
studied in enough paleontological and sedi-
mentological detail to confi dently determine the
black shale biofacies (sensu Arthur and Sage-
man, 1994), many others have been reported as
black shales or black mudrocks, usually (but not
always) with a list of taxa observed in the sedi-
ment. Almost all of the localities are dated by
planktic graptolites. Our observations of many
successions of graptolitic black shales of vari-
ous ages in many parts of the world show that
these strata normally lack any evidence for the
presence of benthic faunas, either in the form of
body fossils or bioturbation. Thus, unless the lit-
erature report noted the presence of bioturbation
or in situ benthic fossils, the presence of which
would indicate dysoxic or oxic conditions in the
bottom waters (e.g., Rhoads and Morse, 1971;
Arthur and Sageman, 1994; Boyer and Droser,
2009), these graptolitic strata were assumed to
lack benthic faunas, and thus, to indicate deposi-
tion under anoxic bottom waters. The combined
biofacies and geochemical study conducted
by Boyer et al. (2011) suggested that detailed
biofacies studies can reliably distinguish black
Melchin et al.
1648 Geological Society of America Bulletin, November/December 2013
shales deposited under anoxic, dysoxic, or oxic
depositional conditions, but that geochemical
proxies are required to distinguish anoxic sub-
surface ocean waters from ferruginous (Fe-rich)
or euxinic (H2S-bearing) waters (Poulton and
Canfi eld, 2011). We note, however, that some
forms of bioturbation or microbenthos can
only be detected by detailed examination (e.g.,
Boyer et al., 2011). As a result, it is certainly
possible that some of the reported graptolitic
black shales from which no benthic faunas have
been reported actually contain subtle, hitherto
un detected evidence of benthos that would indi-
cate dysoxic conditions.
Strata that were described in the literature
as having black shales (usually graptolitic, but
lacking benthos) interbedded with strata with in
situ benthos are described here as intermittent
black shales and suggest the presence of inter-
mittently anoxic subsurface waters. In some
cases, the information provided in the literature
did not permit us to determine whether interbed-
ded benthic fossils were in situ or transported. In
these cases, we took the conservative approach
and assumed that the presence of benthic fos-
sils indicated episodes of oxygenated bottom
waters. In Table 1, which lists our reported
localities, we also report the “data quality” or
level of detail in which each of the black shale
successions has been described.
Nitrogen Isotope Analysis
Selection of study sections for nitrogen iso-
tope analysis was based on a combination of
previously available lithostratigraphic, biostrati-
graphic, and chemostratigraphic data, low level
of thermal alteration of the organic matter, and
the desire to capture a range of depositional and
paleolatitudinal conditions, in comparison with
the available data from Nevada (LaPorte et al.,
2009). The Vinini Creek section documented
by LaPorte et al. (2009, their Fig. 2) represents
a very deep-water succession (base of slope or
ocean fl oor; Finney et al., 2007), situated ~10°
south of the paleoequator on the western mar-
gin of Laurentia. Our new data are from a deep,
distal ramp succession that alternated between
black shale and carbonate deposition in the
Katian on the northern Laurentian margin (~8°
north of the paleoequator) in the central Cana-
dian Arctic Islands (see Melchin and Holmden,
2006a, 2006b), and an open-marine siliciclastic
shelf succession from the Prague Basin in peri-
Gondwanan Europe that shows glacially infl u-
enced sedimentation in the Hirnantian (Štorch,
2006; Mitchell et al., 2011). Conodont alteration
index (CAI) data are available from all of these
regions. The Vinini Creek section has a CAI of
1 (Sweet, 2000), the Canadian Arctic sections
TABLE 1. SOURCES OF DATA FOR PALEOGEOGRAPHIC RECONSTRUCTION OF BLACK SHALE DISTRIBUTION
Locality Region Late Katian Mid-Hirnantian Early
Rhuddanian Depositional setting Data secnerefeR*ytilauq
1 Scotland, S. Uplands Intermittent Intermittent Black shale Base of slope/
accretionary prism Very good Armstrong and Coe (1997); Hammarlund et al. (2012); this study
2a Nevada (shelf) Monitor
Range Intermittent Oxic/dysoxic Black shale or
intermittent Deep shelf basin Very good Finney et al. (1999); Štorch et al. (2011); this study
2b Nevada (deep) Vinini Creek Black shale Intermittent No data Base of slope/
backarc basin Very good Finney et al. (1999, 2000, 2007); Štorch et al. (2011);
3 Idaho Trail Creek Black shale Oxic/dysoxic No data Base of slope Good Goldman et al. (2007); this study
4 SE Alaska Black shale
or Intermittent No data Black shale Base of slope Fair Churkin and Carter (1970)
5 N. Yukon Black shale Intermittent Black shale Deep shelf basin Very good LaPorte et al. (2009); this study
6a Arctic Is. (shallow)
Cornwallis/Truro Is. Intermittent Oxic/dysoxic Black shale Shelf Very good Melchin and Holmden (2006a, 2006b); this study
6b Arctic Is. (deep) Bathurst Is. Black shale Oxic/dysoxic Black shale Deep shelf basin Good Melchin (1989); this study
7 NE Siberia (Kolyma) Black shale
or Intermittent Intermittent Black shale or
intermittent Deep shelf/basin Fair Koren’ et al. (1983); Zhang and Barnes (2007); Koren’ and Sobolevskaya (2008)
)9002(.latesdnallahC;)0891(tteggeLdoogyreVepolsnisaBe
lahskcalBcixosyd/cixOtnettimretnIselaW8
9 N. England Oxic/dysoxic Oxic/dysoxic Black shale Deep slope Good Leggett (1980); Rickards (1988)
10 Poland (Holy Cross) Oxic/dysoxic Oxic/dysoxic Black shale Deep shelf Good Podhalanska and Trela (2007); Masiak et al. (2003); Kremer and Kaźmierczak (2005)
11 Poland (peri-Baltic) Oxic/dysoxic Oxic/dysoxic Black shale Deep shelf basin Good Podhalanska (1999, 2003)
12 Poland (Carpathian) Intermittent Oxic/dysoxic Black shale ? Fair Tomczyk (1963)
13 S. Sweden Oxic/dysoxic Oxic/dysoxic Black shale Deep shelf Good Bergström et al. (1999); Koren’ et al. (2003a)
14 Bornholm, Sweden Oxic/dysoxic Oxic/dysoxic Intermittent Deep shelf Very good Pederson (1989); Hammarlund et al. (2012)
15 Oslo, Norway Oxic/dysoxic Oxic/dysoxic Black shale or
intermittent Deep shelf Good Cocks (1988)
16 Novaya Zemlya Oxic/dysoxic Oxic/dysoxic Intermittent Slope? Good Gogin et al. (1997); Koren’ and Sobolevskaya (1999); Baarli et al. (2003)
17 North Urals Oxic/dysoxic No data Black shale or
intermittent Deep shelf/slope Poor Baarli et al. (2003); Puchkov (2009, 2010)
18 South Urals No data No data Black shale Base slope/
backarc basin Good Koren’ and Rickards (1996, 2004); Puchkov (2009, 2010)
(
a
y
a
k
sve
l
ob
o
S;)3
00
2(.l
at
e
vokase
T
;
)
9
99
1
(
a
y
a
ksvel
o
bo
S
dna’
n
e
r
oK
d
o
oG
fle
h
sp
e
eDt
nett
i
mr
e
t
nI
c
i
x
o
s
y
d/
c
i
xO
t
nett
i
mret
nI
r
y
m
i
aT
9
12011)
20 Norilsk area No data No data Black shale or
intermittent Shelf Fair Tesakov et al. (2003); Kanygin et al. (2010)
21 Gorny Altai Intermittent Oxic/dysoxic Black shale Slope/forearc basin Good Sennikov et al. (2008); Glorie et al. (2011)
22 Kazakhstan Black shale Intermittent Black shale Slope/backarc basin Good Apollonov et al. (1988); Popov et al. (2009)
(continued)
Environmental changes in the Late Ordovician–early Silurian: Review and new insights from black shales and nitrogen isotopes
Geological Society of America Bulletin, November/December 2013 1649
have a CAI of 2 (Melchin et al., 1991; Dewing
and Obermajer, 2009), and the Levín section
in the Prague Basin has a CAI of 3 (Ferretti,
1998). Data from the Triassic of western Can-
ada show that strata with a CAI of ~3 (Orchard,
2007) retain a strong primary N isotope signal
(Sephton et al., 2002). CAI values between 1
and 3 are indicative of thermal alteration condi-
tions within the oil window (Bustin et al., 1990).
Samples for N isotope analysis were broken
into gravel-sized pieces to expose fresh surfaces.
Pieces contaminated by modern plant traces on
fracture surfaces were carefully removed before
powdering in a tungsten carbide swing-mill
device. Preparation of the bulk organic matter
used for the measurements of δ13Corg and δ15NTN
involved accurately weighing ~2 g of powdered
rock, followed by digestion in HCl to remove
carbonate minerals. The residue, consisting of
organic matter, silicates, and oxides, was thor-
oughly rinsed, dried, and then repowdered to
achieve sample homogeneity. Total organic car-
bon (TOC) was determined by comparing the
sum of the ion beam voltages for the masses 44,
45, and 46 between the samples and an internal
organic carbon standard with a known concen-
tration of carbon. Total nitrogen contents (TN)
were determined similarly to TOC using masses
28 and 30 and the nitrogen concentration of
the standard. The uncertainty in TOC and TN
is ±5% and ±25% (1σ), respectively, based on
replicate analyses.
The δ13Corg and δ15NTN analyses were per-
formed using a Flash Elemental Analyzer (EA)
coupled to a Thermo Fisher Delta Plus XL
instrument operating in continuous fl ow mode.
The analyses are reported in the standard delta
(δ) notation refl ecting the per mil variation in
13C/12C and 15N/14N ratios relative to Vienna
Peedee belemnite (VPDB) and atmospheric N2,
respectively. The reference materials used for
the δ13Corg and δ15NTN measurements were cali-
brated against international standards: L-SVEC
(δ13C = –46.6‰ VPDB) and IAEA-CH6 (δ13C =
–10.45‰ VPDB) for δ13Corg; USGS–25 (δ15N =
–30.4‰ atm. N2) and IAEA-305A (δ15N =
39.8‰ atm. N2) for δ15NTN. The 1σ external
precision for δ13Corg and δ15NTN is ±0.1‰ and
±0.4‰, respectively, based on repeated analyses
of caffeine as an internal standard. Molar C/N
ratios were calculated from TOC and TN con-
centrations.
RESULTS
Black Shale Distribution
During the lower-middle P. pacifi cus Zone,
late Katian time slice (Fig. 4A; Table 2), black
shales formed mainly in tropical-subtropical
TABLE 1. SOURCES OF DATA FOR PALEOGEOGRAPHIC RECONSTRUCTION OF BLACK SHALE DISTRIBUTION (continued)
Locality Region Late Katian Mid-Hirnantian Early
Rhuddanian Depositional setting Data secnerefeR*ytilauq
23 Uzbekistan No data No data Intermittent Base of slope/
forearc basin Good Koren’ and Melchin (2000); this study
24a S. China (shallow) Black shale Oxic/dysoxic/
suboxic/dysoxic Black shale Shelf Very good Chen et al. (2000); Zhan and Jin (2007); Yan et al. (2009, 2012)
24b S. China (deep) Black shale Intermittent Black shale Deep shelf basin Very good Chen et al. (2000); Zhan and Jin (2007); Yan et al. (2009, 2012)
25 S. China (slope) Oxic/dysoxic Oxic/dysoxic Intermittent Deep slope Good Zhang et al. (2007)
roelahskcalB
c
ixo
s
yd
/
c
i
xOci
x
osyd/c
i
xOtebiT62 intermittent Shelf Fair Mu and Ni (1983)
27 Prague Basin Oxic/dysoxic Oxic/dysoxic Black shale Shelf/deep shelf basin Very good Brenchley and Štorch (1989); Štorch and Mergl (1989); Štorch (1990, 2006); this study
28 Thuringia Oxic/dysoxic Oxic/dysoxic Black shale Deep shelf basin Good Stein (1965); Schauer (1971); Jaeger (1977)
29 Carnic Alps Oxic/dysoxic Oxic/dysoxic Black shale Shelf/deep shelf basin Very good Jaeger et al. (1975); Schönlaub et al. (2011); Štorch and Schönlaub (2012)
30 S. France Oxic/dysoxic No record Black shale Deep shelf basin Good Štorch and Feist (2008)
31 N. France/Brittany No data Oxic/dysoxic Black shale Shelf/deep shelf basin Good Piçarra et al. (2002, 2009)
32 SW Spain/Portugal Oxic/dysoxic Oxic/dysoxic Black shale Deep shelf Good Jaeger and Robardet (1979); Piçarra et al. (1995)
33 Central Spain Oxic/dysoxic Oxic/dysoxic Intermittent Shelf Good Gutiérrez-Marco and Štorch (1998); Štorch et al. (1998)
34 NW Spain No data Oxic/dysoxic Black shale Shelf Good Gutiérrez-Marco and Robardet (1991); Gutiérrez-Marco et al. (2010)
35 Pyrenees No data Oxic/dysoxic Black shale Deep shelf/deep shelf
basin Fair this study; J. Roqué (2011) personal commun.)
36 Sardinia/Corsica Oxic/dysoxic Oxic/dysoxic Black shale Shelf/deep shelf basin Good Barca and Jaeger (1992); Barca et al. (1996); Štorch and Serpagli (1993); Leone
et al. (2009)
37 Serbia/Bulgaria Oxic/dysoxic Oxic/dysoxic Black shale ?Deep shelf basin Fair Mihajlovič (1974); Sačanski (1993, 1994); Sačanski and Tenov (1993); Lakova and č
Sačanski (2004)
38 Argentine Precordillera No data Oxic/dysoxic Black shale Deep shelf? Good Cuerda et al. (1988); Mitchell et al. (1998)
39 Saudi Arabia No data Oxic/dysoxic Black shale Shelf Good Vaslet (1990); Lüning et al. (2000); Miller and Melvin (2005); Zalasiewicz et al. (2007
)7002(lledyoL;)5002(.lategninüLdoogyreVfle
h
Sela
h
sk
c
alBcixosyd/
c
ixOatado
Nna
droJ04
)0002,3991,88
91(dnargeLdooGflehSela
h
sk
c
alBcixosyd/cixOatadoNa
i
regl
A
14
42 Morocco Oxic/dysoxic Oxic/dysoxic Black shale Shelf Good Destombes and Willefert (1988); Loi et al. (2010); this study
)
2
1
02
(
ll
e
d
y
o
L;)
300
2
(.
l
at
e
g
n
inü
L
d
o
ogy
r
eV
fl
ehS
el
ahs
k
c
a
l
Bc
i
x
o
syd/cixOc
i
x
o
syd/
c
ix
O
ayb
iL
3
4
44 Mauritania No data Oxic/dysoxic Black shale Shelf Good Paris et al. (1998); Underwood et al. (1998); Ghienne (2003)
45 Victoria Australia Black shale
or Intermittent Intermittent Black shale or
intermittent Base slope/
backarc basin Fair Vandenberg et al. (1984); Fergusson (2003)
*Data quality: very good—detailed lithological, faunal, and geochemical data; good—detailed lithological and faunal data; fair—general lithological and limited faunal data; poor—very general lithological data only, with
limited age control.
Melchin et al.
1650 Geological Society of America Bulletin, November/December 2013
paleolatitudes, where they were associated with
two physiographic settings: (1) the western
margins of paleocontinents (e.g., the western
margin of Laurentia and Siberia) and regions
that may have been sites of divergent ocean cur-
rents (Herrmann et al., 2004), and (2) within
restricted epicratonic seas, such as the Yangtze
Platform (Chen et al., 2004; Yan et al., 2012)
(Tables 1 and 2). The black shales of the Mel-
bourne Trough in SW Australia, as well as those
preserved in Kazakhstan, may refl ect deposition
in a semi-enclosed, backarc basin (Fergusson,
2003; Koren’ et al., 2003b; Popov et al., 2009).
As far as we are aware, there are no regions
anywhere in the world that show a record
of continuous deposition of black shales or
demonstrably anoxic sediments through the
mid-Hirnantian (middle M. extraordinarius
Zone to lower M. persculptus Zone). Although
a number of regions show black shales in the
basal strata of the M. extraordinarius Zone,
black shales only occur interbedded with strata
containing in situ shelly fossils in a few regions
in the mid-Hirnantian, and all of those regions
were semi-enclosed basins or regions that may
have experienced divergent ocean current pat-
terns (Herrmann et al., 2004). The majority of
localities that were sites of black shale deposi-
tion during the late Katian, such as in the now
well-studied sections in SE China (Gorjan
et al., 2012; Yan et al., 2012; Zhou et al., 2012),
switched to oxygenated or possibly dysoxic
sediments by mid-Hirnantian time (Fig. 4B;
Tables 1 and 2). Many formerly shallow shelf
sites, however, experienced nondeposition, ero-
sion, or both during most or all of the Hirnantian
glacial interval (e.g., Finney et al., 1997; Berg-
ström et al., 2006).
Black shales appear to have become much
more widespread in the earliest Rhuddanian
than they were in the late Katian or the mid-
Hirnantian (Fig. 4C). Sediments showing evi-
dence of deposition under anoxic conditions
occur at all paleolatitudes where there is a
well-preserved, well-dated record, and also in
a very wide range of paleogeographic settings
(Table 2). In addition, the available data sug-
gest that in most of these regions, the onset of
anoxic conditions took place in late Hirnantian
time, approximately coincident with the major
phase of glacial retreat (Fig. 3). These data also
suggest that in many parts of the world, anoxia
persisted, either intermittently or continuously,
through much of the Rhuddanian, and even
into mid-Aeronian time. For example, in both
the Yangtze Platform succession and succes-
sions of peri-Gondwanan Europe, from the
onset of black shale deposition during the latter
part of the M. persculptus Zone interval, black
shales gradually increased their lateral extent
within their respective basins, reaching their
peak distribution in the early Aeronian, and, in
some parts of the basins, they persisted continu-
ously for at least 5 m.y. (Štorch, 2006; Fan et al.,
2011). Thus, based on the fi rst-order pattern of
occurrences of black shale, the latest Hirnan-
tian–Rhuddanian interval appears globally to
represent a time of much more widely distrib-
uted dysoxia-anoxia in both deep-shelf (below
the pycnocline) and basinal settings. These data
support the conclusion of Page et al. (2007) that
the latest Hirnantian–Rhuddanian interval may
be classed as an oceanic anoxic event, com-
parable in scope to those described from the
Mesozoic (e.g., Schlanger and Jenkyns, 1976;
Jenkyns, 2010; see Discussion).
Nitrogen Isotopes
The nitrogen isotope records in the studied
sections from Arctic Canada exhibit a pattern of
correspondence with lithological records of sea-
level change and black shale (anoxic sediment)
deposition. Unfortunately, the Katian portion of
the Truro Island section (Fig. 5; Table S1 [see
footnote 1]) was not systematically sampled
with isotopic analysis in mind. Therefore, the
carbonates that show evidence of oxygenation
and/or shallowing are not as well represented
in the sampling as are the laminated carbonates
and dolomitic shales that contain graptolites. On
the other hand, the uppermost P. pacifi cus Zone
strata at this section, as well as the Hirnantian
and lower Rhuddanian, were systematically
sampled in all lithologies, as were the late Katian
through latest Hirnantian strata at Eleanor Lake
(Fig. 6; Table S2 [see footnote 1]) and the late
Hirnantian and Rhuddanian strata at Cape Phil-
lips South (Fig. 7; Table S3 [see footnote 1]).
In the Katian portion of the study sections,
δ15N varies between approximately –0.5‰ and
1.5‰–2.0‰, with generally more negative val-
ues in laminated, graptolitic black shales and
more positive values in bioturbated carbonates.
All three sections show a negative shift in δ15N
values in the highest interval of the graptolitic
black shales, just below the onset of positive
δ13C excursions that indicate the beginning
of the Hirnantian glacial episode. This is very
similar to the pattern seen in the Katian strata at
Vinini Creek (LaPorte et al., 2009).
The stratigraphic record through the con-
tact between the Katian and the Hirnantian
differs among the three Arctic study sections.
The succession is most complete at the deeper-
water Truro Island section, where there is an
abrupt change, within only a few millimeters
of sediment, from a graptolite fauna consist-
ing entirely of species of Diplograptina, to
one comprising only Neograptina. Among the
normalograptid species are Normalograptus
mirnyensis, a species previously reported only
from early Hirnantian and younger strata, and
Metabolograptus ojsuensis, a species known
to span the Katian-Hirnantian boundary. The
occurrence of the former species, together
with the sudden, total change in taxonomic
composition of the fauna (e.g., Mitchell et al.,
2007), indicates that this level occurs within the
lower part of the M. extraordinarius Zone, even
though the index species has not been found.
This graptolite faunal turnover occurs 35 cm
above the base of an ~3 m interval of strata that
records the Hirnantian carbon isotope excursion
in both the sedimentary inorganic and organic
carbon fractions of the sediment (Melchin and
Holmden, 2006a). Given the magnitude of the
onset of the δ13C excursion and the relatively
low diversity of the graptolite faunas, we use
the level of the onset of the δ13C excursion to
mark the base of the Hirnantian in this section
(Fig. 5). The return to baseline δ13C values at
the end of the Hirnantian glaciation is marked
by a coincident return to black shales with
graptolites, this time indicative of the M. per-
sculptus Zone (Melchin et al., 1991).
At the shallower-water Eleanor Lake sec-
tion, there is no direct evidence of graptolites
of the M. extraordinarius Zone, and the inter-
val of lithologic change from graptolitic shales
to shallower-water siltstones is marked by a
strongly bioturbated horizon, suggesting that
there may be a brief hiatus at that level. The
overlying ~2 m of calcareous siltstones and silty
limestones show the Hirnantian carbon isotope
excursion, overlain by black, calcareous shales
with M. persculptus Zone graptolites (Melchin
and Holmden , 2006a).
Figure 4. Paleogeographic maps showing distribution of localities at which black shales
have been documented to occur in at least one of the three studied time slices: (A) Late
Katian (mid-pacifi cus Zone); (B) mid-Hirnantian (upper extraordinarius–lower persculptus
Zone); (C) lower Rhuddanian (middle ascensus–acuminatus Zone). Base maps are courtesy
of R.C. Blakey (http://www.cpgeosystems.com/globaltext2.html). A and B are 450 Ma paleo-
geographic reconstructions; C is a 440 Ma reconstruction. For numbered locality names
and data sources, see Table 1. Paleogeographic locations of localities are taken from Gold-
man et al. (2011) and Rasmussen and Harper (2011a).
Environmental changes in the Late Ordovician–early Silurian: Review and new insights from black shales and nitrogen isotopes
Geological Society of America Bulletin, November/December 2013 1651
Early Rhuddanian
X
X
Mid-Hirnantian
X
X
X
X
X
oxic-dysoxic
no data
interbedded black shale
black shale
black shale possibly interbedded
Late Katian
X
X
X
X
X
X
X
X
X
X
X
11
11
14
14
11
14
1
2a2b
3
4
5
6a
6b
7
89
10
12
13
15
17
16
19 20
22
27
37
29
30,36
31 34
28
35
33
42
41
43
40
45
39
32
23
24a
24b 25
26
21
18
38
44
1
2a2b
3
4
5
6a
6b
7
89
10
12
13
15
17
16
19 20
22
27
37
29
30,36
31 34 28
35
33
42
41
43
40
45
39
32
23
24a
24b 25
26
21
18
38
44
1
2a
2b
3
4
5
6a
6b
7
89
10
12
13
15
17
16
19
20
22
27 37
29
30,36
31 34 28
35
33
42
41
43
40
45
39
32
23
24a
24b
25
26
21
18
38
44
A
B
C
Melchin et al.
1652 Geological Society of America Bulletin, November/December 2013
At the shallow-water Cape Phillips South
section, there is an erosional surface with tens
of centimeters of relief and evidence of shallow
karstic dissolution and infi lling in the carbon-
ate immediately underlying the lowest black,
shaley strata containing M. persculptus Zone
graptolites, indicating the presence of a signifi -
cant hiatus and an episode of subaerial erosion.
Here, we present the carbon isotope and percent
total organic carbon data from the late Hirnan-
tian–Rhuddanian interval at this section (from
Melchin and Holmden, 2006b), along with new
nitrogen isotope data (Fig. 7).
The stratigraphic interval between the base of
the Hirnantian and the top of the positive δ13C
excursion at Truro Island and Eleanor Lake
records a positive δ15N excursion. The mean
δ15N value from samples from the P. pacifi cus
Zone from Eleanor Lake and Truro Island is
+1.0‰. If this is taken as a pre-Hirnantian base-
line value, then the magnitude of the peak δ15N
excursion is +2.8‰–2.9‰, because peak values
are +3.8‰ and 3.9‰ at Truro Island and Elea-
nor Lake, respectively.
It is particularly noteworthy that at Truro
Island, the section where the base of the Hirnan-
tian is well constrained by graptolite data, the
biostratigraphic level of the lower Hirnantian
change in graptolite faunas occurs within the
rising limb of the δ15N excursion. The pattern at
Eleanor Lake is consistent with this, although,
as noted already, it lacks supportive graptolite
data in the basal Hirnantian strata.
The δ15N profi les in the Arctic study sections
extend upwards into the late Hirnantian and
early Rhuddanian, ages that are not preserved in
the sedimentary succession at Vinini Creek. As
noted already, the end of the Hirnantian glacial
interval is marked by a sharp decline to substan-
tially more negative values in the δ13C profi le
and the fi rst occurrence of dark, laminated, cal-
careous shales with late Hirnantian graptolites.
All three of the Canadian Arctic sections record
this transition, and two of the sections extend
into the lower Rhuddanian. The values for lat-
est Hirnantian and early Rhuddanian (combined
M. persculptus and Akidograptus ascensus
zones) are 0‰ to +0.5‰ at Truro Island and
Eleanor Lake and –2‰–0‰ at Cape Phillips
South, which are signifi cantly lower than the late
Katian baseline values for these sections. Taken
as a whole, sedimentary δ15N values decline by
4‰ compared to the peak values observed dur-
ing the Hirnantian glaciation.
The δ15N record from the early-mid-Rhudda-
nian at Cape Phillips South, our most complete
section through that interval, shows a similar
pattern as that seen in the late Katian: positive
shifts in δ15N values in light-colored, carbonate
strata that show evidence for deposition beneath
oxygenated bottom waters and/or shallowing,
and negative shifts in δ15N values in black shale
sediments showing evidence for deposition
beneath more oxygen-depleted bottom waters.
In addition to these new data from Arctic
Canada, we also report new nitrogen isotope
data from across the Katian-Hirnantian transi-
tion at two sections in the Prague Basin (Fig. 8;
Table S4 [see footnote 1]). The δ15N values
are high through the entire Katian–Hirnantian
interval, varying between +2‰ and +4‰. In
the Canadian Arctic sections, such high values
are only observed during the interval of the
Hirnantian glaciation. Absent from the Prague
section are the low δ15N values associated with
black shales in both the Vinini Creek and Arctic
Canada sections.
DISCUSSION
N Isotopes
Studies of late Cenozoic upwelling and
Mediter ranean sapropel successions suggest
that although there is some signifi cant spatial
variability in δ15N values in organic matter due
to differences in degrees of oxidation of organic
matter prior to burial, long-term variations within
sedimentary successions mainly refl ect varia-
tions in the primary δ15N signal (e.g., Higgins
TABLE 2. NUMBER OF LOCALITIES WITH OXIC, INTERMITTENT, AND ANOXIC STRATA FOR EACH DEPOSITIONAL SETTING AND TIME SLICE
Oxic/dysoxic Intermittently anoxic* Anoxic or intermittent†Anoxic§No data/no record Total
Late Katian
3161–15flehS 0113–15nisabflehspeeD 6–––15flehspeeD 4–––13epolspeeD 3–21––epolsfoesaB 2–1–1–gnittescra/epolS 42–11–gnittescra/epolsesaB 62–112niatrecnU 841173702latoT
Mid-Hirnantian
211–––11flehS 21129nisabflehspeeD 6––––6flehspeeD 41–––3epolspeeD 52––21epolsfoesaB 2––11gnittescra/epolS 31––11gnittescra/epolsesaB 4–––13niatrecnU 84600753latoT
Early Rhuddanian
31–0121–flehS 11–011––nisabflehspeeD 6–312–flehspeeD 4–211–epolspeeD 321–––epolsfoesaB 2–2–––gnittescra/e
polS 4–2–2–gnittescra/epolsesaB 5–3–2–niatrecnU 84233580latoT
*Intermittently anoxic—interbedded black shales and oxic strata.
†Anoxic or int.—black shales possibly interbedded with oxic strata.
§Anoxic—continuous black shales through time slice.
Environmental changes in the Late Ordovician–early Silurian: Review and new insights from black shales and nitrogen isotopes
Geological Society of America Bulletin, November/December 2013 1653
pacificus pers.
50
45
asc.
TRURO
ISLAND
ac.
40
35
fastigatus extr.
metres
30
-30 -28 -26 13
010
-2 2
0
%TOC
δ13Ccarb
(‰ - VPDB)
δ13Corg (‰ - VPDB)
δ15NTN
(‰ - air)
20
- Silty
Limestone
-
-
Calcareous
Siltstone
Black Laminated
Limestone Black Shale
- Black Dolomitic
Shale
-
-
Black Calcareous
Shale
- Dolostone
- Limestone
- Concretion - Graptolites
- Laminated
Dolostone
- Dolomitic
Siltstone
Katian Hirnantian Rhud.
*
Figure 5. Lithostratigraphy, biostratigraphy, carbon isotope chemostratigraphy (from whole-rock carbonate and organic matter), total or-
ganic carbon (TOC), and nitrogen isotope chemostratigraphy of the section at Truro Island, Arctic Canada. Nitrogen isotope data are new;
otherwise, data are from Melchin and Holmden (2006a, their Fig. 2). Truro Island section is located at 75.306°N, 97.130°W. See Melchin
and Holmden (2006a, their Fig. 1) for location map. Abbreviations are as follows: Rhud.—Rhuddanian; extr.—extraordinarius; pers.—per-
sculptus; asc.—ascensus; ac.—acuminatus; VPDB—Vienna Peedee belemnite.
Melchin et al.
1654 Geological Society of America Bulletin, November/December 2013
et al., 2010; Möbius et al., 2011; Robinson
et al., 2012). Where there is isotopic change due
to organic matter oxidation, it normally results
in enrichment of δ15N values by ~2‰ or more
(e.g., Möbius et al., 2011). Since the δ15N values
recorded in our study sections, as well as Vinini
Creek, are very low in comparison with most
records from the modern ocean, and are consis-
tent with those recorded in well-preserved sap-
ropels (e.g., Higgins et al., 2010) and Mesozoic
black shale successions (e.g., Ganeshram et al.,
2000; Kuypers et al., 2004; Junium and Arthur,
2007), it is unlikely that the values in our study
have been extensively altered prior to burial.
Most recent studies of the effects of thermal
alteration and low-grade metamorphism sug-
gest that there is little or no signifi cant alteration
of the sedimentary δ15N signal up to low-grade
metamorphism (Ader et al., 2006; Boudou
et al., 2008), although one study (Jia, 2006)
did indicate evidence of increasing δ15N values
with increasing levels of metamorphism. As
noted earlier herein, none of our study sections
showed evidence of elevated δ15N values that
may be associated with high levels of thermal
alteration.
The development of anoxia in western and
northern Laurentia during the late Katian and
latest Hirnantian–early Rhuddanian may have
been the result of upwelling, possibly combined
with partial basin restriction, particularly in the
Arctic Canadian localities (Melchin and Holm-
den, 2006a; Finney et al., 2007). Our nitrogen
isotope data indicate, however, that the changes
in communities of primary producers (LaPorte
et al., 2009) were not analogous to those of
modern glacial-interglacial cycles in upwelling
systems. In Quaternary upwelling systems, sedi-
ments in interglacial intervals, particularly early
interglacials, exhibit relatively high δ15N values,
which are generally in the range of 5‰–10‰.
This is the result of the N isotope fractionation
that takes place below the pycnocline as a result
of denitrifi cation (e.g., Ganeshram et al., 2000).
In contrast, sediments in glacial intervals are
commonly at least 2‰–3‰ lower (Ganeshram
et al., 2000; Möbius et al., 2011). This is inter-
preted as a refl ection of variations in the rates
of denitrifi cation in oxygen minimum zones and
surface sediments, with higher rates of denitri-
fi cation in interglacial intervals (Altabet et al.,
1995; Ganeshram et al., 2000). Similar patterns
of δ15N enrichment in early interglacial sedi-
ments (transgressive black shales) and lower
values in later interglacial and glacial intervals
were observed in Pennsylvanian sediments from
the North American interior.
As documented by LaPorte et al. (2009), and
further supported by our new data, the strati-
graphic pattern of variation seen in the δ15N val-
ues of the Late Ordovician preglacial through
glacial to postglacial transitions is opposite to
the pattern in the Cenozoic glacial-interglacial
upwelling successions. In the late Katian and
early Rhuddanian, which are interpreted as
preglacial and postglacial, respectively, the
δ15N values are not as high as they are in the
late Cenozoic preglacial, postglacial, and inter-
glacial episodes, and they are also lower than
those of the peak glacial intervals in both the
Hirnantian and Cenozoic. In addition, the late
Katian and late Hirnantian–early Rhuddanian
values are lower than any seen in late Cenozoic
upwelling systems. On the other hand, the late
Katian and early Rhuddanian black shales of
northern and western Laurentia show a similar
pattern to those seen in the late Cenozoic sap-
ropels of the Mediterranean and also some of
the Mesozoic marine black shales associated
with oceanic anoxic events. In the late Cenozoic
sapropel cycles, δ15N values in the anoxic sedi-
ments have values around 0‰ to –2‰, whereas
those in the interbedded, oxygenated sediments
are 2‰–7‰ (e.g., Meyers and Bernasconi,
2005; Higgins et al., 2010). Although Möbius
et al. (2010) showed that these δ15N values may
be signifi cantly infl uenced by early diagenesis,
Higgins et al. (2010) provided compelling evi-
dence that the oxic-anoxic values showed a
strong primary signal. The low values seen in the
anoxic, organic-rich sapropels are interpreted to
be the result of development of a strong, stable
de nitrifi cation zone within the water column
during anoxic episodes, in which denitrifi cation
was essentially complete, so that there was little
or no fl ux of fi xed nitrogen to the surface waters.
As a result, productivity in the surface waters
relied on nitrogen fi xed by cyanobacteria, which
yields organic matter with δ15N values between
–2‰ and 0‰. During more oxygenated inter-
vals, δ15N values are higher, indicating a signifi -
cant fl ux of recycled, partially denitrifi ed, fi xed
nitrogen to the surface waters.
A similar pattern is seen in the oxic-anoxic
cycles associated with Mesozoic oceanic anoxic
events (e.g., Ganeshram et al., 2000; Kuypers
et al., 2004; Junium and Arthur, 2007), and these
are also interpreted as resulting from a reduction
Figure 6. Lithostratigraphy,
biostratigraphy, carbon iso-
tope chemostratigraphy (from
whole-rock carbonate and or-
ganic matter), total organic
carbon (TOC), and nitrogen
isotope chemostratigraphy of
the section at Eleanor Lake,
Arctic Canada. Nitrogen iso-
tope data are new; otherwise,
data are from Melchin and
Holmden (2006a, their Fig. 2).
Eleanor Lake section is located
at 75.373°N, 94.106°W. See
Melchin and Holmden (2006a,
their Fig. 1) for location map.
See Figure 5 for lithologi-
cal legend. Abbreviations are
as follows: fast.—fastigata;
extr.—extraordinarius; pers.—
persculptus; VPDB—Vienna
Peedee belemnite.
50
55
ELEANOR LAKE
fast. pacificus extr. pers.
metres
-30 -28 -26 -2
-24 0 0510
202
4
%TOC
δ13Ccarb
(‰ - VPDB)
δ13Corg (‰ - VPDB)
δ15NTN
(‰ - air)
Katian Hirnantian
Environmental changes in the Late Ordovician–early Silurian: Review and new insights from black shales and nitrogen isotopes
Geological Society of America Bulletin, November/December 2013 1655
in the upwelling of fi xed N to the surface waters
and the increased role of N-fi xing cyanobacteria
within the plankton communities during anoxic
intervals. In these studies, this interpretation is
supported by organic biomarkers indicating that
an increased biomass of cyanobacteria is pre-
served within the bulk organic matter.
Our results show that for the Nevada and
Canadian Arctic sections, almost all of the
samples that are interpreted to represent deposi-
tion under anoxic conditions show δ15N values
of –1‰ to +1‰. These values are consistent
with organic matter produced by a mixed com-
munity of cyano bacteria and algae in which the
predomi nant source of bioavailable N is sup-
plied by cyano bacterial fi xation. Recent organic
biomarker data from the Vinini Creek section
(hopane-sterane ratios and 3β-methylhopane;
Rohrssen et al., 2012) support this interpretation,
indicating that the waters of the photic zone were
populated by phytoplankton communities with a
high proportion of cyanobacteria during the inter-
vals of black shale deposition. We also observe
fl uctuations to more positive values within the
late Katian and early Rhuddanian successions
in those portions of the sections showing inter-
bedded, organic-poor, bioturbated carbonate
strata. This suggests that the black shale intervals
represent times of incursion of more intensely
denitri fi ed waters into the deep-shelf settings rep-
resented by the Cornwallis Island sections. These
incursions alternated with episodes of more oxy-
genated waters with a less-pronounced infl uence
of cyanobacterial N fi xation.
In the Canadian Arctic sections, the sedi-
ments deposited during the peak glacial interval
of the early-mid-Hirnantian show signifi cantly
higher δ15N values (2‰–4‰) than those of the
late Katian. These values are within the range of
modern ocean waters under well-oxygenated to
suboxic conditions. This accords well with the
observation that the sediments in this interval
show sedimentological evidence of shallowing
and oxygenation (Melchin and Holmden, 2006a).
The magnitudes of the positive δ15N excursions
recorded in the Arctic Hirnantian sections are
comparable to those recorded through oxic-
anoxic cycles associated with Mesozoic oceanic
anoxic events (e.g., Junium and Arthur, 2007;
Paris et al., 2010). The fact that a positive shift in
δ15N values was also observed in the deep-water
succession in Nevada suggests that this shift in
the balance of primary producers in Hirnantian
time may have been a widespread phenomenon,
as hypothesized by LaPorte et al. (2009). The
smaller magnitude of the shift may be an indica-
tion of a less localized N cycle at Vinini Creek,
i.e., one that refl ects a more regional ocean basin
signature, in contrast to the signature of a deep
cratonic setting (e.g., Arctic Canada) where mix-
ing with the global ocean was more restricted.
Alternatively, the smaller shift at Vinini Creek
could be a sign that effi ciency of denitrifi cation
remained strong through the Hirnantian glacial
interval due to patterns of ocean circulation or
physiography of the local basin.
The late Katian strata in the Prague Basin, in
a higher-paleolatitude, clastic, deep-shelf set-
ting, do not show evidence of anoxia, either in
the character of the sediments or in the N iso-
tope record. The δ15N values fl uctuate between
2‰ and 4‰ throughout both the late Katian and
the early Hirnantian, suggesting that the level of
oxygenation and rate of input of fi xed N to the
primary producers remained relatively constant,
although the range of sample-to-sample varia-
tions in values is higher in the early Hirnantian
than in the late Katian. The lack of evidence for
predominance of cyanobacterial fi xation, indi-
cating signifi cant upwelling of reduced N as a
primary nutrient source for phytoplankton in
the Bohemian samples, highlights the regional
nature of anoxia in the world’s oceans at that
time, and the corresponding development of the
30
35
40
45
metres
persculptus ascensus acuminatus atavus acinaces
0 5 10 15 -2 0 2-2-4 0-32 -30 -28
δ15NTN
(‰ - air)
%TOC
δ13Ccarb
(‰ - VPDB)
δ13Corg
(‰ -
VPDB)
CAPE
PHILLIPS
SOUTH
Hirnantian Rhuddanian
Figure 7. Lithostratigraphy,
biostratigraphy, carbon iso-
tope chemostratigraphy (from
whole-rock carbonate and or-
ganic matter), total organic
carbon (TOC), and nitrogen
isotope chemostratigraphy of
the section at Cape Phillips
South, Arctic Canada. Nitrogen
isotope data for the whole sec-
tion are new; otherwise, data
are from Melchin and Holm-
den (2006b, their Fig. 2). Cape
Phillips South section is located
at 75.542°N, 94.531°W. See
Melchin and Holmden (2006b,
their Fig. 1) for location map.
See Figure 5 for lithological
legend. VPDB—Vienna Peedee
belemnite.
Melchin et al.
1656 Geological Society of America Bulletin, November/December 2013
plankton systems dominated by upwelling from
oxygen-depleted waters in the late Katian. This
supports our conclusions based on the compila-
tion of data on black shale distribution for the
late Katian, i.e., that black shale occurrences
were essentially controlled by regional variations
in productivity and ocean circulation, mainly
within the paleotropics. It appears that there
were also strong regional variations in the com-
position of the primary producer communities.
It is also noteworthy that comparisons of
the paired records of δ13C and δ15N of our late
Katian–Rhuddanian successions show that
intervals of positive shift in δ15N also show
positive δ13C excursions. With respect to these
paired records, our successions do not resemble
those of late Cenozoic sapropels, which show
negative δ13C shifts at levels of positive δ15N
excursions (Meyers and Bernasconi, 2005).
From the point of view of the combined C and
N isotope changes, therefore, neither late Ceno-
zoic upwelling systems nor sapropel-producing
settings provide a satisfactory analogue for con-
ditions that gave rise to the Late Ordovician and
early Silurian black shales.
We also note that similar patterns of change in
δ15N values through a succession of interbedded
black shales and oxic limestone and mudstones
were observed in the Katian and upper Rhudda-
nian–Aeronian successions in the subsurface of
Latvia (Kiipli and Kiipli, 2013). Those authors
also observed lower δ15N values in the black
shales (generally –1.1‰ to +0.2‰) than the
lighter-colored strata (generally 1.9‰–5.0‰),
although the lower Katian black shale values
were 0.5‰–2.4‰. Those authors also inter-
preted the lower δ15N values in the black shales
to represent a dominance of cyanobacterial N
fi xation, as a result of deep-water anoxia. They
also suggested that the results indicated anoxia
extending into the photic zone. Evidence of
widespread cyanobacterial fi xation of N within
the photic zone is not, however, indicative of
photic zone anoxia, because cyanobacterial
fi xation of N is a commonly occurring process
in modern, oxygenated surface waters that are
depleted in fi xed N (e.g., Brandes et al., 2007).
In order to demonstrate photic zone anoxia, it
would be necessary to document the presence
of anoxic photosynthetic organisms, such as
green sulfur-reducing bacteria, which can be
inferred by identifi cation of the presence of the
biomarker isorenieratene (e.g., Sinninghe Damsté
and Köster, 1998; Armstrong et al., 2009b).
Hirnantian Glacial Interval: Carbon and
Sulfur Isotopes, Carbon Burial, and the
Record of Anoxia
As discussed in the review section earlier
herein, there is considerable controversy sur-
rounding the interpretation of the Hirnantian
positive carbon isotope excursion (Kump et al.,
1999; Melchin and Holmden, 2006a; Mun-
necke et al., 2010; Holmden et al., 2012). Many
authors have attributed this positive excursion
primarily to increased burial of organic carbon
during the glacial interval (e.g., Brenchley et al.,
2003; Hammarlund et al., 2012; Jones and Fike,
2013). However, Melchin and Holmden (2006a)
and LaPorte et al. (2009) demonstrated signifi -
cant regional differences in shape and magni-
tude of the excursions (Fig. 3), which argue for
strong infl uences of local carbon cycling in the
circulation-restricted settings of shallow epeiric
seas, combined with changes in the isotopic
value of global weathering fl ux to the oceans,
and local variation in the isotopic composition
of allochems (Kump et al., 1999; Melchin and
Holmden, 2006a; LaPorte et al., 2009; Jones
et al., 2011; see also Metzger and Fike, 2013).
LaPorte et al. (2009) provided evidence that the
true shift in the Hirnantian ocean C reservoir
was just 2.7‰ ± 0.4‰, based on study of three
deep-water sections. Our goal in this section is
to critically examine the strength and consis-
tency of the evidence supporting the alterna-
tive, organic carbon burial interpretation, and
the closely allied assumption that the studied
records of C-cycle changes genuinely refl ect
global ocean conditions.
Coincidence of C and S Isotopic Changes
As noted in the review section, Zhang et al.
(2009) and Hammarlund et al. (2012) suggested
that a positive excursion in δ34S values in pyrite
refl ected a decrease in the oceanic SO4 reser-
voir as a result of increased pyrite burial under
conditions of deep-sea anoxia during the early-
mid-Hirnantian. Hammarlund et al., in particu-
lar, suggested that the deep-sea anoxia of the
early-mid-Hirnantian was caused by increased
sequestration of organic matter in the deep
ocean, which also resulted in the positive δ13C
excursion that is globally observed in this strati-
graphic interval. The prediction of this model
is that within each of the studied sections, C,
S-pyrite, and S-sulfate isotopic changes should
be strongly correlated and in phase, sample
by sample. The two sections studied thus far
in China do exhibit the predicted coincident
changes in δ13Corg and δ34SPy (e.g., Zhang et al.,
2009, their Fig. 2; Yan et al., 2009, their Figs. 2
and 3). However, of the three sections studied by
Hammarlund et al. (2012), one (in a section near
1
2
3
4
5
6
7
8
meters
meters
Zadní Třebáň
0
5
10
15
20
25
30 Levín
L
P
U
lower (L) & upper (U) diamictites
Pernik Bed (P)
sandstone
mudstone & shale
δ15NTN
(‰ - air)
δ15NTN
(‰ - air)
δ13Corg
(‰ - VPDB)
δ13Corg
(‰ - VPDB)
23423
4
-30 -28 -28-29 -27-26 -24
laticeps ojsuensis
Katian Hirnantian
Kosov Formation
Králov Dvůr
o
*°
Figure 8. Lithostratigraphy, biostratigraphy, carbon isotope chemostratigraphy (from
whole-rock carbonate and organic matter), total organic carbon (TOC), and nitrogen isotope
chemo stratigraphy of the sections at Levín and Zadní Třebáň, near Prague, Czech Republic.
Nitrogen isotope data are new; otherwise, data and location information are from Mitchell
et al. (2011, their Fig. 2). Asterisk marks the level of appearance of Metabolograptus ojsuensis ;
square indicates lowest occurrence of Hirnantia fauna in the Prague region (Mergl, 2011).
VPDB—Vienna Peedee belemnite.
Environmental changes in the Late Ordovician–early Silurian: Review and new insights from black shales and nitrogen isotopes
Geological Society of America Bulletin, November/December 2013 1657
Plöcken Pass in the Carnic Alps) shows no posi-
tive shift in δ34SPy at the level of the Hirnantian
carbon isotope excursion (see also Shönlaub
et al., 2011), although there are high δ34SPy val-
ues both in the underlying upper Katian and the
overlying upper Hirnantian strata. At Billegrav,
Denmark, the peak in δ34SPy is in strata inter-
preted to represent the lower Hirnantian, but
the peak in δ13Corg values does not occur until
the overlying upper Hirnantian strata. In neither
case do these patterns match the predictions of
a model in which the S isotope excursion is the
result of a decrease in the global SO4 reservoir
due to an increase in global anoxia that resulted
from increased sequestration of organic matter
in the deep ocean.
Zhang et al. (2009) suggested that a test of
their hypothesis that the positive δ34SPy excur-
sion was the result of pyrite burial under con-
ditions of deep-water anoxia was to conduct
paired analyses of δ34S from pyrite and δ34S
from carbonate-associated sulfate (CAS). If
those patterns show parallel trends, then that
would support the hypothesis that the positive
excursions are the result of deep-water anoxia
and high rates of pyrite burial. Jones and Fike
(2013) conducted a study of paired δ34SPy and
δ34SCAS in the Hirnantian strata at Anticosti
Island and found that the δ34SCAS curve did not
show a positive shift at the level of the strong
δ34SPy excursion. They suggested that their data
did not support the presence of a deep, anoxic
ocean during the Hirnantian glacial interval.
Rather, they suggested that the positive δ34SPy
excursion was the result of changes in the pat-
terns and rates of sulfur isotope fractionation
expressed by microbial sulfur cycling. They also
suggested, building on the hypothesis of Stanley
(2010), that cooling of deep-ocean waters dur-
ing the Hirnantian may have resulted in reduced
rates of aerobic decomposition of organic mat-
ter, and increasing burial of organic matter in
oxygenated, deep-ocean sediments.
In addition, we note, as discussed earlier, that
our nitrogen isotope data, and that of LaPorte
et al. (2009), suggest that at least along the west-
ern and northern margins of Laurentia, rates
of denitrifi cation below the chemocline were
somewhat lower during the early-mid-Hirnan-
tian than during the late Katian or early Rhud-
danian. This is not consistent with increased
deep-water anoxia during the Hirnantian glacial
interval.
It is also noteworthy that results of study of
Mesozoic oceanic anoxic events do not show
a consistent relationship between positive δ34S
excursions and deep-ocean anoxia. In fact,
in some sections, the largest of the Mesozoic
oceanic anoxic events, the Cenomanian-Turonian
event (oceanic anoxic event [OAE] 2), coincides
with a negative δ34S shift during the peak of the
anoxic event in the deep ocean (Hetzel et al.,
2009). Therefore, positive δ34S excursions can-
not be taken as signals that consistently imply
increases in global anoxia in the deep ocean.
Shelf-Basin Gradients in C and S
Isotope Records
Melchin and Holmden (2006a) and LaPorte
et al. (2009) showed that there are strong shelf-
basin gradients in C isotope values resulting from
local-scale C-cycling processes. The data pre-
sented by Yan et al. (2009), Zhang et al. (2009),
and Hammarlund et al. (2012) indicate that there
are signifi cant changes in the magnitude of the
positive shift in δ34SPy values between shallower-
and deeper-water sections, similar to those previ-
ously described for C isotopes. Among the two
Iapetus Ocean localities documented by Ham-
marlund et al., the highest δ34SPy values approach
+40‰ in the on-shelf section (Billegrav) but
only ~+5‰ in the oceanic realm site (Dob’s
Linn). Likewise, the Nanbazi and Honghuayuan
sections, which are in what has been interpreted
as a shallower-water part of the Yangtze Platform
Basin (e.g., Zhan and Jin, 2007), show a higher
magnitude of δ34SPy excursion than the deeper-
water Wangjiawan section (Zhang et al., 2009;
Yan et al., 2009). This consistency in the pattern
of declining δ34SPy values from shallower- to
deeper-water sites in two different marine basins
suggests the possibility that the pattern could
have been widespread during the Hirnantian. If
this shore-to-basin δ34SPy gradient does prove to
be a general feature of the S system, then this
pattern of spatial heterogeneity in S cycling
would be inconsistent with the assumption of an
S-cycle model in which the Chinese and Iapetus
sections were deposited in paleobasins that were
fully connected to the global ocean (Hammar-
lund et al., 2012). On the other hand, if it were
common for seawater in the basin-proximal sec-
tions to be restricted from mixing with the ocean,
then the high δ34SPy values recorded in these sec-
tions could be due to the progressive depletion of
the water mass in dissolved SO4 as it circulated
toward the shallowest and innermost regions
of the paleobasin, analogous to the interpreta-
tion offered by Shen et al. (2002) in their study
of the McArthur Basin, and consistent with the
evidence for local C-cycling and Ca-cycling
effects reported during the Hirnantian glaciation
(Melchin and Holmden, 2006a; LaPorte et al.,
2009; Holmden et al., 2012).
Record of Black Shale Preservation in
Deep-Water Sediments
The data presented in this paper on black
shale distributions also clearly show that the
early-mid-Hirnantian glacial interval was a
time in which black shales were very restricted
in their geographic distribution at all oceanic
depths that are preserved in the rock record
(see also Melchin and Mitchell, 1991; Melchin
and Holmden, 2006a; Page et al., 2007) (Fig.
3B). The few basins that do exhibit evidence of
oxygen depletion during this interval appear to
have experienced only intermittent rather than
continuous anoxia. In addition, the late Katian
to mid-Hirnantian successions in South China
(Yan et al., 2012), Scotland (Hammarlund et al.,
2012), and Nevada (LaPorte et al., 2009), which
record the most nearly continuous occurrence
of oxygen-depleted sediments, show that the
organic carbon content of the sediment depos-
ited during the Hirnantian glacial interval was,
on average, signifi cantly lower than that of the
late Katian or the early Rhuddanian.
The strata of the Vinini Creek section in
Nevada lie within the upper plate of the Rob-
erts Mountains allochthon. Regionally, the base
of the Vinini Formation succession includes
abundant greenstones of primarily within-plate
affi nities that were likely extruded as part of a
seamount complex within what may have been
a backarc basin (Emsbo, 1993; Finney et al.,
2000, 2007). The upper member of the Vinini
Formation conformably overlies an ~1.5-km-
thick Middle Ordovician lowstand fan complex
that is coeval with the Laurentian Sauk-Tippi-
canoe supersequence boundary and appears
to have been deposited in a continental rise to
ocean fl oor setting adjacent to the western shelf
margin of Laurentia (Finney et al., 2000; Jones
Crafford, 2008). The strata of the upper member
are primarily shale, laminated carbonate mud-
stone, and fi ne to coarse calcarenite grainstone.
These latter two lithologies appear to represent
distal turbidites. Benthic fauna and burrows are
rare to absent, but graptolites, sponge spicules,
and radiolaria are common (Finney et al., 1999;
Mitchell and Štorch, authors’ personal observa-
tion, 2011). Redox-sensitive N isotopes through
the interval of the Hirnantian glaciation in this
section shift to positive δ15N values at the base
of the M. extraordinarius Zone, precisely coor-
dinating with glacio-eustatically controlled sea-
level fall (LaPorte et al., 2009; Holmden et al.,
2013). LaPorte et al. interpreted this change
in δ15N values to refl ect a shift to less strongly
anoxic conditions below the surface mixed
layer of the ocean. The Vinini succession also
shows lower rather than higher concentrations
of organic matter in the Hirnantian sediments as
compared with those of the upper Katian. Nei-
ther of these features match the predictions of
high rates of organic matter burial and increased
anoxia in the early Hirnantian deep ocean.
The Late Ordovician to early Silurian Moffat
Shale Group succession sampled at Dob’s Linn
Melchin et al.
1658 Geological Society of America Bulletin, November/December 2013
lies within the Southern Uplands terrane, which
is a complexly deformed, southeast-verging,
imbricate thrust system bounded by the Southern
Uplands fault to the NW and the Iapetus suture
on the SE (e.g., Fortey et al., 2000; Phillips et al.,
2003; Needham, 2004). The tectonic origin of
the Southern Uplands terrane is controversial,
but the most strongly supported hypothesis is
that it represents an accretionary prism complex
that formed oceanward of a NW-dipping subduc-
tion zone located along the Laurentian margin of
the Iapetus Ocean (e.g., Needham, 2004; Phillips
et al., 2003; Sawaki et al., 2010; Stone and Merri-
man, 2004; Strachan, 2012). Alternative models
include backarc and marginal-basin settings
(see Armstrong and Owen, 2001; Phillips et al.,
2003). The Hartfell and Birkhill Shales resemble
hemipelagic sediments and appear to have been
deposited distal to the Gala Group fan complex,
which prograded over the Moffat Group strata in
the region of Dob’s Linn during the early Silu-
rian (Webb et al., 1993; Floyd, 2001; Strachan,
2012). These features of the tectonic setting and
facies character suggest that the late Katian to
early Rhuddanian strata at Dob’s Linn record
deposition under oceanic conditions, below the
depth of the pycnocline-chemocline throughout
this time interval.
The early-mid-Hirnantian pattern of dysoxia/
anoxia contrasts even more strongly with the
early Rhuddanian interval than with the late
Katian. The early Rhuddanian shows black
shale deposition at all paleolatitudes and water
depths below the pycnocline where there is a
well-preserved sedimentological record. Thus,
although we do not have quantitative data on the
carbon contents from most of our documented
black shale localities, the available record of
black shale distribution patterns suggests that
global rates of organic carbon burial appear to
have been much greater in the early Rhuddanian
than during the glacial interval.
Some authors have argued that the lowered
sea level during the Hirnantian glacial interval
would have shifted organic matter burial to the
deep sea, where the sedimentary record has since
been largely destroyed by tectonic processes
(e.g., Brenchley et al., 2003; Hammarlund et al.,
2012). However, as noted previously, several of
the sections from which we do have sedimen-
tary records through the Hirnantian appear to
have been deposited in base of slope, backarc
basin, settings, which likely are representative
of deep-ocean conditions (e.g., Nevada, Scottish
Southern Uplands, Gorny Altai, Lachlan fold
belt) (Table 1), and none shows highly organic-
rich black shales through this interval, much
less evidence of increased carbon burial, which
is what would be expected if maximum pro-
ductivity had been shifted to beyond the shelf
edge. Indeed, in our global survey of black shale
occurrences, we did not fi nd any sections that
switched from a more oxic depositional setting
during the late Katian to a less oxic setting in
the Hirnantian. Thus, if increased carbon burial
was an important contributing component to the
Hirnantian positive carbon isotope excursion,
then the reservoir for the sequestered carbon
remains to be identifi ed.
A plausible solution to the Hirnantian TOC
conundrum may be found in the work of Hedges
and Keil (1995) and Tsandev et al. (2010), who
suggested that during times of glacial lowstand,
signifi cant quantities of organic matter derived
from terrestrial and shelf erosion would be
transported directly to the deep sea by turbid-
ity currents and deposited in slope and sub-
marine fan settings. This process might have
resulted in signifi cant sequestration of organic
matter in thick prodelta, slope, and turbidite
successions under oxygenated bottom-water
conditions, particularly if rates of organic mat-
ter decomposition were slowed as a result of
low seawater temperatures during the glacial
interval (Stanley, 2010; Jones and Fike, 2013),
where the increase in organic carbon burial rate
would be masked by an increase in sedimenta-
tion rate, i.e., by a dilution effect. Accordingly,
a change in the depositional mode for organic
matter burial during the Hirnantian may have
contributed to the +2.7‰ increase in the global
δ13C value of the ocean C reservoir as suggested
by LaPorte et al. (2009), together with changes
in fl uxes of weathered carbonate and organic
matter from the continents (Kump et al., 1999;
Melchin and Holmden, 2006a).
Rhuddanian Oceanic Anoxic Event
We have documented the distribution of
black shales specifi cally for the early Rhudda-
nian (Fig. 4C), although their distribution was
similar in the late Hirnantian, and many sedi-
mentary successions show a continuous record
of black shale deposition across the Hirnantian-
Rhuddanian boundary. It is important to note
that although black shales occurred at a wide
range of localities, paleolatitudes, and deposi-
tional settings in the early Rhuddanian, we do
not have data indicating whether or not anoxic,
dysoxic, or euxinic conditions were widespread
between these studied localities. Nevertheless,
our compilation of black shale distributions
clearly suggests that the oceans were more sus-
ceptible to oxygen depletion in deeper waters
at all studied paleolatitudes in the Rhuddanian
than in the earlier Hirnantian or late Katian (Fig.
4C). It is on this basis that we can consider some
possible causes of this particularly widespread
development of oxygen depletion.
The early Rhuddanian represents a period of
apparently rapid global warming, although evi-
dence suggests that there remained some glacial
ice at high paleolatitudes (e.g., Finnegan et al.,
2011; Le Heron et al., 2013). This interval of
relative warming, sea-level rise, and widespread
oxygen depletion coincides with the second
phase of a major mass extinction event. It is
surprising, therefore that this interval has not
received as much research focus as the Katian-
Hirnantian transition and has not been the
specifi c subject of quantitative, global model-
ing studies.
Early Rhuddanian black shales occur not only
at all paleolatitudes and on all paleocontinents
where there are suffi ciently well-preserved and
dated records to permit their identifi cation, but
also in a wide range of depositional environ-
ments. These include distal shelf settings (Arc-
tic Canada), semi-enclosed coastal embayments
(North Africa and Jordan), deep epicontinental
basins (South China, eastern Poland), and basin
slope settings (Wales), as well as apparently
abyssal environments preserved in accretion-
ary prisms (Southern Uplands, Scotland) and
forearc and backarc basins (Kazakhstan, South
Urals) (Table 1). The wide geographic spread,
as well as depth range, suggests that this interval
does, indeed, represent an oceanic anoxic event,
as suggested by Page et al. (2007), hereafter
referred to as the Rhuddanian oceanic anoxic
event. We note, however, that unlike some Meso-
zoic oceanic anoxic events, this interval does not
show a positive C isotope excursion, although
even Mesozoic oceanic anoxic events are not
consistent in this regard (e.g., Jenkyns, 2010).
The evidence suggests that many of the fac-
tors promoting the development of black shales
in the late Katian (discussed earlier) can also
be attributed to the early Rhuddanian. They do
not, however, help explain why dysoxia/anoxia
was more widespread in the early Rhuddanian
than in the late Katian. We next consider several
potential causes of the great extent of Rhudda-
nian black shales.
Sea Level in the Latest Hirnantian–
Early Rhuddanian
The early Rhuddanian is widely recognized
as a time of transgression to highstand condi-
tions due to post-peak-glacial sea-level rise (e.g.,
Page et al., 2007; Johnson, 2010), beginning in
the late Hirnantian. Very few studies document-
ing regional or global sea-level changes attempt
to interpret the records continuously through the
Late Ordovician to early Silurian (see Munnecke
et al., 2010). However, most sea-level compila-
tions indicate that sea level rose gradually or
episodically through the Rhuddanian, but did
not reach levels as high as those of the Katian
Environmental changes in the Late Ordovician–early Silurian: Review and new insights from black shales and nitrogen isotopes
Geological Society of America Bulletin, November/December 2013 1659
(e.g., Haq and Schutter, 2008). Therefore, it is
unlikely that the much greater extent of black
shales seen in the early Rhuddanian, as com-
pared with the late Katian, can be attributed to
higher global sea levels, per se, although shorter-
term transgression-highstand cycles almost cer-
tainly played a key role in their formation (e.g.,
Armstrong et al., 2009b). Signifi cant transgres-
sion of the incised glacial lowstand may have
contributed to strong sediment condensation on
clastic shelves such as the Gondwanan margin,
Iberia, and Perunica, thus contributing to organic
enrichment of shales, if not anoxia itself.
Black Shales on the Gondwanan Margin
As noted earlier herein (see review), the
spread of organic-rich black shales in Gond-
wana in the early Rhuddanian is widely attrib-
uted to a combination of increased weathering
inputs of nutrients to the oceans from formerly
glaciated regions aided by higher temperatures,
combined with increased levels of salinity strati-
fi cation due to increased freshwater input (e.g.,
Page et al., 2007; Armstrong et al., 2009b).
These combined factors, together with the
increased effi ciency of P recycling under low-
oxygen bottom-water conditions, could have
resulted in the spread of anoxia observed in the
peri-Gondwanan regions adjacent to the North
African–Arabian margin. There is a question as
to whether the inputs of nutrients and freshwater
from glacial melting could have been sustained
over suffi ciently long geological periods to pro-
duce the record of anoxia (over more than 1 m.y.
in duration) seen at the great majority of the
peri-Gondwanan localities (e.g., Prague Basin,
Thuringia, Carnic Alps, S. France, SW and SE
Spain, Portugal, Sardinia, E. Serbia, Bulgaria, E
Taurides of Turkey; see Table 1). However, as
noted already, once anoxia is established within
a particular marine shelf or basin, the enhanced
P recycling will contribute to preserving the state
of anoxia over longer geological time intervals.
In addition, there is clumped-isotope evidence
that some glacial ice remained on the continents
during the Rhuddanian (e.g., Finnegan et al.,
2011), as well as stratigraphic evidence for
the persistence of ice in South America (Díaz-
Martínez and Grahn, 2007; Díaz-Martínez et al.,
2011) and likely parts of North Africa (Moreau,
2011; Le Heron et al., 2013). This, combined
with the fact that global sea level appears to
have risen gradually (and probably episodically)
through the Rhuddanian (e.g., Munnecke et al.,
2010), suggests that there may have been suc-
cessive episodes of glacial retreat and release
of glacial sediments, nutrients, and freshwater.
Considerably more research needs to be done to
document the nature and frequency of episodic
sea-level change in the Rhuddanian.
Black Shales in the Paleotropics
The processes described herein provide
hypotheses for the development of anoxia in
high-latitude shelf and deep-water margins of
Gondwana. In addition, the localities where
black shales were deposited in the late Katian,
likely mainly regions of upwelling and/or basin
restriction, are marked by a return to apparently
persistent anoxic conditions in the early Rhud-
danian (see discussion; Fig. 4C). We reiterate
here that many of the regions from which our
black shale data have come are not suffi ciently
well studied to be certain that these successions
actually represent deposition under fully anoxic
(as opposed to dysoxic) bottom waters. Bearing
this in mind, however, one of the most striking
aspects of the early Rhuddanian oceanic anoxic
event is the spread of dysoxia/anoxia into other
mid- and low-latitude regions where deep waters
were oxygenated both in the late Katian and in
the Hirnantian glacial interval. No recent, quan-
titative ocean circulation modeling studies have
been conducted for the early Rhuddanian time
interval. Therefore, any suggestions regarding
the causal factors resulting in the distribution of
dysoxia/anoxia in this time interval are neces-
sarily speculative. Nevertheless, we feel that it
is worthwhile to consider some of the possible
causal factors as a means to leading to further
insights into our understanding of this interval
in Earth history.
Modeling studies that have examined deep-
ocean circulation in the Late Ordovician have
suggested that the principal source of deep
waters for the Southern Hemisphere and tropical
oceans of the world was the southern, high-lati-
tude ocean region adjacent to the Gondwanan
margin (Poussart et al., 1999; Herrmann et al.,
2004). In late Katian and mid-Hirnantian time,
waters of the shelves adjacent to these regions
were oxygenated, as indicated by the oxidation
state of sediments deposited in peri-Gondwanan
localities (Figs. 4A and 4B), suggesting that the
adjacent deeper waters were likely well oxygen-
ated as well. These downwelling waters were
likely the main source of oxygen for much of
the world’s deep oceans through the late Katian
to mid-Hirnantian (Poussart et al., 1999; Herr-
mann et al., 2004). However, as noted earlier,
in the early Rhuddanian, many basins along the
Gondwanan margin were sites of deposition of
black shales (Fig. 4C; Table 1), indicating that
the water mass structure of the southern ocean
adjacent to the northern Gondwanan margin
must have been confi gured differently in the
Rhuddanian than it was during the Katian. Melt-
ing ice sheets, increased nutrient input, and
changes in moisture transport may have fresh-
ened the waters of the peri-Gondwanan region,
thus inhibiting deep-water formation through
salinity stratifi cation (e.g., Broecker et al.,
1989). Ocean downwelling may have relocated
northward to areas occupied by more saline
surface waters, and because these waters would
be warmer, they would also be less saturated
with oxygen, and their capacity to ventilate the
oceans would be correspondingly reduced. We
therefore hypothesize that a major mode-shift in
the thermohaline circulation regime contributed
to the spread of anoxia in the world’s oceans
during the Rhuddanian.
This speculative hypothesis will require test-
ing through integration of detailed lithological,
faunal, and geochemical studies at many locali-
ties around the world, combined with rigorous
climate-ocean modeling studies.
Spread of Land Plants during
the Hirnantian?
Another possible source of difference
between late Katian and early Rhuddanian cir-
cumstances may have involved the expansion
of land plants and consequent changes in their
contribution of terrestrially derived nutrients to
conti nental shelves. Vecoli et al. (2011) provided
evidence that the diversity of cryptospores may
have increased signifi cantly through the Hirnan-
tian. If their results apply to the globe gener-
ally, and if the increase cryptospore diversity is
indicative of increasing ecological diversity of
land plants, then the retreat of glacial ice from
Gondwana and warming of higher-paleolatitude
regions may have fostered a signifi cant increase
in the spread of land plants in regions affected
by climate amelioration. This may have had a
signifi cant impact on rates of weathering, lead-
ing to pulses of increased P input into the oceans
and increased global productivity (Lenton
et al., 2012).
Black Shale Episodes and Plankton Biotas
Recent studies have shown that several dif-
ferent environmental processes played impor-
tant roles in generating the two phases of the
Hirnantian mass extinction. These include: sea-
level change and associated changes in habitable
shelf area; global temperature changes and shifts
in climate belts; changes in ocean chemistry,
particularly oxygenation; patterns of nutrient
cycling; and changes in global paleogeography
(see review section). Although each of these dif-
ferent factors, and the interaction between them,
likely played a role in controlling changing bio-
diversity through the Hirnantian mass extinction,
the relative importance of each of the factors
was likely different for different groups of taxa
and at different times through the Hirnantian.
Quantifying the relative contribution of differ-
ent environmental/ecological processes remains
a signifi cant research challenge. In this section,
Melchin et al.
1660 Geological Society of America Bulletin, November/December 2013
we discuss the specifi c role of changing patterns
of dysoxia/anoxia and changing communities of
primary producers on the biodiversity patterns
of graptolites, in light of our new black shale dis-
tribution and N isotope data.
It is now widely agreed that the maximum
abundance and diversity of graptolite faunas
occur in black shales. It is hypothesized that
the preferred habitat for most pelagic grapto-
lites was the eutrophic waters associated with
upwelling centers in deep-shelf and slope set-
tings and deep, epicontinental seas, although
some species were clearly adapted to a well-
oxygenated near-surface and shallower epicra-
tonic biotope (see review in Cooper et al., 2012).
Moreover, it has been specifi cally suggested
that many graptolites lived in association with
the dysoxic denitrifi cation zone in high-pro-
ductivity waters. (e.g., Berry and Wilde, 1987;
Finney et al., 2007). It is therefore signifi cant
that at both the Vinini Creek section in Nevada
and the Truro Island section in Arctic Canada,
the timing of an episode of pronounced change
in the graptolite faunas, including a decline in
diversity and even more profound change in
the patterns of taxonomic dominance (Mitchell
et al., 2007; Štorch et al., 2011; authors’ per-
sonal observations, 2011), occurred precisely
within the stratigraphic interval in which there
is an increase in δ15N values, which we interpret
to represent the fundamental shift in the nitro-
gen cycling regime that occurred in response
to changes in intensity of denitrifi cation. The
fact that the change in the graptolites at these
two sections does not occur at the same level in
relation to the onset of the Hirnantian carbon
isotope excursion (unlike Vinini Creek, at Truro
Island the onset of the Hirnantian carbon isotope
excursion occurs below the change in graptolite
faunas) supports the suggestion that the shifts
in the nutrient cycling and composition of pri-
mary producer communities are critical factors
in this episode of graptolite faunal turnover.
In addition, the change in graptolite faunas is
marked by a replacement of diplograptine grap-
tolites typical of the late Katian paleotropics
by taxa that appear to be derived from higher-
paleolatitude settings (e.g., neograp tines), such
as Bohemia and Iberia (quantitatively docu-
mented by Goldman et al., 2011), where the late
Katian sediments do not indicate strong oxygen
depletion and intense denitrifi cation. This sup-
ports the hypothesis that the eutrophic, strongly
denitrifying waters were a preferred habitat for
many of the late Katian graptolites of the paleo-
tropics, and that the loss of this habitat, together
with the fundamental change in the community
of primary producers, in the early Hirnantian
may have been an important factor in their pro-
found extinction through this interval (Chen
et al., 2005; Finney et al., 2007; Mitchell et al.,
2007; Sadler et al., 2011). Graptolites show pro-
gressive increases in diversity through the late
Hirnantian–early Rhuddanian, although punctu-
ated by highly variable rates of origination and
extinction, which may be associated with the
environmental instability or cyclicity accompa-
nying the return of widespread oxygen-deple-
tion following the end of the Hirnantian glacial
maximum (Chen et al., 2005).
Unfortunately, none of our studied sections
has yielded a good record of fossil micro phyto-
plankton (acritarchs). However, recent studies
of the patterns of acritarch abundances from
late Katian to late Hirnantian time elsewhere
indicate that the Katian-Hirnantian boundary
was marked by an event of signifi cant fl oral
turnover, with important changes in phyto-
plankton assemblage composition and, in some
cases, slight increases in diversity (Vecoli,
2008; Delabroye et al., 2011). This is consis-
tent with the patterns seen in the distribution
of black shales and our N isotope data, which
suggest that global ocean waters became more
oxygenated, particularly in the tropics, allowing
the phytoplankton communities to become less
dependent on the supplies of fi xed nitrogen from
cyanobacteria living in the photic zone. On the
other hand, the transition through the Hirnantian
into the Rhuddanian appears to show a striking
extinction event in the acritarchs (Vecoli, 2008;
Delabroye et al., 2011), associated with the
onset of the Rhuddanian oceanic anoxic event.
This suggests that this event included another
signifi cant shift in the phytoplankton assem-
blages—this time returning to greater depen-
dence on cyanobacterial N fi xation, as indicated
by our N isotope data.
Whereas the onset of the Rhuddanian oce-
anic anoxic event coincides, at least generally,
with the second phase of the Hirnantian mass
extinction event, the persistence of more wide-
spread oxygen depletion through the Rhudda-
nian, and even into the mid-Aeronian in many
regions (e.g., Štorch, 2006; Page et al., 2007;
Fan et al., 2011), is accompanied by a rela-
tively slow recovery of diversity in many groups
of organisms following the Hirnantian mass
extinction (e.g., Sheehan, 2001). Even the grap-
tolites, which show a relatively rapid recovery
in species diversity in the Rhuddanian, exhibit
a signifi cant delay in the return of morphologic
disparity (Melchin et al., 2011; Bapst et al.,
2012). The fact that morphological diversifi ca-
tion lagged behind species richness suggests
that the early Rhuddanian witnessed a similarly
slow recovery of ecological interactions within
the plankton (Bapst et al., 2012).
Hammarlund et al. (2012) proposed that the
development and spread of deep-ocean anoxia
in the early Hirnantian resulted in extinctions
among deep-shelf benthos as well as planktonic
and nektonic organisms. They suggested, in par-
ticular, that the spread of deep-sea anoxia would
result in extinctions of deeper-water plankton
that hovered in the dysoxic transition zone.
However, the patterns seen in the extinctions of
graptolites at the base of the Hirnantian in the
paleotropics do not match the predictions of this
hypothesis. In the paleotropics, graptolite faunas
in relatively shallow-water localities (Arctic
Canada, some South China localities) show a
very abrupt turnover to very low-diversity lower
Hirnantian faunas (e.g., Chen et al., 2005). On
the other hand, at deeper-water localities, where
the graptolite faunas contain a higher propor-
tion of deep-water taxa, the faunal turnover is
more gradual in terms of the progressive loss of
taxa (Chen et al., 2005; Mitchell et al., 2007).
If development of anoxia in deep water was
the primary kill mechanism, the reverse pat-
tern would be expected, with higher rates of
survivorship at shallower-water localities less
directly affected by the inferred deep-water
anoxia. We note, however, that even though
this may not have been an important extinction
mechanism for graptolites at the beginning of
the Hirnantian, it could have been a signifi cant
factor affecting graptolites in the late Hirnan-
tian, when graptolite faunas appear to have been
diversifying within less oxygen-depleted ocean
waters, which then returned to more strongly
anoxic conditions in many regions.
As noted in the review section, Logan et al.,
(1995) and Butterfi eld (2009, 2011) proposed
that ecological drivers may have also played a
role in controlling the development and main-
tenance of anoxia in open-ocean waters. Butter-
fi eld (2009, 2011) suggested that evidence from
organic biomarkers and nitrogen isotopes indi-
cates that phytoplankton communities during
some Phanerozoic oceanic anoxia events may
have returned to cyanobacterially dominated
communities similar to those present during
the Proterozoic. Our N isotope data support this
hypothesis, indicating that the black shale inter-
vals, particularly in the late Katian and early
Rhuddanian, were characterized by the presence
of abundant cyanobacteria. These cyanobacteria
may have contributed to the eutrophication of
waters of the photic zone, resulting from their
small size and slow settling rates, enhancing
the rate and effi ciency of recycling of nutrients
within the surface waters.
CONCLUSIONS
The past 25 yr has been a period of remarkable
growth in our knowledge of the environmental
and biotic changes that took place in Late Ordo-
Environmental changes in the Late Ordovician–early Silurian: Review and new insights from black shales and nitrogen isotopes
Geological Society of America Bulletin, November/December 2013 1661
vician to earliest Silurian time. New insights are
leading to a better understanding of the long-
term linkages between the processes that control
tectonics, climate, ocean circulation, and chang-
ing biodiversity. Review of the recent literature
leads to the following conclusions pertaining to
the environmental and biotic changes that took
place through Katian to Rhuddanian time:
(1) Lithostratigraphic and chemostratigraphic
evidence clearly shows that the peak glacial
expansion of the Ordovician–Silurian took place
in early-mid-Hirnantian time and that several
additional episodes of glaciation extended into
the early and possibly late Silurian. Chemo-
stratigraphic and sequence-stratigraphic evi-
dence indicates that some Gondwanan glacia-
tion also occurred in Sandbian–Katian time, and
possibly earlier in the Ordovician, but poor pres-
ervation and age control in many pre-Hirnantian
successions in North Africa have hindered posi-
tive identifi cation of Sandbian–Katian glacial
deposits.
(2) Synthesis of geochemical, lithostrati-
graphic, and biostratigraphic data suggests that
the Hirnantian glaciation occurred in two major
phases, of which the second was more exten-
sive and produced a larger sea-level drawdown.
The onset of the fi rst Hirnantian ice advance on
Gondwana coincides approximately with the
base of the Hirnantian Stage, which is marked
by the advent of the Neograptina-dominated
graptolite faunas of the M. extraordinarius
Zone, and in some (but not all) regions by an
abrupt positive shift in δ13C. A brief but marked
transgression occurred during the mid-Hirnan-
tian interglacial, which is now securely placed
near the end of the M. extraordinarius Zone,
based on sequence-stratigraphic, chemostrati-
graphic, and biostratigraphic constraints. This
interglacial was followed by the larger, second
glacial advance and the most strongly developed
expression of the Hirnantian carbon isotope
excursion at many localities, which reached
peak δ13C values during the early part of the
M. persculptus Zone. It is important to note,
however, that this coincidence does not in itself
suggest a particular cause for the relatively large
isotopic excursion during this interval, other
than that the changes were likely related in some
way to the consequences of the glacial episode.
(3) Geochemical evidence suggests that
increased rates of weathering of basalts or the
newly uplifted Taconic and Caledonian moun-
tains, or both, may have led to increased draw-
down of atmospheric CO2 and global cooling,
thus triggering the development of the Gond-
wanan glaciers during the Ordovician. In addi-
tion, modeling evidence suggests that changing
sea level and plate positions through the Late
Ordovician may have led to changes in latitu-
dinal heat transport that contributed to the gla-
cial development. Orbitally controlled changes
in solar insolation may have been a factor in
governing inferred cyclical changes in climate
and glacio-eustasy, but there remain signifi cant
challenges in development of a suffi ciently pre-
cise chronology to verify the cycle durations
and identify the specifi c orbital forcing mecha-
nisms. Nevertheless, on present evidence, it
appears that the Hirnantian glacial interval may
have consisted of two major advance-retreat
cycles and fi ve included subcycles of ~200 k.y.
in duration.
(4) The Late Ordovician mass extinction
event similarly took place in two main stages,
one through the Katian-Hirnantian boundary
interval during the onset of the early Hirnantian
glaciation, and the second at the end of the later
peak glacial phase in late Hirnantian time. Sev-
eral extinction mechanisms have been proposed
that have received recent support, including:
loss of habitat due to sea-level fall, particularly
the draining of large epicontinental seas at the
onset of peak glaciation; changing climate and
major shifts in the distribution of global climate
belts; changes in patterns of ocean circulation
and oxygenation, leading to changes in nutrient
cycling; spread of anoxia; and changes in con-
tinental positions, resulting in loss of regional
faunal differentiation. We suggest that a com-
plex interplay among these different processes
probably resulted in the different patterns of
diversity change during each of the two phases
of the Hirnantian extinction. However, assess-
ment of the relative importance of these dif-
ferent mechanisms will remain an important
subject of future research.
(5) Our new compilation of data on the
global distribution of black shales reveals that
black shales were widespread in the late Katian,
particularly in the paleotropics, but were not
ubiqui tous. Further, this compilation indicates
that almost all of the regions of late Katian black
shale deposition appear to have been either sites
of oceanic upwelling or occurred within poten-
tially restricted or semirestricted settings, such
as foreland or backarc basins. We suggest that
several factors may have made oceanic upwell-
ing zones and semirestricted marine basins
more prone to the development of anoxia in the
Ordovician as compared with the late Cenozoic,
including: lower atmospheric oxygen levels,
leading to lower equilibrium saturation levels of
O2 in surface waters; warmer sea-surface tem-
peratures, which resulted in lower O2 solubility
in areas of peri-Gondwanan downwelling; and
higher global sea levels, allowing for the lateral
spread of oxygen minimum zones in the oceans.
In addition, once established, the presence
of bottom-water anoxia resulted in increased
rates of benthic P regeneration, thus enabling
increased production of organic carbon in the
photic zone and increased deoxygenation of
deep waters by organic matter remineralization.
Changes in the rate of deposition of biominer-
alized P through the Paleozoic and differences
in the rate of organic matter export from cyano-
bacterial versus algal-dominated phytoplankton
communities may have also been important fac-
tors controlling nutrient fl ux through the Late
Ordovician–early Silurian.
(6) Our black shale distribution data, which
are interpreted to represent localities spanning
a range of settings from deep shelf to base of
continental slope, indicate that anoxia was less
widespread at a wide range of depths in the
early-mid-Hirnantian than during either the late
Katian or the succeeding early Rhuddanian. In
addition, those successions that represent the
deep-water settings in which anoxia did occur
in the early-mid-Hirnantian show lower concen-
trations of organic matter compared with black
shales in the underlying or overlying strata.
Thus, we see no direct evidence that the posi-
tive Hirnantian δ13C excursion was the result of
high rates of organic carbon burial in the deep
ocean despite suggestions that this was the
likely cause. However, it is possible there was an
increase in the rate of burial of organic matter in
prodelta slope and submarine fan environments
during the early-mid-Hirnantian lowstand,
where its presence would be masked by high
rates of sediment dilution.
(7) Black shales occur at all studied paleo-
latitudes and a wide range of water depths, in
shelf to continental rise settings, in the early
Rhuddanian. This time interval can, therefore,
be characterized as an oceanic anoxic event,
comparable to those of the Mesozoic. In addi-
tion to the regions interpreted to be upwelling
centers and semirestricted basins (as in the late
Katian), black shales occur both in the higher-
paleolatitude regions of Gondwana and the peri-
Gondwanan terranes, as well as other paleo-
tropical sites.
Development of black shales in the high-
paleolatitude Gondwanan and peri-Gondwanan
regions may be interpreted to have resulted
from a strong infl ux of nutrients released with
sediments by the retreat of the Gondwanan ice
sheets, possibly combined with strong oceanic
stratifi cation induced by high rates of freshwater
input from melting ice. Once established, anoxia
could then be sustained by enhanced P regenera-
tion and recycling under anoxic bottom-water
conditions. Spread of early land plants may have
also been a factor affecting the release of nutri-
ents from terrestrial soils.
Modeling of ocean circulation indicates that
the Gondwanan margin may have been the
Melchin et al.
1662 Geological Society of America Bulletin, November/December 2013
primary source of waters downwelling into the
deep ocean in the Late Ordovician. This fi nding
is consistent with the lack of evidence for anoxia
in the shales of Katian and early Hirnantian age
deposited in deep-shelf and slope settings in this
region. However, the transition to much more
widely distributed black shales in the Rhud-
danian suggests that the deep-water circulation
regime operated differently than it did in the
Katian. Melting ice sheets and changes in mois-
ture transport may have freshened the waters of
the region, thus inhibiting deep-water forma-
tion through salinity stratifi cation, a scenario
that could also lead to development of anoxia
at relatively shallow water depths. Ocean down-
welling may have relocated to areas of warmer
and more saline surface waters containing less
dissolved oxygen, and, as such, their capacity to
ventilate the oceans would be correspondingly
reduced. We suggest that a major mode-shift in
the thermo haline circulation regime contributed
to the spread of anoxia in the world’s oceans dur-
ing the Rhuddanian.
(8) Our new nitrogen isotope data, together
with those of LaPorte et al. (2009), allow us
to infer that the deep oceans in the late Katian
and late Hirnantian–Rhuddanian were severely
depleted in fi xed N, which means that very
little bioavailable nitrogen could be returned to
the photic zone by oceanic upwelling to sup-
port algal productivity. This led to an increase
in cyanobacteria in the photic zone, as these
organisms could produce their own fi xed nitro-
gen. Under the nitrogen-limiting conditions, the
cyanobacteria could more effectively compete
with the algae for reactive P in the water and
other nutrients. Thus, algal productivity became
dependent on cyanobacterial productivity dur-
ing times of widespread anoxia and denitrifi ca-
tion in deep-ocean waters (LaPorte et al., 2009).
However, there was a shift toward higher δ15N
values in the early-mid-Hirnantian, suggesting
a weakening of the intensity of denitrifi cation
and therefore oxygen depletion in these regions,
resulting in a shift in composition of the phyto-
plankton communities. At several studied sites,
this interval was accompanied by a brief resur-
gence in the abundance of the Diplograptina. In
contrast, data from a higher-paleolatitude setting
in peri-Gondwanan Europe through the Katian-
Hirnantian transition show that in those regions,
the ocean remained oxygenated with a continu-
ous supply of fi xed N to the surface waters.
(9) It is signifi cant that in our paleotropical,
Laurentian sections, the timing of an episode
of pronounced change in the graptolite faunas,
including a decline in diversity and even more
profound change in the patterns of taxonomic
dominance, occurred precisely within the strati-
graphic interval in which there is an increase
in δ15N values, which we interpret to represent
a change in the contribution of cyanobacterial
N fi xation to the phytoplankton productivity.
In addition, the change in graptolite faunas is
marked by a replacement of graptolites typical
of the late Katian paleotropics with taxa that
appear to be derived from higher-paleolatitude
settings, where the late Katian sediments do not
indicate strong oxygen depletion and intense
denitrifi cation. This supports the hypothesis that
the eutrophic, strongly denitrifying waters were
a preferred habitat for many of the late Katian
graptolites of the paleotropics and that the loss
of this habitat, and the accompanying change
in the community of primary producers in the
early Hirnantian, may have been an important
factor in their profound extinction through this
interval.
ACKNOWLEDGMENTS
We are grateful to Ron Blakey for providing us
with high-resolution 450 Ma and 440 Ma paleogeo-
graphic maps and also to Seth Finnegan for sharing
his isotopic data set. This paper has benefi ted sig-
nifi cantly from critical reviews by S. Finnegan and
H.A. Armstrong. Financial support for this research
was provided by Natural Sciences and Engineer-
ing Research Council (Canada) Discovery Grants to
Melchin and Holmden, U.S. National Science Foun-
dation grant EAR-0418790 to Mitchell and Melchin,
and the Czech Science Foundation grant 205/09/0619
to Štorch. Holmden thanks Dinka Beisic for technical
support in the Saskatchewan Isotope Laboratory and
for performing the N isotope analyses. Jim Rosen is
thanked for electronics support on the mass spectrom-
eters. This paper is a contribution to the International
Geoscience Programme Project 591: The Early to
Middle Paleozoic Revolution.
REFERENCES CITED
Achab, A., Asselin, E., Desrochers, A., Riva, J.F., and Farley,
C., 2011, Chitinozoan biostratigraphy of a new Upper
Ordovician stratigraphic framework for Anticosti Is-
land, Canada: Geological Society of America Bulletin,
v. 123, p. 186–205, doi:10.1130/B30131.1.
Ader, M., Cartigny, P., Boudou, J.-P., Petit, E., Oh, J.H., and
Javoy, M., 2006, Nitrogen isotopic evolution of carbo-
naceous matter during metamorphism methodology
and preliminary results: Chemical Geology, v. 232,
p. 152–169, doi:10.1016/j.chemgeo.2006.02.019.
Ainsaar, L., Kaljo, D., Martma, T., Meidla, T., Mannik, P.,
Nōlvak , J., and Tinn, O., 2010, Middle and Upper
Ordovician carbon isotope chemostratigraphy in Balto-
scandia: A correlation standard and clues to environ-
mental history: Palaeogeography, Palaeoclimatology,
Palaeoecology, v. 294, p. 189–201, doi:10.1016
/j.palaeo.2010.01.003.
Algeo, T.J., and Ingall, E., 2007, Sedimentary Corg:P ratios ,
paleocean ventilation, and Phanerozoic atmospheric
pO2: Palaeogeography, Palaeoclimatology, Palaeoecol-
ogy, v. 256, p. 130–155, doi:10.1016/j.palaeo.2007
.02.029.
Allison, P.A., Wignall, P.B., and Brett, C.E., 1995, Palaeo-
oxygenation: Effects and recognition, in Bosence,
D.W.J., and Allison, P.A., eds., A Review of Marine
Palaeo environmental Analysis from Fossils: Geological
Society of London Special Publication 83, p. 97–112.
Alroy, J., 2008, Dynamics of origination and extinction in the
marine fossil record: Proceedings of the National Acad-
emy of Sciences of the United States of America, v. 105,
p. 11,536–11,542, doi:10.1073/pnas.0802597105.
Alroy, J., 2010, Geographical, environmental and intrinsic
biotic controls on Phanerozoic marine diversifi ca-
tion: Palaeontology, v. 53, p. 1211–1235, doi:10.1111
/j.1475-4983.2010.01011.x.
Altabet, M.A., Francois, R., Murray, D.W., and Prell, W.L.,
1995, Climate-related variations in denitrification
in the Arabian Sea from sediment N-15/N-14 ratios:
Nature, v. 373, no. 6514, p. 506–509, doi:10.1038
/373506a0.
Alvarez, L.W., Alvarez, W., Asaro, F., and Michel, H.V.,
1980, Extraterrestrial cause for the Cretaceous-Tertiary
boundary extinction: Science, v. 208, p. 1095–1108,
doi:10.1126/science.208.4448.1095.
Álvaro, J.J., Vennin, E., Villas, E., Destombes, J., and Viz-
caïno, D., 2007, Pre-Hirnantian (latest Ordovician)
benthic community assemblages: Controls and replace-
ments in a siliciclastic-dominated platform of the eastern
Anti-Atlas, Morocco: Palaeogeography, Palaeoclima-
tology, Palaeoecology, v. 245, p. 20–36, doi:10.1016
/j.palaeo.2005.09.035.
Apollonov, M.K., Koren’, T.N., Nikitin, I.F., Paletz, L.M.,
and Tsai, D.T., 1988, Nature of the Ordovician-Silurian
boundary in south Kazakhstan, USSR: Bulletin of the
British Museum of Natural History (Geology), v. 43,
p. 145–154.
Armstrong, H.A., 2007, On the cause of Ordovician glacia-
tion, in Williams, M., Haywood, A.M., Gregory, F.J.,
and Schmidt, D.N., eds., Deep-Time Perspectives on
Climate Change: Marrying the Signal from Computer
Models and Biological Proxies: The Micropalaeonto-
logical Society Special Publications: London, Geologi-
cal Society of London, p. 101–121.
Armstrong, H.A., and Coe, A.L., 1997, Deep-sea sediments
record the geophysiology of the Late Ordovician gla-
ciation: Journal of the Geological Society of London,
v. 154, p. 929–934, doi:10.1144/gsjgs.154.6.0929.
Armstrong, H.A., and Owen, A.W., 2001, Terrane evolution
of the paratectonic Caledonides of northern Britain:
Journal of the Geological Society of London, v. 158,
p. 475–486, doi:10.1144/jgs.158.3.475.
Armstrong, H.A., Turner, B.R., Makhlouf, I.M., Weedon,
G.P., Williams, M., Al Smadi, A., and Abu Salah, A.,
2005, Origin, sequence stratigraphy and depositional
environment of an Upper Ordovician (Hirnantian) de-
glacial black shale, Jordan: Palaeogeography, Palaeo-
climatology, Palaeoecology, v. 220, p. 273–289, doi:
10.1016/j.palaeo.2005.01.007.
Armstrong, H.A., Turner, B.R., Makhlouf, I.M., Weedon,
G.P., Williams, M., Al Smadi, A., and Abu Salah, A.,
2006, Reply to “Origin, sequence stratigraphy and
depositional environment of an Upper Ordovician
(Hirnantian) deglacial black shale, Jordan”: Palaeo-
geography, Palaeoclimatology, Palaeoecology, v. 230,
p. 356–360, doi:10.1016/j.palaeo.2005.10.005.
Armstrong, H.A., Baldini, J.U.L., Challands, T.J., Grocke,
D.R., and Owen, A.W., 2009a, Response of the Inter-
tropical Convergence Zone to Southern Hemisphere
cooling during Upper Ordovician glaciation: Palaeo-
geography, Palaeoclimatology, Palaeoecology, v. 284,
p. 227–236, doi:10.1016/j.palaeo.2009.10.001.
Armstrong, H.A., Abbott, G.D., Turner, B.R., Makhlouf,
I.M., Muhammad, A.B., Pedentchouk, N., and Peters,
H., 2009b, Black shale deposition in an Upper Ordovi-
cian–Silurian permanently stratifi ed, peri-glacial basin,
southern Jordan: Palaeogeography, Palaeoclimatol-
ogy, Palaeoecology, v. 273, p. 368–377, doi:10.1016
/j.palaeo.2008.05.005.
Arthur, M.A., and Sageman, B.B., 1994, Marine black
shales: Depositional mechanisms and environments
of ancient deposits: Annual Review of Earth and Plan-
etary Sciences, v. 22, p. 499–551, doi:10.1146/annurev
.ea.22.050194.002435.
Arvidson, R.S., Mackenzie, F.T., and Guidry, M., 2006,
MAGic: A Phanerozoic model for the geochemical
cycling of major rock-forming components: American
Journal of Science, v. 306, p. 135–190, doi:10.2475/ajs
.306.3.135.
Astini, R.A., and Dávila, F.M., 2004, Ordovician back arc
foreland and Ocloyic thrust belt development on the
western Gondwana margin as a response to Precor-
dillera terrane accretion: Tectonics, v. 23, TC4008,
doi:10.1029/2003TC001620.
Environmental changes in the Late Ordovician–early Silurian: Review and new insights from black shales and nitrogen isotopes
Geological Society of America Bulletin, November/December 2013 1663
Astini, R.A., Collo, G., and Martina, F., 2007, Ordovician
K-bentonites in the upper-plate active margin of west-
ern Gondwana (Famatina Ranges): Stratigraphic and
palaeogeographic signifi cance: Gondwana Research,
v. 11, p. 311–325, doi:10.1016/j.gr.2006.05.005.
Baarli, B.G., Johnson, M.E., and Antoshkina, A.I., 2003, Si-
lurian stratigraphy and paleogeography of Baltica, in
Landing, E., and Johnson, M.E., eds., Silurian Lands
and Shelf Margins Exclusive of North America: New
York State Museum Bulletin 492, p. 3–34.
Bambach, R.K., 2006, Phanerozoic biodiversity mass ex-
tinctions: Annual Review of Earth and Planetary Sci-
ences, v. 34, p. 127–155, doi:10.1146/annurev.earth.33
.092203.122654.
Bapst, D.W., Bullock, P.C., Melchin, M.J., Sheets, H.D., and
Mitchell, C.E., 2012, Graptoloid diversity and disparity
became decoupled during the Ordovician mass extinc-
tion: Proceedings of the National Academy of Sciences
of the United States of America, v. 109, p. 3428–3433,
doi:10.1073/pnas.1113870109.
Barca, S., and Jaeger, H., 1992, New geological and bio-
stratigraphical data on the Silurian in SE-Sardinia:
Close affi nity with Thuringia: Bolletino della Societá
Geologica Italiana, v. 108, p. 565–580.
Barca, S., Durand-Delga, M., Rossi, P., and Štorch, P.,
1996, Les micaschistes panafricains de Corse et leur
couverture paléozoique: Leur interprétation au sein de
l’orogene varisque sud-européen: Comptes Rendus de
l’Academie des Sciences de Paris, v. 322, p. 981–989.
Barnes, C.R., and Williams, S.H., eds., 1991, Advances in
Ordovician Geology: Geological Survey of Canada
Paper 90–9, 336 p.
Bassett, M.G., 1985, Towards a “common language” in stra-
tigraphy: Episodes, v. 8, p. 87–92.
Bazhenova, T.K., 2009, Evolution of oil and gas genera-
tion in the Earth’s history and petroleum prediction in
sedimentary basins: Russian Geology and Geophysics,
v. 50, p. 308–319, doi:10.1016/j.rgg.2009.03.008.
Berger, W.H., and Thierstein, H.R., 1979, On Phanerozoic
mass extinctions: Naturwissenschaften, v. 66, p. 46–47,
doi:10.1007/BF00369357.
Berggren, W.A., and Hollister, C.D., 1977, Plate tectonics
and paleocirculation—Commotion in the ocean: Tec-
tonophysics, v. 38, p. 11–48, doi:10.1016/0040-1951
(77)90199-8.
Bergman, N.M., Lenton, T.M., and Watson, A.J., 2004,
COPSE: A new model of biogeochemical cycling
over Phanerozoic time: American Journal of Science,
v. 304, p. 397–437, doi:10.2475/ajs.304.5.397.
Bergström, S.M., Huff, W.D., Koren’, T.N., Larsson, K.,
Ahlberg, P., and Kolata, D.R., 1999, The 1997 core
drilling through Ordovician and Silurian strata at
Röstånga, S. Sweden: Preliminary stratigraphic assess-
ment and regional comparison: GFF, v. 121, p. 127–
135, doi:10.1080/11035899901212127.
Bergström, S.M., Saltzman, M.M., and Schmitz, B., 2006,
First record of the Hirnantian (Upper Ordovician) δ13C
excursion in the North American midcontinent and its
regional implications: Geological Magazine, v. 143,
p. 657–678, doi:10.1017/S0016756806002469.
Bergström, S.M., Chen, X., Gutiérrez-Marco, J.C., and
Dronov, A., 2009, The new chronostratigraphic clas-
sifi cation of the Ordovician System and its relations to
major regional series and stages and to δ13C chemo-
stratigraphy: Lethaia, v. 42, p. 97–107, doi:10.1111
/j.1502-3931.2008.00136.x.
Bergström, S.M., Young, S., and Schmitz, B., 2010, Katian
(Upper Ordovician) δ13C chemostratigraphy and se-
quence stratigraphy in the United States and Balto-
scandia: A regional comparison: Palaeogeography,
Palaeoclimatology, Palaeoecology, v. 296, p. 217–234,
doi:10.1016/j.palaeo.2010.02.035.
Bergström, S.M., Kleffner, M., Schmitz, B., and Cramer,
B.D., 2011, Revision of the position of the Ordovi-
cian-Silurian boundary in southern Ontario: Regional
chrono stratigraphic implications of δ13C chemostratig-
raphy of the Manitoulin Formation and associated
strata: Canadian Journal of Earth Sciences, v. 48,
p. 1447–1470, doi:10.1139/e11-039.
Berner, R.A., 1991, A model for atmospheric CO2 over
Phanero zoic time: American Journal of Science,
v. 291, p. 339–376, doi:10.2475/ajs.291.4.339.
Berner, R.A., 2006, GEOCARBSULF: A combined model for
Phanerozoic atmospheric O2 and CO2: Geochimica et
Cosmochimica Acta, v. 70, p. 5653–5664, doi:10.1016
/j.gca.2005.11.032.
Berner, R.A., 2009, Phanerozoic atmospheric oxygen: New
results using the GEOCARBSULF model: American
Journal of Science, v. 309, p. 603–606, doi:10.2475
/07.2009.03.
Berry, W.B.N., 1998, The Arabian Sea: A modern analogue for
North African–Southern European Silurian organic-rich
graptolite-bearing shales?: Temas Geológico-Mineros
ITGE, v. 23, p. 57–59.
Berry, W.B.N., 2010, Black shales: An Ordovician perspec-
tive, in Finney, S.C., and Berry, W.B.N., eds., The Ordo-
vician Earth System: Geological Society of America
Special Paper 466, p. 141–147.
Berry, W.B.N., and Wilde, P., 1978, Progressive ventila-
tion of the oceans—An explanation for the distribu-
tion of the Lower Paleozoic black shales: American
Journal of Science, v. 278, p. 257–275, doi:10.2475
/ajs.278.3.257.
Berry, W.B.N., Wilde, P., and Quinby-Hunt, M.S., 1987, The
oceanic non-sulfi dic oxygen minimum zone: A habitat
for graptolites?: Bulletin of the Geological Society of
Denmark, v. 35, p. 103–114.
Beuf, S., Biju-Duval, B., de Charpall, O., Rognon, P., Gariel,
O., and Bennacaf, A., 1971, Les grès du Paléozoïque
inférieur au Sahara, sédimentation et disconti nuitiés,
evolution structurale d’un craton: Institute Francais
Pétrole, Science et technique du Pétrole, 18, Éditions
Technip, 464 p.
Bjerrum, C.J., Bendtsen, J., and Legarth, J.J.F., 2006,
Modeling organic carbon burial during sea level
rise with reference to the Cretaceous: Geochemistry
Geophysics Geosystems, v. 7, p. 1–24, doi:10.1029
/2005GC001032.
Boucot, A.J., Rong, J., and Scotese, C.R., 2003, Pre-Hirnan-
tian Ashgill climatically warm event in the Mediterra-
nean region: Lethaia, v. 36, p. 119–131, doi:10.1080
/00241160310001245.
Boudou, J.P., Schimmelmann, A., Ader, M., Mastalerz, M.,
Sebilo, M., and Gengembre, L., 2008, Organic nitro-
gen chemistry during low-grade metamorphism: Geo-
chimica et Cosmochimica Acta, v. 72, p. 1199–1221,
doi:10.1016/j.gca.2007.12.004.
Boulila, S., Galbrun, B., Miller, K.G., Pekar, S.F., Browning,
J.V., Laskar, J., and Wright, J.D., 2011, On the origin
of Cenozoic and Mesozoic “third-order” eustatic se-
quences: Earth-Science Reviews, v. 109, p. 94–112,
doi:10.1016/j.earscirev.2011.09.003.
Boyer, D.L., and Droser, M.L., 2009, Palaeoecological pat-
terns within the dysaerobic biofacies: Examples from
Devonian black shales of New York State: Palaeo-
geography, Palaeoclimatology, Palaeoecology, v. 276,
p. 206–216, doi:10.1016/j.palaeo.2009.03.014.
Boyer, D.L., Owens, J.D., Lyons, T.W., and Droser, M.L.,
2011, Joining forces: Combined biological and
geochemical proxies reveal a complex but refi ned
high-resolution palaeo-oxygen history in Devonian
epeiric seas: Palaeogeography, Palaeoclimatology, Pa-
laeoecology, v. 306, p. 134–146, doi:10.1016/j.palaeo
.2011.04.012.
Brandes, J.A., Devol, A.H., and Deutsch, C., 2007, New
developments in the marine nitrogen cycle: Chemical
Reviews, v. 107, p. 577–589, doi:10.1021/cr050377t.
Brenchley, P.J., 1988, Environmental changes close to the
Ordovician-Silurian boundary: Bulletin of the Brit-
ish Museum of Natural History (Geology), v. 43,
p. 377–385.
Brenchley, P.J., 1989, The Late Ordovician extinction, in
Donovan, S.K., ed., Mass Extinctions: Processes and
Evidence: London, Belhaven Press, p. 104–132.
Brenchley, P.J., and Cullen, B., 1984, The environmental
distribution of associations belonging to the Hirnantia
fauna—Evidence from North Wales and Norway, in
Bruton, D.L., ed., Aspects of the Ordovician System:
Oslo, Norway, Universitetsforlaget, Palaeontologi-
cal Contributions from the University of Oslo, v. 295,
p. 113–125.
Brenchley, P.J., and Newall, G., 1984, Late Ordovician envi-
ronmental changes and their effect on faunas, in Bru-
ton, D.L., ed., Aspects of the Ordovician System: Oslo,
Norway, Universitetsforlaget, Palaeontological Contri-
butions from the University of Oslo, v. 295, p. 65–79.
Brenchley, P.J., and Štorch, P., 1989, Environmental changes
in the Hirnantian (Upper Ordovician) of the Prague
Basin, Czechoslovakia: Geological Journal, v. 24,
p. 165–181, doi:10.1002/gj.3350240302.
Brenchley, P.J., Romano, M., Young, T.P., and Štorch, P.,
1991, Hirnantian glaciomarine diamictites—Evidence
for the spread of glaciation and its effect on Upper
Ordo vician faunas, in Barnes, C.R., and Williams,
S.H., eds., Advances in Ordovician Geology: Geologi-
cal Survey of Canada Paper 90–9, p. 325–336.
Brenchley, P.J., Marshall, J.D., Carden, G.A.F., Robertson,
D.B.R., Long, D.G.F., Meidla, T., Hints, L., and Ander-
son, T.F., 1994, Bathymetric and isotopic evidence for
a short-lived Late Ordovician glaciation in a green-
house period: Geology, v. 22, p. 295–298, doi:10.1130
/0091-7613(1994)022<0295:BAIEFA>2.3.CO;2.
Brenchley, P.J., Marshall, J.D., and Underwood, C.J., 2001,
Do all mass extinctions represent an ecological crisis?
Evidence from the Late Ordovician: Geological Jour-
nal, v. 36, p. 329–340, doi:10.1002/gj.880.
Brenchley, P.J., Carden, G.A., Hints, L., Kaljo, D., Marshall,
J.D., Martma, T., Meidla, T., and Nõlvak, J., 2003, High-
resolution stable isotope stratigraphy of Upper Ordovi-
cian sequences: Constraints on the timing of bioevents
and environmental changes associated with mass ex-
tinction and glaciation: Geological Society of America
Bulletin, v. 115, p. 89–104, doi:10.1130/0016-7606
(2003)115<0089:HRSISO>2.0.CO;2.
Brenchley, P.J., Marshall, J.D., Harper, D.A.T., Buttler,
C.J., and Underwood, C.J., 2006, A Late Ordovician
(Hirnantian) karstic surface in a submarine channel,
recording glacio-eustatic sea-level changes: Meifod,
central Wales: Geological Journal, v. 41, p. 1–22, doi:
10.1002/gj.1029.
Broecker, W.S., Kennett, J.P., Flower, B.P., Teller, J.T.,
Trumbore, S., Bonani, G., and Wolfl i, W., 1989, Rout-
ing of meltwater from the Laurentide ice sheet dur-
ing the Younger Dryas cold episode: Nature, v. 341,
p. 318–321, doi:10.1038/341318a0.
Brooks, J., and Fleet, A.J., eds., 1987, Marine Petroleum
Source Rocks: Geological Society of London Special
Publication 26, 444 p.
Buggisch, W., and Astini, R.A., 1993, The Late Ordovician
ice age: New evidence from the Argentine Precordi-
llera, in Findley, R.H., Unrug, R., Banks, M.R., and
Veevers, J.J., eds., Gondwana Eight: Assembly, Evolu-
tion and Dispersal: Rotterdam, Netherlands, Balkema,
p. 439–447.
Buggisch, W., Joachimski, M.M., Lehnert, O., Bergström,
S.M., Repetski, J.E., and Webers, G.F., 2010, Did
intense volcanism trigger the fi rst Late Ordovician
icehouse?: Geology, v. 38, p. 327–330, doi:10.1130
/G30577.1.
Bustin, R.M., Barnes, M.A., and Barnes, W.C., 1990, Deter-
mining levels of organic diagenesis in sediments and
fossil fuels, in Mcllreath, I.A., and Morrow, D.W., eds.,
Diagenesis: Geoscience Canada Reprint Series 4: On-
tario, Canada, Runge Press, p. 205–226.
Butterfi eld, N.J., 2009, Oxygen, animals and oceanic ven-
tilation: An alternative view: Geobiology, v. 7, p. 1–7,
doi:10.1111/j.1472-4669.2009.00188.x.
Butterfi eld, N.J., 2011, Animals and the invention of the
Phanerozoic Earth system: Trends in Ecology & Evo-
lution, v. 26, p. 81–87, doi:10.1016/j.tree.2010.11.012.
Calner, M., 2008, Silurian global events—At the tipping
point of climate change, in Elewa, A.M.T., ed., Mass
Extinctions: Berlin, Springer-Verlag, p. 21–58.
Came, R.E., Eiler, J.M., Veizer, J., Azmy, K., Brand, U.,
and Weidman, C.R., 2007, Coupling of surface tem-
peratures and atmospheric CO2 concentrations during
the Palaeozoic Era: Nature, v. 449, p. 198–201, doi:
10.1038/nature06085.
Caputo, M.V., 1998, Ordovician-Silurian glaciations and
global sea-level changes, in Landing, E., and John-
son, M.E., eds., Silurian Cycles: Linkages of Dynamic
Stratigraphy with Atmospheric, Oceanic, and Tectonic
Changes; James Hall Centennial Volume: Albany, New
York, New York State Museum, p. 15–25.
Caputo, M.V., and Crowell, J.C., 1985, Migration of
glacial centres across Gondwana during Paleozoic
Melchin et al.
1664 Geological Society of America Bulletin, November/December 2013
Era: Geological Society of America Bulletin, v. 96,
p. 1020–1036, doi:10.1130/0016-7606(1985)96<1020:
MOGCAG>2.0.CO;2.
Challands, T.J., Armstrong, H.A., Maloney, D.P., Davies,
J.R., Wilson, D., and Owen, A.W., 2009, Organic-car-
bon deposition and coastal upwelling at mid-latitude
during the Upper Ordovician (late Katian): A case
study from the Welsh Basin, UK: Palaeogeography,
Palaeoclimatology, Palaeoecology, v. 273, p. 395–410,
doi:10.1016/j.palaeo.2008.10.004.
Chen, X., 1984, Infl uence of the Late Ordovician glacia-
tion on basin confi guration of the Yangtze Platform
in China: Lethaia, v. 17, p. 51–59, doi:10.1111/j.1502
-3931.1984.tb00665.x.
Chen, X., and Mitchell, C.E., 1996, Stratigraphic evidences
on the Taconic and Guangxian orogeny: Journal of
Stratigraphy, v. 20, p. 305–313.
Chen, X., Xiao, C., and Chen, H., 1987, Wufengian (Ashgil-
lian) graptolite faunal differentiation and anoxic envi-
ronment in South China: Acta Palaeontologica Sinica,
v. 26, p. 326–344.
Chen, X., Rong, J.Y., Mitchell, C.E., Harper, D.A.T., Fan,
J.X., Zhan, R.B., Zhang, Y.D., Li, R.Y. and Wang,
Y., 2000, Late Ordovician to earliest Silurian grapto-
lite and brachiopod biozonation from the Yangtze
region, South China, with a global correlation: Geo-
logical Magazine, v. 137, p. 623–650, doi:10.1017
/S0016756800004702.
Chen, X., Melchin, M.J., Fan, J., and Mitchell, C.E., 2003,
Ashgillian graptolite fauna of the Yangtze region and
the biogeographical distribution of diversity in the lat-
est Ordovician: Bulletin de la Société Géologique de
France, v. 174, p. 141–148, doi:10.2113/174.2.141.
Chen, X., Rong, J.Y., Li, Y., and Boucot, A.J., 2004, Fa-
cies patterns and geography of the Yangtze region,
South China, through Ordovician and Silurian transi-
tion: Palaeogeography, Palaeoclimatology, Palaeo-
ecology, v. 204, p. 353–372, doi:10.1016/S0031-0182
(03)00736-3.
Chen, X., Melchin, M.J., Sheets, H.D., Mitchell, C.E., and
Fan, J.-X., 2005, Patterns and processes of latest Ordo-
vician graptolites extinction and recovery based on
data from South China: Journal of Paleontology, v. 79,
p. 842–861, doi:10.1666/0022-3360(2005)079[0842:
PAPOLO]2.0.CO;2.
Cherns, L., and Wheeley, J.R., 2007, A pre-Hirnantian (Late
Ordovician) interval of global cooling—The Boda
event re-assessed: Palaeogeography, Palaeoclimatol-
ogy, Palaeoecology, v. 251, p. 449–460, doi:10.1016
/j.palaeo.2007.04.010.
Cherns, L., and Wheeley, J.R., 2009, Early Palaeozoic cool-
ing events: Peri-Gondwana and beyond, in Bassett,
M.G., ed., Early Palaeozoic Peri-Gondwanan Terranes:
New Insights from Tectonics and Biogeography: Geo-
logical Society of London Special Publication 325,
p. 256–278.
Churkin, M.J., and Carter, C., 1970, Early Silurian Grapto-
lites from Southeastern Alaska and their Correlation
with Graptolitic Sequences in North America and the
Arctic: U.S. Geological Survey Professional Paper
653, 51 p.
Cocks, L.R.M., 1985, The Ordovician-Silurian boundary:
Episodes, v. 8, p. 98–100.
Cocks, L.R.M., 1988, The Ordovician-Silurian boundary in
the Oslo region, Norway: Bulletin of the British Mu-
seum of Natural History (Geology), v. 43, p. 81–84.
Cocks, L.R.M., and Rickards, R.B., eds., 1988, A Global
Analysis of the Ordovician-Silurian Boundary: Brit-
ish Museum (Natural History) Bulletin 43 (Geology
Series), 394 p.
Cooper, R.A., and Sadler, P.M., 2010, Facies preference pre-
dicts extinction risk in Ordovician graptolites: Paleo-
biology, v. 36, p. 167–187, doi:10.1666/09043.1.
Cooper, R.A., and Sadler, P.M., 2012, The Silurian Period ,
in Gradstein, F.M., Ogg, J.G., and Smith, A.G., eds., A
Geologic Time Scale 2012: Elsevier Press, p. 489–523.
Cooper, R.A., Rigby, S., Loydell, D.K., and Bates, D.E.B.,
2012, Palaeoecology of the Graptoloidea: Earth-
Science Reviews, v. 112, p. 23–41, doi:10.1016
/j.earscirev.2012.01.001.
Cramer, B.D., Brett, C.E., Melchin, M.J., Männik, P., Kleff ner,
M.A., McLaughlin, P.I., Loydell, D.K., Munnecke, A.,
Jeppsson, L., Corradini, C., Brunton, F.R., and Saltzman,
M.R., 2011, Revised correlation of Silurian Provin-
cial Series of North America with global and regional
chronostratigraphic units and δ13Ccarb chemostratigra-
phy: Lethaia, v. 44, p. 185–202, doi:10.1111/j.1502
-3931.2010.00234.x.
Cuerda, A.J., Rickards, R.B., and Cingolani, C., 1988, A
new Ordovician-Silurian boundary section in San Juan
Province, Argentina, and its defi nitive graptolite fauna:
Journal of the Geological Society of London, v. 145,
p. 749–757, doi:10.1144/gsjgs.145.5.0749.
de Jong, K., Xiao, W., Windley, B.F., Masago, H., and Lo,
C.-h., 2006, Ordovician 40Ar/39Ar phengite ages from
the blueschist-facies Ondor Sum subduction-accretion
complex (Inner Mongolia) and implications for the
early Paleozoic history of continental blocks in China
and adjacent areas: American Journal of Science,
v. 306, p. 799–845, doi:10.2475/10.2006.02.
Delabroye, A., and Vecoli, M., 2010, The end-Ordovician
glaciation and the Hirnantian Stage: A global review
and questions about Late Ordovician event stratigra-
phy: Earth-Science Reviews, v. 98, p. 269–282, doi:
10.1016/j.earscirev.2009.10.010.
Delabroye, A., Munnecke, A., Vecoli, M., Copper, P., Tribo-
villard, N., Joachimski, M.M., Desrochers, A., and
Servais, T., 2011, Phytoplankton dynamics across the
Ordovician/Silurian boundary at low palaeolatitudes:
Correlations with carbon isotopic and glacial events:
Palaeogeography, Palaeoclimatology, Palaeoecology,
v. 312, p. 79–97, doi:10.1016/j.palaeo.2011.09.011.
Desrochers, A., Farley, C., Achab, A., Asselin, E., and Riva,
J.F., 2010, A far-fi eld record of the end Ordovician
glaciation: The Ellis Bay Formation, Anticosti Island,
eastern Canada: Palaeogeography, Palaeoclimatol-
ogy, Palaeoecology, v. 296, p. 248–263, doi:10.1016
/j.palaeo.2010.02.017.
Destombes, J., 1968, Sur la nature glaciaire des sediments
du groupe du deuxieme Bani, Ashgill superieur de
l’Anti-Atlas, Maroc: Compte Rendus Hedbomadaires
des Seances de l’Academie des Sciences, Paris, ser. D,
v. 267, p. 684–686.
Destombes, J., and Willefert, S., 1988, The Ordovician-
Silurian boundary in Morocco, in Cocks, L.R.M., and
Rickards, R.B., eds., A Global Analysis of the Ordo-
vician-Silurian Boundary: Bulletin of the British Mu-
seum of Natural History (Geology), v. 43, p. 165–170.
Dewing, K., and Obermajer, M., 2009, Lower Paleozoic
thermal maturity and hydrocarbon potential of the
Canadian Arctic archipelago: Bulletin of Canadian
Petroleum Geology, v. 57, p. 141–166, doi:10.2113
/gscpgbull.57.2.141.
Díaz-Martínez, E., and Grahn, Y., 2007, Early Silurian gla-
ciation along the western margin of Gondwana (Peru,
Bolivia and northern Argentina): Palaeogeographic and
geodynamic setting: Palaeogeography, Palaeoclima-
tology, Palaeoecology, v. 245, p. 62–81, doi:10.1016
/j.palaeo.2006.02.018.
Díaz-Martínez, E., Vavrdová, M., Isaacson, E., and Grahn,
Y., 2011, Early Silurian vs. Late Ordovician glaciation
in South America, in Gutiérrez-Marco, J.C., Rábano, I.,
and García-Bellido, D., eds., Ordovician of the World:
Cuadernos del Museo Geominero, Instituto Geológico
y Minero de España, Madrid, Spain, v. 14, p. 127–134.
Dover, J.H., Berry, W.B.N., and Ross, R.J., Jr., 1980, Ordo-
vician and Silurian Phi Kappa and Trail Creek Forma-
tions, Pioneer Mountains, Central Idaho—Stratigraphic
and Structural Revisions, and New Data on Graptolite
Faunas: U.S. Geological Survey Professional Paper
1090, p. 1–54.
Droser, M.L., Bottjer, D.J., Sheehan, P.M., and McGhee,
G.R., 2000, Decoupling of taxonomic and ecologic
severity of Phanerozoic marine mass extinctions:
Geology, v. 28, p. 675–678, doi:10.1130/0091-7613
(2000)28<675:DOTAES>2.0.CO;2.
Elrick, M., Reardon, D., Labor, W., Martin, J., Desrochers,
A., and Pope, M., 2013, Orbital-scale climate change
and glacioeustasy during the early Late Ordovician
(pre-Hirnantian) determined from δ18O values in
marine apatite: Geology, v. 41, p. 775–778, doi:10.1130
/G34363.1.
Emsbo, P., 1993, Cottonwood and Meadow Canyons, in
Finney, S.C., Perry, B.D., Emsbo, P., and Madrid,
R.J., eds., Crustal Evolution of the Great Basin and
Sierra Nevada: Boulder, Colorado, Cordilleran/Rocky
Mountain Section, Geological Society of America, and
Reno, Department of Geological Sciences, University
of Nevada , p. 197–230.
Fan, J.X., Peng, P., and Melchin, M.J., 2009, Carbon isotopes
and event stratigraphy near the Ordovician-Silurian
boundary, Yichang, South China: Palaeogeography,
Palaeoclimatology, Palaeoecology, v. 276, p. 160–169,
doi:10.1016/j.palaeo.2009.03.007.
Fan, J.X., Melchin, M.J., Chen, X., Wang, Y., Zhang, Y.D.,
Chen, Q., Chi, Z.L., and Chen, F., 2011, Biostra-
tigraphy and geography of the Ordovician-Silurian
Lungmachi black shales in South China: Science
China–Earth Science, v. 54, p. 1854–1863.
Fatka, O., and Mergl, M., 2009, The “microcontinent” Pe-
runica: Status and story 15 years after conception, in
Bassett, M.G., eds., Early Palaeozoic Peri-Gondwana
Terranes: New Insights from Tectonics and Biogeogra-
phy: Geological Society of London Special Publication
325, p. 65–101.
Fergusson, C.L., 2003, Ordovician-Silurian accretion tec-
tonics of the Lachlan fold belt, southeastern Australia:
Australian Journal of Earth Sciences, v. 50, p. 475–
490, doi:10.1046/j.1440-0952.2003.01013.x.
Ferretti, A., 1998, Late Ordovician conodonts from the
Prague Basin, Bohemia, in Szaniawski, H., ed., Pro-
ceedings of the Sixth European Conodont Symposium
(ECOS VI): Acta Palaeontologica Polonica, v. 58,
p. 123–139.
Ferretti, A., Melchin, M.J., and Negri, A., 2012, Are there
black shales and black shales?: Bollettino della Società
Paleontologica Italiana, v. 51, p. 149–150.
Finlay, A.J., Selby, D., and Grocke, D.R., 2010, Tracking
the Hirnantian glaciation using Os isotopes: Earth and
Planetary Science Letters, v. 293, p. 339–348, doi:
10.1016/j.epsl.2010.02.049.
Finnegan, S., Bergmann, K., Eiler, J.M., Jones, D.S., Fike,
D.A., Eisenman, I., Hughes, N.C., Tripati, A.K., and
Fischer, W.W., 2011, The magnitude and duration of
Late Ordovician–Early Silurian glaciation: Science,
v. 331, p. 903–906, doi:10.1126/science.1200803.
Finnegan, S., Heim, N.A., Peters, S.E., and Fischer, W.W.,
2012, Climate change and the selective signature of
the Late Ordovician mass extinction: Proceedings
of the National Academy of Sciences of the United
States of America, v. 109, p. 6829–6834, doi:10.1073
/pnas.1117039109.
Finney, S.C., and Berry, W.B.N., 1997, New perspectives
on graptolite distributions and their use as indica-
tors of platform margin dynamics: Geology, v. 25,
p. 919–922, doi:10.1130/0091-7613(1997)025<0919:
NPOGDA>2.3.CO;2.
Finney, S.C., Cooper, J.D., and Berry, W.B.N., 1997, Late
Ordovician mass extinction: Sedimentologic, cy-
clostratigraphic, and biostratigraphic records from
platform and basin successions: Central Nevada:
Brigham Young University Geology Studies, v. 42,
p. 79–102.
Finney, S.C., Berry, W.B., Cooper, J.D., Rippardan, R.L.,
Sweet, W.C., Jacobson, S.R., Soufi ane, A., Achab, A.,
and Noble, P.J., 1999, Late Ordovician mass extinction:
A new perspective from stratigraphic sections in cen-
tral Nevada: Geology, v. 27, p. 215–218, doi:10.1130
/0091-7613(1999)027<0215:LOMEAN>2.3.CO;2.
Finney, S.C., Noble, P.J., and Cluer, J.K., 2000, Lower
Paleo zoic stratigraphy and structure of central Nevada :
Comparisons and contrasts between the lower and
upper plates of the Roberts Mountains thrust, in
Schweickert , R.A., Lahren, M.M., Karlin, R., Howle,
J., and Smith, K., eds., Great Basin and Sierra Nevada :
Geological Society of America Field Guide 2,
p. 279–300.
Finney, S.C., Berry, W.B.N., and Cooper, R.A., 2007, The
infl uence of denitrifying seawater on graptolite extinc-
tion and diversifi cation during the Hirnantian (latest
Ordovician) mass extinction event: Lethaia, v. 40,
p. 281–291, doi:10.1111/j.1502-3931.2007.00027.x.
Floyd, J.D., 2001, The Southern Uplands terrane: A strati-
graphical review: Transactions of the Royal Society
of Edinburgh–Earth Sciences, v. 91, p. 349–362, doi:
10.1017/S0263593300008233.
Environmental changes in the Late Ordovician–early Silurian: Review and new insights from black shales and nitrogen isotopes
Geological Society of America Bulletin, November/December 2013 1665
Fortey, R.A., 1989, There are extinctions and extinctions:
Examples from the Lower Palaeozoic: Philosophical
Transactions of the Royal Society of London, ser. B,
Biological Sciences, v. 325, p. 327–355, doi:10.1098
/rstb.1989.0092.
Fortey, R.A., and Cocks, L.R.M., 2005, Late Ordovician
global warming—The Boda event: Geology, v. 33,
p. 405–408, doi:10.1130/G21180.1.
Fortey, R.A., Harper, D.A.T., Ingham, J.K., Owen, A.W.,
Parkes, M.A., Rushton, A.W.A., and Woodcock, N.H.,
2000, A revised correlation of Ordovician rocks in the
British Isles: The Geological Society of London Spe-
cial Report 24, p. 1–83.
Foster, D.A., and Gray, D.R., 2000, Evolution and structure
of the Lachlan fold belt (orogen) of eastern Australia:
Annual Review of Earth and Planetary Sciences, v. 28,
p. 47–80, doi:10.1146/annurev.earth.28.1.47.
Frakes, L.A., 1979, Climates throughout Geological Time:
Amsterdam, Netherlands, Elsevier, 310 p.
Frakes, L.A., Francis, J.E., and Syktus, J.I., 1992, Climate
Modes of the Phanerozoic: Cambridge, UK, Cam-
bridge University Press, 274 p.
Ganeshram, R.S., Pedersen, T.F., Calvert, S.E., McNeill,
G.W., and Fontugne, M.R., 2000, Glacial-interglacial
variability in denitrifi cation in the world’s oceans:
Causes and consequences: Paleoceanography, v. 15,
p. 361–376, doi:10.1029/1999PA000422.
Ghienne, J.-F., 2003, Late Ordovician sedimentary environ-
ments, glacial cycles, and post-glacial transgression
in the Taoudeni Basin, West Africa: Palaeogeography,
Palaeoclimatology, Palaeoecology, v. 189, p. 117–145,
doi:10.1016/S0031-0182(02)00635-1.
Ghienne, J.F., 2011, The Late Ordovician glacial record:
State of the art, in Gutiérrez-Marco, J.C., Rábano, I.,
and García-Bellido, D., eds., Ordovician of the World:
Cuadernos del Museo Geominero, Instituto Geológico
y Minero de España, Madrid, Spain, v. 14, p. 13–19.
Ghienne, J.-F., Le Heron, D.P., Moreau, J., Denis, M., and
Deynoux, M., 2007, The Late Ordovician glacial sedi-
mentary system of the North Gondwana platform, in
Hambrey, M., Christoffersen, P., Glasser, N., Janssen, P.,
Hubbard, B., and Siegert, M., eds., Glacial Sedimentary
Processes and Products: International Association of
Sedimentologists Special Publication 39, p. 295–319.
Glorie, S., De Grave, J., Buslov, M.M., Zhimulev, F.I., Izmer,
A., Vandoorne, W., Ryabinin, A., Van den Haute, P.,
Vanhaecke, F., and Elburg, M.A., 2011, Formation and
Palaeozoic evolution of the Gorny-Altai-Altai-Mongolia
suture zone (south Siberia): Zircon U/Pb constraints on
the igneous record: Gondwana Research, v. 20, p. 465–
484, doi:10.1016/j.gr.2011.03.003.
Gogin, I.J., Koren’, T.N., Pegel’, T.V., and Sobolevskaya,
R.F., 1997, Atlas zonalnykh kompleksov veduscikh
grup rannepaleozoiskoi fauny severa Rossii: St. Peters-
burg, Izdatelstvo. VSEGEI, 205 p.
Goldman, D., and Bergström, S.M., 1997, Late Ordovician
graptolites from the North American midcontinent:
Palaeon tology, v. 40, p. 965–1010.
Goldman, D., Mitchell, C.E., Maletz, J., Riva, J.F.V., Leslie,
S.A., and Motz, G.J., 2007, Ordovician graptolites and
conodonts of the Phi Kappa Formation in the Trail
Creek region of central Idaho: A revised, integrated
biostratigraphy: Acta Palaeontologica Sinica, v. 46,
p. 155–162.
Goldman, D., Mitchell, C.E., Melchin, M.J., Fan, J.X.,
Wu, S.Y., and Sheets, H.D., 2011, Biogeography and
mass extinction: Extirpation and re-invasion of Nor-
malograptus species (Graptolithina) in the Late Ordo-
vician palaeotropics: Proceedings of the Yorkshire
Geological Society, v. 58, p. 227–246, doi:10.1144
/pygs.58.4.300.
Gorjan, P., Kaiho, K., Fike, D.A., and Chen, X., 2012, Car-
bon- and sulfur-isotope geochemistry of the Hirnantian
(Late Ordovician) Wangjiawan (Riverside) section,
South China: Global correlation and environmental
event interpretation: Palaeogeography, Palaeoclima-
tology, Palaeoecology, v. 337, p. 14–22, doi:10.1016
/j.palaeo.2012.03.021.
Gutiérrez-Marco, J.C., and Robardet, M., 1991, Découverte
de la zone à Parakidograptus acuminatus (base du
Llandovery) dans le Silurien du Synclinorium de Tru-
chas (Zone asturo-léonaise, Nord-Ouest de l’Espagne):
Conséquences stratigraphiques et paléogéographiques
au passage Ordovicien-Silurien: Comptes Rendus de
l’Academie des Sciences de Paris, v. 312, p. 729–734.
Gutiérrez-Marco, J.C., and Štorch, P., 1998, Graptolite bio-
stratigraphy of the Lower Silurian (Llandovery) shelf
deposits of the western Iberian Cordillera, Spain:
Geological Magazine, v. 135, p. 71–92, doi:10.1017
/S0016756897007802.
Gutiérrez-Marco, J.C., Ghienne, J.-F., Bernárdez, E., and
Hacar, M.P., 2010, Did the Late Ordovician African
ice sheet reach Europe?: Geology, v. 38, p. 279–282,
doi:10.1130/G30430.1.
Hambrey, M.J., and Harland, W.B., 1981, Earth’s Pre-Pleisto-
cene Glacial Record: Cambridge, UK, Cambridge Uni-
versity Press, 1004 p.
Hammarlund, E.U., Dahl, T.W., Harper, D.A.T., Bond,
D.P.G., Nielsen, A.T., Bjerrum, C.J., Schovsbo, N.H.,
Schönlaub, H.P., Zalasiewicz, J.A., and Canfi eld, D.E.,
2012, A sulfi dic driver for the end-Ordovician mass ex-
tinction: Earth and Planetary Science Letters, v. 331,
p. 128–139, doi:10.1016/j.epsl.2012.02.024.
Hamoumi, N., 1999, Upper Ordovician glaciation spreading
and its sedimentary record in Moroccan North Gond-
wana margin: Acta Universitatis Carolinae: Geologica,
v. 43, p. 111–114.
Haq, B.U., and Schutter, S.R., 2008, A chronology of Paleo-
zoic sea-level changes: Science, v. 322, p. 64–68, doi:
10.1126/science.1161648.
Harris, N.B., ed., 2005, The Deposition of Organic Carbon-
Rich Sediments: Models, Mechanisms and Conse-
quences: Society for Sedimentary Geology Special
Publication 82, 282 p.
Hedges, J.I., and Keil, R.G., 1995, Sedimentary organic-
matter preservation—An assessment and speculative
synthesis—Closing comment: Marine Chemistry, v. 49,
p. 137–139, doi:10.1016/0304-4203(95)00013-H.
Herrmann, A.D., Haupt, B.J., Patzkowsky, M.E., Seidov, D.,
and Slingerland, R.L., 2004, Response of Late Ordovi-
cian paleoceanography to changes in sea level, conti-
nental drift, and atmospheric pCO2: Potential causes
for long-term cooling and glaciation: Palaeogeography,
Palaeoclimatology, Palaeoecology, v. 210, p. 385–401,
doi:10.1016/j.palaeo.2004.02.034.
Herrmann, A.D., Macleod, K.G., and Leslie, S.A., 2010, Did
a volcanic mega-eruption cause global cooling dur-
ing the Late Ordovician?: Palaios, v. 25, p. 831–836,
doi:10.2110/palo.2010.p10-069r.
Herrmann, A.D., Leslie, S.A., and MacLeod, K.G., 2011,
Did intense volcanism trigger the fi rst Late Ordovi-
cian icehouse?: Comment: Geology, v. 39, e237, doi:
10.1130/G31758C.1.
Hetzel, A., Bottcher, M.E., Wortmann, U.G., and Brum-
sack, H.J., 2009, Paleo-redox conditions during OAE
2 refl ected in Demerara Rise sediment geochemistry
(ODP Leg 207): Palaeogeography, Palaeoclimatol-
ogy, Palaeoecology, v. 273, p. 302–328, doi:10.1016
/j.palaeo.2008.11.005.
Higgins, M.B., Robinson, R.S., Carter, S.J., and Pearson,
A., 2010, Evidence from chlorin nitrogen isotopes for
alternating nutrient regimes in the Eastern Mediterra-
nean Sea: Earth and Planetary Science Letters, v. 290,
p. 102–107, doi:10.1016/j.epsl.2009.12.009.
Hinnov, L.A., and Hilgen, F.J., 2012, Cyclostratigraphy
and astrochronology, in Gradstein, F.M., Ogg, J.G.,
and Smith, A.G., eds., A Geologic Time Scale 2012:
Elsevier , p. 63–83.
Holland, S.M., and Patzkowsky, M.E., 2007, Gradient ecol-
ogy of a biotic invasion: Biofacies of the type Cincin-
natian series (Upper Ordovician), Cincinnati, Ohio
region, USA: Palaios, v. 22, p. 392–407, doi:10.2110
/palo.2006.p06-066r.
Holland, S.M., and Patzkowsky, M.E., 2012, Sequence ar-
chitecture of the Bighorn Dolomite, Wyoming, USA:
Transition to the Late Ordovician icehouse: Journal of
Sedimentary Research, v. 82, p. 599–615, doi:10.2110
/jsr.2012.52.
Holmden, C., Creaser, R.A., Muehlenbachs, K., Leslie,
S.A., and Bergström, S.M., 1998, Isotopic evidence
for geochemical decoupling between ancient epeiric
seas and bordering oceans: Implications for secu-
lar curves: Geology, v. 26, p. 567–570, doi:10.1130
/0091-7613(1998)026<0567:IEFGDB>2.3.CO;2.
Holmden, C., Panchuk, K., and Finney, S.C., 2012, Tightly
coupled records of Ca and C isotope changes during
the Hirnantian glaciation in an epeiric sea setting:
Geochimica et Cosmochimica Acta, v. 98, p. 94–106,
doi:10.1016/j.gca.2012.09.017.
Holmden, C., Mitchell, C.E., LaPorte, D.F., Patterson, W.P.,
Melchin, M.J., and Finney, S.C., 2013, Nd isotope rec-
ords of Late Ordovician sea-level change—Implica-
tions for glaciation frequency and global stratigraphic
correlation: Palaeogeography, Palaeoclimatology,
Palaeoecology, doi:10.1016/j.palaeo.2013.05.014 (in
press).
Ingall, E.D., Bustin, R.M., and Van Cappellen, P., 1993, In-
fl uence of water column anoxia on the burial and pres-
ervation of carbon and phosphorus in marine shales:
Geochimica et Cosmochimica Acta, v. 57, p. 303–316,
doi:10.1016/0016-7037(93)90433-W.
Jaeger, H., 1977, Das Silur/Lochkov-Profi l im Franken berger
Zwischengebirge (Sachsen): Freiberger Forschung-
shefte, ser. C, v. 326, p. 45–59.
Jaeger, H., and Robardet, M., 1979, Le Silurien et le
Dévonien basal dans le nord de la province de Seville
(Espagne): Géobios, v. 12, p. 687–714, doi:10.1016
/S0016-6995(79)80097-2.
Jaeger, H., Havlíček, V., and Schönlaub, H.P., 1975, Bio-
stratigraphie der Ordovizium/Silur Grenze in den
Südalpen—Ein Beitrag zur Diskussion um die Hirnantia-
Fauna: Verhandlungen der Geologischen Bundesan stalt,
v. 1975, p. 271–289.
Jenkyns, H.C., 2010, Geochemistry of oceanic anoxic events:
Geochemistry Geophysics Geosystems, v. 11, p. 1–30,
doi:10.1029/2009GC002788.
Jia, T., 2006, Nitrogen isotope fractionations during pro-
gressive metamorphism: A case study from the Paleo-
zoic Cooma metasedimentary complex, southeastern
Australia: Geochimica et Cosmochimica Acta, v. 70,
p. 5201–5214, doi:10.1016/j.gca.2006.08.004.
Jiang, L., Schofi eld, O.M.E., and Falkowski, P.G., 2005,
Adaptive evolution of phytoplankton cell size: Ameri-
can Naturalist, v. 166, p. 496–505.
Jimenéz-Sánchez, A., and Villas, E., 2010, The bryozoan
dispersion into the Mediterranean margin of Gond-
wana during the pre-glacial Late Ordovician: Palaeo-
geography, Palaeoclimatology, Palaeoecology, v. 294,
p. 220–231, doi:10.1016/j.palaeo.2009.11.027.
Jin, J., and Zhan, R., 2008, Late Ordovician Orthide and
Billingsellide Brachiopods from Anticosti Island, East-
ern Canada: Diversity Change through Mass Extinc-
tion: Ottawa, Canada, National Research Council of
Canada Research Press, 151 p.
Johnson, M.E., 2010, Tracking Silurian eustasy: Alignment
of empirical evidence or pursuit of deductive reason-
ing?: Palaeogeography, Palaeoclimatology, Palaeo-
ecology, v. 296, p. 276–284, doi:10.1016/j.palaeo.2009
.11.024.
Jones, D.S., and Fike, D.A., 2013, Dynamic sulfur and car-
bon cycling through the end-Ordovician extinction re-
vealed by paired sulfate-pyrite delta S-34: Earth and
Planetary Science Letters, v. 363, p. 144–155, doi:
10.1016/j.epsl.2012.12.015.
Jones, D.S., Fike, D.A., Finnegan, S., Fischer, W.W., Schrag,
D.P., and McCay, D., 2011, Terminal Ordovician car-
bon isotope stratigraphy and glacioeustatic sea-level
change across Anticosti Island (Quebec, Canada): Geo-
logical Society of America Bulletin, v. 123, p. 1645–
1664, doi:10.1130/B30323.1.
Jones Crafford, A.E., 2008, Paleozoic tectonic domains of
Nevada: An interpretive discussion to accompany the
geologic map of Nevada: Geosphere, v. 4, p. 260–291,
doi:10.1130/GES00108.1.
Junium, C.K., and Arthur, M.A., 2007, Nitrogen cycling dur-
ing the Cretaceous, Cenomanian-Turonian oceanic an-
oxic event II: Geochemistry Geophysics Geosystems,
v. 8, p. 1–18, doi:10.1029/2006GC001328.
Jux, U., and Manze, U., 1979, Glazialeustatisch gesteuerte
Sedimentationsablufeauf dem kaledonischen Schelf
(Mittelschweden) an der WendeOrdovizium-Silur: Neues
Jahrbuch für Geologie und Paläontologie, Monatshefte,
v. 1979, p. 155–180.
Kaljo, D., and Martma, T., 2011, Carbon isotope trend in
the Mirny Creek area, NE Russia, its specifi c features
and possible implications of the uppermost Ordovi-
Melchin et al.
1666 Geological Society of America Bulletin, November/December 2013
cian stratigraphy, in Gutiérrez-Marco, J.C., Rábano, I.,
and García-Bellido, D., eds., Ordovician of the World:
Cuadernos del Museo Geominero, Instituto Geologico
y Minero de España, Madrid, Spain, v. 14, p. 267–273.
Kaljo, D., Kiipli, T., and Martma, T., 1998, Correlation of
carbon isotope events and environmental cyclicity in
the East Baltic Silurian, in Landing, E., and Johnson,
M.E., eds., Silurian Cycles: Linkages of Dynamic
Stratigraphy with Atmospheric, Oceanic, and Tectonic
Changes, James Hall Centennial Volume: Albany, New
York, New York State Museum, p. 297–312.
Kaljo, D., Hints, L., Hints, O., Mannik, P., Martma, T., and
Nōlvak, J., 2011, Katian prelude to the Hirnantian
(Late Ordovician) mass extinction: A Baltic perspec-
tive: Geological Journal, v. 46, p. 464–477.
Kaljo, D., Mannik, P., Martma, T., and Nōlvak, J., 2012,
More about the Ordovician-Silurian transition beds at
Mirny Creek, Omulev Mountains, NE Russia: Carbon
isotopes and conodonts: Estonian Journal of Earth Sci-
ences, v. 61, p. 277–294, doi:10.3176/earth.2012.4.07.
Kanygin, A., Dronov, A., Timokhin, A., and Gonta, T.,
2010, Depositional sequences and palaeoceanographic
change in the Ordovician of the Siberian craton:
Palaeo geography, Palaeoclimatology, Palaeoecology,
v. 296, p. 285–296, doi:10.1016/j.palaeo.2010.02.014.
Kidder, D.L., and Worsley, T.R., 2010, Phanerozoic large ig-
neous provinces (LIPs), HEATT (haline euxinic acidic
thermal transgression) episodes, and mass extinctions:
Palaeogeography, Palaeoclimatology, Palaeoecology,
v. 295, p. 162–191, doi:10.1016/j.palaeo.2010.05.036.
Kiipli, E., and Kiipli, T., 2013, Nitrogen isotopes in kuker-
site and black shale implying Ordovician-Silurian
seawater redox conditions: Oil Shale, v. 30, p. 60–75,
doi:10.3176/oil.2013.1.06.
Klemme, H.D., and Ulmishek, G.F., 1991, Effective petro-
leum source rocks of the world—Stratigraphic distri-
bution and controlling depositional factors: American
Association of Petroleum Geologists Bulletin, v. 75,
p. 1809–1851.
Klug, C., Kroger, B., Kiessling, W., Mullins, G.L., Servais,
T., Frýda, J., Korn, D., and Turner, S., 2010, The De-
vonian nekton revolution: Lethaia, v. 43, p. 465–477,
doi:10.1111/j.1502-3931.2009.00206.x.
Koren’, T.N., and Melchin, M.J., 2000, Lowermost Silurian
graptolites from the Kurama Range, eastern Uzbekistan:
Journal of Paleontology, v. 74, p. 1093–1113, doi:10.1666
/0022-3360(2000)074<1093:LSGFTK>2.0.CO;2.
Koren’, T.N., and Rickards, R.B., 1996, Taxonomy and
Evolution of Llandovery Biserial Graptoloids from the
Southern Urals, Western Kazakhstan: Palaeontological
Association Special Papers in Palaeontology 54, 103 p.
Koren’, T.N., and Rickards, R.B., 2004, An unusually diverse
Llandovery (Silurian) diplograptid fauna from the
Southern Urals of Russia and its evolutionary signifi -
cance: Palaeontology, v. 47, p. 859–918, doi:10.1111
/j.0031-0239.2004.00411.x.
Koren’, T.N., and Sobolevskaya, R.F., 1999, Facies and faunal
diversity across the Ordovician-Silurian boundary in
the Asian part of Russia and adjacent territories: Acta
Universitatis Carolinae: Geologica, v. 43, p. 213–215.
Koren’, T.N., and Sobolevskaya, R.F., 2008, The regional
stratotype section and point for the base of the Hirnan-
tian Stage (the uppermost Ordovician) at Mirny Creek,
Omulev Mountains: Northeast Russia: Estonian Jour-
nal of Earth Sciences, v. 57, p. 1–10, doi:10.3176
/earth.2008.1.01.
Koren’, T.N., Oradovskaya, M.M., Sobolevskaya, R.F.,
Pylma , L.J., and Chugaeva, M.N., 1983, The Ordo-
vi cian and Silurian Boundary in the Northeast of
the USSR: Leningrad, Nauka, Publishers Leningrad
Branch, Interdepartmental Stratigraphic Committee of
USSR, 205 p. (in Russian).
Koren’, T.N., Ahlberg, P., and Nielsen, A.T., 2003a, The
post-persculptus and pre-ascensus graptolite fauna in
Scania, south-western Sweden: Ordovician or Silu-
rian?, in Ortega, G., and Aceolaza, G.F., eds., Proceed-
ings of the 7th International Graptolite Conference &
Field Meeting of the International Subcommission on
Silurian Stratigraphy, Volume 18: Tucumn, Argentina,
Comunicarte Editorial, p. 133–138.
Koren’, T.N., Popov, L.E., and Degtjarev, K.E., Kovalevsky,
O., and Modzalevskay, T., 2003b, Kazakhstan in the
Silu rian, in Landing, E., and Johnson, M.E., eds., Silu-
rian Lands and Seas, Paleogeography Outside of Lauren-
tia: New York State Museum Bulletin 493, p. 323–343.
Kozlu, H., and Ghienne, J.-F., 2012, Ordovician, in
Göncüoglu, M.C., and Bozdogan, N., eds., Paleozoic
of Northern Gondwana and its Petroleum Potential:
A Field Workshop, 09–14 September 2012, Kayseri,
Turkey; “Paleozoic of Eastern Taurides” Guidebook
Volume: Ankara, Turkey, Turkish Association of Pe-
troleum Geologists Special Publication 7, p. 42–60.
Kremer, B., and Kamierczak, J., 2005, Cyanobacterial mats
from Silurian black radiolarian cherts: Phototrophic
life at the edge of darkness?: Journal of Sedimentary
Research, v. 75, p. 897–904, doi:10.2110/jsr.2005.069.
Krug, A.Z., and Patzkowski, M.E., 2007, Geographic varia-
tion in turnover and recovery from the Late Ordovi-
cian mass extinction: Paleobiology, v. 33, p. 435–454,
doi:10.1666/06039.1.
Kump, L.R., and Arthur, M.A., 1999, Interpreting carbon-
isotope excursions: Carbonates and organic matter:
Chemical Geology, v. 161, p. 181–198, doi:10.1016
/S0009-2541(99)00086-8.
Kump, L.R., Arthur, M.A., Patzkowsky, M.E., Gibbs, M.T.,
Pinkus, D.S., and Sheehan, P.M., 1999, A weathering
hypothesis for glaciation at high atmospheric pCO2
during the Late Ordovician: Palaeogeography, Palaeo-
climatology, Palaeoecology, v. 152, p. 173–187, doi:
10.1016/S0031-0182(99)00046-2.
Kumpulainen, R.A., 2007, The Ordovician glaciation in
Eritrea and Ethiopia, NE Africa, in Hambrey, M.J.,
Christoffersen, P., Glasser, N.F., and Hubbard, B., eds.,
Glacial Sedimentary Processes and Products: Interna-
tional Association of Sedimentologists Special Publi-
cation 39, p. 295–319.
Kuypers, M.M.M., van Breugel, Y., Schouten, S., Erba, E.,
and Damste, J.S.S., 2004, N2-fi xing cyanobacteria sup-
plied nutrient N for Cretaceous oceanic anoxic events:
Geology, v. 32, p. 853–856, doi:10.1130/G20458.1.
Lakova, I., and Sačanski, V., 2004, Cryptospores and trilete
spores in oceanic graptolite-bearing sediments (Saltar
Formation) across the Ordovician-Silurian boundary in
the West Balkan Mountains, Bulgaria: Review of the
Bulgarian Geological Society, v. 65, p. 151–156.
Landing, E., 2011, Time-specifi c black mudstones and global
hyperwarming on the Cambrian-Ordovician slope and
shelf of the Laurentia palaeocontinent: Palaeogeogra-
phy, Palaeoclimatology, Palaeoecology, v. 367–368,
p. 256–272, doi:10.1016/j.palaeo.2011.09.005.
LaPorte, D.F., Holmden, C., Patterson, W.P., Loxton, J.D.,
Melchin, M.J., Mitchell, C.E., Finney, S.C., and Sheets,
H.D., 2009, Local and global perspectives on carbon
and nitrogen cycling during the Hirnantian glaciation:
Palaeogeography, Palaeoclimatology, Palaeoecology,
v. 276, p. 182–195, doi:10.1016/j.palaeo.2009.03.009.
Lefebvre, V., Servais, T., Francois, L., and Averbuch, O.,
2010, Did a Katian large igneous province trigger the
Late Ordovician glaciation? A hypothesis tested with a
carbon cycle model: Palaeogeography, Palaeoclimatol-
ogy, Palaeoecology, v. 296, p. 310–319, doi:10.1016
/j.palaeo.2010.04.010.
Leggett, J.K., 1980, British Lower Palaeozoic black shales
and their palaeooceanographic signifi cance: Journal of
the Geological Society of London, v. 137, p. 139–156,
doi:10.1144/gsjgs.137.2.0139.
Legrand, P., 1988, The Ordovician-Silurian boundary in the
Algerian Sahara, in Cocks, L.R.M., and Rickards, R.B.,
eds., A Global Analysis of the Ordovician-Silu rian
Boundary: Bulletin of the British Museum of Natural
History (Geology), v. 43, p. 171–176.
Legrand, P., 1993, Graptolites d’âge ashgillien dans la ré-
gion de Chirfa (Djado, République du Niger): Bulletin
des Centres de Recherches Exploration-Production
Elf-Aquitaine, v. 17, p. 435–442.
Legrand, P., 2000, Une région de référence pour la limite
Ordo vicien-Silurien: l’Oued In Djerane, Sahara algérien:
Comptes Rendus de l’Académie des Sciences, Série 2:
Sciences de la Terre et des Planètes, v. 330, p. 61–66.
Le Heron, D.P., and Dowdeswell, J.A., 2009, Calculating
ice volumes and ice fl ux to constrain the dimensions
of a 440 Ma North African ice sheet: Journal of the
Geological Society of London, v. 166, p. 277–281,
doi:10.1144/0016-76492008-087.
Le Heron, D.P., Craig, J., and Etienne, J.L., 2009, Ancient
glaciations and hydrocarbon accumulations in North
Africa and the Middle East: Earth-Science Reviews,
v. 93, p. 47–76, doi:10.1016/j.earscirev.2009.02.001.
Le Heron, D.P., Armstrong, H.A., Wilson, C., Howard, J.P.,
and Gindre, L., 2010, Glaciation and deglaciation of
the Libyan Desert: The Late Ordovician record: Sedi-
mentary Geology, v. 223, p. 100–125, doi:10.1016
/j.sedgeo.2009.11.002.
Le Heron, D.P., Meinhold, G., Page, A., and Whitham, A.,
2013, Did lingering ice sheets moderate anoxia in the
early Palaeozoic of Libya?: Journal of the Geological
Society of London, v. 170, p. 327–339, doi:10.1144
/jgs2012-108.
Lehnert, O., Eriksson, M.J., Calner, M., Joachimski, M.,
and Buggisch, W., 2007, Concurrent sedimentary
and isotopic indications for global climatic cooling in
the Late Silurian: Acta Palaeontologica Sinica, v. 46,
p. 249–255.
Lehnert, O., Mannik, P., Joachimski, M.M., Calner, M., and
Fryda, J., 2010, Palaeoclimate perturbations before
the Sheinwoodian glaciation: A trigger for extinc-
tions during the ‘Ireviken Event’: Palaeogeography,
Palaeoclimatology, Palaeoecology, v. 296, p. 320–331,
doi:10.1016/j.palaeo.2010.01.009.
Lenton, T.M., Crouch, M., Johnson, M., Pires, N., and Dolan,
L., 2012, First plants cooled the Ordovician: Nature
Geoscience, v. 5, p. 86–89, doi:10.1038/ngeo1390.
Leone, F., Loi, A., Pillola, G.L., and Štorch, P., 2009, The
Late Ordovician (Hirnantian) deposits in the Domus-
novas area, SW Sardinia, in Corradini, C., Ferretti, A.,
and Štorch, P., eds., Silurian of Sardinia: Rendiconti
della Societa Paleontologica Italiana, v. 3, p. 227–237.
Lille, U., 2003, Current knowledge on the origin and struc-
ture of Estonian kukersite kerogen: Oil Shale, v. 20,
p. 253–263.
Logan, G.A., Hayes, J.M., Hieshima, G.B., and Summons,
R.E., 1995, Terminal Proterozoic reorganization of bio-
geochemical cycles: Nature, v. 376, no. 6535, p. 53–56,
doi:10.1038/376053a0.
Loi, A., Ghienne, J.F., Dabard, M.P., Paris, F., Botquelen, A.,
Christ, N., Elaouad-Debbaj, Z., Gorini, A., Vidal, M.,
Videt, B., and Destombes, J., 2010, The Late Ordovi-
cian glacio-eustatic record from a high-latitude storm-
dominated shelf succession: The Bou Ingarf section
(Anti-Atlas, southern Morocco): Palaeogeography,
Palaeoclimatology, Palaeoecology, v. 296, p. 332–358,
doi:10.1016/j.palaeo.2010.01.018.
Loydell, D., 2007, Graptolites from the Upper Ordovician
and Lower Silurian of Jordan: Palaeontological Asso-
ciation Special Papers in Palaeontology 78, 66 p.
Loydell, D.K., 2012, Graptolite biostratigraphy of the E1–
NC174 core, Rhuddanian (lower Llandovery, Silurian),
Murzuq Basin, Libya: Bulletin of Geosciences, v. 87,
doi:10.3140/bullgeosci.1311.
Loydell, D.K., Butcher, A., Fryda, J., Lüning, S., and Fowler,
M., 2009, Lower Silurian “hot shales” in Jordan: A new
depositional model: Journal of Petroleum Geology, v. 32,
p. 261–270, doi:10.1111/j.1747-5457.2009.00447.x.
Lüning, S., Craig, J., Loydell, D.K., Štorch, P., and Fitches,
B., 2000, Lower Silurian ‘hot shales’ in North Africa
and Arabia: Regional distribution and depositional
model: Earth-Science Reviews, v. 49, p. 121–200, doi:
10.1016/S0012-8252(99)00060-4.
Lüning, S., Archer, R., Graig, J., and Loydell, D.K., 2003,
The Lower Silurian hot shales and double hot shales in
North Africa and Arabia, in Salem, M.J., Oun, K.M.,
and Seddiq, H.M., eds., The Geology of Northwest
Libya (Ghadamis, Jifarah, Tarabulus and Sabratah
Basins ), Volume 3: Tripoli, Libya, Earth Science Soci-
ety of Libya, p. 91–105.
Lüning, S., Shahin, Y.M., Loydell, D., Al-Rabi, H.T., Masri,
A., Tarawneh, B., and Kolonic, S., 2005, Anatomy of
a world-class source rock: Distribution and deposi-
tional model of Silurian organic-rich shales in Jordan
and implications for hydrocarbon potential: American
Association of Petroleum Geologists Bulletin, v. 89,
p. 1397–1427, doi:10.1306/05250505014.
Lüning, S., Loydell, D.K., Štorch, P., Shahin, Y., and Craig,
J., 2006, Origin, sequence stratigraphy and deposi-
tional environment of an Upper Ordovician (Hirnan-
tian) deglacial black shale, Jordan—Discussion:
Environmental changes in the Late Ordovician–early Silurian: Review and new insights from black shales and nitrogen isotopes
Geological Society of America Bulletin, November/December 2013 1667
Palaeogeography, Palaeoclimatology, Palaeoecology,
v. 230, p. 352–355, doi:10.1016/j.palaeo.2005.10.004.
Lyons, T.W., Anbar, A.D., Severmann, S., Scott, C., and
Gill, B.C., 2009, Tracking euxinia in the ancient
ocean: A multiproxy perspective and Proterozoic
case study: Annual Review of Earth and Planetary
Sciences, v. 37, p. 507–534, doi:10.1146/annurev
.earth.36.031207.124233.
Marshall, J.D., and Middleton, P.D., 1990, Changes in ma-
rine isotopic composition in the Late Ordovician gla-
ciation: Journal of the Geological Society of London,
v. 147, p. 1–4, doi:10.1144/gsjgs.147.1.0001.
Masiak, M., Podhalanska, T., and Stempien-Salek, M., 2003,
Ordovician-Silurian boundary in the Bardo syncline,
Holy Cross Mountains, Poland—New data on fossil
assemblages and sedimentary succession: Geological
Quarterly, v. 47, p. 311–330.
McCracken, A.D., and Nowlan, G.S., 1988, The Gamachian
Stage and Fauna 13, in Landing, E., New York State
Museum Bulletin, no. 462 (1986); Proceedings of the
Canadian Paleontology and Biostratigraphy Seminar,
Albany, September 26–29 1988, p. 71–79.
McGhee, G.R., Jr., Sheehan, P.M., Bottjer, D.J., and Droser ,
M.L., 2004, Ecological ranking of Phanerozoic bio-
diversity crises: Ecological and taxonomic severities
are decoupled: Palaeogeography, Palaeoclimatology,
Palaeo ecology, v. 211, p. 289–297, doi:10.1016/j.palaeo
.2004.05.010.
McGhee, G.R., Jr., Sheehan, P.M., Bottjer, D.J., and Droser ,
M.L., 2012, Ecological ranking of Phanerozoic bio-
diversity crises: The Serpukhovian (early Carbonif-
erous) crisis had a greater ecological impact than the
end-Ordovician: Geology, v. 40, no. 2, p. 147–150,
doi:10.1130/G32679.1.
Melchin, M.J., 1989, Llandovery graptolite biostratigraphy
and paleobiogeography, Cape Phillips Formation, Ca-
nadian Arctic Islands: Canadian Journal of Earth Sci-
ences, v. 26, p. 1726–1746, doi:10.1139/e89-147.
Melchin, M.J., 2008, Restudy of some Ordovician-Silurian
boundary graptolites from Anticosti Island, Canada,
and their biostratigraphic signifi cance: Lethaia, v. 41,
p. 155–162, doi:10.1111/j.1502-3931.2007.00045.x.
Melchin, M.J., and Holmden, C., 2006a, Carbon isotope
chemostratigraphy in Arctic Canada: Sea-level forc-
ing of carbonate platform weathering and implications
for Hirnantian global correlation: Palaeogeography,
Palaeoclimatology, Palaeoecology, v. 234, p. 186–200,
doi:10.1016/j.palaeo.2005.10.009.
Melchin, M.J., and Holmden, C., 2006b, Carbon iso-
tope chemostratigraphy of the Llandovery in Arctic
Canada: Implications for global correlation and sea-
level change: GFF, v. 128, p. 173–180, doi:10.1080
/11035890601282173.
Melchin, M.J., and Mitchell, C.E., 1991, Late Ordovician
extinction in the Graptoloidea, in Barnes, C.R., and
Williams, S.H., eds., Advances in Ordovician Geology:
Geological Survey of Canada Paper 90–9, p. 143–156.
Melchin, M.J., McCracken, A.D., and Oliff, F.J., 1991,
The Ordovician-Silurian boundary on Cornwallis and
Truro Islands, Arctic Canada: Preliminary data: Cana-
dian Journal of Earth Sciences, v. 28, p. 1854–1862,
doi:10.1139/e91-165.
Melchin, M.J., Holmden, C., and Williams, S.H., 2003, Cor-
relation of graptolite biozones, chitinozoan biozones,
and carbon isotope curves through the Hirnantian, in
Albanesi, G.L., Beresi, M.S., and Peralta, S.H., eds.,
Ordovician from the Andes, Volume 17: Tucumán,
Argen tina, Comunicarte Editorial, p. 101–104.
Melchin, M.J., Mitchell, C.E., Naczk-Cameron, A., Fan,
J.X., and Loxton, J., 2011, Phylogeny and adaptive
radiation of the Neograptina (Graptoloida) during
the Hirnantian mass extinction and Silurian recovery:
Proceedings of the Yorkshire Geological Society, v. 58,
p. 281–309, doi:10.1144/pygs.58.4.301.
Melchin, M.J., Sadler, P.M., and Cramer, B.D., 2012, The
Silurian Period, in Gradstein, F.M., Ogg, J.G., and
Smith, A.G., eds., A Geologic Time Scale 2012: Elsevier ,
p. 525–558.
Melott, A.L., and Thomas, B.C., 2009, Late Ordovician geo-
graphic patterns of extinction compared with simulations
of astrophysical ionizing radiation damage: Paleobiol-
ogy, v. 35, p. 311–320, doi:10.1666/0094-8373-35.3.311.
Melott, A.L., Lieberman, B.S., Laird, C.M., Martin, L.D.,
Medvedev, M.V., Thomas, B.C., Cannizzo, J.K.,
Gehrels , N., and Jackman, C.H., 2004, Did a gamma-
ray burst initiate the Late Ordovician mass extinction?:
International Journal of Astrobiology, v. 3, p. 55–61,
doi:10.1017/S1473550404001910.
Mergl, M., 2011, Earliest occurrence of the Hirnantia fauna
in the Prague Basin (Czech Republic): Bulletin of
Geosciences, v. 86, no. 1, p. 63–70, doi:10.3140/bull
.geosci.1245.
Metzger, J.G., and Fike, D.A., 2013, Techniques for as-
sessing spatial heterogeneity of carbonate 13C values:
Implications for craton-wide isotope gradients: Sedi-
mentology, doi:10.1111/sed.12033.
Meyer, K.M., and Kump, L.R., 2008, Oceanic euxinia in
Earth history: Causes and consequences: Annual Re-
view of Earth and Planetary Sciences, v. 36, p. 251–
288, doi:10.1146/annurev.earth.36.031207.124256.
Meyers, P.A., and Bernasconi, S.M., 2005, Carbon and
nitro gen isotope excursions in mid-Pleistocene
sapropels from the Tyrrhenian Basin: Evidence for
climate-induced increases in microbial primary pro-
duction: Marine Geology, v. 220, p. 41–58, doi:10.1016
/j.margeo.2005.07.003.
Middleton, P.D., Marshall, J.D., and Brenchley, P.J., 1991,
Evidence for isotopic change associated with Late
Ordovician glaciation, from brachiopods and marine
cements of central Sweden, in Barnes, C.R., and Wil-
liams, S.H., eds., Advances in Ordovician Geology:
Geological Survey of Canada Paper 90–9, p. 313–323.
Mihajlovič, M., 1974, Silurski graptoliti istočne Serbije
(Njihov stratigrafski pologaj): Bulletin du Museum
d’Histoire Naturelle, Beograd, ser. A, v. 29, 247 p. (In
Serbian.)
Miller, M.A., and Melvin, J., 2005, Signifi cant new bio-
stratigraphic horizons in the Qusaiba member of the
Silurian Qalibah Formation of central Saudi Arabia,
and their sedimentological expression in a sequence
stratigraphic context: GeoArabia, v. 10, p. 49–92.
Mitchell, C.E., and Bergström, S.M., 1991, New graptolite
and lithostratigraphic evidence from the Cincinnati re-
gion, U.S.A., for the defi nition and correlation of the
base of the Cincinnatian Series (Upper Ordovician),
in Barnes, C.R., and Williams, S.H., eds., Advances
in Ordovician Geology: Geological Survey of Canada,
Paper 90–9, p. 59–77.
Mitchell, C.E., Brussa, E.D., Toro, B.A., and Astini, R.A.,
1998, Late Ordovician graptolites from the Empozada
Formation, Argentine Precordillera, an outer shelf,
cool water, peri-Gondwanan assemblage?, in Gutiér-
rez-Marco, J.C. and Rábano, I., eds, Proceedings Sixth
International Graptolite Conference of the GWG (IPA)
and the SW Iberia Field Meeting 1998 of the IUGS
Subcommission on Silurian Stratigraphy, Madrid,
1998: Instituto Tecnologico Geominero de España Vol-
ume 23, p. 224–226.
Mitchell, C.E., Sheets, H.D., Belscher, K., Finney, S.C.,
Holmden, C., LaPorte, D.F., Melchin, M.J., and Pat-
terson, W.P., 2007, Species abundance changes during
mass extinction and the inverse Signor-Lipps effect:
Apparently abrupt graptolite mass extinction as an arti-
fact of sampling: Acta Palaeontologica Sinica, v. 46,
p. 340–346.
Mitchell, C.E., Štorch, P., Holmden, C, Melchin, M.J., and
Gutiérrez-Marco, J.C., 2011, New stable isotope data
and fossils from the Hirnantian Stage in Bohemia and
Spain: Implications for correlation and paleoclimate,
in Gutiérrez-Marco, J.C., Rábano, I., and García-Bel-
lido, D., eds., Ordovician of the World: Cuadernos del
Museo Geominero, Instituto Geologico y Minero de
España, v. 14, Madrid, Spain, p. 371–378.
Mitchell, C.E., Melchin, M.J., Holmden, C., and Finney,
S.C., 2012, An alternative international correlation of
the Late Ordovician carbon isotope chemostratigraphic
record: Implications for latest Ordovician temperature
history: Geological Society of America Abstracts with
Programs, v. 44, no. 7, p. 124.
Möbius, J., Lahajnar, N., and Emeis, K.C., 2010, Diagenetic
control of nitrogen isotope ratios in Holocene sapropels
and recent sediments from the Eastern Mediterranean
Sea: Biogeosciences, v. 7, p. 3901–3914, doi:10.5194
/bg-7-3901-2010.
Möbius, J., Gaye, B., Lahajnar, N., Bahlmann, E., and Emeis,
K.C., 2011, Infl uence of diagenesis on sedimentary
delta N-15 in the Arabian Sea over the last 130 kyr:
Marine Geology, v. 284, p. 127–138, doi:10.1016
/j.margeo.2011.03.013.
Monod, O., Kozlu, H., Ghienne, J.-F., Dean, W.T., Gnay,
Y., Le Herissé, A., Paris, A., and Robardet, M.,
2003, Late Ordovician glaciation in southern Turkey:
Terra Nova, v. 15, p. 249–257, doi:10.1046/j.1365
-3121.2003.00495.x.
Montenegro, A., Spence, P., Meissner, K.J., Eby, M.,
Melchin, M.J., and Johnston, S.T., 2011, Climate simu-
lations of the Permian-Triassic boundary: Ocean acidi-
fi cation and the extinction event: Paleoceanography,
v. 26, PA3207, doi:10.1029/2010PA002058.
Moreau, J., 2011, The Late Ordovician deglaciation sequence
of the SW Murzuq Basin (Libya): Basin Research, v. 23,
p. 449–477, doi:10.1111/j.1365-2117.2010.00499.x.
Mu, E., and Ni, Y., 1983, Uppermost Ordovician and low-
ermost Silurian graptolites from the Xainza area of
Xizang (Tibet) with discussion on the Ordovician-
Silurian boundary: Palaeontologica Cathayana, v. 1,
p. 155–179.
Munnecke, A., Calner, M., Harper, D.A.T., and Servais, T.,
2010, Ordovician and Silurian sea-water chemistry,
sea level, and climate: A synopsis: Palaeogeography,
Palaeoclimatology, Palaeoecology, v. 296, p. 389–413,
doi:10.1016/j.palaeo.2010.08.001.
Nardin, E., Godderis, Y., Donnadieu, Y., Le Hir, G., Blakey,
R.C., Puceat, E., and Aretz, M., 2011, Modeling the
early Paleozoic long-term climatic trend: Geological
Society of America Bulletin, v. 123, no. 5–6, p. 1181–
1192, doi:10.1130/B30364.1.
Needham, D.T., 2004, Deformation in Moffat Shale detach-
ment zones in the western part of the Scottish Southern
Uplands: Geological Magazine, v. 141, p. 441–453,
doi:10.1017/S0016756804009203.
Negri, A., Ferretti, A., Wagner, T., and Meyers, P.A., 2009,
Phanerozoic organic-carbon–rich marine sediments:
Overview and future research challenges: Palaeo-
geography, Palaeoclimatology, Palaeoecology, v. 273,
p. 218–227, doi:10.1016/j.palaeo.2008.10.002.
Newell, N.D., 1963, Crises in the history of life: Sci-
entific American, v. 208, p. 76–92, doi:10.1038
/scientifi camerican0263-76.
Newell, N.D., 1967, Revolutions in the history of life, in
Albritton , C., ed., Uniformity and Simplicity: Geologi-
cal Society of America Special Paper 89, p. 63–91.
Oliver, G.J.H., 2001, Reconstruction of the Grampian
episode in Scotland: Its place in the Caledonian orog-
eny: Tectonophysics, v. 332, p. 23–49, doi:10.1016
/S0040-1951(00)00248-1.
Orchard, M.J., 2007, A proposed Carnian-Norian boundary
GSSP at Black Bear Ridge, northeast British Colum-
bia, and a new conodont framework for the boundary
interval: Albertiana, v. 36, p. 130–141.
Orth, C.J., Gilmore, L.R., Quintana, L.R., and Sheehan, P.M.,
1986, Terminal Ordovician extinction: Geochemical
analysis of the Ordovician-Silurian boundary, Anticosti
Island, Quebec: Geology, v. 14, p. 433–436, doi:10.1130
/0091-7613(1986)14<433:TOEGAO>2.0.CO;2.
Ozaki, K., Tajima, S., and Tajika, E., 2011, Conditions re-
quired for oceanic anoxia/euxinia: Constraints from a
one-dimensional ocean biogeochemical cycle model:
Earth and Planetary Science Letters, v. 304, p. 270–
279, doi:10.1016/j.epsl.2011.02.011.
Page, A.A., Zalasiewicz, J.A., Williams, M., and Popov,
L.E., 2007, Were transgressive black shales a negative
feedback modulating glacioeustasy in the early Palaeo-
zoic icehouse?, in Williams, M., Haywood, A.M.,
Gregory, F.J., and Schmidt, D.N., eds., Deep-Time
Perspectives on Climate Change: Marrying the Signal
from Computer Models and Biological Proxies: The
Micropalaeontological Society Special Publications:
London, Geological Society of London, p. 123–156.
Palastanga, V., Slomp, C.P., and Heinze, C., 2011, Long-
term controls on ocean phosphorus and oxygen in a
global biogeochemical model: Global Biogeochemical
Cycles, v. 25, 2011, doi:10.1029/2010GB003827.
Pancost, R.D., Freeman, K.H., Herrmann, A.D., Patzkowsky,
M.E., Ainsaar, L., and Martma, T., 2013, Reconstructing
Late Ordovician carbon cycle variations: Geochimica
Melchin et al.
1668 Geological Society of America Bulletin, November/December 2013
et Cosmochimica Acta, v. 105, p. 433–454, doi:10.1016
/j.gca.2012.11.033.
Paris, F., Deynoux, M., and Ghienne, J.-F., 1998, Décou-
verte de Chitinozoaires la limite Ordovicien-Silurien en
Mauritanie; implications paléogéographiques: Comptes
Rendus de l’Academie des Sciences de Paris, v. 326,
p. 499–504.
Paris, G., Beaumont, V., Bartolini, A., Clemence, M.E., Gar-
din, S., and Page, K., 2010, Nitrogen isotope record
of a perturbed paleoecosystem in the aftermath of the
end-Triassic crisis, Doniford section, SW England:
Geochemistry Geophysics Geosystems, v. 11, p. 1–15,
doi:10.1029/2010GC003161.
Parrish, J.T., 1982, Upwelling and petroleum source with
reference to the Palaeozoic: American Association of
Petroleum Geologists Bulletin, v. 66, p. 750–774.
Pedersen, G.K., 1989, The sedimentology of Lower Palaeo-
zoic black shales from the shallow wells Skelbro 1 and
Billegrav 1, Bornholm, Denmark: Bulletin of the Geo-
logical Society of Denmark, v. 37, p. 151–173.
Phillips, E.R., Evans, J.A., Stone, P., Horstwood, M.S.A.,
Floyd, J.D., Smith, R.A., Akhurst, M.C., and Barron,
H.F., 2003, Detrital Avalonian zircons in the Lauren-
tian Southern Uplands terrane, Scotland: Geol ogy,
v. 31, p. 625–628, doi:10.1130/0091-7613(2003)031
<0625:DAZITL>2.0.CO;2.
Piçarra, J.M., Štorch, P., Gutiérrez-Marco, J.C., and Oliveira
J.T., 1995, Characterization of the Parakidograptus
acuminatus graptolite biozone in the Silurian of the
Barrancos region (Ossa Morena zone, south Portugal):
Comunicações do Instituto Geológico e Mineiro, v. 81,
p. 3–8.
Piçarra, J.M., Robardet, M., Bourahrouh, A., Paris, F.,
Pereira, Z., Le Menn, J., Gourvennec, R., Oliveira, T.,
and Lardeux, H., 2002, Le passage Ordovicien-Silurien
et la partie inferieure du Silurien (Sud-Est du Massif ar-
moricain, France): Comptes Rendus de l’Academie des
Sciences de Paris, v. 334, p. 1177–1183, doi:10.1016
/S1631-0713(02)01864-3.
Piçarra, J.M., Robardet, M., Oliveira, J.T., Paris, F., and Lar-
deux, H., 2009, Graptolite faunas of the Llandovery
phtanites at Les Fresnaies (Chalonnes-sur-Loire,
southeastern Armorican Massif): Palaeontology and
biostratigraphy: Bulletin of Geosciences, v. 84, p. 41–
50, doi:10.3140/bull.geosci.1085.
Podhalanska, T., 1999, The Upper Ordovician and the Lower
Silurian in the Peribaltic Depression; stratigraphy and
development: Acta Universitatis Carolinae: Geologica,
v. 43, p. 221–224.
Podhalanska, T., 2003, Late Ordovician to Early Silurian
transition and the graptolites from Ordovician/Silurian
boundary near the SW rim of the East European craton
(northern Poland), in Ortega, G., and Aceñolaza, G.F.,
eds., Proceedings of the 7th International Graptolite Con-
ference & Field Meeting of the International Subcom-
mission on Silurian Stratigraphy, Volume 18: Tucumán,
Argentina, Comunicarte Editorial, p. 165–171.
Podhalanska, T., and Trela, W., 2007, Stratigraphy and sedi-
mentary record of the Lower Silurian succession in the
southern Holy Cross Mountains, Poland: Acta Palaeon-
tologica Sinica, v. 46, p. 397–401.
Pope, M.C., 2004, Cherty carbonate facies of the Montoya
Group, southern New Mexico and western Texas, and
its regional correlatives: A record of Late Ordovician
paleoceanography on southern Laurentia: Palaeo-
geography, Palaeoclimatology, Palaeoecology, v. 210,
p. 367–384, doi:10.1016/j.palaeo.2004.02.035.
Pope, M.C., and Steffen, J.B., 2003, Widespread, prolonged
late Middle to Late Ordovician upwelling in North
America: A proxy record of glaciation?: Geology, v. 31,
p. 63–66, doi:10.1130/0091-7613(2003)031<0063:
WPLMTL>2.0.CO;2.
Popov, L.E., Bassett, M.G., Zhemchuzhnikov, V.G., Holmer,
L.E., and Klishevich, I.A., 2009, Gondwanan faunal
signatures from early Palaeozoic terranes of Kazakhstan
and Central Asia: Evidence and tectonic implications,
in Bassett, M.G., ed., Early Palaeozoic Peri-Gondwana
Terranes: New Insights from Tectonics and Biogeogra-
phy: Geological Society of London Special Publication
325, p. 23–64.
Potter, P.E., Maynard, J.B., and Depetris, P.J., 2005, Mud
and Mudstones: Berlin, Springer, 314 p.
Poulton, S.W., and Canfi eld, D.E., 2011, Ferruginous condi-
tions: A dominant feature of the ocean through Earth’s
history: Elements, v. 7, p. 107–112.
Poussart, P.F., Weaver, A.J., and Barnes, C.R., 1999, Late
Ordovician glaciation under high atmospheric CO2:
A coupled model analysis: Paleoceanography, v. 14,
p. 542–558.
Puchkov, V.N., 2009, The evolution of the Uralian orog-
eny, in Murphy, J.B., Keppie, J.D., and Hynes, A.J.,
eds., Ancient Orogens and Modern Analogues: Geo-
logical Society of London Special Publication 327,
p. 161–195.
Puchkov, V.N., 2010, Geology of the Urals and Cis-Urals
(Actual Problems of Stratigraphy, Tectonics, Geo-
dynamics and Metallogeny): Ufa, DesignPoligraph-
Service, 288 p. (in Russian).
Rasmussen, C.M.O., and Harper, D.A.T., 2011a, Did the
amalgamation of continents drive the end Ordovician
mass extinctions?: Palaeogeography, Palaeoclimatol-
ogy, Palaeoecology, v. 311, p. 48–62, doi:10.1016
/j.palaeo.2011.07.029.
Rasmussen, C.M.O., and Harper, D.A.T., 2011b, Interroga-
tion of distributional data for the end Ordovician crisis
interval: Where did disaster strike?: Geological Jour-
nal, v. 46, p. 478–500.
Raup, D.M., and Sepkoski, J.J., Jr., 1982, Mass extinctions
in the marine fossil record: Science, v. 215, p. 1501–
1503, doi:10.1126/science.215.4539.1501.
Raup, D.M., and Sepkoski, J.J., Jr., 1986, Periodic extinction
of families and genera: Science, v. 231, p. 833–836.
Rhoads, D. S., and Morse, J. W., 1971, Evolutionary and
ecological signifi cance of oxygen-defi cient marine ba-
sins: Lethaia, v. 4, p. 413–428.
Rickards, R.B., 1988, Base of the Silurian in the Lake Dis-
trict and Howgill Fells, Northern England: Bulletin of
the British Museum (Natural History) Historical Se-
ries, v. 43, p. 53–57.
Robardet, M., and Doré, F., 1988, The Late Ordovician
diamictic formations from southwestern Europe: North
Gondwana glaciomarine deposits: Palaeogeography,
Palaeoclimatology, Palaeoecology, v. 66, p. 19–31, doi:
10.1016/0031-0182(88)90078-8.
Robinson, R.S., Kienast, M., Albuquerque, A.L., Altabet, M.,
Contreras, S., Holz, R.D., Dubois, N., Francois, R., Gal-
braith, E., Hsu, T.C., Ivanochko, T., Jaccard, S., Kao,
S.J., Kiefer, T., Kienast, S., Lehmann, M.F., Martinez,
P., McCarthy, M., Möbius, J., Pedersen, T., Quan, T.M.,
Ryabenko, E., Schmittner, A., Schneider, R., Schneider-
Mor, A., Shigemitsu, M., Sinclair, D., Somes, C., Studer,
A., Thunell, R., and Yang, J.Y., 2012, A review of nitro-
gen isotopic alteration in marine sediments: Paleocean-
ography, v. 27, PA4203, doi:10.1029/2012PA002321.
Rohrssen, M., Love, G.D., Fischer, W., Finnegan, S., and
Fike, D.A., 2012, Lipid biomarkers record fundamen-
tal changes in the microbial community structure of
tropical seas during the Late Ordovician Hirnantian
glaciation: Geology, v. 41, p. 127–130, doi:10.1130
/G33671.1.
Rong, J.Y., and Chen, X., 1987, Faunal differentiation, bio-
facies and lithofacies patterns of Late Ordovician (Ash-
gillian) in southern China: Acta Palaeontologica Sinica,
v. 26, p. 507–535.
Rong, J.Y., Boucot, A.J., Harper, D.A.T., Zhan, R.B., and
Neuman, R.B., 2006, Global analyses of brachiopod
faunas through the Ordovician and Silurian transi-
tion: Reducing the role of the Lazarus effect: Canadian
Journal of Earth Sciences, v. 43, p. 23–39, doi:10.1139
/e05-089.
Rosenau, N.A., Herrmann, A.D., and Leslie, S.A., 2012,
Conodont apatite δ18O values from a platform margin
setting, Oklahoma, USA: Implications for initiation of
Late Ordovician icehouse conditions: Palaeogeography,
Palaeoclimatology, Palaeoecology, v. 315, p. 172–180,
doi:10.1016/j.palaeo.2011.12.003.
Rothwell Group, LP., 2011, PaleoGIS/Arcview 4.0, PALEO-
MAP Project: Arlington, Texas, University of Texas at
Arlington.
Sačanski, V.V., 1993, Boundaries of the Silurian System in
Bulgaria defi ned by graptolites: Geologica Balcanica,
v. 23, p. 25–33.
Sačanski, V.V., 1994, Age assessment of the Cerecel and Sir-
man formations in Sofi a Stara Planina Mountain (Ordo-
vician, Ashgill): Review of the Bulgarian Geological
Society, v. 55, p. 83–90.
Sačanski, V.V., and Tenčov, J., 1993, Lithostratigraphic sub-
division of the Silurian deposits in the Svoge anticline:
Review of the Bulgarian Geological Society, v. 54,
p. 71–81.
Sadler, P.M., Cooper, R.A., and Melchin, M.J., 2011, Se-
quencing the graptoloid clade: Building a global diver-
sity curve from local range charts, regional composites
and global time-lines: Proceedings of the Yorkshire
Geological Society, v. 58, p. 329–343, doi:10.1144
/pygs.58.4.296.
Saltzman, M.R., 2005, Phosphorus, nitrogen, and the redox
evolution of the Paleozoic oceans: Geology, v. 33,
p. 573–576, doi:10.1130/G21535.1.
Saltzman, M.R., and Young, S.A., 2005, Long-lived glacia-
tion in the Late Ordovician? Isotopic and sequence-
stratigraphic evidence from western Laurentia: Geology,
v. 33, p. 109–112, doi:10.1130/G21219.1.
Sawaki, Y., Shibuya, T., Kawai, T., Komiya, T., Omori, S.,
Iizuka, T., Hirata, T., Windley, B.F., and Maruyama, S.,
2010, Imbricated ocean-plate stratigraphy and U-Pb
zircon ages from tuff beds in cherts in the Ballantrae
complex, SW Scotland: Geological Society of America
Bulletin, v. 122, p. 454–464, doi:10.1130/B26329.1.
Schauer, M., 1971, Biostratigraphie und Taxionomie der
Graptolithen des tieferen Silurs unter besonderer
Berücksichtigung der tektonischen Deformation: Frei-
berger Forschungshefte, ser. C, v. 273, p. 1–185.
Schlanger, S.O., and Jenkyns, H.C., 1976, Cretaceous oce-
anic anoxic events: Causes and consequences: Geolo-
gie en Mijnbouw, v. 55, p. 179–184.
Schönian, F., and Egenhoff, S.O., 2007, A Late Ordovi-
cian ice sheet in South America: Evidence from the
Cancairi tillites, southern Bolivia, in Linnemann, U.,
Nance, R.D., Kraft, P., and Zulauf, G., eds., The Evolu-
tion of the Rheic Ocean: Geological Society of Amer-
ica Special Paper 423, p. 525–548.
Schönlaub, H.P., Ferretti, A., Gaggero, L., Hammarlund, E.,
Harper, D.A.T., Histon, K., Priewalder, H., Spötl, C.,
and Štorch, P., 2011, The Late Ordovician glacial event
in the Carnic Alps (Austria), in Gutiérrez-Marco, J.C.,
Rábano, I., and García-Bellido, D., eds., Ordovician of
the World: Cuadernos del Museo Geominero, Instituto
Geologico y Minero de España: Madrid, Spain, v. 14,
p. 515–526.
Schopf, T.J.M., 1974, Permo-Triassic extinctions: Relation
to sea-fl oor spreading: The Journal of Geology, v. 82,
p. 129–143, doi:10.1086/627955.
Sell, B.K., 2011, Intense volcanism and Ordovician ice-
house climate, in Gutiérrez-Marco, J.C., Rábano, I.,
and García-Bellido, D., eds., Ordovician of the World:
Cuadernos del Museo Geominero, Instituto Geologico
y Minero de España, Madrid, Spain, v. 14, p. 527–536.
Sennikov, N.V., Yolkin, E.A., Petrunina, Z.E., Gladkikh,
L.A., Obut, O.T., Izokh, N.G., and Kipriyanova, T.P.,
2008, Ordovician-Silurian Biostratigraphy and Paleo-
geography of the Gorny Altay: Novosibirsk, Russia,
Publishing House SB RAS, 156 p.
Sephton, M.A., Amor, K., Franchi, I.A., Wignall, P.B., New-
ton, R., and Zonneveld, J.P., 2002, Carbon and nitro-
gen isotope disturbances and an end-Norian (Late
Triassic) extinction event: Geology, v. 30, p. 1119–1122,
doi:10.1130/0091-7613(2002)030<1119:CANIDA>2.0
.CO;2.
Sepkoski, J.J., Jr., 1986, Global bioevents and the question
of periodicity, in Walliser, O., ed., Global Bio-Events,
Volume 8: Berlin, Springer-Verlag, p. 47–61.
Severmann, S., and Anbar, A.D., 2009, Reconstructing
paleo redox conditions through a multitracer approach:
The key to the past is the present: Elements, v. 5,
p. 359–364, doi:10.2113/gselements.5.6.359.
Sheehan, P.M., 1973, The relation of the Late Ordovician
glaciation to the Ordovician-Silurian changeover in
North American brachiopod faunas: Lethaia, v. 6,
p. 146–154, doi:10.1111/j.1502-3931.1973.tb01188.x.
Sheehan, P.M., 1975, Brachiopod synecology in a time of
crisis (Late Ordovician–Early Silurian): Paleobiology,
v. 1, p. 205–212.
Sheehan, P.M., 1979, Swedish Late Ordovician marine
benthic assemblages and their bearing on brachiopod
zoogeography, in Gray, J., and Boucot, A.J., eds., His-
Environmental changes in the Late Ordovician–early Silurian: Review and new insights from black shales and nitrogen isotopes
Geological Society of America Bulletin, November/December 2013 1669
torical Biogeography, Plate Tectonics, and the Chang-
ing Environment: Corvallis, Oregon, Oregon State
University Press, p. 61–73.
Sheehan, P.M., 2001, The Late Ordovician mass extinction:
Annual Review of Earth and Planetary Sciences, v. 29,
p. 331–364, doi:10.1146/annurev.earth.29.1.331.
Shen, Y.N., Canfi eld, D.E., and Knoll, A.H., 2002, Middle
Proterozoic ocean chemistry: Evidence from the
McArthur Basin, northern Australia: American Jour-
nal of Science, v. 302, p. 81–109, doi:10.2475/ajs
.302.2.81.
Shields, G.A., Carden, G.A.F., Veizer, J., Meidla, T., Rong,
J.Y., and Li, R.Y., 2003, Sr, C, and O isotope geochem-
istry of Ordovician brachiopods: A major isotopic
event around the Middle-Late Ordovician transition:
Geochimica et Cosmochimica Acta, v. 67, p. 2005–
2025, doi:10.1016/S0016-7037(02)01116-X.
Siddall, M., Kaplan, M.R., Schaefer, J.M., Putnam, A.,
Kelly, M.A., and Goehring, B., 2010, Changing infl u-
ence of Antarctic and Greenlandic temperature records
on sea-level over the last glacial cycle: Quaternary
Science Reviews, v. 29, p. 410–423, doi:10.1016
/j.quascirev.2009.11.007.
Simberloff, D.S., 1974, Permo-Triassic extinctions: Effects
of area on biotic equilibrium: The Journal of Geology,
v. 82, p. 267–274, doi:10.1086/627962.
Sinninghe Damsté, J.S., and Köster, J., 1998, A euxinic
southern North Atlantic Ocean during the Cenoma-
nian/Turonian oceanic anoxic event: Earth and Plan-
etary Science Letters, v. 158, p. 165–173, doi:10.1016
/S0012-821X(98)00052-1.
Skevington, D., 1974, Controls infl uencing the composition
and distribution of Ordovician graptolite faunal prov-
inces, in Rickards, R.B., Jackson, D.E., and Hughes,
C.P., eds., Graptolite Studies in Honour of O.M.B.
Bulman: London, The Palaeontological Association,
p. 59–73.
Slomp, C.P., and Van Cappellen, P., 2007, The global ma-
rine phosphorus cycle: Sensitivity to oceanic circula-
tion: Biogeosciences, v. 4, p. 155–171, doi:10.5194
/bg-4-155-2007.
Sobolevskaya, R.F., 2011, Atlas of the Palaeozoic Fauna—
Taimyr. Part II. The Ordovician-Silurian Graptolites:
FGUP, I.S. Gramberg VNIIOkeangeologia, 283 p. (in
Russian).
Stanley, S.M., 2010, Relation of Phanerozoic stable isotope
excursions to climate, bacterial metabolism, and ma-
jor extinctions: Proceedings of the National Academy
of Sciences of the United States of America, v. 107,
p. 19,185–19,189, doi:10.1073/pnas.1012833107.
Stanley, S.M., and Powell, M.G., 2003, Depressed rates of
origination and extinction during the late Paleozoic
ice age: A new state for the global marine ecosystem:
Geology, v. 31, p. 877–880, doi:10.1130/G19654R.1.
Stein, V., 1965, Stratigraphische und palontologische Unter-
suchungen im Silur des Frankenwaldes: Neues Jahr-
buch für Geologie und Paläontologie: Abhandlungen,
v. 121, p. 111–200.
Stone, P., 1995, Geology of the Rhins of Galloway district:
Memoir of the British Geological Survey, sheets 1 and
3 (Keyworth, England), 102 p.
Stone, P., and Merriman, R.J., 2004, Basin thermal his-
tory favours an accretionary origin for the Southern
Uplands terrane, Scottish Caledonides: Journal of the
Geological Society of London, v. 161, p. 829–836,
doi:10.1144/0016-764903-170.
Stone, P., Rigby, S., and Rushton, A.W.A., 2003, Advances
in Scottish graptolite biostratigraphy: An introduction:
Scottish Journal of Geology, v. 39, p. 11–15.
Štorch, P., 1990, Upper Ordovician–Lower Silurian se-
quences of the Bohemian Massif, central Europe: Geo-
logical Magazine, v. 127, p. 225–239.
Štorch, P., 2006, Facies development, depositional settings
and sequence stratigraphy across the Ordovician-Silu-
rian boundary: A new perspective from the Barrandian
area of the Czech Republic: Geological Journal, v. 41,
p. 163–192.
Štorch, P., and Feist, R., 2008, Lowermost Silurian grapto-
lites of Montagne Noire, France: Journal of Paleontol-
ogy, v. 82, p. 938–956.
Štorch, P., and Mergl, M., 1989, Krlodvor/Kosov boundary
and the Late Ordovician environmental changes in the
Prague Basin (Barrandian, Bohemia): Sborník geo-
logických věd, Geologie, v. 44, p. 117–153.
Štorch, P., and Schönlaub, H.-P., 2012, Ordovician-Silurian
boundary graptolites of the southern Alps, Austria:
Bulletin of Geosciences. doi:10.3140/bull.geosci.
Štorch, P., and Serpagli, E., 1993, Lower Silurian graptolites
from southwestern Sardinia: Bolletino della Societa
Paleontologia Italiana, v. 32, p. 3–57.
Štorch, P., Gutiérrez-Marco, J.C., Sarmiento, G.N., and
Rábano, I., 1998, Upper Ordovician and Lower Silu-
rian of Corral de Calatrava, southern part of the Central
Iberian zone, in Gutiérrez-Marco, J.C., and Rábano, I.,
eds., Proceedings of the Sixth International Graptolite
Conference of the GWG (IPA) and the SW Iberia Field
Meeting 1998 of the International Subcommission
on Silurian Stratigraphy (ICS-IUGS): Madrid, Temas
Geológico-Mineros ITGE, v. 23, p. 319–325.
Štorch, P., Mitchell, C.E., Finney, S.C., and Melchin, M.J.,
2011, Uppermost Ordovician (upper Katian-Hirnan-
tian) graptolites of north-central Nevada, USA: Bul-
letin of Geosciences, v. 86, p. 301–386.
Strachan, R.A., 2012, Mid-Ordovician to Silurian subduction
and collision: Closure of the Iapetus Ocean, in Wood-
cock, N.H., and Strachan, R.A., eds., Geological History
of Britain and Ireland (2nd ed.): Wiley-Blackwell, 423 p.
Strauss, H., 2006, Anoxia through Time, in Neretin, L.N.,
ed., Past and Present Water Column Anoxia: NATO
Science Series, IV: Earth and Environmental Sciences,
Volume 64, p. 3–19.
Su, W.B., Huff, W.D., Ettensohn, F.R., Liu, X.M., Zhang,
J.E., and Li, Z.M., 2009, K-bentonite, black-shale and
fl ysch successions at the Ordovician-Silurian transi-
tion, South China: Possible sedimentary responses to
the accretion of Cathaysia to the Yangtze block and its
implications for the evolution of Gondwana: Gond-
wana Research, v. 15, p. 111–130, doi:10.1016/j.gr
.2008.06.004.
Sutcliffe, O.E., Dowdeswell, J.A., Whittington, R.J., Theron,
J.N., and Craig, J., 2000, Calibrating the Late Ordovi-
cian glaciation and mass extinction by the eccentric-
ity cycles of Earth’s orbit: Geology, v. 28, p. 967–970,
doi:10.1130/0091-7613(2000)28<967:CTLOGA>2.0
.CO;2.
Sweet, W.C., 2000, Conodonts and biostratigraphy of Upper
Ordovician strata along a shelf to basin transect
in central Nevada: Journal of Paleontology, v. 74,
p. 1148–1160, doi:10.1666/0022-3360(2000)074<1148:
CABOUO>2.0.CO;2.
Tasáryová, Z., Pruner, P., Manda, Janoušek, V., Schnabl, P.,
Štorch, P., Frýda, J., Šifnerová, K., and Erban, V., 2012,
Perunica microplate in Silurian period: Implications
from basalt geochemistry, palaeomagnetism and faunas
(Prague Basin, Tepl-Barrandian Unit, Bohemian Mas-
sif), In BRGM & SGF: Géologie de la France, v. 2012,
p. 213–214.
Tesakov, Y.I., Predtetchenskyj, N.N., Khromych, V.G.,
Berger, A.Y., and Kovalevskaya, E.O., 2003, Stratigra-
phy and paleogeography of the Silurian of East Siberia,
in Landing, E., and Johnson, M.E., eds., Silurian Lands
and Shelf Margins Exclusive of North America: New
York State Museum Bulletin 492, p. 345–400.
Thompson, C.K., and Kah, L.C., 2012, Sulfur isotope evi-
dence for widespread euxinia and a fl uctuating oxy-
cline in Early to Middle Ordovician greenhouse oceans:
Palaeogeography, Palaeoclimatology, Palaeoecology,
v. 313, p. 189–214, doi:10.1016/j.palaeo.2011.10.020.
Tomczyk, H., 1963, Ordowik i sylur w podlou Zapadliska
Przedkarpackiego: Rocznik Polskiego Towarzystwa
Geologicznego, v. 33, p. 289–320.
Trabucho-Alexandre, J., Hay, W.W., and de Boer, P.L., 2012,
Phanerozoic environments of black shale deposition
and the Wilson cycle: Solid Earth, v. 3, p. 29–42, doi:
10.5194/se-3-29-2012.
Trotter, J.A., Williams, I.S., Barnes, C.R., Lecuyer, C., and
Nicoll, R.S., 2008, Did cooling oceans trigger Ordo-
vician biodiversification? Evidence from conodont
thermometry: Science, v. 321, p. 550–554, doi:10.1126
/science.1155814.
Tsandev, I., and Slomp, C.P., 2009, Modeling phosphorus
cycling and carbon burial during Cretaceous oceanic
anoxic events: Earth and Planetary Science Letters,
v. 286, p. 71–79, doi:10.1016/j.epsl.2009.06.016.
Tsandev, I., Slomp, C.P., and Van Cappellen, P., 2008, Glacial-
interglacial variations in marine phosphorus cycling:
Implications for ocean productivity: Global Biogeo-
chemical Cycles, v. 22, doi:10.1029/2007GB003054.
Tsandev, I., Rabouille, C., Slomp, C.P., and Van Cappellen,
P., 2010, Shelf erosion and submarine river canyons:
Implications for deep-sea oxygenation and ocean
productivity during glaciation: Biogeosciences, v. 7,
p. 1973–1982, doi:10.5194/bg-7-1973-2010.
Underwood, C.J., Crowley, S.F., Marshall, J.D., and Brenchley,
P.J., 1997, High-resolution carbon isotope stratig raphy
of the basal Silurian stratotype (Dob’s Linn, Scotland)
and its global correlation: Journal of the Geological
Society of London, v. 154, p. 709–718, doi:10.1144
/gsjgs.154.4.0709.
Underwood, C.J., Deynoux, M., and Ghienne, J.F., 1998,
High palaeolatitude (Hodh, Mauritania) recovery of
graptolite faunas after the Hirnantian (end Ordovician)
extinction event: Palaeogeography, Palaeoclimatology,
Palaeoecology, v. 142, p. 91–105, doi:10.1016/S0031
-0182(98)00070-4.
Van Cappellen, P., and Ingall, E.D., 1994, Benthic phosphorus
regeneration, net primary production, and ocean anoxia:
A model of the coupled marine biogeochemical cycles of
carbon and phosphorus: Paleoceanography, v. 9, p. 677–
692, doi:10.1029/94PA01455.
Vandenberg, A.H.M., Rickards, R.B., and Holloway, D.J.,
1984, The Ordovician-Silurian boundary at Darraweit
Guim, central Victoria: Alcheringa, v. 8, p. 1–22, doi:
10.1080/03115518408619607.
Vandenbroucke, T.R.A., Armstrong, H.A., Williams, M.,
Paris, F., Zalasiewicz, J.A., Sabbe, K., Nōlvak, J., Chal-
lands, T.J., Verniers, J., and Servais, T., 2010, Polar
front shift and atmospheric CO2 during the glacial maxi-
mum of the early Paleozoic icehouse: Proceedings of
the National Academy of Sciences of the United States
of America, v. 107, p. 14,983–14,986, doi:10.1073
/pnas.1003220107.
van Staal, C.R., and Hatcher, R.D., Jr., 2010, Global setting
of Ordovician orogenesis, in Finney, S.C., and Berry,
W.B.N., eds., The Ordovician Earth System: Geologi-
cal Society of America Special Paper 466, p. 1–11.
van Staal, C.R., Vujovich, G.I., Currie, K.L., and Naipauer,
M., 2011, An Alpine-style Ordovician collision com-
plex in the Sierra de Pie de Palo, Argentina: Record
of subduction of Cuyania beneath the Famatina arc:
Journal of Structural Geology, v. 33, p. 343–361,
doi:10.1016/j.jsg.2010.10.011.
Vaslet, D., 1990, Upper Ordovician glacial deposits in Saudi
Arabia: Episodes, v. 13, p. 147–167.
Vecoli, M., 2008, Fossil microphytoplankton dynamics across
the Ordovician-Silurian boundary: Review of Palaeo-
botany and Palynology, v. 148, p. 91–107, doi:10.1016
/j.revpalbo.2006.11.004.
Vecoli, M., Delabroye, A., Spina, A., and Hints, O., 2011,
Cryptospore assemblages from Upper Ordovician
(Katian-Hirnantian) strata of Anticosti Island, Que-
bec, Canada, and Estonia: Palaeophytogeographic
and palaeo climatic implications: Review of Palaeo-
botany and Palynology, v. 166, p. 76–93, doi:10.1016
/j.revpalbo.2011.05.006.
Videt, B., Paris, F., Rubino, J.-L., Boumendjel, K., Dabard ,
M.-P., Loi, A., Ghienne, J.-F., Marante, A., and Gorini,
A., 2010, Biostratigraphical calibration of third order
Ordovician sequences on the northern Gondwana plat-
form: Palaeogeography, Palaeoclimatology, Palaeo-
ecology, v. 296, p. 359–375, doi:10.1016/j.palaeo.2010
.03.050.
Villas, E., Vennin, E., Álvaro, J.J., Hammann, W., Herrera,
Z.A., and Piovano, E.L., 2002, The Late Ordovician
carbonate sedimentation as a major triggering factor
of the Hirnantian glaciation: Bulletin de la Société
Géologique de France, v. 173, no. 6, p. 569–578, doi:
10.2113/173.6.569.
Vujovich, G.I., van Staal, C.R., and Davis, W., 2004, Age
constraints on the tectonic evolution and provenance of
the Pie de Palo complex, Cuyania composite terrane,
and the Famatinian orogeny in the Sierra de Pie de
Palo, San Juan, Argentina: Gondwana Research, v. 7,
p. 1041–1056, doi:10.1016/S1342-937X(05)71083-2.
Wang, K., Chatterton, B.D.E., Attrep, M.J., and Orth, C.J.,
1992, Iridium abundance maxima at the latest Ordo-
Interglacial
Interglacial
Hirnantian
HirnantianMid Mid
complanatus Zone
chitinozoan
zones
0246
-32 -30 -28 -26 -24
-4 -5 -6 -7
Interglacial
Interglacial
Hirnantian
Hirnantian
Mid
Mid
~~
~
~
~
100 m
δ18Owater
-1 0 1 2 3
Anticosti Is.
Other
late Boda
warming
early Boda
warming
mid Boda
cooling
complanatus Zone
pacificus Zone
complexus Zone
extraordinarius
Zone
persculptus
Zone HICE
Elkhorn
Paroveja
Moe
ascensus Zone
Sea Surface
Temperature (˚C)
30 32 34 36 38
Whitewater
Waynesville
Upper Second Bani
Lower Second Bani
Open Marine Conti-
nental
upper
offshore
lower offshore
shoreface
restricted marine
eustuarine
fluvio-glacial
Eustatic Sea level curve
-31 -29 -27 -25
M. persculptus
& Hirnantia fauna
-23
M. ojsuensis
& Hirnantia
fauna
HICE
Elkhorn
δ13Corg (‰)
HA
diamictites
Králov Dvůr
Králuv Dvůr
Zelkovice Zelkovice
KosovKosov
pacificus Zone
Bohemia20,21
acuminatus Zone
vesiculosus Zone
cyphus Zone
KosovKosov
Ascensus Zone
ascensus Zone
20 m
Anti-Atlas, Morocco22 Clumped Isotope Δ47 Paleothermometry23
HICE
Elkhorn
HA
HB
Paroveja
Moe
SaludusSaludusHallikuHallikuJonstorpJonstorpTuduinnaliTuduinnali
10 m
Viljandi, Estonia18,19 Kardla, Estonia18,19
Halliku
SaludusSaludus
KuldigaKuldiga
HallikuHalliku
Chitinozoan
Zones
T. anticostiensis C. scabra
B. gamachianaB. gamachianaC. rugataT. bergstroemi
PorkuniPirguVormsi
Regional
Stages
S.
tau.
10 m
δ13Ccarb(‰)
43210 765
δ13Ccarb(‰)
43210
δ13Ccarb(‰)
43210-1
Anna
HallikuHalliku AdilaAdila ArinaArina Ōhne
Ōhne
*
*
Kaugatuma, Estonia18,19
black shalegray shale siltstone cherty limestone
dolomitic mudstone argillaceous wackestone
calcareous mudstone packstone grainstone bioherms & grainstone limestone breccia-conglomerate
interbedded shale
& limestone
*
*Diplograptina
recurrence *FAD of Persculptus Zone
graptolites
Di
upper limit
common
Diplograptina
channel sandstone
sandstone tempestites
& mudstone
covered
International correlation of Late Ordovician stratigraphic sequences, chemostratigraphy & biostratigraphy
Melchin et al. Figure S1
Dicellograptus
anceps
Dicello.
complexus
Dicellograptus
ornatus
Styracograptus
uncinatus
Graptolite ZonesEpoch/Age
Llandovery
Rhuddanian
Akidograptus ascensus
Parakidograptus acuminatus
Cystograptus vesiculosus
Coronograptus cyphus
Metabolograptus persculptus
Metabolograptus
extraordinarius
Paraortho-
graptus
pacificus
Paraortho-
graptus
pacificus
Dicellograptus
gravis
Pleurograptus
linearis
443
445
Dicellograptus
complanatus
HirnantianKatian
AGE
(MA)
441
Late Ordovician
447
443.8
445.2
20 m
Mirny Creek, Siberia15,16,17
shelly
beds
Q-70
Q-67
Di
*
pacificus Zone
complexus Zone
Elkhorn
HICE
extraordinarius
Zone
persculptus
Zone
δ13Ccarb(‰)
43210 5-2 -1
BBA
A
C
C
D
DE
E
EB
EB
2 m
-33 -31 -29
Birkhill Shale
Birkhill Shale Hartfell Shale
Hartfell Shale
Dob’s Linn, Scotland12,13,14
δ13Corg (‰)
acuminatus
Zone
vesiculosus Zone
cyphus Zone
Di
*
*
*
HICE
ascensus Zone
Whirlpool SS
conodont
zones
TR1
TR2
TR4
TR5
Grindstone Mbr.
Laframboise *
20 m
Ellis Bay Fm.Ellis Bay Fm. Fox Point Fm.
Fox Point Fm.
43210-1
δ13Ccarb(‰)
Velleda
Lousy
Cove
S.C.
Mill Bay Vauréal Fm.Vauréal Fm.
-2 0 2 4
East End
West End
δ13Ccarb(‰)
TR5
Lafram.
TR4
TR2
TR1
B. gamachiana
20 m
Prinsta
TR3
S. taugoudeaui
H. crickmayi
A. ellisbayensis
HICE
Elkhorn
persculptus
Zone
Whitewater
Elkhorn
C6
C5
Waynesville
A. ordovicicusA. superbus
A. grandis
?
?
-2 -1 0 1 2
-1 0 1 2 3
HA
HB
HICE
Cincinnati, Ohio
Midcontinent Region8,9
Bruce Penninsula
Ontario
Elkhorn
QueenstonQueenston
Elkhorn
Unde-
fined
WhitewaterLiberty
Waynesville
Arnheim Elkhorn
Unde-
fined
WhitewaterLibertyWaynesvilleArnheim
δ13Ccarb(‰)
δ13Ccarb(‰)
δ13Ccarb(‰)
20 m
20 m
Anticosti Is., Que.10,11
HB
Whitewater
20 m
4 m
Manitoulin
TR3
extraordinarius
Zone
Hirnantia
Sp. A
Jin & Zhan
(2008)
Wangjiawan, Hubei1,2
Wufeng Fm.Wufeng Fm. Lungmachi Fm.Lungmachi Fm.
Hirnantia
beds
Hirnantia
beds
1.0 m
2.0 m
Lungmachi Fm.Lungmachi Fm.Wufeng Fm.
Wufeng Fm.
Kuanyinqiao Fm.
shelly
beds
shelly
beds
δ13Corg (‰)
-28.0 -26.0-30.0
δ13Corg (‰) δ34S (‰)
-20 0 20
-30 -29 -28
vesiculosus
Zone
cyphus
Zone
Nanbazi, Guizhou3
Hirnantia
beds
Hirnantia
beds
Di
Di
Fan et al.
2009
Gorjan et al.
2012
*
**
*
HICE
-30 -28 -26 -24
0 2 4
4.0 m
Vinini Creek, Nevada4,5 Blackstone River, Yukon4,5 Eleanor Lake, Arctic Canada6,7
Vinini FormationVinini Formation
*
*
*
*
Di
**
*
δ15Ntot (‰)
-1 0 1 2
δ13Corg (‰)
-32 -31 -30 -29 -28
δ13Corg (‰)
HICE
Elkhorn
δ15Ntot (‰)
δ13Corg (‰)
-4 -6 -8
εNd (443 Ma)
εNd (443 Ma)
HB
DR2013352 M.J. Melchin et al.
SIFigure1.InternationalcorrelationofLateOrdovicianstratigraphicsequences,
chemostratigraphy&biostratigraphybasedondatafromtherecentliterature.Superscripts
attachedtothelocalitynamesrefertothenumbereddatasourceslistedbelow.Timescalebased
onGTS2012(Cooper&Sadler2012,Melchinetal.2012).Hirnantianshadedlightblueand
narrowdeeporangebandinthemidHirnantiancorrespondstothemidHirnantianinterglacial
age.Seetextforfurtherdiscussion.
1.FanJunxuan,PengPing’anandMelchin,M.J.,2009,Carbonisotopesandeventstratigraphy
neartheOrdovician‐Silurianboundary,Yichang,SouthChina:Palaeogeography,
Palaeoclimatology,Palaeoecology,v.276,no.1‐4,p.160‐16
2.Gorjan,P.,Kaiho,K.,Fike,D.A.,andChenXu,2012,Carbon‐andsulfur‐isotopegeochemistryof
theHirnantian(LateOrdovician)Wangjiawan(Riverside)section,SouthChina:Global
correlationandenvironmentaleventinterpretation:Palaeogeography,Palaeoclimatology,
Palaeoecology,v.337‐338,p.14‐22.
3.YanDetian,ChenDiazhao,WangQingchen,andWangJianguo,2012,Predominanceof
stratifiedanoxicYangtzeSeainterruptedbyshort‐termoxygenationduringtheOrdo‐Silurian
transition:ChemicalGeology,v.291,p.69‐78.
4.LaPorte,D.F.,Holmden,C.,Patterson,W.P.,Loxton,J.D.,Melchin,M.J.,Mitchell,C.E.,Finney,S.
C.,andSheets,H.D.,2009,Localandglobalperspectivesoncarbonandnitrogencycling
duringtheHirnantianglaciation:Palaeogeography,Palaeoclimatology,Palaeoecology,v.276,
no.1‐4,p.182‐195.
5.Holmden,C.Mitchell,C.E.,LaPorte,D.F.,Patterson,W.P.,Melchin,M.J.,andFinney,S.C.,2013.
NdisotoperecordsoflateOrdoviciansea‐levelchange—implicationsforglaciationfrequency
andglobalstratigraphiccorrelation.Palaeogeography,Palaeoclimatology,Palaeoecology,
10.1016/j.palaeo.2013.05.014.
6.Melchin,M.J.,andHolmden,C.,2006,CarbonisotopechemostratigraphyinArcticCanada:Sea‐
levelforcingofcarbonateplatformweatheringandimplicationsforHirnantianglobal
correlation:Palaeogeography,Palaeoclimatology,Palaeoecology,v.234,p.186‐200.
7.Thisstudy.
8.Bergström,S.M.,Young,S.,andSchmitz,B.,2010,Katian(UpperOrdovician)δ13C
chemostratigraphyandsequencestratigraphyintheUnitedStatesandBaltoscandia:A
regionalcomparison:Palaeogeography,Palaeoclimatology,Palaeoecology,v.296,p.217‐234.
9.Bergström,S.M.,Kleffner,M.,Schmitz,B.,andCramer,B.D.,2011,Revisionofthepositionof
theOrdovician–SilurianboundaryinsouthernOntario:regionalchronostratigraphic
implicationsofδ13CchemostratigraphyoftheManitoulinFormationandassociatedstrata:
CanadianJournalofEarthSciences,v.48,no.11,p.1447‐1470.
10.Desrochers,A.,Farley,C.,Achab,A.,Asselin,E.,andRiva,J.F.,2010,Afar‐fieldrecordofthe
endOrdovicianglaciation:TheEllisBayFormation,AnticostiIsland,EasternCanada:
Palaeogeography,Palaeoclimatology,Palaeoecology,v.296,no.3‐4,p.248‐263.
11.Achab,A.,Asselin,E.,Desrochers,A.,Riva,J.F.,andFarley,C.,2011,Chitinozoan
biostratigraphyofanewUpperOrdovicianstratigraphicframeworkforAnticostiIsland,
Canada:GeologicalSocietyofAmericaBulletin,v.123,no.1‐2,p.186‐205.
12.Underwood,C.J.,Crowley,S.F.,Marshall,J.D.,andBrenchley,P.J.,1997,High‐resolution
carbonisotopestratigraphyofthebasalSilurianstratotype(Dob'sLinn,Scotland)andits
globalcorrelation:JournaloftheGeologicalSociety,v.154,no.4,p.709‐718.
13.Armstrong,H.A.,andCoe,A.L.,1997,Deep‐seasedimentsrecordthegeophysiologyofthe
lateOrdovicianglaciation:JournaloftheGeologicalSocietyofLondon,v.154,p.929‐934.
14.Melchin,M.J.,Holmden,C.andWilliams,S.H.,2003.Correlationofgraptolitebiozones,
chitinozoanbiozones,andcarbonisotopecurvesthroughtheHirnantian.In:G.L.Albanesi,
M.S.BeresiandS.H.Peralta(Editors),OrdovicianfromtheAndes.INSUEGO,SerieCorrelación
Geológica.ComunicarteEditorial,Tucumán,Argentina,pp.101‐104.
15.Koren’,T.N.,Oradovskaya,M.M.,Pylma,L.J.,Sobolevskaya,R.F.,andChugaeva,M.N.,1983,
TheOrdovicianandSilurianboundaryinthenortheastoftheU.S.S.R.,Leningrad,Nauka
Publishers,205p.:
16.Koren’,T.N.,andSobolevskaja,R.F.,2008,Theregionalstratotypesectionandpointforthe
baseoftheHirnantianStage(theuppermostOrdovician)atMirnyCreek,OmulevMountains,
NortheastRussia:EstonianJournalofEarthSciences,v.57,no.1,p.1‐10.
17.Kaljo,D.,Männik,P.,Martma,T.,andNõlvak,J.,2012,MoreabouttheOrdovician‐Silurian
transitionbedsatMirnyCreek,OmulevMountains,NERussia:carbonisotopesand
conodonts:EstonianJournalofEarthSciences,v.61,no.4,p.277‐294.
18.Kaljo,D.,Hints,L.,Martma,T.,Nõlvak,J.,andOraspõld,A.,2004,LateOrdoviciancarbon
isotopetrendinEstonia,itssignificanceinstratigraphyandenvironmentalanalysis:
Palaeogeography,Palaeoclimatology,Palaeoecology,v.210,no.2‐4,p.165‐185.
19.Ainsaar,L.,Kaljo,D.,Martma,T.,Meidla,T.,Männik,P.,Nõlvak,J.,andTinn,O.,2010,Middle
andUpperOrdoviciancarbonisotopechemostratigraphyinBaltoscandia:Acorrelation
standardandcluestoenvironmentalhistory:Palaeogeography,Palaeoclimatology,
Palaeoecology,v.294,no.3–4,p.189‐201.
20.Štorch,P.,2006,Faciesdevelopment,depositionalsettingsandsequencestratigraphyacross
theOrdovician–Silurianboundary:anewperspectivefromtheBarrandianareaoftheCzech
Republic:GeologicalJournal,v.41,no.2,p.163‐192.
21.Mitchell,C.E.,Štorch,P.,Holmden,C.,Melchin,M.J.,andGutierrez‐Marco,J.C.,2011,New
stableisotopedataandfossilsfromtheHirnantianStageinBohemiaandSpain:implications
forcorrelationandpaleoclimate,inGutierrez‐Marco,J.C.,Rábano,I.,andGarcía‐Bellido,D.,
eds.,OrdovicianoftheWorld:Madrid,CuadernosdelMuseoGeominero,14.Instituto
GeológicoyMinerodeEspaña,p.371‐378.
22.Loi,A.,Ghienne,J.F.,Dabard,M.P.,Paris,F.,Botquelen,A.,Christ,N.,Elaouad‐Debbaj,Z.,
Gorini,A.,Vidal,M.,Videt,B.,andDestombes,J.,2010,TheLateOrdovicianglacio‐eustatic
recordfromahigh‐latitudestorm‐dominatedshelfsuccession:TheBouIngarfsection(Anti‐
Atlas,SouthernMorocco):Palaeogeography,Palaeoclimatology,Palaeoecology,v.296,no.3‐4,
p.332‐358
23.Finnegan,S.,Bergmann,K.,Eiler,J.M.,Jones,D.S.,Fike,D.A.,Eisenman,I.,Hughes,N.C.,
Tripati,A.K.,andFischer,W.W.,2011,TheMagnitudeandDurationofLateOrdovician‐Early
SilurianGlaciation:Science,v.331,no.6019,p.903‐906.
Meterage Calcite TOC TN 13Ccarb 18Ocarb 14Corg 15NTN
m wt.% wt.% wt.% (‰, V-PDB) (‰, V-PDB) (‰, V-PDB) (‰, AIR)
30.0 70.4 1.4 0.041 -1.59 -7.92 -29.87 2.05
31.8 59.3 1.9 0.050 -0.16 -4.80 -29.94 1.84
34.3 93.5 0.2 0.012 0.54 -5.34 -29.24 0.50
35.8 88.8 0.4 0.023 -0.47 -9.58 -29.61 1.56
36.9 90.8 0.1 0.014 0.55 -6.34 -29.83 1.08
37.2 80.2 0.5 0.015 0.65 -5.99 -30.27 2.32
39.0 76.9 0.8 0.038 -0.26 -6.35 -30.23 0.98
40.2 82.8 0.4 0.012 0.04 -6.20 -30.15 2.07
41.3 98.3 0.1 0.003 -0.52 -5.17 -29.85 1.52
41.6 78.5 1.1 0.048 -1.50 -6.28 -30.09 1.55
43.7 67.5 2.9 0.113 0.21 -5.47 -29.98 0.81
44.7 96.0 0.2 0.010 -1.24 -7.59 -29.46 1.67
45.5 92.6 0.4 0.017 0.34 -5.27 -30.25 1.34
45.7 31.7 3.3 0.135 0.90 -5.84 -29.73 0.91
46.0 70.8 3.5 0.091 0.61 -5.74 -30.47 0.44
46.2 29.2 4.5 0.155 0.75 -5.28 -28.01 1.29
46.4 22.6 12.1 0.247 0.56 -5.24 -27.13 0.67
46.5 41.4 4.9 0.177 1.26 -7.20 -27.45 1.73
46.6 46.4 1.8 0.094 1.00 -6.56 -27.24 2.51
46.8 57.9 0.7 0.049 1.63 -6.33 -27.77 2.56
47.0 73.0 0.7 0.035 1.38 -7.43 -27.72 3.11
47.5 44.6 0.7 0.065 0.82 -5.05 -28.45 3.81
47.9 49.1 0.8 0.064 0.53 -5.07 -28.62 3.37
48.4 51.7 0.8 0.063 0.61 -5.12 -28.66 3.38
49.1 77.4 0.8 0.040 0.27 -5.27 -28.64 1.38
49.4 50.8 1.9 0.103 0.29 -5.31 -27.93 1.41
49.6 21.3 8.3 0.300 0.31 -5.47 -29.57 0.18
50.0 8.7 4.9 0.179 -1.24 -6.31 -30.43 0.41
51.0 17.7 4.3 0.158 -1.46 -5.84 -30.36 0.45
51.5 21.0 5.4 0.185 -0.97 -6.25 -30.34 0.47
52.0 61.3 4.8 0.174 -1.28 -5.26 -29.94 0.87
52.5 57.3 3.8 0.143 -1.48 -4.83 -30.34 0.88
Meterage Calcite TOC TN 13Ccarb 18Ocarb 14Corg 15NTN
m wt.% wt.% wt.% (‰, V-PDB) (‰, V-PDB) (‰, V-PDB) (‰, AIR)
49.50 92.7 0.26 0.013 -0.01 -3.28 -29.67 1.19
50.00 57.4 1.39 0.045 -1.18 -7.54 -30.07 1.04
50.50 80.1 2.51 0.067 -0.82 -7.59 -30.46 1.00
51.00 58.0 1.50 0.039 -0.58 -7.56 -30.25 0.50
51.50 83.7 1.64 0.043 -0.40 -7.35 -30.55 0.53
52.00 86.4 0.53 0.023 -0.37 -3.24 -29.95 1.22
52.50 88.5 0.32 0.015 -0.13 -3.20 -30.46 1.21
52.90 56.6 1.28 0.007 -1.60 -11.66 -26.68 -
53.00 94.8 0.13 0.007 -0.06 -3.42 -29.71 1.21
53.50 81.3 1.04 0.035 -1.25 -6.49 -30.29 0.72
54.00 87.9 0.51 0.022 -0.68 -5.72 -30.06 0.72
54.50 97.6 0.12 0.005 -2.09 -8.72 -30.35 -0.41
54.90 98.4 0.10 0.004 -0.63 -8.06 -29.30 0.76
55.00 33.2 9.46 0.273 0.77 -4.96 -28.53 0.13
55.10 51.3 3.43 0.123 0.68 -6.32 -28.31 0.51
TABLE S1. CARBON AND NITROGEN ISOTOPE, % CALCITE, % TOTAL ORGANIC CARBON (TOC) & % TOTAL
NITROGEN (TN) DATA FROM TRURO, CANADIAN ARCTIC
TABLE S2. CARBON AND NITROGEN ISOTOPE, % CALCITE, % TOTAL ORGANIC CARBON (TOC) & % TOTAL
NITROGEN (TN) DATA FROM ELEANOR LAKE, CANADIAN ARCTIC
Supp data tables1-4.xlsx
55.20 48.5 6.26 0.192 2.19 -4.04 -27.61 0.55
55.30 62.0 0.31 0.028 0.48 -9.36 -26.34 2.34
55.35 70.6 0.09 no data 1.69 -5.77 -24.05 no data
55.40 79.6 0.06 0.010 -0.07 -6.20 -24.85 3.79
55.60 78.4 0.11 0.010 0.75 -6.19 -25.12 2.94
55.80 67.0 0.09 no data 1.51 -4.91 -23.88 no data
55.90 64.6 0.08 0.020 1.53 -5.14 -24.21 3.38
56.20 55.6 0.07 no data 1.25 -3.52 -24.54 no data
56.40 53.4 0.05 0.030 1.17 -4.12 -25.23 2.98
56.80 49.8 0.05 0.020 0.61 -4.61 -24.43 3.34
56.55 56.6 0.09 no data 0.62 -3.90 -25.97 no data
56.65 53.8 0.08 no data 0.34 -4.41 -24.88 no data
56.90 63.1 0.05 no data 0.69 -3.64 -24.71 no data
57.00 37.0 0.06 no data 0.43 -4.55 -25.54 no data
57.20 44.5 0.05 0.020 0.88 -4.24 -26.31 3.4
57.40 36.6 0.25 0.030 1.50 -4.57 -28.55 2.22
57.50 38.7 0.33 0.020 0.88 -4.96 -28.30 3.45
57.55 10.9 0.43 0.050 -3.01 -13.92 -29.26 3.86
57.60 29.0 2.14 0.080 -3.98 -14.23 -28.62 2.32
57.70 57.9 1.25 0.030 0.10 -6.59 -29.05 2.34
57.80 20.2 3.21 0.140 -0.34 -7.73 -29.41 0.54
57.90 no data 11.94 no data no data no data -29.83 no data
58.00 32.6 9.65 0.340 -2.05 -12.94 -30.27 0.39
58.10 no data 2.97 no data no data no data -30.13 no data
58.20 no data 6.18 no data no data no data -30.22 no data
58.30 8.6 7.24 0.250 -1.04 -10.37 -30.44 0.23
58.70 28.3 4.45 0.160 -0.11 -8.45 -30.65 0.23
59.00 41.9 3.73 0.130 1.77 -7.17 -30.74 0.48
Meterage Calcite TOC TN d13Ccarb d18Ocarb d14Corg d15NTN
m wt.% wt.% wt.% (‰, V-PDB) (‰, V-PDB) (‰, V-PDB) (‰, AIR)
25.4 81.3 0.3 0.015 -0.74 -6.11 -28.89 1.98
25.5 64.9 1.0 0.043 -0.34 -5.59 -28.48 0.73
25.6 9.5 11.1 0.409 1.24 -6.48 -29.02 -0.71
25.8 71.9 2.1 0.066 0.67 -5.00 -29.21 -0.57
26.0 91.1 2.2 0.060 0.72 -5.70 -29.32 -1.17
26.3 81.3 2.1 0.071 1.43 -5.87 -29.39 -0.12
26.5 19.6 19.6 0.620 1.29 -9.70 -29.24 -1.34
26.8 17.4 6.7 0.272 -0.94 -8.10 -29.73 -0.32
27.0 91.6 2.4 0.050 0.59 -5.17 -30.20 -1.55
27.3 54.6 6.3 0.249 0.41 -6.67 -30.31 -0.57
27.5 95.0 1.7 0.046 -1.06 -6.06 -30.49 -0.90
27.8 16.5 9.1 0.358 -1.50 -6.59 -30.48 -0.21
28.0 95.1 1.9 0.033 -0.91 -6.67 -30.50 -1.57
28.3 15.2 11.6 0.392 -0.52 -7.53 -30.42 -0.90
28.5 90.7 1.5 0.049 -0.69 -5.43 -30.73 -1.18
28.8 55.1 2.7 0.085 0.34 -5.34 -30.72 -1.69
29.0 95.4 1.1 0.026 -1.67 -5.83 -30.76 -1.61
29.3 41.1 2.1 0.139 -0.22 -10.00 -30.69 -0.51
29.5 66.5 1.0 0.025 -0.36 -5.76 -30.89 -1.75
29.8 88.8 1.5 0.032 -0.97 -5.64 -30.90 -1.40
30.0 56.9 1.3 0.045 -0.47 -5.11 -30.74 -0.84
30.5 97.0 0.7 0.017 -5.17 -6.51 -30.99 -0.87
31.0 85.5 1.3 0.054 -0.71 -5.41 -30.18 0.35
TABLE S3. CARBON AND NITROGEN ISOTOPE, % CALCITE, % TOTAL ORGANIC CARBON (TOC) & % TOTAL
NITROGEN (TN) DATA FROM CAPE PHILLIPS SOUTH, CANADIAN ARCTIC
31.5 98.3 0.2 0.010 -1.06 -5.37 -30.82 -1.47
32.0 86.2 2.1 0.067 -1.34 -5.63 -30.71 -1.37
32.5 93.5 2.1 0.034 -0.83 -5.46 -30.89 -1.29
33.0 82.0 4.2 0.047 -0.35 -5.61 -30.52 -0.68
33.5 99.9 0.0 0.000 -0.08 -5.37 -30.30 -0.40
34.0 89.8 2.6 0.067 -1.65 -6.68 -30.28 -0.87
34.5 92.2 1.4 0.046 -0.98 -6.85 -30.36 -0.65
35.0 88.9 0.9 0.022 -0.54 -5.92 -30.25 -0.82
36.0 84.4 3.3 0.081 -0.54 -6.83 -30.42 -0.77
36.5 87.3 1.7 0.023 -0.38 -6.23 -30.22 -0.66
37.0 70.5 2.0 0.036 0.03 -7.44 -30.12 -0.35
38.0 89.4 0.6 0.033 -0.34 -6.31 -30.32 -0.79
38.5 65.8 1.3 0.070 -0.30 -7.25 -29.96 -0.34
39.0 83.4 1.3 0.042 0.67 -7.73 -30.24 0.65
39.5 97.6 0.4 0.014 -0.32 -6.42 -30.50 0.34
40.0 90.1 1.0 0.032 1.23 -6.06 -30.13 1.33
40.5 97.7 0.2 0.008 -1.41 -5.60 -30.35 0.32
41.0 96.4 0.2 0.007 -1.92 -6.65 -29.64 0.55
41.5 92.3 0.2 0.014 0.05 -5.75 -29.13 2.87
41.7 19.9 5.3 0.163 -1.49 -2.92 -29.86 -1.26
42.0 60.3 6.0 0.142 -1.50 -6.39 -30.04 -1.43
42.5 94.6 0.2 0.016 0.41 -5.02 -30.62 -1.10
43.0 6.5 4.8 0.319 -0.33 -6.31 -29.99 -1.45
43.5 56.4 5.6 0.136 -2.10 -6.09 -30.07 -1.58
43.7 97.4 0.3 0.010 -2.38 -5.41 -29.58 -0.42
44.2 43.9 5.7 0.152 0.31 -5.61 -30.22 -1.25
44.6 99.7 0.1 0.002 -1.99 -5.70 -29.48 0.80
45.2 93.5 0.3 0.021 -0.38 -5.62 -29.22 2.64
sample Meterage Calcite TOC TN d13Ccarb d18Ocarb d14Corg d15NTN
number m wt.% wt.% wt.% (‰, V-PDB) (‰, V-PDB) (‰, V-PDB) (‰, AIR)
Levin (samples from 2008 interpolated into 2006 measured section
2006-33 29.00 10.20 0.18 -26.62
2006-32 28.00 9.74 0.15 -27.34
2006-31 27.10 11.16 0.15 -27.68
2006-30 26.10 7.69 0.17 -27.30
2006-29 25.10 6.93 0.19 -27.65
2006-28 24.10 8.30 0.15 -27.29
2006-27 23.10 7.02 0.17 -27.40
2006-26 22.10 12.62 0.17 -27.71
2006-25 21.10 8.01 0.20 -27.40
2006-24 20.05 10.85 0.16 -27.45
2006-23 19.15 5.65 0.19 -27.50
2006-22 18.00 7.52 0.16 -27.69
2006-21 17.00 9.51 0.19 -27.31
2006-20 16.00 6.00 0.19 -27.44
2006-19 15.00 6.12 0.16 -27.89
2006-18 14.00 12.07 0.16 -27.89
2006-17 13.00 10.34 0.16 -27.61
2006-16 11.90 6.76 0.22 -27.85
2006-15 10.90 11.42 0.18 -28.33
2008-50 10.54 5.38 0.30 0.052 -29.00 3.26
2008-49 10.23 5.68 0.47 0.066 -29.51 3.15
N not measured for 2006 samples
TABLE S4. CARBON & NITROGEN ISOTOPE, % TOTAL ORGANIC CARBON (TOC) & TOTAL NITROGEN (TN) DATA FROM LEVIN
AND ZADNÍ TŘEBÁŇ SECTIONS, CZECH REPUBLIC
N not measured for 2006 samples
isotopes in carbonate
not measured
2008-48 9.99 4.24 0.40 0.067 -29.10 3.05
2008-47 9.90 6.47 0.35 0.051 -29.07 3.15
2006-14 9.90 5.38 0.30 -28.53
2008-46 9.35 11.16 0.26 0.028 -28.15 3.36
2008-45 8.80 6.54 0.20 0.027 -27.91 3.64
2008-44 8.22 9.48 0.20 0.015 -28.04 4.06
2008-43 7.80 7.98 0.19 0.028 -27.83 3.35
2006-13 7.40 5.09 0.18 -27.88
2008-21 7.39 6.31 0.25 0.054 -28.71 3.48
2008-22 7.29 5.94 0.30 0.061 -29.01 3.27
2008-23 7.06 4.35 0.26 0.053 -29.16 3.29
2008-24 6.83 5.91 0.27 0.050 -28.67 3.37
2008-25 6.60 5.28 0.37 0.056 -28.69 2.82
2006-12 6.50 7.90 0.27 -28.48
2008-26 6.37 4.79 0.29 0.055 -28.74 3.51
2008-27 6.15 4.59 0.33 0.050 -28.81 3.45
2008-28 5.92 8.35 0.32 0.052 -28.67 3.62
2008-29 5.69 4.73 0.32 0.052 -28.70 3.25
2006-11 5.50 4.14 0.26 -28.77
2008-30 5.46 3.97 0.38 0.057 -29.08 3.21
2008-31 5.23 9.06 0.27 0.055 -28.92 2.91
2008-32 5.01 6.93 0.33 0.045 -28.76 2.25
2008-33 4.78 7.52 0.32 0.054 -29.03 3.14
2008-34 4.55 6.21 0.41 0.066 -29.22 2.73
2006-10 4.50 5.24 0.40 -28.78
2008-35 4.49 6.99 0.32 0.044 -27.98 2.94
2008-36 4.46 16.40 0.21 0.027 -28.17 3.40
2008-37 4.32 6.53 0.35 0.059 -28.39 2.88
2008-38 4.07 9.42 0.39 0.049 -22.53 2.99
2006-9 4.00 5.60 0.24 -26.33
2008-39 3.92 8.76 0.23 0.045 -25.85 2.93
2008-40 3.82 8.53 0.22 0.045 -27.68 2.75
2008-41 3.57 9.21 0.24 0.046 -27.68 2.75
2008-42 3.45 12.75 0.22 0.051 -27.17 3.15
2006-8 3.40 8.50 0.16 -25.73
2008-1 3.32 12.47 0.21 0.058 -28.05 2.89
2006-1 3.20 22.40 0.10 -26.03
2008-2 3.20 21.04 0.16 0.047 -26.97 2.98
2008-3 3.13 52.66 0.12 0.029 -25.91 3.12
2006-2 3.10 60.79 0.07 -27.18
2008-4 3.07 51.72 0.11 0.031 -27.95 3.29
2006-3 3.00 7.09 0.07 -25.34
2008-5 3.00 12.23 0.13 0.049 -25.31 3.01
2008-6 2.94 5.94 0.10 0.063 -29.32 2.95
2008-7 2.81 9.31 0.19 0.060 -29.34 3.06
2006-4 2.60 7.24 0.12 -25.58
2008-8 2.57 10.85 0.20 0.063 -29.35 3.00
2008-9 2.32 6.41 0.15 0.065 -29.61 2.90
2008-10 2.07 6.29 0.13 0.058 -27.15 2.81
2006-5 2.00 4.30 0.12 -26.57
2008-11 1.82 5.29 0.17 0.063 -26.10 2.71
2008-12 1.58 16.39 0.23 0.047 -29.45 2.87
2008-13 1.33 5.94 0.19 0.056 -28.08 2.72
2008-14 1.08 5.32 0.15 0.062 -29.59 3.12
2006-6 1.00 4.65 0.11 -27.46
2008-15 0.83 4.45 0.17 0.065 -29.45 3.16
2008-16 0.58 4.33 0.16 0.066 -29.49 3.08
2008-17 0.34 3.38 0.20 0.069 -29.49 3.50
Pernik Bed
upper diamictite bed
lower diamictite bed
2008-18 0.09 3.59 0.21 0.061 -29.47 3.08
2006-7 0.00 3.97 0.11 -28.57
2008-19 -0.53 6.73 0.17 0.051 -28.12 2.93
2008-20 -1.15 5.06 0.16 0.067 -29.59 3.36
Zadní Třebáň
2008-7 7.80 2.91 0.18 - -27.53 -
2008-6 7.10 3.21 0.19 - -27.48 -
2008-1 6.55 3.21 0.14 - -27.86 -
2008-2 6.46 4.89 0.27 0.051 -28.53 2.09
2008-3 6.30 6.36 0.25 0.052 -27.48 2.56
2008-4 6.05 5.66 0.29 0.059 -28.37 3.20
2008-5 5.90 5.67 0.25 0.054 -28.25 3.59
2008-8 5.70 5.83 0.31 0.063 -28.33 3.39
2008-9 5.50 5.36 0.28 0.052 -28.30 2.60
2008-10 5.30 4.78 0.28 0.050 -28.13 2.56
2008-11 5.10 6.43 0.26 0.059 -28.28 3.22
2008-12 4.90 5.95 0.30 0.060 -28.14 3.41
2008-13 4.70 5.24 0.36 0.066 -28.02 3.35
2008-14 4.55 6.47 0.29 0.062 -29.04 2.38
2008-15 4.50 7.46 0.28 0.052 -27.70 3.53
2008-16 4.42 4.64 0.25 0.047 -28.17 3.80
2008-17 4.38 3.66 0.21 - -28.45 -
2008-18 4.32 5.70 0.25 0.044 -27.46 2.48
2008-19 4.20 5.32 0.25 0.056 -28.05 3.38
2008-20 4.00 6.02 0.25 0.054 -27.93 3.64
2008-21 3.80 4.53 0.23 0.048 -27.98 3.38
2008-22 3.60 5.12 0.23 0.044 -27.98 2.47
2008-23 3.40 5.53 0.23 0.049 -27.75 3.56
2008-24 3.20 6.35 0.28 0.046 -27.73 2.51
2008-25 3.00 6.26 0.21 0.052 -27.29 3.93
2008-26 2.80 5.13 0.29 0.064 -27.55 2.92
2008-27 2.60 5.15 0.28 0.067 -27.62 3.48
2008-28 2.40 5.70 0.29 0.067 -27.58 3.28
2008-29 2.20 4.68 0.39 0.070 -28.40 3.28
2008-30 2.00 6.78 0.35 0.064 -28.37 3.87
2008-31 1.95 8.33 0.26 0.056 -27.70 3.35
2008-38 1.88 4.84 0.23 0.057 -27.66 2.36
2008-32 1.85 8.97 0.17 0.061 -27.98 3.01
2008-33 1.83 8.76 0.19 0.063 -27.86 2.90
2008-34 1.80 10.40 0.24 0.064 -27.77 3.02
2008-35 1.75 50.80 0.14 0.029 -28.50 2.76
2008-36 1.60 50.35 0.12 0.030 -28.24 3.28
2008-37 1.55 6.86 0.18 0.058 -28.28 3.20
2008-39 1.50 17.18 0.19 0.052 -28.48 3.10
2008-40 1.30 5.05 0.17 0.056 -28.66 3.08
2008-41 1.10 7.00 0.17 0.044 -28.42 2.87
2008-42 0.90 7.93 0.17 0.062 -28.57 3.27
2008-43 0.70 5.84 0.15 0.059 -28.88 2.86
2008-44 0.50 5.94 0.17 0.064 -28.84 3.11
2008-45 0.30 17.30 0.22 0.058 -28.82 2.86
2008-46 0.00 4.97 0.18 0.064 -28.72 3.04
- below limit for accurate measurement
burrowed calc mdst Pernik
lower Pernik
Upper Diamictite Bed
Lower Diamictite Bed
top Pernik Bed with M. ojsuensis
prob. fault repeat #32-33
Chondrites bearing bed
Mucronaspis beds
Mucronaspis beds
Melchin et al.
1670 Geological Society of America Bulletin, November/December 2013
vician mass extinction horizon, Yangtze Basin, China:
Terrestrial or extraterrestrial?: Geology, v. 20, p. 39–
42, doi:10.1130/0091-7613(1992)020<0039:IAMATL
>2.3.CO;2.
Wang, T., Hong, D., Jahn, B., Tong, Y., Wang, Y., Han, B.,
and Wang, X., 2006, Timing, petrogenesis, and set-
ting of Paleozoic synorogenic intrusions from the Altai
Mountains, Northwest China: Implications for the tec-
tonic evolution of an accretionary orogen: The Journal
of Geology, v. 114, p. 735–751, doi:10.1086/507617.
Wang, X., and Chai, Z., 1989, Terminal Ordovician mass
extinction and discovery of iridium anomaly—An ex-
ample from the Ordovician-Silurian boundary section,
eastern Yangtze Gorges area, China, in Progress of
Geosciences of China 1985–1988, Volume III: Beijing,
Geological Publishing House, p. 11–16.
Webb, B.C., Rushton, A.W.A., and White, D.E., 1993, Clas-
sic Areas of British Geology: Moffatdale and the Up-
per Ettrick Valley: Description of the Solid Geology
of Parts of Sheets NT 10, 11, 20 and 21: London, Her
Majesty’s Stationery Offi ce for the British Geological
Survey, scale 1:25,000.
Wignall, P.B., 1991, Model for transgressive black shales?:
Geology, v. 19, p. 167–170.
Wignall, P.B., 1994, Black Shales: Oxford, UK, Oxford Uni-
versity Press, 127 p.
Wilde, P., 1987, Model of progressive ventilation of the
late Precambrian–early Paleozoic ocean: American
Journal of Science, v. 287, p. 442–459, doi:10.2475
/ajs.287.5.442.
Wilde, P., 1991, Oceanography in the Ordovician, in Barnes,
C.R., and Williams, S.H., eds., Advances in Ordovician
Geology: Geological Survey of Canada Paper 90–9,
p. 283–298.
Wilde, P., and Berry, W.B.N., 1984, Destabilization of the
oceanic density structure and its signifi cance to marine
“extinction” events: Palaeogeography, Palaeoclimatol-
ogy, Palaeoecology, v. 48, p. 143–162, doi:10.1016
/0031-0182(84)90041-5.
Wilde, P., Berry, W.B.N., Quinby-Hunt, M.S., Orth, C.J.,
Quintana, L.R., and Gilmore, J.S., 1986, Iridium abun-
dances across the Ordovician-Silurian stratotype: Sci-
ence, v. 233, p. 339–341, doi:10.1126/science.233
.4761.339.
Wilde, P., Lyons, T.W., and Quinby-Hunt, M.S., 2004, Or-
ganic carbon proxies in black shales: Molybdenum:
Chemical Geology, v. 206, p. 167–176, doi:10.1016
/j.chemgeo.2003.12.005.
Williams, G.E., 1991, Milankovitch-band cyclicity in bed-
ded halite deposits contemporaneous with Late Ordovi-
cian–Early Silurian glaciation, Canning Basin, Western
Australia: Earth and Planetary Science Letters, v. 103,
p. 143–155, doi:10.1016/0012-821X(91)90156-C.
Williams, S.H., 1982, The Late Ordovician graptolite fauna
of the Anceps Bands at Dob’s Linn, southern Scotland:
Geologica et Palaeontologica, v. 16, p. 29–56.
Williams, S.H., 1983, The Ordovician-Silurian boundary
graptolite fauna of Dob’s Linn, southern Scotland:
Palaeon tology, v. 26, p. 605–639.
Williams, S.H., and Barnes, C.R., eds., 1988, Program and
Abstracts: Fifth International Symposium on the Ordo-
vician System, August 9–12, 1988: Memorial Univer-
sity, St. John’s, Newfoundland, 117 p.
Witzke, B.J., 1987, Models for circulation patterns in epi-
continental seas applied to Paleozoic facies of North
America craton: Paleoceanography, v. 2, p. 229–248,
doi:10.1029/PA002i002p00229.
Yan, D.T., Chen, D.Z., Wang, Q.C., Wang, J.G., and Wang,
Z.Z., 2009, Carbon and sulfur isotopic anomalies across
the Ordovician-Silurian boundary on the Yangtze plat-
form, South China: Palaeogeography, Palaeoclima-
tology, Palaeoecology, v. 274, p. 32–39, doi:10.1016
/j.palaeo.2008.12.016.
Yan, D.T., Chen, D.Z., Wang, Q.C., and Wang, J.G., 2012,
Predominance of stratifi ed anoxic Yangtze Sea inter-
rupted by short-term oxygenation during the Ordo-
Silurian transition: Chemical Geology, v. 291, p. 69–78,
doi:10.1016/j.chemgeo.2011.09.015.
Yapp, C.J., and Poths, H., 1992, Ancient atmospheric CO2
pressures inferred from natural goethites: Nature,
v. 355, p. 342–344, doi:10.1038/355342a0.
Young, G.A., Minter, W.E.L., and Theron, J.N., 2004, Geo-
chemistry and palaeogeography of Upper Ordovician
glaciogenic sedimentary rocks in the Table Mountain
Group, South Africa: Palaeogeography, Palaeoclima-
tology, Palaeoecology, v. 214, p. 323–345.
Young, S.A., Saltzman, M.R., Bergström, S.M., Leslie, S.A.,
and Chen, X., 2008, Paired 13Ccarb and 13Corg records of
Upper Ordovician (Sandbian-Katian) carbonates in
North America and China: Implications for paleocean-
ographic change: Palaeogeography, Palaeoclimatol-
ogy, Palaeoecology, v. 270, p. 166–178, doi:10.1016
/j.palaeo.2008.09.006.
Young, S.A., Saltzman, M.R., Foland, K.A., Linder, J.S., and
Kump, L.R., 2009, A major drop in seawater Sr-87/Sr-86
during the Middle Ordovician (Darriwilian): Links to
volcanism and climate?: Geology, v. 37, p. 951–954.
Young, S.A., Saltzman, M.R., Ausich, W.I., Desrochers, A.,
and Kaljo, D., 2010, Did changes in atmospheric CO2
coincide with latest Ordovician glacial-interglacial
cycles?: Palaeogeography, Palaeoclimatology, Palaeo-
ecology, v. 296, p. 376–388, doi:10.1016/j.palaeo.2010
.02.033.
Zalasiewicz, J., Williams, M., Miller, M., Page, A., and
Blackett, E., 2007, Early Silurian (Llandovery) grap-
tolites from central Saudi Arabia: First documented
record of Telychian faunas from the Arabian Peninsula:
GeoArabia, v. 12, p. 15–36.
Zhan, R.B., and Jin, J., 2007, Ordovician–Early Silurian
(Llandovery) Stratigraphy and Palaeontology of the
Upper Yangtze Platform, South China: Beijing, Sci-
ence Press, 169 p.
Zhang, S.X., and Barnes, C.R., 2002, A New Llandovery
(Early Silurian) Conodont Biozonation and Conodonts
from the Bescie, Merrimack, and Gun River Formations,
Anticosti Island, Quebec: Paleontological Society Mem-
oir 57 (Journal of Paleontology, v. 76, supplement), 46 p.
Zhang, S.X., and Barnes, C.R., 2007, Late Ordovician to
Early Silurian conodont faunas from the Kolyma ter-
rane, Omulev Mountains, northeast Russia, and their
paleobiogeographic affi nity: Journal of Paleontology,
v. 81, p. 490–512, doi:10.1666/05077.1.
Zhang, T., Shen, Y., Zhan, R., Shen, S., and Chen, X., 2009,
Large perturbations of the carbon and sulfur cycle
asso ciated with the Late Ordovician mass extinction in
South China: Geology, v. 37, p. 299–302, doi:10.1130
/G25477A.1.
Zhang, T.G., Trela, W., Jiang, S.Y., Nielsen, J.K., and Shen,
Y.N., 2011, Major oceanic redox condition change cor-
related with the rebound of marine animal diversity
during the Late Ordovician: Geology, v. 39, p. 675–
678, doi:10.1130/G32020.1.
Zhang, Y.D., Chen, X., Yu, G.H., Goldman, D., and Liu, X.,
2007, Ordovician and Silurian Rocks of Northwest
Zhejiang and Northeast Jiangxi Provinces, SE China:
Hefei, University of Science and Technology of China
Press, 189 p.
Zhou, L., Su, J., Huang, J.H., Yan, J.X., Xie, X.N., Gao, S.,
Dai, M.N., and Tonger, 2011, A new paleoenviron-
mental index for anoxic events—Mo isotopes in black
shales from Upper Yangtze marine sediments: Science
China–Earth Sciences, v. 54, no. 7, p. 1024–1033,
doi:10.1007/s11430-011-4188-z.
Zhou, L., Wignall, P., Su, J., Feng, Q., Xie, S., Zhao, L., and
Huang, J., 2012, U/Mo ratios and δ98/95Mo as local and
global redox proxies during mass extinction events:
Chemical Geology, v. 324–325, p. 99107.
Žigaitė, Z., Joachimski, M.M., Lehnert, O., and Brazaus-
kas, A., 2010, δ18O composition of conodont apatite
indicates climatic cooling during the middle Pridoli:
Palaeogeography, Palaeoclimatology, Palaeoecology,
v. 294, p. 242–247, doi:10.1016/j.palaeo.2010.03.033.
SCIENCE EDITOR: J. BRENDAN MURPHY
MANUSCRIPT RECEIVED 1 NOVEMBER 2012
REVISED MANUSCRIPT RECEIVED 5 JUNE 2013
MANUSCRIPT ACCEPTED 14 AUGUST 2013
Printed in the USA
